Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberCN1781268 A
Publication typeApplication
Application numberCN 200480011254
PCT numberPCT/US2004/007848
Publication date31 May 2006
Filing date15 Mar 2004
Priority date27 Mar 2003
Also published asCA2520150A1, DE602004027897D1, EP1614234A1, EP1614234A4, EP1614234B1, US6958986, US20040028018, WO2004095734A1
Publication number200480011254.8, CN 1781268 A, CN 1781268A, CN 200480011254, CN-A-1781268, CN1781268 A, CN1781268A, CN200480011254, CN200480011254.8, PCT/2004/7848, PCT/US/2004/007848, PCT/US/2004/07848, PCT/US/4/007848, PCT/US/4/07848, PCT/US2004/007848, PCT/US2004/07848, PCT/US2004007848, PCT/US200407848, PCT/US4/007848, PCT/US4/07848, PCT/US4007848, PCT/US407848
Inventors约瑟夫·比布·凯恩
Applicant哈里公司
Export CitationBiBTeX, EndNote, RefMan
External Links: SIPO, Espacenet
Wireless communications system with enhanced time slot
CN 1781268 A
Abstract  translated from Chinese
无线通信网络(10)可包括多个移动节点(12),所述每个移动节点包括一个无线收发器(14)和一个用于控制无线收发器的控制器(18)。 The wireless communication network (10) may include a plurality of mobile nodes (12), wherein each mobile node comprising a wireless transceiver (14) and a controller (18) for controlling the wireless transceiver. 控制器(18)还可用于调度相应的半永久时隙,以便与相邻移动节点建立通信链路,用于在它们之间传送数据,所述数据具有不同的优先级。 The controller (18) may also be used for scheduling respective semi-permanent time slot to establish a communication link with a neighboring mobile nodes, for transferring data therebetween, the data having different priorities. 控制器(18)还可确定每个通信链路的每个数据优先级的相应链路使用度量,并根据链路使用度量和数据优先级,调度按需分配的时隙,以便与相邻移动节点建立额外的通信链路,用于在它们之间传送数据。 The controller (18) can be determined for each data in each communication link using link corresponding priority measure, and according to the link using metrics and data priority scheduling on-demand time slots to an adjacent mobile Additional nodes establish a communication link for transmitting data between them. 在一些实施例中,无线通信网络(10)还提供增强的干扰回避和/或减轻特征。 In some embodiments, a wireless communication network (10) also provide enhanced interference avoidance and / or mitigation features.
Claims(10)  translated from Chinese
1.一种无线通信网络,包括:多个移动节点,所述每个移动节点包括一个无线收发器和一个用于控制所述收发器的控制器,所述控制器还用于:调度相应的半永久时隙,以便与相邻移动节点建立通信链路,用于在它们之间传送数据,所述数据具有不同的优先级,确定每个通信链路的每个数据优先级的相应链路使用度量,和根据链路使用度量和数据优先级,调度按需分配的时隙,以便与相邻移动节点建立额外的通信链路,用于在它们之间传送数据。 A wireless communication network, comprising: a plurality of mobile nodes, each mobile node comprising a wireless transceiver and a controller for controlling said transceiver, said controller further for: scheduling the corresponding semi-permanent time slot to establish a communication link with neighboring mobile nodes for transmitting data therebetween, the data having different priorities, each communication link is determined for each data the priority of use of respective link metrics, and priority, scheduling demand assigned time slots based on link utilization metrics and data, in order to establish a communication link with a neighboring additional mobile nodes for transmitting data therebetween.
2.按照权利要求1所述的无线通信网络,其中所述控制器确定每个额外通信链路的每个数据优先级的链路使用度量,并根据每个额外通信链路的每个数据优先级的链路使用度量和数据优先级,重新分配按需分配的时隙。 2. The according to claim 1, wherein the wireless communication network, wherein said controller determines each of the data for each additional communication link priority link utilization metrics and data according to each priority for each additional communication link Use the link-level metrics and data priority, re-allocate time slots allocated on demand.
3.按照权利要求2所述的无线通信网络,其中所述控制器通过下述操作重新分配按需分配时隙:根据与之相关的相应链路使用度量,指定要被重新分配的按需分配时隙;根据失去的分配给它的相应指定的按需分配时隙,关于每个数据优先级估计移动节点之间降低容量的链路使用度量;和根据估计的降低容量的链路使用度量,重新分配指定的按需分配时隙。 According to Claim 2, wherein the wireless communication network, wherein said controller re-demand time slot allocation by the following operations: According to the corresponding link associated with the use of metrics, designated to be reallocated on demand slot; according lost assigned to its corresponding designated demand assigned time slot, on each data priority level estimates between the mobile node reduced capacity link utilization metrics; and based on the estimated reduced capacity link utilization metrics, reassign the DAMA slots.
4.按照权利要求2所述的无线通信网络,其中所述控制器通过下述操作重新分配按需分配时隙:根据与之相关的相应链路使用度量,指定要被重新分配的按需分配时隙;根据增加的分配给它的相应指定按需分配时隙,关于每个数据优先级估计移动节点之间增大容量的链路使用度量;和根据估计的增大容量的链路使用度量,重新分配指定的按需分配时隙。 According to claim 2, wherein the wireless communication network, wherein said controller re-demand time slot allocation by the following operations: based on the respective associated link utilization metrics, designated to be re-allocated according to need slot; assigned to it according to the increase of demand assignment of time slots corresponding designated, on the priority of each data between the mobile node estimates increase the link capacity using the metric; and based on the estimated increased capacity link utilization metrics redistribution designated slots on demand.
5.按照权利要求1所述的无线通信网络,其中根据与相应优先级对应的在至少一个在先半永久时隙内发送的数据的数量,确定每个链路使用度量。 5. according to claim 1, wherein the wireless communication network, wherein the number of data according to priority corresponding to the respective at least one prior semi-permanent time slot transmitted, determining each link utilization metric.
6.一种无线通信网络的通信方法,所述无线通信网络包括多个移动节点,所述方法包括:调度相应的半永久时隙,以便在相应的各对移动节点之间建立通信链路,用于在它们之间传送数据,其中数据具有不同的优先级;确定每个通信链路的每个数据优先级的相应链路使用度量;和根据链路使用度量和数据优先级,调度按需分配时隙,以便在各对移动节点之间建立额外的通信链路,用于在它们之间传送数据。 6. A communication method for a wireless communication network, the wireless communication network comprising a mobile node, the plurality method comprising: scheduling respective semi-permanent time slot to establish a communication link between each of the respective mobile node, for transferring data therebetween, wherein data having different priorities; communications link is determined for each data priority level for each respective link utilization metrics; and based on link utilization metrics and data priority, scheduling DAMA slots, in order to establish an additional communication link between each pair of mobile nodes for transmitting data therebetween.
7.按照权利要求6所述的方法,其中所述确定包括确定每个额外通信链路的每个数据优先级的链路使用度量;还包括根据每个额外通信链路的每个数据优先级的链路使用度量和数据优先级,重新分配按需分配的时隙。 7. A method according to claim 6, wherein said determining comprises determining for each additional communication link for each data priority link utilization metrics; according to each data further includes for each additional communication link priority The link uses metrics and data priority redistribution DAMA slots.
8.按照权利要求7所述的方法,其中重新分配还包括:根据与之相关的相应链路使用度量,指定要被重新分配的按需分配时隙;根据失去的分配给它的相应指定的按需分配时隙,关于每个数据优先级估计相应移动节点对之间降低容量的链路使用度量;和根据估计的降低容量的链路使用度量,重新分配指定的按需分配时隙。 8. A method according to claim 7 or claim, wherein the reallocation further comprising: based on the corresponding link associated with the use of metrics, specify demand assigned time slots to be reallocated; assigned to it based on the loss of the corresponding specified DAMA slots on each data priority measure is estimated to reduce the capacity of the link between the use of the corresponding mobile node; and based on the estimated reduced capacity link utilization metrics, reassign the DAMA slots.
9.按照权利要求7所述的方法,其中重新分配还包括:根据与之相关的相应链路使用度量,指定要被重新分配的按需分配时隙;根据增加的分配给它的相应指定按需分配时隙,关于每个数据优先级估计相应移动节点对之间增大容量的链路使用度量;和根据估计的增大容量的链路使用度量,重新分配指定的按需分配时隙。 9. A method according to claim 7, wherein the reallocation further comprising: based on the corresponding link associated with the use of metrics, specify demand assigned time slots to be reallocated; assigned to it according to the increase of the respective designated by should be allocated time slots for each data priority measure is estimated to increase the capacity of the link between the use of the corresponding mobile node; and according to the estimated increase link capacity usage metrics, reassign the DAMA slots.
10.按照权利要求6所述的方法,其中所述确定包括根据与相应优先级对应的在至少一个在先半永久时隙内发送的数据的数量,确定每个链路使用度量。 10. A method according to claim according to claim 6, wherein said determining comprises the number of data according to priority corresponding to the respective at least one prior semi-permanent time slot transmitted, determining each link utilization metric.
Description  translated from Chinese
具有增强时隙分配的无线通信系统 Slot allocation with enhanced wireless communication system

背景技术 BACKGROUND

时分多址访问(TDMA)是用于在无线移动通信系统之间建立通信链接的访问方案的一个例子。 Time division multiple access (TDMA) is an example of establishing a communications link between a wireless mobile communication system access scheme used. 在一系列的时间帧内建立无线移动通信系统之间的通信链接。 Establishing a communications link between a wireless mobile communication system in a series of time frames. 每个时间帧被分成时隙,每个无线移动通信系统被分配至少一个时隙。 Each time frame is divided into time slots, each of the wireless mobile communication system being assigned at least one time slot.

无线移动通信系统一般使用全向天线,以致一个移动通信系统发射的信息被所有其它移动通信系统接收。 Wireless mobile communication system is generally used omnidirectional antenna, so that a mobile communication system transmitted information is received by all other mobile communication systems. 当移动通信系统在固定频率下工作时,它们必须在它们各自的时隙内依次发射,以便避免信道干扰。 When the mobile communication system operates at a fixed frequency, they must in turn transmit within their respective time slots, in order to avoid channel interference.

为了改进两个无线通信系统之间的通信链路的质量,可以使用定向天线。 In order to improve the quality of the communication link between two wireless communication systems, a directional antenna may be used. 定向天线在覆盖范围有限的所需区域中提供增大的天线增益,同时降低朝向剩余区域的天线增益。 Directional antenna provides an increased antenna gain in a desired area of limited coverage while reducing the antenna gain towards the remaining area.

Pritchett的美国专利No.5767807公开用于在无线通信系统的网络内建立通信链接的相控阵天线。 U.S. Patent No.5767807 discloses Pritchett for establishing a communications link in a wireless communication system network phased array antenna. 相控阵天线包括有选择地控制天线图的寄生单元。 Phased array antenna comprising selectively controlling the antenna pattern of the parasitic elements. 当所有寄生单元处于高阻抗状态时,相控阵天线辐射全向信号,当响应开关电路使选定数目的寄生单元处于低阻抗状态时,相控阵天线辐射定向信号。 When all the parasitic elements in a high impedance state, phased array antenna radiation omnidirectional signal in response to the switching circuit so that when a selected number of parasitic elements in a low-impedance state, phased array antenna radiation directional signals.

更具体地说,Pritchertt′807专利公开固定启动无线通信系统从固定接收无线通信系统获得在网络中工作的无线通信系统的列表和每个无线通信系统的对应时隙列表。 More specifically, Pritchertt'807 patent discloses a wireless communication system to obtain a fixed starting time slot corresponding to a list of operating in a wireless communication network system and a list of each of the wireless communication system from a fixed receiving wireless communication system. 随后根据该列表创建表格,以便在无线通信系统之间调度时隙。 Then create a table based on the list, so that between the wireless communication system scheduling slots.

为借助定向天线工作的无线通信系统(尤其是当无线通信系统移动时)调度时隙是复杂的。 With directional antenna for wireless communication systems (especially when the wireless mobile communication system) scheduling slot is complicated. 在这样的动态网络中,移动通信系统不断进入和脱离网络。 In such a dynamic network, mobile communication systems continue into and out of the network. 此外,需要用于干扰检测和干扰回避的程序。 In addition, the need for interference detection and interference avoidance procedures.

借助优化的链路状态路由(OLSR),链路状态信息能够以各种服务质量(QoS)量度,包括带宽、延迟和数据丢失的可能性来量化链路的状态。 With Optimized Link State Routing (OLSR), link state information can be in a variety of quality of service (QoS) metrics, including the possibility of bandwidth, delay and data loss to quantify the status of the link. 每个路由器保持完整的布局信息,并定期通过淹没(flooding),向网络中的所有其它节点广播所述链路状态信息。 Each router maintain the integrity of the layout information, and regularly by drowning (flooding), the network broadcast to all other nodes of the link status information. 从而,需要向定向天线移动通信系统中的OLSR协议报告链路质量的过程。 Thus, the need to report to the directional antenna mobile communication system OLSR protocol link quality process.

发明内容 SUMMARY

本发明的一个目的是按照响应移动无线网络中的通信链路需求的变化的方式,调度时隙并减轻干扰的影响。 An object of the present invention is the mobile radio network in response to changes in the communication link needs way, scheduling time slots and mitigate the effects of interference.

根据本发明的该目的及其它目的,特征和优点由可包括多个移动节点的无线通信网络实现,所述移动节点均包括一个无线收发器和用于控制所述收发器的一个控制器。 According to this object of the present invention and the other objects, features and advantages of the wireless communication network may include a plurality of mobile nodes implementation, the mobile node comprises a wireless transceiver and a controller for controlling the transceiver. 控制器还可用于调度相应的半永久时隙,以便与相邻移动节点建立通信链路,用于在它们之间传送数据,所述数据具有不同的优先级(priority level)。 The controller may also be for scheduling a respective semi-permanent time slot to establish a communication link with a neighboring mobile nodes, for transferring data therebetween, the data having different priority (priority level). 控制器还可确定每个通信链路的每个数据优先级的相应链路使用度量(metrics),并根据链路使用度量和数据优先级,调度按需分配的时隙(demand assignedtime slot),用于与相邻移动节点建立额外的通信链路,用于在它们之间传送数据。 The controller can also be determined for each data in each communication link corresponding priority link usage metrics (metrics), and according to the link using metrics and data priority, scheduling time slots allocated on demand (demand assignedtime slot), for establishing additional communication links with the neighboring mobile nodes for transmitting data therebetween.

更具体地说,控制器可确定每个额外通信链路的每个数据优先级的链路使用度量,并根据每个额外通信链路的每个数据优先级的链路使用度量和数据优先级,重新分配按需分配的时隙。 More specifically, the controller may determine whether each data for each additional communication link priority link utilization metrics and data according to each link for each additional communication link uses priority metrics and data priority redistribution DAMA slots. 控制器可通过根据与之相关的相应链路使用度量,指定要被重新分配的按需分配时隙,并根据失去的分配给它的相应按需分配时隙,关于每个数据优先级估计移动节点之间降低容量的(reduced capacity)链路使用度量,重新分配按需分配的时隙。 The controller can be based on the respective associated link utilization metrics, specify demand assigned time slots to be reallocated, and according to the lost-demand assigned to its respective slot, the priority data is estimated with respect to each mobile reduce capacity between nodes (reduced capacity) link utilization metrics, re-allocate time slots allocated on demand. 随后可根据估计的降低容量的链路使用度量,重新分配指定的按需分配时隙。 Link utilization metrics may then be reduced based on the estimated capacity, reassign the demand assignment slot.

另外,控制器还可根据增加的分配给它的相应指定的按需分配时隙,关于每个数据优先级估计移动节点之间增大容量的链路使用度量。 In addition, the controller may be assigned to it according to the increase of the respective designated demand assigned time slot, on each data priority level estimates between the mobile node increases the capacity link utilization metrics. 这样,控制器可根据估计的增大容量的(increased capacity)链路使用度量,重新分配指定的按需分配时隙。 Thus, the controller can be increased based on the estimated capacity (increased capacity) link utilization metrics, reassign the DAMA slots.

可根据与相应优先级对应的在至少一个在先半永久时隙内发送的数据的数量,确定每个链路使用度量。 According to the number of data corresponding to the appropriate priority in at least one semi-permanent time slot earlier sent to determine the use of metrics for each link. 另外,控制器还可包括在传送之前保存数据的队列,可根据队列中与相应优先级对应的数据的数量,确定每个链路使用度量。 In addition, the controller may also include data stored in a queue before transmission, according to the number of priority queues corresponding to respective data, determining each link utilization metric. 另外,控制器可根据每个通信链路的链路使用度量的加权和,确定每个通信链路的集合(aggregate)链路度量。 In addition, the controller may set each communication link (aggregate) link metric in accordance with the link metric for each communication link using a weighted sum, is determined.

有利的是,通信链路和额外的通信链路可以是定向通信链路,每个节点还可包括与收发器连接,并由控制器控制以便建立定向通信链路的相控阵天线。 Advantageously, the communication link and additional communication links may be directional communication links, each node may also include a connection to the transceiver, controlled by the controller in order to establish the directional communication link phased array antenna. 例如,无线通信网络可以是移动自组织网络(MANET)。 For example, a wireless communication network can be a mobile ad hoc network (MANET).

本发明的另一方面涉及一种无线通信网络,它包括多个移动节点,每个移动节点包括一个无线收发器和一个控制所述无线收发器的控制器。 Another aspect of the present invention relates to a wireless communications network, comprising a plurality of mobile nodes, each mobile node comprising a wireless transceiver and a controller controlling the wireless transceiver. 控制器还可通过确定某一相邻移动节点和在时隙内发射的其它潜在干扰移动节点的相对位置,并根据所述相对位置确定对定向通信链路的潜在干扰,调度到所述相邻移动节点的定向通信链路。 The controller may also determine whether a neighboring mobile node by the relative positions of other potentially interfering mobile node and transmitted in a slot, and determining potential interference to the directional communication link according to said relative position, scheduled to the adjacent directional communication link between the mobile node. 此外,如果所述潜在干扰低于某一阈值,那么在所述时隙内,所述定向通信链路可被调度。 Further, if the potential interference is below a certain threshold, then in the time slot, the directional communication link may be scheduled.

更具体地,控制器可根据到所述相邻移动节点和每个潜在干扰移动节点的相应距离和角度,确定所述相对位置。 More specifically, the controller according to the mobile node and each potentially interfering mobile node of the adjacent respective distance and angle, determining the relative position. 还可根据地平面衰减算法和自由空间衰减算法至少之一确定所述相对位置。 Also base plane attenuation algorithm and at least one free space attenuation algorithm to determine the relative positions.

另外,控制器可根据估计的信号路径损耗,确定所述潜在干扰。 Further, the controller according to the signal path loss estimate, determining the interference potential. 还可根据来自潜在干扰节点发射的信号的旁瓣(side lobe)和旁瓣抑制算法,确定所述潜在干扰。 Also based on signals from potentially interfering nodes transmit sidelobe (side lobe) and side lobe suppression algorithm to determine the potential interference.

在确定所述相对位置之前,控制器可接收来自所述相邻移动节点的建立定向通信链路的请求。 Prior to determining the relative position, the controller may receive a request from the mobile node to establish the directional communication link of said adjacent. 在调度定向通信链路之后,控制器还可间歇地确定所述相邻移动节点和其它潜在干扰节点的相对位置,根据间歇确定的相对位置,确定对定向通信链路的潜在干扰,如果潜在干扰高于所述阈值,那么在新时隙内调度定向通信链路。 After scheduling the directional communication link, the controller may also intermittently determine the relative location of the mobile node and other potentially interfering nodes in the neighbor, based on the relative position of the intermittent determination, determine potential interference to the directional communication link, if the potential interference above the threshold, then the directional communication link scheduling in a new time slot.

另外,控制器可具有保存信号-干扰比值的查寻表,控制器还可根据保存的信号-干扰比值确定干扰。 In addition, the controller may have saved the signal - interference ratio lookup table, the controller can also be stored in accordance with the signal - interference ratio determined interference. 另外,移动节点还可包括测量其各自位置的位置确定装置,控制器可根据测得的相邻移动节点和潜在干扰移动节点的位置,确定所述相对位置。 In addition, the mobile node may further comprise measuring the position of their respective position determining means, the position of the controller can be measured based on the mobile node and the potentially interfering mobile nodes adjacent to, determining the relative position.

根据本发明的另一方面,无线电通信网络可包括在分配的时隙期间,在它们之间建立通信链路,并根据使用要求重新分配所分配的时隙的多个移动节点。 According to another aspect of the invention, a radio communication network may be included in time slot assignment, the communications link is established between them, and a plurality of mobile nodes according to the requirement to redistribute the allocated time slot. 特别地,发起(initiating)移动节点可识别与相邻移动节点共用的可用时隙,根据所述时隙内,发起移动节点和相邻移动节点的链路使用率,对可用时隙排序(rank),并把包括可用时隙及其排序的时隙重新分配请求传送给接收移动节点。 In particular, to initiate (initiating) the mobile node and the neighboring mobile node may identify available time slots shared, based on said slot, initiating mobile node and neighboring mobile node link utilization, sort (rank the available time slots ), and to include the sort of available time slots and slot reallocation request to receive the mobile node.

接收移动节点可依次接收时隙重新分配请求,根据可用时隙内,接收移动节点和邻近接收移动节点的移动节点的链路使用率,对可用时隙排序,并根据发起移动节点和接收移动节点生成的排序,产生可用时隙的组合排序。 Receiving mobile node may in turn receive the time slot re-allocation request, based on the available time slots, the receiving mobile node and mobile nodes neighboring the receiving mobile node link utilization, ordering the available time slots, and in accordance with initiating mobile node and the receiving mobile node generating sort, produce a combined sorting available timeslots. 接收移动节点随后可根据组合排序,重新分配至少一个可用时隙,用于在发起移动节点和接收移动节点之间建立通信链路。 Receiving mobile node may then sorted according to the combination, re-allocating at least one available time slot for establishing a communication link between the initiating mobile node and the receiving mobile node.

更具体地说,发起移动节点和接收移动节点都可确定与每个可用时隙相关的通信需求度量,并把与之相关的通信需求度量高于较高使用阈值的任何可用时隙排除在重新分配的考虑之外。 More specifically, the initiating mobile node and the receiving mobile node can determine the time slot associated with each of the available communication demand metric and the metric associated with the communication needs using any of the available slots is higher than the higher threshold excluded re consider allocating outside. 此外,发起移动节点和接收移动节点都还可确定与每个可用时隙相关的信号-干扰比(SIR)值,并把与之相关的SIR值高于较高SIR阈值的任何可用时隙排除在重新分配的考虑之外。 In addition, initiating mobile node and the receiving mobile node may also determine available timeslots associated with each signal - to-interference ratio (SIR) value, and the SIR associated with a higher value than any available slot SIR threshold Exclude In addition to considering redistribution. 另外,发起移动节点和接收移动节点都可确定与每个可用时隙相关的SIR值,并据此对时隙排序。 Further, initiating mobile node and the receiving mobile node can determine the SIR value associated with each available time slot, and accordingly sorted slot.

有利的是,发起移动节点和接收移动节点可传送具有多种优先级的数据。 Advantageously, the initiating mobile node and the receiving mobile node may transmit data with a variety of priorities. 从而,如果将在可用时隙内传送的数据的优先级等于或高于当前正在所述可用时隙内传送的数据的优先级,那么接收移动节点可只重新分配所述可用时隙。 Thus, if the priority is equal to or higher than the transfer of data within the available time slots of the currently available data time slot transmission priority, then the receiving mobile node may only re-allocation of the available time slots. 发起移动节点和接收移动节点都可估计每个可用时隙的降低容量的链路使用度量,并据此对可用时隙排序。 Initiating mobile node and the receiving mobile node can estimate the metric for each link using the reduced capacity of available time slots and available time slots sorted accordingly.

此外,发起移动节点和接收移动节点都可根据在可用时隙内传送的数据的平均数量,对可用时隙排序。 In addition, initiating mobile node and the receiving mobile node based on the average amount of data can be transmitted within the available time slots, ordering the available time slots. 同样,移动节点可包括保存将通过通信链路传送的数据的队列,发起移动节点和接收移动节点都可根据保存在队列中的与相应通信链路相关的数据的数量,对每个可用时隙排序。 Similarly, the mobile node may include the transmission queue stored data via a communication link, the initiating mobile node and the receiving mobile node can be based on the number associated with the respective communication link data stored in the queue, for each available time slot sorting.

在无线通信网络中提供本发明的另一有利方法,在所述无线通信网络中,一对移动节点建立通信链路,用于在至少一个时隙内在它们之间传送分组。 The present invention is provided in a wireless communication network to another advantageous method, in the wireless communication network, a mobile node to establish a communication link between the at least one time slot for their inherent transport packets. 所述一对移动节点中的至少一个移动节点可确定所述至少一个时隙内,与通信链路相关的链路质量值,和所述通信链路是具有第一质量还是第二质量。 Said pair of mobile nodes at least one mobile node may determine the at least one time slot, the communication link associated with the link quality value, and the communications link having a first quality or second quality. 如果在所述至少一个时隙内,通信链路具有第一质量,那么该对移动节点可合作在第一时段内把该通信链路分配给新的时隙。 If at least one time slot within the communication link having a first mass, then the mobile node may cooperate in the first period to the communication link assigned to the new time slot. 如果在所述至少一个时隙内,通信链路具有第二质量,那么它们还可合作在短于第一时段的第二时段内,把该通信链路分配给新的时隙。 If the time slot, the communication link having at least one of said second mass, then they can also cooperate in the second period is shorter than the first period, the allocation of the communication link to a new time slot.

从而,如果链路质量严重恶化,那么时隙可被相当快速地分配,以使对通信的破坏降至最小。 Thus, if the link quality seriously deteriorated, then the slot may be allocated relatively quickly, so that the destruction of the communication is minimized. 另一方面,如果链路质量处于临界状态但是仍然可用,那么可更长地延迟时隙的重新分配,这可降低试图同时重新分配时隙的各对竞争节点之间的冲突。 On the other hand, if link quality is marginal but still usable, it may be longer to delay the re-allocation of time slots, which can simultaneously reduce the attempt to redistribute the conflict among contending nodes for slots.

更具体地,所述一对移动节点中的所述至少一个移动节点可根据SIR和分组接收差错值(PREV)至少之一,确定链路质量值。 More specifically, the mobile node in the pair of the at least one mobile node according to at least one of SIR and packet reception error value (PREV), determining a link quality value. 此外,如果SIR介于第一SIR阈值和高于第一SIR阈值的第二SIR阈值之间,并且PREV介于第一差错阈值和高于第一差错阈值的第二差错阈值之间,那么确定通信链路具有第一质量。 In addition, if the SIR is between a first SIR threshold and a second SIR threshold is between the first threshold value is higher than the SIR, and the PREV is between a first error threshold higher than the first error threshold and a second error threshold value, it is determined a communication link having a first mass. 此外,如果SIR小于第一SIR阈值,并且PREV大于第二差错阈值,那么通信链路也被确定为具有第一质量。 In addition, if the SIR is less than the first SIR threshold and the PREV is greater than the second error threshold value, then the communication link has been identified as having a first mass.

另外,如果SIR小于第一SIR阈值,并且PREV介于第一差错阈值和第二差错阈值之间,那么确定通信链路具有第二质量。 Further, if the SIR is less than between a first SIR threshold and the PREV is between a first error threshold and the second error threshold value, it is determined that the communication link having a second mass. 此外,如果PREV低于第一差错阈值,那么通信链路也被确定为具有第二质量。 In addition, if the PREV is below the first error threshold value, then the communication link has been identified as having a second mass.

所述至少一个时隙可以是至少一个按需分配时隙,所述一对移动节点还可在半永久时隙内建立额外的通信链路。 The at least one time slot may be at least one on-demand time slot, the pair of mobile nodes may also establish an additional communication link within a semi-permanent time slot. 这样,该对移动节点中的至少一个移动节点中可确定半永久时隙内,与所述额外的通信链路相关的半永久链路质量值,并根据半永久链路质量值,确定在半永久时隙内,所述额外通信链路是具有第一质量还是具有第二质量。 Thus, the mobile node at least one mobile node may determine the semi-permanent time slot, with the additional communication link is related to semi-permanent link quality value, and according to a semi-permanent link quality value, determined in the semi-permanent time slot within , the additional communication link having a first quality and having a second mass. 从而,如果在半永久时隙内,额外通信链路具有第一质量或第二质量,那么该对移动节点可合作在第二时段内把额外的通信链路分配给新的半永久时隙。 Thus, if in a semi-permanent time slot, an additional communication link having a first or second mass by mass, then the mobile node may cooperate within the second period to allocate additional communication link to a new semi-permanent time slot.

如上所述,所述多个移动还可包括与收发器连接的全向天线,该对移动节点可使用在它们之间建立全向通信链路。 As described above, the mobile may also include a plurality of transceivers connected to the omnidirectional antenna, the mobile node may use to establish between them an omnidirectional communication link. 这样,如果通信链路具有第二质量,并且与全向通信链路相关的全向质量值持续预定时间低于全向链路质量阈值,那么该对移动节点还可立即解除分配所述至少一个时隙。 Thus, if the communication link having a second mass, and an omni-directional communication link with the associated full duration time is below a predetermined value to the whole mass to the link quality threshold, then the mobile node may immediately deallocating the at least one slots.

本发明的通信方法用于一种无线通信网络,例如上面简要说明的无线通信网络,其中调度相应的半永久时隙,以便在相应的各对移动节点之间建立通信链路,用于在它们之间传送数据,其中数据具有不同的优先级。 Communication method of the present invention is used in a wireless communication network, such as a wireless communication network described briefly above, wherein the scheduling respective semi-permanent time slot to establish a communication link between each of the respective mobile node, for which the transferring data between, wherein the data having different priorities. 该方法还包括确定每个通信链路的每个数据优先级的相应链路使用度量,并根据链路使用度量和数据优先级,调度按需分配时隙,以便在各对移动节点之间建立额外的通信链路,用于在它们之间传送数据。 The method further comprises determining for each data priority level for each communication link of the respective link utilization metrics and link utilization metrics and data according to priority, scheduling time slots according to need, so that between each pair of the mobile nodes establish additional communication link, for transmitting data between them.

本发明的另一种通信方法可包括在某一时隙内,调度从第一移动节点到第二相邻移动节点的定向通信链路。 Another communication method of the present invention may include within a certain time slot, scheduling a first mobile node from the mobile node to the second neighboring directional communications link. 更具体地,这可通过确定第二相邻移动节点和在该时隙内进行发射的其它潜在干扰移动节点的相对位置,并根据所述相对位置确定对定向通信链路的潜在干扰来实现。 More specifically, this can be determined by the relative positions of the second neighboring mobile node and other potentially interfering mobile nodes transmitting in the time slot, and determining potential interference to the directional communication link is achieved in accordance with the relative position. 这样,如果干扰低于某一阈值,那么在该时隙内,定向通信链路可被调度。 Thus, if the interference is below a certain threshold, then the time slot, the directional communication link may be scheduled.

本发明的另一通信方法可包括识别在发起移动节点和相邻移动节点之间共用的可用时隙,并根据该时隙内,发起移动节点和相邻移动节点的链路使用率,对所述可用时隙排序。 Another communication method of the present invention may include identifying the initiating mobile node and neighboring mobile nodes sharing between the available timeslots, and in accordance with the time slot, initiating mobile node and mobile nodes neighboring link utilization, for the sort described available time slots. 还可从发起移动节点向接收移动节点传送包括可用时隙及其排序的时隙重新分配请求,时隙重新分配请求可在接收移动节点被接收。 May also be transmitted from the initiating mobile node to the receiving mobile node including the available time slots and sorting slot reallocation request, the time slot re-allocation request may be received in the receiving mobile node. 可根据所述可用时隙内,接收移动节点和与接收移动节点相邻的移动节点的链路使用率,对可用时隙排序,根据发起移动节点和接收移动节点产生的可用时隙的排序,可产生可用时隙的组合排序。 According to said available slot, receiving link utilization and the receiving mobile node and mobile nodes neighboring mobile nodes, sort the available time slots, produced according to the initiating mobile node and the receiving mobile node of available timeslots is sorted, can produce a combined sorting available timeslots. 此外,根据组合排序,可重新分配至少一个可用时隙,用于在发起移动节点和接收移动节点之间建立通信链路。 Further, according to the combination sort, can be re-allocating at least one available time slot for establishing a communication link between the initiating mobile node and the receiving mobile node.

本发明的另一通信方法可包括在一对移动节点之间建立通信链路,用于在至少一个时隙内,在它们之间传送分组,确定在所述至少一个时隙内,与该通信链路相关的链路质量值,和确定该通信链路是具有第一质量还是第二质量。 Another communication method of the present invention may include the establishment of a mobile node between a pair of communication links, for at least one time slot, the packet transfer between them, determine the time slot, and the at least one communication the relevant link quality value of the link, and determining the communication link having a first or second mass by mass. 该方法还可包括如果在所述至少一个时隙内,通信链路具有第一质量,那么在第一时段内把通信链路分配给新的时隙。 The method may further comprise, if within said at least one time slot, the communication link having a first mass, the first time period of the communication links assigned to the new time slot. 另外,如果在所述至少一个时隙内,通信链路具有第二质量,那么可在比第一时段短的第二时段内,把通信链路重新分配给新的时隙。 Further, if at least one time slot within the communication link having a second mass, can be shorter than the first period in the second period, the communications link is re-allocated to the new time slot.

附图说明 Brief Description

图1图解说明根据本发明的无线移动自组织网络。 Figure 1 illustrates a wireless mobile ad hoc networks based on the present invention.

图2是图解说明根据本发明的无线移动节点的更详细方框图。 Figure 2 is a diagram illustrating a wireless mobile node in accordance with the present invention, a more detailed block diagram.

图3图解说明根据本发明的一帧时隙。 3 illustrates a frame according to the present invention the slot.

图4图解说明根据本发明的相对于图2中图解说明的网络图的可用时隙的调度。 Figure 4 illustrates the present invention in accordance with the scheduling for the network illustrated in Figure 2 diagram illustrates the available timeslots.

图5是根据本发明的半永久时隙和可用时隙的调度的顶级状态图。 Figure 5 is a semi-permanent time slots and available time slots of the present invention the top of the scheduling state diagram.

图6图解说明根据本发明的半永久时隙调度过程。 Figure 6 illustrates a semi-permanent time slot scheduling process in accordance with the present invention.

图7图解说明根据本发明的为新的通信链路调度的半永久时隙。 Figure 7 illustrates a new communication link according to the present invention, the scheduling of semi-permanent time slot.

图8图解说明根据本发明的可用时隙调度过程。 Figure 8 illustrates the scheduling of available time slots in accordance with the process of the present invention.

图9图解说明根据本发明的正被加入通信链路中的可用时隙。 9 illustrates a communication link to be added to an available slot in accordance with the present invention being.

图10和11图解说明根据本发明的,正在根据来自相控阵天线的多个同时天线波束,为新的通信链路调度的半永久时隙。 Figures 10 and 11 illustrate according to, from being based on the plurality of phased array antenna, while the antenna beam, for the new communication link scheduling of semi-permanent time slot of the present invention.

图12和13是图解说明根据本发明的用于根据全向链路质量值,在移动节点之间建立定向通信链路的方法的流程图。 12 and FIG. 13 is a flowchart illustrating an omni-directional link quality value in accordance with established methods directional communication link between a mobile node used in the present invention.

图14-16是图解说明根据本发明的用于根据链路利用率,分配按需分配时隙的方法的流程图。 14-16 is a flow diagram illustrating a method according to the present invention is used in accordance with the link utilization, the slot allocation according to need.

图17和18是图解说明根据本发明的数据优先化方法的流程图。 17 and FIG. 18 is a flowchart illustrating the present invention in accordance with data prioritization method.

图19和20是图解说明根据本发明的确定分组接收误差值,并根据分组接收误差值调整链路使用的方法的流程图。 Figures 19 and 20 are diagrams illustrating the packet reception error value is determined in accordance with the present invention, according to the packet reception error value and adjusting link uses a flowchart of a method.

图21是图解说明本发明的无线通信网络中的两对移动节点的干扰回避情形的示意方框图。 Figure 21 is a diagram illustrating a wireless communication network of the present invention two pairs of interference avoidance schematic block diagram of the mobile node circumstances.

图22和23是图解说明根据本发明的干扰回避方法的流程图。 Figures 22 and 23 is a flowchart illustrating a method of avoiding the interference of the present invention.

图24和25是图解说明根据本发明的重新分配时隙的方法的流程图。 24 and FIG. 25 is a flowchart illustrating re-allocation of time slots in accordance with a method of the present invention.

图26-28是图解说明根据本发明的基于链路质量,在不同的时段内重新分配时隙的方法的流程图。 Figure 26-28 is a flowchart illustrating the link quality according to the present invention is based, in a different temporal slot reallocation method.

图29是图解说明根据本发明的估计链路质量并向路由协议报告链路质量的方法的流程图。 FIG 29 is a flowchart illustrating reporting link quality to a routing protocol based on the estimated link quality the method of the present invention.

具体实施方式 DETAILED DESCRIPTION

现在参考表示本发明的优选实施例的附图,更充分地说明本发明。 Referring now to the accompanying drawings showing a preferred embodiment of the present invention embodiment, to more fully illustrate the present invention. 但是,可用许多不同的形式具体体现本发明,本发明不应被理解为局限于这里陈述的实施例。 However, the available embodied in many different forms of the present invention, the present invention should not be construed as limited to the embodiments set forth herein. 相反,提供这些实施例以致本公开将是彻底和完整的,并且将向本领域的技术人员充分传送本发明的范围。 Instead, these embodiments are provided so that this disclosure will be thorough and complete, and will fully skilled in the art transfer scope of the invention. 相同的附图标记表示相同的部件。 The same reference numerals denote the same parts.

首先参考图1-2,无线移动通信网络10包含多个无线移动节点12a-12h。 Referring first to FIGS. 1-2, a wireless mobile communication network 10 comprises a plurality of wireless mobile nodes 12a-12h. 每个移动节点12a-12h包含收发器14,与收发器连接的定向天线16,和与收发器连接的控制器18。 Each mobile node 12a-12h comprises a transceiver 14, a controller connected to the directional antenna and the transceiver 16, and 18 connected with the transceiver.

控制器18包括为每个时间帧调度相应的半永久时隙,以便与每个相邻移动节点建立通信链路,同时在每个时间帧中保留至少一个可用时隙的半永久时隙单元18a。 The controller 18 includes a frame scheduling respective semi-permanent time slot for each time, in order to establish a communication link with each neighboring mobile node, while retaining at least one available time slot of the semi-permanent time slot unit 18a in each time frame. 可用时隙单元18b根据链路通信需求,调度所述至少一个可用时隙,以便也服务于与相邻移动节点的通信链路。 Available time slot unit 18b according to the at least one available time slot link communications demand, scheduling, and to also serve the communication link neighboring mobile node. 另外,控制器18包括在与每个相邻移动节点的通信期间,使定向天线对准(aim)每个相邻移动节点的天线对准单元18c。 In addition, the controller 18 comprises a communication with each neighboring mobile node during the make directional antenna alignment (aim) of each neighboring mobile node antenna alignment unit 18c.

并行操作能够减少时隙分配延迟。 Parallel operations can reduce time slot allocation delay. 因此,半永久时隙单元18a可关于各个时间帧,启动一个或多个半永久时隙请求,以便建立与每个相邻移动节点的通信链路,并在每个时间帧中保留至少一个可用时隙,同时处理从相邻移动节点接收的多个半永久时隙请求。 Thus, semi-permanent time slot unit 18a may be on each time frame, start one or more semi-permanent time slot requests to establish a communication link with each neighboring mobile node, and to leave at least one available time slot in each time frame while processing a plurality of semi-permanent time slots from neighboring mobile node receives the request. 可用时隙单元18b可根据链路通信需求,启动一个或多个可用时隙请求,以便也服务于与相邻移动节点的通信链路,同时处理从相邻移动节点接收的多个可用时隙请求。 Available time slot unit 18b according to a communication link for launching one or more available time-slot request to also serve the mobile node and the adjacent communication link, simultaneously processing a plurality of available time slots received from a neighboring mobile node request.

换句话说,在处理多个接收的请求的时候,一个节点能够具有它启动的一个或多个待定需求可用请求和半永久请求。 In other words, when processing a plurality of requests received, and a node can have one or more pending demand available it starts requests and semi-permanent requests. 有时,这会导致向一个以上的邻居临时分配指定的时隙。 Sometimes this leads to more than one neighbor temporarily assigned designated timeslots. 但是,如下更详细所述,借助指示一个相邻节点对该时隙的选择的确认消息,可消除这种冲突。 However, as described in more detail below, by means of a neighboring node indicating the selection of the acknowledgment message time slot, to eliminate such conflicts.

可利用两种不同的方法提供可靠的确认消息。 Reliable confirmation messages may be provided using two different methods. 发起移动节点向接收移动节点传送时隙请求,接收移动节点向发起移动节点传送答复。 Initiating mobile node to the receiving mobile node transmits the time slot request, the receiving mobile node transmits the reply to the initiating mobile node. 发起移动节点向接收移动节点传送确认,如果没有收到所述确认,那么接收移动节点再次传送答复。 Initiating mobile node to the receiving mobile node delivery confirmation, if the acknowledgment is not received, the receiving mobile node transmits the reply again. 另一方面,接收移动节点可向发起移动节点传送肯定应答,如果没有收到所述肯定应答,发起移动节点再次传送所述确认。 On the other hand, to the receiving mobile node may transmit an acknowledgment initiating mobile node, if not received the acknowledgment, the initiating mobile node transmits again confirmed.

如果两个节点同时相互启动时隙请求,那么应可靠地处理时隙请求冲突。 If two nodes simultaneously start mutual slot request, the request should be reliably handle slot conflict. 当同时向另一移动节点发送请求和接收来自所述另一移动节点的请求,而没有收到对应的答复时,控制器18等待一段时间以便重新发送另一请求。 When simultaneously sending a request to another mobile node and receiving a request from the other mobile node without receiving a corresponding reply, the controller 18 waits for a period of time in order to resend another request. 在该段时间内,可处理到来的时隙请求。 In that period of time, you can handle the arrival of slot requests. 在该段时间结束之后,如果没有从所述另一节点收到任何请求,或者如果未对所述另一节点进行任何时隙分配,那么可向该节点发送新的请求。 After the end of that period, if not received any request from the other node, or if you do not make any other node on the slot allocation, then you can send a new request to the node. 如果延迟的请求到达队列的前端,那么控制器18检查是否已对该节点进行了时隙分配。 If the delayed request reaches the front of the queue, the controller 18 checks whether the node of the slot assignment. 如果是,那么延迟的请求被丢弃。 If so, the delayed request is discarded. 另外,如果请求预定给在该请求到达队列的前端之前不再是邻居的节点,那么延迟的请求被丢弃。 Further, if a predetermined request to the request before it reaches the front of the queue is no longer a neighbor node, then the delayed request is discarded.

控制器18包括检测用于与相邻移动节点通信的时隙中的干扰的干扰检测单元18d。 The controller 18 includes a detector for the mobile node and the adjacent communication slot in the interference of the interference detection unit 18d. 控制器18根据检测的干扰,协调时隙的调度。 The controller 18 according to the detected interference, coordinated scheduling time slot. 干扰检测单元18d可测量信号干扰比和/或分组差错率。 Interference detection unit 18d measures the signal to interference ratio and / or a packet error rate. 分组差错率可基于循环冗余检验(CRC)失效。 Packet error rate may fail based on cyclic redundancy check (CRC). 另外,干扰检测单元18d可比较检测的干扰与阈值。 In addition, the interference detection unit 18d may compare detected interference to a threshold. 最好,控制器根据检测的干扰,持续指定的时隙交换节点之间的通信顺序,并在交换通信的顺序之后,根据检测的干扰,协调新时隙的调度。 Preferably, the controller detects the interference, the communication sequence between the switching node for a specified time slot, and after switching the order of communication, based on the detected interference, to coordinate scheduling of new time slot.

通信(traffic)协调单元18e通过根据链路通信需求,把时隙分配给时隙单元,协调与每个相邻移动节点的通信。 Communications (traffic) coordination unit 18e through the communication link according to the needs, to allocate time slots to the time slot unit, to coordinate communication with each neighboring mobile node. 控制器18根据分配的时隙,协调时隙的调度。 The controller 18 according to the assigned time slot, the time slot scheduling coordination. 通信协调单元18e可根据增大的链路通信需求,分配一组大量的时隙,和/或根据增大的链路通信需求,向相邻的移动节点请求一组大量的时隙。 Communication coordination unit 18e may increase according to the link communications demand, assign a set of a large number of time slots, and / or based on an increased link communications demand, to the neighboring mobile node requests a set of a large number of time slots. 另外,通信协调单元18e可增大时隙的最大数目,重新分配时隙,和/或根据增大的链路通信需求,分配半时隙。 Further, the communication unit 18e may increase the maximum coordination number of slots, reallocate time slots, and / or based on an increased link communications demand, assign half slot. 这种增大的链路通信需求可包括流式数据和/或高速率传感器数据。 Such increased link communications demand may include streaming data and / or high rate sensor data.

无线移动节点12a-12h在移动环境中工作。 Wireless mobile node 12a-12h work in a mobile environment. 这些系统可以是地面系统和/或空中系统,从而它们不断进入和脱离网络10。 These systems may be ground system and / or air system, so that they continue into and out of the network 10. 定向天线16可以是相控阵天线,抛物面天线或喇叭天线。 Directional antenna 16 may be a phased array antenna, a parabola antenna or a horn antenna. 通过定向天线16的发射使RF信号能够沿所需方向集中。 By transmitting directional antenna 16 enables the RF signal can be concentrated in the desired direction.

通过有选择地控制一对无线移动通信系统之间的天线图的方向以便在其间建立通信链路,可在相同的预定半永久时隙内在其它无线通信系统之间建立另外的通信链路。 By selectively controlling the direction of the antenna between a pair of wireless mobile communication system diagram for establishing a communication link therebetween, additional communication link can be established between the same semi-permanent time slot on a predetermined other wireless communication systems. 如图1中所示,这由移动节点12c与12e之间在时隙1中工作的通信链路27,和移动节点12a与12b之间同样在时隙1中工作的通信链路29所示。 As shown in Figure 1, which between the mobile nodes 12c and 12e in communication slots by working in a link 27, and also between mobile nodes 12a and 12b in communication slots of a working link 29 shown . 本发明的该特征有利地允许更好地利用无线移动通信网络10的资源。 The features of the present invention advantageously allows better use of the wireless mobile communication network 10 resources.

控制器18根据每个时间帧内的时隙的总数,限制该时间帧内每个无线移动节点12a-12h的通信链路的数目。 The controller 18 according to the total number of time slots within each time frame, the time limit for each wireless mobile node frame number of the communication links 12a-12h of. 把通信链路的数目局限为时间帧内时隙的总数的一小部分的优点显著简化了与邻近节点的时隙的调度。 Advantage of a small portion of the total number of the limitations of the communication link for the time slot of the frame significantly simplifies the scheduling of time slots adjacent node.

每个时间帧内每个无线移动节点12a-12h的通信链路的数目小于或等于N,每个时间帧内的时隙的总数等于或大于2N-1。 The number of communication links 12a-12h within each time frame of each wireless mobile node is less than or equal to N, each of the total number of time slots within the time frame is equal to or greater than 2N-1. 除了简化时隙的调度之外,这种分布式调度避免了冲突。 In addition to simplifying the scheduling of time slots outside the distributed scheduling avoids conflicts.

分布式调度允许任意两对无线移动节点,例如12a和12b调度半永久时隙,而不必与任何其它无线移动节点通信。 Distributed scheduling allows any two pairs of wireless mobile nodes, such as 12a and 12b scheduling semi-permanent time slot without having to communicate with any other wireless mobile node. 换句话说,不存在关于所有无线移动节点12a-12h的调度半永久时隙的集中式主/从类型的协调。 In other words, there is no centralized master scheduling of all the wireless mobile nodes 12a-12h semi-permanent time slot of the / from the type of coordination. 由于按照分布方式调度无线移动节点12a-12h之间的时隙,因此在无线移动通信网络10中不存在单一的故障点。 Scheduling time slots according to a distributed manner due to the wireless mobile nodes 12a-12h between, there is no single point of failure in the wireless mobile communication network 10.

控制器18可区分通信链路的优先顺序,并根据优先化中断通信链路之一,使半永久时隙可用于建立与新的邻近移动节点的通信链路。 The controller 18 can distinguish the communication link priorities, and the priority of the interrupt according to one of the communication links, so that semi-permanent time slots may be used to establish a new neighboring mobile node to the communication link. 下面将更详细地谈到通信链路的优先化。 Turning in more detail below prioritized communications link. 另外,控制器18还可区分通信链路的优先顺序,并根据该优先化调度至少一个可用时隙。 In addition, the controller 18 may also prioritize the communication link to distinguish, and at least one available time slot based on the priority scheduling.

如果通信链路的数目小于N,那么控制器18还可把半永久时隙之一预定为可用时隙。 If the number of communication links is less than N, then the controller 18 may also the one semi-permanent time slot predetermined for the available timeslots. 对于现有通信链路,这有利地根据需要支持通信链路需求。 For existing communication links, which advantageously supports communication link demands as required. 但是,如果通信链路的数目再次等于N,那么控制器18可把按需分配的时隙重新调度给半永久时隙,如下更详细所述。 However, if the number N is equal to the communication link again, the controller 18 may reschedule the demand assigned time slots to the semi-permanent time slot, as described in more detail below.

每个通信链路由发起移动节点,例如节点12a,和接收移动节点,例如节点12b形成,发起移动节点把可用的半永久时隙的列表传送给接收移动节点。 Each communication link initiated by the mobile node, such as node 12a, and the receiving mobile node, such as node 12b are formed, the initiating mobile node transmits a list of available semi-permanent time slots to the receiving mobile node. 接收移动节点12b随后把半永久时隙之一的选择传送给发起移动节点。 Receiving mobile node 12b then transferred to select one of the semi-permanent time slot to the originating mobile node. 发起移动节点12a随后向接收移动节点确认所选半永久时隙的选择。 Initiating mobile node 12a then confirms selection of the selected semi-permanent time slot to receive the mobile node.

每个移动节点还包括一个与收发器14连接的全向天线20,用于与其它相邻的移动节点交换位置信息。 Each mobile node further comprises a transceiver 14 connected to the omnidirectional antenna 20 for exchanging positional information with other neighboring mobile nodes. 可交换的其它信息包括资源要求和可能的新的相邻节点的存在检测。 Other information includes detecting the presence of exchangeable resource requirements and possible new neighboring nodes. 另外,相控阵天线16可同时产生多个天线波束,其中在预定的半永久时隙内,控制器18使相控阵天线对准多个相邻的移动节点。 In addition, the phased array antenna 16 may simultaneously generate multiple antenna beams, wherein the semi-permanent time slot within a predetermined, the controller 18 controls the phase array antenna aligning a plurality of adjacent mobile nodes.

干扰检测单元18d检测并避免波束宽度内被分配相同时隙的共线节点对的干扰。 Interference detection unit 18d detects and avoids being assigned the same time slot within the beamwidth collinear node pair interference. 例如,参见图1,节点12a和12e分别在它们的相同的分配时隙1的一半内向节点12b和12c发射。 For example, referring to Figure 1, nodes 12a and 12e, respectively, of their emission in the same half of an allocated slot inward node 12b and 12c. 在天线波束宽度足够宽的情况下,节点12b和12c可同时收到来自节点12a和12e的发射。 In the antenna beamwidth wide enough, the node 12b and 12c may simultaneously receive transmissions from the nodes 12a and 12e. 干扰检测单元18d可在时隙使用期间,在物理层测量信号干扰比(SIR)。 During the interference detection unit 18d may be used in a time slot, the physical layer in the measurement signal to interference ratio (SIR). 另一方面,可根据CRC检验失效,在链路层测量分组差错率。 On the other hand, according to the CRC check fails, the measured packet error rate at the link layer. 如果这些测量结果违反了规定极限,那么可断言该时隙是不良的。 If the results of these measurements in breach of the limit, then you can assert that slot is bad.

但是,由于衰减可导致单一时隙未能通过该测试,因此最好是如果时隙的n次试验中的m次受到衰变,那么断言该时隙中干扰过大。 However, due to attenuation can cause a single slot failed to pass the test, so it is best if the n trials in slot m times by the decay, then the assertion that slot interference is too large. 此时,控制器18尝试避免干扰。 At this time, the controller 18 attempts to avoid the interference. 对于该时隙来说,位于链路两端的Tx/Rx的顺序可被转换。 For the time slot, the order of Tx / Rx at both ends of the link may be converted. 如果这样的转换失败,那么可协调新的时隙。 If this conversion fails, then you can coordinate the new time slot. 当然应随机进行这两种变化,以降低两个节点对试图同时进行相同变化,从而仍然冲突的可能性。 Of course, these two should be randomly varied to reduce the likelihood of two nodes simultaneously trying to change the same, thus still conflicting.

通信协调单元18e管理由流式视频或高速率传感器数据产生的不均衡的通信负荷。 Communication management coordination unit 18e uneven traffic load by streaming video or high rate sensor data generated. 提供协调机制以便允许每个半双工链路在通信量的任何Tx/Rx分离中分配时隙。 Provide coordination mechanisms to allocate time slots to allow each half-duplex link traffic in any Tx / Rx separation. 另外,时隙的最大数目可被增大到高于最小值的数值,以产生更多的需求时隙。 Further, the maximum number of time slots may be increased to above the minimum value, to generate more demand time slots. 子时隙化(subslotting)允许有效地增大或降低时隙的最大数目,因为节点可从半永久分配时隙“偷取”子时隙,重新分配给需求时隙。 Sub-slot of (subslotting) allows to efficiently increase or decrease the maximum number of slots, since the node from the semi-permanently assigned time slots "steal" subslots, reassigned to demand time slots. 此外,预约协议可和链路调度协议一起用于指示为适应高速率流,通过在沿着源节点到目的节点的路径上的每个节点请求和分配一组大批时隙和/或子时隙,相对于沿着该路径的每个节点的高速率流的资源分配。 In addition, appointments and agreements can be used to indicate the link scheduling protocol together to adapt to the high-speed flow rate through the requests and assign each node on the source node to the destination node along the path of a group of a large number of slots and / or sub-slots , resource allocation with respect to the high rate of each node along the path of the stream. 对于预约的资源来说,独立的队列和队列服务规程是必要的,以确保高速率流所需的容量被释放。 For reserved resources are concerned, separate queues and a queue service procedures are necessary to ensure a high rate of flow required capacity is released.

本发明的目的还在于提供一种为多个移动节点12a-12h建立通信链路的方法,其中每个移动节点包含收发器14,与所述收发器连接的相控阵天线16,和与所述收发器连接的控制器18。 Object of the present invention is to provide a plurality of mobile nodes 12a-12h establishing a communications link, wherein each mobile node comprising a transceiver 14, a phased array antenna connected to the transceiver 16, and with the a controller connected to said transceiver 18. 该方法包括为每个移动节点12a-12h,调度每个时间帧的一个相应半永久时隙,以便建立与相邻移动节点的通信链路,并在每个时隙帧中保留至少一个可用时隙。 The method comprises 12a-12h, each time frame of a scheduling respective semi-permanent time slot to establish a communication with neighboring mobile node link, and retained in each slot in a frame at least one available time slot for each mobile node .

所述至少一个可用时隙最好被安排成根据链路通信需求,服务与相邻移动节点的通信链路。 Said at least one slot is preferably arranged to communicate according to a communication link link demand, the mobile node with an adjacent service available. 在与每个相邻的移动节点12a-12h通信期间,使相控阵天线16瞄准每个相邻的移动节点12a-12h。 During each neighboring mobile node 12a-12h communication, the phase array antenna 16 is aimed at each neighboring mobile node 12a-12h. 每个时间帧可具有多达N个半永久时隙和至少2N-1个可用时隙。 Each time frame may have up to N semi-permanent time slots and at least 2N-1 available time slots.

该方法还可包括启动对相应时间帧的一个或多个半永久时隙请求,以建立与每个相邻移动节点的通信链路,并在每个时间帧中保留至少一个可用时隙,同时处理从相邻移动节点接收的多个半永久时隙请求,启动至少一个可用时隙请求以便还根据链路通信需求,服务于与相邻移动节点的通信链路,同时处理从相邻移动节点接收的多个可用时隙请求。 The method may further comprise a respective start time of a frame or a plurality of semi-permanent time slot request to establish a communication link with each neighboring mobile node, and to leave at least one available time slot in each time frame, while processing receiving a plurality of semi-permanent time slot requests from neighboring mobile nodes, initiating at least one available time slot request to also link communication in accordance with demand, the communication link serving the neighboring mobile nodes, while processing received from the neighboring mobile node and Request more available slots.

在与每个相邻的移动节点12a-12h通信期间,使定向/相控阵天线16对准每个相邻的移动节点12a-12h,在用于与相邻的移动节点通信的时隙中检测干扰,根据检测的干扰协调新时隙的调度。 In-12a 12h period and each adjacent mobile node communication, so that the directional / phased array antenna 16 aligned with each neighboring mobile node 12a-12h, in time slots for communication with neighboring mobile nodes in detecting interference, the interference detection coordinate scheduling of new time slot. 干扰检测单元18d可测量信号干扰比和/或分组差错率。 Interference detection unit 18d measures the signal to interference ratio and / or a packet error rate. 分组差错率可基于循环冗余检验(CRC)失效。 Packet error rate may fail based on cyclic redundancy check (CRC). 另外,干扰检测单元18d可比较检测的干扰与阈值。 In addition, the interference detection unit 18d may compare detected interference to a threshold. 最好,控制器根据检测的干扰,持续指定的时隙交换节点之间的通信顺序,还可在交换通信的顺序之后,根据检测的干扰,协调新时隙的调度。 Preferably, the controller detects the interference, the communication sequence between the switching node for a specified time slot, but also after switching the order of communication, based on the detected interference, to coordinate scheduling of new time slot. 另外,通过根据链路通信需求,分配用于调度的时隙,可协调与每个相邻移动节点12a-12h的通信。 Further, according to the link communications demand assigned time slots for scheduling, may coordinate communication with each neighboring mobile node 12a-12h of.

该方法还包括使每个节点按优先顺序排列通信链路,并根据优先化中断通信链路之一,以使半永久时隙可用于建立与新的相邻移动节点的通信链路。 The method also includes each node prioritize the communication links, and based on the priority of the interrupt one of the communication links, so that the semi-permanent time slots may be used to establish a new neighboring mobile node to the communication link. 另外,根据链路需求,正被调度以便服务特定通信链路的可用时隙可被重新分配给另一通信链路。 Further, according to the link demand, being to serve a particular communication link scheduling of available time slots may be reassigned to another communication link. 这有利地允许任何移动节点适应通信链路需求方面的变化。 This advantageously allows any mobile node to adapt to changing needs of the communication link.

现在将更详细地说明半永久时隙和可用时隙的调度。 Now described semi-permanent time slots and available time slots scheduled in more detail. 使定向天线16转向接收移动节点12a-12h的有关细节将被省略,因为本领域的技术人员易于理解本发明的该特征。 So that steering the directional antenna 16 receiving mobile node 12a-12h of the details will be omitted, since those skilled in the art will readily appreciate that the features of the present invention.

为了便于讨论,将假定定向天线16是相控阵天线。 For ease of discussion, will be assumed that the directional antenna 16 is a phased array antenna. 本领域的技术人员易于认识到,相控阵天线16包括可被调整以便在所需方向产生可转向的天线波束的多个天线元件和相应的移相器。 Those skilled in the art will readily recognize that the phased array antenna 16 may be adjusted so as to comprise a plurality of antenna elements to produce steerable antenna beam and the corresponding phase shifters in the desired direction. 相控阵天线16操纵或扫描天线图,而不必物理移动天线。 Phased array antenna 16 to manipulate or scan antenna diagram, without having to physically move the antenna.

另外为了便于讨论,做出与无线移动通信网络10相关的许多假设。 Also for ease of discussion, a number of assumptions made with a wireless mobile communication network 10 related. 首先,存在一个单一频带,该频带是由所有无线移动节点12a-12h共享的高数据速率通道。 First, there is a single frequency band, the band is composed of all the wireless mobile nodes 12a-12h sharing the high data rate channel. 这种传输通道在所有无线移动节点12a-12h之间被分时用于发射和接收。 This transmission path between all the wireless mobile nodes 12a-12h are used for transmitting and receiving time division. 所及传输时隙被预先安排。 Clinics transmission slots are pre-arranged.

另外假定提供一个独立的低数据速率开销通道(overheadchannel)。 Also assume that a separate low data rate overhead channel (overheadchannel). 该开销通道可被用于节点发现,网络进入,和各种其它数据链路控制开销信息,包括资源请求的交换。 The overhead channel can be used for node discovery, Internet access, and a variety of other data link control overhead information, including the exchange of resource requests. 通过全向天线20提供这种开销通道。 This overhead by providing a full 20-channel antenna. 在所有节点还知道精确的全球定时基准。 In all nodes also know the exact global timing reference. 在下面的讨论中,术语无线移动节点和无线移动通信系统12a-12h是可互换的。 In the following discussion, the term wireless mobile nodes and wireless mobile communications systems 12a-12h are interchangeable.

无线移动通信网络10还包括定位和跟踪移动节点,以致当预定的时隙可用时,相控阵天线16可被准确指向的能力。 Wireless mobile communication network 10 also includes a locating and tracking mobile nodes so that when a predetermined time slot is available, a phased array antenna 16 may be the ability to accurately pointed. 如上所述,这里将不提供有关指向/跟踪的详细讨论。 As mentioned above, there will be no more discussion of the pointing / tracking available.

另外假定相控阵天线16具有零波束宽度。 Also assume that the phased array antenna beam width 16 with zero. 该假定将在后面放宽。 This assumption will be relaxed later. 从而,我们能够假定指定移动节点的发射将只被它正试图向其发射的相邻移动节点接收。 Thus, we can assume that a given mobile node will only be transmitting it is trying to receive transmitted to its neighboring mobile nodes. 这允许关于时隙调度的一组不太严格的约束。 This allows a less restrictive set of constraints on the scheduling of time slots. 每个通信链路可用数字标记,所述数字表示发射和接收数据的预定时隙。 Each communication link is available digital signature, the digital representation of the predetermined time slot to transmit and receive data.

所述约束如下所述。 The constraints are as follows. 没有任何节点具有一个以上的用相同时隙编号标记的通信链路。 No node has more than one communication links with the same slot number tag. 指定时隙分配将被应用于两个节点之间的半双工链路,并被这两个节点交替用于发射和接收。 Specifies the time slot assignment will be applied to half-duplex link between two nodes, the two nodes and is alternately used for transmission and reception. 这两个约束意味着一个移动节点分配给其相邻节点之一的时隙受该节点分配给其它链路的在先时隙约束。 This constraint means that the two time slots allocated one mobile node to one of its neighboring nodes by the previous time slot assigned to the node constraint other link.

图1中图解说明了相控阵天线16的时隙的调度,图1表示了具有基于预定时隙的链路连通性的网络10。 Illustrated in Figure 1 illustrates a phased array antenna slot 16 of scheduling, Figure 1 shows a time slot having a predetermined based on the connectivity of the network link 10. 时隙被调度以致无线移动节点12a-12h知道何时使它们各自的相控阵天线16指向相邻的无线移动节点。 Slots are scheduled so that the wireless mobile nodes 12a-12h know when to bring their respective phased array antenna 16 pointing neighboring wireless mobile node.

通信链路被假定为双向的,并且可按照半双工方式使用,其中每个时隙编号代表一个时隙和在该时隙中发生的每个方向上的传输机会。 Communication link is assumed to be bidirectional and can be used in accordance with the half-duplex mode, where each time slot number represents a time slot and a transmission opportunity in each direction occurring in that time slot on. 术语Nframe将被用于表示某一帧内的最大链路索引或者时隙的最大数目。 The term Nframe will be a maximum number of intra maximum link index or time slots used to represent. 就本例来说,Nframe=6。 For this example, Nframe = 6.

图3图解说明一帧典型的时隙。 Figure 3 illustrates a typical time slot. 在最简单的公式表示中,每个信号出现时间或帧具有n个时隙,n的值被设置为Nframe。 In the simplest formula, each epoch or frame has n slots, the value of n is set to Nframe. 图3中还表示了时隙如何被用于连接标记为节点A和B的节点的链路。 Figure 3 also shows how the time slots are used for the link connecting the node labeled A and B. 每个时隙被分成两个微小时隙22a、22b。 Each time slot is divided into two small slots 22a, 22b. 第一个微小时隙22a(例如该时隙的一半)被用于从节点A到节点B的传输。 The first small slot 22a (e.g., half of the time slot) is used for transmission from node A to node B. 随后,该链路的方向被反转,第二个微小时隙22b被用于从节点B到节点A的传输。 Subsequently, the direction of the link is reversed, the second slot 22b is used for small from node B to node A's transmission.

在传输周期内,可传送多个分组。 In the transmission period, multiple packets may be transmitted. 如图所示,每个微小时隙22a、22b还包含根据下面考虑因素选择的一个保护时间24a、24b。 As shown, each small slot 22a, 22b also contains a selection from the following considerations a guard time 24a, 24b. 任意一对节点之间的最大距离确定必须适应的最大传播延迟。 Determining the maximum distance between any pair of nodes must be adapted to the maximum propagation delay. 100英里的最大距离对应于约0.5毫秒的传播延迟。 The maximum distance of 100 miles corresponds to about 0.5 ms of propagation delay. 为每个微小时隙22a、22b分配保护时间,以适应所有各对节点之间的传播延迟的不确定性和不相等的传播延迟。 For each small slot 22a, 22b assigned guard time to adapt to spread among all the pairs of nodes delay uncertainty and unequal propagation delay.

在100英里的最大范围下,需要0.5毫秒的保护时间。 In the 100 miles to the maximum extent, the need to protect the time of 0.5 milliseconds. 关于100英里的最大范围的保护时间分配意味着需要使微小时隙22a、22b约为2-4毫秒,以使通道效率损失降至最小。 The maximum range of about 100 miles of protected time allocation implies the need to make the tiny slot 22a, 22b is about 2-4 milliseconds, so to minimize the loss of channel efficiency. 例如,如果假定通信链路上的数据速率为0Mb/s,并且最大距离为100英里,那么4毫秒的微小时隙意味着200000比特/微小时隙(250微小时隙/秒)。 For example, if it is assumed the data rate on a communications link for 0Mb / s, and a maximum distance of 100 miles, then a 4 msec time slot implies 200,000 bits micro / micro time slot (slot 250 minute / second). 从而,微小时隙会包含25000比特保护时间和175000比特的遗漏数据。 Thus, the small slot will contain 25,000 bits guard time and 175,000 bits of missing data.

控制器18还可偏向每个建立的链路,以便当调度可用时隙时分配优先级。 The controller 18 may also bias each established link to assign priority when the available time slot scheduling. 如下更详细所述,在每个帧内提供半永久(SP)时隙和可用或按需分配(DA)的时隙。 As described in greater detail below, provide semi-permanent (SP) time slots and available or demand assignment (DA) time slots in each frame. 目的是增大同一时间几个节点之间的时隙的复用。 Purpose is to increase the number of time slots between the same node multiplexing. 虽然图1中的移动节点10在节点和通信链路的总数方面受到限制,不过存在并行使用时隙的许多情况。 Although in Figure 1 the mobile node 10 is limited in terms of the total number of nodes and communication links, but in many cases the use of parallel slots exist. 例如,时隙1和2均同时在3个不同的通信链路上被使用,时隙6只在一个链路上被使用。 For example, slots 1 and 2 are simultaneously on 3 different communication links are used, the time slot 6 is used on one link. 所有其它时隙被分配给两个通信链路。 All other time slots are assigned to two communication links. 我们可把指示平均复用水平的复用因子定义为网络中的时隙分配的总数(Nframe)与分配的时隙的数目(Num_Slots_Assigned)的比值:R=Num_Slots_AssignedNframe...(1)]]>对于图1中的例证网络10,复用方法提供R=14/6=2.333的复用因子,指示平均来说,在网络调度中,每个时隙存在稍多于2个的同时用户。 We can indicate the number average multiplexing level reuse factor is defined as the total number of slots allocated in the network (Nframe) with the assigned slot (Num_Slots_Assigned) ratio: R = Num_Slots_AssignedNframe ... (1)]]> For the example of FIG. 1 in the network 10, provided with a multiplexing method of R = 14/6 = 2.333 reuse factor, indicating average, network scheduling, each time slot of the presence of slightly more than two simultaneous users. 显然关于任何具体的调度算法计算的复用因子与网络规模和布局高度相关。 Clearly on any specific scheduling algorithm of the reuse factor highly correlated with network size and layout. 完全相当的市场段应考虑各种网络规模和布局。 Completely considerable market segment should consider all network size and layout.

通过注意每个节点至少需要和该节点具有的邻居一样多的时隙,即该节点需要至少等于其度(degree)的许多时隙,能够确定任意图的Nframe的值的下限。 Note that each node by at least as much, and the node has neighbors slot, that is, the node needs at least equal to its degree (degree) of many time slots can be determined arbitrarily limit values Nframe map. 从而,Nframe必须至少和整个图内的最大节点度一样大。 Thus, Nframe must be at least within the maximum node degree as large as the entire FIG. 从而,通过用di表示节点i的节点度,Nframe的值的下限为Nframe≥maxi{di} (2)对于图2中图解说明的例证网络,复用部分被分配具有根据等式2,等于必须使用的时隙的最小数目的Nframe的调度。 Thus, the lower limit represented by di node degree of node i, Nframe values for Nframe≥maxi {di} (2) For example network illustrated in Figure 2, are allocated with portions having multiplexed according to Equation 2, must be equal to Nframe the minimum number of time slots used for scheduling. 注意几个节点,即除节点1之外的所有节点被分配少于完整的一组时隙。 Note that several nodes, namely all nodes except the nodes other than 1 are assigned less than the full set of time slots. 从而,增强的调度算法能够向一些链路分配额外的时隙,而不会在调度上引起冲突。 Thus, enhance the scheduling algorithm can allocate additional slots to some links without causing conflict in scheduling.

下面的讨论主要集中在用于产生链路调度的时隙的调度上。 The following discussion focuses on the scheduling of time slots for generating the link scheduling. 最终必须解决的整个相控阵网络问题的其它部分包括:1)节点和邻居发现,2)网络进入,3)开销通道格式和协议,包括调度更新的协议交换,4)相邻节点的跟踪和定位(可包括相控阵天线16的帮助,和5)动态网络布局的路由算法。 The other part of the phased array network problem must ultimately be addressed include: 1) node and neighbor discovery, 2) Internet access, track 3) overhead channel formats and protocols, including the protocol exchange scheduling updates, 4) and the adjacent nodes location (which may include helping the phased array antenna 16, and 5) the dynamic network topology routing algorithm.

根据本发明的调度时隙的方法基于下述原则。 Based on the principle that the scheduling time slots according to the method of the present invention. 首先,规定数目的时隙被分配成为指定链路安排的半永久(SP)时隙。 First, the required number of slots are allocated as a designated link arrangement of semi-permanent (SP) time slots. 剩余的可用时隙(DA)可根据按需分配,被分配给最需要它们的那些节点/链路。 Remaining available slots (DA) according to DAMA, is assigned to those who need them most nodes / links. 这允许根据需要灵活地改变调度。 This allows the flexibility to change according to the needs of the scheduling. 其次,如上所述,确定关于半永久分配时隙的最大数目的极限。 Secondly, as described above, determine the maximum number of semi-permanently assigned time slots on the limit. 该极限是根据具体的网络选择的一个参数。 This limit is a parameter depending on the network selection. 在每个节点一个SP时隙的情况下,该极限也是关于许可的相邻节点的数目的上限。 In one SP time slot per node case, also the limit on the maximum number of neighbor nodes Licensing.

第三,如上所述,确定每帧的时隙的最大数目的极限。 Third, as described above, determine the maximum number of time slots per frame limit. 该极限也是根据具体网络选择的一个参数。 The limit is based on a specific network selection parameter. 该极限对确定有关等待时间的限度是重要的,因为它确定链路传输机会的最大再访问时间。 The limit for determining the limits of the relevant latency is important, because it determines the maximum uplink transmission opportunity then access time.

第四,选择每帧的时隙的总数Nframe和每帧的半永久分配时隙的最大数目的极限之间的关系,以致半永久分配时隙的调度被大大简化,并且即使在分布式调度的情况下,也可显著避免调度冲突。 Fourth, the relationship between the maximum number of selection of the total number of slots per frame, Nframe, and each frame of the semi-permanently assigned time slots between limits, so that semi-permanently assigned time slots is greatly simplified scheduling, and even in the case of distributed scheduling , can also be significantly avoided scheduling conflict.

通过把每个节点的半永久分配时隙的最大数目限制成每帧的时隙总数的几分之一,大大简化了分布式分配半永久分配时隙的过程。 By limiting the maximum number of semi-permanently assigned time slots per node to a fraction of the total number of slots per frame, which greatly simplifies the process of distributed allocation of the semi-permanently assigned time slots. 半永久分配的数目的上限(于是,许可的相邻节点的最大数目)由N表示。 The upper limit of the number of semi-permanently assigned (Thus, the maximum permitted number of neighboring nodes) represented by N. 我们将考虑Nframe的值,以致:Nframe≥2N-1 (3)假定网络10中的所有节点12a-12h都由定向链路连接,每个节点具有通过跳时实现波速共用,并且指向其相邻节点的单波束相控阵天线16。 We will consider values of Nframe such that: Nframe≥2N-1 (3) assumes that all nodes in the network 10. 12a-12h is connected by directional links, each node having a common velocity to achieve through the jump, and points adjacent thereto Single-beam phased array antenna node 16. 此外,假定邻居的数目等于N,半永久时隙(同时每个邻居分配一个SP时隙)的许可数目的极限被固定。 In addition, the number of neighbors assumed equal to N, semi-permanent time slot (and each neighbor is assigned a time slot SP) limit the number of licenses to be fixed.

如果Nframe的固定值满足等式(3),那么通过关于这些链路中的每个链路,与邻居相互达成协议,所有节点能够为该链路选择一个不同的半永久时隙,而不考虑远离一个中继段以上的其它节点正在选择什么链路。 If the fixed value of Nframe satisfies equation (3), then through each of these links in the link, mutual agreement with the neighbor, all nodes can select a different semi-permanent time slot for the link, regardless away Other nodes are more than one hop to choose what link. 这允许每个节点通过只与相邻节点通信,以非常直接的方式为到该相邻节点的链路选择其半永久时隙。 This allows each node to communicate with neighboring nodes only, in a very straightforward way to link the adjacent node to select its semi-permanent time slot. 可对多达N个相邻节点遵循该过程。 Can for up to N neighbor nodes follow the process.

关键是认识到对于N的固定值,当Nframe的值增大时,对节点的选择不与邻居的时隙选择冲突的时隙的能力的约束较少。 The key is to recognize that, for a fixed value of N, when the value of Nframe increases, fewer constraints on the selection of a node does not conflict with the time slot of the time slot selection neighbors ability. 为新链路选择时隙的节点必须选择当前未被使用,并且邻居没有正在使用的时隙。 Select the slot for the new link node must choose not currently in use, and the neighbors are not being used slot.

如果节点目前具有m个邻居(到所述邻居的这些链路中的每个链路被分配单一时隙),并且正在增加到新的相邻节点的链路,那么相邻节点最多能够使用(N-1)个时隙。 If a node currently has m neighbors (these links to the neighbors of each link is assigned a single time slot), and a new link is added to neighboring nodes, then the neighbor node can be using at most ( N-1) time slots. 从而,如果Nframe大于(m+N-1),那么将存在该节点可分配给新链路的至少一个以上的时隙。 Thus, if Nframe is greater than (m + N-1), then the presence of the node can assign to the new link is at least one or more slots. 该分配过程中的最坏情况是节点已具有(N-1)个邻居,并且正在为第N个相邻节点分配时隙的时候。 The worst case allocation process is the node already has (N-1) neighbors, and the N-th adjacent node is allocated time slots for. 这种情况下,对于要确保可供分配给到第N个邻居的链路的额外时隙来说,Nframe必须满足等式(3)。 In this case, to ensure that available to be assigned to the N-th link neighbor additional time slot, Nframe must satisfy equation (3).

下面将关于在所公开的时隙调度方法中如何利用该性质,进行另外的一些观察评论。 The following will be made regarding the time slot scheduling method disclosed in how to use the nature, some observations were further comments. 首先,节点只需要与邻居协调将为到该邻居的定向链路分配的半永久时隙的选择。 First, the node will only need to coordinate with neighbors to semi-permanent time slot allocated to the neighbor's choice of a directional link. 请求该链路的节点可向所述邻居发送该链路的建议时隙列表。 Node requesting the link can be sent a list of suggestions for the link slot to the neighbors. 这是以这些时隙未被用于SP分配为基础的。 This is not used for SP allocated time slots based. 根据下面要讨论的其它因素,可存在该列表的其它排序,不过这不是必需的。 According to other factors to be discussed below, there may be other ordering of the list, but this is not required. 相邻节点随后能够从该列表中选择它更喜欢的时隙,并返回答复及该选择。 Neighboring nodes is then able to select the time slot it prefers from this list, and returns the reply and the selection. 这允许我们定义一种直截了当、完全分布的半永久时隙调度方法。 This allows us to define a straightforward, fully distributed semi-permanent time slot scheduling method.

如果节点的邻居少于N个,那么在单个链路上可分配其N个许可的半永久时隙中的一个以上的半永久时隙。 If a neighbor node is less than N, then N number of licenses can be assigned its semi-permanent time slot of one or more semi-permanent time slot on a single link. 但是,这种情况下,不能保证无冲突地通过近邻间的节点协调完成所有N个分配。 However, this case is not guaranteed to complete without conflict coordinate distribution through all N neighbor nodes between. 例如,如果N=6,并且一个节点只具有3个近邻,但是这些近邻均具有6个邻居,那么该节点能够只向与其3个近邻的每个链路分配一个时隙。 For example, if N = 6, and a node having only 3 neighbors but each of these neighbors have six neighbors, then the node can assign only one time slot to each link to its three neighbors. 为了简化我们的算法,我们将不允许每个链路一个以上的SP时隙的调度。 In order to simplify our algorithm, we will not allow more than one per link SP time slot scheduling. 但是,所有未用的时隙可被分配为可用时隙。 However, all unused slots can be assigned to an available slot.

对于具有非常大量的节点的某些网络(其中潜在近邻的数目将远远大于极限N),也存在要处理的布局控制(topology control)问题。 For a very large number of nodes having a certain network (where the number of potential neighbors will be much larger than the limit N), there are also the control layout to be processed (topology control) problems. 节点将面临从潜在近邻中选择产生最佳网络布局的那些近邻的问题。 Node will face the choice of potential neighbors from those produced in neighboring best network layout problems. 这种布局控制问题还与优化能量高效网络的概念相关。 This layout control problem is also associated with the concept of optimizing the energy efficiency of the network. 在潜在近邻的数目远远大于极限N的情况下,可使用布局控制功能来选择要连接的相邻节点。 In case the number of potential neighbors is much larger than the limit N, use the layout controls to select adjacent nodes to be connected.

如果把(3)许可的最小值分配给Nframe,那么每个节点将被允许具有最多N个半永久时隙和总共(2N-1)个时隙分配。 If (3) the minimum value assigned permission to Nframe, then each node will be allowed to have a maximum of N semi-permanent time slots and a total of (2N-1) time slots are assigned. 将基于最佳适应通信负载地分配按需分配的时隙。 The slot allocation based on demand to best accommodate the traffic load. 当然,分配更大值的Nframe也是一种选择。 Of course, a larger value of Nframe is also assigned an option. 这种情况下,将存在可用于需求分配的更多时隙。 In this case, there will be more time slots available for demand assignment. 可存在对其来说这是配置网络的最佳方式的应用。 Can exist on its application, this is the best way to configure the network.

如同半永久时隙一样,节点只需要与某一近邻协调将为到该近邻的定向链路分配的可用时隙的选择。 As a semi-permanent time slot as a neighbor node only needs to coordinate with the neighbor's orientation will be to link the distribution of the available time slots to choose from. 这意味着某一近邻将通过所述定向链路,向所述近邻发送时隙分配的请求,并通过同一链路接收分配的准许或者请求的拒绝。 This means that a neighbor will be directed through the link, send to the neighbor slot allocation request, and receives the allocated grant or reject the request through the same link.

向近邻请求可用时隙DA的分配的节点将根据感知的对该链路上的额外容量的需要向近邻请求可用时隙DA的分配。 Neighboring nodes to request the allocation of available slots of DA will be available for distribution to neighboring slots DA request extra capacity according to the perceived needs of the link. 这可由基于短期和长期测量的高链路利用(队列形成)推动。 This can be based on short-term and long-term high link utilization measurements (queue formation) push. 该请求将包含所请求的时隙的数目和量度,所述量度指示将附着到所述请求上的优先级。 The request will contain the number of slots requested and a metric, said metric indication attached to the priority of the request on. 所述量度可把队列长度表示成对时隙分配的需要的尺度。 The measure can be expressed the need to queue length scales pairs of slot allocation.

接收请求的节点还可接收来自其它相邻节点的请求,所述其它相邻节点可能争夺相同时隙的分配。 The node receiving the request may also receive requests from other neighbor nodes, the other neighboring nodes may be assigned the same time slot contention. 为了简化时隙,在考虑下一分配之前,节点必须完成处理一个可用时隙DA分配的一个线程。 To simplify time slot, before considering the next allocation, a node must complete processing one available time slot DA allocation of a thread. 这些分配可以不必持续长期的时间,因为由于布局变化的结果,它们不断受到预占,从而被重新分配成半永久时隙,或者由于改变通信量需求,而受到重新分配。 These allocations may not necessarily be a long-term duration of time, because the results due to changes in the layout, they continue to be camped on, so as to be reassigned to a semi-permanent time slot, or due to changes in traffic demand, subject to re-allocation.

现在将讨论近邻和链路发现。 Neighbors will now be discussed and link discovery. 在与潜在的相邻节点建立定向链路之前,分布式链路调度算法需要全向开销通道对必须与所述潜在的相邻节点发生的某些协议交换的支持。 Before establishing a directional link with potential neighboring nodes, the distributed link scheduling algorithm requires support some of the protocols that must occur with a potential neighbor nodes exchange the whole of the overhead channel. 这样的消息包括REQ_SPTS,REQ_SPTS请求到该节点的定向链路上的半永久时隙的分配。 Such message comprises REQ_SPTS, REQ_SPTS semi-permanent time slot allocation request to the orientation of the node on the link.

除了直接支持这里定义的协议的支持协议消息交换之外,全向开销通道必须支持近邻和链路发现的功能。 In addition to supporting protocol message exchange directly support the protocol defined here, omni-directional overhead channel must support the function of neighbor and link discovery. 这通常是通过由每个节点通过全向天线20进行的周期性全向发射实现的,所述周期性全向发射向在范围内移动的任何其它节点通知这两个节点可以是相邻节点。 This is usually done by the whole performed by each node via the antenna 20 to the transmitting periodic full realization, the periodic full notification these two nodes to transmit to any other node within the range of movement may be a neighboring node. 几种自组织路由协议(包括OLSR)定义了这样的支持协议。 Several self-organizing routing protocols (including OLSR) defines such a support agreement. 这些先前定义的协议可被修改,以支持这种分布式链路调度算法。 These agreements previously defined can be modified to support this distributed link scheduling algorithm. 这种协议的主要功能是发现新的潜在相邻节点,并把这些报告给布局控制功能。 The main function of this protocol is to discover new potential neighbor nodes, and these reports to the layout control.

节点和链路发现的一种方法包括每个节点通过控制通道定期发射信标消息,把它的存在和它的位置通知相邻的节点。 A method for node and link discovery includes each node periodically transmitting beacon messages over the control channel to notify neighbor nodes of its presence and its position. 另外定期发射链路状态消息,把其信标近邻(BN列表)和其PA近邻节点(PAN列表)的身份以及分配给这些节点的时隙通知相邻节点。 Also periodically transmit link status messages to its beacon neighbors (BN list) and its PA neighbor nodes (PAN list) as well as the identity of these nodes are assigned to slots notify neighbors.

该算法的链路发现部分不断比较双向信标近邻(BBN)列表与PAN列表,查看在BBN列表上是否存在不在PAN列表上的任意节点。 Link found that some of the algorithm constantly compared bi beacon neighbors (BBN) list with the PAN list to see if any node on the list if PAN is not present on the BBN list. 任何这样的相邻节点变成确定PA链路是否可能的链路测试的候选者。 Any such neighbor node becomes possible to determine whether the link PA link test candidate. 根据这种方法,在交换控制消息之后,测试定向链路以确定可靠的通信是否可能。 According to this method, after the exchange of control messages the directional link is tested to determine if reliable communication is possible. 如果通信是可靠的,那么新的相邻节点被添加到PAN列表中。 If communication is reliable, the new neighbor node is added to the PAN list.

这在测试时隙中确认通信,但是不一定在可半永久地分配给链路的时隙中。 This confirms that the communication in the test time slot, but not necessarily be semi-permanently assigned time slots to the links. 一种方法是按照这种方式来进行,或者另一种方法是等待,直到SP时隙被分配并在该时隙中测试通信。 One approach is to perform in this way, or another method is to wait until the SP time slot is assigned and test communications in the time slot.

如果布局控制功能不必进行布局优化,那么布局控制功能可以是非常简单的功能。 If the layout of the controls is not necessary to optimize the layout, so the layout of the control function can be very simple function. 该功能的目的是获得PAN列表中的节点的列表,关于这些链路的可靠性的信息,和有关网络布局的信息,并使用该信息来确定PAN列表上的哪些节点应成为PA近邻。 The purpose of this feature is to obtain a list of nodes in the PAN list, the information about the reliability of these links, and the information about the network topology, and use this information to determine which nodes on the PAN list should become PA neighbors. 这是如果存在诸如PA近邻的数目不允许PAN列表中的所有节点都成为PA近邻之类约束条件,那么应优化网络布局的功能。 That is if there is a list of all the nodes in the PAN, such as the number of PA neighbors are allowed to become PA neighbors such constraints, you should optimize the function of the network layout.

就提出的Nframe的固定值和N的固定值(每个节点的半永久时隙的最大数)的约束条件来说,存在与网络布局利用有一定关系的潜能。 To a fixed value of Nframe raised and fixed value of N (the maximum number of semi-permanent time slots per node) constraints, the existence of the network layout utilizing a certain relationship potential. 如果这些值被选择成很小的数字,那么情况肯定是这样。 If these values are selected to a small number, then the situation is certainly so. 例如,如果在Nframe=5的情况下选择N=3,那么当对于任何节点,我们不能具有3个以上的近邻时,难以期待良好连接的网络布局,除非在增加新的PA近邻节点之前,智能布局控制功能仔细地利用该布局。 For example, if the case Nframe = 5 choices N = 3, then when any node, we can not have three or more neighbors, the network layout is difficult to expect a good connection, unless, prior to the addition of new PA neighbor nodes, intelligent Layout control functions carefully using this layout. 对于大型网络来说尤其如此。 This is especially true for large networks.

从而,布局控制功能应创建近邻优先级(NP)列表,它是按照作为潜在PA近邻的需要性的顺序排序的PAN列表。 Thus, the layout of the control function should create neighbor priority (NP) list, which is the PAN list sorted in the order as a potential PA needs of neighbors. 该列表将指导优先级顺序,其中潜在的PA近邻被安排时隙。 This list will guide the priority order in which potential PA neighbors are arranged slots. 但是,我们的首要问题是可能具有15个节点的小型网络的问题。 However, our primary problem is likely to have 15 nodes of a small network. 这种情况下,我们可指定N具有在5-8范围内的值,并且仍然具有低的等待时间。 In this case, we have to specify the value of N in the range of 5-8, and still have low latency. 由于考虑5-8个相邻节点将允许几乎所有可能的近邻成为PA近邻,因此存在任何布局利用问题的可能性很小。 In consideration of 5-8 adjacent nodes will allow almost all possible neighbors to become PA neighbors, so there is no possibility of utilization of small layout.

布局控制功能的另一目的是产生布局改变事件,布局改变事件使链路调度器进程改变状态,并对SP时隙进行重新分配过程。 Another object of the layout of the control function is to generate the layout change events, the layout changing event that the link scheduler process to change state, and the SP slot reallocation process.

现在将讨论顶级调度算法结构。 Will now discuss the structure of the top-level scheduling algorithm. 以使调度过程的复杂性降至最小,同时利用上面概述的整个方法的目的,用公式表示所述调度过程。 So that the scheduling process is to minimize the complexity, while taking advantage of the overall process outlined above object, using the equation represents the scheduling process. 控制该调度的关键是在每个节点保持准确的数据结构,所述数据结构反映分配给与每个相邻节点的链路的未来时隙的时隙调度的状态。 The key controlling this scheduling is maintaining an accurate data structure at each node, said data structure to reflect the state of time slots assigned to each adjacent node of the scheduling of the future time slot of the link.

提出两种数据结构:时隙分配DB和链路消息DB。 Propose two data structures: slot assignment DB and a link message DB. 表1中列举了关于信号出现时间(epoch)中的指定时隙的数据结构中的链路的可能状态。 Listed in Table 1 on the epoch time (epoch) of the specified time slot in the data structure may link state. 表1描述了每种可能状态,并给出了该状态的符号。 Table 1 describes each possible state and gives the notation that state. 表2表示了例证的时隙分配DA和指示对于Nframe=9(N=5)来说的时隙的内容,每种状态的状态分配,和每个时隙的例证的分配近邻ID。 Table 2 shows examples of time slot assignment DA and instructions for Nframe = 9 (N = 5) for the contents of the time slot, the state of distribution of each state, and each time slot assignment illustration neighbor ID.

在该例子中,4个近邻已被分配SP时隙,从而在这些约束条件下,可连接一个额外的近邻。 In this example, 4 neighbors have been assigned SP time slots so that these constraints can be connected to an additional neighbor. 存在一个自由时隙,该自由时隙可被分配为DB时隙或者和将被分配为SP时隙的DB时隙一起提供,如果新的相邻节点是可能的话。 There is a free slot, the free time slots can be allocated as DB and the slot or slots to be allocated as a DB time slot SP provided together, if the new neighbor node is possible. 后面在详细的协议说明中将讨论链路消息DB的使用。 Use link message DB discussed later in detail in the description of the agreement. 该例子还表示子时隙,例如每个时隙2个子时隙的使用。 This example also said sub-slots, each slot such as the use of two sub-slots.

这是一种将和DA分配一起使用以便提供更细的粒度的概念。 This is a way to allocate and DA used together to provide a finer granularity of the concept. 这种情况下该含义会是时隙k的分配,子时隙1会是对奇数编号帧上时隙k的链路的分配。 In this case the meaning will be allocated slot k, sub-slot 1 would be assigned to slot k on the odd numbered frames of the link. 相反,子时隙2会指示偶数编号帧上该时隙的分配。 In contrast, sub-slot 2 would indicate that the distribution of the even-numbered frame slot.

表1 Table 1

表2 Table 2

图5中表示了链路调度协议的顶级状态图。 Figure 5 shows a top-level state diagram of the link scheduling protocol. 图5表示负责维护和修改时隙分配数据库的两个独立过程30和32。 Figure 5 shows two independent processes responsible for the maintenance and modification of slot allocation database 30 and 32. 左侧是维护和分配半永久(SP)时隙的过程,即过程30的状态图。 On the left is the maintenance and distribution of semi-permanent (SP) time slots process, that process state diagram 30. 该过程优先于右侧的过程32进行的分配,过程32负责分配可用(DA)时隙。 This occurs in preference to the right side 32 of the distribution, the process 32 is responsible for assigning the available (DA) time slots. 在过程路径中31,可被捕捉的时隙如下所示:自由的、DA分配的、和在被DA分配的过程中。 In the process path 31, the time slot can be captured as follows: free, DA allocated, and in process of being DA allocated. 类似地,在过程路径33中,可被捕捉的时隙如下所示:自由的、DA分配的和还需要被重新分配。 Similarly, in the process path 33, the time slot can be captured as follows: free, DA allocated and also need to be reallocated.

该数据库可被控制为锁定数据库,以致对于任何指定的时隙分配状态,在指定的时间,这两种调度过程中只有一种能够修改所述状态。 The database may be controlled to lock the database, such that for any specified time slot assignment state, at a specified time, both the scheduling process only one capable of modifying the state. 一旦过程之一开始修改特定时隙分配的状态,那么该状态被锁定,在所述状态被释放之前,另一过程不能对其进行修改。 Once the process begins to modify the state of a particular time slot allocation, then the state is locked before being released in the state, another process can not be modified.

任何时候,DB中的每个时隙处于如表1中所示七种状态之一。 At any time, DB each time slot in one of seven states as shown in Table 1. 可用时隙被认为处于自由状态,即,它们未被分配给到其相邻节点之一的链路,因为调度冲突妨碍了分配或者因为该时隙最近变成自由状态并且还未被调度。 Available time slots are said to be in a free state, i.e., they are not assigned to a link to one of its neighboring nodes, because a scheduling conflict prevented assignment or because the time slot has recently become a free state and has not yet been scheduled.

如上所示,处于自由状态的时隙可被调度为SP时隙或者DA时隙。 As indicated above, the time slot in the free state may be scheduled as SP time slot or a DA time slot. 已被分配为SP分配的时隙只能由维持SP时隙的过程修改。 Slots have been allocated for the SP allocation can only be modified by the process of maintaining the SP slot. 如果网络布局改变或者如果更理想的布局是可能的,那么该过程可解除分配该时隙。 If the network topology changes or if a more desirable layout is possible, then the process can be de-allocated time slot. 在这样的时隙被返回自由状态之前,维护和分配DA时隙的过程不能修改其状态。 Prior to this time slot is returned in a free state, maintenance and distribution of DA slot process can not modify its status.

另外,具有指示它在正被SP分配过程中的DB状态的任何时隙不能被DA分配过程分配。 Further, it being indicative of any SP slot assignment process DB state can not be allocated DA assignment process. 这包括指示SP请求和答复消息已被发送的状态。 This includes an indication of the state of SP request and reply message has been sent. 但是,如果时隙的状态是被DA分配,那么它可被DA分配过程重新分配。 However, if the state of the time slot is allocated DA, then it may be reallocated DA assignment process. 如果网络上的加载指示需要DA时隙的重新分配,那么DA分配过程可重新分配处于DA分配状态的时隙。 If the load instruction needs to reallocate DA slot on the network, then the DA can be re-allocated slot allocation process in DA assignment status.

相反,分配SP时隙的过程具有优先级。 In contrast, SP slot allocation process has priority. 除了分配自由时隙之外,它可捕捉和重新分配已被DA分配或者在正被DA分配过程中的所有时隙。 In addition to the distribution of free time slots, which can capture and redistribution have been DA allocated or all the time slots being allocated in the DA process. 这样做是为了提供一种确保在一帧Nframe时隙内,至少单个SP时隙被分配给每个相邻节点的简单过程。 This is done to provide a way to ensure Nframe time slots in a frame, at least a single simple process SP time slot assigned to each neighbor node. 只有当链路丢失或者如果布局控制功能确定特定链路不应再在将与相邻节点建立的前N个链路的列表中时,SP分配时隙才被返回自由状态。 Only when the link is lost or if the layout of the control function to determine a list of specific link should not be established in the first N adjacent node when the link, SP slot allocation was only to return in a free state.

图5图解说明该过程是如何工作的。 Figure 5 illustrates how the process works. SP时隙分配过程在分配时隙方面具有极大的灵活性。 SP slot allocation process with a great deal of flexibility in the allocation of slots. 和DA过程相比,它可捕捉更多的时隙来分配,它可捕捉已被DA分配的或者在正被DA分配过程中的时隙。 DA process and compared it can capture more time slots allocated, which may have been captured or DA time slots allocated in the process of being allocated the DA. SP过程能够从布局控制功能和协议消息接收各种事件以便处理,包括布局变化事件。 SP process can control the layout of the function and protocol messages from a variety of events in order to receive treatment, including layout change events.

这样的事件可包括到近邻的链路的中断,新近邻的发现,从相邻节点收到SP分配请求消息,应发生布局变化,以便增加到某一近邻的链路,中断某一链路或者完成这两者的发现。 Such events may include links to interrupt neighborhood, new neighbors find that received from a neighboring node SP allocation request message, the layout changes should occur in order to link to a neighbor, or interruption of a link both complete discovery. 布局变化通知将携带描述需要发生的布局变化的数据。 Layout change notification will carry data describing the layout changes that need to occur.

如果事件描述链路的中断,那么必须采取的唯一动作是把时隙分配DB中的恰当时隙状态改为“自由”。 If the event described in the link is interrupted, then the only action is to be taken in the proper slot assignment DB slot status to "freedom." 如果要增加链路,那么该过程更复杂。 If you want to increase the link, then the process is more complicated. 这种情况下,SP时隙分配过程启动与新近邻的协议消息交换,并修改时隙分配DB。 In this case, SP slot allocation process initiation protocol message exchange with the new neighbors, and modify the slot assignment DB. 这最终导致这两个节点在关于分配给该链路的SP时隙的时隙分配方面达到一致。 This eventually led to the two nodes to reach agreement on the allocation of time slots allocated to that link aspects of SP slot. 只有单个SP时隙将被分配给与近邻的每条链路,以便简化协议。 Only a single SP time slot will be allocated to each link neighbors, in order to simplify the protocol. 下面说明该协议的其它细节。 The following description of the other details of the agreement.

分配DA时隙的过程遵循类似的程序。 DA slot assignment process follows similar procedures. DA时隙分配过程必须计算DA时隙需要,并将其与分配的时隙进行比较,以确定是否需要新的时隙再分配。 DA slot assignment process must calculate the DA time slot needs and compare with the slot allocation, to determine if a new time slot reallocation. 如果启动了DA时隙的再分配,那么它还将导致与相邻节点的一系列协议消息交换,以便对重新分配的时隙达成协议。 If you start the DA slot redistribution, then it will lead to a series of protocol messages exchanged with neighboring nodes to agree on the time slot reallocation. DA时隙分配过程可以只重新分配处于自由状态或者未被SP分配的时隙。 DA slot assignment process may reassign only time slots in the free state or not SP assigned. 下面将更详细地讨论协议细节和确定何时需要DA时隙再分配的过程。 Agreement will be discussed in detail and determine when the process of redistribution of DA slot detail below.

现在将讨论把半永久时隙分配给定向链路。 Will now be discussed to semi-permanent time slot allocated to a directional link. 在分配N个半永久时隙的方法的说明中,假定N是固定的,并且相对于网络规模和环境被智能选择。 In the description of assigning N semi-permanent time slot method, assuming N is fixed with respect to the network size and environment is a smart choice. 另外假定Nframe=2N-1。 Also assume Nframe = 2N-1. Nframe也可被设置成高于此的任意值,以便提供额外的按需时隙,如果认为这对特定网络和通信环境有用的话。 Nframe could also be set to any value higher than this, and to provide additional on-demand time slots if that this specific network and the communication environment is useful so.

布局控制功能提供几种重要的功能。 Layout controls provide several important functions. 近邻优先级(NP)列表由布局控制功能产生,并被用于指示时隙的分配的优选PA相邻节点。 Neighbor priority (NP) list generated by the layout control function, and is used in the preferred PA neighbor nodes indicating time slots allocated.

如果NP列表的长度为N或更小,那么布局控制功能将产生对SP时隙分配过程的布局改变事件,使其尝试得到对所有这些相邻节点的时隙分配。 If the length of the NP list is N or smaller, then the layout control function will generate the layout of the SP slot allocation process change event to try to get time slot assignments to all of these neighbor nodes. 如果NP列表的长度大于N,那么它将产生对SP时隙分配过程的布局改变事件,从而获得对NP列表上优先级最高的N个节点中的每一个的时隙分配。 If the length of the NP list is greater than N, then it generates the layout SP slot assignment process change event, so as to obtain the list of the highest priority NP N nodes in each of the slot allocation.

由于网络动态的缘故,NP列表不断变化。 As the network dynamic sake, NP list is constantly changing. 当PA链路中断时,该节点从NP列表中被除去,该链路的时隙随后被重新分配。 PA link is interrupted when the node is removed from the NP list, then slot the link is re-allocated. 这由布局控制功能启动,布局控制功能向SP时隙分配过程发送链路删除事件。 This control function is activated by the layout, layout control feature to send a link to delete events to the SP slot allocation process. 从而,SP时隙和分配给该链路的任何DA时隙变得适合于重新分配给PA列表上的另一节点。 Thus, SP time slot and any DA time slots allocated to that link become adapted to be reassigned to another node on the PA list.

在已知NP列表的当前状态的情况下,当时隙变得可用时的第一种选择是把时隙分配给额外的PA相邻节点,如果这是可能的话。 In the known current state of the NP list, the first option becomes available when the slot is the allocated slot to additional PA neighbor nodes if that is possible. 如果不能增加任何额外的相邻节点,那么可基于DA重新分配时隙。 If not add any additional adjacent nodes, then re-allocate time slots may be based on DA.

图6表示了SP时隙分配过程的状态图。 Figure 6 shows a state diagram of the SP slot assignment process. 为了管理协议消息处理,创建链路调度消息DB,如表3中所示。 In order to manage the protocol message processing, to create the link scheduling message DB, as shown in Table 3. 这保持从当下一SP消息到达以便处理时,要使用的在先协议交换所需的状态。 This keeps the SP from the moment a message arrives for processing, prior to use of the swap agreement needed to state. 空闲(idle)过程进行事件管理,因为在允许状态改变成其它状态之一之前,它检查收到的事件。 Idle (idle) incident management process, because the event before allowing the state to change to one of the other state, it checks received.

这些操作包括检查接收的消息,确定它们是否与DB的当前状态一致。 These operations include checking the received message, determine whether they are consistent with the current state of the DB. 如果消息与DB不一致,那么它被丢弃。 If the message is inconsistent with the DB, it is discarded. 某些超时可指示DB状态需要被重置。 Some overtime may indicate DB state needs to be reset. 该过程实现该功能。 The process to achieve this functionality.

表3 Table 3

存在如表4中列举的在SP时隙分配协议中所需的四种基本消息类型。 The existence of four basic message types as required in the SP slot assignment protocol listed in Table 4. 这些消息类型的使用是无需解释的,并且与前面的讨论一致。 Use of these message types are self-explanatory, and is consistent with the previous discussion.

表4 Table 4

图7中表示了SP时隙分配的一个例子。 Figure 7 shows an example of SP time slot assignment. 节点1和2都具有3个近邻,同时对每个链路表示了SP时隙分配。 Node 1 and 2 have three neighbors, also said the SP slot allocation for each link. 地是,它们可在它们之间增加额外的链路。 Manner, they can add an additional link between them. 链路调度协议将发现用于SP分配的可接受时隙。 Link scheduling protocol will find an acceptable time slot for the SP allocation. 表5中表示了对应的协议消息交换。 Table 5 shows the corresponding protocol message exchange.

节点1通过发送REQ_SPTS(L+4、5、6、7))和至少N个候选时隙的列表,启动交换。 Node 1 by sending REQ_SPTS (L + 4,5,6,7) list), and at least N candidate slots and start swapping. 该列表可包括全部的自由时隙和DA时隙。 The list may include all free and DA time slots time slots. 节点1正在把时隙1、2和3用于对其近邻的SP分配,从而其列表L包含其它时隙4、5、6和7。 Node 1, 2 and 3 are the slot for the SP allocation of its neighbors, so its list L contains the other time slots 4,5,6 and 7. 当请求消息被发送时,对时隙和链路调度消息数据结构进行恰当的改变。 When a request message is transmitted, to the time slot and link scheduling message data structures appropriate changes. 节点2正在把时隙4、5和6用作到其3个近邻的链路的SP分配,从而它选择时隙7作为将用于新链路的唯一时隙。 Node 2 is the slot 5 and 6 are used to allocate SP link its three neighbors, so it chose the time slot 7 as the only time slot will be used for new links. 它在答复消息中发送该选择。 It sends a reply message to the selection.

当答复消息被发送时,也对时隙和链路调度消息数据结构进行恰当的改变。 When the reply message is sent, but also on the time slot and link scheduling message data structures appropriate changes. 最后,当确认被发送或收到时,恰当时隙的状态被改成“SP分配给链路(1,2)。”另外注意如果节点1和2已选择4个相邻节点,那么它们仍然能够找到借助其在它们之间建立链路的公共时隙,如果它们与它们的至少两个近邻使用相同的时隙。 Finally, when the acknowledgment is sent or received, the state is changed to the appropriate slot "SP allocated to link (1,2)." Also note that if nodes 1 and 2 have chosen four adjacent nodes, so they are still establishing a link can be found by means of their common slot between them, if they are at least two neighbors with the same time slot.

表5 Table 5

开发了描述图6中所需的过程的一些初始伪代码。 FIG developed some initial pseudocode describing the processes required 6. 存在可能发生的必须由SP时隙分配过程34处理的各种事件。 Events must process 34 by the SP slot assignment process may occur in the presence of. 如表6中所示在空闲过程中进行事件管理。 As shown in Table 6 for event management in the idle process. 表示了四类事件:接收的消息,检查超时,来自布局控制的链路增加通知,和链路故障或链路删除。 Shows four types of events: the received message, check timeouts, link from the layout control increases notification, and link failure or link deletion.

首先相对于链路调度消息DB检查接收的消息,以确保该消息与DB的当前状态一致。 First, with respect to the link scheduling message DB checks the received messages to ensure that the message is consistent with the current state of the DB. 例如,如果我们向某一近邻发送请求,那么预期的下一消息是答复。 For example, if we send a request to a neighbor, then the next message is expected to reply. 为了简化该分布式协议,每次只允许SP协议消息交换的一个线程。 To simplify this distributed protocol, allowing only SP protocol message exchanges a thread. 通过在启动链路增加转变(transition)之前或者在处理REQ SPTS消息之前,检查DB以了解是否正在进行其它SP消息交换,在程序中强制实现这一点。 Before starting the chain by increasing the transition (transition) or before processing REQ SPTS message, check DB to see if other SP message exchange in progress in the program enforces this.

如果由于另一SP协议线程目前正在进行中,不能启动链路增加,那么通过放弃和对预期其它进程完成时的稍后时间重新调度,链路增加将被延期。 If another SP protocol thread is currently in progress, the link does not start to increase, then give up and later by other processes on the expected completion of the re-scheduling, link the increase will be postponed. 允许进行多次尝试,以处理试图同时增加链路的几个节点之间的潜在冲突。 Allow multiple attempts to deal with at the same time trying to increase the potential conflict between several nodes link. 这并不意味着处理不可靠RF链路的问题。 This does not mean handling unreliable RF link problem. 应通过使用和使用ARQ和重发来恢复丢失/出错的消息的开销通道的链路协议,来解决后一问题。 By using and use should ARQ and retransmission to recover lost / error messages link protocol overhead channel to solve the latter problem.

从而,分布式调度协议可假定消息将不会被丢失。 Thus, the distributed scheduling protocol can assume that messages will not be lost. 这为协议的简化创造了条件。 This creates the conditions for the simplified protocol. 当布局控制从NP列表中选择某一相邻节点,以便连接为新邻居时,它发出布局改变(链路增加)事件,布局改变事件(在空闲进程中的一致性检查之后)导致转变到SP时隙分配进程中的链路增加状态。 (After the idle process consistency check) when the control to select a layout from the NP list of neighboring nodes to connect to the new neighbors, it sends the layout change (increase link) events, the layout change events leading to transition to SP slot allocation process to increase the link status.

表6 Table 6

表7中表示了链路增加进程的伪代码。 Table 7 shows the process of increasing link pseudo-code. 这启动只需要这两个相邻节点之间的SP时隙分配和协议消息交换的协调的进程。 This boot requires only two SP time slot assignment and protocol message exchange between the neighboring nodes coordination process. 请求链路的节点把REQ_SPTS消息连同该链路的可接受时隙的列表一起发送给候选相邻节点。 Requesting node links to REQ_SPTS message together with the list of acceptable time slots for the link is sent along to the candidate neighbor node.

候选时隙的列表必须包含至少N个时隙,所述至少N个时隙包括至少一个半永久时隙SP。 The list of candidate time slots must contain at least N time slots, said at least N time slots including at least one semi-permanent time slot SP. 该列表还可包括可能的所有N-1个可用DA时隙。 The list may further comprise all possible N-1 available DA time slots. 可用或按需时隙目前可被暂时分配用于按需通信。 Demand slots currently available or can be temporarily allocated for on-demand communications. 该列表将按照优先级排序,以指示在当前的可用时隙分配中产生最少混乱的时隙优先选择。 The list will be sorted by priority, to indicate that minimal confusion slots available in the current slot allocation priority selection. 换句话说,正被使用的符号是时隙不是SP时隙,除非已被分配给通信链路。 In other words, a positive sign is the time slot is not being used SP time slot unless already allocated to a communication link. 2N-1个时隙中的任意一个可以是SP时隙。 2N-1 time slots may be any one SP time slot. 从而,发送的N个时隙的列表或者是自由时隙或者是可用的DA时隙。 Thus, a list of N time slots sent either free time slots or DA time slots are available. 这些可以是N-1个SP时隙,不过它们已被分配,不在该列表上。 These can be of N-1 SP time slots, but they have been assigned, is not on the list.

考虑到不可靠的链路和与可能正在同时进行的其它分配的冲突,REQ_SPTS消息可被发送多达MAX_TRIES次。 Considering the unreliable links and conflicts with other distribution may be carried out simultaneously in, REQ_SPTS message can be sent up to MAX_TRIES times. 如果不存在响应REQ_SPTS消息,来自相邻节点的任何REPLY_SPTS消息,那么链路调度消息DB中的超时触发重试。 If the response message does not exist REQ_SPTS any REPLY_SPTS news from a neighbor, then the link scheduling message DB triggers retry timeout. 一旦REQ_SPTS消息被发送,进程返回空闲状态,在空闲状态下,其它事件可被处理。 Once REQ_SPTS message is sent, the process returns to the idle state, idle state, other events can be processed.

表7 Table 7

接收REQ_SPTS消息的近邻将使其SP时隙分配进程转变成处理REQ_SPTS状态。 Receive messages neighbors REQ_SPTS SP slot allocation process will make it into the processing REQ_SPTS state. 处理该消息的程序示于表8中。 Handles the message is shown in Table 8. 该程序获得提供的时隙列表Ls,并选择其优选的时隙Ns。 The procedure for obtaining a list Ls slots provided, and select its preferred time slot Ns.

如果到相邻节点的链路的数目Num_links小于极限N,那么该程序从该列表中选择它更喜欢的时隙。 If the number of neighboring nodes Num_links to link smaller than the limit N, then the program selects the time slot it prefers from this list. 随后发送REPLY_SPTS答复消息和该选择。 REPLY_SPTS reply message and then sends the selection. 如果该链路不能被接受或者如果有另一SP时隙分配正在进行,那么发送否定的REPLY_SPTS答复消息。 If the link is not acceptable or if there is another ongoing SP slot allocation, then send a reply message REPLY_SPTS negative.

选择的时隙将选自其N个可用时隙之一或者其自由时隙之一。 The selected time slot will be selected from one of its N available time slots or one of its free time slots. 可用时隙或者是“自由”时隙或者是可用的DA时隙。 Available time slots or "free" time slot or slots available DA. 如果我们能够增加另一链路,那么将至少存在N个可用时隙。 If we can add another link, then there will be at least N available time slots. 每个节点始终管理其时隙,以致存在可供分配为半永久时隙的N个时隙(N个相邻节点中的每个节点一个半永久时隙,如果存在这么多的相邻节点)。 Each node always manages its time slots so that there is a semi-permanent time slot allocated for the N slots (N adjacent nodes in a semi-permanent time slot for each node, if so many neighboring node exists). 如果它接受该链路,那么它将最多具有N-1个其它相邻节点,每个节点分配一个半永久时隙。 If it accepts the link, then it will have a maximum of N-1 other neighbor nodes, assigning each node a semi-permanent time slot. 该程序还对链路调度消息DB和时隙分配DB中的状态进行恰当的修改。 The program also the link scheduling message DB and the slot assignment DB in a state appropriate modifications.

表8 Table 8

接收的REPLY_SPTS消息被处理,如表9中所示。 REPLY_SPTS message received is processed, as shown in Table 9. 从消息中抽取从相邻节点收到的时隙的选择Ns。 Extracted received from a neighboring node slots choice Ns from the message. 我们还需要节点用指示它将同意使用所分配时隙的肯定或否定CONFIRM消息来确认该答复。 We will also need to agree with the indicated node using the assigned time slots positive or negative CONFIRM message to confirm the answer. 这种三方握手消除了调度进程的结果方面的不确定性。 This three-way handshake scheduling process eliminates the uncertainty of outcomes.

如果REPLY SPTS消息是肯定的答复,那么检查时隙Ns的选择,了解它是否仍然是允许的用于新链路的新SP时隙的分配。 If REPLY SPTS message is a positive response, then check the slot Ns choice, to see if it is still allocated to allow for the new link the new SP slots. 如果它是允许的,那么对时隙分配和链路调度消息数据库中的状态进行恰当修改。 If it is allowed, then the slot assignment and link scheduling message database in a state appropriate modifications. 随后返回肯定的CONFIRM消息。 Then returned to the affirmative CONFIRM message.

如果接收的REPLY_SPTS消息是否定的答复,那么对该Nbr_ID重置时隙分配和链路调度消息数据库。 If REPLY_SPTS message received is negative answer, then resets the Nbr_ID slot assignment and link scheduling message database. 另外,如果Ns的选择不再是许可的,那么对该Nbr_ID,链路调度消息数据库被复位。 In addition, if the choice of Ns is no longer permitted, then the Nbr_ID, link scheduling message database is reset. 随后向相邻节点发送拒绝该链路的否定的CONFIRM消息。 Then refuse to adjacent nodes send the link negative CONFIRM message.

表9 Table 9

表10表示处理CONFIRM消息的程序。 Table 10 shows the processing CONFIRM message procedures. 如果CONFIRM是肯定的,那么认为该链路将被加入该组近邻中。 If CONFIRM is yes, then consider that the link will be added to the group's immediate neighborhood. 节点的链路的数目Num_links被递增。 Number Num_links node link is incremented. 在时隙分配DB中,分配的时隙Ns被标记成SP_Alloc,对于索引Nbr_ID,重置链路调度消息DB中的链路消息状态。 In the slot allocation DB, allocated time slot Ns is marked SP_Alloc, for indexing Nbr_ID, resets the link scheduling message DB link in the message states. 如果消息是否定的CONFIRM,那么对于该Nbr_ID,重置时隙分配和链路调度消息数据库。 If the message is negative CONFIRM, then for the Nbr_ID, reset the slot assignment and link scheduling message database.

表10 Table 10

由于几个原因之一,分配的时隙可能需要被解除分配。 For one of several reasons, slot allocation may need to be deallocated. 如果在正常操作的过程中,链路中断或者变得不可靠,那么布局控制功能被卷入以解决不可靠链路问题。 If during normal operation, the link is down or becomes unreliable, then the layout control functions are involved to resolve the unreliable link problem. 最终,它可产生指导时隙分配进程删除分配给该链路的所有时隙的布局改变(例如,链路删除)事件。 Ultimately, it may generate guide slot assignment process to delete the layout assigned to the link to change all the time slots (e.g., link deletion) event.

表11中表示了该程序中所涉及的步骤。 Table 11 shows the steps involved in the program. 通过从请求与其它节点共用的所有时隙的解除分配的节点发送DELETE_TS消息,该链路被解除分配。 By sending a DELETE_TS message from the requesting slots shared with all other nodes to deallocate a node, the link is de-allocated. 另外,链路调度消息DB和时隙分配DB中的恰当条目被重置。 In addition, the link scheduling message DB and the slot assignment DB are reset in the proper entry.

表11 Table 11

表12表示了处理收到的DELETE_TS消息的程序。 Table 12 shows a program processing received DELETE_TS message. 从该消息抽取解除分配的时隙的列表Ls。 Extracting from the message list of deallocated time slots Ls. 随后时隙分配DB和链路调度消息DB中的恰当状态被重置。 Then slot assignment and link scheduling message DB DB in the appropriate state is reset.

表12 Table 12

总之,分配半永久时隙的功能的目的是连接尽可能多达N个的相邻节点。 In short, the semi-permanent time slot allocation function is to connect as far as possible up to N neighbor nodes. 如果获得N个相邻节点,那么每个相邻节点被分配单个半永久时隙。 If the obtained N adjacent nodes, then each neighboring node is assigned a single semi-permanent time slot. 一旦借助该协议建立了新的链路,那么这两个节点将在新分配的SP时隙中开始工作。 With this agreement, once established a new link, then both nodes will begin work in the newly allocated SP slot.

该操作将测试所述新链路,以确定利用分配的时隙是否能够保持可靠的通信。 This operation will test the new link to determine if the use of the allocated time slots to maintain reliable communication. 这确保不存在于该特定时隙中发生的异常干扰。 This ensures that there is no abnormality in the interference in a particular time slot. 如果该链路被测试为不可靠,那么将告知布局控制功能,以致该时隙可被解除分配并被用于其它用途。 If the link is tested as unreliable, then tells the layout control function, so that the time slot can be deallocated and used for other purposes.

下面将讨论可用(按需)时隙的分配。 The following will discuss the allocation of available (on-demand) slots. 将按照响应不断波动的网络通信量的需求的方式,分配可用时隙。 Will respond according to the fluctuating demands of network traffic way, allocation of available time slots. 同样,假定N被固定,并且相对于网络规模和环境被智能选择。 Similarly, assume that N is fixed, and with respect to the network size and environment is a smart choice. 另外,假定Nframe=2N-1。 Further, assume that Nframe = 2N-1.

为了使可用容量的分配的粒度更细,时隙将被分成ms个子时隙。 In order to make the allocation of available capacity finer granularity, slots will be divided into sub-ms time slot. 对于下述讨论的剩余部分,假定ms=2。 For the remainder of the following discussion, it is assumed ms = 2. 这将通过把子时隙定义为每第ms(或者第二)帧重复的特定时隙分配来实现。 This will be achieved by the definition of sub-timeslots for each section ms (or second) frame repetition of a particular time slot assignment.

只有当为一个节点和一个相邻节点之间的链路分配至少一个半永久时隙时,才允许从所述一个节点到所述一个相邻节点的对可用时隙的请求。 Only when the link is allocated at least one semi-permanent time slots between the adjacent node and a node, one node is allowed from the requesting node to said one of the available slots adjacent. 在链路被分配至少一个半永久时隙之后,每第ms(或者第二)帧,节点可请求单一时隙的周期分配。 After the link is allocated at least one semi-permanent time slot, each of ms (or second) frame, a node may request a periodic allocation of a single time slot. 用于调度可用时隙的消息可通过PA链路发送,以便在需要时隙之前几帧,调度时隙,因为该链路具有每帧至少一个半永久时隙的分配。 Available time slots for scheduling messages can be sent via PA link, so take a few frames before slot scheduling slot, because the link has allocated at least one semi-permanent time slots per frame.

有效分配可用时隙的关键要求是每个链路上通信量要求的测量。 The key requirements for efficient allocation of available time slots is the measurement of the traffic requirements of each link. 将需要两个量度。 Would require two metrics. 首先,测得的通过链路(i,k)发送的平均通信量(用每帧的时隙的数目为单位)将由Tikse表示。 First, measured through the link (i, k) of the average traffic sent (with the number of time slots per frame as a unit) will be expressed Tikse. 该量度将包括通过每帧的一个或多个半永久时隙以及任何可用时隙发送的所有通信量。 This measure will include all traffic sent over one or more semi-permanent time slots per frame as well as any available time slots.

另外,我们还需要保持链路(i,k)的队列状态的当前量度Qik,Qik的值越大,表示需要一个或多个可用时隙的立即分配。 In addition, we also need to keep a link (i, k) of the current queue status measure Qik, the greater the value of Qik, and expressed the need for immediate distribution to one or more of the available slots. 需求的偶然爆发可产生Qik的增大,这随后应触发对按需容易的额外时隙的请求,直到队列大小减小为止。 Needs occasional outbreaks can produce increased Qik, which should then trigger a request for additional slots easily demand until the queue size is reduced so far.

在链路(i,k)上分配的时隙的总数(在ms=2的情况下,量化为时隙的一半)将用Niktot表示。 The total number of link (i, k) on the allocation of time slots (in ms = 2, the quantization half slot) will be denoted by Niktot. 时隙需求被定义成如下所示:Tikdem=f(Tikse,Qik),...(4)]]>它是测量的通信量加上由队列大小指示的所需的估计附加容量的函数。 Time slot demand is defined as follows: Tikdem = f (Tikse, Qik), ... (4)]]> It was measured traffic plus the size of the queue indicated by the estimated additional capacity needed function. 从而,该链路上所需的时隙的数目Tikneed如下所示:Tikneed=max(Tikdem,Tkidem)...(5)]]>分配给该链路的度量如下所示:MikDA=Tikneed-Niktot+B,...(6)]]> Thus, the required number of time slots on the link Tikneed as follows: Tikneed = max (Tikdem, Tkidem) ... (5)]]> metric assigned to this link is as follows: MikDA = Tikneed- Niktot + B, ... (6)]]>

它是应通过DA时隙分配机制分配给该链路的额外时隙的估计数目的量度。 It is estimated that an additional number of slots to be allocated to the link through the DA slot allocation mechanism metrics. B是可能名义上被设置成约为时隙的1/4-1/2,以便向每个链路分配足够的过量容量,从而避免大排队的偏项。 B is may nominally be set to about slot 1 / 4-1 / 2, so that a sufficient excess capacity to each link assignment, thereby avoiding a large queued bias term. 虽然我们正在举例说明使用在(4)中定义的度量的方法,不过各种其它形式的度量也可被用作分配DA时隙的基础。 While we are illustrating the use of the (4) The method as defined in metric, but a variety of other forms of metric could also be used as a basis for the allocation of DA time slots.

图8表示了DA时隙分配进程36的状态图。 Figure 8 shows the DA slot assignment process 36. The state of FIG. 状态图和协议交换与SP时隙进程的状态图和协议交换类似。 State diagram and the protocol exchange with the SP slot processes and protocols to exchange similar state diagram. 为了简化协议消息处理,在任何时候只能进行DA时隙分配的单一线程。 In order to simplify the protocol message processing, at any time only a single thread of DA time slot allocation. 空闲的进程进行事件管理,因为在允许状态改变到其它状态之一之前,它检查收到的状态。 Idle event management processes, as in the state prior to allowing a state change to one of the other state, it checks received.

这些操作包括下述内容。 These operations include the following. 检查接收的消息,确定它们是否与DB的当前状态一致。 Check the received message, determine whether they are consistent with the current state of the DB. 如果消息与DB不一致,那么它被丢弃。 If the message is inconsistent with the DB, it is discarded. 某些超时可指示DB状态需要被重置。 Some overtime may indicate DB state needs to be reset. 该进程实现该功能。 The process to achieve this function. 它还确定在已知节点的通信负载需求的情况下,DA时隙分配是否最佳。 It is also determined in the case where the communication load requirements of known nodes, DA slot assignment is optimal. 如果它确定新的DA时隙是否必须被加入到特定的链路中,那么它可导致转换变增加DA时隙状态。 If it is determined whether a new DA time slot must be added to a particular link, then it may lead to an increase DA slot state variable converted.

在DA时隙分配协议中需要四种基本消息类型,如下表13中所示。 In the DA time slot assignment protocol requires four basic message types, as shown in Table 13. 它们与在SP时隙分配中使用的消息类型非常类似。 They message types used in the SP slot allocation is very similar. 这些消息类型的使用是自明的,并且与SP时隙分配进程的在先说明一致。 Use of these message types are self-evident, and the SP slot allocation process described earlier agreement.

表13 Table 13

图9中表示了DA时隙分配的一个例子。 9 shows an example of DA time slot allocation. 节点1希望为其链路(1,2)增加额外的DA时隙分配。 Node 1 wants its link (1,2) additional DA time slot allocation. 对应的协议消息交换示于表5中。 Corresponding protocol message exchange is shown in Table 5 below. 节点1通过发送指示它能够支持所有时隙5和6及子时隙4.2的分配的REQ_DATS(L=(4.2,5,6)),启动所述交换。 By sending node 1 indicates that it can support all the slots 5 and 6, and sub-slot allocation of 4.2 REQ_DATS (L = (4.2,5,6)), activating the exchange. 该列表可包括所有自由时隙和DA时隙,后者较少需要。 The list may include all free and DA time slots time slots, which is less required.

当请求消息被发送时,对时隙和链路调度消息数据结构进行恰当的改变。 When a request message is transmitted, to the time slot and link scheduling message data structures appropriate changes. 节点2正在把时隙1、3和6用作其到其3个近邻的链路的SP分配,把子时隙2.1和3.2用作DA分配。 Node 2 is the slot 1, 3 and 6 as its three neighbors to its links SP assignment, handle 2.1 and 3.2 as a DA slot allocation. 它可选择子时隙4.2或者时隙5的两个子时隙。 It can select two sub-sub-slot 4.2 slots or slot 5. 它选择并在答复消息中发送该选择。 It selects and sends a reply message to the selection.

当发送答复消息时,对时隙和链路调度消息数据结构进行恰当改变。 When sending a reply message, for the time slot and link scheduling message data structures appropriate change. 最后,当发送或收到确认时,恰当时隙的状态被改变成“子时隙4.2被DA分配给链路(1,2)。”表14 Finally, when the transmission or acknowledgment is received, the state of the appropriate time slots are changed to "sub-slot 4.2 DA allocated to link to be (1,2)." Table 14

在每个网络节点使用下述方法为到相邻节点的定向链路分配(N-1)个可用时隙。 (N-1) available time slots in a directional link allocated to each network node using the following method to the adjacent nodes. 利用这些措施,每个节点将连续保持其被分配半永久时隙的每个链路的链路度量MikDA。 Using these measures each node will continuously maintain its semi-permanent time slot is allocated link each link metric MikDA. 每个节点将使用该链路度量来指示对到每个相邻节点的额外传输时隙的需要。 Each node will use this link metric to indicate the additional transmission time slots to each neighbor node needs. MikDA的最大值指示最需要额外的按需时隙分配的链路。 Maximum indication MikDA most require additional demand link slot allocation. MikDA的正值指示所需的额外时隙的数目,MikDA的负值指示可被放弃以供重新分配的时隙的数目。 A positive value indicates the number of additional time slots required for MikDA, the negative value indicates MikDA may be abandoned for the number of re-allocation of time slots.

作为度量,MikDA被保持,如果最大的链路度量指示需要额外的子时隙分配,并且如果存在可用作自由时隙或者用作对其它链路的过量DA分配(同样由较小度量指示)的子时隙,那么该进程转变到增加DA时隙状态,启动找出DA子时隙分配的进程。 As a metric, MikDA is maintained, if the largest link metric indicates the need for an additional sub-slot allocation and if there can be used as free slots or as excess DA allocation to other links (again by a small metric indication) sub-slot, then the process transitions to increase DA slot state, start the process to find DA sub-slot allocation.

如同半永久时隙一样,节点只需要与近邻协调将为到该近邻的定向链路分配的DA时隙的选择。 As a semi-permanent time slot as coordination with neighboring nodes only need to directional link will assign the neighborhood of DA slots to choose from. 这意味着近邻将通过定向链路,向该近邻发送时隙分配请求,并通过相同的链路接收分配的批准或者请求的拒绝。 This means that the neighbor will, transmitted to neighbors via a directional link time slot assignment request, and receiving an allocation request for approval or refusal by the same link.

开发了描述图8中所需进程的一些初级伪代码。 Develop a description of some of the primary pseudo-code in Figure 8 is required for the process. 存在可能发生的必须上DA时隙分配进程处理的各种事件。 May exist on a variety of events must occur DA slot allocation process handled. 如表6中所示在空闲进程中进行事件管理。 As shown in Table 6 for event management in the idle process.

表示了四种类型的事件:1)接收的消息,2)检查超时,3)链路度量的重新计算和4)DA时隙需要和DA时隙删除。 Represents the four types of events: 1) received message, 2) check timeouts, 3) recalculation of link metrics, and 4) DA time slot needs and DA time slot deletion. 首先相对于链路调度消息DB检查接收的消息,以确保该消息与DB的当前状态一致。 First, with respect to the link scheduling message DB checks the received messages to ensure that the message is consistent with the current state of the DB. 例如,如果我们向某一近邻发送请求,那么预期的下一消息是答复。 For example, if we send a request to a neighbor, then the next message is expected to reply.

为了简化该分布式协议,每次只允许SP协议消息交换的一个线程。 To simplify this distributed protocol, allowing only SP protocol message exchanges a thread. 通过在启动链路增加转变(transition)之前或者在处理REQ_DATS消息之前,检查DB以了解是否正在进行其它SP消息交换,在程序中强制实现这一点。 Before starting the chain by increasing the transition (transition) or before processing REQ_DATS message, check DB to see if other SP message exchange in progress in the program enforces this. 如果由于另一SP协议线程目前正在进行中,不能启动额外的时隙,那么所述额外时隙将不被进行。 If another SP protocol thread is currently in progress, you can not start the extra slot, then the additional slots will not be carried out.

在重新计算链路度量和DA时隙需要的下一机会,自然地可重新调度额外时隙。 In recalculated link metrics and DA time slots needed next opportunity, naturally re-scheduling extra slots. 将根据预先设置的时间表,定期重新计算链路度量。 According to a preset schedule, periodically recalculated link metric. 链路度量大于某一阈值Mzx_metric_threshold的链路是获得新的DA子时隙的候选者。 Link metric greater than a certain threshold Mzx_metric_threshold link is to obtain a new DA sub-slot candidates.

具有超过该阈值的最大度量的链路可被选为将被分配新的DA子时隙的下一链路。 The link with the largest metric exceeds the threshold value may be selected to be assigned a new DA sub-slot of the next link. 当新的DA子时隙需要被分配,并且如果它满足上述条件,那么在DA时隙分配进程中发生到增加DA时隙状态的转变。 When a new DA sub-slot needs to be allocated and if it satisfies the above condition, then the transition occurs to increase DA DA slot state in the slot assignment process.

表15 Table 15

表16中表示了额外DA时隙进程的伪代码。 Table 16 shows the pseudo-code additional DA slot process. 这启动只要求这两个相邻节点之间的时隙分配和协议消息交换的协调的进程。 This starts the process only requires a coordinated time slot assignment and protocol message exchanges between two adjacent nodes. 请求链路的节点把REQ_DATS和所述链路的可接受时隙的列表一起发送给候选相邻节点。 Requesting node and the link to the link list REQ_DATS acceptable time slot is transmitted together to the candidate neighbor node.

候选时隙的列表必须包含所有自由子时隙和具有低于某一阈值Min_metric_threshold的度量的所有DA子时隙。 The list of candidate time slots must contain all free sub-slots and all DA sub-slots having less than a certain threshold Min_metric_threshold metric. DA时隙可能被同时暂时分配给其它DA通信。 DA time slots may be simultaneously temporarily assigned to other DA traffic. 该列表将按照优先级排序,以指示在当前的按需时隙分配中导致最少混乱的子时隙优先选择。 The list ordered by priority, to indicate the sub-slots leads to a minimum of confusion preference in the current on-demand time slot allocation. 优选权排序将是首先是自由时隙,随后是从具有最小度量的子时隙,一直到具有小于Min_metric_threshold的最大度量的子时隙。 Preferred weight for the first free slot will be followed from the sub-slot has the minimum metric, until the sub-slot has the largest metric less than Min_metric_threshold.

为了简化该分布式协议,每次只允许DA协议消息交换的一个线程。 To simplify this distributed protocol, allowing only DA protocol message exchanges a thread. 这在空闲程序中被强制实现。 This is enforced in the idle procedure. REQ_DATS消息只被发送一次,但是如果相邻节点目前正在处理另一DA协议交换,那么它可能不成功。 REQ_DATS message is sent only once, but if the adjacent node is currently processing another DA protocol exchange, then it may not succeed. 这种情况下,节点最终将收到否定的REPLY_DATS消息。 In this case, the node will eventually receive a negative REPLY_DATS message. 如果下一次评估链路度量时,该链路具有最大的度量,那么这种情况下将再次进行增加DA时隙的尝试。 If the next assessment link metric, the link with the largest metric, then in this case will again attempt to increase the DA slots. 一旦REQ_DATS消息被发送,那么进程就返回空闲状态,在空闲状态下,其它事件可被处理。 Once REQ_DATS message is sent, then the process will return to the idle state, idle state, other events can be processed.

表16 Table 16

接收REQ_DATS的近邻将使其DA时隙分配进程转变到REQ_SPTS状态。 Receive REQ_DATS neighbors DA slot allocation process will make the transition to REQ_SPTS state. 表17中表示了处理该消息的程序。 Table 17 shows the procedure to process the message. 该程序获得提供的子时隙的列表Ls,并选择其优选的子时隙Ns。 The procedure for obtaining a list of sub-slots provided Ls, and selects its preferred sub-slot Ns. 接受的子时隙是列表Ls上或者在时隙分配DB中被标记为自由的,或者被DA分配的具有小于Min_metric_threshold的链路度量的第一子时隙。 The sub-slot accepted is the list Ls in the slot assignment DB or is marked as free, DA allocated or are in the link metric less than Min_metric_threshold first sub-slots. 随后连同该选择发送REPLY_DATS答复消息。 Then, along with the selection REPLY_DATS send a reply message. 如果该链路不能被接受或者如果有另一DA时隙分配正在进行中,那么发送否定的REPLY_DATS答复消息。 If the link is not acceptable or if there is another ongoing DA slot allocation, then send a reply message REPLY_DATS negative. 该程序还对链路调度消息DB和时隙分配DB中的状态进行恰当修改。 The program also link scheduling message DB and the slot assignment DB in the appropriate state changes.

表17 Table 17

接收的REPLY_DATS消息如表18中所示被处理。 REPLY_DATS message received as shown in Table 18 are processed. 从该消息中抽取从相邻节点接收的子时隙Ns的选择。 Receiving from a neighbor node extracted sub-slot choice Ns from the message. 我们要求该节点用指示它将同意使用所分配时隙的肯定或否定CONFIRM消息确认该答复。 We request that the node with the instructions will agree to use the allocated time slots positive or negative CONFIRM message to confirm the answer. 如SP分配进程中所示,这种三方握手消除了调度进程的结果方面的不确定性。 SP allocation process as shown in the results of this three-way handshake eliminates scheduling process uncertainties.

如果REPLY_DATS消息是肯定的答复,那么检查子时隙Ns的选择,了解它是否仍然是允许的用于新链路的新DA子时隙的分配。 If the answer is yes REPLY_DATS news, check subslot Ns choice, to see if it is still allocated to allow for the new link the new DA sub-slots. 如果它是允许的,那么对时隙分配和链路调度消息数据库中的状态进行恰当修改。 If it is allowed, then the slot assignment and link scheduling message database in a state appropriate modifications. 随后返回肯定的CONFIRM消息。 Then returned to the affirmative CONFIRM message.

如果接收的REPLY_DATS消息是否定的答复,那么对该Nbr_ID重置时隙分配和链路调度消息数据库。 If REPLY_DATS message received is negative answer, then resets the Nbr_ID slot assignment and link scheduling message database. 另外,如果Ns的选择不再是许可的,那么对该Nbr_ID,重置链路调度消息数据库。 In addition, if the choice of Ns is no longer permitted, then the Nbr_ID, reset the link scheduling message database. 随后向相邻节点发送拒绝该链路的否定的CONFIRM消息。 Then refuse to adjacent nodes send the link negative CONFIRM message.

表18 Table 18

表19表示处理CONFIRM消息的程序。 Table 19 shows a program processing CONFIRM messages. 如果CONFIRM是肯定的,那么选择的子时隙将被加入到对到Nbr_ID的链路的分配中。 If the CONFIRM is positive, then the choice of sub-slot to be added to the allocation of the link to Nbr_ID in. 分配的时隙Ns在时隙分配DB中被标记成DA_Alloc,对于索引Nbr_ID,重置链路调度消息DB中的链路消息状态。 Ns allocated time slot in the time slot assignment DB are marked DA_Alloc, for index Nbr_ID, resets the link scheduling message DB in the link message state. 如果消息是否定的CONFIRM,那么对该子时隙,重置时隙分配和链路调度消息数据库。 If the message is negative CONFIRM, then the sub-slot, reset the slot allocation and link scheduling message database.

表19 Table 19

由于几个原因之一,分配的时隙可能需要被解除分配。 For one of several reasons, slot allocation may need to be deallocated. 如果在正常操作的过程中,链路中断或者变得不可靠,那么布局控制功能被卷入以解决不可靠链路问题。 If during normal operation, the link is down or becomes unreliable, then the layout control functions are involved to resolve the unreliable link problem. 最终,它可产生指导时隙分配进程删除分配给该链路的所有时隙的布局改变(例如,链路删除)事件。 Ultimately, it may generate guide slot assignment process to delete the layout assigned to the link to change all the time slots (e.g., link deletion) event.

表11中表示了该程序中所涉及的步骤。 Table 11 shows the steps involved in the program. 通过从请求与其它节点共用的所有时隙的解除分配的节点发送DELETE_TS消息,该链路被解除分配。 By sending a DELETE_TS message from the requesting slots shared with all other nodes to deallocate a node, the link is de-allocated. 另外,链路调度消息DB和时隙分配DB中的恰当条目被重置。 In addition, the link scheduling message DB and the slot assignment DB are reset in the proper entry.

表20 Table 20

表21表示处理收到的DELETE_TS消息的程序。 Table 21 shows the processing procedure DELETE_TS messages received. 从该消息抽取将被解除分配的子时隙Ls。 Extraction will be deallocated subslot Ls from the message. 随后时隙分配DB和链路调度消息DB中的恰当状态被重置。 Then slot assignment and link scheduling message DB DB in the appropriate state is reset.

表21 Table 21

链路调度算法还适用于由相控阵天线16产生的多个同时波束。 Link scheduling algorithm is also applicable to the plurality of phased array antenna 16 by the generated beam simultaneously. 设想对其中每个节点采用多个天线波束和独立的接收器,例如多波束相控阵天线(或者其它类型的多重定向天线)的系统的扩展。 Vision of where each node using multiple antenna beams and independent receivers, such as the expansion of multi-beam phased array antenna (or other types of multi-redirect antenna) systems. 此外,假定所有节点并不都必须具有相同数目的波束,即,节点k具有Bk个波束。 Furthermore, assume that all nodes do not all have to have the same number of beams, i.e., node k has Bk beams. 这等同于在任意时隙可能的Bk个并行链路。 This is equivalent to Bk parallel links possible at any time slot.

我们正在扩展前面的讨论(前面的讨论设想单一的指向波束),以允许Bk个波束在大于Bk的一组相邻节点之间被分时。 We are extending the previous discussion (previous discussion envisaged single point beam) to allow the Bk beams between Bk is larger than a set of adjacent nodes are sharing. 即使节点均具有不同数目的波束,所有节点必须使用共同的时隙格式和对于每个波束来说,每帧的时隙的数目等于Nframe的帧。 Even if the nodes have different numbers of beams, all nodes must use a common time slot format and for each beam, the number of time slots per frame is equal to Nframe frame.

把在任意节点k,其Bk波束任意之一上的半永久(SP)分配时隙的数目的上限(于是,每个波束许可的相邻节点的最大数目)看做由Nbeam表示。 In any of the node k, the upper limit (thus, the maximum number of adjacent nodes of each beam permission) seen as semi-permanent (SP) time slots assigned to any number on one of its Bk beams denoted by Nbeam. Nbeam的值只取决于每帧的时隙的数目,而不取决于波束的数目。 Value depends only on the number of Nbeam time slots per frame, without depending on the number of beams. 如(3)中一样,我们将规定Nbeam必须满足下述等式:Nframe≥2·Nbeam-1 (7)假定网络中的所有节点由定向链路连接,其中节点k具有通过跳时实现波速共用,并且指向其相邻节点的Bk个波束。 Such as (3), we will provide Nbeam must satisfy the following equation: Nframe≥2 · Nbeam-1 (7) assuming that all nodes in the network are connected by directional links, where node k has realized through common jumping velocity and point Bk beams of its neighbor nodes. 此外,假定每个波束允许的近邻的数目等于Nbeam,每个波束允许的半永久时隙的许可数目的固定极限(在每个近邻被分配一个SP时隙的情况下)。 Further, assume the number of neighbors allowed per beam is equal to Nbeam, (in the case of each neighbor is assigned a time slot SP) for each license number of beams allowed semi-permanent time slot of a fixed limit.

如果在每个相邻节点,对于每个波束来说,Nbeam的固定值满足(7),那么通过关于这些链路中的每个链路,与近邻的相互协定,所有节点能够为该链路及其每个波束选择不同的半永久时隙,而不考虑距离一个中继段以上的其它节点正在选择什么颜色。 If each adjacent node, for each beam is, Nbeam fixed value satisfies (7), then through each of these links in the link, mutual agreements with neighboring, all nodes that can link and each beam to select a different semi-permanent time slots, regardless of other nodes hop from one or more of what color is being selected. 这允许每个节点通过只与相邻节点通信,以非常直接的方式为每个波束选择其Nbeam个半永久时隙。 This allows each node through adjacent nodes only communicate with each beam to select its Nbeam semi-permanent time slots in a very direct way. 通过遵循该策略,每个节点能够支持至少Nk=Bk·Nbeam(8)个近邻,并且均被分配单一的SP时隙,同时每个波束被分配不高于Nbeam个这种时隙。 By following this strategy, each node is capable of supporting at least Nk = Bk · Nbeam (8) nearest neighbors, and were allocated a single SP time slots, and each beam is allocated a not more than Nbeam such time slots.

只要(7)被满足,就可支持每个波束Nbeam个近邻的验证是直接根据关于单波束情况的观察结果的验证得出的。 As long as (7) is satisfied, we can support each beam Nbeam a neighbor verification is based on verification on direct observation of single beam case derived. 从而,如果所有Bk个波束使它们的SP时隙按照相同的方式被调度,那么显然可被支持的相邻节点的数目是波束的数目与每波束的近邻数目的乘积,得到等式(8)。 So that the number, if all Bk beams have their SP time slots are scheduled in the same manner, it is clear that adjacent nodes that can be supported is the product of the number of neighbors per beam number of the beam, to obtain the equation (8) .

图10中表示了每个节点具有不同数目的波束的两个节点之间的SP时隙分配的例子。 Figure 10 shows an example of SP time slot assignment between two nodes each node having a different number of beams. 在该例子中,节点1具有2个波束,节点2具有3个波束。 In this example, node 1 has 2 beams, node 2 has 3 beams. 虽然这两个节点具有不同数目的波速,不过两个节点都必须使用相同的帧结构。 Although these two nodes have different numbers of velocity, but two nodes must use the same frame structure. 在本例中,Nframe=5时隙/帧。 In this example, Nframe = 5 time slots / frame. 根据(7)和(8),这允许节点1最多具有6个邻居,节点2最多具有9个邻居。 According to (7) and (8), this allows node 1 with a maximum of 6 neighbors and node 2 having a maximum of 9 neighbors.

最初,这两个节点都具有1个节点,小于在(7)和(8)的约束下,它们被允许的近邻的最大数目。 Initially, both nodes having a node, the maximum number is less than in (7) and (8) constraint, the neighbors they are allowed. 为每个链路表示了SP波束/时隙分配。 Represents the SP beam / slot assignment for each link. 这些节点可在它们之间增加额外的一条链路,同时仍然满足(7)和(8)的约束。 These nodes can add an additional between them a link, while still meeting (7) and (8) of the constraint. 链路调度协议将得到用于每个节点的SP分配的可接受的波束/时隙,并且它基本上按照和单波束情况下它所采用的相同方式工作。 Same way link scheduling protocol will be allocated for each node SP acceptable beam / time slot, and it substantially under and in accordance with its use of a single beam case work.

表22中表示了对应的协议消息交换。 Table 22 shows the corresponding protocol message exchange. 节点1通过发送REQ_SPTS(L=1,2,3)以及至少Nbeam个候选时隙的列表,启动所述交换。 The switching node 1 by sending a REQ_SPTS (L = 1,2,3) and at least Nbeam candidate time slot list, start. 注意3个波束ID由a、b和c表示,时隙编号由波束ID上的下标表示。 Note that three beams ID by a, b and c, said slot number represented by the subscript on the beam ID. 节点1必须识别它已在波束a上使用所有3个许可的SP时隙,但是在其波束b上,它只分配了3个许可的SP时隙中的2个。 Node 1 must recognize that it has permission to use all 3 SP time slots on beam a, but on its beam b, only three licenses assigned SP time slot 2.

从而,它把3个SP时隙(在波束b上可用)的列表发送给节点2。 Thus, it is the 3 SP time slots (available on beam b) a list sent to the node 2. 该列表可包括该波束上的所有自由时隙和DA时隙。 The list may include all free slots on the beam and the DA slots. 当发送请求消息时,对时隙和链路调度消息数据结构进行恰当改变。 When sending a request message to the time slot and link scheduling message data structures appropriately changed. 节点2先前已为其到其8个近邻的链路,在波束a和b上所有可用的SP时隙。 Node 2 has previously been to its 8 neighbors for their links, all available on beams a and b SP time slots.

从而,波束c是能够接收新的SP分配的唯一波束。 Thus, beam c is the only beam capable of receiving a new SP allocation. 当它从节点1收到REQ_SPTS(L=(1,2,3))时,它选择波束/时隙c3作为将为新链路(先前已分配c1和c2作为SP时隙)的唯一波束/时隙。 When it receives REQ_SPTS (L = (1,2,3)), while it selects beam / time slot for the new link as c3 (previously allocated c1 and c2 as SP time slots) is the only beam from node 1 / slots. 它在答复消息中发送该选择。 It sends a reply message to the selection. 当发送答复消息时,还对波束/时隙和链路调度消息数据结构进行恰当改变。 When sending a reply message, but also on the beam / time slot and link scheduling message data structures appropriate to change. 最后,当发送或收到确认时,恰当时隙的状态被改变为“被SP分配给链路(1,2)。”表22 Finally, when the transmission or acknowledgment is received, the state of the appropriate time slots are changed to "SP allocated to link to be (1,2)." Table 22

实现多波束调度算法/协议所需的改变简单并如下所示。 Multi-beam scheduling algorithm / protocol required to change simple and as follows. 把波束ID作为变量加入时隙DB和链路调度消息DB的状态中。 The beam ID as a variable joining slot and link scheduling message DB DB state. 使用(7)和(8)作为确定是否能够调度新的SP时隙的标准。 Use (7) and (8) as to determine whether to schedule a new SP time slot standards. 我们为网络指定参数Nframe和Nbeam的值。 We specify the values of the parameters for the network Nframe and Nbeam.

为了新潜在的近邻提供新的SP时隙分配,该算法必须首先找出近邻的数目小于Nbeam的波束。 In order to provide a new potential neighbors new SP slot allocation, the algorithm must first find the number of neighbors is less than Nbeam beam. 该波束随后可被用于增加该新邻居。 The beam can then be used to add the new neighbor. 节点向其近邻发送的REQ_SPTS消息将规定当前未被SP分配的波束的Nbeam可用时隙。 REQ_SPTS node to its neighbors to send a message to the provisions of the current distribution of the beam is not SP Nbeam available slot.

在收到REQ_SPTS消息的情况下,节点必须找出它的近邻的数目小于Nbeam的波束之一。 In the case of receipt of REQ_SPTS message, the node must find the number of its neighbors is less than one beam Nbeam of. 该波束随后可被用于增加新邻居。 The beam may then be used to add new neighbor. 通过比较接收的REQ_SPTS消息中的Nbeam个时隙的列表与所选波束中当前未被分配的Nbeam个时隙,可发现为这两个列表共用的至少一个时隙。 By comparing the received REQ_SPTS message in the list of Nbeam time slots in the selected beam Nbeam time slots not currently allocated, can be found for at least one time slot common to both lists. 该时隙可被选为将在REPLY_SPTS消息中发送的时隙。 The slot may be selected as the time slot to send in the REPLY_SPTS message. 一旦始发节点收到REPLY_SPTS消息,那么两个节点将已选择它们的波束和共同的时隙分配。 Once REPLY_SPTS originating node receives a message, then both nodes will have selected their beam and the common slot allocation.

该例子含蓄地假定单一频带被用于每个波束。 This example implicitly assumed that a single frequency band is used for each beam. 这种情况下,一个节点可具有通过相同频带无干扰地同时通信的几个波束。 In this case, a node may have the same frequency band without interference by several beams simultaneously communicating. 实践中难以支持这种无干扰操作。 In practice it is difficult to support this operation without interference. 就在不同的频带中工作的每个波束(即图10中的波束a、b和c均使用不同的频带)来说,可类似地用公式表示该问题。 At each beam working in different frequency bands (i.e., in Figure 10 the beam a, b and c are the use of different frequency bands), it can be formulated similarly to the problem. 根据调度算法,我们理应对SP时隙的分配应用相同的约束条件。 According to the scheduling algorithm, we should use the SP slot assignment same constraints. 但是,在实际分配时隙/波束组合的情况下,我们需要找出这样一种分配,以致这两个节点正在使用相同的波束(等同于使用相同的频带)以及相同的时隙。 However, in the case of the actual distribution time slot / beam combinations we need to find an allocation such that the two nodes are using the same beam (equivalent to using the same band) as well as the same time slot. 从调度的观点来看,这等同于使每个波束/时隙组合不同。 From the point of view of scheduling, which is equivalent to making each beam / time slot combination different. 从而,可用时隙的数目是波束的数目与帧大小的乘积。 Thus, the number of available slots is the number of beams multiplied by the frame size. 这种情况下,向潜在近邻分配SP时隙的约束条件由下式给出:B·Nframe≥2·N-1, (9)其中B表示波束的数目。 In this case, allocation of SP time slots to potential neighbors is given by the constraint condition: B · Nframe≥2 · N-1, (9) where B denotes the number of beams. 这种有关邻居数目的约束稍微比(7)和(8)的约束更严格,因为要求共用SP时隙的节点还必须使用相同的波束/频率通道以及相同的时隙。 This constraint related to the number of neighbors of a little more than (7) and (8) the more stringent constraints, because they require a common node SP slot must also use the same beam / frequency channel and the same time slot. 对于Nframe=5和B=3的例子来说,那么(8)的约束条件允许每个节点8个近邻,而(7)和(8)的约束条件将允许每个节点9个近邻。 For Nframe = 5 and B = 3 in the example, then (8) constraints allows each node to eight neighbors, and (7) and (8) of the constraint will allow 9 neighbors for each node.

图10中的例证问题具有2个节点,每个节点具有3个波束,每个波束在不同的频带中工作,即,波束a、b和c均使用不同的频带。 Figure 10 illustration of the problem has two nodes, each having three beams, each beam in a different frequency band, i.e., beams a, b and c are the use of different frequency bands. 另外假定帧大小为5。 Also assume that the frame size is five. 两个节点都已向相邻节点提交7个SP时隙,从而根据(9),它们都可利用一个SP时隙增加一个额外的近邻,所述一个SP时隙允许它们在其间建立链路。 Two nodes are submitted 7 SP time slots to neighbor nodes, according to (9), they can take advantage of a SP slot to add an additional neighbor, the one SP time slot allowing them to establish a link between them. 图中指示出了提交的SP时隙,在表23中示出了建立SP时隙分配和新链路所需的消息交换。 Figure indicates an SP time slot presented in Table 23 illustrates the establishing of a new SP time slot assignment and link required message exchange. 节点1通过向节点2发送REQ_SPTS(L=(a4,a5,b3,b4,b5,c3,c4,c5))启动消息交换,节点2必须包括它先前未分配为SP时隙的8个波束/时隙组合。 Node 1 to Node 2 by sending REQ_SPTS (L = (a4, a5, b3, b4, b5, c3, c4, c5)) start message exchange, node 2 which must include a previously unallocated time slot for the SP 8 beam / slot combinations. 在本例中,节点2已分配未被节点1使用的7个波束/时隙组合(它们在于REQ_SPTS消息中接收的8个波束/时隙组合的列表中)。 In this case, the node is not the node 2 has been allocated 7 beam / time slot combinations used for one (which in that list REQ_SPTS message received 8 beam / time slot combinations in). 从而,依据(9),存在它可选择以供分配的至少一个剩余波束/时隙组合(c5)。 Thus, according to (9), there is at least one remaining beam / time slot combination that it can select for allocation (c5). 这是分配给节点1和2之间的链路的SP波束/时隙组合,如图11和表23中所示。 This is assigned to the link between nodes 1 and 2 the SP beam / time slot combination, 11 and 23 shown in Table.

表23 Table 23

现在参见图12和13,决定何时适合于在两个移动节点12之间建立定向链路的一种有利的方法是首先确定节点之间的全向链路的质量,使所述决定基于该质量值。 Referring now to Figures 12 and 13, is adapted to decide when to establish between two mobile nodes 12 of a directional link an advantageous method is to first determine the node between omni link quality, make the decision based on the quality values. 即,定向信号的信号强度一般大于在相同功率水平下发射的定向信号的信号强度,因为定向通信链路的能量集中在一个方向上,而不在360°内展开。 That is, directional signals typically signal strength greater than the signal strength at the same power level transmitted directional signals, because the energy of the directional communication link is focused in one direction, but not in the 360 ° swing. 从而,在建立定向通信链路之前,通过要求节点具有足够的全向通信链路的质量值,这提供定向通信链路也将具有所需质量水平的良好指示。 Thus, before establishing the directional communication link, by requiring nodes to have a sufficient quality of the full value of the communication link, which provides the directional communication link will also have a good indication of a desired quality level.

实际上,全向链路质量是维持使用定向天线的节点的连接移动自组织网络的关键问题。 In fact, the whole is to maintain the link quality using a directional antenna nodes connected mobile ad-hoc network of key issues. 能够间歇或者持续相当长时间地导致定向链路和全向链路任意之一或者这两者损失的链路减损有多种多样。 Intermittent or sustained quite a long time can lead to all-directional links and link to any one or both of the link impairment losses have varied. 最好对这两种链路类型,持续测量链路质量,应采用机制来对抗链路质量方面的变化,并通过链路层通知,把链路质量通知网络层,不过在一些实施例中可以使用其它配置。 Preferably these link types, link quality is continuously measured, the mechanism should be used to combat changes in the link quality and, through the link layer notification, to notify the network layer link quality, but may in some embodiments using other configurations.

这样,最好如上所述,通过首先在方框121,在指定的一对移动节点12之间建立全向链路,开始建立定向链路的进程(方框120)。 Thus, it is preferable as described above, at block 121 by first, between a pair of mobile nodes specified in the establishment of an omni-directional link 12, begins the process of establishing a directional link (block 120). 当通过收到某一近邻的Link_HELLO分组,首先检测到该近邻时,发生上述步骤。 When received by a neighbor Link_HELLO grouping, first detected in the neighborhood, the occurrence of the above steps. 根据本发明的这一方面,在全向链路质量足够高之前,将不允许链路调度器启动建立定向链路的进程。 According to this aspect of the invention, the omni-directional link quality is sufficiently high before, the link scheduler will not be allowed to start the process of establishing a directional link.

控制器18如果确定全向链路的质量值(方框122)。 If the controller 18 determines the value of omni-directional link quality (block 122). 通过利用OLSR作为例证的路由协议,以便与OLSR规范相符,我们将把链路质量量度(由变量N_quality表示)定义为介于0和1之间的数,这里1代表最高质量。 As exemplified by the use of OLSR routing protocol to match the OLSR specification, we will link quality metric (N_quality represented by variable) is defined as a number between 0 and 1, where 1 represents the highest quality.

根据OLSR规范中计算质量的一种方法,这可被用作关于OLSR分组的分组接收差错值的估计值。 A method in accordance with the calculated mass of the OLSR specification, this may be used as an estimate for the packet OLSR packet reception error value.

应按照获得通道质量的可能的最准确估计值的方式实现估计通道质量的进程。 Shall obtain the channel quality estimate of the most accurate possible way to achieve the estimated channel quality of the process. 一种可能性是使用例如来自802.11卡的信噪比估计值,如果它们易于获得并且与分配给特定节点的时隙相关。 One possibility is to use e.g. SNR estimates from 802.11 card, if they are easily obtained and related to the time slot assigned to a particular node.

如果不能容易地使用信噪比估计值,那么可使用和OLSR一起使用的方法,该方法根据OLSR分组的成功接收来估计质量。 If the SNR estimation can not easily use value, and can be used with OLSR using the method based on the successful reception of OLSR packets to estimate quality. 对于全向链路,每秒从每个近邻收到类型为Link_HELLO,定向近邻和通道质量反馈的几个分组。 For the omni link several per second received from each neighbor of type Link_HELLO, neighbor and directional channel quality feedback packet. 对于指定的一组配置参数,每秒从每个相邻节点传送固定数目的这些分组。 For a given set of configuration parameters, a fixed number of the second transmission packets from each neighbor node. 从而,每秒传送已知数目的这些分组,接收的数目可被计数。 Thus, a known number of these packets per second transmission, the number of receivers can be counted.

计算全向质量值N_qualityomniLk的算法如下所示。 Omni quality value calculation algorithm N_qualityomniLk below. 对于来自节点k的每个成功分组接收,来自节点k的链路的N_qualityomniLk被更新为:N_qualityomniLk=(1-α)·N_qualityomniLk+α...(10)]]>对于丢失的来自节点k的每个分组传输,来自节点k的链路的N_qualityomniLk被更新为:N_qualityomniLk=(1-α)·N_qualityomniLk...(11)]]>在第一个接收的Link_HELLO之前,来自节点k的链路的N_qualityomniLk值的初始条件被设置为0,随后利用正确接收的Link_HELLO,根据(10)对其进行更新。 For each successful packet reception from node k, N_qualityomniLk link from node k is updated as: N_qualityomniLk = (1- & alpha;) & CenterDot; N_qualityomniLk + & alpha; ... (10)]]> For missing from Each packet transmission node k, N_qualityomniLk link from node k is updated as: N_qualityomniLk = (1- & alpha;) & CenterDot; N_qualityomniLk ... (11)]]> before the first one received Link_HELLO, from initial conditions N_qualityomniLk value link node k is set to 0, then the use of correctly received Link_HELLO, according to (10) to update it. 参数α的选择有效地设置一阶滤波器(或者指数加权平均值)的时间常数,以便计算成功分组传输的可能性的估计值。 The selection parameter α effectively sets a first order filter (or exponentially weighted average) time constant, in order to calculate the estimated value of the likelihood of successful packet transmission. 该参数值的正确设置取决于预期的每秒的消息数,它是所需的响应性和估计值的精度之间的折衷。 Correctly set the parameter value depends on the expected number of messages per second, which is a compromise between responsiveness and accuracy of the estimated value of between required. 本领域的技术人员会认识到,α的值越大,导致对于收到的每个新样本变化越大。 Those skilled in the art will recognize, the greater the value of α, for each new sample received cause larger changes. 这以增大对链路状态不正确分类的可能性的代价,导致链路状态的变化的快速识别。 This is at the cost of increasing the possibility of a link status is incorrect classification, resulting in rapid identification of the link status changes.

对于链路Lk,将设置变量Can_AllocLk,变量Can_AllocLk最好为TRUE,以便为初始SP时隙分配,或者为归因于干扰减轻的任何未来的DA分配或再分配,启动定向链路时隙分配。 For link Lk, will set the variable Can_AllocLk, variable Can_AllocLk best is TRUE, so that the initial SP slot assignment, or due to the interference mitigation for any future distribution or redistribution of DA, start directional link slot allocation. 即,在方框123和124,如果全向通信链路的质量值大于第一质量阈值(指示Can_AllocLk为TRUE),那么链路调度器将只建立与相邻移动节点的定向通信链路,从而结束图解说明的方法。 That is, at block 123 and 124, if an omnidirectional communication link quality value is greater than a first quality threshold (indicating Can_AllocLk is TRUE), then the link scheduler will only establish a directional communication link with neighboring mobile nodes, thus End method illustrated. 可如前所述建立定向通信链路。 Directional communication link may be established as previously described.

当然,在建立定向链路之后,通过确定/监视质量值(方框131),该方法可随意继续(方框130)。 Of course, after the establishment of the directional link, by determining / monitoring quality value (block 131), the method can easily continue (block 130). 这种情况下,根据下述关系,每次确定N_qualityomniLk时,修改变量Can_AllocLk:N_qualityomniLk<TomniL;Set Can_AllocLk=FALSE;...(12)]]>N_qualityomniLk>TomniH;Set Can_AllocLk=TRUE;and...(13)]]>TomniL≤N_qualityomniLk≤TomniH;]]>不修改Can_AllocLk(14)为了避免估计N_qualityomniLk值时,由统计波动造成的链路拍动,在(12)-(14)中引入滞后,以防止Can_AllocLk变量的修改,除非N_qualityomniLk的足够变化被估计,如方框132所示。 In this case, according to the following relationship, each time determining N_qualityomniLk, modify variables Can_AllocLk: N_qualityomniLk & lt; TomniL; Set Can_AllocLk = FALSE; ... (12)]]> N_qualityomniLk> TomniH; Set Can_AllocLk = TRUE; and .. . (13)]]> TomniL & le; N_qualityomniLk & le; TomniH;]]> does not modify Can_AllocLk (14) In order to avoid an estimated N_qualityomniLk value, link by statistical fluctuations flapping in (12) - the introduction of lag (14) , to prevent modification Can_AllocLk variable unless a sufficient change N_qualityomniLk is estimated, as shown in block 132. 选择第一阈值TomniH,第二阈值TomniL和参数α以确保该行为。 Select the first threshold TomniH, the second threshold parameter α TomniL and to ensure that the behavior. 阈值的例证值为TomniL和TomniH不过也可使用其它值。 Threshold values TomniL illustration and TomniH but other values may be used. 当然,如果质量值低于第二阈值TomniL,那么定向链路的使用可被临时暂停,或者链路被恰当地中断(方框133),从而结束图解说明的方法(方框134)。 Of course, if the quality value is below a second threshold TomniL, then the use of the directional link may be temporarily suspended, or the link is properly interrupted (block 133), thereby ending the method (block 134) illustrated.

应注意下面将进一步讨论的程序可基于使用Can_AllocLk变量作为门限,如上所述。 Note that the program should be further discussed below can be used as a threshold based on the use Can_AllocLk variables described above. 即,在时隙可被分配给定向链路Lk之前,该变量应具有值Can_AllocLk=TRUE.]]>当对于链路Lk,收到第一个Link_HELLO分组时,该变量最初被设置为值Can_AllocLk=FALSE.]]>当然,可使用其它方法来确定何时适合于建立和/或继续使用定向通信链路。 That is, in the time slot can be allocated to a directional link before Lk, the variable should have the value Can_AllocLk = TRUE.]]> When For link Lk, the first Link_HELLO received packet, the variable is initially set to a value Can_AllocLk = FALSE.]]> Of course, you can use other methods to determine when it is appropriate to establish and / or continue to use the directional communication link.

现在参见图14-16,更详细地说明按需分配时隙的分配。 Referring now to Figures 14-16, on demand assigned time slots described in more detail. 最好按照响应网络通信量的波动需求的方式分配DA时隙。 DA slot allocation in accordance with the best response to fluctuations in demand for network traffic ways. 如上所述,变量Can_AllocLk可被用于确定何时首次适合于建立定向通信链路。 As described above, variables can be used to determine when Can_AllocLk first orientation suitable for establishing a communication link. 即,良好的全向通道质量可被用作新DA时隙的分配的前提条件。 That is, good omni channel quality may be used as a new DA time slot assigned prerequisite. 从而,对于链路Lk,在时隙的请求者或者接收时隙分配请求的节点分配新的DA时隙之前,变量Can_AllocLk应为TRUE。 Before thus, for the link Lk, the allocation of new DA time slot or in the requester node receives a request slot allocated slot, variable Can_AllocLk should be TRUE.

类似地,最好还在响应干扰条件或其它链路退化,重新分配DA定向时隙之前,要求变量Can_AllocLk为TRUE。 Similarly, the best response is still interference conditions or other link degradation, redistribution of DA directional slots before asking Can_AllocLk variable to TRUE. DA时隙的分配的下述说明假定Can_AllocLk为TRUE,但是不是在所有实施例中都要求这样。 DA slot assignment following description assumes Can_AllocLk is TRUE, but not all embodiments require so.

根据本发明的这一方面,只有当为一个节点和一个相邻节点之间的链路分配一个半永久时隙时,才允许从所述一个节点到所述相邻节点的对按需时隙的请求。 According to this aspect of the invention, only when a semi-permanent time slot allocated for the link between a node and an adjacent node before allowing the adjacent node to said one node from the on-demand time slot request. 在链路被分配至少一个半永久时隙之后,节点可请求额外时隙的周期分配。 After the link is allocated at least one semi-permanent time slots, a node may request a periodic allocation of an additional time slot. 用于调度DA时隙的消息可通过全向链路或者定向链路发送。 DA slots for scheduling messages may be sent to the full or directional link by link.

更特别地,每个控制器18最好包括数据队列18f,用于保存将通过指定的通信链路发送的数据。 More particularly, each controller 18 preferably includes a data queue 18f, for storing the data transmitted by specifying a communication link. 在如上所述,在方框141在各对移动节点12之间调度相应的半永久时隙的情况下,开始分配按需分配时隙的第一种方法(图14)(方框140)。 The first method (FIG. 14) (block 140) as described above, at block 141 in the scheduling of the mobile node 12 between the respective semi-permanent time slots, the start-demand time slot allocation. 在方框142,可确定在半永久时隙期间先前通过通信链路发送的数据的数量,以及保存数据队列中的数据的数量(方框143),并在方框144,据此确定每个通信链路的链路使用度量。 At block 142, the data may be determined during a semi-permanent time slot previously transmitted via a communication link number, and the number of data stored in the data queue (block 143), and at block 144, each communication is determined accordingly link into the use of metrics. 随后在方框145,根据链路使用度量,可调度按需分配的时隙(下面将进一步说明),从而在方框146结束图解说明的方法。 Then in block 145, according to the link using metrics that can be allocated on demand scheduling slot (further described below), which ends at block 146 the method illustrated.

特别地,对于DA时隙的有效分配来说,每个链路上的通信量需要的测量特别重要。 In particular, for efficient allocation of DA time slots, the need to measure traffic on each link is particularly important. 如上所述,这最好根据通过指定链路发送的数据的数量和在队列18f中正在变得倒退(back up)的数据来实现。 As described above, this is preferably based on the number of transmitted data and by specifying the link in the queue 18f are becoming backward (back up) data is achieved. 就前者来说,测得的通过节点i和k之间的链路发送的平均通信量(以每个信号出现时间的时隙的数目为单位)将由Tikse表示。 In the former, the measured average traffic through the node link transmission between i and k (in number of time slots per epoch as a unit) will be expressed Tikse. 该量度将包括通过每个信号出现时间的一个或多个半永久时隙以及任何按需时隙发送的所有通信量。 This measure will include the time of occurrence of each signal through one or more semi-permanent time slots as well as any on-demand time slots of all traffic sent.

还为节点i和k之间的链路保持队列状态的当前量度Qik。 Also between nodes i and k Link maintain the current queue status measure Qik. 较大的Qik值指示需要立即分配一个或多个DA时隙。 Qik value indicates a greater need to assign one or more of the DA slots immediately. 需求的偶然爆发会引起Qik增大,这随后应触发对DA容量的额外时隙的请求,直到队列规模降低为止。 Demand will cause occasional outbreaks Qik increases, which should then trigger DA slot for additional capacity requests, until reduced until the queue size.

在节点i和k之间的链路上分配的时隙的总数由Niktot表示。 The total number of the link between nodes i and k allocated time slot is denoted by Niktot. 时隙需求被定义为:Tikdem=f(Tikse,Qik)...(15)]]>这是测量的通信量加入由队列规模指示的估计的所需额外容量的函数。 Time slot demand is defined as: Tikdem = f (Tikse, Qik) ... (15)]]> This is added to the estimated traffic measurement indicated by the queue size of the extra capacity required function. 该函数可采取多种形式,还会受对保留容量的新请求影响。 This function can take many forms, but also affected by the request for a new retention capacity.

在一些实施例中,最好允许链路度量受通信量需求的优先级影响。 In some embodiments, better to allow the link metric is affected by the priority of the traffic demand. 即,某一时隙期间,对指定通信链路的需求的等级可被进一步分成对几种优先级类别的通信量的预期需求的数量。 That is, during a certain time slot, the communication link needs to specify the level can be further divided into a number of expected demand for several traffic priority class of. 这使得能够更有效地分配时隙以满足优先化通信量的需求。 This enables more effective allocation of time slots to meet the prioritized traffic demand.

从而,参考图15,通过类似于参考在图14中的方框141图解说明的步骤描述的那样,调度SP地洞,开始这种方法(方框151)。 Thus, with reference to FIG. 15, as similar to the reference block 141 through step illustrated in FIG. 14 described in the description, scheduling SP hole, this method starts (block 151). 随后在方框152,可关于每个通信链路,对多个数据优先级或类别中的每一个,确定链路使用度量。 Then at block 152, for each communication link may be, for each of the plurality of data priority or category, determining link utilization metrics. 这可利用上述度量之一或者两者(即先前传送的数据的数量和/或保存在数据队列18f中的数据的数量),或者其它适当的使用度量来实现。 This metric can be used one or both of the above (i.e., the number of the number of previously transmitted data and / or the data stored in the data queue 18f), or other suitable metrics used to achieve.

例如,如果通信量按照从最低到最高的顺序被区分优先顺序为p个优先级类别,p1,p2,p3,…,pp,那么对优先级类别pj的时隙需求为Tikdem,Pj=f(Tikse,Pj,QikPj)...(16)]]>对每个优先级类别计算独立的需求度量。 For example, if the traffic in accordance with the order from the lowest to the highest are prioritized into p priority classes, p1, p2, p3, ..., pp, then the time slot demand for priority class pj is Tikdem, Pj = f ( Tikse, Pj, QikPj) ... (16)]]> computing needs independent measure of each priority category. 随后利用来自两个方向的度量,计算该链路上对于该优先级类别所需的时隙的数目Tikneed,Pj:最好是使最高优先极需求度量Tikneed,Pp增大一个偏项B(例如,约为时隙的1/4),以确保分配足够的容量,从而避免大排队。 Then use metrics from both directions, calculated on the link for the time slot of the priority class of the desired number Tikneed, Pj: it is preferable that the highest priority demand metric pole Tikneed, Pp increases a bias term B (e.g. about 1/4 of the time slot) to ensure the allocation of sufficient capacity to avoid big queues. 随后,集合度量可被任意确定(方框153)为所有优先级类别的集合需求,即:Tikneed=Σj=1PTikneed,Pj...(18)]]> Subsequently, the set of metrics may be arbitrarily determined (block 153) for all priority classes of the set of requirements, namely: Tikneed = & Sigma; j = 1PTikneed, Pj ... (18)]]>

随后在方框154可据此及根据优先级调度DA时隙,从而结束该方法(方框155)。 Then accordingly and according to priority scheduling DA slot, thus ending the process (Block 155) in block 154.

需求度量被分配给该链路,它是对容量的需求和先前分配给该链路的时隙的数目的函数。 Demand metric is assigned to the link, which is a function of the number and capacity requirements of the slots previously assigned to the link. 我们用MikDA(TikPj,Niktot)来表示它,指示它取决于估计的对所有优先级类别的需求和分配时隙的数目。 We use MikDA (TikPj, Niktot) to indicate that it, indicating the number of all the priority class needs and it depends on the allocated slot estimate. 该方法规定当较高优先级通信量的比例增大,以及当需要更多时隙时,该度量增大。 This method provides a high-priority traffic when the ratio increases, and when you need more slots, this measure increases. 另外,如果分配时隙的数目被增大,那么该度量被降低。 Further, if the number of assigned time slots is increased, then the metric is reduced. 从而,如果某一链路具有过量的时隙,那么该度量为负。 Thus, if a link has excess time slots, then the metric is negative. 当然,本领域的技术人员会认识到在各种应用中可以使用其它度量。 Of course, those skilled in the art will recognize that various applications may use other metrics.

下述算法可被用于计算链路度量。 The following algorithm can be used to calculate the link metric. 对于下面的例子,为了简便起见,我们假定三种优先级类别,优先级类别p3最重要。 For the following example, for simplicity, we assume that three kinds of priority classes, the most important priority class p3. 如果Tikneed≤Niktot,]]>那么设置MikDA(TikPj,Niktot)=Tikneed-Niktot;...(19)]]>否则如果Tikneed-Tikneed,P1≤Niktot,]]>那么MikDA(TikPj,Niktot)=max(Ks,Tikneed-Niktot);...(20)]]>否则如果Tikneed-Tikneed,P1-Tikneed,P2≤Niktot,]]>那么MikDA(TikPj,Niktot)=max(2·Ks,Ks+Tikneed-Tikneed,P1-Niktot);...(21)]]>否则MikDA(TikPj,Niktot)=2·Ks+Tikneed-Tikneed,P1-Tikneed,P2-Niktot...(22)]]>在上面的算法中,选择恒定参数Ks≥1以便对每种优先级类别允许和度量中所需的一样多的范围(range)。 If Tikneed & le; Niktot,]]> then set the MikDA (TikPj, Niktot) = Tikneed-Niktot; ... (19)]]> Otherwise, if Tikneed-Tikneed, P1 & le; Niktot,]]> So MikDA (TikPj, Niktot) = max (Ks, Tikneed-Niktot); ... (20)]]> else if Tikneed-Tikneed, P1-Tikneed, P2 & le; Niktot,]]> then MikDA (TikPj, Niktot) = max (2 & CenterDot; Ks, Ks + Tikneed-Tikneed, P1-Niktot); ... (21)]]> Otherwise MikDA (TikPj, Niktot) = 2 & CenterDot; Ks + Tikneed-Tikneed, P1-Tikneed, P2-Niktot ... (22)] ]> In the above algorithm, in order to select Ks≥1 constant parameter for each priority class and metrics required to allow as much of the range (range). 例如,如果Ks=1,那么如果需要额外的时隙,并且存在类型P3通信量需求,那么集合度量将在范围2<MikDA(TikPj,Niktot)≤3]]>中。 For example, if Ks = 1, then if additional time slots, and the presence of Type P3 traffic demand, then the set of metrics will be in the range of 2 & lt; MikDA (TikPj, Niktot) & le; 3]]> in. 同样地,如果最高的通信量类别为类别P2,那么集合度量将在范围1<MikDA(TikPj,Niktot)≤2]]>中。 Likewise, if the highest traffic category category P2, then the measure will be in the range of set 1 & lt; MikDA (TikPj, Niktot) & le; 2]]> in. 否则,如果最高的通信量类别为类别P1,那么集合度量将在范围0<MikDA(TikPj,Niktot)≤1]]>中。 Otherwise, if the highest traffic category for category P1, then the set of metrics will be in the range 0 & lt; MikDA (TikPj, Niktot) & le; 1]]> in. 通过为Ks选择更大的值,该范围可被进一步扩大。 By selecting a larger value for Ks, the range can be further expanded. 对于Ks=2,利用上述程序,从最低到最高的3种优先级类别的度量范围为0-2、2-4和4-6。 For Ks = 2, using the above procedures, from the lowest to the highest of three kinds of measurement range priority category for 0-2,2-4 and 4-6.

如上计算的度量MikDA(TikPj,Niktot)提供链路是否需要额外的时隙分配(即,MikDA(TikPj,Niktot)是否为正)和哪个优先级类别要求所述分配的基础。 As a measure of computing MikDA (TikPj, Niktot) whether the provision of a link requires additional slot allocation (ie, MikDA (TikPj, Niktot) is positive) basis and which claims priority class assignment. 它还使新时隙能够以最高的优先级被分配在具有指示对时隙的最大需要的最高度量的那些链路上。 It also enables a new time slot can be assigned the highest priority in the time slot having the indication of the maximum need for those links on the highest metric. 另外,它还指示是否存在分配给该链路的过量容量(即,MikDA(TikPj,Niktot)为负)。 In addition, it indicates whether there is excess capacity assigned to the link (i.e., MikDA (TikPj, Niktot) is negative).

要注意的关键点是当使用上述方法时,链路度量将指示对于时隙的优先化相对需求。 The key point to note is that when using the above method, the link metric indicates the priority for the slot of the relative demand. 大于0的最大度量指示与任何其它链路相比,最需要用于较高优先级的通信量的额外容量。 Greater than 0 indicates the maximum measure than any other link, the need for additional capacity higher priority traffic. 小于0的度量指示链路具有过量的容量。 The measure is less than 0 indicates that the link has excess capacity. 最小的链路度量指示具有最大的过量容量的链路。 Smallest link metric indicates the link with the greatest excess capacity. 该链路从而理应是在另一链路上重新调度容量(如果需要的话)的最佳候选者。 The link is thus supposed to be rescheduled on another link capacity (if required) the best candidate.

下面将参考图16说明例证的方法,该方法将在每个网络节点12被用于把用于定向链路的按需时隙分配给相邻节点。 Described below with reference to FIG. 16 illustration, the method will be used to assign 12-demand time slots for directional links to neighbor nodes in each network node. 该方法开始于方框160,通过利用在图14和15任意之一或者两者中概述的方法,在方框161,移动节点12将连续保持其被分配半永久时隙的每个链路的链路度量MikDA(TikPj,Niktot)。 The method begins at block 160, FIG. 14 and 15 by using any one of or both the methods outlined in, at block 161, the mobile node 12 will maintain a continuous chain which is semi-permanently assigned time slots for each link Road measure MikDA (TikPj, Niktot).

每上节点12将使用该度量来向每个相邻节点指示对额外传输时隙的需求。 Each node 12 will use this metric to indicate the need for additional transmission time slots to each neighbor node. MikDA(TikPj,Niktot)的最大值指示按照优先级排序,最需要额外的DA时隙分配的链路。 MikDA (TikPj, Niktot) indicates the maximum ordered by priority, most require additional DA slot assignment link. MikDA(TikPj,Niktot)>2·Ks]]>的值指示对最高优先级类别P3的额外时隙的需要。 MikDA (TikPj, Niktot)> 2 & CenterDot; value Ks]]> indicates the need for additional time slots of the highest priority class P3's. 类似地,如果Ks<MikDA(TikPj,Niktot)≤2·Ks,]]>那么需要额外的时隙来服务优先级类别P2。 Similarly, if Ks & lt; MikDA (TikPj, Niktot) & le; 2 & CenterDot; Ks,]]> then require additional time slots to serve priority class P2. 最后,如果0<MikDA(TikPj,Niktot)≤Ks,]]>那么需要额外的时隙来服务于优先级类别P1。 Finally, if 0 & lt; MikDA (TikPj, Niktot) & le; Ks,]]> you need additional slots to serve the priority class P1.

在度量MikDA(TikPj,Niktot)被保持的情况下,在方框162,具有大于0的最大链路度量的加应被选为额外时隙分配的首选。 In the case of the preferred metric MikDA (TikPj, Niktot) are maintained, at block 162, with the largest link metric greater than zero should be selected as additional additive slot allocation. 如果存在可用作自由时隙或者用作对其它链路的过量DA分配(同样由小度量指示)的时隙或者正被使用但是具有较低优先级的时隙,那么进程转变到增加DA时隙状态,启动查找DA时隙分配的进程。 If there can be used as free slots or as excess DA allocation of time slots to other links (again indicated by a small metric) or is being used but have a lower priority slot, then the process transitions to increase DA slot state, start to find DA slot assignment process.

存在可用在时隙再分配进程中的几种其它度量。 There are available slots redistribution process in several other metrics. 在方框163,可计算增大的容量链路使用度量MikDA(TikPj,Niktot+1),以估计在增加额外的时隙之后,链路度量如何变化。 At block 163, calculate the increased capacity link utilization metrics MikDA (TikPj, Niktot + 1), to estimate the additional time slots after, how the link metric changes. 这可用于确定是否应请求另一时隙,以及该请求的优先级如何。 This can be used to determine whether the request of the other time slots, and how the priority of the request. 它可指示该链路应被分配另一时隙,或者另一链路现有具有较大的度量。 It may indicate that the link should be assigned to another slot, or another existing link with a larger measure. 在后一情况下,下一时隙应被分配给具有最大度量的链路。 In the latter case, the next time slot to be allocated to the link with the largest metric.

类似地,在方框164,可计算降低的容量链路使用度量MikDA(TikPj,Niktot-1),以估计如果从该链路取走一个时隙,链路使用度量将为多少。 Similarly, at block 164, calculate a reduced capacity link utilization metrics MikDA (TikPj, Niktot-1), in order to estimate if a time slot is removed from the link, the link will use the measure of how much. 这可用于从一个链路到另一链路的时隙的重新分配。 This can be used to redistribute from one link to another link time slots. 即,如同本领域技术人员理解的那样,如果增大的和/或降低的链路使用度量在可接受的限度内,那么在方框165,为再分配指定的按需分配时隙随后被重新分配,从而结束图解说明的方法(方框166)。 That is, as those skilled in the art will appreciate, if the increased and / or decreased link utilization metrics are within acceptable limits, then at block 165, designated as time slot is then re-demand redistribution distribution, thus ending the illustrated method (Block 166).

如同半永久时隙一样,节点12只需要与某一近邻协调将为到该近邻的定向链路分配的DA时隙的选择。 As a semi-permanent time slot as node 12 will require coordination with a neighbor to the neighbor directional link DA slot allocation choices. 这意味着某一近邻将通过定向链路向该近邻发送时隙分配请求,并通过相同的链路接收分配的批准或者请求的拒绝,如下进一步所述。 This means that a neighbor will send the time slot assignment request via a directional link to the neighbor, and receiving an allocation request for approval or refusal by the same link, as described further below.

最好根据预定的时间表定期重新计算链路度量。 The best link metric recalculated periodically according to a predetermined schedule. 链路度量大于某一阈值Max_metric_threshold的链路是获得新的DA容量的候选者。 Link metric greater than a certain threshold Max_metric_threshold link is to obtain a new DA capacity of the candidates. 该度量可指示对多个时隙的需要,可在一个请求中同时请求它们。 This metric may indicate the need for a plurality of slots, which can request them in a request. 具有超过所述阈值的最大度量的链路喜欢分配新的DA时隙。 Link having a maximum value exceeding the threshold metric like the new DA time slot allocation. 但是,如果多个链路度量超过阈值Max_metric_threshold,那么某一节点可同时向多个相邻节点请求DA时隙。 However, if multiple link metrics exceed the threshold Max_metric_threshold, then a node may request DA time slots simultaneously to a plurality of neighboring nodes. 链路度量的负值指示该链路具有过量的容量,不过在各个实施例中也可采用其它形式的度量。 A negative value indicates that the link has a link metric of excess capacity, but in various embodiments may be employed other forms of metric. 该容量可被放弃,以满足具有更大链路度量的其它链路的需求。 The capacity can be abandoned in order to meet the needs of other links have a greater link metrics.

使用多个数据优先级还提供区分数据优先顺序的灵活性,从而优先级较高的数据可首先被传送。 Using multiple data priority also provides flexibility to differentiate data prioritization, and thus higher-priority data can be transmitted first. 当然,指定链路可具有在每个定向信号出现时间中分配的几个时隙。 Of course, a given link may have allocated in each directional epoch in several time slots. 从而,到相邻节点k的定向链路Lk可具有分配给它的时隙i和j,这两个时隙可具有不同的链路质量。 Thus, to the adjacent node k may have a directional link Lk time slot i allocated to it and j, the two time slots may have a different link quality. 根据干扰水平,时隙之一可能质量较高,而另一时隙可能质量较差。 According to the level of interference, one slot may be high quality, and the other time slots may be of poor quality. 一般不适于通过质量较差的时隙传送重要通信。 Generally not suitable for transfer important communications by the poor quality of the slot.

现在将参考图17说明根据指定时隙内的链路服务质量(QoS),区分数据的优先顺序的一种有利方法。 17 will now be described with reference to FIG link quality of service (QoS) within the specified timeslot, an advantageous method to distinguish the priority of the data. 开始于方框170,在方框171,在指定的一对移动节点12之间调度多个时隙。 Begins at block 170, at block 171, between the pair of mobile nodes specified scheduling a plurality of slots 12. 同样,如前所述,这一般包括一个或多个SP时隙,以及一个或多个DA时隙。 Similarly, as previously described, which typically include one or more SP time slots, and one or more DA time slots. 同样如上所述,在方框172,还例证确定在每个时隙内,与通信链路相关的相应链路质量值。 As also described above, at block 172, also in the illustration to determine each time slot, the communication link of the respective link quality values. 这可根据SIR、分组接收差错值等来实现。 This can be based on SIR, packet reception error values, etc. to achieve. 这样,在方框173,根据数据的优先级和质量值,数据被区分优先顺序,以便在时隙内被传送,从而结束图解说明的方法(方框174)。 Thus, at block 173, according to the priority and the quality value of the data, the data is prioritized to be transmitted during a time slot, thus ending the method (block 174) illustrated.

更具体地说,控制器18最好把与优先级中最高的一个优先级对应的数据分配给具有与之相关的最高质量值的时隙。 More specifically, the controller 18 is best to assign the highest priority and a priority of the corresponding data to the slot has associated with the highest quality value. 例如,最高的优先级可能对应于网络控制数据,因为这种数据一般要求即时传送和动作,以便减轻干扰,把丢失的链路重新分配给新时隙等。 For example, the highest priority may correspond to network control data, as this data transfer and generally require immediate action to mitigate interference, the missing link to a new time slot reallocation like.

另外参见图18,说明区分数据优先顺序的一种例证方法。 See also FIG. 18, described distinguish one illustrative data prioritization method. 开始于方框180,在方框181,首先最好依据数据优先级对等待传送的数据分级,随后在方框182,依据数据被接收的顺序进一步对所述数据分级。 Begins at block 180, in block 181, the first priority of the best on the basis of the data transmitted to wait for data classification, followed by block 182, the order on the basis of data received further to the data classification. 来自这两个步骤的最高分级的数据随后可被选择(方框183),并在方框184,在第一个可用时隙内被发送,对于所述第一可用时隙来说,质量值高于关于该数据优先级的相应最小质量阈值。 The highest-ranking data from these two steps may then be selected (block 183), and 184, is transmitted in the first available slot in the block, for the first available time slot, the weight value higher than the corresponding minimum quality threshold on the priority of the data. 该进程随后继续进行,直到所有数据都被发送为止(方框185),从而在方框186结束图解说明的方法。 The process then continues until all data has been transmitted so far (block 185), so that the method illustrated at block 186 ends. 当然,在实际操作中,上述步骤将被频繁重复,但是为了清楚起见,它们被图解表示成在方框186终止。 Of course, in practice, the above steps will be repeated frequently, but for clarity, they are illustrated as terminating at Block 186.

应注意数据队列18f事实上可被实现成独立的数据优先级队列,如果实现了基于质量值(即QoS)的优先化。 Note that the data queue 18f in fact be implemented as independent data priority queue, if implemented based on quality values (ie QoS) prioritization. 此外,一些修改是合乎需要的,以便考虑到分配给特定链路的每个时隙的不同链路质量。 In addition, some modifications may be desirable to take into account the different link quality of each link is assigned to a particular time slot. 同样利用上述例子,可能不希望通过质量可疑时隙传送OLSR网络控制通信量。 Also using the above example, you may not want to transmit OLSR network control traffic through quality suspicious slots.

在节点j关于经由时隙i,其到节点k的链路估计的质量量度由N_qualityiLk表示。 Via a slot on the node j i, its link to node k is represented by the estimated quality measure N_qualityiLk. 类似的,在节点k关于该时隙的并将在通道质量反馈分组中发射的估计质量由N_quality_RxiLk表示。 Similarly, the estimated quality emitted at node k of the slot and on the channel quality feedback packet indicated by N_quality_RxiLk. 随后,根据这两个估计值,时隙i的质量可被估计为QualityiLk=Min(N_qualityiLk,N_quality_TxiLk).]]>如果网络控制通信量被赋予最高优先级,那么当分配给定向链路Lk的时隙i出现时,关于该定向链路Lk的任何排队网络控制通信量可以首选使用该时隙,假定关于时隙的质量的一些条件被满足。 Then, based on these two estimates, the quality of time slot i may be estimated as QualityiLk = Min (N_qualityiLk, N_quality_TxiLk).]]> If network control traffic is assigned the highest priority, then when allocated to a directional link Lk of time slot i appear on any of the queuing network control traffic directional link Lk may prefer to use the time slot, assuming certain conditions regarding the quality of slots are met.

对于网络控制通信的类别来说,这样的条件如下所示。 For network control communication particular category, such conditions are as follows. 发送给节点k的分组在第一时隙i中被传送,并且如果链路Lk的质量满足 Packets sent to node k is transmitted in the first time slot i, and if the quality of the link Lk satisfies 那么所述分组被分配给定向链路Lk,其中选择阈值TNC以便为网络控制通信确保足够的链路质量。 Then the packet is assigned to a directional link Lk, which selection threshold TNC control communication network in order to ensure sufficient link quality. 否则,发送给节点k的分组可在分配给具有最高链路质量QualityiLk的定向链路Lk的时隙中被发送。 Otherwise, the packets sent to node k may be transmitted in the time slot allocated to a directional link Lk has the highest link quality QualityiLk in.

此外,对于任务数据通信的一些类别,最好施加类似的时隙质量条件。 In addition, for some categories of data communication task, preferable to apply a similar slot quality conditions. 对于任意任务数据类别Cm,可施加如下所述的时隙质量条件。 For any of the task data category Cm, slot quality conditions as described below may be applied. 发送给节点k的分组可在分配给满足QualityiLk>TCm]]>的定向链路Lk的第一时隙i内被传送,其中选择阈值TCm以便为类别Cm的任务数据通信,确保足够的链路质量。 The packets sent to node k may be allocated to meet QualityiLk> TCm]]> a directional link Lk of a first time slot i is transmitted, wherein in order to select the threshold TCm Cm task category data communication, to ensure sufficient link quality. 否则,发送给节点k的分组可在分配给具有最高链路质量QualityiLk的定向链路Lk的时隙内被传送。 Otherwise, the packet transmitted to node k may be transmitted during a time slot allocated to a directional link Lk has the highest link quality QualityiLk.

本领域的技术人员会认识到,上述方法的优点是具有更长的信号出现时间,在每个信号出现时间内,几个时隙被分配给每个链路Lk。 Those skilled in the art will recognize that the advantage of the above method is a signal having a longer time of occurrence, within each epoch, several time slots are allocated to each link Lk. 即,当由于干扰的缘故,一个时隙临时地或者永久地质量较差时,链路恶化的机会很小。 That is, when the interference due to the reason, a time slot of poor quality temporarily or permanently, the opportunity to link the deterioration is small. 这种情况下,其它的分配时隙可具有更高的质量,并且对保持关键通信的QoS来说关系重大。 In this case, the other assigned slot may have a higher quality, and for maintaining QoS for critical communications is a significant relationship. 如果单个时隙被分配给某一链路,那么如果由于干扰的缘故,所述单个分配时隙开始恶化,那么新时隙的分配可能存在延迟。 If a single time slot is allocated to a link, then if due to interference reason, the single dispensing slot began to deteriorate, then assign a new time slot may be a delay.

最好对定向链路以及全向链路进行链路质量检测。 The best directional and omni-directional link links link quality testing. 然而,对于定向链路来说,这更复杂,因为每个时隙可能具有不同的链路质量。 However, for a directional link, it is more complicated, because each time slot may have a different link quality. 除了引起链路中断的正常传播效应之外,定向链路经受由来自正在复用相同时隙的其它节点对的干扰引起的质量恶化。 In addition to normal propagation effects caused by interruption of the link outside, subjected to a directional link quality from deterioration by the other nodes are multiplexed on the same time slot due to noise. 从而,需要单独估计并保存每个时隙的链路质量(即使几个时隙被分配给相同的相邻节点)。 Thus, the need for separate and store the estimated link quality of each time slot (even if several time slots are allocated to the same neighbor node).

此外,即使就关于每个时隙单独估计的链路质量来说,几个时隙内某一链路的集合链路质量一般是传送给路由协议的唯一信息,和OLSR的情况一样。 Moreover, even on the link quality on individual estimates for each slot, set within a few slots link quality is generally only a link information to the routing protocol, and situations like OLSR. 由于对某些时隙,通信量需求的变化和干扰约束条件的变化(由几何开关的变化引起)所需的对链路的频繁时隙重新分配,使该进程复杂化。 Due to changes in certain time slots, changes in traffic demand and interference constraints (induced by changes in the geometry of the switch) needed to link the frequent reassignment of time slots, so that the process is complicated.

在链路层的所有该活动事实上对路由协议来说是透明的。 All of this activity at the link layer routing protocol in fact is transparent. 对于到某一近邻的任意指定链路,一般同路由协议报告的唯一信息是到该近邻的链路的链路质量。 For any given link to a neighbor, generally only information reported is the same routing protocol to link quality of the link neighborhood. 下面说明本发明的可在链路层用于支持链路质量检测,干扰避免和减轻的各种特征,以及它们与分配新时隙的关系。 The following description of the invention may be used at the link layer to support link quality detection, interference avoidance and mitigation of various features, and their relationship to allocate a new time slot.

现在将参考图19和20说明一种特别有利的确定定向链路质量的方法。 19 and now described a particularly advantageous method for determining the quality of a directional link 20 with reference to FIG. 如果信噪比估计值不能容易地被用于估计每个单个时隙的质量,那么可以使用上面所述的估计全向链路质量的方法(即,使用全向开销分组(overhead packet)的成功接收)。 If the SNR estimation value can not be easily used to estimate quality for each individual time slot, then the estimate can be used above all according to the link quality of the process (i.e., using the whole successfully overhead packet (overhead packet) of reception). 就定向时隙来说,通信量主要是任务数据通信量。 On directional time slot, the traffic is primarily the task of data traffic.

具体地说,开始于方框190,在每个时隙中,源节点将传送几个这样的数据分组,但是接收节点不知道传送了多少个数据分组。 Specifically, begins at block 190, in each time slot, the source node transmits several such data packets, but the receiving node does not know how many data packets transferred. 为了使接收节点知道在每个时隙中,传送了多少个数据分组,在方框191,在该时隙中,连同数据分组一起,发射器还传送特殊的开销分组PKT_CT。 To enable the receiving node to know in each time slot, the number of transmitted data packets, at block 191, in the time slot, along with the data packets together, the transmitter also transmits a special overhead packet PKT_CT. 开销分组提供该时隙内传送的分组的计数(包括它自己)。 Overhead Packet provides count packet transmitted within the time slot (including itself). 从而,如果收到了PKT_CT分组,那么接收节点将知道在一个时隙中总共传送了多少个分组。 Thus, if the received packet PKT_CT, the receiving node will know that in a time slot in total number of packets transmitted. 我们把在该时隙中传送的分组的数目表示为niT。 We put the number of transmitted packet in the time slot denoted as niT. 在方框192,接收节点还对正确接收的分组的数目(表示为niR)计数。 At block 192, the receiving node is also the number of correctly received packets (denoted as niR) count.

根据在时隙i中接收的分组,可估计时隙i的分组接收差错值。 The packet received in time slot i, the time slot i may be estimated packet reception error value. 最好,网络被这样配置,以致始终传送至少一个分组(即,PKT_CT分组)。 Preferably, the network is configured so that always transmitting at least one packet (i.e., PKT_CT packet). 这样,如果在指定时隙中没有收到任何分组,那么时隙i的分组接收差错值被估计为0。 Thus, the absence of any packets in the specified time slot, then the packet reception error value for time slot i is estimated as 0. 如果在方框193,PKT_CT分组被正确接收,那么它提供niT的值,时隙i的分组接收差错值被估计为niR/niT,从而在方框196结束图解说明的方法。 If at block 193, PKT_CT packet is received correctly, it provides the value of niT, and the packet reception error value for time slot i is estimated as niR / niT, so that the method illustrated at block 196 ends.

可能发生的另一种情况是如果PKT_CT分组未被正确接收,并且niT的值未知。 Another situation that can happen is if PKT_CT packet is not received correctly, and niT the value is unknown. 这种情况下,时隙i的正确分组接收的概率被估计为被正确接收的分组占据的时隙的分数(方框195),即BiR/TSi。 In this case, the probability of correct packet reception for time slot i is estimated as the packet is received correctly fraction occupied slot (block 195), i.e. BiR / TSi. 这里,BiR表示在该时隙内,正确接收的分组中的字节的数目,TSi以字节为单位表示时隙长度。 Here, BiR represents the number of packets in the time slot, the correctly received bytes, TSi represents the slot length in bytes. 从而,时隙i的分组接收差错值的估计值 Thus, the estimated value of the packet reception error value for time slot i 被计算为p~iC={niR/niT;(niT_known)BiR/TSi;(niT_not_known)0;(no_pkis_received)...(23)]]>开始于方框200,该方法可随意包括在方框201,利用分组接收差错值来确定/更新分配给链路Lk的每个时隙i的定向链路质量值(LQV)N_qualityiLk。 Is calculated as p ~ iC = {niR / niT; (niT_known) BiR / TSi; (niT_not_known) 0; (no_pkis_received) ... (23)]]> begins at block 200, the method can easily be included in the box 201, the use of packet reception error value to determine / update directional link quality value is assigned to each slot of the i link Lk (LQV) N_qualityiLk. 对于信号出现时间中的每个时隙,每个信号出现时间这些值被更新一次。 For epochs in each time slot, each epoch time these values are updated once. 在每个时隙之后,分配给链路i的定向链路的N_qualityiLk值被更新为:N_qualityiLk=(1-β)·N_qualityiLk+β·p~iC...(24)]]>在时隙被分配给来自节点k的链路之后的第一个接收时隙之前,该链路的N_qualityiLk值的初始条件被设置为1。 After each time slot i allocated to a directional link links the N_qualityiLk values are updated as: N_qualityiLk = (1- & beta;) & CenterDot; N_qualityiLk + & beta; & CenterDot; p ~ iC ... (24)]] > before the first receive timeslots are allocated to the link from node k after the initial conditions N_qualityiLk value of the link is set to 1. 随后对于该信号出现时间的时隙i之后的每个后续信号出现时间,根据(24)更新N_qualityiLk值。 The signal is then for each subsequent epoch after the time slot i occurrence time, according to the update N_qualityiLk value (24).

参数β的选择有效地设置相对于估计成功分组传输的概率的一阶滤波器的定向链路信号出现时间的长度的时间常数。 The selection parameter β effectively sets a time constant with respect to the estimated length of a directional link signal packet transmission success probability of a first order filter of the time of emergence. 参数β的正确设置是所需的响应性和估计精度之间的折衷。 Set parameter β is a compromise between the desired responsiveness and the accuracy of estimation between. 较大的β值导致信号出现时间内的较大可能变化。 Large β values result in large changes in time of the signal may occur. 这以增大错误分类链路状态的可能性为代价,更快速地识别链路状态方面的变化。 This is in order to increase the likelihood of misclassification link state expense, more quickly identify changes in link state area.

一个问题是在确定链路状态发生变化之前,应允许过去多少时间(以信号出现时间为数)。 One problem is that before a determination on the link status changes, how much time should be allowed in the past (in epochs number). 在一段时间之后,如果时隙i变得非常差,那么N_qualityiLk≈0.]]>类似地,如果时隙i已变得非常好,那么N_qualityiLk≈1.]]>困难在于判断良好和低劣之间的转变,和何时宣告应重新分配时隙。 After a period of time, if the time slot i became very poor, so N_qualityiLk & ap;. 0]]> Similarly, if the time slot i has become very good, so N_qualityiLk & ap;. 1]]> The difficulty is that the judgment of good and poor change between and when declared should be re-allocated slot.

一种特别有利的完成所述确定的方法是使用滞后。 A particularly advantageous method of completing the determination is to use hysteresis. 根据本发明的这一方面,一般来说,在方框201,目的节点根据分组接收差错值确定与通信链路相关的链路质量值N_qualityiLk。 According to this aspect of the invention, in general, at block 201, a communication link with the destination node determines the link quality value associated N_qualityiLk The packet reception error value. 如果链路质量值N_qualityiLk低于第一阈值,那么在方框203,源节点和目标节点将在该时隙内停止使用所述通信链路。 If the link quality value N_qualityiLk lower than the first threshold value, then at block 203, the source node and the destination node will stop using the time slot within the communication link.

此外,在方框204,如果链路质量值N_qualityiLk持续预定的时间仍然低于第一阈值(在其它实施例中也可使用其它阈值),那么,源节点和目的节点可确定新的时隙,以便在它们之间建立无线通信链路,从而结束图解说明的方法(方框206)。 In addition, at block 204, if the link quality value N_qualityiLk predetermined time duration is still below a first threshold value (in other embodiments, other thresholds may be used), then the source node and the destination node may determine a new time slot, In order to establish a wireless communication link between them, thus ending the illustrated method (Block 206). 另一方面,在方框207,如果链路质量值N_qualityiLk先前已低于第一阈值,并且链路的使用已被停止,那么在方框208和209,如果链路质量值增大到高于第一阈值的第二阈值之上,源节点和目的节点将在该时隙内继续使用该通信链路。 On the other hand, at block 207, if the link quality value N_qualityiLk previously been below a first threshold, and use of the link has been stopped, then at block 208 and 209, if the link quality value increases above a above the second threshold value of the first threshold, the source node and the destination node will continue using the communication link during the time slot.

举例来说,如果在时隙i内,来自节点k的链路的N_qualityiLk值的范围为0-1,那么对于本例来说,我们将借助某些网络控制功能的滞后,将其量化成三个水平。 For example, if the time slot i, the value range N_qualityiLk link from node k is 0-1, then for the present example, we will build lag certain network control functions, to be quantized into three levels. 变量TS_QualiLk将代表N_qualityiLk变量的量化值。 Variables representing the quantized value TS_QualiLk N_qualityiLk variables. 新变量可以取值为代表三种链路质量状态的GOOD、MID和BAD。 The new value of the variable can represent three link quality status of GOOD, MID, and BAD. 每次修改N_qualityiLk时,可根据下面所示修改TS_QualiLk变量。 Each time you modify N_qualityiLk, according to modify TS_QualiLk variables shown below.

N_qualityiLk>TQH;]]>设置TS_QualiLk=GOOD;...(25)]]>TS_QualiLk=GOOD]]>并且如果TQM≤N_qualityiLk≤TQH,]]>那么设置TS_QualiLk=GOOD;...(26)]]>TS_QualiLk=GOOD]]>并且如果TQL≤N_qualityiLk≤TQM,]]>那么设置TS_QualiLk=MID;...(27)]]>TS_QualiLk=MID]]>并且如果TQL≤N_qualityiLk≤TQH,]]>那么设置TS_QualiLk=MID;...(28)]]>TS_QualiLk=BAD]]>并且如果TQM≤N_qualityiLk≤TQH,]]>那么设置TS_QualiLk=MID;...(29)]]>TS_QualiLk=BAD]]>并且如果TQL≤N_qualityiLk≤TQM,]]>那么设置TS_QualiLk=BAD;...(30)]]>N_qualityiLk=TQL;]]>设置TS_QualiLk=BAD;...(31)]]>如同本领域的技术人员将理解的那样,在上述关系(25)-(31)中引入滞后,以便防止估计的时隙状态的过度波动。 N_qualityiLk> TQH;]]> Set TS_QualiLk = GOOD; ... (25)]]> TS_QualiLk = GOOD]]> And if TQM & le; N_qualityiLk & le; TQH,]]> then set the TS_QualiLk = GOOD; ... (26) ]]> TS_QualiLk = GOOD]]> And if TQL & le; N_qualityiLk & le; TQM,]]> then set the TS_QualiLk = MID; ... (27)]]> TS_QualiLk = MID]]> And if TQL & le; N_qualityiLk & le; TQH,] ]> then set TS_QualiLk = MID; ... (28)]]> TS_QualiLk = BAD]]> And if TQM & le; N_qualityiLk & le; TQH,]]> then set the TS_QualiLk = MID; ... (29)]]> TS_QualiLk = BAD]]> And if TQL & le; N_qualityiLk & le; TQM,]]> then set the TS_QualiLk = BAD; ... (30)]]> N_qualityiLk = TQL;]]> Set TS_QualiLk = BAD; ... (31)] ]> As those skilled in the art will appreciate, in the above relation (25) - (31) introducing hysteresis to prevent excessive fluctuations in the estimated time slot states. 阈值的例证数值是TQL=0.2,]]>TQM=0.5,]]>TQH=0.8,]]>不过根据指定应用,也可使用其它数值。 Illustration value threshold value is TQL = 0.2,]]> TQM = 0.5,]]> TQH = 0.8,]]> However, according to a given application, other values may also be used. 这里应注意虽然为了方便起见,这里使用的例证质量范围从0到1,不过也可使用其它范围。 Note here that although for convenience of illustration herein mass range from 0 to 1, but other ranges may also be used. 当对于指定时隙,TS_QualiLk=BAD]]>时,在该时隙内,调度器最好不通过链路传送任何通信,除非不存在可用于具有MID或GOOD质量的链路的其它时隙。 When for a given time slot, TS_QualiLk = BAD]]> when, in the time slot, the scheduler preferably does not transmit any traffic over the link, may be used unless there is no other time slots having MID or GOOD quality link.

另外应注意如上所述的指数加权的平均值的使用适合于测量全向链路和定向链路的链路质量。 Also it should be noted as described above using the weighted average of the index suitable for measuring the quality of the link to the link and full directional link. 存在能够导致这两种链路中任意之一中断的各种链路损害。 Damage can lead to the presence of any one of the various links of the two interrupt link. 最好关于这两种链路类型,持续测量链路质量,还应使用各种机制对链路质量的改变作出反应。 The best of these two link types, continuous measurement of link quality, should also be used to change a variety of mechanisms to respond to the link quality.

本领域的技术人员会认识到,各种现象会影响链路质量。 Those skilled in the art will recognize, various phenomena may affect link quality. 例如,两个节点之间的距离可能变得如此之大,以致失去视线(LOS)。 For example, the distance between two nodes may become so large that the loss of sight (LOS). 这种情况下,链路被丢失,直到在未来某一时间,节点在距离方面变得更近为止。 In this case, the link is lost until at some future time, the nodes become closer in distance it up. 如果节点移动到障碍物之后,并且长时间被遮蔽,那么观察到相同的效果。 If, after node moves to the obstacle, and a long time is obscured, then the same effect is observed. 这种情况下,全向链路和定向链路都具有很低的质量。 In this case, omni and directional links will have very low quality links. 这最好会触发半永久时隙分配的取消和/或OLSR路由中链路的消除。 This will trigger the best semi-permanent time slot allocated to cancel and / or elimination of OLSR routing link.

另一种现象是链路可在良好质量和低劣质量之间快速摆动(也称为“摇摆”)。 Another phenomenon is that a link may rapidly oscillate (also known as "swing") between good quality and poor quality. 当节点快速通过小型障碍物时会发生这种现象。 When a node quickly through the small obstacles occur this phenomenon. 这种情况下,全向链路和定向链路都会摇摆。 In this case, the full swing and directional links will forward link. 如果利用过小的样本尺寸估计链路质量,也可能发生这种现象。 If you use too small a sample size to estimate link quality, this phenomenon may also occur. 这种情况需要稍微更仔细地处理。 This situation requires a little more careful handling. 在这种情形持续的时候,链路可能过差,以致不能被用于发送通信。 In this case, when continued, the link may be too poor that it can not be used to send a communication. 但是,时隙分配可被保持一定时间,以便确定高质量链路是否可被恢复。 However, the slot assignment can be to maintain a certain time, in order to determine whether the link quality can be restored.

此外,在特定时隙中的定向链路质量较低时,全向链路可具有高质量。 Additionally, when a directional link quality in particular lower slot, the omni-directional link may have a high quality. 在分配的其它时隙中,与另一节点的定向链路上的通信也能够是高质量的。 In other assigned time slot, the communication link to another node on the orientation and can also be of high quality. 这表示存在来自正在低质量的单一定向时隙中工作的其它用户的特定干扰。 This indicates the presence of other interfering user specific single directional time slot from being of low quality in work. 间歇性干扰也能够导致在该时隙中链路的摇摆。 Intermittent interference can lead to swings in the time slot of the link. 这种情况下,时隙应被重新调度以消除干扰。 In this case, the slot should be rescheduled to eliminate the interference. 如果其它时隙可用或者在通信被缓存的时候,能够快速获得新的时隙分配,那么仍然能够发生路由。 Or if other time slots are available in the cache when communication is possible to quickly obtain a new time slot assignment, then the route can still occur.

响应上述现象,可采用几种动作。 In response to the above phenomenon, several actions may be employed. 例如,在链路层,可确定在单一中继段中不再能够到达相邻节点。 For example, at the link layer, may be determined in a single hop is no longer able to reach the neighboring node. 这种情况下,近邻表被更新以反映该新状态,任何定向时隙被解除分配并被标记为“自由”。 In this case, neighbor tables are updated to reflect the new state, and any directional time slots are de-allocated and marked as "free." 如果在所述近邻在范围内的时候,由于干扰过大的缘故,某一时隙被宣告不良,那么链路调度协议将尝试分配无干扰的新时隙。 If the neighbor is in the range of time, due to the interference of the reason is too large, a time slot is declared bad, then the link scheduling protocol will try to allocate a new time slot without interference.

在路由层,不必把链路层的这些动作中的一些通知路由协议,除非它们影响连通性。 In the routing layer, these actions do not have to link layer routing protocols in some notification, unless they affect connectivity. 如果两个节点之间的时隙被重新调度,但是这两个节点仍然被连接并且能够交换任务数据,那么通过该链路的路线仍然是良好的。 If the slots between two nodes are rescheduled but the two nodes are still connected and can exchange mission data, the routes through that link are still good. 如果某一相邻节点不再是可达到的,那么路由协议应找到不利用该链路的新路线。 If a neighboring node is no longer attainable, then the routing protocol should find a new route does not take advantage of the link.

现在转到干扰的问题,本领域的技术人员会认识到在以时隙复用为目标的任何调度算法中,都可能发生干扰。 Turning now to the problem of interference, those skilled in the art will recognize that any scheduling algorithm in the target time slot reuse, the interference may occur. 一般来说,根据本发明提供两种用于降低指定时隙中,来自其它用户的潜在和实际干扰的策略。 In general, provided both for reducing the specified timeslot, the potential and actual interference from other users of the policy according to the present invention. 一种干扰回避策略用于初始时隙调度和某些抢先干扰回避重新调度。 A distraction avoidance strategy for initial time slot scheduling and some preemptive interference avoidance rescheduling. 干扰减轻策略用于回避策略不能及时起作用以防止实际干扰的情况。 Interference mitigation strategies for avoiding policy does not act promptly to prevent actual interference. 干扰减轻策略将产生重新调度时隙,从而消除干扰的效果。 Interference mitigation strategies will produce rescheduled slots, thus eliminating the interference effect.

图21中图解说明了例证的干扰避免情形。 Illustrated in Figure 21 illustrates examples of the interference avoidance scenario. 在该情形下,在为对相邻移动节点n的传输调度时隙之前,节点m评估可能的干扰。 In this case, the transmission schedule for the adjacent slot before the mobile node n, node m evaluate possible interference. 如上所述,所述评估可由从节点n到节点m的在它们之间建立初始/附加时隙的请求引起。 As described above, the evaluation causing establish initial / additional time slot therebetween may request from node n to node m. 节点m能够搜索它的通过定向近邻分组接收的信息的表格,确定其相邻节点中的哪一个在相同时隙中安排了传输。 Node m can search its neighbor table by directional information received packet, to determine which of its neighbor nodes in a transmission arranged in the same time slot. 在举例说明的例子中,节点k和j是利用相同时隙的潜在干扰节点(图解表示为空心圆)。 In the example illustrated, nodes k and j are potentially interfering nodes using the same time slot (graphical representation as open circles).

更具体地说,在节点m的来自节点k的干扰经过距离dkm,并以偏离节点m用于指向节点n的天线段的视轴(由箭头210指示)θmk的角度被接收。 More specifically, the interference from node m from node k through dkm, and departing from node m to node n for the boresight (indicated by arrows 210) θmk angle received by the antenna section. 另外,当向节点j发射时,在节点k以偏离天线的视轴(由箭头211指示)θkm的角度朝着节点m发射潜在干扰信号。 Further, when transmitting to node j, node k to the deviation from the antenna boresight (indicated by arrows 211) θkm angle toward node m transmit potentially interfering signals.

可根据几个因素确定所产生的干扰是否足够强,足以使节点m为到节点n的传输调度该时隙。 Several factors may be determined according to whether the generated interference is strong enough, is sufficient to node m to the node n of time slots to the transmission schedule. 现在参见图22,一种方法首先(方框220)在方框222确定相邻移动节点n和在该时隙内发射的其它潜在干扰移动节点(这里,节点j和k)的相对位置。 Referring now to Figure 22, a first method (block 220) determines whether the mobile node n and other potentially interfering mobile nodes transmitting during the time slot (here, the nodes j and k) the relative position of the adjacent block 222.

随后在方框222,根据这些相对位置确定对节点m和n之间的定向通信链路的潜在干扰。 Then at block 222, to determine potential interference with the directional communication link between nodes m and n of these relative positions. 确定所述潜在干扰的一种方法是根据到所述相邻移动节点n和潜在干扰移动节点k、j的相应距离和角度,计算位置,如下进一步所述。 A method for determining the interference potential is based on the mobile node n and the potentially interfering mobile nodes k to the adjacent, j corresponding distances and angles, calculates the position, as described further below. 随后在方框224,只有当干扰低于阈值时,才在该时隙内调度定向通信链路,从而结束图解说明的方法(方框225)。 Then at block 224, only if the interference is below the threshold, the scheduling directional communication link during the time slot, thereby ending the method (block 225) illustrated.

除了节点的相对位置之外,还可根据可由节点m使用的抑制算法或滤波器在天线旁瓣实现的抑制,确定潜在干扰。 In addition to the relative position of nodes, but also in the antenna side lobe suppression achieved, according to identify potential interference suppression algorithm or filter used by node m. 即,干扰信号可被距离dkm的路径损耗和在节点m和k的天线旁瓣抑制到各不相同的程度。 That is, the interference signal may be path loss and distance dkm nodes m and k in the antenna side lobe suppression to different extent. 这将由关于每个潜在的干扰节点计算的度量MK,TiI量度。 It will be on the metric for each potentially interfering node calculation of MK, TiI measure. 该度量时隙Ti中从节点k接收的干扰水平(单位为dB)。 The measure of the level of interference received from Ti slot node k (in units of dB).

在它们的HELLO和定向近邻更新分组中从相邻节点接收的信息现在将允许计算到节点的距离和在每个时隙中用于传输的波束的视轴方向。 In their HELLO and Directional neighbor update packet received from the neighboring node information will now be calculated to allow the visual axis direction and the distance between nodes in each time slot for transmission of beams. 对于本例来说,我们将假定所有节点具有作为离视轴的角度的函数的天线增益模式,由函数Ga(θ)给出。 For this example, we will assume that all nodes have as a function of the angle from the boresight of the antenna gain pattern is given by a function Ga (θ). 如果指定环境中的准确传播损耗未知,那么可估计作为dnp的逆函数的损耗,其中d是从干扰节点k到关心的节点的距离,np是一般在2-4的范围中的恒定传播损耗指数。 If you specify the exact propagation loss in the environment is unknown, it can be estimated as an inverse function of dnp loss, where d is the distance from the interfering node k to the node of interest, np is generally in the range of 2-4 in the constant propagation loss exponent .

作为例证模型,路径衰减可被计算成由地平面衰减等式和自由空间衰减等式给出的路径衰减的最大值。 As an example model, the path attenuation can be calculated from ground level to the maximum attenuation equation and the equation given free space attenuation path attenuation. 对于地平面等式来说,衰减为:Lkmp=40·log(dkm)-20·log(hi)-20·log(hr),...(32)]]>其中路径长度dkm(从节点k到节点m的距离)和发射器及接收器的相应高度ht、hz的单位都为米。 For the ground plane equation, the attenuation is: Lkmp = 40 & CenterDot; log (dkm) -20 & CenterDot; log (hi) -20 & CenterDot; log (hr), ... (32)]]> where path length dkm (from node distance corresponding to the height ht k to node m) and the transmitter and the receiver, hz of units of meters. 自由空间衰减等式给出的路径衰减为:Lkmfs=32.45+20·log(dkm)-20·log(f),...(33)]]>其中RF射频f的单位为GHz。 Free space attenuation path attenuation is given by the equation: Lkmfs = 32.45 + 20 & CenterDot; log (dkm) -20 & CenterDot; log (f), ... (33)]]> where f is the RF GHz. 估计的总路径损耗为两者的最大值,即:Lkmtot=max(Lkmp,Lkmfs)....(34)]]>本领域的技术人员会认识到,实际的损耗将是时变的,并且由于障碍物或树叶的缘故,明显不同于该计算结果,但是该恒定值可被用作评估潜在干扰源的影响中的合理近似。 The total estimated path loss is the maximum value of the two, namely: Lkmtot = max (Lkmp, Lkmfs) .... (34)]]> Those skilled in the art will recognize, the actual loss will be time-varying, and because of obstructions or foliage sake, significantly different from the calculation results, but this constant value can be used to assess the impact of potential sources of interference in a reasonable approximation. 从而可根据对于时隙Ti内从节点k发射的信号,在来自潜在干扰节点k的路径上的节点m的接收天线的信号损耗,把理论度量计算为:Mk,TiI=-Lkmtot+Ga(θkm),...(35)]]> Thereby according to the signal for time slot Ti transmitted from node k, the signal loss at the receiving antenna on the path from the potentially interfering node k to node m, the theoretical metric calculated as: Mk, TiI = -Lkmtot + Ga (& theta ; km), ... (35)]]>

其中θkm是朝向节点m的信号方向的偏离节点k的视轴的角度。 Wherein θkm is the angle deviation from the node k of the signal direction toward node m boresight. 该度量可在节点m被用于计算干扰影响以便把时隙Ti分配给任何其它相邻节点。 This metric can be used to calculate the interference at node m to the assigned time slot Ti to any other neighbor nodes.

上述度量没有体现在节点m的干扰信号的旁瓣抑制的影响,因为这取决于正在考虑在时隙Ti内调度哪个相邻节点。 Said metric does not reflect an interference signal sidelobe suppression of node m in the impact, because it depends on which neighbor node is being considered in the scheduling time slot Ti. 但是,当为该时隙考虑任何相邻节点,例如节点n时,可根据Mk,TiI计算包括在节点m的旁瓣抑制的新度量,如下所示:Mk,TiI,mn=Mk,TiI+Ga(θmk)....(36)]]>从而在节点m接收的来自在时隙Ti中通信的节点对k、j之间的链路L的干扰的最大干扰水平由从节点k或者节点j接收的最大水平给出,即:NL,Timn=maxk,j(Mk,TiI,mn,Mj,TiI,mn)....(37)]]>这样,时隙Ti中在节点m收到的总干扰为从在时隙Ti中工作的所有链路L接收的干扰的总和,即,NTimn=10·log[ΣL10NL,Timn/10]....(38)]]>多数情况下,该量值的近似值就足够了。 However, when considering any neighbor node for this time slot, such as node n, according to Mk, TiI calculated including the new metric of node m in the sidelobe suppression, as follows: Mk, TiI, mn = Mk, TiI + Ga (& theta; mk) .... (36)]]> whereby node m in time slot Ti received from a node in the communication link maximum interference level for interference k, j L between the slave node by k or node j receives the maximum level is given, namely: NL, Timn = maxk, j (Mk, TiI, mn, Mj, TiI, mn) .... (37)]]> Thus, in time slot Ti total interference is the sum of node m receives interference from all links L operating in time slot received and Ti, i.e., NTimn = 10 & CenterDot; log [& Sigma; L10NL, Timn / 10] .... (38)] ]> In most cases, the magnitude of the approximation is sufficient. 就任意时隙中的少量潜在干扰源来说,两个或更多的干扰源不可能具有几乎相等的功率。 On a small number of potential sources of interference for any time slot, two or more sources of interference can not have nearly equal power. 这种情况下,我们可把干扰的近似值计算为:NTimn≈maxL(NL,Timn)....(39)]]>从而,由于实际的信号-干扰比将高度取决于到所需目标节点的距离与到干扰节点的距离的比值,我们可把时隙Ti的理论SIR度量计算为:SIRmnTi=NTimn+Lmntot....(40)]]>上述度量的单位为dB,它们可被用于在最初为任何相邻节点调度时隙时提供干扰回避。 In this case, we can calculate the approximate value of interference as: NTimn & ap; maxL (NL, Timn) .... (39)]]> Therefore, since the actual signal - to-interference ratio to a target node will be highly dependent on the desired the distance between the ratio of the distance to the interfering node, we can put the theoretical SIR metric for time slot Ti calculated: SIRmnTi = NTimn + Lmntot .... (40)]]> said units of measure dB, which may be used to provide interference avoidance in initially scheduling slots for any adjacent nodes. 另外,有利的是它们可被用于间歇检查,确定当前调度的时隙是否正在开始经历或者可能经历增大的干扰水平。 Further, it is advantageous that they can be used intermittently checking to determine whether the time slot is currently scheduled start experiencing or likely to experience an increased level of interference. 图23中图解说明了该方法,所述方法开始于方框230,在方框231,节点m间歇地确定相邻移动节点n和其它潜在干扰节点的相对位置。 Figure 23 illustrates that the method begins at block 230, at block 231, the relative location of the mobile node n and other potentially interfering nodes of node m intermittently determining adjacent. 应注意由于节点是移动的,因此在这期间,潜在干扰节点可能已发生变化。 It should be noted since the node is moving, so during this period, a potential interfering node may have changed. 节点j和k可能仍然是潜在干扰节点也可能不再是潜在干扰节点,在这期间,其它潜在干扰节点可能已进行节点m的范围内。 Nodes j and k may still be potentially interfering nodes may no longer be potentially interfering nodes, during which the other potentially interfering nodes may have be in the range of node m.

无论怎样,一旦确定了对于新的潜在干扰节点的新的相对位置,那么如前所述,在方框232,据此确定对定向通信链路的潜在干扰。 In any case, once the new potentially interfering nodes for the new relative position, as previously described, at block 232, accordingly determine potential interference to the directional communication link. 如果在方框233,潜在干扰高于上面提及的干扰阈值,那么在方框234,在新时隙内调度定向通信链路,从而结束图解说明的方法(方框235)。 If at block 233, potential interference above the interference threshold mentioned above, then 234, scheduling directional communications link in a new time slot in the block, thereby ending the method (block 235) illustrated. 当然,如上所述,可以使用不同的阈值以便引入滞后。 Of course, as mentioned above, you can use different thresholds in order to introduce hysteresis.

上述干扰等式为当确定时隙的恰当分配时最好考虑的时隙优先化提供一种有用的模型。 When the above-described interference equations appropriate for the slot allocation determination preferable to consider the time slot prioritization provides a useful model. 当然,实际上每次要确定干扰时,计算每个上述度量需要大量的处理资源。 Of course, in fact, every time you want to determine the interference is calculated for each of the measure requires a lot of processing resources. 这样,如同本领域的技术人员会认识到的那样,预先计算输入/输出组的一组范围或者比值,并把它们保存在控制器18中的查寻表18g中以便计算SIR是有利的。 Thus, as those skilled in the art will recognize that the above, calculated in advance a set of input / output range or ratio of the group, and saves them in the controller 18 in the look-up table 18g in order to calculate the SIR is advantageous.

现在参见图24,更详细地说明SP和DA时隙再分配。 Now 24, SP and DA time slot reallocation described in further detail see Fig. 首先,我们考虑节点m和其相邻节点n之间新的SP或DA时隙的初始调度(参见图21)。 First, we consider a new SP or DA time slot between the node m and the initial scheduling of its neighbor nodes n (see Figure 21). 首先,这里假定当分配新的SP时隙时,任何“自由”时隙或DA时隙可用于分配,同时“自由”时隙的优先级高于DA时隙。 First, assuming the new SP when assigning time slots, any "free" time slot or the DA slots available for allocation, and the "free" time slot has priority over the DA slots. DA时隙最好也被排序,需求度量值最小的DA时隙的再分配优先级最高。 DA slot at best be sorted, the highest demand metric minimal redistribution DA slot priority. 除了如上所述,需求度量大于某一数量的现有时隙分配不可用于再分配之外,当分配新的DA时隙时,遵循类似的进程。 Except as described above, the demand metric larger than a certain number of time slot assignment is available for redistribution of existing outside when a new DA time slot allocation, follow a similar process.

如上所述,通过把估计的干扰水平和优先化的通信需求度量看作优先化度量,还可扩大该初始调度进程。 As described above, the estimated interference level and prioritize communications needs of a measure seen as a priority measure, but also to expand the initial scheduling process. 从而,就这两种度量来说,对潜在时隙排序的进程更复杂。 Thus, on these two metrics, the potential slot sorting process is more complicated.

一般来说,时隙的重新分配开始于方框240,随后如上所述,在方框241确定初始时隙。 Generally speaking, re-allocation of time slots begins at block 240, then as described above, at block 241 to determine the initial slot. 根据本例,在方框242,发起移动节点n将识别与其相邻移动节点共用的可用时隙,并在方框243,根据发起移动节点和相邻移动节点在时隙内的链路使用率,对可用时隙排序。 According to the present embodiment, at block 242, initiating mobile node n will identify the mobile node common to adjacent time slots available, and at block 243, the originating mobile node and neighboring mobile nodes within a slot link utilization sorting the available time slots. 随后在方框244,发起移动节点n把包括可用时隙及其排序的时隙分配请求传送给接收移动节点m。 Then in block 244, initiating mobile node n to include the sort of available time slots and slot allocation request is transmitted to the receiving mobile node m.

这样,在方框245,接收移动节点m将依次接收时隙再分配请求,并根据接收移动节点和与接收移动节点相邻的移动节点在可用时隙内的链路使用率,对可用时隙排序。 Thus, at block 245, the receiving mobile node m will in turn receive the time slot reallocation request, and according to the mobile node and the receiving mobile node and the receiving mobile node adjacent link utilization within the available time slot, the available time slots sorting. 此外,在方框246,节点m随后根据发起移动节点和接收移动节点的可用时隙的排序,产生可用时隙的组合排序。 In addition, at block 246, node m then initiating mobile node and the receiving mobile node of available timeslots is sorted, ordering the available time slots to produce a combination. 随后在方框247,接收移动节点m可根据组合排序,重新分配一个或多个可用时隙,以便在发起移动节点和接收移动节点之间建立通信链路,从而在方框248结束图解说明的方法。 Then at block 247, the receiving mobile node m may sort according to the combination, reassign one or more available time slots to establish a communication link between the initiating mobile node and the receiving mobile node, at block 248 so that the end of the illustrated Methods.

参考图25,在一些实施例中,上述再分配方法还可有利地包括各种其它特征。 With reference to Figure 25, in some embodiments, the above-described redistribution method may advantageously include various other features. 开始于方框250,在方框251,如上所述,发起移动节点n和接收移动节点m均可确定与每个可用时隙相关的通信需求度量,并在方框252和253,把与之相关的通信需求度量高于较高的使用阈值的任何可用时隙排队在再分配的考虑之外。 Begins at block 250, at block 251, described above, the initiating mobile node n and receiving mobile node m can be determined for each available time slot associated with the communication demand metric, and at block 252 and 253, with the any available communication time slot demand metric is higher than a high usage threshold queuing outside redistribution consideration.

类似地,在方框254,发起移动节点n和接收移动节点m还都可确定与每个可用时隙相关的SIR值,并在方框255和253,把与之相关的SIR值高于较高的SIR阈值的任何可用时隙排队在再分配的考虑之外。 Similarly, at block 254, initiating mobile node n and receiving mobile node m can also determine the available time slots associated with each SIR value, and at block 255 and 253, the associated SIR value than in the more high SIR threshold any available time slots queuing outside redistribution consideration. 另外应注意在时隙的排序中,发起移动节点n和接收移动节点m还使用SIR值(图24中的方框243、245)。 Also to be noted in the slot ordering, initiating mobile node n and receiving mobile node m also use the SIR value (block 243, 245 in FIG. 24).

如前所述,这些排序还可考虑到增大的和/或降低的链路使用度量。 As mentioned earlier, the sorting may take into account the increased and / or decreased link utilization metrics. 此外,同样如上所述,发起移动节点n和接收移动节点m均可根据其中传送的数据的平均数量和/或保存在它们的队列18f中的数据的数量,对可用时隙排序。 Further, similarly as described above, initiating mobile node n and receiving mobile node m can be based on the average number of transmission in which the number of data and / or stored in their queues 18f data, sorting the available time slots.

此外,在方框256和253,由于发起移动节点n和接收移动节点m最好传送具有多个优先级的数据,因此如果将在可用时隙内传送的数据并不具有和在所述可用时隙内当前正被传送的数据相等或者更多的优先级,那么接收移动节点m能够禁止可用时隙的重新分配。 In addition, at block 256 and 253, due to initiating mobile node n and receiving mobile node m is preferably transmitted data having a plurality of priorities, and therefore if the data to be transmitted within the available time slot does not have and is available in the Data is currently being transmitted within the gap equal to or more priority, then the receiving mobile node m can prohibit re-allocation of available time slots. 否则,如前所述,链路可被重新分配(方框257),从而在方框258结束图解说明的方法。 Otherwise, as described above, the link can be re-assigned (block 257), thereby ending the method at block 258 illustrated.

更具体地说,现在将提供实现时隙的上述分级的例证算法。 More specifically, the classification will now be provided to achieve the above examples slot algorithm. 但是,在给出对可被重新分配给节点m和n之间的链路的潜在时隙评定等级的算法之前,我们首先定义项MetLimmn,以确保没有任何时隙将从另一链路被重新分配给节点m和n之间的链路,除非需要传送更高优先级的通信为止,如上参考方框256所述。 However, before that can be given to re-assigned to the link between nodes m and n potential slots rating algorithm, we first define the term MetLimmn, to ensure that there is no link to another time slot will be re assigned to the link between nodes m and n, unless a higher priority needs to be transmitted until the communication, as described above with reference to the block 256. 同样采用具有三个优先级类别的例子,被定义为MetLimmn:IfMmnDA(TmnPj,Nmntot)/Ks>2thenMetLimmn=2,...(41)]]>ElseifMmnDA(TmnPj,Nmntot)/Ks>1thenMetLimmn=1,...(42)]]>Else MetLimmn=0. (43)为了举例说明整个方法,首先研究当不考虑干扰度量时,对时隙排序的方法。 Using the same example with three priority categories are defined as MetLimmn: IfMmnDA (TmnPj, Nmntot) / Ks> 2thenMetLimmn = 2, ... (41)]]> ElseifMmnDA (TmnPj, Nmntot) / Ks> 1thenMetLimmn = 1 , ... (42)]]> Else MetLimmn = 0. (43) To illustrate the overall process, when the research does not consider the first measure of interference, time slots sorting method. 首先根据下述时隙优先化方法,对未被分配给节点m和n的所有其它时隙(即,在节点m和其未示出的相邻节点r之一之间分配的那些时隙)排序,所述时隙优先化方法考虑了分配新时隙中的通信需求的优先级,但是不包括干扰的影响。 First, according to the following time slot prioritization method, is not assigned to the nodes m and n all other time slots (i.e., those time slots between node m and one of its neighbor nodes r, not shown, assigned) sorting, the time slot prioritization method considers the new time slot allocation priority of communication needs, but does not include the effect of interference.

特别地,当要求在节点m和n之间分配一个新时隙时,根据自由时隙和DA时隙产生可用时隙的列表。 Particularly, when the request to allocate a new time slot between the node m and n, generates a list of available time slots and free time slots in accordance with the DA slots. 随后在把潜在时隙的列表发送给接收节点m之前,在请求或发送节点n对这些时隙排序。 Then send the list of potential time slots to the receiving node before m, sort request or sending node n in these slots. 利用优先化通信需求度量MmrDA(TmrPj,(Nmrtot-1))对可用时隙排序。 Utilizing prioritized communication needs metric MmrDA (TmrPj, (Nmrtot-1)) to sort the available time slots. 注意该度量中(Nmrtot-1)的使用,如果从节点m和r之间的链路取走某一时隙,那么它指示所述度量的值。 Note that the metrics (Nmrtot-1) is used, if a slot is removed from the link between nodes m and r, it indicates that the value of the metric.

下述方法被推荐用于所述排序。 The following method is recommended for the sorting. 等级最高的时隙是自由的时隙。 The highest level of slots are free slots. 按照通信需求度量MmrDA(TmrPj,(Nmrtot-1))的逆序,对剩余的时隙排序。 According to the communication needs of metrics MmrDA (TmrPj, (Nmrtot-1)) of reverse, sort the remaining slots. 如果要分配的新时隙是DA时隙,那么除去传送相等或更高优先级通信的所有时隙,即MmrDA(TmrPj,(Nmrtot-1))>MetLimmr....(44)]]>这确保在重新分配进程中,只有低优先级的通信被丢失。 If you want to assign a new time slot is DA slot, then remove the transfer of equal or higher priority for all time slots for communication, namely MmrDA (TmrPj, (Nmrtot-1))> MetLimmr .... (44)]]> This ensures that the redistribution process, only the low-priority traffic is lost.

如上所述,(根据上述标准的)时隙的排序列表包括在请求节点n发送的REQ消息中。 As described above, (according to the above criteria) sorted list of time slots comprises REQ message sent by the requesting node n. 接收节点m接收REQ消息,获得排序时隙的列表,并通过利用优先通信需求,根据上述方法对这些时隙排序。 The receiving node m receives the REQ message, to obtain a list of sorting time slot, and by using the preferential communication needs, according to the above method to sort these slots. 随后,根据这两个节点确定的排序,剩余的时隙被给予组合排序,同样如上所述,选择具有最佳组合排序的时隙。 Then, based on these two nodes to determine the sorting, the remaining time slots are given a combination of sorting, the same as described above, select the time slot has the best combination of sort.

如果度量MikDA(TikPj,(Niktot+1))>0,]]>这指示即使在分配第一时隙之后,还需要另一时隙。 If the metric MikDA (TikPj, (Niktot + 1))> 0,]]> This indicates that even after the first time slot allocation, but also the other time slots. 随后把该度量与到其它近邻的链路的度量进行比较,以了解是否应请求第二时隙,或者另一链路现在是否更迫切地需要一个额外的时隙。 Then sends the measure to measure with other neighboring links are compared to see whether they should request a second time slot, or whether another link is now more urgent need for an extra slot. 在后一情况下,下一时隙分配最好被给予具有最大度量的链路。 In the latter case, the next time slot allocation is preferably given to the link with the largest metric.

在一些实施例还可有利地增强上述方法,以便考虑到如上所述的通信优先级。 In some embodiments of the method described above may also advantageously be enhanced in order to take into account the priority of communication as described above. 下述方法被推荐用于对具有干扰的潜在时隙排序。 The following methods are recommended for potential slot has sort of interference. 为每个相邻节点和每个时隙保持接收的干扰水平Mn,TiI。 For each neighbor node and each time slot holds the received interference levels Mn, TiI. 这些干扰水平最好随着新的位置更新被定期更新,例如大约每秒一次。 These interference level best to update with the new location is updated on a regular basis, for example, about once per second.

当要求在节点m和n之间分配新时隙时,根据自由时隙和DA时隙创建可用时隙的列表。 When asked to assign a new time slot between nodes m and n, create a list of available time slots and slots under the Freedom of DA slots. 随后在向接收节点m发送潜在时隙的列表之前,在请求节点n对这些时隙排序或者区分优先次序。 Then before sending the list of potential time slots to the receiving node m, the requesting node n of these slots sort or prioritize. 从而,为了对时隙排序,在接收每个近邻的天线抑制之后,可评估总的接收干扰功率。 Thus, in order to sort the slot, after receiving antenna suppression for each neighbor can be assessed total received interference power. 本领域的技术人员会认识到,随后可计算估计的总的接收干扰功率NTimn。 Those skilled in the art will recognize, then calculate the estimated total received interference power NTimn.

随后可利用优先化的通信需求度量MmrDA(TmrPj,(Nmrtot-1))和干扰度量NTimn对分配给相邻节点的可用时隙排序。 Then be used to prioritize communication needs metric MmrDA (TmrPj, (Nmrtot-1)) and a measure of interference NTimn sorted allocated to adjacent nodes available slot. 注意该度量中(Nmrtot-1)的使用,它指示如果从节点m和r之间的链路取走某一时隙时,所述度量的值。 Note that the metrics (Nmrtot-1) is used, which indicates if a slot is removed from the link between nodes m and r, the metric value. 对于该排序来说,最好排序最前的时隙是自由时隙或者是具有小于预定阈值的优先化通信需求度量MmrDA(TmrPj,(Nmrtot-1))≤TDAL]]>的DA时隙。 For the sorting, the best ordering frontmost free slot or slots prioritized communication needs less than a predetermined threshold value metric MmrDA (TmrPj, (Nmrtot-1)) & le; TDAL]]> of the DA slots. 换句话说,这些时隙或者是未正在使用的(自由)时隙或者是几乎从未被使用的时隙。 In other words, these are time slots or are not being used (free) time slots or time slots are almost never used. 随后根据估计的干扰功率NTimns水平(NTimn的最低值被赋予最高等级),对这些时隙排序。 Then NTimns interference power level based on the estimated (minimum value NTimn been given the highest rating), sort of these slots.

按照等级排序的下一时隙是具有低于较高阈值的优先化通信需求度量MmrDA(TmrPj,(Nmrtot-1))≤TDAH,]]>并且同时对于该时隙来说具有很高的SIRmnTi(由超过某一阈值SIRmnTi>TSIRH]]>指示)的剩余DA时隙。 According to the ranking of a next slot is less than the higher threshold value having a priority of communication needs metric MmrDA (TmrPj, (Nmrtot-1)) & le; TDAH,]]> and also has high SIRmnTi for this time slot, (by more than a certain threshold SIRmnTi> TSIRH]]> instructions) DA remaining slots. 按照优先化通信需求度量对这些时隙排序,最小的需求度量指示最高的等级。 According to prioritize communications needs sorting measure of these slots, the smallest metric indicates the highest level of demand. 这些时隙被排列在先前计算的那些时隙之后。 These slots are arranged in the slot after those previously calculated.

按照SIRmnTi的最高值,对满足mmrDA(TmrPj,(Nmrtot-1))≤TDAH]]>但是不满足SIRmnTi>TSIRH]]>的剩余DA时隙排序。 SIRmnTi according to the highest value, and to satisfy mmrDA (TmrPj, (Nmrtot-1)) & le; TDAH]]> but does not satisfy SIRmnTi> TSIRH]]> remaining DA time slots sorted. 这些时隙被排列在先前计算的那些时隙之后。 These slots are arranged in the slot after those previously calculated. 此外,剩余的DA时隙被排列在先前计算的那些时隙之后。 Further, the remaining DA time slots are arranged in the slot after those previously calculated. 根据优先化通信需求度量对这些时隙排序,最小的优先化通信需求度量指示最高的等级。 Measure based on the priority of the communication requirements of these slots sort, prioritize communications needs of the smallest metric indicates the highest level. 当然,要认识到上述排序方法是例证性的,在本发明的范围内还可使用其它方法。 Of course, be appreciated that the above sorting method is illustrative, within the scope of the present invention may also use other methods.

如果时隙预测的信号-干扰比为:SIRmnTi<TSIRL....(45)]]>那么最好把先前的优先级排序中的任何时隙排除在再分配的考虑之外。 If the slot predicted signal - interference ratio: SIRmnTi & lt; TSIRL .... (45)]]> it is best to prioritize in the previous time slot exclude any consideration of redistribution. 这把对其来说,干扰节点可能导致过低的SIR的任何时隙排除在考虑之外。 This put it, the interfering node may result in too low SIR of any timeslot eliminated from consideration. 此外,如果要分配的新时隙是DA时隙,那么传送相同或更高优先级通信量的所有时隙最好被除去,即:MmrDA(TmrPj,(Nmrtot-1))>MetLimmr....(46)]]>这确保在再分配进程中,只有优先级较低的通信量被丢失。 In addition, if you want to assign a new time slot is DA slot, then transfer all slots are the same or higher priority traffic is best removed, namely: MmrDA (TmrPj, (Nmrtot-1))> MetLimmr ... . (46)]]> This ensures that the redistribution process, only lower priority traffic is lost.

如上所述,(根据上述标准的)时隙的排序列表包括在请求节点发送的REQ消息中。 As described above, (according to the above criteria) sorted list of time slots comprises REQ message sent by the requesting node. 接收REQ消息的接收节点m获得排序时隙的列表,并利用它计算的需求和干扰度量,按照如上所述的方法对这些时隙排序。 Receiving node m receives the REQ message to obtain the list of sorted time slot, and the use of demand and interference metrics it calculates, according to the method described above to sort these slots. 这最好包括除去未达到SIR、干扰节点距离和干扰节点角度阈值的任何时隙,或者在接收节点不可用的任何时隙。 This preferably includes removing any timeslot SIR, interfering node distance, and interfering node angle thresholds are not reached, or at any slot receiving node unavailable. 随后剩余的时隙被赋予由这两个节点确定的排序得到的组合排序,并选择具有最佳的组合排序的时隙。 Then the remaining time slots are given composition is determined by sorting the sort obtained both nodes, and selecting a time slot having the best combination sort.

如果MikDA(TikPj,(Niktot+1))>0,]]>这指示即使在分配第一时隙之后,仍然需要另一时隙。 If MikDA (TikPj, (Niktot + 1))> 0,]]> This indicates that even after the first time slot assignment, still needs another slot. 随后把该度量与到其它近邻的链路的度量进行比较,以了解是否应请求第二时隙,或者另一链路现在是否更迫切地需要一个额外的时隙。 Then sends the measure to measure with other neighboring links are compared to see whether they should request a second time slot, or whether another link is now more urgent need for an extra slot. 在后一情况下,下一时隙分配最好被给予具有最大度量的链路。 In the latter case, the next time slot allocation is preferably given to the link with the largest metric.

在分配某一时隙之后,作为干扰回避和减轻方法的一部分,将不断地监视该时隙。 After a certain time slot allocation, interference avoidance and mitigation as part of the method, will continue to monitor the time slot. 另外,不断测量全向链路质量。 Further, an omni-directional link quality is continually measured. 如果全向链路的质量变坏(指示LOS的损失或者到另一节点的距离过大),那么宣告该链路中断,并把链路的中断通知路由协议(例如OLSR)。 If the full link to quality deterioration (indicating loss of LOS distance to another node or too large), then declare the link is interrupted, and to link the interrupt notification routing protocols (such as OLSR). 这将指示在网络层重新进行路由选择。 This would indicate that the network layer re-routing. 只有全向链路的质量仍然良好,那么必须依然关于干扰监视各个时隙。 Only omni link quality is still good, it must remain on the interference monitor each time slot.

下面的讨论举例说明当全向链路质量良好时,如何评估定向时隙上的潜在或现有干扰。 The following discussion illustrates, when omni link quality is good, how to assess the potential or existing interference on directional slot. 将存在链路质量指示符表示到指定相邻节点的全向链路质量良好,但是分配给该节点的定向时隙之一非常不可靠的情况。 The presence of link quality indicator represents the whole to the specified neighboring node to the link quality is good, but one of the nodes assigned to the slot orientation is very unreliable. 这将是未被如上所述的干扰回避技术避免的该时隙上的过多干扰的指示。 This would not be as described above avoid excessive interference on the time slot interference avoidance technology indication.

能够导致这种干扰的情况有几种,它们一般由移动引起。 Such interference can result in the case there are several, they are generally caused by movement. 例如,这些情况包括所需信号的到达方向的快速变化,干扰信号的到达方向的快速变化,干扰源的数目不断增大,以及节点和其目标节点及干扰节点之间的距离的变化。 For example, these situations include rapid changes in direction of arrival of the desired signal, rapid change of the direction of arrival of the interfering signal, increasing number of interference sources change in the distance between, and the node and its target node and the interfering nodes.

下面将参考图26,说明根据本发明的干扰监视/减轻方法的概要。 Below with reference to FIG. 26, described in accordance with the present invention, the interference monitoring / mitigation approach outline. 开始于方框260,如上所述,在一个或多个时隙内在该对移动节点m和n之间建立了通信链路之后(方框261),在方框262,这两个节点中的至少一个确定在所述时隙期间,与所述通信链路相关的链路质量值。 Begins at block 260, as described above, in one or more of the time slot between the pair of mobile nodes m and n after the establishment of a communication link (block 261), at block 262, the two nodes determining at least one during the time slot associated with the communication link link quality value. 随后在方框263和264,该节点确定通信链路是否具有第一(临界)质量或第二(低劣)质量。 Then in block 263 and 264, the node determines whether the communication link having a first (critical) quality or second (poor) quality.

根据本发明,在方框265,如果在指定时隙内,通信链路具有临界质量,那么移动节点合作以便在第一时段内把通信链路重新分配给一个新时隙。 According to the present invention, at block 265, if within the specified time slot, the communication link having a critical mass, then the mobile node cooperation for the first time period to re-allocate the communication link to a new time slot. 此外,在方框266,如果在指定时隙内,通信链路具有低劣的质量,那么节点合作以便在短于经一时段的第二时段内,把通信链路重新分配给一个新时隙。 In addition, at block 266, if within the specified time slot, the communication link having a poor quality, then the nodes cooperate in order through the second period is shorter than a period, the communication link reassigned to a new time slot. 即,如果链路质量严重恶化,那么时隙可被快速分配(即,在第二个较短的时段内),以使对通信的破坏降至最小。 That is, if link quality is seriously deteriorated, then the slot may be quickly assigned (i.e., within the second shorter time period), so that the destruction of the communication is minimized. 另一方面,如果链路质量处于临界状态但是仍然可用,那么可以更长地延迟时隙的重新分配(即,在较长的第一时段内),这可减少试图同时重新分配时隙的各对竞争节点之间的冲突。 On the other hand, if link quality is marginal but still usable, it may be longer delay reallocate time slots (i.e., within the longer first time period), which can simultaneously reduce the attempt to redistribute the respective slots conflict between competing nodes.

图27中更详细地图解说明了在方框262-264一般性说明的步骤。 Figure 27 illustrates in more detail the steps at block 262-264 the general description. 具体地,在方框271,如上所述,最好根据SIR和分组接收差错值(PREV),确定链路质量值。 Specifically, at block 271, as described above, according to the best SIR and packet reception error value (PREV), to determine the link quality value. 如下进一步所述,在方框272和274,如果SIR在第一SIR阈值和第二SIR阈值(它高于第一质量阈值)之间,并且PREV在第一差错阈值和第二差错阈值(它高于第一差错阈值)之间,那么通信链路最好被确定为具有临界质量,随后结束该方法的这一方面(方框278)。 As further described below, at block 272 and 274, if the SIR of the first SIR threshold and a second SIR threshold (which is higher than the first quality threshold) between, and the PREV first error threshold and the second error threshold (which in higher than the first error threshold) between, then a communication link is preferably determined to have a critical mass, then the end of this aspect of the method (block 278).

类似地,在方框274和273,如果SIR小于第一SIR阈值,并且PREV大于第二差错阈值,那么通信链路最好也被确定为具有临界质量。 Similarly, at block 274 and 273, if the SIR is less than the first SIR threshold and the PREV is greater than the second error threshold, the communication link is also preferably determined to have a critical mass. 另外,在方框275和276,如果SIR小于第一SIR阈值,并且PREV介于第一差错阈值和第二差错阈值之间,那么通信链路被确定为质量低劣。 Further, at block 275 and 276, if the SIR is less than the first SIR between the threshold and PREV is between a first error threshold and the second error threshold value, then the communication link is determined to be of poor quality. 另外,如果质量因子低于第一差错阈值,那么通信链路被确定为质量低劣。 Further, if the quality factor is below the first error threshold value, then the communication link is determined to be of poor quality.

上述方法特别适合于按需分配时隙。 The above-described method is particularly suitable for on-demand time slots. 对于半永久时隙来说,在一些应用中,可能适于相当快速地重新分配这些时隙。 For semi-permanent time slot, in some applications it may be adapted fairly quickly re-allocate these time slots. 这样,现在参见图28,开始于方框280,在方框281初始确定SP时隙之后,类似于上面关于DA时隙所述那样,在方框282,确定半永久时隙内与通信链路相关的半永久链路质量值。 Thus, Referring now to Figure 28, begins at block 280, at block 281 to determine the initial SP time slot after, similar to the above with respect to the DA slots, as at block 282, to determine the semi-permanent time slot associated with the communication link The semi-permanent link quality value. 如果在方框283,根据半永久链路质量值,在半永久时隙内,额外的通信链路被确定为具有临界质量或者低劣的质量,那么在方框284,在相当短的第二时段内,链路被重新分配给新的SP时隙。 If at block 283, according to a semi-permanent link quality value, in the semi-permanent time slot, an additional communication link is determined as having a critical mass or poor quality, then at block 284, the second in a relatively short period of time, link is re-allocated to a new SP time slot. 否则,如同举例所示那样,将继续监视链路质量值。 Otherwise, as shown as an example, will continue to monitor link quality value.

下面将参考用于计算上述变量的特定等式,更详细地说明上面概述的干扰回避和减轻程序。 Below with reference to the above-mentioned variables used to calculate a specific equation, described in more detail with the interference avoidance and mitigation outlined above procedures. 同样,将对每个时隙和每个干扰节点定期重新计算几个关键变量。 Similarly, it will interfere with each slot and each node periodically recalculated several key variables. 这些关键变量包括每个定向链路Lk的每个时隙i的链路质量N_qualityiLk及其量化值TS_QualiLk;时隙I的潜在SIRmnTi;和全向链路状态Can_AllocLk。 These key variables include each time slot of each directional link Lk i N_qualityiLk link quality and the quantized value TS_QualiLk; potential slot I SIRmnTi; and omnidirectional link state Can_AllocLk.

这些变量将是时隙i上的干扰或潜在干扰的指示符。 These variables will be indicators of interference or potential interference on time slot i. 这些变量的微小降低将启动更从容的尝试(即,在相对较长的第一时段内),重新调度更良好的时隙,以便避免潜在的干扰。 Slight lowering these variables will initiate a more leisurely attempt (i.e., within the first relatively long time period), rescheduling a better time slot, in order to avoid potential interference. 由于干扰节点对可能也已发现该条件,因此从容的补偿(backoff)将降低两个节点对同时重新调度,并且可能引起另一冲突的可能性。 Since the interfering node may also have been found for this condition, and therefore comfortably compensation (backoff) will reduce both nodes simultaneously re-scheduling, and may cause the possibility of another conflict.

这样,如果干扰节点对决定首先重新调度,这可消除重新调度的需要。 Thus, if the interfering node pair decide to reschedule first, this may eliminate the need for rescheduling. 对于该“慢速”补偿方法(即,在第一时段内),将在下一ΔTR秒内的任意时间,以概率prsslow启动重新调度操作。 For this "slow" compensation method (i.e., within the first period), at any time within the next ΔTR seconds, the probability prsslow start rescheduling operation. 如果重新调度操作未被启动,那么当再次重新计算变量时,将再次进行测试(大约每秒一次)。 If you re-scheduling operation is not started, then when again recalculate variables will be tested again (about once per second). 如果测试再次失败,那么将在下一ΔTR秒内的任意时间,以概率prsslow启动重新调度操作。 If the test fails again, then at any time within the next ΔTR seconds probability prsslow start rescheduling operation. 只要潜在干扰条件存在,该测试过程就将继续下去。 As long as potential interference condition exists, the testing process will continue.

现有严重恶化的检测将要求更即时的重新调度,以避免现有的过多干扰。 Serious deterioration of existing detection will require more immediate rescheduling to avoid existing excessive interference. 这种情况下,“快速”补偿方法是在较短的第二时段内启动时隙的重新调度。 In this case, "fast" compensation method is to start in a short time slot a second time re-scheduling. 例如,它可在下一定向信号出现时间之内,并以概率prsfast启动重新调度操作。 For example, it may occur at the next time of the signal, and start with probability prsfast reschedule operation. 如果未在该信号出现时间内启动重新调度,那么以概率prsfast在下一信号出现时间内启动重新调度,等等。 If the time to start re-scheduling, then the probability prsfast start rescheduling time signal appears in the next, and so does not appear in the signal.

不进行当前时隙分配的任何重新分配和对重新分配进行慢速或者快速补偿之间的决定最好基于下述标准。 Decided to re-allocation and re-allocation of slow or fast without any compensation between the best of the current slot allocation based on the following criteria. 首先,如果用于节点m和n之间的链路的时隙i满足下述两个条件任意之一,那么不考虑时隙的重新分配:SIRmnTi>TSIRMandTS_QualiLk=GOOD OR MID;and...(47)]]>SIRmnTi>TSIRLandTS_QualiLk=GOOD....(48)]]>此外,如果用于节点m和n之间的链路的时隙i满足下述两个条件任意之一,那么使用对时隙的重新分配的慢速补偿(如果它是用于节点m和n之间的链路的唯一时隙,那么使用快速补偿):TSIRL≤SIRmnTi≤TSIRMandTS_QualiLk=MID;and...(49)]]>SIRmnTi<TSIRLandTS_QualiLk=GOOD....(50)]]>另外,如果用于节点m和n之间的链路的时隙i满足下述两个条件任意之一,那么使用对时隙的重新分配的快速补偿: First, if the time slot for the link between nodes m and n i meet any one of the following two conditions, then do not consider re-allocation of slots: SIRmnTi> TSIRMandTS_QualiLk = GOOD OR MID; and ... ( 47)]]> SIRmnTi> TSIRLandTS_QualiLk = GOOD .... (48)]]> Also, any one of the following two conditions slot i If the link between nodes m and n for the meet, then use Slow compensation for the reallocation of slots (slots only if it is used for the link between nodes m and n, then use the fast compensation): TSIRL & le; SIRmnTi & le; TSIRMandTS_QualiLk = MID; and ... (49 )]]> SIRmnTi & lt; TSIRLandTS_QualiLk = GOOD .... (50)]]> In addition, any one of the following two conditions slot i If the link between nodes m and n for the meet, then the use of the Rapid re-allocated time slot of compensation:

SIRmnTi<TSIRLandTS_QualiLk=MID;and...(51)]]>TS_QualiLk=BAD....(52)]]>如果下述条件被满足,那么时隙被解除分配,并且路由协议(例如OLSR)被告知节点m和n之间的链路已被中断:持续至少TB秒,TS_QualiLk=BAD]]>并且Can_AllocLk=FALSE...(53)]]>如上简要所述,一旦通过快速补偿或者慢速补偿确定重新调度是必需的,那么如果时隙i是SP时隙,某些预防措施是合乎需要的。 SIRmnTi & lt; TSIRLandTS_QualiLk = MID; and ... (51)]]> TS_QualiLk = BAD .... (52)]]> If the following condition is satisfied, then the time slot to be deallocated, and the routing protocol (e.g., OLSR) was informed link between nodes m and n have been interrupted: for at least TB seconds, TS_QualiLk = BAD]]> and Can_AllocLk = FALSE ... (53)]]> As briefly mentioned, once compensated by fast or slow Speed Compensation determine rescheduling is required, so if the time slot i is the SP slot, certain precautions are desirable. 由于SP时隙的重要性,重要的是该时隙未被恶化。 Because of the importance of SP time slots it is important that the time slot is not deteriorated. 如果该时隙是在节点m和n之间分配的唯一时隙,那么即使下面列举的条件指示慢速补偿令人满意,也最好进行快速补偿。 If the slot is the only slot between nodes m and n is assigned, then even if the conditions listed below indicate slow satisfactory compensation, it is best to quickly compensate.

另一种情况是当节点m和n具有分配的几个时隙时。 Another case is when nodes m and n have several time slots having assigned. 如果对SP时隙不同地指示慢速补偿,那么节点m能够立即把应使它们的DA时隙之一成为SP时隙通知节点n,从而使当前时隙承担DA时隙的不太关键的任务。 If the SP in different time slots indicate slow compensation, then the node m can immediately should enable them to become one of the DA slot slot SP notify node n, so that the current time slot assume less critical tasks DA slot . 如果分配给节点m和n的其它时隙都不具有足以重命名为SP时隙的质量,那么使用快速补偿来重新调度SP时隙i。 If you are assigned to the nodes m and n are not other slots have enough SP slot rename quality, then use the quick compensation to reschedule the SP slot i. 如果时隙i是DA时隙或者可以使之成为DA时隙,那么使用依据下述标准确定的快速或慢速补偿决定。 If the slot is DA slot or i can make it a DA time slot, then the use of the decision on the basis of fast or slow the compensation determined by the following criteria.

特别地,在进行新时隙分配之前,正被重新分配的时隙不被解除分配。 Specifically, before performing a new time slot assignment, the slot being re-allocated is not de-allocated. 在获得新的时隙之后,旧的有问题的或者低劣的时隙可被解除分配。 After obtaining a new time slot, the old questionable or poor slot may be deallocated. 如果另一时隙可用于节点m和n之间的链路,那么最好停止“低”质量(TS_QualiLk=BAD)]]>时隙内分组的转发。 If another slot can be used for the link between nodes m and n, then it is best to stop the "low" quality (TS_QualiLk = BAD)]]> forward packets within the slot.

具有优化链路状态路由(OLSR)路由算法的简单接口可包括在本发明中,并且下面参考图2和29进行讨论。 Simple interface with optimized link state routing (OLSR) routing algorithm may be included in the present invention, and with reference to FIG. 2 and the following 29 discussion. 本领域的技术人员已知,就OLSR来说,链路状态信息能够量化链路的状态。 Known to persons skilled in the art, it is OLSR, link state information can quantify the status of the link. 完整的布局信息被保持,并且定期向网络中的所有其它节点广播近邻布局信息,以允许它们构成完整的网络布局。 Complete layout information is maintained and regularly to all other nodes in the network broadcasts neighbor layout information to allow them to form a complete network layout. 应使OLSR不知道围绕时隙的分配和解除分配,以满足通信量需求和干扰的减轻的多数活动。 Should make OLSR do not know most of the activities surrounding the allocation and deallocation of slots to meet traffic demands and interference mitigation. 当存在分配给与某一相邻节点的链路的几个时隙,例如到相邻节点k的链路具有m个时隙i1,i2,…,im时,出现一个问题。 When there are several time slots allocated to a link of the neighboring nodes, e.g., a link to an adjacent node k has m time slots i1, i2, ..., im, a problem arises. 如果由于干扰的缘故,一个或多个,但是不是全部所述时隙被确定为质量低劣,那么可启动时隙的重新调度。 If the interference due to the reason, one or more, but not all the time slots are determined to be of poor quality, it can be rescheduled start time slots. 只有给该相邻节点的至少一个时隙仍然质量较高,那么所述节点仍然是被连接的近邻,路由表中不需要任何改变。 Only to the adjacent node the at least one time slot is still higher quality, then the neighbor node is still being connected, without any change in the routing table. 从而,不需要使OLSR对此作出反应。 Thus, unnecessary to OLSR react. 在重新调度期间,链路确实遭到一些容量的损失(这可能关系重大),但是不存在OLSR应进行的任何事情。 During the re-scheduling, link does suffer some loss of capacity (which may be at stake), but anything OLSR should not exist. 如果所有时隙质量都低,那么必须允许OLSR作出反应,并通过其它链路重新发送数据。 If all slots are low quality, then OLSR must be allowed to respond, and re-transmits data through other links. 以分配给到节点k的每个时隙的最大质量的形式,定义了该链路的下述链路质量量度:N_qualityLk=maxAll_TS_for_Lk(N_qualitylLk)...(54)]]>从而每次到节点k的链路的质量量度N_qualityLk变化时,可把该质量量度报告给OLSR。 In the form of maximum quality is assigned to each time slot to the node k, the following definition of a link quality metric for this link: N_qualityLk = maxAll_TS_for_Lk (N_qualitylLk) ... (54)]]> to each node so that k link quality metric N_qualityLk change, you can put the quality measure reported to OLSR. 该质量量度将替换OLSR一般根据接收的HELLO分组计算的质量量度。 The quality measure will replace OLSR HELLO generally based on the quality of the received packet is calculated measure. 它将替换标准的OLSR质量量度,但是随后它将由OLSR按照完全相同的方式使用。 It will replace the standard OLSR quality measure, but then it will be used by OLSR in exactly the same way. 如同在OLSR中通常进行的那样,可对其应用滞后函数,以降低OLSR观察到的“链路摇摆”。 As as is conventionally done in OLSR, a function may be to apply the hysteresis to reduce OLSR observed "link swing." 如果链路质量变得过差,那么该链路将被宣告“停止”,OLSR将在其HELLO和TC更新分组中自动发送新的状态信息,并重新计算路线。 If the link becomes too poor quality, then the link will be declared "Stop", OLSR packets will be updated automatically send new status information in its HELLO and TC, and recalculate the route.

更具体地说,控制器18(图2)最好包括发现路线,并利用路由协议,例如OLSR把通信发送给相邻节点的通信路由器18i。 More specifically, the controller 18 (FIG. 2) preferably includes a route discovery and use of routing protocols, such as OLSR send communications to neighboring nodes communication router 18i. 时隙调度单元18a/18b调度时隙,以便与每个相邻移动节点建立通信链路,天线对准单元18c使定向天线在与每个相邻移动节点相关的通信期间对准所述每个相邻移动节点。 Time slot scheduling unit 18a / 18b scheduling time slots to establish a communication link with each neighboring mobile node, an antenna aligning unit 18c so that the directional antenna during the communication associated with each neighboring mobile node aligned with each of said neighboring mobile nodes. 这里,提供链路质量估计器18h,以便根据为通信链路安排的每个时隙的质量,估计链路质量,并把估计的链路质量报告给通信路由器18i。 Here, there is provided a link quality estimator 18h, according to the quality of each time slot scheduled for the communication link, to estimate link quality, and to report estimated link quality to the communications router 18i.

如上所述,估计的链路质量可被定义为通信链路调度的每个时隙的最大质量,路由器18i可根据估计的链路质量,把通信传送给相邻的节点。 As described above, the estimated link quality may be defined as the maximum quality of each time slot scheduled communication link, and the router 18i may be based on the estimated link quality to transmit a communication to neighboring nodes. 只要用于该链路的至少一个时隙的最大质量高于预定阈值,那么最好保持该通信链路,当该链路的估计链路质量低于预定阈值时,路由器18i开始新的路线发现。 As long as the link for at least one time slot of maximum mass above a predetermined threshold, then it is best to maintain the communication link, when the estimated link quality of the link is lower than a predetermined threshold value, the router 18i to start a new route discovery . 路由器18i可对从链路质量估计器报告的估计链路质量使用滞后函数。 18i router can estimate link quality report from the link quality estimator using hysteresis function.

现在参见图29,说明本发明的方法的一般步骤。 Referring now to Figure 29, the general steps of the method of the present invention will be described. 该方法开始于方框290,根据上面讨论的过程,分别在方框291和292,调度时隙并对准天线。 The method begins at block 290, depending on the process discussed above, respectively, in blocks 291 and 292, and scheduling slot antenna alignment. 在方框293,估计链路质量,并如方框294中所示,向通信路由器18i或者OLSR报告链路质量,之后在方框295结束该方法。 At block 293, the estimated link quality, and as shown in block 294, the communications router 18i or OLSR reporting link quality, then at block 295 the method ends.

从而本发明为相控阵网络提供一种完全分布的链路调度算法和协议。 The present invention thus provides a link scheduling algorithms and protocols fully distributed network for phased array. 应注意所述算法/协议细节的上述描述设想了每个节点单个定向波束的情况,在用于该访问的分配时隙内,它被时发并且指向相邻的节点。 The algorithm should be noted that the above description / details of the agreement envisages a single directional beam of each node, and in the allocation of slots for the visit, and the point when it issued an adjacent node. 但是,该方法可被用于每个节点任意数目的指向波束的情况。 However, this method may be used an arbitrary number of nodes pointing to each beam.

Referenced by
Citing PatentFiling datePublication dateApplicantTitle
CN101193054B28 Nov 200630 Mar 2011华为技术有限公司A method and system for data transmission and node devices and coordination device
WO2008064598A1 *1 Nov 20075 Jun 2008Huawei Technologies Co., Ltd.A method and system for transmitting data, and a node device and coordination device
WO2016011669A1 *25 Jul 201428 Jan 2016华为技术有限公司Communication device and method for resource allocation
Classifications
International ClassificationH04B7/04, H04L1/00, H04J3/16, H04W84/18, H04L12/28, H04L1/20, H04L12/56, H04B7/26, H04W16/28, H04B7/212, H04W74/04, H04W76/02, H04W16/14, H04W72/12, H04W72/10
Cooperative ClassificationH04W72/085, H04W72/10, H04W72/0446, H04W74/04, H04L1/0041, H04B7/18506, H04W72/1242, H04B7/2643, H04W76/02, H04L1/203, H04W84/18, H04L1/0061
European ClassificationH04L1/20, H04L1/00B3, H04L1/00B7E, H04B7/26T, H04W72/08D
Legal Events
DateCodeEventDescription
31 May 2006C06Publication
26 Jul 2006C10Request of examination as to substance
27 Jan 2010C02Deemed withdrawal of patent application after publication (patent law 2001)