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.


  1. Advanced Patent Search
Publication numberCN1750514 B
Publication typeGrant
Application numberCN 200510109819
Publication date18 May 2011
Filing date23 Jun 2005
Priority date23 Jun 2004
Also published asCN1750514A, DE602005006478D1, EP1617608A1, EP1617608B1, US7616575, US20050286426
Publication number200510109819.2, CN 1750514 B, CN 1750514B, CN 200510109819, CN-B-1750514, CN1750514 B, CN1750514B, CN200510109819, CN200510109819.2
InventorsBD兹尔, JD帕德海, 小RP德拉弗斯
Export CitationBiBTeX, EndNote, RefMan
External Links: SIPO, Espacenet
System and method for quality routing using weighted cumulative expected transmission time metric
CN 1750514 B
A system and method for link quality routing uses a weighted cumulative expected transmission time path metric. A system for routing in a multi-hop ad hoc network, and a method for measuring the link quality of a route in the network, include assignment of a weight to a link in accordance with an expected transmission time of a packet over the link, and a combining of individual link weights for a route into a path metric. The path metric accounts for interference among links that use a shared channel. In the calculation of the path metric, the expected transmission times of links that interfere with one another are added, while the expected transmission times for non-interfering links are considered separately.
Claims(15)  translated from Chinese
1.在多跳特定网络里,一种用于测量从源节点到目标节点的路由的链路质量的系统, 所述路由包括多个链路,所述系统包括:用于根据链路的性能,为路由里的每个链路分配权值的装置,其中,链路的性能包括在链路上传送分组以使分组成功穿过链路的所期望时间的测量值乘以链路的带宽测量值,所述带宽测量值为通过把固定的分组大小除以原始数据率而得到的商;及用于对为路由分配的链路权值进行加权平均得到路径度量的装置,其中所述路径度量说明了使用共享信道的链路之间的干扰。 1. In a multi-hop ad hoc network, the method for measuring the link quality route from the source node to the destination node system comprises a plurality of links of the route, the system comprising: means for performance of the link in accordance with , where the means of routing weights assigned for each link, wherein the link performance including transport packets on the link so that the packet successfully across the link measurement value is multiplied by a desired time link bandwidth measurement the value of the bandwidth measured value by the fixed packet size divided by the original data rate quotient obtained; and for the link weights assigned to the route by the weighted average path metric device, wherein the path metric It describes using a shared channel interference between the links.
2.如权利要求1所述的系统,其中所述多跳特定网络包括一个或多个具有多个无线电收发器的节点。 2. The system of claim 1, wherein said multi-hop network includes one or more particular nodes having multiple radio transceivers.
3.如权利要求1所述的系统,其中所述多跳特定网络包括一个或多个具有多个同种无线电收发器的节点。 3. The system according to claim, wherein said multi-hop network includes one or more particular nodes having a plurality of radio transceiver isotype.
4.如权利要求1所述的系统,其中在链路上传送分组以使分组成功穿过链路的所期望时间的测量包括ETX测量。 4. The system of claim 1, wherein the packet transmitted on the link so that a desired packet successfully across the link comprises a measurement time ETX measurement.
5.如权利要求1所述的系统,其中带宽测量值是凭经验来确定的。 5. The system of claim 1, wherein the bandwidth measurement is determined empirically.
6.如权利要求1所述的系统,其中带宽测量值是使用分组对技术来确定的。 6. The system according to claim 1, wherein the bandwidth measurement is determined using the packet-pair technique.
7.如权利要求1所述的系统,其中路径度量包括加权的平均值,{I-β)*Υ4Ψί +p*ma^Xj其中η是路由中链路的数量,Wi是链路i的权值,Xi是信道j上权值的总和,K是信道的数量,及β是符合β ( 1的参数。 7. The system of claim 1, wherein the path metric comprises a weighted mean value, {I-β) * Υ4Ψί + p * ma ^ Xj where η is the number of links in the route, Wi is the weight for link i value, Xi is the sum of the weights of the channel j, K is the number of channels, and β is in line with the parameters β (1 in.
8.如权利要求7所述的系统,其中信道j上的权值的总和Xj由Xj = YWi \<j<k跳跃i在信道j上来计算,因此,总的路径吞吐量由瓶颈信道来决定,其中瓶颈信道是具有最大Xj值的信道。 8. The system of claim 7, wherein the sum Xj right channel on the value of j by Xj = YWi \ <j <k i in channel j jump up calculation, therefore, the total throughput path determined by the bottleneck channel where the bottleneck channel is a channel having a maximum value of Xj.
9.如权利要求7所述的系统,其中链路i的权值包括在链路i上传送分组以使分组成功穿过链路的所期望时间的测量值,乘以链路i的带宽测量值。 9. The system of claim 7, wherein the weight value of the link i comprises transport packets across the link i so that the packet successfully across the link, the desired measurement time, multiplied by the link bandwidth measurement i value.
10.在多跳特定网络里,一种用于测量从源节点到目标节点的路由的链路质量的方法, 所述路由包括多个链路,所述方法包括:根据链路的性能,为路由里的每个链路分配权值,其中,链路的性能包括在链路上传送分组以使分组成功穿过链路的所期望时间的测量值乘以链路的带宽测量值,所述带宽测量值为通过把固定的分组大小除以原始数据率而得到的商;及对为路由分配的链路权值进行加权平均得到路径度量,其中所述路径度量说明了使用共享信道的链路之间的干扰。 10. In particular multi-hop network, a link quality measurement from the source node to the destination node routing method used, the route comprising a plurality of links, the method comprising: the performance of the link, as Each link in the routing weight distribution, wherein the link performance including transport packets on the link so that the packet successfully across the link, the measured value of the desired bandwidth of the link multiplied by the time measurement value, the Bandwidth measured value by a fixed packet size divided by the original data rate quotient obtained; and the link weights assigned to the route by the weighted average path metric, wherein the path metric illustrates the use of the shared channel link interference between.
11.如权利要求10所述的方法,其中所述多跳特定网络包括一个或多个具有多个无线电收发器的节点。 11. The method of claim 10, wherein said multi-hop network includes one or more particular nodes having multiple radio transceivers.
12.如权利要求10所述的方法,其中所述多跳特定网络包括一个或多个具有多个同种无线电收发器的节点。 12. The method of claim 10, wherein said multi-hop network includes one or more particular nodes having a plurality of the same kind of radio transceivers.
13.如权利要求10所述的方法,其中所述带宽测量值是凭经验确定的。 13. The method of claim 10, wherein the bandwidth measurement is determined empirically.
14.如权利要求10所述的方法,其中所述带宽测量值是使用分组对技术确定的。 14. The method of claim 10, wherein the bandwidth measurement is determined using the packet-pair technique.
15.如权利要求10所述的方法,其中对为路由分配的链路权值进行加权平均得到路径度量包括:为路由里的链路计算期望传输时间的总和; 为一个或多个链路使用的每个信道确定共享信道的链路的子集; 为每个子集里的链路计算期望传输时间的总和; 确定具有最大总和的子集;及通过用(i-β)乘以路由中链路的期望传输时间的总和加上β乘以具有最大总和的子集的期望传输时间的总和来计算路径度量,其中β是符合β ^ 1的参数。 15. The method of claim 10, wherein the link weights for the route assigned to the weighted average to obtain the path metric comprises: routing link in calculating the sum of the expected transmission time; using one or more links determining for each channel of the shared channel subset of links; for each link in a subset of the total transmission time calculating expected; determining a subset having a maximum sum; and by (i-β) multiplied by the chain route the sum of the expected transmission time plus β multiplied by the sum of the path having the maximum sum of the subset to calculate a desired transmission time path metric, wherein β is in line with the parameter β ^ 1.
Description  translated from Chinese

使用加权的累计期望传输时间度量进行高质路由的系统和 Cumulative expected transmission time metric weighted routing system will be high-quality and

方法 Method

[0001] 相关申请 [0001] RELATED APPLICATIONS

[0002] 本专利申请要求2004年6月23日申请的、申请号为60/582,360的美国临时专利申请为优先权,并引用在此作为参考。 [0002] This patent application claims the June 23, 2004 application, U.S. Provisional Patent Application No. 60 / 582,360 Application of priority, and incorporated herein by reference.

技术领域 Technical Field

[0003] 本发明一般地涉及计算机网络,尤其是涉及在多跳(multi-hop)无线网络里用于高质路由协议的度量。 [0003] The present invention relates generally to computer networks, and more particularly to a multi-hop (multi-hop) wireless network in routing protocols for quality metrics.

背景技术 Background

[0004] 路由是指选择穿越计算机网络的一个路径的问题,消息将沿这个路径从源节点传输到目标节点。 [0004] The route is to choose a path through the computer network of the problem, the message will be transmitted along the path from the source node to the destination node. 多跳特定(ad hoc)网络中的路由提出了特殊的挑战。 Multihop specific (ad hoc) network routing presents special challenges. 特定网络是自组织网络,在这些网络里,节点共同地维持网络连通性。 Specific networks are self-organizing networks, these networks, the nodes together to maintain network connectivity. 特定网络里的节点都装备了无线电收发器,诸如符合IEEE 802. 11标准的网络接口卡。 In particular network nodes are equipped with a radio transceiver, such as compliance with IEEE 802. 11 standard network interface cards. 这些无线电收发装置允许在节点之间通信而不需要集中式网络管理机构或固定的网络基础结构。 These radios allow communication between nodes without requiring centralized network administration or fixed network infrastructure. 由于每个无线电收发装置都具有有限的有效范围,因而两个远距离的节点必须穿过多跳路径来通信。 Since each device has a radio transceiver limited effective range, two distant nodes must thus pass through the multi-hop communication path. 在这个路径中的中间节点通过转发分组来充当路由器。 In this path to forward packets through intermediate nodes to act as a router.

[0005] 无线共同体(community) “网状”网络是一种越来越重要的多跳无线网络,可用来提供到共同体的宽带英特网访问。 [0005] Wireless Community (community) "mesh" network is an increasingly important multi-hop wireless networks can be used to provide broadband Internet access to the community. 在这些网络里,节点或是静止的或是最低限度可移动的, 并且一般不依赖于电池电源。 In these networks, the nodes either stationary or minimally mobile and typically do not depend on battery power. 用于无线共同体网络的路由技术聚焦在提高网络容量或单独传输的性能上,而不是应付灵活性或使功率用量最小化。 Routing techniques for wireless community networks focused on improving the performance of the network transmission capacity or individually, instead of dealing with flexibility or to minimize power usage.

[0006] 次优的网络容量是与多跳无线网络相关的普遍问题。 [0006] Suboptimal network capacity is a common problem with multi-hop wireless network dependent. 在这些网络里,随着跳跃数量的增加,吞吐量快速下降。 In these networks, the increase in the number as jumping, rapid decline in throughput. 在802. 11网络里,这个原因部分是由于802. IlMAC所固有的不公平,这个固有的不公平可以拖延多个跳跃上的分组流。 In the 802.11 network, this is due in part 802. IlMAC inherent unfairness, the inherent unfairness can delay packet flows on multiple jumps. 此外,这些网络一般在每个节点上只使用了可用频谱的一小部分和单个无线电收发装置来发送和接收分组。 Further, these networks typically used on each node, and a small portion of the available spectrum a single radio transceiver to transmit and receive only the packets.

[0007] 改善多跳无线网络网络容量的一个方法是为节点装备多于一个的无线电收发装置。 One method [0007] Multi-hop wireless networks to improve network capacity is more than one node equipped with a radio transceiver. 这个解决方法具有几个优点。 This solution has several advantages. 具有多个无线电收发装置的节点可以同时发送和接收。 Having a plurality of radio node devices can send and receive simultaneously. 而且,具有两个无线电收发装置的节点可以同时在两个信道上发送,使得网络可以利用较大部分的频谱。 Further, two nodes having radios can transmit simultaneously in the two channels, so that the network can use a larger portion of the spectrum. 此外,多个不同种类的无线电收发装置的使用可以提高耐用性、连通性和性能,所述不同种类的无线电收发装置在不同频带(例如在5(ihZ的802. Ila和在2. 4Ghz的802. llb/g)上操作,具有不同带宽、波段(range)和衰减特性。最后,802. 11网络接口卡是快速减价可用的现成商品元件。 In addition, multiple users with different types of radios can improve durability, connectivity and performance, the different types of radios in different frequency bands (for example, 5 (ihZ of 802. Ila and in 2. 4Ghz 802 . llb / g operate), and with different bandwidths, the band (range) and attenuation characteristics. Finally, 802.11 network interface cards are readily available commodity components quickly Deals.

[0008] 用于多跳无线网络的大部分路由协议集中在选择具有最少中间跳数量的路径上。 [0008] The majority of routing protocols for multi-hop wireless networks focused on selecting a minimum number of intermediate hops path. 在这些网络里申请最短路径路由的缺点是非常公知的。 Shortcomings in the application shortest path routing in these networks are very well known. 选择最少跳数量的路径会导致差的性能,部分是因为这些路径通常在包括远距离节点之间缓慢的或者有损耗的衰减的无线链路。 Selecting a minimum number of hops of the path will result in poor performance, in part because these paths typically include slow or lossy wireless links between distant attenuation nodes. 在具有多个无线电收发装置节点的网络里一般较差地执行最短路径路由技术。 In a network having a plurality of radio transceiver nodes in generally poor implementation of the shortest path routing technology. 这些可 These

4通过以下两个示例来说明。 4 The following two examples to illustrate. 第一,设定一个网络,其中每个节点都具有802. Ila和802. Ilb 无线电。 First, set up a network in which each node has 802. Ila and 802. Ilb radio. 通常802. Ilb无线电比802. Ila无线电具有较长波段。 Usually 802. Ilb radio than 802. Ila radio has a longer band. 因此,如果使用最短路径路由,那么该网络里的大量通信量将在较慢的802. Ilb链路上执行。 Therefore, if you use the shortest path routing, so that the network will perform a lot of traffic on a slow link 802. Ilb. 现在假设该网络里的节点更换为具有两个802. Ilb无线电,一个转向信道1,另一个转向信道11。 Now suppose that the network node 802. Ilb replaced with two radios, a steering channel 1, another steering channel 11. 假设在这个网络里具有两跳(三个节点)。 Assumed to have a two-hop (three nodes) in the network. 完整地穿越信道1或信道11的路径将比在两个不同信道上的两跳的路径的吞吐量差得多。 Complete path through the channel 1 or channel 11 is much more than the difference in the throughput of the two-hop path two different channels. 因此选择不能保证跳是位于不同信道上的路径的最短路径算法将产生次优的性能。 So choose not guarantee hop is the shortest path algorithm in different channels paths will produce suboptimal performance.

[0009] 基于链路高质的路由是最短路径路由的一个公知替换。 [0009] The link quality-based routing is a well-known replacement shortest path routing. 研究人员已经提出了用于多跳无线网络的几个链路高质度量,但这些计划集中在具有同种的单一无线电节点的网络上。 Researchers have proposed a number of link quality metrics for multi-hop wireless networks, but these plans to focus on the network with a single radio nodes of the same species. 一个示例是在DSJ De Couto,D. Aguayo, J. Bicket,和R. Morris 的“A High-Throughput Path Metric for Multi-Hop Wireless Routing,"ACMM0BIC0M(2003 ^ 9月)中所描述的ETX度量,引用在此作为参考。 One example is in DSJ De Couto, D. Aguayo, J. Bicket, and R. Morris's "A High-Throughput Path Metric for Multi-Hop Wireless Routing," ACMM0BIC0M (2003 ^ September) ETX metric described, incorporated herein by reference. ETX需要经过链路发送单路传送分组来测量期望的传输时间,该传送包括重发。 ETX need to go through link transmission unicast packets to measure a desired transmission time, the transmission comprises a retransmission. 所述路径度量是用于所述路径里每个链路的ETX值的总和。 The path metric is the sum of the path in the ETX values for each link. 路由协议选择具有最小路径度量的路径。 Routing protocol selection path having the smallest path metric.

[0010] 回顾ETX对理解本发明在此公开的实施例是有用的。 [0010] Review ETX understanding of embodiments of the invention disclosed herein are useful. ETX的引出始于在向前和向后方向(分别用Pf和Pr表示)里潜在的分组丢失可能性的测量。 ETX begins with measurements of lead in the forward and backward directions (respectively represented by Pf and Pr) in a potential packet loss probability. 首先,计算分组发送不成功的的可能性。 First, the possibility of sending a packet unsuccessful calculation. 对于发送成功的分组,所述802. 11协议请求所述分组的成功肯定。 For packet transmission successful, the 802.11 protocol request is successful sure the packet. 让P表示分组传送从χ到y不成功的可能性。 Let P possibility packets from χ to y unsuccessful representation. 那么: Then:

[0011] P = I-(I-Pf)* (I-Pr) [0011] P = I- (I-Pf) * (I-Pr)

[0012] 所述802. IlMAC将为这些发送不成功的重发分组。 [0012] the 802. IlMAC will send these unsuccessful retransmission packet. 让s (k)表示在k次尝试之后, 分组从X到y传递成功的可能性。 Let s (k) represents the after k attempts, the packet is passed from X to y likelihood of success. 那么: Then:

[0013] S(IO==Pk-1Ml-P) [0013] S (IO == Pk-1Ml-P)

[0014] 最后,用ETX表示需要成功地将分组从χ传递到y的期望的传送时间: [0014] Finally, ETX expressed the need to successfully pass the packet from χ to y desired delivery time:

O) 1 O) 1

[0015] ETX = Yk*s{k) = ~ [0015] ETX = Yk * s {k) = ~

k=\ ^-P k = \ ^ -P

[0016] 由于ETX在同种的单一无线电网络环境里性能较好,ETX不需要在上述的多无线电方案里选择好的路由。 [0016] Since the ETX radio network environment with a single species in a good performance, ETX does not need to choose a good route in the above-mentioned multi-radio program. 在第一种方案里,每个节点具有802. Ila无线电和802. Ilb无线电,ETX为了两个原因将在802. Ilb链路上路由大量的通信量。 In the first scenario, each node has a radio and 802. Ilb 802. Ila radio, ETX two reasons for the road by a large number of traffic at 802. Ilb link. 第一,ETX只考虑链路上的丢失比率,而不考虑他们的带宽。 First, ETX only considers loss ratio on the link, regardless of their bandwidth. 第二,在使全球资源使用最小化的尝试里,只要较短路径上的丢失比率不是显著的高于较长路径上的丢失比率,ETX被设计为给较短路径优于较长路径的选择权。 Second, the global attempt to minimize the use of resources, as long as the loss ratio on the shorter path is not significantly higher than the loss ratio on the long path, ETX is designed to give a shorter route than the longer path selection Right. 在第二个方案里,每个节点具有两个802. Ilb无线电,由于ETX没有将任可的优先选择提供给信道传输路径,ETX再次选择次优的路径。 In the second scenario, each node has two radio 802. Ilb, since ETX can not be any preference to a channel transmission paths, ETX choose suboptimal path again. 因此,像其它现有的路由技术和路径度量一样,ETX不能从多个无线电的可用性和链路干扰的存在和多跳网络里变化的带宽中获取全部的利益。 So, like other existing routing techniques and path metrics, like, ETX can not radio link interference from multiple usability and presence and multi-hop network in varying bandwidth get all the benefits.


[0017] 根据本发明一方面,提供了多跳特定网络中的一种用于测量从源节点到目标节点的路由的链路质量的系统,所述路由包括多个链路,所述系统包括:用于根据链路的性能, 为路由里的每个链路分配权值的装置,其中,链路的性能包括在链路上传送分组以使分组成功穿过链路的所期望时间的测量值乘以链路的带宽测量值,所述带宽测量值为通过把固定的分组大小除以原始数据率而得到的商;及用于对为路由分配的链路权值进行加权平均得到路径度量的装置,其中所述路径度量说明了使用共享信道的链路之间的干扰。 [0017] According to an aspect of the present invention, there is provided a multi-hop network for a particular link quality measurement route from the source node to the destination node system comprises a plurality of links of the route, the system comprising measuring time for the desired link performance according to, for the route in each link means assigned weights, wherein the link performance including transport packets on the link so that the packet successfully across the link: bandwidth measurement value is multiplied by the value of the link, the bandwidth of the measured value by the fixed packet size divided by the original data rate quotient obtained; and for the link weights assigned to the route by the weighted average path metric The device, wherein the path metric illustrates the use of a shared channel interference between the links.

[0018] 根据本发明另一方面,提供了多跳特定网络中的一种用于测量从源节点到目标节点的路由的链路质量的方法,所述路由包括多个链路,所述方法包括:根据链路的性能,为路由里的每个链路分配权值,其中,链路的性能包括在链路上传送分组以使分组成功穿过链路的所期望时间的测量值乘以链路的带宽测量值,所述带宽测量值为通过把固定的分组大小除以原始数据率而得到的商;及对为路由分配的链路权值进行加权平均得到路径度量,其中所述路径度量说明了使用共享信道的链路之间的干扰。 [0018] According to another aspect of the present invention, there is provided a multi-hop network in one particular measure link quality from the source node to the destination node routing method, comprising a plurality of links, the method of the route comprising: the performance of the link, the routing in the weights assigned to each link, which link performance, including transport packets on the link so that the packet crossing the measured value is multiplied by the expected time link bandwidth measurement value of the link, the bandwidth of the measured value by the fixed packet size divided by the original data rate quotient obtained; and the link weights assigned to the route by the weighted average path metric, wherein the path measure illustrates the use of a shared channel interference between the links.

[0019] 本发明的几个方面提出了一种使用加权的累计期望传输时间路径度量、根据链路质量进行路由的系统和方法。 Several aspects [0019] The present invention proposes a weighted cumulative expected transmission time path metric system and method for routing based on link quality. 一种用于在多跳特定网络里进行路由的系统,及用于测量网络中路由的链路质量的方法,包括根据链路上分组的期望传输时间对链路加权的分配,及将一路由的个别链路权值权值合并到路径度量中。 A method for multi-hop routing in a network of a particular system, and means for measuring the link quality of the route network, including the transmission time according to a desired allocation of uplink packet link weights, and all the way by the The individual link weights weights into the path metric. 路径度量说明了使用一共享信道的链路之间的干扰。 The path metric illustrates the use of a shared channel interference between the links. 在计算路径度量时,增加彼此干扰的链路的期望传输时间,而分别考虑非干扰链路的期望传输时间。 When calculating the path metric increase expected transmission time of each link interference, and non-interfering links are considered desirable transmission time. 本发明的实施例将在多跳网络里进行,所述网络里的节点装备了一个或多个无线电收发器。 Embodiments of the invention will be carried out in a multi-hop network in the network where the nodes are equipped with one or more radio transceivers.

[0020] 本发明可整个的或部分的在软件里、硬件里及其组合中实施。 [0020] The present invention may be in the software, hardware, and combinations implemented in whole or part. 网络里计算设备预期被配置为实现所述系统和执行与本发明相关的方法。 The network computing device is expected to be configured to implement the system and execute methods associated with the present invention.

附图说明 Brief Description

[0021] 图1是根据本发明的实施例,说明为计算ETT测量的丢失比率而由探针发送器节点采取的步骤的流程图。 [0021] Figure 1 is an embodiment of the present invention, a flow chart illustrating the steps for calculating the loss ratio ETT measurement taken by the probe and the node transmitter.

[0022] 图2是根据本发明的实施例,说明由探针接收器对于丢失比率计算采取的步骤的流程图。 [0022] FIG. 2 is an embodiment of the present invention, a flowchart illustrating a probe receivers loss ratio calculated for the steps taken.

[0023] 图3是根据本发明的实施例,说明在为每个链路计算带宽时由探针发送器采取的初始步骤的流程图。 [0023] Figure 3 is an embodiment of the present invention, when calculating a flowchart bandwidth for each link is taken by a probe sender initial step instructions.

[0024] 图4是根据本发明的实施例,说明由探针接收器对于带宽计算采取的步骤的流程图。 [0024] Figure 4 is an embodiment of the present invention, a flowchart illustrating the steps of the receiver by the probe taken by the bandwidth calculation.

[0025] 图5是根据本发明的实施例,说明由探针回复接收器对于带宽计算采取的步骤的流程图。 [0025] Figure 5 is an embodiment of the present invention, a flowchart illustrating steps taken by the receiver calculated by the bandwidth probe response.

[0026] 图6是根据本发明的实施例,说明一给定路径的WCETT的计算中涉及的步骤的流程图。 [0026] FIG. 6 according to an embodiment of the present invention, a flow chart illustrating a given path WCETT calculation steps involved.

[0027] 图7是说明可在其中结合本发明实施例的示例性网络环境的示意图。 [0027] FIG. 7 is a schematic diagram in which the connection with the exemplary embodiments of the invention described network environment.

[0028] 图8是说明可用于本发明一示例性实施例中的示例性计算机设备的示意图。 [0028] FIG. 8 is a diagram that can be used with an exemplary embodiment of the present invention, an exemplary computer equipment instructions.


[0029] 在以下的描述里,将呈现本发明的一些实施例。 [0029] In the following description, the will to present some embodiments of the present invention. 为了说明的目的,阐明特定的结构和细节来对实施例提供一个彻底的理解。 For illustrative purposes, to clarify the specific structure and detail of the embodiments provide a thorough understanding. 然而,对本领域的技术人员而言,显而易见的是也可以不需要这特定细节来实施本发明。 However, those skilled in the art, it is also apparent that the specific details may not be needed to practice the invention. 此外,为了不使描述的实施例晦涩,可以忽略或简化公知的特征。 In addition, in order not to obscure the embodiments described can be ignored or simplified well-known features.

[0030] 本发明的实施例包括一个新的路由度量,即加权的累计期望传输时间(WCETT),用来在多跳网络的源节点和目标节点之间选择高吞吐量路径。 [0030] Embodiments of the present invention includes a new routing metric that is weighted cumulative expected transmission time (WCETT), is used to select high-throughput path between source and destination nodes multi-hop networks. 尽管它不仅限于此,度量在无线网络里尤其具有优势,所述无线网络里的至少一些节点装备了多个同种的无线电收发装置并且所述网络里的一些链路彼此干扰。 Although it is not limited to this, metrics in a wireless network is particularly advantageous, at least some of the nodes of the wireless network in the radio apparatus equipped with a plurality of the same kind and number of links in the network interfere with one another. WCETT度量能够选择信道各异的路径。 WCETT measure can select a channel different path. 基于固定大小的分组的期望传输时间(ETT),对每个链路分配权值以成功的在所述链路上发送。 Based on expectations of a fixed size packet transmission time (ETT), to assign weights to the success of each link are transmitted on the link. 所述ETT是所述链路的丢失比率和带宽的函数。 The ETT is a function of the link loss ratio and bandwidth. 在本发明的可选实施例里,可使用链路质量的测量而不是ETT。 In an alternative embodiment of the present invention, the link quality measurement may be used instead of ETT. 个别链路权值被合并成WCETT路径度量,所述度量明确说明了使用同一信道的链路之间的干扰。 Individual link weights are combined into WCETT path metric, the metric explicitly use the same channel interference between the links. 可使用控制参数β来调节信道多样性和路径长度之间的权衡。 Can be used to adjust the control parameter β weigh channel diversity and path length between. 可以调节WCETT的计算或者使给定流的吞吐量最大化,或者使其对其它流的影响最小。 WCETT calculation can be adjusted or so to maximize the throughput of a given flow, or its impact on other flow is minimal.

[0031] 本发明的一些实施例被设计为应用在无线网络里,该无线网络里的节点具有遵循802. 11的无线电。 [0031] Some embodiments of the present invention is designed for applications in the wireless network, the wireless network node has to follow in the 802.11 radio. 然而,其它实施例可应用在基于其它无线技术的网络里。 However, other embodiments can be applied to other wireless technologies, based on the network. 本发明可与一个或多个其它技术或机制一起使用以提高多跳无线网络里的网络容量,诸如剥离(striping) 技术或定向天线的使用。 The present invention can be used with one or more other techniques or mechanisms to improve the multi-hop wireless network in network capacity, such as stripping (striping) using technology or directional antenna. 此外,由于在此描述了用于无线环境的实施例,因此本发明提供了一种更通用和广泛的应用技术,以便将路由总成本的测量和路由瓶颈组件的测量合并,以生成路径度量。 Furthermore, since the embodiment described herein for the wireless environment, the present invention provides a more versatile and widely used technique for the measurement of the total cost routing and routing bottleneck component measuring combined to generate path metric.

[0032] 为了计算WCETT路径度量,首先需要为给定的链路计算ETT。 [0032] In order to calculate WCETT path metric, we first need to compute ETT for a given link. 在一个实施例里,通过将链路期望传输的时间(如ETX)乘以链路带宽的测量来得到ETT,以得到发送分组所花费的时间的测量。 In one embodiment, a desired transmission link by the time (e.g., ETX) of the link bandwidth multiplied by the measurement obtained ETT, to obtain a measure of the time it takes to send the packet. 以ETX为例,所述ETT变成: In ETX example, the ETT becomes:

[0033] ETT = ETX* S/B [0033] ETT = ETX * S / B

[0034] 其中S是固定的分组大小(如IOM字节),B表示链路的带宽(原始的数据率)。 [0034] where S is a fixed packet size (such as IOM bytes), B indicates that the link bandwidth (raw data rate). 其它实施例可使用期望传输时间的求解(derivation)而不是ETX求解来计算ETT。 Other embodiments may be used to solve a desired transmission time (derivation) rather than solving ETX calculated ETT. 所述期望传输时间(ETT)不限于使用ETX来测量,可以使用从发送成功概率得到的其它计算方法。 The desired transmission time (ETT) is not limited to the use of ETX to measure, you can use other methods to calculate the probability of success is sent from the get.

[0035] 计算ETT需要知道各个链路的正反丢失比率(分别是Pf和P,)和带宽。 [0035] calculate ETT need to know the pros and cons of each link loss ratio (respectively Pf and P,) and bandwidth. 用来确定Pf和P,值的一个方法是使用广播分组技术,诸如是上文所参考的、由De Couto等人所公开的技术。 Pf and used to determine the P, a method is to use the value of a broadcast packet technique such as the above referenced by the De Couto et al disclosed technique. 在这个方法里,每个节点周期地(例如每秒一次)发出一个广播分组。 In this method, each node periodically (e.g., once per second) sends a broadcast packet. 注意的是所述广播分组不被802. IlMAC重发。 Note that the broadcast packet 802. IlMAC not be retransmitted. 在滑动时窗(例如是10秒)期间,节点追踪从每个相邻节点接收到的探针数量,并将该信息包括在它们自己的探针里。 In the sliding window (for example, 10 seconds), the node tracks received from each neighboring node to the number of probes, and include that information in their own probe inside. 节点可以从它在时间窗口里从相邻节点接收到的探针数量直接计算出&,并且该节点可以使用在来自相邻节点的最后一个探针里接收到的关于它自己的信息来计算Pf。 Node can be calculated directly from it receives from neighboring nodes at the time window to the number of probes &, and the node can be used in the last probe from a neighbor in the received information on its own to calculate Pf . 其它确定Pf和已值的方法在不离开本发明的精神和范围时也可使用。 Other Pf and methods of determining value when they are not departing from the spirit and scope of the present invention may also be used. 例如,可以对网络接口卡(NIC)上的重发计数器作出询问来发现正反丢失比率。 For example, you can ask for retransmission counter to the network interface card (NIC) to discover the pros and cons on the loss ratio.

[0036] 图1示出了由探针发送器在为ETF测量计算丢失比率时所采取的步骤。 [0036] Figure 1 illustrates steps by a probe sender in calculating the loss ratio for the ETF measurement taken. 节点等待一个周期的时间(丢失探针周期)(步骤101)。 Node waits a period of time (loss of probe cycle) (step 101). 所述节点计算从前一次发送的探针开始从每个相邻节点接收到的探针数量(步骤10,并创建包括这个信息的探针(步骤10。该节点也为到每个相邻节点的链路计算丢失比率(步骤107)。所述节点广播所述探针(步骤109)并返回步骤101等待。 The probe was once computing node starts to receive transmitted from each neighboring node Number of the probe (step 10, including the probe information and creating (step 10. The node is also adjacent to each node Link loss ratio calculated (step 107). The node broadcasts the probe (step 109) and returns to step 101 to wait.

[0037] 图2示出了由接收器对丢失比率计算而采取的步骤。 [0037] FIG. 2 shows the steps by the receiver to calculate the loss ratio taken. 节点接收探针(步骤201)。 Node receives a probe (step 201). 所述节点查找和已经被发送器接收到的探针有关的信息(步骤20。它更新从发送器接收到的探针数量(步骤20并返回步骤201。 Find and has been receiving information about the sender to probe (step 20. It updates received from the transmitter to the number (step 20 probe and returns to step 201 of the node.

[0038] 计算带宽是一个更复杂的问题。 [0038] calculate the bandwidth is a more complex issue. 一种可能是将每个802. 11无线电的带宽限制为一个给定的固定值。 One possibility is that each of the 802.11 radio bandwidth is limited to a given fixed value. 另一个可能是允许无线电为每个分组自动地选择带宽。 Another possibility is to allow each packet radio bandwidth is automatically selected. 这个特征,通称为自动比率(autorate),被许多现代802. 11卡支持。 This feature, known as automatic ratio (autorate), by many modern 802.11 cards. 然而,在自动比率算法不能为公众熟知以及带宽信息不能被驱动程序支持的802. 11卡的情况下,带宽的精确测量只能凭经验来获得。 However, in the case of automatic ratio algorithm can not be known by the public as well as the bandwidth of the information can not be supported by the driver 802.11 cards, accurate measurement bandwidth can be obtained empirically.

[0039] 在一个实施例中,使用熟知的分组对技术来获得带宽的经验测量。 [0039] In one embodiment, using well-known techniques to obtain the bandwidth for packet empirical measurements. 例如,在S. Keshav 所著的“A Control-Theoretic Approach to Flow Control,”ACMSIGC0MM(1991 年9月)中对所述分组对技术进行了描述,并引用在此以作参考。 For example, S. Keshav book "A Control-Theoretic Approach to Flow Control," ACMSIGC0MM in (September 1991) to the packet technology are described, and incorporated herein by reference. 每个节点周期地(例如每分钟)对每个相邻节点发送出两个紧接着的(backto back)探针分组。 Each node periodically (eg, every minute) for each adjacent node sends out (backto back) probe packet two immediately. 第一个探针分组很小(例如是137字节),而第二个分组很大(例如是1137字节)。 The first probe packet is small (e.g., 137 bytes), and the second packet is large (e.g. 1137 bytes). 每个相邻节点测量在第一个分组和第二个分组的接收之间的时间差,并将所述值传回给发送器。 Each neighboring nodes measuring time of the first packet and the second packet is received difference between and the value back to the sender. 发送器取一组连续样本的最小值(例如10个连续样本的最小值),然后通过把第二个探针分组的大小除以最小样本来估计带宽。 The transmitter takes a set minimum value (e.g., 10 consecutive samples minimum) consecutive samples, then by the second probe packet size divided by the sample to estimate the minimum bandwidth. 这个估计忽视了几个影响分组传递时间的因素,但对于实质上不同带宽的链路之间的区分,这是足够精确的。 This estimate ignores the several factors affect packet delivery time, but for distinguishing between the link bandwidth substantially different, which is sufficiently accurate.

[0040] 图3示出了由探针发送器在为每个链路计算带宽时采取的步骤。 [0040] Figure 3 illustrates steps by a probe sender in calculating the bandwidth for each link is taken. 节点等待一个周期(分组对探针周期(Pktpair Probe Period))(步骤301)。 Node waits for one cycle (cycle probe packet (Pktpair Probe Period)) (step 301). 它为相邻节点创建探针对(步骤30。所述节点然后紧接着发送所述探针对(步骤30,然后返回步骤301。 It creates a probe adjacent nodes (step 30. The node then immediately send the probe on (step 30, and then returns to step 301.

[0041] 图4示出了由探针接收器在带宽计算中采取的步骤。 [0041] FIG. 4 shows the steps taken by the probe the receiver bandwidth calculation. 节点接收探针(步骤401)。 Node receives a probe (step 401). 如果探针不是一对探针中的第一个探针(步骤40,那么所述节点放弃所述探针(步骤405)并返回到步骤401。否则,所述节点接收第二探针(步骤407)。如果这个探针不是一对探针中的第二个(步骤409),那么所述节点放弃所述探针(步骤40并返回至步骤401。 否则,所述节点计算在第一探针和第二探针的接收之间的延迟(步骤411)。然后节点创建探针回复(步骤41,并将所述探针回复发送给发送器(步骤41。然后节点返回到步骤401。 If the probe is not a pair of probes in the first probe (step 40, then give up the node of the probe (step 405) and returns to step 401. Otherwise, the node receives a second probe (step 407). If the probe is not a pair of probes in the second (step 409), then the node to abandon the probe (step 40 and returns to step 401. Otherwise, the node calculates a first probe delay (step 411) received between the needle and the second probe. Then node to create a probe reply (step 41, and the reply of the probe to the transmitter (step 41. Then node returns to step 401.

[0042] 图5示出了由探针回复接收器在带宽计算时采取的步骤。 [0042] FIG. 5 shows the steps taken by the probe reply to the receiver when the bandwidth is calculated. 节点接收探针回复(步骤501)。 Node receives a probe reply (step 501). 它将延迟值转换成带宽样本(步骤50。如果这个带宽值比最大带宽大(步骤505),所述节点就放弃这个样本(步骤507)。否则,它将更新到目前为止最小的延迟值(步骤509)。如果依然没有收集到全部的一组样本(步骤511),那么节点返回到步骤501。否则,该节点为这个链路更新带宽测量并将样本数量设为0 (步骤51,然后返回到步骤501。 It converts the delay value to the bandwidth sample (step 50. If the bandwidth values greater than the maximum bandwidth (step 505), the node will abandon this sample (step 507). Otherwise, it updates the minimum delay value so far ( step 509). If you still have not collected all of a set of samples (step 511), then the node returns to step 501. Otherwise, the node for the link bandwidth measurement and update the number of samples is set to 0 (step 51, then return to step 501.

[0043] 一旦计算了从节点χ到节点y的每个链路i的ETT,即ETTi,就必须将沿着一路径的跳跃的个别ETT链路权值合并成一度量WCETT,该度量反映该路径的整个需求。 [0043] Once calculated from node to node y χ ETT for each link i, that ETTi, you have to combine the individual ETT link along jump right into a value of a path metric WCETT, the measure reflects the overall demand path. 对于WCETT路径度量有三个基本的设计目标。 For WCETT path metric There are three basic design goals. 第一,当选择包含在路由路径中的链路时,路径度量应当考虑链路带宽以及物理层的丢失比率。 First, when you select the link included in the routing path, the path loss ratio should be considered a measure of link bandwidth and physical layer. 例如,在802. 11网络中,由于802. IlMAC结合了ARQ机制,因而在无线链路上的分组传输时间取决于这两者因素。 For example, in a 802.11 network, since the combination of the ARQ mechanism 802. IlMAC, and thus the packet transmission time on a wireless link depends on both factors.

[0044] 第二,应当增加合并个别链路权值的度量。 [0044] Second, the merger should be increased measure of individual link weights. 也就是说,如果将一跳跃增加到现有路径里,路径的成本不应该降低,而是应该增加。 That is, if a jump to an existing path, the path should not reduce costs, but should be increased. 对于这个需求的一个判断方法是,附加跳跃的遍历包括了附加资源的消耗。 The requirement for a judgment method, the additional jumping traversal includes additional resource consumption. 另一个判断方法是,跳跃的附加增加了沿路径的总延迟。 Another judge is to jump an additional increase in the total delay along the path. To

8于TCP连接,这将产生增加的往返时间并因此降低吞吐量。 8 TCP connections, which will have increased round-trip time and hence reduced throughput.

[0045] 第三,由于运行在相同信道上的链路间的干扰,路径度量应该明确解决吞吐量的下降。 [0045] Third, due to the interference between the link run the same channel, the path metric should clearly address degradation in throughput. 相似地,它也应当解决沿着一路径的、不在相同信道上操作、彼此没有干扰的链路的情况。 Similarly, it should be addressed along a path, not on the same channel operation, without interfering with each other in the case of the link. 因此,由不同信道上的跳跃所组成的路径优先于所有跳跃在同一信道上的路径。 Therefore, the path from the jump on different channels composed precedence over all paths in the same channel hopping.

[0046] 为了符合这些设计目标,当附加链路增加到现有路径中时WCETT值应当增加。 [0046] In order to meet these design goals, while additional link to an existing path when WCETT value should be increased. 可通过把WCETT设为路径上所有跳跃的ETT的总和来确保这一特性。 WCETT set by the sum of all ETT jumping on the path to make sure this feature. 然而,这样却不正确地假设了所有链路都是干扰的。 However, this is not correct to assume that all links are interference. WCETT的另一个目标是反映信道差异的影响。 Another goal is to reflect the impact of WCETT channel differences. 由于没有区分不同信道上的跳跃之间的区别,因此简单地累计ETT将不能保证该特性。 Since there is no distinguish between jumping between different channels, so simply accumulated ETT will not guarantee this feature. 因此WCETT度量需要附加项。 So WCETT measure requires additional items.

[0047] 首先我们假设,如果路径上的两个跳跃在同一信道上,那么它们通常彼此干扰。 [0047] First, we assume that if the two jumps on the path on the same channel, they usually interfere with each other. 这个假设对于校短的路径一般精确,但对较长路径则稍微有点不容乐观。 This assumption is generally shorter path for school accurate, but a little bit longer path is not optimistic. 当路径上的两个跳跃互相干扰时,在一个时间里只可操作一个跳跃。 When two jumps paths interfere with each other, at a time only operable one jump. 这可以通过将干扰跳跃上的分组传输时间加在一起来获得。 This can be achieved by interfering packet transmission time jumping on added together to obtain. 为了概括起见,考虑具有η个跳跃的路径,并假设所述系统总共具有k 个信道,定义&如下: To summarize sake, consider having η hop path, and assume that the system has a total of k channels, & defined as follows:

[0048] [0048]

Xj= Y^ETT ,\< j<k Xj = Y ^ ETT, \ <j <k

跳i在信道j上 I jump on channel j

[0049] Xj是信道j上跳跃的传输时间的总和。 [0049] Xj channel j is the sum of the transmission time jump. 总路径吞吐量将由具有最大Xj值的瓶颈信道来决定。 The total path throughput will be determined by having the biggest bottleneck channel Xj values. 基于这个,一个建议是可以使用该最大值Xj作为WCETT的定义。 Based on this, a suggestion is to use the maximum value as defined WCETT of Xj. 通过这样定义, 所述度量将支持具有更多不同信道的路径。 By this definition, the measure will support paths with more different channels. 然而,可以看到,由于使用非瓶颈信道的附加跳跃不会影响度量的值,因而这个度量值将不能随着更多的跳跃加入到路径而一直增加。 However, it can be seen that the use of non-bottleneck channel does not affect the value of the additional skip measure, therefore this measure will not jump as more is added to the path has been increased.

[0050] 先前参考的两个路径度量的预期特性(ETF的总和以及最大Xj)可以通过采用它们的权值平均数来合并成一个满意的度量: [0050] the desired characteristics of the two previous reference path metric (the sum of the ETF and the maximum Xj) by using the value of the average of their right to merge into a single measure of satisfaction:

[0051 ] WCETT = (IP)^YjETTi+β* max Xj [0051] WCETT = (IP) ^ YjETTi + β * max Xj

ϊ=1 众 ϊ = 1 public

[0052] 因此,WCETT度量是两个数值的加权平均值:沿路径的所有跳跃的ETT的总和,及瓶颈信道上ETT的总和。 [0052] Thus, WCETT measure is a weighted average of the two values: the sum of all ETT jump along the path, and the sum of the bottleneck channel ETT. 第二个数值迫使不同信道路径的选择。 The second value forced to select a different channel paths. β是符合β 的可调参数。 β is in line with the tunable β. 选择β为0.5时可以给不同信道和ETT的总和提供相等的权值。 Select β of 0.5 can provide equal to the sum of the weights of different channels and the ETT. 如果将β选择为0,WCETT仅基于链路的丢失比率和带宽来选择链路,而不考虑不同信道。 If β is selected as 0, WCETT only link-based loss ratio and bandwidth to select a link, regardless of the different channels. 在一个实施例里,β的选择可以自动的基于当前网络带宽。 In one embodiment where, β selection automatically based on the current network bandwidth.

[0053] 加权平均的WCETT表达式可以用两种方法来解释。 [0053] The weighted average WCETT expression can be interpreted in two ways. 第一,它可以被看作是全局利益和利己主义之间的权衡。 First, it can be seen as a trade-off between global interests and egoism. 第一项是网络里沿着所有跳跃的传输时间的总和。 The first item is the sum of the network along all the transmission time jump. 这反映了沿这个路径的总的网络资源消耗。 This reflects along this path of total network resource consumption. 第二项反映了将最大限度地影响路径吞吐量的一组跳跃。 The second reflects the impact will maximize the throughput of a group jump path. 于是可以把加权平均值看作试图是平衡这两者。 The weighted average can then be viewed as trying to balance the two. 也可以将所述等式看作是吞吐量和延迟之间的权衡。 The equation can also be seen as a trade-off between throughput and latency. 可以将第一个项看作是路径等待时间的测量。 The first item can be regarded as the path wait time measurement. 由于第二项表示瓶颈跳跃的影响, 因此它可以被看作为路径吞吐量的测量。 Since the second term represents the bottleneck effect jumps, so it can be viewed as a path throughput measurements. 因此,也可以将加权平均值看作为试图平衡这两者ο Therefore, it can be seen as trying to balance the weighted average of both ο

[0054] 图6示出了对于一给定路径、在WCETT计算中涉及的步骤。 [0054] FIG. 6 shows that for a given path, the steps involved in WCETT calculation. 对于路径上的每个链路,节点利用丢失比率和带宽来计算ETT (步骤601)。 For each link on the path node and bandwidth utilization loss ratio is calculated ETT (step 601). 节点计算ETT的总和(步骤60。然后根据信道将链路分组并为每个组计算ETT总和(步骤60幻。所述节点查找具有最大ETT总和的组(步骤607)。最后,节点为每个路径计算WCETT,具体是由(l-β)乘以ETT的总和,再加上β乘以信道组里ETT总和的最大值。 Node computes ETT sum (step 60. Then according to the channel link packets and calculate ETT sum (step 60 magic. Find the node having the maximum ETT sum set (step 607) for each group. Finally, for each node path calculation WCETT, specifically ETT sum by (l-β) multiplied, plus β multiplied by the maximum ETT sum channel group.

[0055] 图7示出了可以实施本发明实施例的示例性网络环境。 [0055] FIG. 7 shows an exemplary embodiment of a network environment may be an embodiment of the present invention. 所述示出的网络部分地是多跳无线网络。 The network shown in part a multi-hop wireless network. 如所描述的,所述网络包括无线节点701、703、705、707、709、711、713、715、 717、719。 As described, the network includes wireless nodes 701,703,705,707,709,711,713,715, 717,719. 作为典型,所描述的无线网络不能独立于其它网络而运行;如示出的,所述网络通过网关设备723、721连接至通常是较大网络的第二网络,例如因特网,所述网关设备通过有线通信装置725连接至第二网络。 As typical, the wireless network can not be described independently of the operation of other networks; As shown, the network connection through the gateway device typically 723,721 to a second network of a larger network, such as the Internet, via the gateway device wired communication means 725 is connected to the second network. 所述破折线表示现有的无线电链路。 The broken line indicates existing radio link. 诸如节点701这样的节点是计算设备,可具有许多通用或专用的配置和结构,这对本领域的技术人员是公知的。 Such a node, such as node 701 is a computing device, may have a number of generic or specific configuration and structure, which is skilled in the art are well known. 通常,这些设备将至少具有处理器和存储器分级体系,及有线和无线网络接口硬件。 Typically, these devices will have at least a processor and a memory hierarchy, and wired and wireless network interface hardware. 尤其是,这个网络里的节点可以具有一个或多个数字分组无线电收发器,以便穿越频率信道而没有限制的进行无线通信,例如,这些包括在网络接口卡里的无线点收发器可以适用于802. Ila 和802. Ilg IEEE 标准。 In particular, this network node can have one or more digital packet radio transceivers, so the crossover frequency channel without restriction for wireless communications, for example, which includes a network interface card of wireless point transceiver can be applied to 802 . Ila and 802. Ilg IEEE standard.

[0056] 由于无线电具有有限的无线通信范围,因而这个网络里的许多对节点不能直接进行通信,而是必须依赖所述网络里的一个或多个合作节点作为中介节点来从源节点向目标节点转发数据。 [0056] Since the radio has a limited wireless communication range, and thus the number of nodes in the network can not communicate directly, but must rely on the network in one or more co-nodes as an intermediary node from the source node to the destination node forwarding data. 因而,源节点一般将数据分组传送给它可以直接通信的相邻节点。 Thus, the source node typically transmits the packet data to neighboring nodes it can communicate directly. 所述相邻节点顺序的将所述分组传送给它的相邻节点中的一个,如此,直到将所述分组传送给它的最后目标。 Said adjacent node order the packet to its neighboring nodes a, and so on until the packet to its final goal. 分组发送所越过的每条链路称为“跳跃”。 The packet transmission across each link called "jump." 分组从源节点到最终目标节点所经历的链路组是所述分组的路由或路径。 Packet from the source node to the final destination node through a link set is the route or path a packet. 可以通过以分布方式运行在多跳网络的几个节点上的路由协议来发现路由。 You can find the route runs through a distributed manner on several multi-hop network node routing protocol. 在图7中,描述了两个示例性路由。 In Figure 7, we described two exemplary routes. 节点S1707将分组传送给中介节点R1709,中介节点R1709再将分组传送给中介节点R2703,中介节点R2703将所述分组传送给目标节点D 711。 Node S1707 packet transmitted to the intermediary node R1709, R1709 intermediary node packet to intermediate node then R2703, R2703 intermediary node the packet to the destination node D 711. 这些节点之间的实心黑色箭头线表示包括这个路由的数据通信链路。 Solid black arrow lines between these nodes represent the data including the route of the communication link. 节点S2717向中介节点R3719发送去往外部网络的分组,中介节点R3719将所述分组传送给网关723。 Node S2717 sends packets destined for the external network to intermediate node R3719, R3719 intermediary node will deliver the packet to the gateway 723.

[0057] 在本发明的实施例里,WCETT度量与MR_LQSR(多无线电链路高质源路由)一起实施,该MR-LQSR是运行在网状连通层(MCL)路由框架里的链路质量源路由(LQSR)协议的一个修改版本。 [0057] In an embodiment of the present invention in, WCETT metric MR_LQSR (multi-radio link quality source routing) together with the implementation of the MR-LQSR is run in a mesh link quality source communication layer (MCL) routing framework of A modified version of the route (LQSR) protocol. 2004年2月幻日申请的美国专利申请号为10/784,687的“System and Method for Link Quality Source Routing”公开了LQSR的一些特征,上述申请与本申请具有共同的发明人,并在此引用以作参考,以及在2003年6月30日申请的美国专利申请号为10/610,397 的“Methodand System for Providing a Virtual Protocol Interlayer,,中公开了MCL —些特征,上述申请与本申请是共同的发明人,并在此引用以作参考。这些特征中的部分也在RP Draves,J. Padhye,禾口BD Zill 的"Comparison of Routing Metrics for StaticMulti-Hop Wireless Networks" Technical R印ort (技术报告)MSR-TR-2004-18, Microsoft Research (微软公司研究报告)(2004年3月)中有论述,并在此引用以作参考。 U.S. Patent Application No. February 2004 magic date for applications is 10 / 784,687 for "System and Method for Link Quality Source Routing" feature LQSR are disclosed, said with the present application have a common inventor, and in this cited by reference, and in US Patent Application No. 30 June 2003 Application 10 / 610,397 for "Methodand System for Providing a Virtual Protocol Interlayer ,, disclosed MCL - these features above with the present application common inventor, and hereby incorporated by reference. These features are part of RP Draves, J. Padhye, Hekou BD Zill's "Comparison of Routing Metrics for StaticMulti-Hop Wireless Networks" Technical R India ort ( Technical Report) MSR-TR-2004-18, Microsoft Research (Microsoft Research Report) () have discussed in March 2004, and incorporated herein by reference.

[0058] MCL实现了一个处于层2(链路层)和层3(网络层)之间的中间层内的虚拟网络适配器。 [0058] MCL implements a virtual network adapter in the middle layer Layer 2 (link layer) and Layer 3 (network layer) inside. 对较高层应用,MCL看起来是另一个以太网链路,具有自己的48位虚拟以太网地址,不同于底层物理适配器的层2地址。 For higher-level applications, MCL appears to be another Ethernet link, with its own 48-bit virtual Ethernet address is different from the underlying physical adapter layer 2 address. MCL对较低层则看起来是另一个运行在物理链路上的协议。 MCL lower layer protocol appears to have been another run on a physical link. 多个物理网络适配器可以复合成单一虚拟链路,网络协议可以在该单一虚拟链路上运行而无须修改。 Multiple physical network adapters combined into a single virtual link, network protocols can run without modification on the single virtual link.

[0059] MCL适配器使用MR-LQSR对分组进行路由。 [0059] MCL adapter using MR-LQSR the packet routing. 对于给定的节点,通过合并链路权值来形成路径度量,MRLQSR协议发现节点的相邻节点,向节点与相邻节点之间的链路分配权值,将这个信息传播给网络里的其它节点,并使用所述链路权值来查找到达给定目标的好路径。 For a given node, to form a path metric by combining the link weights, MRLQSR protocol discovery of an adjacent node, assign weights to the link between the node and the adjacent node, will disseminate this information to the network of other node, and use the link to find the right value for the given target path to good. 读者可以参考以上所引用的相关专利申请和举证的技术报告来获得有关MCL和LQSR 的其它细节。 Readers can refer to the technical report related patent applications cited above and the burden of proof to obtain additional details about the MCL and LQSR. 然而本发明不限于基于MCL和LQSR的实施。 However, based on the implementation of the present invention is not limited to the MCL and LQSR. 在本发明的其它实施例里可使用其它路由协议和框架。 In other embodiments of the present invention may be used in other routing protocols and frameworks.

[0060] 为了实施WCETT,与每个链路的丢失比率和带宽或ETT —起传送信道号是必须的。 [0060] In order to implement WCETT, with each link of the loss ratio and the bandwidth or ETT - from the delivery channel number is necessary. 在一个实施例里,度量值的8位用来编码为抽象的信道号。 In one embodiment, the metric of eight used to encode abstract channel number. 传送信道号的其它方法也是可能的。 Other methods of transmission channel numbers are also possible.

[0061] 在本发明描述的上下文中,术语“一” “和“这”及相似情况的使用解释为覆盖单数和复数,除非这里有指定或文本里有清楚的矛盾。术语“包括”、“具有”、和“包含”解释为开放式结束术语(意为“包括,但不限于”),除非有其它标注。这里值的范围的列举仅意指作为一种单个的参考每个落入这个范围之内的分离值的速记方法,除非这里有指定,并且每个分离值都被合并进入本说明书,就像在此分别举证的一样。这里使用的任何和所有的示例或示例性语言(“诸如”,“例如”,“如”等)仅仅意图更好的说明本发明,并没有在本发明的范围上指定限制,除非有其它请求。说明书里没有语言应当解释为指定任何非请求的元素为本发明的实施的本质。 [0061] In the context of the present invention described herein, the terms "a", "and" this "and use similar situations construed to cover both the singular and the plural, unless there is designated or text there is a clear contradiction. The term" comprising "," have "and" with "be interpreted as an open ended terms (meaning" including, but not limited to "), unless otherwise marked. ranges of values listed here only meant as a reference for each individual to fall into this shorthand method of separating value within the range, unless there is specified, and each separate value are merged into the present description, as in the proof of this are the same as used here, any and all examples, or exemplary language (" such as "," for example "," such as ", etc.), is intended merely to better illustrate the present invention, there is no specified limit on the scope of the invention, unless otherwise requested. There is no language specification should be construed as any element designated unsolicited the nature of the implementation of the present invention.

[0062] 参见图8,示出了用来实现本发明的示例性系统,包括计算设备,诸如计算设备800。 [0062] Referring to Figure 8, there is shown used to implement the exemplary system of the present invention, includes a computing device, such as computing device 800. 在基本的配置里,计算设备800 —般包括至少一个处理单元802和系统存储器804。 In the basic configuration, the computing device 800 - as including at least one processing unit 802 and system memory 804. 根据准确的配置和计算设备的类型,系统存储器804可以是易失性的(诸如RAM)、非易失的(诸如ROM、快速存储器等)或这两个的结合。 Depending on the type and the exact configuration of the computing device, system memory 804 may be volatile (such as RAM), non-volatile (such as ROM, flash memory, etc.) or a combination of both. 系统存储器804—般包括操作系统805,一个或多个应用程序806,并也可以包括程序数据807。 System memory 804 typically includes an operating system 805, one or more application programs 806, and program data 807 may also include. 这个基本的配置在图8中由虚线808中的这些部件来示出。 This basic configuration shown in FIG. 8 by a dotted line 808 shows these components come.

[0063] 计算设备800也可以具有额外特征或功能。 [0063] Computing device 800 may also have additional features or functionality. 例如,计算设备800可包括额外的数据存储设备(可移动和/或不可移动),例如磁盘,光盘或磁带。 For example, computing device 800 may include additional data storage devices (removable and / or non-removable) such as disks, optical disks or tapes. 这个额外存储器在图8中由可移动存储器809和不可移动存储器810来示出。 This additional memory in FIG. 8 by removable storage 809 and non-removable memory 810 is shown. 计算机存储媒体可包括为了信息的存储,在任何方法和技术中实现的易失和非易失的,可移动和不可移动媒体,所述信息诸如是计算机可读指令、数据结构、程序模块或其它数据。 Computer storage media for storing information may include, in any of the methods and techniques implemented in volatile and nonvolatile, removable and non-removable media, the information such as computer readable instructions, data structures, program modules, or other data. 系统存储器804、可移动存储器809和非可移动存储器810是计算机存储媒体的所有示例。 System memory 804, removable storage 809 and non-removable storage 810 are all examples of computer storage media. 计算机存储媒体包括,但不限于,RAM、 R0M、EEPR0M闪存或其它存储器技术、CD-ROM,数字化视频光盘(DVD)或其它光学存储器、磁带盒、磁带、磁盘存储器或其它磁性存储设备,或任何其它可以用来存储期望信息并可以被计算设备800访问的媒体。 Computer storage media include, but are not limited to, RAM, R0M, EEPR0M flash memory or other memory technology, CD-ROM, digital video disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any the other can be used to store the desired information and which can be accessed by computing device 800 media. 任何这些计算机存储媒体可以是设备800的一部分。 Any such computer storage media may be part of device 800. 计算设备800也可具有输入设备812,诸如键盘、鼠标、笔、声音输入设备、触摸输入设备等。 Computing device 800 may also have input device 812, such as a keyboard, mouse, pen, voice input device, touch input devices. 也可以包括诸如是显示器、扬声器、打印机等的输出设备814。 It may also be included, such as a display, speakers, printer, etc. The output device 814. 所有的这些设备都是本领域公知的并不需要在这详细论述。 All these devices are known in the art need not be discussed in detail in this.

[0064] 计算设备800也包括允许所述设备与其它计算机设备818例如在网络或无线网状网络上进行通信的通信连接816。 [0064] The computing device 800 may also include allowing the device and other computers to communicate with a communication device 818 such as 816 connections on a network or a wireless mesh network. 通信连接816是通信媒体的一个示例。 Communication connection 816 is one example of communication media. 通信媒体一般包含计算机可读指令、数据结构、程序模块或在诸如是载波或其它传输机构的已调制数据信号里的其他数据,并包括任何信息传递媒体。 Communication media generally embodies computer readable instructions, data structures, program modules or as a carrier wave or other transport mechanism modulated data signal in the other data, and includes any information delivery media. 术语“已调制数据信号”意指具有一个或多个特征集的信号,或以这种方式改变以便在信号里编码信息。 The term "modulated data signal" means a signal having one or more of its characteristics set or changed in such a manner to encode information in the signal. 以举例的方式,但不限制,通信媒体包括诸如是有线网络或直接有线连接的有线媒体,和诸如是声学的、RF,红外线的和其它无线媒体及其它无线媒体。 By way of example but not limitation, communication media includes wired media such as a wired network or direct-wired connection, and such as are acoustic, RF, infrared and other wireless media and other wireless media. 这里使用的术语“计算机可读媒体”包括存储媒体和通信媒体。 As used herein, the term "computer-readable media" includes storage media and communication media.

[0065] 根据本发明,在一个实施例里,应用程序806进一步包括应用程序820来执行网状网络功能。 [0065] According to the present invention, in one embodiment, applications 806 further include an application 820 to perform a mesh network functions. 应用程序820表示的功能可以由包括在计算设备800中的、用来建立和维持特定网络的附加输入设备812、输出设备814和通信连接816来进一步支持。 Functionality of the application can be represented by the 820 included in the computing device 800, and to establish and maintain a specific network of 812 additional input devices, output devices and communication connections 814 816 for further support.

[0066] 已经在此描述了本发明的优选实施例,所述实施例包括为发明人所知的用来实施本发明的最佳方式。 [0066] have been described in the preferred embodiment of the present invention, the embodiment of the invention includes a person known to the best embodiment of the present invention. 在这些优选实施例上的变化对阅读了前述描述的本领域的普通技术人员而言是显而易见的。 Changes in these cases the preferred embodiment of ordinary skill in reading the foregoing description of the art in terms of the obvious. 发明人期望技术人员合理的使用这些变化,并且发明人提供的本发明除这里指定的描述之外可以实施。 The inventors expect skilled and rational use of these changes, and in addition to specified herein described embodiment of the present invention may be provided by the invention. 因此,在法律允许的范围内,本发明包括在所附权利要求里引用的要点的所有变化和替换。 Therefore, within the scope permitted by law, the present invention includes all changes and modifications in the appended claims in reference to the main points. 此外,在其可能的变化中的上述要素的任何可能的结合都包含在本发明之内,除非这里有指定或在上下文中有明显的矛盾。 In addition, any possible combination of these elements in its possible changing are included in the present invention, unless there is a clear contradiction specified or in context.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
CN1483266A31 Aug 200117 Mar 2004诺基亚网络公司Multicast route in AD-HOC network
US2004/0025018 Title not available
US630455624 Aug 199816 Oct 2001Cornell Research Foundation, Inc.Routing and mobility management protocols for ad-hoc networks
International ClassificationH04L12/28, H04L12/56
Cooperative ClassificationH04W40/02, H04W16/14, H04W84/18, H04L45/00, H04L45/124
European ClassificationH04W40/02, H04L45/00, H04L45/124
Legal Events
22 Mar 2006C06Publication
11 Jul 2007C10Entry into substantive examination
18 May 2011C14Grant of patent or utility model
20 May 2015C41Transfer of patent application or patent right or utility model
20 May 2015ASSSuccession or assignment of patent right
Effective date: 20150430
10 Aug 2016EXPYTermination of patent right or utility model