CA2170893C - Method and device for controlling time slot contention to provide fairness between a plurality of types of subscriber units in a communication system - Google Patents
Method and device for controlling time slot contention to provide fairness between a plurality of types of subscriber units in a communication system Download PDFInfo
- Publication number
- CA2170893C CA2170893C CA002170893A CA2170893A CA2170893C CA 2170893 C CA2170893 C CA 2170893C CA 002170893 A CA002170893 A CA 002170893A CA 2170893 A CA2170893 A CA 2170893A CA 2170893 C CA2170893 C CA 2170893C
- Authority
- CA
- Canada
- Prior art keywords
- time slot
- subscriber unit
- packet
- predetermined
- subscriber
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W74/00—Wireless channel access, e.g. scheduled or random access
- H04W74/08—Non-scheduled or contention based access, e.g. random access, ALOHA, CSMA [Carrier Sense Multiple Access]
- H04W74/0833—Non-scheduled or contention based access, e.g. random access, ALOHA, CSMA [Carrier Sense Multiple Access] using a random access procedure
Abstract
The present invention provides a unique access control scheme (100) for assigning a plurality of time slots in a fair manner among subscriber units (508, 514) using a single time slot in a frame and subscriber units (508, 514) using multiple time slots in a frame in a communication system.
Description
2 t 7 a ~ ~ PCT/U595/05949 METHOD AND DEVICE FOR PROVIDING TIME SLOT CONTENTION
FAIRNESS BETWEEN SUBSCRIBER UNITS
Field of the Invention The present invention relates generally to control of time slot usage in a communication system and, more particularly, to control of time slot contention for fairness in 10 a communication system.
Background In time-division multiple access (TDMA) communication 15 systems, each frame is an interval of T seconds, and each frame is divided into n discrete time slots. Thus, subscriber units can communicate with each other on a basis of non-overlapping transmission bursts. Since there is no overlap, a same carrier frequency may be assigned to all subscriber units 20 using a same base site.
The TDMA technique is characterized by duration of the time frame and the time slot within the frame. Each time slot typically consists of a guard time, a preamble, and the 25 information to be transmitted. Typically, the preamble contains system information such as synchronization, control and routing information. The guard time and the preamble generally depend on the organization of the system. The information to be transmitted occupies a predetermined 30 number of bits.
Generally, any time slot in a frame is available for any subscriber unit. In such a system there is a need for a fair method of simultaneously allocating time slots between users WO 96/04720 2 1 7 a ~ 9 ~ PCT/US95/0~9~19 ~
as the number and location of available time slots within a frame changes.
Brief Descriptions of the Drawings FIG. 1 is a flow chart of steps of a method for utilizing a novel scheme for controlling contention for a plurality of idle time slots to provide fairness between a plurality of types of 10 subscriber units in a time division multiplex communication system in accordance with the present invention.
FIG. 2 iS a flow chart of one embodiment of the steps of the novel scheme for controlling time slot contention in 15 accordance with the present invention.
FIG. 3 is a flow chart of one embodiment of the unique contention-based access scheme for determining the spreading factor N in accordance with the present invention.
FIG. 4 is a flow chart of another embodiment of steps of a method for controlling time slot contention to provide fairness between a plurality of types of subscriber units communicating with a base site in a time division multiplex 25 communication system in accordance with the present invention FIG. 5 is a block diagram of one embodiment of a time division multiplex communication system for controlling time 30 slot contention to provide fairness among transmissions for a plurality of traffic types of subscriber units in accordance with the present invention.
FIG. 6 is a block diagram of one embodiment of a process 35 organization for a radio porVbase site in accordance with the 1 WO 96/04720 2 ~ 7 Q ~ ~ ~ PCT/US95105949 present invention wherein the radio porVbase site and the radio port control unit/base site control unit are separate units.
FIG. 7 is a block diagram of a time division multiplex communication system for controlling time slot contention to provide fairness among transmissions for a plurality of traffic types of subscriber units in accordance with the present Invention.
FIG. 8 is a schematic drawing of an exemplary simplified TDMA frame structure that contains 8 time slots and shows a l:ypical uplink packet structure.
FIG. 9 is a flow chart showing steps for one embodiment of an uplink initial access control scheme executed by a subscriber unit in accordance with the method of the present invention.
FlGs. 10, 11, and 12 are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by a single-slot subscriber unit in accordance with the method of the present invention.
FlGs. 13, 14, 15 and 16 are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by the multi-slot subscriber unit in accordance with the method of the present invention.
FIG. 17 is a schematic showing, for a block diagram of elements operating in accordance with the present invention, one implementation of slow channel bits and variables exchanged between a subscriber unit, a radio port and a radio 0 port control unit in accordance with the present invention.
2~ 70~q~
WO 9610-1720 PCT/US9S/059~9~t FlGs. 18 and 19 show a flow chart for steps in one embodiment for an uplink processing scheme executed by a radio port in accordance with the method of the present I nventlon .
FIG. 20 is a flow chart for one embodiment of a downlink processing scheme executed by a radio port in accordance with the present invention.
FIG. 21 is a flow chart for one embodiment of an uplink processing scheme executed by a radio port control unit in accordance with the present invention.
Detailed Description of a Preferred Embodiment The present invention allows subscriber units using a single time slot in a frame and subscriber units using multiple time slots in the frame to share time slots of a physical 20 channel in a communication system. The invention provides an access protocol that uses a contention scheme for assigning a plurality of time slots in a fair manner among multiple traffic types.
A channel protocol is a procedure by which a group of nameless users contending for a channel all agree on a set of rules for acquiring a transmission channel. Where all users abide by the protocol, the channel is sequentially assigned to subsequent users in an orderly fashion. A collision channel is a channel in which there is a simple central relay or base site in which messages that do not satisfy a predetermined parity check condition are negatively acknowledged with a return to a transmitting user. If one user begins to transmit and another user transmits an overlapping message, a message collision r-35 occurs and neither message is intelligible. If both users ~ WO 96/04720 2 t 7 Q ~ 9 ~ PCT/US951059~9 retransmit, a collision occurs again. Typically, a conflict resolution algorithm is used to determine which user is allowed to transmit first, i.e., to determine a retransmission time for the users such that it is unlikely that their 5 retransmissions will collide. In a slotted channel time is dlivided into intervals called slots, where the duration of a slot is the same as the duration of a segment of a packet. A
segment contains information transmitted during one time slot. Thus, packets can only collide with other packets that 10 are transmitted in a same time slot.
FIG. 1, numeral 100, is a flow chart of steps of a method for utilizing a novel scheme for controlling, at a base site, contention for a plurality of idle time slots to provide fairness 15 between a plurality of types of subscriber units in a time division multiplex communication system in accordance with the present invention. The method includes the steps of: A) contending, by the plurality of types of subscriber units, on an idlle packet time slot by transmission of a predetermined 20 number of control bits (102); and B) utilizing, by the base site, the predetermined number of control bits from the subscriber UllitS and supervisory counter bits generated at the base site in accordance with a novel scheme to determine distribution of the time slot, and where the subscriber unit uses multiple 25 time slots, to determine distribution of remaining idle packet time slots (104).
FIG. 2, numeral 200, is a flow chart of one embodiment of the steps of the novel scheme for controlling time slot 30 contention in accordance with the present invention. The novel scheme utilizes parallel performance by the plurality of subscriber units and the base site. Each subscriber unit performs the steps of: A) utilizing a transmission retry counter for storing a transmission retry count and, prior to 35 transmission of a packet on an uplink, inserting the retry count -WO 96/04720 ~ 1 7 ~ 8 q 3 PCTIUS95/0~9~19~
into a packet header together with a traffic type identifier (202) and B) utilizing a random number selector for randomly selecting a value between one and a spreading factor N, where N, N an integer, and a channel bitmap are obtained from the 5 base site, for a wait time and converting the value of the wait time to a number of time slots for waiting until retransmission (204). Meanwhile, the base site control unit performs the steps of: C) utilizing an averaging unit for receiving the transmissions of the packets from the subscriber 10 units and for determining an average retry count for a predetermined time interval (206) and D) utilizing a comparing unit (208) for: D1) where a number of packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing 15 the average retry count with a predetermined threshold value, and utilizing a unique contention-based access scheme to determine the spreading factor N, and D2) where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, 20 automatically increasing N by a predetermined increment number and inserting N in a downlink packet together with the channel bitmap, setting a predetermined number of control bits based on the traffic type of the subscriber unit in a downlink for controlling contention for time slots among the plurality 25 of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
FIG. 3, numeral 300, is a flow chart of one embodiment of the unique contention-based access scheme for determining 30 the spreading factor N in accordance with the present invention. The scheme typically comprises the steps of: A) where the average retry count is greater than or equal to the predetermined threshold value plus a first predetermined constant, increasing N by a value equal to a predetermined 35 percentage of the average retry count (302); B) where the ~ W096/04720 2 t 7 0 8~ 3 PCTIUS9510594~
average retry count is less than or equal to the predetermined threshold value minus a second predetermined constant, decreasing N by a predetermined decrease amount (304~; and C) where the average retry count lies between the predetermined 5 threshold value minus the second predetermined constant and the predetermined threshold value plus the first predetermined constant, maintaining a current value of N (306).
In the preferred embodiment, the predetermined number 10 of control bits is two and the control bits indicate four types of ownership of a time slot. In one implementation, the four types of ownership are: A) a current time slot is available for contention requests; B) the current time slot is owned by a single time slot subscriber unit; C) the current time slot is 15 owned by the multiple time slot subscriber unit that sent a channel request in the time slot; and D) the current time slot is owned by a multiple time slot subscriber unit that sent a channel request in a different time slot.
A traffic type determines a maximum number of time slots per time division multiplex frame. For example, in the preferred embodiment of this invention, a first traffic type may use one time slot per frame, and a second traffic type may be selected to use up to 8 time slots per frame. A plurality of control bits, being two control bits in this example, are broadcast in a downlink for controlling contention among the first traffic type subscriber units, contention among the multiple time slot subscriber units, and between first and second traffic type subscriber units. The two control bits indicate ownership of the time slots. In this example, the control bits have the following four digital values:
00, indicating that a current time slot is available for contention requests;
01, indicating that the current time slot is owned by a single time slot subscriber unit;
WO 96/04720 2 ~ ~ 0 8 9 3 PCT/US9';/OS949 ~
10, indicating that the current time slot is owned by the multiple time slot subscriber unit that sent a channel request in this time slot; and 11, indicating that the current time slot is owned by a 5 multiple time slot subscriber unit that sent the channel requested in a different time slot.
Further, a set of controi bits is sent in an uplink indicating whether a packet segment transferred in each time 10 slot is part of a packet transmitted by a single slot user or a multiple slot user. A base site, typically a radio port, contention scheme differentiates between single time slot channel requests and multiple time slot channel requests and sets the two control bits in the downlink accordingly. A
15 subscriber unit contention scheme, executed by the subscriber unit, provides for contention for channel access, and upon the subscriber unit obtaining channel access, allows a single time slot user to transmit on a same time slot that was won in the contention and allows a multiple time slot user to transmit on 20 all time slots designated for the multiple time slot user, as is described in more detail below.
An uplink packet is typically divided into one-slot segments. For a multiple time slot packet, i.e., an N of N
25 packet, N a positive integer, a first byte of the non-header segment is used as the segment number. For a single time slot packet, i.e., a 1 of N packet, transmitted segments are acknowledged via a Stop-and-Wait automatic acknowledgment scheme. Thus, only one bit sequence number is needed. This 30 one bit sequence number is accomplished through the use of even and odd segments. The subscriber unit alternately marks a new segment as odd or even. In the event that an uplink segment is correctly received by the base site, but the downlink time slot in the following frame is not decoded by 35 the subscriber unit, the subscriber unit retransmits the old ~ WO 96/04720 2 t 7~ 9 3 PCT/US95105949 segment and the base site identifies and discards the duplicate segment by identifying that its segment type (odd or even) is identical to that of the original.
FIG. 4, numeral 400, is a flow chart of another embodiment of steps of a method for controiling time slot contention to provide fairness between a plurality of types of subscriber units communicating with a base site in a time division multiplex communication system in accordance with the present invention. The method includes the steps of: A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits (402); B) utilizing, by the base site, the predetermined number of control bits from the subscriber units to determine the type of subscriber unit to win contention and assigning the time slot to a winning subscriber unit in accordance with the determination wherein, where the winning subscriber unit is a multiple time slot user type, further assigning the remaining of the idle packet time slots ~404); C) starting, by a base site control unit, at least a first supervisory counter timer to control a length of time that the winning subscriber unit holds the time slot (406); D) utilizing, by the base site, a predetermined number of control bits to broadcast to the plurality of types of subscriber units the type of subscriber unit currently assigned to the time slot (408);
I_) utilizing, by the winning subscriber unit, a predetermined number of control bits to determine a current status of the next time slot prior to transmitting (412); and F) transmitting, by the winning subscriber unit, a packet segment and recyling to step D until one of: a packet is completed and one supervisory counter timer expires (410).
FIG. 5, numeral ~00, is a block diagram of one embodiment of a time division multiplex communication 35 system for controlling time slot contention to provide fairness WO 96/04720 2 ~ 7 0 8 9 ~ PCT/US95/ûS9~9 among transmissions for a plurality of traffic types of subscriber units in accordance with the present invention. The system includes a plurality of subscriber units (508, 514, ...) and a base site (502) that perform the steps set forth below in 5 parallel. Each subscriber unit (508, 514, ...) includes a transmission retry counter (510, 516, ...) and a random number selector (512, 518, ...). The transmission retry counter (510, 516) is utilized for storing a transmission retry count and, prior to transmission of a packet on an uplink, for inserting 10 the retry count into a packet header together with a traffic type identifier. The random number selector (512, 518, ...) is operably coupled to receive a spreading factor N and a channel bitmap from the base site and is utilized for randomly selecting a value between one and N for a wait time and 15 converting the value of the wait time to a number of time slots for waiting until retransmission. A unique contention-based access control scheme is used by the base site for prioritizing transmissions of the subscriber units The base site (502) includes an averaging unit (504) and a comparing 20 unit (506). The averaging unit (504) is used for receiving the transmissions of the packets from the subscriber units and for determining an average retry count for a predetermined time interval. The comparing unit (506) is operably coupled to the averaging unit (504) and is used for: A) where a number of 25 packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing the average retry count with a predetermined threshold value, and utilizing the unique contention-based access scheme to determine the spreading factor N, N an 30 integer, and B)where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, automatically increasing N
by a predetermined increment number and inserting N in a downlink packet together with the channel bitmap, setting a 35 predetermined number of control bits based on the traffic type WO 96/04720 ~ 1 7 0 8 9 3 PCTIUS9~/05949 of the subscriber unit in a downlink for controlling contention for time slots among the plurality of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
The unique contention-based access scheme for dletermining the spreading factor N utilized by the time division multiplex communication system is typically the scheme described above and shown in FIG. 3. In a preferred 10 embodimenl: the predetermined number of control bits is two and the control bits indicate four types of time slot ownership, where the four types are as described above.
Thus, a subscriber unit that desires to transmit a packet 15 on the uplink first contends for the channel access. The subscriber unit monitors the downlink and locates a packet time slot by reading the slow channel. When a packet time slot is found, the subscriber unit determines a status of the channel before transmitting an initial access request, i.e., an 20 uplink packet header, on that channel. A status of busy or idle bits is described herein as B/l. The subscriber unit transmits on the uplink time slot when the status is Idle (B/l=00) and, where the status is Busy (B/l ~ 00), waits a random period of time, R1. After transmitting the packet header, the subscriber 25 Ul1it reads the slow channel bits in the corresponding time slot of the next downlink packet frame to determine whether it has successfully gained access to the channel. Where the downlink WEI, i.e., word error indication, is zero and the ECHO bits are matched, the subscriber unit is allowed to continue where the 30 B/l bits are 01 for a 1-of-N user, or 10 for a N-of-N user.
Where the B/l bits are 11, the current time slot is assigned to another N-of-N user who has just won contention on another time slot prior to this time slot. Where a subscriber unit fails to seize the channel, i.e., one of: WEI ~ 0; uplink ECHO
35 downlink ECHO; B/l = 00; and B/l = 11, the transmission WO96/04720 2 t 7 Q ~ ~ 3: PCT/US9510S9i91~
process is halted and remains idle for a random amount of time, R2, before attempting to access the uplink again.
Where a subscriber unit successfully gains access to the 5 channel, it continues to send packet segments and reads the WEI and ECHO bits on the downlink slow channel of the following frame to determine the result of each transmission.
Where the WEI is set, the subscriber unit retransmits the segment. The base site, typically a radio port, maintains a 10 busy status on the time slot until it receives the DISCONNECT
segment, i.e., a last segment, of the packet. An 1-of-N user stays on the same time slot where it won the contention, while an N-of-N user may transmit on all the time slots designated for it, i.e., the time slots designated B/l = 10 or 11.
15 The transmission process is aborted where the ECHO bits do not match or the time slot is no longer a packet slot, or the time slot becomes idle.
FIG. 6, numeral 600, is a block diagram of one 20 embodiment of a process organization for a radio port/base site in accordance with the present invention wherein the radio porVbase site and the radio port control unit/base site control unit are separate units. This organizational diagram is provided for an overview of the implementation of the present 25 invention. Thus, a more detailed description of the invention follows this paragraph. Upon subscriber unit initial uplink access (604) at a first subscriber unit SU (602) of a plurality of subscriber units, uplink packet transmission is executed by the subscriber unit (602). Where the subscriber unit is a 30 single time slot user (606), the subscriber unit transmits a packet on 1 of N time slots per frame to the radio porVbase site RP (610) as set forth with more particularity in FlGs 10, 11, and 12. Where the subscriber unit is a multiple time slot user (608), the subscriber unit transmits a packet in a 35 multiplicity of time slots per frame specified by the radio ~ WO 96/04720 2 1 7 0 8 9 3 PCT/US95/059~9 porVbase site (610), as set forth with more particularity in FlGs 13, 14, 15, and 16. The radio port/base site (610) is responsive to the subscriber units and to the radio port control unit/base site control unit, performs uplink processing (612) 5 as described with more particularity with respect to FlGs. 18 and 19, and performs downlink processing (614) as described with more particularity with respect to FIG. 20. The radio port control uniVbase site control unit RPCU (616) provides processing (618) of each time slot of data or segment of data 10 received from the radio porVbase site, as described with more particularity with respect to FIG. 21.
FIG. 7, numeral 700, is a block diagram of a time division multiplex communication system for controlling time slot 15 contention to provide fairness among transmissions for a plurality of traffic types of subscriber units in accordance with the present invention. The radio port control uniVbase site control unit (702) includes supervisory counters (704), i.e., timers, that control a length of time that a subscriber unit 20 holds a time slot. The radio port/base site (706) includes a time slot manager (708) for assigning the time slots to subscriber units in accordance with a time slot winner determination by a contention controller and the contention controller (710) for determination of the time slot winner in 25 accordance with the unique contention-based access scheme of the present invention as described more fully above. The plurality of subscriber units includes at least one single time slot user, i.e., 1-of-N, having an ECHO bits generator (714) for generating ECHO bits that are used as described above and a 30 transmission queue (716), Tx queue and a retransmission queue, ReTx queue (730), and at least one multiple time slot user, i.e., N-of-N, that includes a counter (720), and ECHO bits generator, a Tx queue (724), a retransmission queue/ReTx queue (726), and an acknowledgment queue/Ack queue (728).
35 These elements are utilized as described herein.
WO ~610J720 2 ~ 7 0 8 ~?3 PCT/1159~/059~9~
FIG. 8, numeral 800, is a schematic drawing of an exemplary simplified TDMA frame structure that contains 8 time slots and shows a typical uplink packet structure. The 5 subscriber unit transmits on the uplink and receives on the downlink. A devoted out-of-band signaling channel such as the slow channel in a Personal Communications System, for example the PACS, (Personal Access Communication System), may be selected to be used for transmission of contention 10 control bits, thereby maximizing bandwidth utilization on the fast channel. An in-band signaling channel such as the fast channel in PACS may be used for transmission of user information. The slow channel is a small portion of the PACS
time slot which typically carries the control information such 15 as time slot usage (802), echo bits (804), and access type (806). The fast channel is a larger portion of the PACS time slot which generally carries user information and a minimum amount of control information such as user identification, packet length and packet number. In the preferred embodiment 20 the downlink slow channel bits typically include: PCI, packet - channel bits that are used by the base site to indicate that a current time slot is a packet time slot; B/l, busy/ldle bits that are used by the base site to indicate the status of the uplink to all subscriber units; ECHO bits, bits that are generated 25 randomly and transmitted by a subscriber unit on an uplink and echoed back by a base site to the subscriber unit on the downlink in the next frame; and WEI, a word error indication bit that is used by the base site to indicate to the subscriber units whether the uplink time slot in the previous frame was 30 received correctly. The uplink slow channel bits typically include: PCI (802), packet channel indication bits that are used by the subscriber unit to indicate that the current time slot is a packet time slot; ECHO bits (804), bits that are generated randomly and transmitted by the subscriber unit on an uplink 35 and echoed back by a base site to the subscriber unit on the ~ WO 96/04720 2 1 7 0 ~ 9 3 PCT/US95/05949 downlink in the next frame; and SEG bits (806), bits that are used by the subscriber unit to indicate which type of segment is being transmitted. For example, SEG bits may be encoded as follows: 000 for a first segment of a 1-of-N packet; 001 for a 5 rrliddle odd segment of a 1-of-N packet; 010 for a middle even segment of a 1-of-N packet; 011 for a disconnect segment of a 1-of-N packet; 100 for a first segment of an N-of-N packet;
101 for a middle of an N-of-N packet; 110 for reserved; and 111 for a disconnect segment of an N-of-N packet.
FIG. 9, numeral 900, is a flow chart showing steps for one embodiment of an uplink initial access control scheme executed by a subscriber unit in accordance with the method of the present invention. In the idle mode a subscriber unit 15 monitors the downlink for possible messages addressed to itself. Upon desiring to transmit a new packet (902), the subscriber unit reads the slow channel on a downlink time slot to determine whether it contains vaild information, i.e., cyclical redundancy code (CRC) is successful (904) and the 20 time slot is a packet time slot (906). Where the CRC is invalid or time slot is a non-packet time slot, the subscriber unit waits a first random period of time (908), R1 slots, before contending again on the same or a different time slot (904).
Where the CRC is valid and the time slot is a packet time slot, the subscriber unit reads the slow channel bits to d~termine the busy/idle, B/l, status of the time slot (910).
Where the time slot is busy, the subscriber unit delays R1 time slots (908) and recycles to contending again on a new time slot (904). Where the channel is idle, the subscriber unit may transmit a request segment. The subscriber unit transmits a packet header (912) to request channel access, waits exactly one frame (914), and checks the downlink CRC again (916). The packet header contains information about the packet and the subscriber unit that sent it.
WO 96/04720 2 ~ 7 0 8 ~ 3 PCT/US95/059~9~
Where a cyclical redundancy code (CRC) fails, the subscriber unit delays R2 time slots (932), R2 a second random period of time, and recycles to determining the status on a 5 new time slot (904). Where the cyclical redundancy code is successful, the subscriber unit determines whether the same time slot,i.e., the time slot where the subscriber unit transmitted on in the previous frame, is still a packet time slot (918). Where the time slot is no longer a packet time 10 slot, the subscriber unit delays R1 time slots (908), and recycles to contending again on another time slot (904). Where the time slot is still a packet time slot, the WEI bit is checked (916) for collision detection. That is, where two or more subscriber units try to transmit request segments at the same 15 time and the radio porVbase site does not capture anyone's segment,i.e., there is a collision, the WEI bit in the next downlink is set equal to one; and where only one subscriber unit's segment is received correctly by the radio porVbase site, the WEI bit in the next downlink is set equal to zero.
20 Where the WEI bit is 1, the subscriber unit recycles to delaying R1 time slots (908). Where the WEI bit is 0, the subscriber unit checks for matching ECHO bits (922). That is, where a subscriber unit is sending an access request segment or a normal segment, it randomly assigns a number to an ECHO
~5 field. A typical ECHO field is three bits. The radio port/base site is required to echo this field back on the same time slot in the following frame of the downlink. All subscriber units that are transmitting check this field against their random number that was previously transmitted. Where the subscriber 30 unit's random number is a non-match for the value in the ECHO
field, then the subscriber unit waits another random period of time R2 time slots and contents again on a new time slot. The value of R1 and R2 are determined using a novel scheme described in FlGs 1, 2, and 3.
~ WO 9~/047~ PCT/~J59S/05949 Where ~he ECHO bits are matching and the subscriber unit is a single time slot user, the subscriber unit checks the B/l bits (928). Checking B/l bits confirms the result of the subscriber unit's channel access request. A value of 00 5 indicates that there was a collision and no subscriber unit is assigned to the time slot. A value of 11 indicates that the time slot will be reserved for another user as soon as next frame. Where the B/l bits ~ 01, the subscriber unit recycles to delaying R1 time slots and contends for channel access again 10 as previously described. Where the B/l bits =01, the subscriber unit begins to transmit segments on the time slot.
Where the ECHO bits are matching and the subscriber unit is a multiple time slot user, the subscriber unit checks the B/l bits (926). Where the B/l bits ~ 10, the subscriber unit recycles to 15 delaying R1 time slots (908). Where the B/l bits = 10, the subscriber unit begins to transmit segments using one or more idle time slot per frame.
FlGs 10, 11, and 12 numeral 1000, 1100, and 1200 20 respectively, are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by a single-slot subscriber unit in accordance with the method of the present invention. The subscriber unit, after gaining access to the channel (930), checks the B/l bits (1002) to 25 determine whether the time slot is busy, i.e., B/l = 01. Where the B/l bits ~ 01, the subscriber unit aborts the transmission and recycles to delaying R1 time slots (908) as previously described. Where the B/l bits =01, the subscriber unit begins to transmit segment on the time slot (1004). The segment 30 transmission process, shown in FIG. 12, works as follows: all segments of a packet are intially stored in the transmission queue, Tx queue. The ReTx is initially empty. The subscriber unit first checks the retransmission queue, ReTx queue (1202).
Where the ReTx queue is non-empty, segment at the front of 35 the ReTx queue is transmitted (1206). Where the ReTx is empty WO 96/04720 ~ 7 Ci 8 9 3 PCT/US95/OS9 19 and the Tx queue is non-empty, the subscriber unit transmits a segment at the front of the Tx queue (1208) and moves the segment to the ReTx queue (1210). Where both the ReTx and Tx queues are empty, the subscriber unit begins to transmit DISCONNECT segment (1024) as shown in FIG. 11 and described I ater.
After transmitting a segment, the subscriber unit waits exactly one frame (1006) and reads/decodes the downlink time slot (101 0),i.e., the time slot where the subscriber unit transmitted on in the previuos frame. Where the downlink time slot is error-free, i.e., a CRC is successful, ~he subscriber unit checks the PCI bits to determine whether l:he time slot is still a packet slot (1014). Where the time slot is a non packet time slot, the subscriber unit delays R1 time slots (908) and proceeds as previously described. Where the time slot is still a packet time slot, the subscriber unit checks the WEI bit (1008). The WEI bit on the downlink indicates whether the uplink segment on the previous frame was received correctly by the base site/radio port. Where the WEI bit is 1, i.e., the previous transmitted segment was received incorrectly, the subscriber unit proceeds to check the B/l status (1002) in order to transmit the same segment again,i.e., the segment which is still in the ReTx queue. Where the WEI is 0, the subscriber unit compares ECHO bits (1020), and where there is a non-match, delays R1 time slots (908) and proceeds as previously described. Where the ECHO bits match, the subscriber unit discards the segment in the ReTx queue and proceeds to check the B/l status in order to transmit a new segment.
Where the downlink time slot contains error, i.e., the CRC
is invalid, the subscriber unit skips transmitting on the current time slot and waits exactly one frame (1008) and recycles to decode the downlink again until the CRC is valid.
~ WO 96/04720 2 1 7 0 8 9 3 PCT/US95/05949 The subscriber unit then determines whether the time slot is still a packet slot (1016). Where the time slot is a non packet time slot, the subscriber unit delays R1 time slots (908) and proceeds as previously described. Where the time slot is still 5 a packet time slot, the subscriber unit proceeds immidiately to checking the B/l status in order to transmit a segment without checking the previous transmission result since the subscriber unit did not transmit in the last frame.
Where both the ReTx queue and the Tx queue are empty when it is time to transmit a segment (1024), the subscriber unit transmits a DISCONNECT segment (1102), waits exactly one frame (1104), and decodes the downlink. The subscriber unit recycles to transmitting DISCONNECT segment until one 15 of: a CRC is invalid (1106), the CRC is valid but the time slot becomes non packet slot (1108), and the CRC is valid and the time slot is still a packet and the WEI is 0 (1110). Where one of these events occurs the subscriber unit completes the uplink packet transmission (11 12) FlGs 13, 14, 15, and 16, numeral 1300, 1400, 1500, and 1600 respectively, are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by the multi-slot subscriber unit in accordance with the 25 method of the present invention. Each subscriber unit of the multi-slot subscriber unit traffic type has three data queues: a transmission, Tx, queue, for storing segments of the original packet without the DISCONNECT segments; a retransmission, ~eTx, queue for storing segments awaiting to be 30 retransmitted; and an acknowledgment queue, Ack queue, for storing the segments which are waiting for acknowledgment.
Where the subscriber unit successfully gains access to - the uplink (932), the subscriber unit begins to transmit packet 35 segment (1302). The ReTx queue is checked (1502) to -WO 96/04720 2 1 ~ Q ~ ~ 3 PCTIUS95/OS9~9~
determine whether there is a segment to retransmit. Where there is a segment to retransmit, i.e., the ReTx queue is non empty, the subscriber unit transmits a segment at the front of the ReTx queue (1~06) and moves the segment from the ReTx 5 queue to the Ack queue (1514). Where the ReTx queue is empty, the subscriber unit checks the Tx queue to determine whether there is a new segment to transmit (1504). Where there is a new segment to transmit, i.e., the Tx queue is non empty, the subscriber unit transmits a segment at the front of the Tx 10 queue (1508) and moves the segment from the Tx queue to the Ack queue (1512). Where both the ReTx queue and the ReTx queue are empty and the Ack queue is non empty, the subscriber unit transmits a "fill" segment. Where both the ReTx queue and the ReTx queue are empty and the Ack queue is 15 also empty, the subscriber unit proceeds to transmitting DISCONNECT segment (1304) as shown in FIG.14 and is described below.
Where a normal segment (1302) or a "fill" segment 20 (1306) was transmitted, the subscriber unit waits until the beginning of the next time slot (1308) to determine the transmission result of the segment sent in the current time slot of the previous frame. Where no segment was transmitted in the current time slot of the previous frame, the subscriber 25 unit performs CRC check (1312) and determines whether the current time slot is still a packet time slot (1316). Where the CRC is invalid, or the CRC is valid but the time slot becomes a non packet time slot, the subscriber unit recycles to waiting until the next time slot. Where the CRC is valid and the time 30 slot is still a packet time slot, the subscriber unit proceeds to checking the B/l bits (1320). Where there was a segment, k, transmitted in the current time slot of the previous frame, the subscriber unit determines transmission result and performs neccessary actions (1314) as shown in FIG. 16 and is described 35 in the following: The subscriber unit performs CRC check ~ WO 96/04720 2 1 7 0 8 ~ 3 PC'T/U595/059J9 (1602) and determines whether the current time slot is still a packet time s.ot (1606). Where the CRC is invalid, or the CRC
is valid but the time slot becomes a non packet time slot, the subscriber unit moves segment k from the Ack queue to the 5 ReTx queue (1604) and recycles to waiting until the next time slot (1308). Where the CF2C is valid and the time slot is still a packet time slot, the subscriber unit checks the WEI bit (1608) and the ECHOs (1610). Where the WEI is 1, the subscriber unit moves segment k from the Ack queue to the ReTx queue (1614) 10 and proceeds to checking the B/l bits (1320). Where the WEI is 0 and the ECHOs match, the subscriber unit deletes segment k from the Ack queue (1612) and proceeds to checking the B/l bits (1320). Where the WEI is 0 and the ECHOs are non match, the subscriber unit aborts the transmission (1318) and 15 proceeds to delay R1 slot (908) as previously described.
The subscriber unit checks the B/l (1320) to determine whether the time slot is reserved for it. Where the B/ls are either '00' or '01', i.e., B/l ~ 1x, the subscriber unit proceeds to 20 the next time slot (1308). Where the B/l bits are either '10' or '11 ', i.e., B/l = 1 x, the subscriber unit recycles to transmitting a new segment (1302).
Where the ReTx queue, the ReTx queue, and the Ack queue 25 are all empty, the subscriber unit transmits a DISCONNECT
segment (1402), waits until next time slot (1406), and decodes the downlink (1408). Where the CRC is invalid, or the CF2C is valid but the time slot becomes a non packet time slot, the subscriber unit finishes the packet transmission. Where 30 the CRC is valid and the time slot is still a packet time slot, the subscriber unit determines whether there was a - DISCONNECT segment transmitted in the same time slot of the previous frame (1412). Where there was no DISCONNECT
segment transmitted in the same time slot of the previous 35 frame, the subscriber unit proceeds to check the B/l bits WO 96/04720 ~ 1~ 7 Q 8: 9 ~ PCT/US9S/059 (1404). Where there was a DISCONNECT segment transmitted in the same time slot of the previous frame. the subscriber unit checks the WEI bit (1414) to determine the transmission result. Where the WEI is 0, the subscriber unit finishes the 5 packet transmission (1416). Where the WEI is 1, the subscriber unit proceeds to check the B/l bits (1404). Where the B/ls are either '00' or '01', the subscriber unit proceeds to the next time slot (1406). Where the B/l bits are either ~10~ or '1 1', i.e., B/l = 1 x, the subscriber unit recycles to transmitting 10 another DISCONNECT segment (1402).
FIG. 17, numeral 1700, iS a schematic showing, for a block diagram of elements operating in accordance with the present invention, one implementation of slow channel bits and 15 variables exchanged between a subscriber unit, a radio port and a radio port control unit in accordance with the present invention. The subscriber unit (1702) sends a segment (1708) that includes PCI bits, ECHO bits, and SEG bits and receives from the radio port a segment (1710) that includes PCI bits, 20 ECHO bits, B/l bits, and a WEI bit. The radio port (1704) sends a segment (1712) that includes PCI bits, B/l bits, SEG bits, and a WEI bit and receives from the radio port control unit (1706) a segment (1714) that includes PCI bits and a plurality of control bits, C0, C1, C2.
FIG. 18, numeral 1800, and FIG. 19, numeral 1900, show a flow chart for steps in one embodiment for an uplink processing scheme in accordance with the method of the present invention. The uplink processing scheme is executed 30 by the radio porVbase site, hereafter referenced as the radio port, RP, upon receiving an uplink segment from a subscriber unit (1802). The RP checks a variable M_PCI to determine whether a current time slot is a packet slot (1804). This packet channel indication variable was set when the RP
35 executed a downlink algorithm. Where the slot is not a packet ~ WO 96/04720 2 1 7 0 8 ~ 3 PCT/US95105919 slot, no action is taken by the RP. Where the slot is a packet slot, the RP checks whether there is an N-of-N user currently using the channel (1806). This event is indicated by the variable N-of-N flag. Typically, where there is an N-of-N user 5 using the channel, the N-of-N flag is one, and the RP
determines whether the time slot is idle (1808), i.e., typically whether M_B/I = 00. Where the N-of-N flag is one and the time slot is idle, i.e., M_B/I = 00, the RP sets the M_B/I variable to 11 (1810). Where the N-of-N flag is zero and the B/l status is 10 either 10 or 11 (1820), the B/l status is set to idle (1818), iae., M_B/I = 00. The RP then determines whether a CRC check is valid (1812). Where the CRC check is invalid (1812), the RP
the variable M_WEI is set to one (1814) and a one is stored in a WElpfield (1816). Then the RP delivers an uplink segment to 15 the radio port control uniVbase site control unit (1916).
Where the CRC check is valid (1812), the RP determines a whether the variable M_B/l is equal to 00 (1902), proceeding as set forth in FIG. 19. Where the channe! is idle, i.e., M_B/l 20 =00, the RP checks the SEGA to determine whether the uplink slot is a first time slot of a packet (1904), i.e., a first two bits of SEGA equal 00. This is the only type of uplink time slot that is permitted when the channel is idle. The RP determines whether an N-of-N flag is set (1906). Where the N-of-N flag is 25 clear, the RP determines whether the time slot is from a 1-of-N user or a from a N-of-N user (1908), i.e., whether SEGA = Oxx.
Where the time slot is from a 1-of-N user, the B/l bits are set to 01 (1910), i.e., are set to indicate busy. Where the time slot is from a N-of-N user, the B/l bits are set to 10 to indicate 30 the time slot is busy and the N-of-N flag is set to one (1912).
Where, after the RP determines that the CRC check is valid, the variable M_B/l is unequal to 00 (1902), the RP determines whether SEGA equals xO0 (1918). Where SEGA equals xO0, a one is stored in a WElpfield (1816), and the RP delivers an 35 uplink time slot to the radio port control uniVbase site WO 96/04720 2 t 7 ~ ~ 9 ~. PCT/US95/059~1~
control unit (1916). Where the SEGA is unequal to xO0, the RP
determines whether SEGAequals x11 (1920). Where SEGAis~
equal to x11 (1920), the RP determines whether SEGA equals OXX (1922). Where SEGA equals Oxx, the variable M_B/I is set 5 to 00 (1926), i.e., is set to idle.. Where SEGA is unequal to Oxx, the N-of-N flag is set to zero (1924) and the variable M_B/I is set to 00 (1926), i.e., is set to idle. The ECHO bits received in the variable M_ ECHO are saved, the M_ WEI variable is set to zero, the segment packet SEGP jS updated to SEGA, the B/lp 10 packet bits are updated to B/IA and N/Np is set to the N-of-N
flag (1914). The states of the B/l bits and the N-of-N flag are sent to the radio port control uniVbase/site when the RP
delivers the uplink time slot to the radio port control uniVbase site (1916). Where the RP is expecting a first time 15 slot of a packet and the uplink time slot is not a first time slot of a packet, or the RP is expecting a non-first time slot of a packet and the time slot is a first time slot of a packet (1904 and 1918), the RP processes the packet as if the packet had failed the CRC check.
The RP maintains a busy status of the uplink channel as long as the disconnect segment, SEGA = X11, jS not received.
Where the disconnect segment is received, the RP sets the M_B/I variables equal to 00 to indicate that the uplink is idle.
FIG. 20, numeral 2000, is a flow chart for one embodiment of a downlink processing scheme executed by a radio port in accordance with the present invention. The steps of this method are executed whenever a downlink time slot is 30 received from the radio port control uniVbase site to be transmitted (2002). The RP determines whether PClp bits indicate that the slot is a packet slot (2004). Where the PClp bits indicate that the slot is not a packet slot, the RP inserts the WEI bit computed from the same time slot in the previous 35 uplink frame (2018) and broadcasts the downlink time slot to WO 96/047~0 PCT/US95/OS9~9 the subscriber units (2020). Where the PClp bits indicate that the slot is a packet slot, the RP determines whether a control bit C2 iS set to one (2006). Where C2 is set to 1, an N-of-N
flag is set to zero (2008). The RP then determines whether 5 other control bits, C0 and C1, are set to 01 (2010). Where C0 and C1, are set to 01, the RP stores 00 in the M_B/I (2014).
Where C0 and C1, are set to 10, the RP stores 01 in the M_B/I
(2012). The B/IA bits from the previous uplink transmission are saved in the M_B/I variable and the M_PCI bits are updated 10 to packet bits (2016). Then the RP inserts the WEI bit computed from the same time slot in the previous uplink frame (2018) and broadcasts the downlink time slot to the subscriber units (2020) .
FIG. 21, numeral 2100, is a flow chart for one embodiment of an uplink processing scheme executed by a radio port control unit in accordance with the present invention. Typically, the radio port control uniVbase site c~ntrol unit, i.e., RPCU, maintains an individual counter, 20 Dcounter, for each TDMA time slot, and a common counter, Ncounter, for all the TDMA time slots. Upon receiving a slot or segment from the radio porVbase site (2102), the RPCU first determines whether the current time slot is a packet slot (2106), i.e., whether PClp equals packet. Where the current 25 time slot is not a packet slot, Ncounter is incremented by one (2108). Where the current time slot is a packet slot, the RPCU
determines whether a CRC check is valid (2110), i.e., whether WElp is equal to zero. Where WElp is equal to zero ~2110), the Rl'CU copies the values of the B/l bits and the N/N bits into the 30 variables Busy and Nflag, respectively (2104), the Dcounter is set equal to 0 (2118), and the RPCU determines whether a most significant bit of the segment SEGp is zero (2120), i.e., whether SEGp=Oxx. Where the most significant bit of the segment SEGp is zero, Ncounter is incremented by one (2122).
35 Where WElp is unequal to zero (2110), the RPCU determines W096/W720 2 ~ 7 0 8 9 ~ PCT~S95/059~
whether the channel is busy (2112). Where the channel is busy (2112), Dcounter is incremented by one (2114) and Ncounter is set to zero (2124). Where the channel is idle (2112), Dcounter is set to zero (2121) and Ncounter is set to zero (2124). Upon 5 the Ncounter being set to zero or one (2124, 2108, 2122), the RPCU determines whether the Dcounter has exceeded its maximum value (2126). Where the Dcounter has exceeded its maximum value, control bits C0 and C1 are set to 01 (2130).
Where the Dcounter has not exceeded its maximum value, 10 control bits C0 and C1 are set to 00 (2128). After setting the control bits C0 and C1 (2128, 2130), the RPCU determines whether the Ncounter is greater than or equal to the number of time slots per TDMA frame (2132), i.e., 8. Where the N counter is greater than or equal to the number of time slots per TDMA
15 frame, the RPCU determines whether an Nflag is set to one (2134). Where the Nflag is set to one, the control variable C2 is set to one (2136) and the RPCU inserts C2, C1, and C0 into the slow channel of the corresponding time slot on the next downlink frame (2140). Where the Nflag is not set to one, the 20 control variable C2 iS set to zero (2138) and the RPCU inserts C2, C1, and C0 into the slow channel of the corresponding time slot on the next downlink frame (2140).
Although exemplary embodiments are described above, it 25 will be obvious to those skilled in the art that many alterations and modifications may be made without departing from the invention. Accordingly, it is intended that all such alterations and modifications be included within the spirit and scope of the invention as defined in the appended claims.
Throughout the description above "x" means a "don't care"
bit, as is commonly understood.
FAIRNESS BETWEEN SUBSCRIBER UNITS
Field of the Invention The present invention relates generally to control of time slot usage in a communication system and, more particularly, to control of time slot contention for fairness in 10 a communication system.
Background In time-division multiple access (TDMA) communication 15 systems, each frame is an interval of T seconds, and each frame is divided into n discrete time slots. Thus, subscriber units can communicate with each other on a basis of non-overlapping transmission bursts. Since there is no overlap, a same carrier frequency may be assigned to all subscriber units 20 using a same base site.
The TDMA technique is characterized by duration of the time frame and the time slot within the frame. Each time slot typically consists of a guard time, a preamble, and the 25 information to be transmitted. Typically, the preamble contains system information such as synchronization, control and routing information. The guard time and the preamble generally depend on the organization of the system. The information to be transmitted occupies a predetermined 30 number of bits.
Generally, any time slot in a frame is available for any subscriber unit. In such a system there is a need for a fair method of simultaneously allocating time slots between users WO 96/04720 2 1 7 a ~ 9 ~ PCT/US95/0~9~19 ~
as the number and location of available time slots within a frame changes.
Brief Descriptions of the Drawings FIG. 1 is a flow chart of steps of a method for utilizing a novel scheme for controlling contention for a plurality of idle time slots to provide fairness between a plurality of types of 10 subscriber units in a time division multiplex communication system in accordance with the present invention.
FIG. 2 iS a flow chart of one embodiment of the steps of the novel scheme for controlling time slot contention in 15 accordance with the present invention.
FIG. 3 is a flow chart of one embodiment of the unique contention-based access scheme for determining the spreading factor N in accordance with the present invention.
FIG. 4 is a flow chart of another embodiment of steps of a method for controlling time slot contention to provide fairness between a plurality of types of subscriber units communicating with a base site in a time division multiplex 25 communication system in accordance with the present invention FIG. 5 is a block diagram of one embodiment of a time division multiplex communication system for controlling time 30 slot contention to provide fairness among transmissions for a plurality of traffic types of subscriber units in accordance with the present invention.
FIG. 6 is a block diagram of one embodiment of a process 35 organization for a radio porVbase site in accordance with the 1 WO 96/04720 2 ~ 7 Q ~ ~ ~ PCT/US95105949 present invention wherein the radio porVbase site and the radio port control unit/base site control unit are separate units.
FIG. 7 is a block diagram of a time division multiplex communication system for controlling time slot contention to provide fairness among transmissions for a plurality of traffic types of subscriber units in accordance with the present Invention.
FIG. 8 is a schematic drawing of an exemplary simplified TDMA frame structure that contains 8 time slots and shows a l:ypical uplink packet structure.
FIG. 9 is a flow chart showing steps for one embodiment of an uplink initial access control scheme executed by a subscriber unit in accordance with the method of the present invention.
FlGs. 10, 11, and 12 are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by a single-slot subscriber unit in accordance with the method of the present invention.
FlGs. 13, 14, 15 and 16 are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by the multi-slot subscriber unit in accordance with the method of the present invention.
FIG. 17 is a schematic showing, for a block diagram of elements operating in accordance with the present invention, one implementation of slow channel bits and variables exchanged between a subscriber unit, a radio port and a radio 0 port control unit in accordance with the present invention.
2~ 70~q~
WO 9610-1720 PCT/US9S/059~9~t FlGs. 18 and 19 show a flow chart for steps in one embodiment for an uplink processing scheme executed by a radio port in accordance with the method of the present I nventlon .
FIG. 20 is a flow chart for one embodiment of a downlink processing scheme executed by a radio port in accordance with the present invention.
FIG. 21 is a flow chart for one embodiment of an uplink processing scheme executed by a radio port control unit in accordance with the present invention.
Detailed Description of a Preferred Embodiment The present invention allows subscriber units using a single time slot in a frame and subscriber units using multiple time slots in the frame to share time slots of a physical 20 channel in a communication system. The invention provides an access protocol that uses a contention scheme for assigning a plurality of time slots in a fair manner among multiple traffic types.
A channel protocol is a procedure by which a group of nameless users contending for a channel all agree on a set of rules for acquiring a transmission channel. Where all users abide by the protocol, the channel is sequentially assigned to subsequent users in an orderly fashion. A collision channel is a channel in which there is a simple central relay or base site in which messages that do not satisfy a predetermined parity check condition are negatively acknowledged with a return to a transmitting user. If one user begins to transmit and another user transmits an overlapping message, a message collision r-35 occurs and neither message is intelligible. If both users ~ WO 96/04720 2 t 7 Q ~ 9 ~ PCT/US951059~9 retransmit, a collision occurs again. Typically, a conflict resolution algorithm is used to determine which user is allowed to transmit first, i.e., to determine a retransmission time for the users such that it is unlikely that their 5 retransmissions will collide. In a slotted channel time is dlivided into intervals called slots, where the duration of a slot is the same as the duration of a segment of a packet. A
segment contains information transmitted during one time slot. Thus, packets can only collide with other packets that 10 are transmitted in a same time slot.
FIG. 1, numeral 100, is a flow chart of steps of a method for utilizing a novel scheme for controlling, at a base site, contention for a plurality of idle time slots to provide fairness 15 between a plurality of types of subscriber units in a time division multiplex communication system in accordance with the present invention. The method includes the steps of: A) contending, by the plurality of types of subscriber units, on an idlle packet time slot by transmission of a predetermined 20 number of control bits (102); and B) utilizing, by the base site, the predetermined number of control bits from the subscriber UllitS and supervisory counter bits generated at the base site in accordance with a novel scheme to determine distribution of the time slot, and where the subscriber unit uses multiple 25 time slots, to determine distribution of remaining idle packet time slots (104).
FIG. 2, numeral 200, is a flow chart of one embodiment of the steps of the novel scheme for controlling time slot 30 contention in accordance with the present invention. The novel scheme utilizes parallel performance by the plurality of subscriber units and the base site. Each subscriber unit performs the steps of: A) utilizing a transmission retry counter for storing a transmission retry count and, prior to 35 transmission of a packet on an uplink, inserting the retry count -WO 96/04720 ~ 1 7 ~ 8 q 3 PCTIUS95/0~9~19~
into a packet header together with a traffic type identifier (202) and B) utilizing a random number selector for randomly selecting a value between one and a spreading factor N, where N, N an integer, and a channel bitmap are obtained from the 5 base site, for a wait time and converting the value of the wait time to a number of time slots for waiting until retransmission (204). Meanwhile, the base site control unit performs the steps of: C) utilizing an averaging unit for receiving the transmissions of the packets from the subscriber 10 units and for determining an average retry count for a predetermined time interval (206) and D) utilizing a comparing unit (208) for: D1) where a number of packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing 15 the average retry count with a predetermined threshold value, and utilizing a unique contention-based access scheme to determine the spreading factor N, and D2) where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, 20 automatically increasing N by a predetermined increment number and inserting N in a downlink packet together with the channel bitmap, setting a predetermined number of control bits based on the traffic type of the subscriber unit in a downlink for controlling contention for time slots among the plurality 25 of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
FIG. 3, numeral 300, is a flow chart of one embodiment of the unique contention-based access scheme for determining 30 the spreading factor N in accordance with the present invention. The scheme typically comprises the steps of: A) where the average retry count is greater than or equal to the predetermined threshold value plus a first predetermined constant, increasing N by a value equal to a predetermined 35 percentage of the average retry count (302); B) where the ~ W096/04720 2 t 7 0 8~ 3 PCTIUS9510594~
average retry count is less than or equal to the predetermined threshold value minus a second predetermined constant, decreasing N by a predetermined decrease amount (304~; and C) where the average retry count lies between the predetermined 5 threshold value minus the second predetermined constant and the predetermined threshold value plus the first predetermined constant, maintaining a current value of N (306).
In the preferred embodiment, the predetermined number 10 of control bits is two and the control bits indicate four types of ownership of a time slot. In one implementation, the four types of ownership are: A) a current time slot is available for contention requests; B) the current time slot is owned by a single time slot subscriber unit; C) the current time slot is 15 owned by the multiple time slot subscriber unit that sent a channel request in the time slot; and D) the current time slot is owned by a multiple time slot subscriber unit that sent a channel request in a different time slot.
A traffic type determines a maximum number of time slots per time division multiplex frame. For example, in the preferred embodiment of this invention, a first traffic type may use one time slot per frame, and a second traffic type may be selected to use up to 8 time slots per frame. A plurality of control bits, being two control bits in this example, are broadcast in a downlink for controlling contention among the first traffic type subscriber units, contention among the multiple time slot subscriber units, and between first and second traffic type subscriber units. The two control bits indicate ownership of the time slots. In this example, the control bits have the following four digital values:
00, indicating that a current time slot is available for contention requests;
01, indicating that the current time slot is owned by a single time slot subscriber unit;
WO 96/04720 2 ~ ~ 0 8 9 3 PCT/US9';/OS949 ~
10, indicating that the current time slot is owned by the multiple time slot subscriber unit that sent a channel request in this time slot; and 11, indicating that the current time slot is owned by a 5 multiple time slot subscriber unit that sent the channel requested in a different time slot.
Further, a set of controi bits is sent in an uplink indicating whether a packet segment transferred in each time 10 slot is part of a packet transmitted by a single slot user or a multiple slot user. A base site, typically a radio port, contention scheme differentiates between single time slot channel requests and multiple time slot channel requests and sets the two control bits in the downlink accordingly. A
15 subscriber unit contention scheme, executed by the subscriber unit, provides for contention for channel access, and upon the subscriber unit obtaining channel access, allows a single time slot user to transmit on a same time slot that was won in the contention and allows a multiple time slot user to transmit on 20 all time slots designated for the multiple time slot user, as is described in more detail below.
An uplink packet is typically divided into one-slot segments. For a multiple time slot packet, i.e., an N of N
25 packet, N a positive integer, a first byte of the non-header segment is used as the segment number. For a single time slot packet, i.e., a 1 of N packet, transmitted segments are acknowledged via a Stop-and-Wait automatic acknowledgment scheme. Thus, only one bit sequence number is needed. This 30 one bit sequence number is accomplished through the use of even and odd segments. The subscriber unit alternately marks a new segment as odd or even. In the event that an uplink segment is correctly received by the base site, but the downlink time slot in the following frame is not decoded by 35 the subscriber unit, the subscriber unit retransmits the old ~ WO 96/04720 2 t 7~ 9 3 PCT/US95105949 segment and the base site identifies and discards the duplicate segment by identifying that its segment type (odd or even) is identical to that of the original.
FIG. 4, numeral 400, is a flow chart of another embodiment of steps of a method for controiling time slot contention to provide fairness between a plurality of types of subscriber units communicating with a base site in a time division multiplex communication system in accordance with the present invention. The method includes the steps of: A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits (402); B) utilizing, by the base site, the predetermined number of control bits from the subscriber units to determine the type of subscriber unit to win contention and assigning the time slot to a winning subscriber unit in accordance with the determination wherein, where the winning subscriber unit is a multiple time slot user type, further assigning the remaining of the idle packet time slots ~404); C) starting, by a base site control unit, at least a first supervisory counter timer to control a length of time that the winning subscriber unit holds the time slot (406); D) utilizing, by the base site, a predetermined number of control bits to broadcast to the plurality of types of subscriber units the type of subscriber unit currently assigned to the time slot (408);
I_) utilizing, by the winning subscriber unit, a predetermined number of control bits to determine a current status of the next time slot prior to transmitting (412); and F) transmitting, by the winning subscriber unit, a packet segment and recyling to step D until one of: a packet is completed and one supervisory counter timer expires (410).
FIG. 5, numeral ~00, is a block diagram of one embodiment of a time division multiplex communication 35 system for controlling time slot contention to provide fairness WO 96/04720 2 ~ 7 0 8 9 ~ PCT/US95/ûS9~9 among transmissions for a plurality of traffic types of subscriber units in accordance with the present invention. The system includes a plurality of subscriber units (508, 514, ...) and a base site (502) that perform the steps set forth below in 5 parallel. Each subscriber unit (508, 514, ...) includes a transmission retry counter (510, 516, ...) and a random number selector (512, 518, ...). The transmission retry counter (510, 516) is utilized for storing a transmission retry count and, prior to transmission of a packet on an uplink, for inserting 10 the retry count into a packet header together with a traffic type identifier. The random number selector (512, 518, ...) is operably coupled to receive a spreading factor N and a channel bitmap from the base site and is utilized for randomly selecting a value between one and N for a wait time and 15 converting the value of the wait time to a number of time slots for waiting until retransmission. A unique contention-based access control scheme is used by the base site for prioritizing transmissions of the subscriber units The base site (502) includes an averaging unit (504) and a comparing 20 unit (506). The averaging unit (504) is used for receiving the transmissions of the packets from the subscriber units and for determining an average retry count for a predetermined time interval. The comparing unit (506) is operably coupled to the averaging unit (504) and is used for: A) where a number of 25 packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing the average retry count with a predetermined threshold value, and utilizing the unique contention-based access scheme to determine the spreading factor N, N an 30 integer, and B)where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, automatically increasing N
by a predetermined increment number and inserting N in a downlink packet together with the channel bitmap, setting a 35 predetermined number of control bits based on the traffic type WO 96/04720 ~ 1 7 0 8 9 3 PCTIUS9~/05949 of the subscriber unit in a downlink for controlling contention for time slots among the plurality of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
The unique contention-based access scheme for dletermining the spreading factor N utilized by the time division multiplex communication system is typically the scheme described above and shown in FIG. 3. In a preferred 10 embodimenl: the predetermined number of control bits is two and the control bits indicate four types of time slot ownership, where the four types are as described above.
Thus, a subscriber unit that desires to transmit a packet 15 on the uplink first contends for the channel access. The subscriber unit monitors the downlink and locates a packet time slot by reading the slow channel. When a packet time slot is found, the subscriber unit determines a status of the channel before transmitting an initial access request, i.e., an 20 uplink packet header, on that channel. A status of busy or idle bits is described herein as B/l. The subscriber unit transmits on the uplink time slot when the status is Idle (B/l=00) and, where the status is Busy (B/l ~ 00), waits a random period of time, R1. After transmitting the packet header, the subscriber 25 Ul1it reads the slow channel bits in the corresponding time slot of the next downlink packet frame to determine whether it has successfully gained access to the channel. Where the downlink WEI, i.e., word error indication, is zero and the ECHO bits are matched, the subscriber unit is allowed to continue where the 30 B/l bits are 01 for a 1-of-N user, or 10 for a N-of-N user.
Where the B/l bits are 11, the current time slot is assigned to another N-of-N user who has just won contention on another time slot prior to this time slot. Where a subscriber unit fails to seize the channel, i.e., one of: WEI ~ 0; uplink ECHO
35 downlink ECHO; B/l = 00; and B/l = 11, the transmission WO96/04720 2 t 7 Q ~ ~ 3: PCT/US9510S9i91~
process is halted and remains idle for a random amount of time, R2, before attempting to access the uplink again.
Where a subscriber unit successfully gains access to the 5 channel, it continues to send packet segments and reads the WEI and ECHO bits on the downlink slow channel of the following frame to determine the result of each transmission.
Where the WEI is set, the subscriber unit retransmits the segment. The base site, typically a radio port, maintains a 10 busy status on the time slot until it receives the DISCONNECT
segment, i.e., a last segment, of the packet. An 1-of-N user stays on the same time slot where it won the contention, while an N-of-N user may transmit on all the time slots designated for it, i.e., the time slots designated B/l = 10 or 11.
15 The transmission process is aborted where the ECHO bits do not match or the time slot is no longer a packet slot, or the time slot becomes idle.
FIG. 6, numeral 600, is a block diagram of one 20 embodiment of a process organization for a radio port/base site in accordance with the present invention wherein the radio porVbase site and the radio port control unit/base site control unit are separate units. This organizational diagram is provided for an overview of the implementation of the present 25 invention. Thus, a more detailed description of the invention follows this paragraph. Upon subscriber unit initial uplink access (604) at a first subscriber unit SU (602) of a plurality of subscriber units, uplink packet transmission is executed by the subscriber unit (602). Where the subscriber unit is a 30 single time slot user (606), the subscriber unit transmits a packet on 1 of N time slots per frame to the radio porVbase site RP (610) as set forth with more particularity in FlGs 10, 11, and 12. Where the subscriber unit is a multiple time slot user (608), the subscriber unit transmits a packet in a 35 multiplicity of time slots per frame specified by the radio ~ WO 96/04720 2 1 7 0 8 9 3 PCT/US95/059~9 porVbase site (610), as set forth with more particularity in FlGs 13, 14, 15, and 16. The radio port/base site (610) is responsive to the subscriber units and to the radio port control unit/base site control unit, performs uplink processing (612) 5 as described with more particularity with respect to FlGs. 18 and 19, and performs downlink processing (614) as described with more particularity with respect to FIG. 20. The radio port control uniVbase site control unit RPCU (616) provides processing (618) of each time slot of data or segment of data 10 received from the radio porVbase site, as described with more particularity with respect to FIG. 21.
FIG. 7, numeral 700, is a block diagram of a time division multiplex communication system for controlling time slot 15 contention to provide fairness among transmissions for a plurality of traffic types of subscriber units in accordance with the present invention. The radio port control uniVbase site control unit (702) includes supervisory counters (704), i.e., timers, that control a length of time that a subscriber unit 20 holds a time slot. The radio port/base site (706) includes a time slot manager (708) for assigning the time slots to subscriber units in accordance with a time slot winner determination by a contention controller and the contention controller (710) for determination of the time slot winner in 25 accordance with the unique contention-based access scheme of the present invention as described more fully above. The plurality of subscriber units includes at least one single time slot user, i.e., 1-of-N, having an ECHO bits generator (714) for generating ECHO bits that are used as described above and a 30 transmission queue (716), Tx queue and a retransmission queue, ReTx queue (730), and at least one multiple time slot user, i.e., N-of-N, that includes a counter (720), and ECHO bits generator, a Tx queue (724), a retransmission queue/ReTx queue (726), and an acknowledgment queue/Ack queue (728).
35 These elements are utilized as described herein.
WO ~610J720 2 ~ 7 0 8 ~?3 PCT/1159~/059~9~
FIG. 8, numeral 800, is a schematic drawing of an exemplary simplified TDMA frame structure that contains 8 time slots and shows a typical uplink packet structure. The 5 subscriber unit transmits on the uplink and receives on the downlink. A devoted out-of-band signaling channel such as the slow channel in a Personal Communications System, for example the PACS, (Personal Access Communication System), may be selected to be used for transmission of contention 10 control bits, thereby maximizing bandwidth utilization on the fast channel. An in-band signaling channel such as the fast channel in PACS may be used for transmission of user information. The slow channel is a small portion of the PACS
time slot which typically carries the control information such 15 as time slot usage (802), echo bits (804), and access type (806). The fast channel is a larger portion of the PACS time slot which generally carries user information and a minimum amount of control information such as user identification, packet length and packet number. In the preferred embodiment 20 the downlink slow channel bits typically include: PCI, packet - channel bits that are used by the base site to indicate that a current time slot is a packet time slot; B/l, busy/ldle bits that are used by the base site to indicate the status of the uplink to all subscriber units; ECHO bits, bits that are generated 25 randomly and transmitted by a subscriber unit on an uplink and echoed back by a base site to the subscriber unit on the downlink in the next frame; and WEI, a word error indication bit that is used by the base site to indicate to the subscriber units whether the uplink time slot in the previous frame was 30 received correctly. The uplink slow channel bits typically include: PCI (802), packet channel indication bits that are used by the subscriber unit to indicate that the current time slot is a packet time slot; ECHO bits (804), bits that are generated randomly and transmitted by the subscriber unit on an uplink 35 and echoed back by a base site to the subscriber unit on the ~ WO 96/04720 2 1 7 0 ~ 9 3 PCT/US95/05949 downlink in the next frame; and SEG bits (806), bits that are used by the subscriber unit to indicate which type of segment is being transmitted. For example, SEG bits may be encoded as follows: 000 for a first segment of a 1-of-N packet; 001 for a 5 rrliddle odd segment of a 1-of-N packet; 010 for a middle even segment of a 1-of-N packet; 011 for a disconnect segment of a 1-of-N packet; 100 for a first segment of an N-of-N packet;
101 for a middle of an N-of-N packet; 110 for reserved; and 111 for a disconnect segment of an N-of-N packet.
FIG. 9, numeral 900, is a flow chart showing steps for one embodiment of an uplink initial access control scheme executed by a subscriber unit in accordance with the method of the present invention. In the idle mode a subscriber unit 15 monitors the downlink for possible messages addressed to itself. Upon desiring to transmit a new packet (902), the subscriber unit reads the slow channel on a downlink time slot to determine whether it contains vaild information, i.e., cyclical redundancy code (CRC) is successful (904) and the 20 time slot is a packet time slot (906). Where the CRC is invalid or time slot is a non-packet time slot, the subscriber unit waits a first random period of time (908), R1 slots, before contending again on the same or a different time slot (904).
Where the CRC is valid and the time slot is a packet time slot, the subscriber unit reads the slow channel bits to d~termine the busy/idle, B/l, status of the time slot (910).
Where the time slot is busy, the subscriber unit delays R1 time slots (908) and recycles to contending again on a new time slot (904). Where the channel is idle, the subscriber unit may transmit a request segment. The subscriber unit transmits a packet header (912) to request channel access, waits exactly one frame (914), and checks the downlink CRC again (916). The packet header contains information about the packet and the subscriber unit that sent it.
WO 96/04720 2 ~ 7 0 8 ~ 3 PCT/US95/059~9~
Where a cyclical redundancy code (CRC) fails, the subscriber unit delays R2 time slots (932), R2 a second random period of time, and recycles to determining the status on a 5 new time slot (904). Where the cyclical redundancy code is successful, the subscriber unit determines whether the same time slot,i.e., the time slot where the subscriber unit transmitted on in the previous frame, is still a packet time slot (918). Where the time slot is no longer a packet time 10 slot, the subscriber unit delays R1 time slots (908), and recycles to contending again on another time slot (904). Where the time slot is still a packet time slot, the WEI bit is checked (916) for collision detection. That is, where two or more subscriber units try to transmit request segments at the same 15 time and the radio porVbase site does not capture anyone's segment,i.e., there is a collision, the WEI bit in the next downlink is set equal to one; and where only one subscriber unit's segment is received correctly by the radio porVbase site, the WEI bit in the next downlink is set equal to zero.
20 Where the WEI bit is 1, the subscriber unit recycles to delaying R1 time slots (908). Where the WEI bit is 0, the subscriber unit checks for matching ECHO bits (922). That is, where a subscriber unit is sending an access request segment or a normal segment, it randomly assigns a number to an ECHO
~5 field. A typical ECHO field is three bits. The radio port/base site is required to echo this field back on the same time slot in the following frame of the downlink. All subscriber units that are transmitting check this field against their random number that was previously transmitted. Where the subscriber 30 unit's random number is a non-match for the value in the ECHO
field, then the subscriber unit waits another random period of time R2 time slots and contents again on a new time slot. The value of R1 and R2 are determined using a novel scheme described in FlGs 1, 2, and 3.
~ WO 9~/047~ PCT/~J59S/05949 Where ~he ECHO bits are matching and the subscriber unit is a single time slot user, the subscriber unit checks the B/l bits (928). Checking B/l bits confirms the result of the subscriber unit's channel access request. A value of 00 5 indicates that there was a collision and no subscriber unit is assigned to the time slot. A value of 11 indicates that the time slot will be reserved for another user as soon as next frame. Where the B/l bits ~ 01, the subscriber unit recycles to delaying R1 time slots and contends for channel access again 10 as previously described. Where the B/l bits =01, the subscriber unit begins to transmit segments on the time slot.
Where the ECHO bits are matching and the subscriber unit is a multiple time slot user, the subscriber unit checks the B/l bits (926). Where the B/l bits ~ 10, the subscriber unit recycles to 15 delaying R1 time slots (908). Where the B/l bits = 10, the subscriber unit begins to transmit segments using one or more idle time slot per frame.
FlGs 10, 11, and 12 numeral 1000, 1100, and 1200 20 respectively, are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by a single-slot subscriber unit in accordance with the method of the present invention. The subscriber unit, after gaining access to the channel (930), checks the B/l bits (1002) to 25 determine whether the time slot is busy, i.e., B/l = 01. Where the B/l bits ~ 01, the subscriber unit aborts the transmission and recycles to delaying R1 time slots (908) as previously described. Where the B/l bits =01, the subscriber unit begins to transmit segment on the time slot (1004). The segment 30 transmission process, shown in FIG. 12, works as follows: all segments of a packet are intially stored in the transmission queue, Tx queue. The ReTx is initially empty. The subscriber unit first checks the retransmission queue, ReTx queue (1202).
Where the ReTx queue is non-empty, segment at the front of 35 the ReTx queue is transmitted (1206). Where the ReTx is empty WO 96/04720 ~ 7 Ci 8 9 3 PCT/US95/OS9 19 and the Tx queue is non-empty, the subscriber unit transmits a segment at the front of the Tx queue (1208) and moves the segment to the ReTx queue (1210). Where both the ReTx and Tx queues are empty, the subscriber unit begins to transmit DISCONNECT segment (1024) as shown in FIG. 11 and described I ater.
After transmitting a segment, the subscriber unit waits exactly one frame (1006) and reads/decodes the downlink time slot (101 0),i.e., the time slot where the subscriber unit transmitted on in the previuos frame. Where the downlink time slot is error-free, i.e., a CRC is successful, ~he subscriber unit checks the PCI bits to determine whether l:he time slot is still a packet slot (1014). Where the time slot is a non packet time slot, the subscriber unit delays R1 time slots (908) and proceeds as previously described. Where the time slot is still a packet time slot, the subscriber unit checks the WEI bit (1008). The WEI bit on the downlink indicates whether the uplink segment on the previous frame was received correctly by the base site/radio port. Where the WEI bit is 1, i.e., the previous transmitted segment was received incorrectly, the subscriber unit proceeds to check the B/l status (1002) in order to transmit the same segment again,i.e., the segment which is still in the ReTx queue. Where the WEI is 0, the subscriber unit compares ECHO bits (1020), and where there is a non-match, delays R1 time slots (908) and proceeds as previously described. Where the ECHO bits match, the subscriber unit discards the segment in the ReTx queue and proceeds to check the B/l status in order to transmit a new segment.
Where the downlink time slot contains error, i.e., the CRC
is invalid, the subscriber unit skips transmitting on the current time slot and waits exactly one frame (1008) and recycles to decode the downlink again until the CRC is valid.
~ WO 96/04720 2 1 7 0 8 9 3 PCT/US95/05949 The subscriber unit then determines whether the time slot is still a packet slot (1016). Where the time slot is a non packet time slot, the subscriber unit delays R1 time slots (908) and proceeds as previously described. Where the time slot is still 5 a packet time slot, the subscriber unit proceeds immidiately to checking the B/l status in order to transmit a segment without checking the previous transmission result since the subscriber unit did not transmit in the last frame.
Where both the ReTx queue and the Tx queue are empty when it is time to transmit a segment (1024), the subscriber unit transmits a DISCONNECT segment (1102), waits exactly one frame (1104), and decodes the downlink. The subscriber unit recycles to transmitting DISCONNECT segment until one 15 of: a CRC is invalid (1106), the CRC is valid but the time slot becomes non packet slot (1108), and the CRC is valid and the time slot is still a packet and the WEI is 0 (1110). Where one of these events occurs the subscriber unit completes the uplink packet transmission (11 12) FlGs 13, 14, 15, and 16, numeral 1300, 1400, 1500, and 1600 respectively, are a flow chart showing steps for one embodiment of an uplink packet transmission scheme executed by the multi-slot subscriber unit in accordance with the 25 method of the present invention. Each subscriber unit of the multi-slot subscriber unit traffic type has three data queues: a transmission, Tx, queue, for storing segments of the original packet without the DISCONNECT segments; a retransmission, ~eTx, queue for storing segments awaiting to be 30 retransmitted; and an acknowledgment queue, Ack queue, for storing the segments which are waiting for acknowledgment.
Where the subscriber unit successfully gains access to - the uplink (932), the subscriber unit begins to transmit packet 35 segment (1302). The ReTx queue is checked (1502) to -WO 96/04720 2 1 ~ Q ~ ~ 3 PCTIUS95/OS9~9~
determine whether there is a segment to retransmit. Where there is a segment to retransmit, i.e., the ReTx queue is non empty, the subscriber unit transmits a segment at the front of the ReTx queue (1~06) and moves the segment from the ReTx 5 queue to the Ack queue (1514). Where the ReTx queue is empty, the subscriber unit checks the Tx queue to determine whether there is a new segment to transmit (1504). Where there is a new segment to transmit, i.e., the Tx queue is non empty, the subscriber unit transmits a segment at the front of the Tx 10 queue (1508) and moves the segment from the Tx queue to the Ack queue (1512). Where both the ReTx queue and the ReTx queue are empty and the Ack queue is non empty, the subscriber unit transmits a "fill" segment. Where both the ReTx queue and the ReTx queue are empty and the Ack queue is 15 also empty, the subscriber unit proceeds to transmitting DISCONNECT segment (1304) as shown in FIG.14 and is described below.
Where a normal segment (1302) or a "fill" segment 20 (1306) was transmitted, the subscriber unit waits until the beginning of the next time slot (1308) to determine the transmission result of the segment sent in the current time slot of the previous frame. Where no segment was transmitted in the current time slot of the previous frame, the subscriber 25 unit performs CRC check (1312) and determines whether the current time slot is still a packet time slot (1316). Where the CRC is invalid, or the CRC is valid but the time slot becomes a non packet time slot, the subscriber unit recycles to waiting until the next time slot. Where the CRC is valid and the time 30 slot is still a packet time slot, the subscriber unit proceeds to checking the B/l bits (1320). Where there was a segment, k, transmitted in the current time slot of the previous frame, the subscriber unit determines transmission result and performs neccessary actions (1314) as shown in FIG. 16 and is described 35 in the following: The subscriber unit performs CRC check ~ WO 96/04720 2 1 7 0 8 ~ 3 PC'T/U595/059J9 (1602) and determines whether the current time slot is still a packet time s.ot (1606). Where the CRC is invalid, or the CRC
is valid but the time slot becomes a non packet time slot, the subscriber unit moves segment k from the Ack queue to the 5 ReTx queue (1604) and recycles to waiting until the next time slot (1308). Where the CF2C is valid and the time slot is still a packet time slot, the subscriber unit checks the WEI bit (1608) and the ECHOs (1610). Where the WEI is 1, the subscriber unit moves segment k from the Ack queue to the ReTx queue (1614) 10 and proceeds to checking the B/l bits (1320). Where the WEI is 0 and the ECHOs match, the subscriber unit deletes segment k from the Ack queue (1612) and proceeds to checking the B/l bits (1320). Where the WEI is 0 and the ECHOs are non match, the subscriber unit aborts the transmission (1318) and 15 proceeds to delay R1 slot (908) as previously described.
The subscriber unit checks the B/l (1320) to determine whether the time slot is reserved for it. Where the B/ls are either '00' or '01', i.e., B/l ~ 1x, the subscriber unit proceeds to 20 the next time slot (1308). Where the B/l bits are either '10' or '11 ', i.e., B/l = 1 x, the subscriber unit recycles to transmitting a new segment (1302).
Where the ReTx queue, the ReTx queue, and the Ack queue 25 are all empty, the subscriber unit transmits a DISCONNECT
segment (1402), waits until next time slot (1406), and decodes the downlink (1408). Where the CRC is invalid, or the CF2C is valid but the time slot becomes a non packet time slot, the subscriber unit finishes the packet transmission. Where 30 the CRC is valid and the time slot is still a packet time slot, the subscriber unit determines whether there was a - DISCONNECT segment transmitted in the same time slot of the previous frame (1412). Where there was no DISCONNECT
segment transmitted in the same time slot of the previous 35 frame, the subscriber unit proceeds to check the B/l bits WO 96/04720 ~ 1~ 7 Q 8: 9 ~ PCT/US9S/059 (1404). Where there was a DISCONNECT segment transmitted in the same time slot of the previous frame. the subscriber unit checks the WEI bit (1414) to determine the transmission result. Where the WEI is 0, the subscriber unit finishes the 5 packet transmission (1416). Where the WEI is 1, the subscriber unit proceeds to check the B/l bits (1404). Where the B/ls are either '00' or '01', the subscriber unit proceeds to the next time slot (1406). Where the B/l bits are either ~10~ or '1 1', i.e., B/l = 1 x, the subscriber unit recycles to transmitting 10 another DISCONNECT segment (1402).
FIG. 17, numeral 1700, iS a schematic showing, for a block diagram of elements operating in accordance with the present invention, one implementation of slow channel bits and 15 variables exchanged between a subscriber unit, a radio port and a radio port control unit in accordance with the present invention. The subscriber unit (1702) sends a segment (1708) that includes PCI bits, ECHO bits, and SEG bits and receives from the radio port a segment (1710) that includes PCI bits, 20 ECHO bits, B/l bits, and a WEI bit. The radio port (1704) sends a segment (1712) that includes PCI bits, B/l bits, SEG bits, and a WEI bit and receives from the radio port control unit (1706) a segment (1714) that includes PCI bits and a plurality of control bits, C0, C1, C2.
FIG. 18, numeral 1800, and FIG. 19, numeral 1900, show a flow chart for steps in one embodiment for an uplink processing scheme in accordance with the method of the present invention. The uplink processing scheme is executed 30 by the radio porVbase site, hereafter referenced as the radio port, RP, upon receiving an uplink segment from a subscriber unit (1802). The RP checks a variable M_PCI to determine whether a current time slot is a packet slot (1804). This packet channel indication variable was set when the RP
35 executed a downlink algorithm. Where the slot is not a packet ~ WO 96/04720 2 1 7 0 8 ~ 3 PCT/US95105919 slot, no action is taken by the RP. Where the slot is a packet slot, the RP checks whether there is an N-of-N user currently using the channel (1806). This event is indicated by the variable N-of-N flag. Typically, where there is an N-of-N user 5 using the channel, the N-of-N flag is one, and the RP
determines whether the time slot is idle (1808), i.e., typically whether M_B/I = 00. Where the N-of-N flag is one and the time slot is idle, i.e., M_B/I = 00, the RP sets the M_B/I variable to 11 (1810). Where the N-of-N flag is zero and the B/l status is 10 either 10 or 11 (1820), the B/l status is set to idle (1818), iae., M_B/I = 00. The RP then determines whether a CRC check is valid (1812). Where the CRC check is invalid (1812), the RP
the variable M_WEI is set to one (1814) and a one is stored in a WElpfield (1816). Then the RP delivers an uplink segment to 15 the radio port control uniVbase site control unit (1916).
Where the CRC check is valid (1812), the RP determines a whether the variable M_B/l is equal to 00 (1902), proceeding as set forth in FIG. 19. Where the channe! is idle, i.e., M_B/l 20 =00, the RP checks the SEGA to determine whether the uplink slot is a first time slot of a packet (1904), i.e., a first two bits of SEGA equal 00. This is the only type of uplink time slot that is permitted when the channel is idle. The RP determines whether an N-of-N flag is set (1906). Where the N-of-N flag is 25 clear, the RP determines whether the time slot is from a 1-of-N user or a from a N-of-N user (1908), i.e., whether SEGA = Oxx.
Where the time slot is from a 1-of-N user, the B/l bits are set to 01 (1910), i.e., are set to indicate busy. Where the time slot is from a N-of-N user, the B/l bits are set to 10 to indicate 30 the time slot is busy and the N-of-N flag is set to one (1912).
Where, after the RP determines that the CRC check is valid, the variable M_B/l is unequal to 00 (1902), the RP determines whether SEGA equals xO0 (1918). Where SEGA equals xO0, a one is stored in a WElpfield (1816), and the RP delivers an 35 uplink time slot to the radio port control uniVbase site WO 96/04720 2 t 7 ~ ~ 9 ~. PCT/US95/059~1~
control unit (1916). Where the SEGA is unequal to xO0, the RP
determines whether SEGAequals x11 (1920). Where SEGAis~
equal to x11 (1920), the RP determines whether SEGA equals OXX (1922). Where SEGA equals Oxx, the variable M_B/I is set 5 to 00 (1926), i.e., is set to idle.. Where SEGA is unequal to Oxx, the N-of-N flag is set to zero (1924) and the variable M_B/I is set to 00 (1926), i.e., is set to idle. The ECHO bits received in the variable M_ ECHO are saved, the M_ WEI variable is set to zero, the segment packet SEGP jS updated to SEGA, the B/lp 10 packet bits are updated to B/IA and N/Np is set to the N-of-N
flag (1914). The states of the B/l bits and the N-of-N flag are sent to the radio port control uniVbase/site when the RP
delivers the uplink time slot to the radio port control uniVbase site (1916). Where the RP is expecting a first time 15 slot of a packet and the uplink time slot is not a first time slot of a packet, or the RP is expecting a non-first time slot of a packet and the time slot is a first time slot of a packet (1904 and 1918), the RP processes the packet as if the packet had failed the CRC check.
The RP maintains a busy status of the uplink channel as long as the disconnect segment, SEGA = X11, jS not received.
Where the disconnect segment is received, the RP sets the M_B/I variables equal to 00 to indicate that the uplink is idle.
FIG. 20, numeral 2000, is a flow chart for one embodiment of a downlink processing scheme executed by a radio port in accordance with the present invention. The steps of this method are executed whenever a downlink time slot is 30 received from the radio port control uniVbase site to be transmitted (2002). The RP determines whether PClp bits indicate that the slot is a packet slot (2004). Where the PClp bits indicate that the slot is not a packet slot, the RP inserts the WEI bit computed from the same time slot in the previous 35 uplink frame (2018) and broadcasts the downlink time slot to WO 96/047~0 PCT/US95/OS9~9 the subscriber units (2020). Where the PClp bits indicate that the slot is a packet slot, the RP determines whether a control bit C2 iS set to one (2006). Where C2 is set to 1, an N-of-N
flag is set to zero (2008). The RP then determines whether 5 other control bits, C0 and C1, are set to 01 (2010). Where C0 and C1, are set to 01, the RP stores 00 in the M_B/I (2014).
Where C0 and C1, are set to 10, the RP stores 01 in the M_B/I
(2012). The B/IA bits from the previous uplink transmission are saved in the M_B/I variable and the M_PCI bits are updated 10 to packet bits (2016). Then the RP inserts the WEI bit computed from the same time slot in the previous uplink frame (2018) and broadcasts the downlink time slot to the subscriber units (2020) .
FIG. 21, numeral 2100, is a flow chart for one embodiment of an uplink processing scheme executed by a radio port control unit in accordance with the present invention. Typically, the radio port control uniVbase site c~ntrol unit, i.e., RPCU, maintains an individual counter, 20 Dcounter, for each TDMA time slot, and a common counter, Ncounter, for all the TDMA time slots. Upon receiving a slot or segment from the radio porVbase site (2102), the RPCU first determines whether the current time slot is a packet slot (2106), i.e., whether PClp equals packet. Where the current 25 time slot is not a packet slot, Ncounter is incremented by one (2108). Where the current time slot is a packet slot, the RPCU
determines whether a CRC check is valid (2110), i.e., whether WElp is equal to zero. Where WElp is equal to zero ~2110), the Rl'CU copies the values of the B/l bits and the N/N bits into the 30 variables Busy and Nflag, respectively (2104), the Dcounter is set equal to 0 (2118), and the RPCU determines whether a most significant bit of the segment SEGp is zero (2120), i.e., whether SEGp=Oxx. Where the most significant bit of the segment SEGp is zero, Ncounter is incremented by one (2122).
35 Where WElp is unequal to zero (2110), the RPCU determines W096/W720 2 ~ 7 0 8 9 ~ PCT~S95/059~
whether the channel is busy (2112). Where the channel is busy (2112), Dcounter is incremented by one (2114) and Ncounter is set to zero (2124). Where the channel is idle (2112), Dcounter is set to zero (2121) and Ncounter is set to zero (2124). Upon 5 the Ncounter being set to zero or one (2124, 2108, 2122), the RPCU determines whether the Dcounter has exceeded its maximum value (2126). Where the Dcounter has exceeded its maximum value, control bits C0 and C1 are set to 01 (2130).
Where the Dcounter has not exceeded its maximum value, 10 control bits C0 and C1 are set to 00 (2128). After setting the control bits C0 and C1 (2128, 2130), the RPCU determines whether the Ncounter is greater than or equal to the number of time slots per TDMA frame (2132), i.e., 8. Where the N counter is greater than or equal to the number of time slots per TDMA
15 frame, the RPCU determines whether an Nflag is set to one (2134). Where the Nflag is set to one, the control variable C2 is set to one (2136) and the RPCU inserts C2, C1, and C0 into the slow channel of the corresponding time slot on the next downlink frame (2140). Where the Nflag is not set to one, the 20 control variable C2 iS set to zero (2138) and the RPCU inserts C2, C1, and C0 into the slow channel of the corresponding time slot on the next downlink frame (2140).
Although exemplary embodiments are described above, it 25 will be obvious to those skilled in the art that many alterations and modifications may be made without departing from the invention. Accordingly, it is intended that all such alterations and modifications be included within the spirit and scope of the invention as defined in the appended claims.
Throughout the description above "x" means a "don't care"
bit, as is commonly understood.
Claims (9)
OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method for controlling contention for a plurality of idle time slots, at a base site, to provide fairness between a plurality of types of subscriber units in a time division multiplex communication system, comprising the steps of:
A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits; and B) utilizing, by the base site, the predetermined number of control bits from the subscriber units and supervisory counter bits generated at the base site in accordance with a scheme to determine allocation of the time slot, and where a subscriber unit uses multiple time slots, to determine allocation of remaining idle packet time slots, wherein the predetermined number of control bits include echo bits and traffic type identification bits, and wherein number of traffic type identification bits is at least two and the traffic type identification bits indicate at least four types of time slot ownership in accordance with a predetermined contention scheme.
A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits; and B) utilizing, by the base site, the predetermined number of control bits from the subscriber units and supervisory counter bits generated at the base site in accordance with a scheme to determine allocation of the time slot, and where a subscriber unit uses multiple time slots, to determine allocation of remaining idle packet time slots, wherein the predetermined number of control bits include echo bits and traffic type identification bits, and wherein number of traffic type identification bits is at least two and the traffic type identification bits indicate at least four types of time slot ownership in accordance with a predetermined contention scheme.
2. The method of claim 1 wherein the four types of time slot ownership are:
A) a current time slot is available for contention requests;
B) the current time slot is owned by a single time slot subscriber unit;
C) the current time slot is owned by the multiple time slot subscriber unit that sent a channel request in the time slot; and D) the current time slot is owned by a multiple time slot subscriber unit that sent a channel request in a different time slot.
A) a current time slot is available for contention requests;
B) the current time slot is owned by a single time slot subscriber unit;
C) the current time slot is owned by the multiple time slot subscriber unit that sent a channel request in the time slot; and D) the current time slot is owned by a multiple time slot subscriber unit that sent a channel request in a different time slot.
3. A method for controlling contention for a plurality of idle time slots, at a base site, to provide fairness between a plurality of types of subscriber units in a time division multiplex communication system, comprising the steps of:
A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits; and B) utilizing, by the base site, the predetermined number of control bits from the subscriber units and supervisory counter bits generated at the base site in accordance with a scheme to determine allocation of the time slot, and where the subscriber unit uses multiple time slots, to determine allocation of remaining idle packet time slots, wherein the scheme to determine allocation of the time slot, and where the subscriber unit uses multiple time slots, to determine allocation of the remaining idle packet time slots comprises:
each subscriber unit performing the steps of:
A) utilizing a transmission retry counter for storing a transmission retry count and, prior to transmission of a packet on an uplink, inserting the retry count into a packet header together with a traffic type identifier, B) utilizing a random number selector for randomly selecting a value between one and a spreading factor N, where N, N an integer, and a channel bitmap are obtained from the base site, for a wait time and converting the value of the wait time to a number of time slots for waiting until retransmission, the base site control unit performing the steps of:
C) utilizing an averaging unit for receiving the transmissions of the packets from the subscriber units and for determining an average retry count for a predetermined time interval, D) utilizing a comparing unit for:
D1) where a number of packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing the average retry count with a predetermined threshold value, and utilizing a contention-based access scheme to determine the spreading factor N, and D2) where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, automatically increasing N by a predetermined increment number and inserting N in a downlink packet together with the channel bitmap, setting a predetermined number of control bits based on the traffic type of the subscriber unit in a downlink channel for controlling contention for time slots among the plurality of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits; and B) utilizing, by the base site, the predetermined number of control bits from the subscriber units and supervisory counter bits generated at the base site in accordance with a scheme to determine allocation of the time slot, and where the subscriber unit uses multiple time slots, to determine allocation of remaining idle packet time slots, wherein the scheme to determine allocation of the time slot, and where the subscriber unit uses multiple time slots, to determine allocation of the remaining idle packet time slots comprises:
each subscriber unit performing the steps of:
A) utilizing a transmission retry counter for storing a transmission retry count and, prior to transmission of a packet on an uplink, inserting the retry count into a packet header together with a traffic type identifier, B) utilizing a random number selector for randomly selecting a value between one and a spreading factor N, where N, N an integer, and a channel bitmap are obtained from the base site, for a wait time and converting the value of the wait time to a number of time slots for waiting until retransmission, the base site control unit performing the steps of:
C) utilizing an averaging unit for receiving the transmissions of the packets from the subscriber units and for determining an average retry count for a predetermined time interval, D) utilizing a comparing unit for:
D1) where a number of packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing the average retry count with a predetermined threshold value, and utilizing a contention-based access scheme to determine the spreading factor N, and D2) where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, automatically increasing N by a predetermined increment number and inserting N in a downlink packet together with the channel bitmap, setting a predetermined number of control bits based on the traffic type of the subscriber unit in a downlink channel for controlling contention for time slots among the plurality of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
4. The method of claim 3 wherein the unique contention-based access scheme for determining the spreading factor N comprises the steps of:
A) where the average retry count is greater than or equal to the predetermined threshold value plus a first predetermined constant, increasing N by a value equal to a predetermined percentage of the average retry count, B) where the average retry count is less than or equal to the predetermined threshold value minus a second predetermined constant, decreasing N by a predetermined decrease amount, and C) where the average retry count lies between the predetermined threshold value minus the second predetermined constant and the predetermined threshold value plus the first predetermined constant, maintaining a current value of N.
A) where the average retry count is greater than or equal to the predetermined threshold value plus a first predetermined constant, increasing N by a value equal to a predetermined percentage of the average retry count, B) where the average retry count is less than or equal to the predetermined threshold value minus a second predetermined constant, decreasing N by a predetermined decrease amount, and C) where the average retry count lies between the predetermined threshold value minus the second predetermined constant and the predetermined threshold value plus the first predetermined constant, maintaining a current value of N.
5. A method for controlling time slot contention to provide fairness between a plurality of types of subscriber units communicating with a base site in a time division multiplex communication system, comprising the steps of:
A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits;
B) utilizing, by the base site, the predetermined number of control bits from the subscriber units to determine the type of subscriber unit to win contention and assigning the time slot to a winning subscriber unit in accordance with the determination wherein, where the winning subscriber unit is a multiple time slot user type, further assigning remaining idle packet time slots;
C) starting, by a base site control unit, at least a first supervisory counter timer to control a length of time that the winning subscriber unit holds the time slot;
D) utilizing, by the base site, the predetermined number of control bits to broadcast to the plurality of types of subscriber units the type of subscriber unit currently assigned to the time slot;
E) utilizing, by the winning subscriber unit, a predetermined number of control bits to determine a current status of a next time slot prior to transmitting;
and F) transmitting, by the winning subscriber unit, a packet segment and returning to step D until one of: a packet is completed and one supervisory counter timer expires.
A) contending, by the plurality of types of subscriber units, on an idle packet time slot by transmission of a predetermined number of control bits;
B) utilizing, by the base site, the predetermined number of control bits from the subscriber units to determine the type of subscriber unit to win contention and assigning the time slot to a winning subscriber unit in accordance with the determination wherein, where the winning subscriber unit is a multiple time slot user type, further assigning remaining idle packet time slots;
C) starting, by a base site control unit, at least a first supervisory counter timer to control a length of time that the winning subscriber unit holds the time slot;
D) utilizing, by the base site, the predetermined number of control bits to broadcast to the plurality of types of subscriber units the type of subscriber unit currently assigned to the time slot;
E) utilizing, by the winning subscriber unit, a predetermined number of control bits to determine a current status of a next time slot prior to transmitting;
and F) transmitting, by the winning subscriber unit, a packet segment and returning to step D until one of: a packet is completed and one supervisory counter timer expires.
6. A time division multiplex communication system for controlling time slot contention to provide fairness among transmissions for a plurality of traffic types of subscriber units, comprising:
A) a plurality of subscriber units having a plurality of traffic types, each subscriber unit including:
A1) a transmission retry counter for storing a transmission retry count and, prior to transmission of a packet on an uplink, inserting the retry count into a packet header together with a traffic type identifier, A2) a random number selector, operably coupled to receive a spreading factor N
and a channel bitmap from a base site, for randomly selecting a value between one and N for a wait time and converting the value of the wait time to a number of time slots for waiting until retransmission, B) a base site having a contention-based access control scheme for prioritizing transmissions of the subscriber units including:
B1 ) an averaging unit, for receiving the transmissions of the packets from the subscriber units and for determining an average retry count for a predetermined time interval, B2) a comparing unit, operably coupled to the averaging unit, for:
where a number of packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing the average retry count with a predetermined threshold value, utilizing the contention-based access scheme to determine the spreading factor N, N an integer, and where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, automatically increasing N by a predetermined increment number, and inserting N in a downlink packet together with the channel bitmap, setting a predetermined number of control bits of the downlink packet based on the traffic type of the subscriber unit in a downlink channel for controlling contention for time slots among the plurality of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
A) a plurality of subscriber units having a plurality of traffic types, each subscriber unit including:
A1) a transmission retry counter for storing a transmission retry count and, prior to transmission of a packet on an uplink, inserting the retry count into a packet header together with a traffic type identifier, A2) a random number selector, operably coupled to receive a spreading factor N
and a channel bitmap from a base site, for randomly selecting a value between one and N for a wait time and converting the value of the wait time to a number of time slots for waiting until retransmission, B) a base site having a contention-based access control scheme for prioritizing transmissions of the subscriber units including:
B1 ) an averaging unit, for receiving the transmissions of the packets from the subscriber units and for determining an average retry count for a predetermined time interval, B2) a comparing unit, operably coupled to the averaging unit, for:
where a number of packets received during a predetermined reception interval is greater than or equal to a predetermined number of packets, comparing the average retry count with a predetermined threshold value, utilizing the contention-based access scheme to determine the spreading factor N, N an integer, and where the number of packets received during the predetermined reception interval is less than the predetermined number of packets, automatically increasing N by a predetermined increment number, and inserting N in a downlink packet together with the channel bitmap, setting a predetermined number of control bits of the downlink packet based on the traffic type of the subscriber unit in a downlink channel for controlling contention for time slots among the plurality of traffic types of packet traffic user subscriber units, and transmitting the downlink packet to the subscriber units.
7. The time division multiplex communication system of claim 6 wherein the contention-based access scheme for determining the spreading factor N
includes:
where the average retry count is greater than or equal to the predetermined threshold value plus a first predetermined constant, increasing N by a value equal to a predetermined percentage of the average retry count, where the average retry count is less than or equal to the predetermined threshold value minus a second predetermined constant, decreasing N by a predetermined decrease amount, and where the average retry count lies between the predetermined threshold value minus the second predetermined constant and the predetermined threshold value plus the first predetermined constant, maintaining a current value of N.
includes:
where the average retry count is greater than or equal to the predetermined threshold value plus a first predetermined constant, increasing N by a value equal to a predetermined percentage of the average retry count, where the average retry count is less than or equal to the predetermined threshold value minus a second predetermined constant, decreasing N by a predetermined decrease amount, and where the average retry count lies between the predetermined threshold value minus the second predetermined constant and the predetermined threshold value plus the first predetermined constant, maintaining a current value of N.
8. The time division multiplex communication system of claim 6 wherein the predetermined number of control bits is two and the control bits indicate four types of time slot ownership.
9. The time division multiplex communication system of claim 8 wherein the four types of time slot ownership are:
A) a current time slot is available for contention requests;
B) the current time slot is owned by a single time slot subscriber unit;
C) the current time slot is owned by the multiple time slot subscriber unit that sent a channel request in the time slot; and D) the current time slot is owned by a multiple time slot subscriber unit that sent a channel request in a different time slot.
A) a current time slot is available for contention requests;
B) the current time slot is owned by a single time slot subscriber unit;
C) the current time slot is owned by the multiple time slot subscriber unit that sent a channel request in the time slot; and D) the current time slot is owned by a multiple time slot subscriber unit that sent a channel request in a different time slot.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US08/282,242 US5530700A (en) | 1994-07-29 | 1994-07-29 | Method and device for controlling time slot contention to provide fairness between a plurality of types of subscriber units in a communication system |
| US08/282,242 | 1994-07-29 | ||
| PCT/US1995/005949 WO1996004720A1 (en) | 1994-07-29 | 1995-05-15 | Method and device for providing time slot contention fairness between subscriber units |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2170893A1 CA2170893A1 (en) | 1996-02-15 |
| CA2170893C true CA2170893C (en) | 2000-06-13 |
Family
ID=23080642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002170893A Expired - Fee Related CA2170893C (en) | 1994-07-29 | 1995-05-15 | Method and device for controlling time slot contention to provide fairness between a plurality of types of subscriber units in a communication system |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US5530700A (en) |
| EP (1) | EP0721706A4 (en) |
| CN (1) | CN1131482A (en) |
| AU (1) | AU685282B2 (en) |
| CA (1) | CA2170893C (en) |
| WO (1) | WO1996004720A1 (en) |
Families Citing this family (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6088590A (en) | 1993-11-01 | 2000-07-11 | Omnipoint Corporation | Method and system for mobile controlled handoff and link maintenance in spread spectrum communication |
| US6005856A (en) | 1993-11-01 | 1999-12-21 | Omnipoint Corporation | Communication protocol for spread spectrum wireless communication system |
| US5732076A (en) * | 1995-10-26 | 1998-03-24 | Omnipoint Corporation | Coexisting communication systems |
| JPH09163431A (en) * | 1995-12-12 | 1997-06-20 | Fujitsu Ltd | Radio channel allocating method |
| US6122500A (en) * | 1996-01-24 | 2000-09-19 | Ericsson, Inc. | Cordless time-duplex phone with improved hearing-aid compatible mode |
| US7184426B2 (en) * | 2002-12-12 | 2007-02-27 | Qualcomm, Incorporated | Method and apparatus for burst pilot for a time division multiplex system |
| US9118387B2 (en) | 1997-11-03 | 2015-08-25 | Qualcomm Incorporated | Pilot reference transmission for a wireless communication system |
| WO1999043168A2 (en) * | 1998-02-20 | 1999-08-26 | Koninklijke Philips Electronics N.V. | Method in a selective call system including a primary station with means for sending a qualifying call from which data can be determined at a secondary station |
| KR100381012B1 (en) | 1998-05-04 | 2003-08-19 | 한국전자통신연구원 | Random connection device for reverse common channel in cdma scheme and method therefor |
| JP2002525913A (en) * | 1998-09-11 | 2002-08-13 | シェアウェーブ・インコーポレーテッド | Method and apparatus for controlling communication in a computer network |
| US6539007B1 (en) * | 1998-11-16 | 2003-03-25 | Samsung Electronics Co., Ltd. | System and method for multiple single-phase access in a TDMA wireless network |
| US6937609B1 (en) * | 1999-05-25 | 2005-08-30 | Cingular Wireless Ii, Inc. | Method for improving efficiency in a time sharing network |
| US8064409B1 (en) | 1999-08-25 | 2011-11-22 | Qualcomm Incorporated | Method and apparatus using a multi-carrier forward link in a wireless communication system |
| US6842435B1 (en) * | 1999-09-15 | 2005-01-11 | General Instrument Corporation | Congestion monitoring and power control for a communication system |
| US6621804B1 (en) | 1999-10-07 | 2003-09-16 | Qualcomm Incorporated | Method and apparatus for predicting favored supplemental channel transmission slots using transmission power measurements of a fundamental channel |
| US7068683B1 (en) | 2000-10-25 | 2006-06-27 | Qualcomm, Incorporated | Method and apparatus for high rate packet data and low delay data transmissions |
| US6973098B1 (en) | 2000-10-25 | 2005-12-06 | Qualcomm, Incorporated | Method and apparatus for determining a data rate in a high rate packet data wireless communications system |
| CN100539526C (en) * | 2003-07-31 | 2009-09-09 | 皇家飞利浦电子股份有限公司 | Method, the access point of bandwidth and airtime fairness are provided in wireless network |
| EP1605727A1 (en) * | 2004-06-09 | 2005-12-14 | Koninklijke Philips Electronics N.V. | Integrated circuit and method for time slot allocation |
| US8284707B2 (en) * | 2006-11-07 | 2012-10-09 | Qualcomm Incorporated | Hybrid power-saving mechanism for VoIP services |
| US9201835B2 (en) * | 2007-02-12 | 2015-12-01 | Mushroom Networks, Inc | Access line bonding and splitting methods and apparatus |
| US8717885B2 (en) * | 2007-04-26 | 2014-05-06 | Mushroom Networks, Inc. | Link aggregation methods and devices |
| CN101828347B (en) * | 2007-08-31 | 2013-07-17 | 汤姆森特许公司 | Method and apparatus for communicating in multiple modes |
| EP2294739B1 (en) * | 2008-06-24 | 2019-08-07 | Mushroom Networks Inc. | Inter-office communication method |
| WO2010019083A1 (en) * | 2008-08-15 | 2010-02-18 | Telefonaktiebolaget L M Ericsson (Publ) | Relative time division for network coding |
| US8811200B2 (en) | 2009-09-22 | 2014-08-19 | Qualcomm Incorporated | Physical layer metrics to support adaptive station-dependent channel state information feedback rate in multi-user communication systems |
| US20150117206A1 (en) * | 2013-10-29 | 2015-04-30 | Qualcomm Incorporated | Backhaul management of a small cell using passive estimation mechanism |
| US9942694B2 (en) * | 2013-12-03 | 2018-04-10 | The University Of Tokyo | Information transmission/reception system and terminal apparatus |
| US9590692B2 (en) * | 2014-07-21 | 2017-03-07 | Cisco Technology, Inc. | Distributed appropriation and defense of time slot ownership in a channel hopping schedule |
| DE112018007489B4 (en) * | 2018-05-21 | 2022-03-31 | Mitsubishi Electric Corporation | COMMUNICATION DEVICE, COMMUNICATION METHOD AND COMMUNICATION PROGRAM |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4400585A (en) * | 1979-11-30 | 1983-08-23 | Motorola, Inc. | Method and apparatus for automatically attempting to seize a radio channel in a multichannel communication system |
| US5121387A (en) * | 1988-08-26 | 1992-06-09 | Scientific Atlanta | Link utilization control mechanism for demand assignment satellite communications network |
| US5124985A (en) * | 1988-12-13 | 1992-06-23 | Small Power Communication Systems Research Laboratories Co., Ltd. | Radiocommunication system using time-division digital frames |
| US5276911A (en) * | 1990-11-02 | 1994-01-04 | Motorola, Inc. | Contention reducing technique for a radio frequency communication system |
| US5278833A (en) * | 1991-03-28 | 1994-01-11 | Motorola, Inc. | Method for providing reserved communication access using multiple random access resources |
| US5142533A (en) * | 1991-03-28 | 1992-08-25 | Motorola, Inc. | Method for controlling the scheduling of multiple access to communication resources |
-
1994
- 1994-07-29 US US08/282,242 patent/US5530700A/en not_active Expired - Lifetime
-
1995
- 1995-05-15 CN CN95190696.8A patent/CN1131482A/en active Pending
- 1995-05-15 WO PCT/US1995/005949 patent/WO1996004720A1/en not_active Application Discontinuation
- 1995-05-15 EP EP95920415A patent/EP0721706A4/en not_active Withdrawn
- 1995-05-15 CA CA002170893A patent/CA2170893C/en not_active Expired - Fee Related
- 1995-05-15 AU AU25873/95A patent/AU685282B2/en not_active Ceased
Also Published As
| Publication number | Publication date |
|---|---|
| EP0721706A1 (en) | 1996-07-17 |
| EP0721706A4 (en) | 1998-07-29 |
| CN1131482A (en) | 1996-09-18 |
| CA2170893A1 (en) | 1996-02-15 |
| AU2587395A (en) | 1996-03-04 |
| AU685282B2 (en) | 1998-01-15 |
| WO1996004720A1 (en) | 1996-02-15 |
| US5530700A (en) | 1996-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2170893C (en) | Method and device for controlling time slot contention to provide fairness between a plurality of types of subscriber units in a communication system | |
| US6529525B1 (en) | Method for supporting acknowledged transport layer protocols in GPRS/edge host application | |
| US5541924A (en) | Method and device for channel contention and data transmission for packet-switched subscriber units in a communication system | |
| US7359705B2 (en) | Random access burst transmission scheme | |
| CA2171015C (en) | Method for providing error correction using selective automatic repeat requests in a packet-switched communication system | |
| CA2125389C (en) | A method of effecting random access in a mobile radio system | |
| TWI387379B (en) | Methods of ranging procedure and bandwidth request ranging procedure | |
| CA2171978C (en) | Method and system for providing uplink/downlink collision avoidance in a wireless communication system | |
| US5734867A (en) | Method, device, microprocessor and microprocessor memory for instantaneous preemption of packet data | |
| US5721735A (en) | Method and apparatus for arbitrating access of plural bidding devices to a central device using a goal index | |
| WO1998037668A1 (en) | Multiple access communication network with dynamic access control | |
| WO1998037669A1 (en) | Multiple access communication network with combined contention and reservation mode access | |
| US6633558B1 (en) | Device and method for controlling channel access by access slot reserving in a mobile communication system | |
| CA2171339C (en) | Method and system for resolution of channel access in data transmission systems | |
| US6795420B1 (en) | Radio communication system | |
| KR100304372B1 (en) | Method for changing slot assignmet/slot size in base stati on and packet data transmission method through reverse common channel in mobile station | |
| AU675038C (en) | Method for providing error correction using selective automatic repeat requests in a packet-switched communication system | |
| PT1574104E (en) | A method of controlling services on a mobile phone network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed | ||
| MKLA | Lapsed |
Effective date: 20030515 |