CN116367350A

CN116367350A - Binary countdown medium access control method for wireless scene

Info

Publication number: CN116367350A
Application number: CN202310380823.0A
Authority: CN
Inventors: 许言午; 秦爽; 李晓倩; 冯钢
Original assignee: University of Electronic Science and Technology of China
Current assignee: University of Electronic Science and Technology of China
Priority date: 2023-04-11
Filing date: 2023-04-11
Publication date: 2023-06-30

Abstract

A binary countdown medium access control method for a wireless scene, wherein the competition process comprises the following steps: step 1, in a competition interval, step 101, each node attempting to access generates a k-bit random number; step 102, the node makes corresponding actions in the corresponding contention micro time slot according to each digit of the random number; step 103, generating a mask by the non-sender node; step 104, the node which does not exit the competition is taken as a winner node; step 2, all non-sender nodes send contention feedback information to all winner nodes, and winner nodes with successful contention are taken as winner nodes; and 3, the winner node transmits data. The invention solves the problem of hidden terminals by utilizing the information mastered by the characteristic nodes, and improves the probability of collision-free access.

Description

Binary countdown medium access control method for wireless scene

Technical Field

The invention belongs to the technical field of software, relates to a wireless data transmission technology, and particularly relates to a binary countdown medium access control method for a wireless scene.

Background

A wireless ad hoc network (Wireless Ad hoc Networks, WANET) is an autonomous system of network nodes connected by wireless links. The node relationships in the network are peer-to-peer and wireless communication can take place without the need to establish a fixed infrastructure. In the communication process, the node generates traffic demand and transmits data as a mobile terminal, and physical broadcasting cannot cover all terminals due to the limitation of radio range, and also a multi-hop situation must be considered, that is, the node acts as a router to relay data from other nodes at the same time. The WANET has the advantages of self-establishment, self-organization, self-management and the like, so that the WANET is suitable for being applied to the scenes of military battlefields, disaster search and rescue, environment monitoring, intelligent driving, medical care and the like, and a base station or a fixed infrastructure is not required to be established, so that a wireless network is rapidly deployed.

Because of the high variability of the environment, in WANET, all protocols and coordination functions must be realized in a fully distributed manner, so that the functions of transmission synchronization or collision detection cannot be performed by means of a central controller. Furthermore, due to the distributivity of the environment, collisions between transmissions may occur only on a part of the recipients, which feature further brings about the so-called hidden terminal problem.

The WANET-oriented medium access control (Medium Access Control, MAC) protocol is mainly divided into two types, namely allocation-based and contention-based, in a wireless channel access coordination manner.

The allocation class protocol needs to allocate resources to each node, and the implementation cost of the protocol is generally large. Depending on the allocated resources, it can be classified into FDMA-based, TDMA-based, and the like types.

The allocation type protocol refers to that the divided resources are fixedly allocated to specific nodes, and the division mode generally ensures that the resources used by different nodes are not overlapped. The fixed allocation type protocol has the advantages of ensuring fairness of data transmission of each node and controllable protocol overhead. The disadvantage is that when the network scale is large, the cycle period of the time frame needs to be prolonged, and the network delay is increased; if there are fewer active nodes in the network, excessive idle resources are caused, and channel resources can be fully utilized only when the proportion of active nodes in the network is large enough. And the fixed allocation type protocol has a disadvantage in that it is difficult to implement QoS support and channel utilization is low.

In the contention-type protocol, the node can dynamically contend for resources according to different service requirements and channel occupation conditions. However, when the network load is large, the collision probability generated by the traditional contention-based protocol is greatly increased, the channel access time delay of each node is also greatly increased, and the protocol performance is rapidly reduced. A typical scheme is to use CSMA and RTS/CTS to sense channel conditions and contend for access to the wireless channel, the representative protocol being 802.11p MAC.

The partial concepts of the present invention are explained as follows:

the binary countdown protocol (Binary Countdown Protocols), also known as the master protocol (Dominance Protocols), was originally proposed in the 20 th century, 70, to address wired network medium access conflicts, and has now evolved into an important class of algorithms to address arbitration problems for use in a variety of fields.

Topology-Transparent (Topology-Transparent), meaning that network nodes do not need to master the network Topology. A protocol scheme can operate without the network node having knowledge of the network topology, and is said to be topologically transparent.

The hidden terminal problem (Hidden node problem/Hidden terminal problem) is that in the communication field, a wireless node A sends information to a node B, and a node C does not detect that A also sends information to B, so that A and C send signals to B simultaneously, transmission collision is caused, and finally, the signals sent to B cannot be decoded, and the information transmission of A and C fails.

Time division multiple access (Time division multiple access, TDMA) is a communication technology for implementing a shared transmission medium (typically in the radio domain) or network. It allows multiple users to transmit data in different time slots (time slices) using the same frequency.

"Try-Once-Discard" (TOD), which is a proposed architecture for network control systems, arbitrates Once at a special stage for nodes to complete medium contention access, and then switches to normal transmission, using channels according to the contention result, to transmit more data.

The binary countdown protocol (Binary Countdown Protocols), also known as the master protocol (Dominance Protocols), was originally proposed in the 20 th century 70 to address wired network medium access conflicts, and has now evolved into an important class of algorithms to address arbitration problems. The idea is as follows: competing nodes (almost) simultaneously begin an arbitration process; in the arbitration process, each node transmits a unique bit sequence with fixed length, which represents a priority value; the bit values are encoded as dominant and recessive signals (e.g., high and low levels on the wire), the signal being characterized by the fact that if at least one node transmits a dominant signal, all nodes listening to the medium will observe the dominant signal, otherwise the recessive signal. The competition is performed bit by bit: when a node observes a dominant signal at a bit position where the signal of the node is recessive, stopping the transmission process, namely discarding the competition, otherwise, continuing to transmit the bit sequence; until a node successfully transmits its complete bit sequence, it wins the contention.

One problem to be solved is that wireless network nodes are generally not able to transmit and receive simultaneously, i.e. cannot listen on the medium during transmission. This can be equivalently achieved by transmitting only when the current sequence bit is a dominant signal, and receiving at the recessive signal, noting that interception is not necessary when the current sequence bit of a node is a dominant signal. Another problem is that the competition between several nodes has to start (almost) at the same time, which places high demands on synchronization. In single hop wireless networks, the application of binary countdown is easier, while in multi-hop wireless networks further measures need to be taken due to synchronization challenges and hidden terminal problems.

C. H. Yeh and team work, the first two representative protocols are the broadcast protocol and the CSMA/IC protocol. In the broadcast protocol, a binary back-off protocol is applied prior to transmission of control frames used to schedule data frames on a data channel. The way the broadcast solves the problem of exposing the terminals is to let the nodes negotiate the transmission power of the data frames, while the arbitration of the binary back-off remains unchanged with the single-hop network, at the cost of high control overhead. The carrier sense media access/ID countdown (CSMA/IC) protocol is a binary countdown protocol that includes internal synchronization. The time is structured into one synchronization phase and several "binary contention phases" of the cycle to solve the medium access contention problem. The hidden terminal problem is solved by setting the sensing range to be at least twice the communication range, which requires fine tuning of the transceiver and also exacerbates the exposed terminal problem.

The system and method for binary countdown media access control for multi-hop wireless free networks of patent 200410097719.8 is an implementation of the CSMA/IC protocol described above. In a multi-hop wireless network system, a system and method for performing binary countdown media access control with non-duplicate and unequal length contention codes within two hops. The wireless mobile device performs binary countdown competition according to the competition code to complete medium access, and adjusts the induction range by means of additional power control so as to solve the problem of hidden terminals. Wherein, the acquisition of the bit number of the maximum competition code in the two hops is finished by means of periodical information broadcasting.

In some special application target scenes, such as network building under the condition that the network cannot be preconfigured, for example, a battlefield environment, a plurality of wireless nodes start up in similar time, but the network condition is not known at all, the communication sequence cannot be configured on site, the nodes randomly select time slots to send in a time frame period, the success rate of node sending is very low, the nodes are difficult to be sure that no conflict exists, and in the initial stage of network building, the arbitration is completed quickly under the condition that the information is very deficient, and finally, a conflict-free communication sequence is formed, so that great difficulty exists.

In addition To contention-free pre-allocation, the existing node access method of the infinite ad hoc network mostly performs contention access based on carrier sense multiple access (Carrier Sense Multiple Access, CSMA) and Request To Send/Clear To Send (RTS/CTS) mechanisms, and each contention access may suffer from an indefinite number of transmission collisions. (1) Each collision implies a failure of one access attempt, which in modern wireless ad hoc networks with high speed physical layer transmission capabilities implies a significant loss of bandwidth; (2) with the increase of the number of competing access nodes, the probability of successful access nodes generated by one competition is exponentially reduced, so that the successful access time delay is obviously increased under the scene of higher node density and higher competition intensity, and the service quality is difficult to support.

Disclosure of Invention

In order to overcome the technical defects in the prior art, the invention discloses a binary countdown medium access control method for a wireless scene.

The invention relates to a binary countdown medium access control method for a wireless scene, wherein the time frame period of the wireless scene comprises a plurality of time slots, and the time slots sequentially comprise a competition interval, a hidden terminal elimination interval and a data interval from front to back; the contention interval includes k contention minislots;

the control method comprises a competition process of each time slot in each time frame period, wherein the nodes participated in the competition process are all nodes in a single-hop conflict domain, and the competition process comprises the following steps:

step 1, in the competition interval, the method is carried out,

step 101, each node attempting to access generates a k-bit random number b= { bi }, i= (1, 2 … k);

step 102, the node performs corresponding actions in the corresponding contention micro time slot according to each digit of the random number, specifically:

if bi=1, the node transmits a contention signal CSi for channel contention;

otherwise bi=0, node X listens to the channel;

in the process of monitoring the channel, if other nodes in the channel are monitored to perform channel contention, the nodes abandon the contention in the time slot and exit the contention;

step 103, defining all nodes which exit from contention as non-sender nodes, wherein each non-sender node monitors all contention signals CSI, and when one station is transmitting in a j-th contention micro time slot, the non-sender node generates a mask, wherein the mask is a k-bit character string, the j-th bit is 1, and all other bits are 0;

step 104, the node which does not exit the competition is taken as a winner node to enter step 2;

and 2, step 2.

If only 1 node does not exit the competition after the step 1 is finished, the node directly enters the step 3 as a winner node;

otherwise, go to step 201;

step 201. All non-sender nodes send contention feedback messages to all winner nodes,

the contention feedback information includes a SID field indicating a slot and an HCM field containing a mask generated in step 103;

step 201.

Each winner node performs phase-to-phase calculation on the k-bit random number of each winner node and the mask in each HCM field, and if all the results are non-zero, the contention is considered to be successful;

a winner node with successful contention proceeds to step 3 as a winner node;

step 3, the winner node sends win information FI to other nodes, wherein the FI information comprises time slot state information in a time frame period acquired in the past and time slot information newly occupied by the winner node, and the other nodes update and record the time slot states in the time frame period according to the win information FI;

the winner node transmits data.

Preferably, in the step 201, the time for the non-sender node to send the contention feedback message to all the winner nodes is:

if the node is the node which gives up the contention, the node is sent in a hidden terminal elimination interval;

if the node is the node which performs the node access, the node transmits in the time slot with successful contention.

Preferably, the hidden terminal cancellation interval includes a plurality of cancellation micro time slots, and the node that gives up contention randomly selects one cancellation micro time slot to transmit.

Preferably, the non-sender node further comprises a winner node of other time slots.

Preferably, the contention signal CSi is MAC address information of a node.

Compared with the prior art, the control method provided by the invention has the following technical advantages:

in the competition process, not only the node to be accessed participates in information transfer, but also the node which exits from competition can participate in feedback, so that the problem of hidden terminals is solved by utilizing the information mastered by the characteristic node, the probability of collision-free access is improved, and the access process of the nodes of the whole network is accelerated; under the conditions of any network node density and any network topology, the scheme generates at least one successfully accessed node in the contention access stage on average each contention. I.e. the access delay of the network node is bounded and significantly better than that of a general contention access mechanism, especially in a scene of higher node density.

And secondly, after the competition access of the nodes of the whole network is completed, each node acquires a collision-free transmission channel. In the subsequent conventional transmission, the node does not collide with the transmission on the own channel unless the physical layer channel changes caused by network topology changes or other disasters, is suitable for realizing TOD architecture, can be used for rapidly completing the access of the whole network node in the wireless self-organizing network establishment stage without pre-configuration of unknown network density, and can be switched to the conventional transmission mode to transmit more data after the system is stable.

No additional power control is required to adjust the sensing range and the communication range.

Drawings

Fig. 1 is a schematic diagram of a specific embodiment of a wireless channel according to the present invention divided into a plurality of time frame periods according to a time axis;

FIG. 2 is a schematic side view of one embodiment of a data interval according to the present invention;

FIG. 3 is a schematic diagram of one embodiment of a time slot according to the present invention;

FIG. 4 is a schematic representation of an embodiment of the present invention;

FIG. 5 is a schematic flow chart of step 1 according to the present invention;

FIG. 6 is a schematic diagram showing a specific flow of step 2 according to the present invention;

FIG. 7 is a schematic flow chart of a control method according to the present invention.

Description of the embodiments

The following describes embodiments of the present invention in further detail with reference to fig. 1 to 3.

The invention can be based on a TDMA wireless network protocol, and the wireless channels are divided and organized into a periodic time frame structure according to time axis, each time frame is divided into a certain number of time slots, and the protocol operates according to time with the time frame as the period, as shown in the following figure 1. The time slots are used as the basic unit of radio resource management, and in TDMA architecture, the time slots are synonymous with channels, with a typical value of 8ms for one slot length. A node accesses a collision-free channel, and the node occupies a time slot in a time frame period, and data transmission on the time slot cannot collide with other transmitting nodes.

Time synchronization is not the key point of the scheme, at present, an internal synchronization algorithm tends to be mature, the cost of external synchronization technologies such as time service, high-precision clocks and the like is gradually reduced, and other infrastructures in a plurality of networking environments can be used for auxiliary synchronization, so that the scheme considers that the time synchronization is completed through other means, and the whole network nodes use synchronous time frames.

The center of the scheme is intended to have all nodes successfully access, the standard being "each node acquires at least one time slot within the time frame period, and the node has no collision in the data transmission on its own acquired time slot". In TDMA architecture, when a node successfully accesses a time slot, it will continuously occupy the corresponding time slot in a subsequent time frame period to transmit data, and the node is referred to as the "owner" of the time slot, and also referred to as the node "acquires" the time slot. For example, node C successfully accesses slot No. 5 in time frame period 3, and continues to use slot No. 5 transmission number in subsequent time frame period.

The scheme can normally operate under various conditions, whether a large number of wireless network nodes are simultaneously started to access the network initially or a new node is accessed to a network which has been operated for a period of time. For convenience of description, a generalized case is set: one or more nodes with different numbers simultaneously try to access the network, and the network has the accessed nodes which normally run for a period of time, and the state is accessed; there are also nodes that are attempting to access, the state is "pending access".

When a node is accessed, the time frame period comprises time slots accessed by other nodes and idle time slots, the node receives signaling in the time slots accessed by adjacent nodes, and the instruction comprises information of network control information.

The node selects an idle slot to attempt access. When a node successfully accesses a slot, the slot structure is the same as the general data slot structure, as shown in fig. 2.

And when a node attempts to access an idle slot, it is divided in a "contention access slot" structure, as shown in fig. 3: the contention access slot consists of three intervals: a contention section, a hidden terminal cancellation section, and a data section. The contention interval is used to resolve contention within a hop and consists of a plurality of contention minislots; contention minislots are distinguished from time slots, which are smaller time slices, on the order of tens of microseconds, than time slots.

Each contention mini-slot includes a turnaround time between transmit and receive modes and a time to send a very short message, referred to as a contention signal, denoted CSi, containing the identity ID of the transmitting node, which is the unique identity of each node in the network, without limitation to the manner and length of acquisition, which can be represented, for example, by a 48bit MAC address.

The hidden terminal elimination section is used for solving the hidden terminal problem and consists of a plurality of elimination micro time slots. During this time interval, each obsolete non-sender node will send a hidden terminal clear message during the randomly selected cancellation minislot. Finally, the contender transmits winning information (FI) in the data interval, wherein the winning information comprises state knowledge of each time slot in the time frame period, and if the contender knows that 1 time slot is successfully occupied, the contender comprises n data blocks, and each data block comprises two fields { SID fields: indicating which time slot; HCM field: hidden terminal clear message with its own generated mask }.

In the invention, the time frame period of the wireless scene comprises a plurality of time slots, wherein the time slots sequentially comprise a competition interval, a hidden terminal elimination interval and a data interval from front to back; the contention interval includes k contention minislots;

step 1. In the competition stage, step 1 is carried out in the competition interval,

if bi=1, the node transmits a contention signal CSi for channel contention;

otherwise bi=0, node X listens to the channel;

when a node, for example X, selects an idle slot and attempts contention access, a random number b= { bi }, i= (1, 2 … k) with k bits is generated, where k is the number of contention minislots in a contention interval. For each node, means such as only one competition in a certain period can be limited to slow down the competition.

The action result of the node X in the contention micro time slot i is divided into two cases, if bi=1, the node sends a contention signal CSi; otherwise, node X continues to monitor the channel. If the channel is busy, i.e., it is perceived that one or more other nodes are transmitting on the channel, node X stops further transmissions and relinquishes contention for the time slot, transitioning to a non-sender node.

If no other node is performing channel contention during the channel contention in step 102, the unique outgoing contention signal CSi, although possibly a winner, is not necessarily a winner node

The nodes in the wireless network are in principle unable to know how many nodes contend during the networking phase. For example, node C in the particular embodiment, which is only aware of minislots 4, 5, 7 when listening to itself, does not receive a message throughout the contention phase, and therefore cannot determine that node C must be a winner node.

A single-hop collision domain is defined, i.e. two nodes taken by all nodes in a region, referred to as a single-hop collision domain, are within one-hop transmission range of each other, if each node in the single-hop collision domain generates a different random number bi, only one node survives at the end of the contention interval, step 2 is entered and transmission is continued. A collision may occur if two or more nodes generate one and the same random number, and are the largest of all the generated numbers. A sufficiently large k value can be chosen to maintain a satisfactorily low collision probability and other disaster recovery mechanisms can be redesigned to eliminate the impact.

The non-sender node is defined as including a node in a single-hop conflict domain which is successfully accessed in other time slots and no longer participates in competition and a node which is not successfully accessed and is failed to be converted into a competition receiver in the middle of the competition in the time slot.

A non-sender node Y determines its authorized winner by examining the contention signal received for the first time and successfully decoded. If multiple stations transmit in the same contention mini-slot, Y cannot receive a valid contention signal and then waits for the next contention mini-slot. When one and only one station is transmitting in a contention mini-slot, such as mini-slot i, node Y may receive a valid contention signal CSi. At this time, node Y generates a mask, which is a character string of k bits, the i-th bit being 1, and all other bits being 0.

As in fig. 4, the links between nodes represent link connections. Suppose nodes B, C, H and G wish to contend for a slot in a contention interval. After the first contention mini-slot, node B gives up because it detected the transmission of node C. For the same reason, node H gives up after the third contention mini-slot. At the end of the contention phase, node C and node G survive and are ready to enter step 2.

Wherein each of the C's one-hop neighbors A, B, D, E, H generates a mask that is not written out separately but is not 0 from the result of the C generated random code (11100101). Meanwhile, in the one-hop neighbor node E, K, H of G, K generates a mask 10000000 and a random code 11100100 phase of G to be 1, and H, E generates a mask 00000001 and a random code phase of G to be 0.

A typical process flow for each node during the contention phase of a contention access slot is shown in fig. 5.

Step 2

If only 1 node does not exit the competition after the step 1 is finished, directly entering the step 3.

Otherwise, step 201 is entered.

step 201.

the winner node, which has successfully contended, proceeds to step 3 as the winner node.

After step 1, only one node in a single hop collision domain can survive, unless the same random number is generated and happens to be the largest of all generated numbers, the node that survives after step 1 is referred to as the winner node, and there may be multiple winner nodes. However, the hidden terminal problem specific to multi-hop wireless networks has not been solved. As described above, since the C node and the G node cannot receive the competing signals of each other, they all consider themselves as winners after step 1, and therefore, collision may occur during the data transmission interval. However, their co-neighbors H, E can identify the true winner, i.e., the node with the higher random number.

The processing of this by a non-transmitting node, for example, the neighboring node H, E, is split into two cases.

If the node status is accessed, that is, a slot belonging to the node has been acquired, a contention feedback message is attached to the signaling of the next slot transmission of the node, and the contention feedback message includes two fields { SID field: which slot is indicated, HCM field: hidden terminal clear message with self-generated mask };

if the node state is not accessed, that is, a time slot is not occupied, randomly selecting an elimination micro time slot in an elimination interval of the hidden terminal, and transmitting an HCM field with a mask of the node in the elimination micro time slot. Each winner node compares the random number of each winner node with the contention access time slot and all the HCM mask phases which are received by each time slot of the next time frame period and are related to the contention, and if the result is non-zero, the contention is considered to be successful, and the time slot is acquired; otherwise, the idle time slot is selected again to try to access.

As above, based on the mask information contained in the HCM field, the final computed phase and result of the C node and H, E are not zero, becoming the only winner, while the computed phase and result of the G node are zero, being eliminated. The phase-sum calculation is the result accumulation of multiplying two multi-bit corresponding bits, for example 1101 and 0010 phases and the calculation result is 1×0+1×0+0×1+1×0=0.

Assuming that each hidden terminal cancellation interval includes m cancellation minislots, the process flow of the hidden terminal cancellation phase of a node in one contention access slot is shown in fig. 6.

Step 3, the winner node sends winning information FI to other nodes, wherein the FI information comprises time slot state information in a time frame period acquired in the past and time slot information newly occupied by the winner node, and the other nodes update and record the time slot states in the time frame period according to the winning information FI;

the winner node transmits data.

If all HCM masks and own random number phases received by the winner node in the contention access time slot are non-zero, winner information FI is sent in the time slot data transmission interval, wherein the FI information comprises time slot state information in the time frame period learned in the past and time slot information newly occupied by the winner node. If the neighbor nodes receive the winning information FI and decode correctly, the neighbor nodes update the time frame time slot states in the recording period.

Note that the HCM field received in this time slot may not be all, and that the winning information FI sent by the node in this time slot may collide, since some accessed neighbor nodes will send HCM only in the time slot they occupy.

After going through steps 1 to 3, the winner nodes are very probable to be true winners, for which this data transmission phase can accelerate their access and the maintenance of information of the neighbors; conversely, if a node misbelieves itself to be a winner because the HCM field is not fully acquired, it will only cause the FI to collide at some receiving nodes, and the HCM field transmitted by the subsequent neighbor nodes on their access slots will also eliminate the misunderstanding.

To sum up, the processing flow of the node at any time slot in the time frame period is shown in fig. 7.

The disaster recovery mechanism is established to handle abnormal situations such as the occurrence of an intrusion terminal, channel error code, attack, etc. Then the time slot needs to be re-accessed, and other time slot notices still occupied by the user can be selected; or selecting idle time slots in the time frame period perceived by the user, and starting the contention time slot acquisition process to retry occupation like a newly-accessed network node. As the receiving side: when node X does not receive the signaling of Y in a time slot occupied by a certain neighbor node Y, it may not decode the code, change the node mobile link, etc. due to collision; or receiving the signaling of another node, and setting that the situation occurs for accumulating 2 time frame periods, based on the new state, updating the state information of each time slot in the time frame period held by the node, and announcing the next time slot occupied by the node.

The foregoing description of the preferred embodiments of the present invention is not obvious contradiction or on the premise of a certain preferred embodiment, but all the preferred embodiments can be used in any overlapped combination, and the embodiments and specific parameters in the embodiments are only for clearly describing the invention verification process of the inventor and are not intended to limit the scope of the invention, and the scope of the invention is still subject to the claims, and all equivalent structural changes made by applying the specification and the content of the drawings of the present invention are included in the scope of the invention.

Claims

1. A binary countdown medium access control method for wireless scene is characterized in that,

the time frame period of the wireless scene comprises a plurality of time slots, wherein the time slots sequentially comprise a competition interval, a hidden terminal elimination interval and a data interval from front to back; the contention interval includes k contention minislots;

step 1, in the competition interval, the method is carried out,

if bi=1, the node transmits a contention signal CSi for channel contention;

otherwise bi=0, node X listens to the channel;

and 2, step 2.

otherwise, go to step 201;

step 201.

a winner node with successful contention proceeds to step 3 as a winner node;

the winner node transmits data.

2. The control method as set forth in claim 1, wherein in step 201, the time for the non-transmitting node to transmit the contention feedback message to all the winner nodes is:

3. The control method of claim 2, wherein the hidden terminal cancellation interval includes a plurality of cancellation micro-slots, and the node that gives up contention randomly selects one cancellation micro-slot to transmit.

4. The control method of claim 1 wherein the non-sender node further comprises a winner node of other time slots.

5. The control method of claim 1, wherein the contention signal CSi is MAC address information of a node.