Distributed channel multicast access method based on service characteristics

Abstract

A distributed channel multicast access method based on service characteristics belongs to the field of wireless communication technology. This method takes the distributed multicast network as the research object, and optimizes the distributed multicast access mechanism and method according to the characteristics of the destination set and transmission time changes caused by the diversity of services in the wireless network. The sources in each multicast group in the network compete for the channel in a distributed manner. The winning source dynamically decides whether to access the channel by identifying the service characteristic information and comparing the multicast access equivalent rate with the fixed threshold, thereby completing the reliable multicast transmission of the source to the destination corresponding to the service. This method realizes distributed access to the wireless multicast network with low signaling overhead and improves the network throughput.

Description

Technical Field

[0001] The invention relates to a distributed channel multicast access method based on service characteristics, belonging to the technical field of wireless communications. Background Art

[0002] With the rapid development of wireless communication technology, the diversity of mobile services and traffic have increased rapidly, and the demand for spectrum resources has increased significantly. As an important way to alleviate spectrum shortage and improve spectrum efficiency, multicast communication technology has developed rapidly. When multiple users in the network request the same data, multicast communication is formed from the source to multiple destination nodes. Compared with unicast communication, by allowing users to share channels for data transmission, the access efficiency of multicast communication channels can be significantly improved. Existing literature has studied the multicast channel access method of centralized networks, mainly focusing on the joint design method of the multicast access physical layer and the access control layer (MAC) layer. Through the multicast access method with multi-user centralized control, the source dynamically adjusts the user transmission power and allocates sub-channels according to the status of each destination channel in the multicast group, which has real-time throughput optimization.

[0003] As an important direction for the development of wireless network technology, distributed networks have the characteristics of node autonomy and strong anti-destruction capabilities, and are widely used in cluster system construction. Compared with centralized networks, distributed networks do not rely on central nodes to uniformly manage user channel access, have low networking overhead, and have high network structure flexibility. However, there is competition among multiple users in distributed networks, which affects channel access efficiency. The high time variability of the wireless link between the source and the destination further exacerbates the efficiency problem. Therefore, it is necessary to jointly design the physical layer and MAC layer of the distributed network and study new channel access methods to achieve efficient and reliable channel access.

[0004] At the same time, in order to meet the different service quality requirements of users for different services, the IEEE802.11e standard adopts an enhanced distributed coordinated access mechanism to further consider the transmission service priority and provide differentiated services for users. Despite this, there are still obvious deficiencies in the research on distributed channel multicast access methods based on service characteristics. The author SANTOS MA studied multicast transmission based on service quality requirements in the article "A Novel QoE-Aware Multicast Mechanism for Video Communications over IEEE802.11WLANs" [IEEE Journal on Selected Areas in Communications, 2012, 30(7): 1205-1214], but it did not consider the impact of distributed access channels of multiple multicast groups on system performance in a network containing multiple multicasts. Summary of the invention

[0005] In order to overcome the defects and shortcomings of the prior art, the present invention proposes a distributed channel multicast access method based on service characteristics to maximize the statistical throughput of a wireless distributed multicast network.

[0006] To achieve the above object, the technical solution of the present invention is as follows:

[0007] A distributed channel multicast access method based on service characteristics is implemented by a multicast communication system: the system includes K multicast groups, and the kth multicast group includes a source node S k and M k Destination Node D k,m , k=1,2,...,K, m=1,2,...,M k , each multicast group shares the channel in a competitive manner to perform multicast transmission based on service characteristics; suppose the wireless channel between each source node and the destination node obeys an independent and identically distributed Rayleigh fading model, and the signal-to-noise ratio h k,m The expected Rayleigh fading variance The exponential random distribution of the wireless channel between each source node and the sink node also obeys the block fading model. The channel state is within the coherence time τ d Considering the diversity of services, the types of services transmitted by multicast are divided into J types, and the transmission time of each service packet is recorded as For all source nodes, the probability of occurrence of each type of service is equal, and the transmission time is less than or equal to the channel coherence time, that is, Assume that the expected value of throughput is λ, and calculate the expected value of the maximum throughput of the system, which is the threshold λ * , is the critical value for the source node to decide whether to perform multicast transmission, and is calculated as follows:

[0008] The iterative calculation formula is:

[0009] λ q+1 =φ(λ q ) (1)

[0010] q=1,2,3,... represents the number of iterations;

[0011]

[0012] Where l represents the number of target sink nodes corresponding to the service packet, is the source node S k The probability of successful competition, the symbol Π represents the cumulative multiplication, i represents the index of the i-th item in the cumulative multiplication process, p i Represents the source node S i The probability of contention for a channel, is the probability that a source node successfully competes in a channel competition process, and I[·] is the indicator

[0013] Function, τ0 is the expected time spent in a single channel contention process, expressed as:

[0014]

[0015] Among them, τ′ RTS Indicates the transmission time of the request to send packet, i.e., the RTS packet, τ CTS Indicates the transmission time of the clear-to-send packet, i.e., the CTS packet. k Represents the source node S k The probability of contention for a channel, It represents the probability of idle state in a single channel contention. Indicates the probability of a single channel contention conflict;

[0016] Assume that the initial value λ1=1, and perform iterative calculation according to formula (1) until it converges to the maximum throughput expected value λ * , the iterative algorithm is updated according to the coefficient q, when |λ q+1 -λ q |<ε, the iteration process ends, where ε>0 is the precision threshold of the numerical iteration algorithm, which is 10 -3 ,

[0017] The specific steps of the distributed channel multicast access method based on service characteristics are as follows:

[0018] 1) Each multicast group senses and accesses the channel in a competitive manner. The source node in each multicast group starts in a micro-slot with a duration of δ and starts with a probability p. k ∈[0,1] sends RTS packets to all sink nodes in the group to independently compete for channels. The source nodes that have not sent RTS packets monitor the channel occupancy in the time slot δ through physical layer energy detection, and then proceed to step 2);

[0019] 2) If the source node that has not sent the RTS packet detects that the channel in the time slot is not occupied, no source node sends the RTS packet, the channel is idle, and all source nodes continue to compete for the channel in the next time slot, and return to step 1); if the source node that has not sent the RTS packet detects that the channel is occupied in the time slot, the occupied time is recorded as τ RTS , there are two situations. After completing the two situations, proceed to step 3):

[0020] (1) If two or more source nodes send RTS packets in the time slot, and the corresponding sink node of the sending source node cannot successfully receive the RTS packet, then there is a channel conflict. In the subsequent time slot, the corresponding sink node does not feed back the CTS packet to the source node.

[0021] (2) There is only one source node S in this time slot k Send RTS packet, send source node to corresponding destination node D k,m Successfully received the RTS packet, through the RTS packet to carry the information of the sink node set D of the service to be sent to determine whether it is the target sink node corresponding to the current service packet, if the sink node D k,m Keep silent; if the sink node D k,m ∈D,D k,m Calculate the relationship between the training sequence and the source node S k The signal-to-noise ratio h k,m , and obtain the maximum achievable rate R k,m =log2(1+h k,m ), sink node D k,m According to the rate R k,m Determine the backoff time And continue to monitor the channel. If the channel remains idle during the backoff time, the destination node D k,m To the source node S k Send a CTS packet, where ∈ indicates belonging to the symbol, It means it does not belong to the symbol;

[0022] 3) If there is a sink node that sends a CTS packet to the source node, the source node will RTS The channel occupancy can be monitored within the time, and the feedback time τ=minτ k,m , and accordingly calculate the achievable rate of the target destination node with the worst channel quality in the multicast group At this point, a channel competition process is completed, and the source node S k The node that wins the channel competition is called the winning source node. The source node that has not sent the RTS packet remains silent and then proceeds to step 4). If the source node is τ after sending the RTS packet RTS If the channel is not occupied within the time period, it is known that a channel conflict occurs, and all source nodes continue to compete in the next time slot, and return to step 1), where min represents the minimum value;

[0023] 4) Winning source node S k According to the worst target destination node achievable rate R in its multicast group worst And the number of destinations corresponding to the current service |D|, calculate the equivalent transmission rate R of the multicast service that can be obtained by multicast transmission worst |D| and compare it with the threshold

[0024] λ * For comparison, if R worst |D|≥λ * , then the winning source node S k At the rate Rworst Perform multicast, and then go to step 5), otherwise if R worst |D|<λ * , winning source node S k Abandoning this transmission opportunity, all multicast group source nodes start a new round of channel competition and return to step 1).

[0025] 5) After the transmission is completed, the target destination node with the worst channel quality feeds back an acknowledgment character, i.e., ACK, to the winning source node. The winning source node confirms the success of the multicast transmission by monitoring the channel occupancy, and a multicast transmission process is completed.

[0026] The RTS is the abbreviation of request-to-send in English, which means request to send in Chinese. It is a commonly used data packet in the channel sensing access protocol.

[0027] The CTS is the abbreviation of clear-to-send in English, which means clear to send in Chinese. It is a commonly used data packet in the channel sensing access protocol and is used by the receiving node to respond to the sending node.

[0028] The ACK is the abbreviation of acknowledge character in English, which means confirmation character in Chinese. It is a transmission control character sent by the receiving node to the sending station in the channel sensing access protocol, indicating that the sent data has been confirmed to be received correctly.

[0029] Compared with the prior art, the present invention has the following significant advantages: 1) a distributed channel access method for multicast networks based on service characteristics is designed to achieve optimal global statistical throughput; 2) channel access is determined by a fixed threshold, with low signaling overhead, simple process, clear logic, and easy implementation; 3) the proposed method takes into account the differences in the quality of service of multiple services, solving the problem that the differences in service characteristics limit the effectiveness of existing networking methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the system structure of the present invention.

[0031] Figure 2 The present invention is a flow chart of a distributed channel multicast access method based on service characteristics. DETAILED DESCRIPTION

[0032] The present invention will be further described below with reference to the accompanying drawings and embodiments, but is not limited thereto.

[0033] Example:

[0034] A distributed channel multicast access method based on service characteristics is implemented by a multicast communication system: Figure 1As shown, the system includes K multicast groups, and the kth multicast group includes a source node S k and M k Destination Node D k,m , k=1,2,...,K, m=1,2,...,M k , each multicast group shares the channel in a competitive manner to perform multicast transmission based on service characteristics; suppose the wireless channel between each source node and the destination node obeys an independent and identically distributed Rayleigh fading model, and the signal-to-noise ratio h k,m The expected Rayleigh fading variance The exponential random distribution of the wireless channel between each source node and the sink node also obeys the block fading model. The channel state is within the coherence time τ d Considering the diversity of services, the types of services transmitted by multicast are divided into J types, and the transmission time of each service packet is recorded as For all source nodes, the probability of occurrence of each type of service is equal, and the transmission time is less than or equal to the channel coherence time, that is, Assume that the expected value of throughput is λ, and calculate the expected value of the maximum throughput of the system, which is the threshold λ * , is the critical value for the source node to decide whether to perform multicast transmission, and is calculated as follows:

[0035] The iterative calculation formula is:

[0036] λ q+1 =φ(λ q ) (1)

[0037] q=1,2,3,... represents the number of iterations;

[0038]

[0039] Where l represents the number of target sink nodes corresponding to the service packet, is the source node S k The probability of successful competition, the symbol Π represents the cumulative multiplication, i represents the index of the i-th item in the cumulative multiplication process, p i Represents the source node S i The probability of contention for a channel, is the probability that a source node successfully competes in a channel competition process, and I[·] is the indicator

[0040] Function, τ0 is the expected time spent in a single channel contention process, expressed as:

[0041]

[0042] Among them, τ′ RTS Indicates the transmission time of the request to send packet, i.e., the RTS packet, τ CTSIndicates the transmission time of the clear-to-send packet, i.e., the CTS packet. k Represents the source node S k The probability of contention for a channel, It represents the probability of idle state in a single channel contention. Indicates the probability of a single channel contention conflict;

[0043] Assume that the initial value λ1=1, and perform iterative calculation according to formula (1) until it converges to the maximum throughput expected value λ * , the iterative algorithm is updated according to the coefficient q, when |λ q+1 -λ q |<ε, the iteration process ends, where ε>0 is the precision threshold of the numerical iteration algorithm, which is 10 -3 ,

[0044] The specific steps of the distributed channel multicast access method based on service characteristics are as follows:

[0045] 1) Each multicast group senses and accesses the channel in a competitive manner. The source node in each multicast group starts in a micro-slot with a duration of δ and starts with a probability p. k ∈[0,1] sends RTS packets to all sink nodes in the group to independently compete for channels. The source nodes that have not sent RTS packets monitor the channel occupancy in the time slot δ through physical layer energy detection, and then proceed to step 2);

[0046] 2) If the source node that has not sent the RTS packet detects that the channel in the time slot is not occupied, no source node sends the RTS packet, the channel is idle, and all source nodes continue to compete for the channel in the next time slot, and return to step 1); if the source node that has not sent the RTS packet detects that the channel is occupied in the time slot, the occupied time is recorded as τ RTS , there are two situations. After completing the two situations, proceed to step 3):

[0047] (1) If two or more source nodes send RTS packets in the time slot, and the corresponding sink node of the sending source node cannot successfully receive the RTS packet, then the channel conflicts. In the subsequent time slot, the corresponding sink node does not feedback to the source node.

[0048] CTS package;

[0049] (2) There is only one source node S in this time slot k Send RTS packet, send source node to corresponding destination node D k,m Successfully received the RTS packet, through the RTS packet to carry the information of the sink node set D of the service to be sent to determine whether it is the target sink node corresponding to the current service packet, if the sink node D k,m Keep silent; if the sink node Dk,m ∈D,D k,m Calculate the relationship between the training sequence and the source node S k The signal-to-noise ratio h k,m , to obtain the maximum achievable rate

[0050] R k,m =log2(1+h k,m ), sink node D k,m According to the rate R k,m Determine the backoff time And continue to monitor the channel. If the channel remains idle during the backoff time, the destination node D k,m To the source node S k Send CTS packet, where ∈

[0051] Indicates that it belongs to the symbol, It means it does not belong to the symbol;

[0052] 3) If there is a sink node that sends a CTS packet to the source node, the source node will RTS The channel occupancy can be monitored within the time, and the feedback time τ=minτ k,m , and accordingly calculate the achievable rate of the target destination node with the worst channel quality in the multicast group At this point, a channel competition process is completed, and the source node S k The node that wins the channel competition is called the winning source node. The source node that has not sent the RTS packet remains silent and then proceeds to step 4). If the source node is τ after sending the RTS packet RTS If the channel is not occupied within the time period, it is known that a channel conflict occurs, and all source nodes continue to compete in the next time slot, and return to step 1), where min represents the minimum value;

[0053] 4) Winning source node S k According to the worst target destination node achievable rate R in its multicast group worst And the number of destinations corresponding to the current service |D|, calculate the equivalent transmission rate R of the multicast service that can be obtained by multicast transmission worst |D| and compare it with the threshold λ * For comparison, if R worst |D|≥λ * , then the winning source node S k At the rate R worst Perform multicast, and then go to step 5), otherwise if R worst |D|<λ * , winning source node S k Abandoning this transmission opportunity, all multicast group source nodes start a new round of channel competition and return to step 1).

[0054] 5) After the transmission is completed, the target destination node with the worst channel quality feeds back an acknowledgment character, i.e., ACK, to the winning source node. The winning source node confirms the success of the multicast transmission by monitoring the channel occupancy, and a multicast transmission process is completed.

Claims

1. A distributed channel multicast access method based on service characteristics, implemented by a multicast communication system: the system includes K multicast groups, and the kth multicast group includes a source node S k and M k Destination Node D k,m , k=1,2,...,K, m=1,2,...,M k , each multicast group shares the channel in a competitive manner to perform multicast transmission based on service characteristics; suppose the wireless channel between each source node and the destination node obeys an independent and identically distributed Rayleigh fading model, and the signal-to-noise ratio h k,m The expected Rayleigh fading variance The exponential random distribution of the wireless channel between each source node and the sink node also obeys the block fading model. The channel state is within the coherence time τ d Considering the diversity of services, the types of services transmitted by multicast are divided into J types, and the transmission time of each service packet is recorded as For all source nodes, the probability of occurrence of each type of service is equal, and the transmission time is less than or equal to the channel coherence time, that is, Assume that the expected value of throughput is λ, and calculate the expected value of the maximum throughput of the system, which is the threshold λ * , is the critical value for the source node to decide whether to perform multicast transmission, and is calculated as follows: The iterative calculation formula is: l q+1 =φ(λ q ) (1) q=1,2,3,... represents the number of iterations; Where l represents the number of target sink nodes corresponding to the service packet, is the source node S k The probability of successful competition, the symbol ∏ represents cumulative multiplication, i represents the index of the i-th item in the cumulative multiplication process, p i Represents the source node S i The probability of contention for a channel, is the probability that a source node successfully competes in a channel competition process, I[·] is the indicator function, and τ0 is the expected time spent in a single channel competition process, which can be expressed as: in, τ′ RTS Indicates the transmission time of the request to send packet, i.e., the RTS packet, τ CTS Indicates the transmission time of the clear-to-send packet, i.e., the CTS packet. k Represents the source node S k The probability of contention for a channel, It represents the probability of idle state in a single channel contention. Indicates the probability of a single channel contention conflict; Assume that the initial value λ1=1, and perform iterative calculation according to formula (1) until it converges to the maximum throughput expected value λ * , the iterative algorithm is updated according to the coefficient q, when |λ q+1 -λ q |<ε, the iteration process ends, where ε>0 is the precision threshold of the numerical iteration algorithm, which is 10 -3 , The specific steps of the distributed channel multicast access method based on service characteristics are as follows: 1) Each multicast group senses and accesses the channel in a competitive manner. The source node in each multicast group starts in a micro-slot with a duration of δ and starts with a probability p. k ∈[0,1] sends RTS packets to all sink nodes in the group to independently compete for channels. The source nodes that have not sent RTS packets monitor the channel occupancy in the time slot δ through physical layer energy detection, and then proceed to step 2); 2) If the source node that has not sent the RTS packet detects that the channel in the time slot is not occupied, no source node sends the RTS packet, the channel is idle, and all source nodes continue to compete for the channel in the next time slot, and return to step 1); if the source node that has not sent the RTS packet detects that the channel is occupied in the time slot, the occupied time is recorded as τ RTS , there are two situations. After completing the two situations, proceed to step 3): (1) If two or more source nodes send RTS packets in the time slot, and the corresponding sink node of the sending source node cannot successfully receive the RTS packet, then there is a channel conflict. In the subsequent time slot, the corresponding sink node does not feed back the CTS packet to the source node. (2) There is only one source node S in this time slot k Send RTS packet, send source node to corresponding destination node D k,m Successfully received the RTS packet, through the RTS packet to carry the information of the sink node set D of the service to be sent to determine whether it is the target sink node corresponding to the current service packet, if the sink node D k,m Keep silent; if the sink node D k,m ∈D,D k,m Calculate the relationship between the training sequence and the source node S k The signal-to-noise ratio h k,m , and obtain the maximum achievable rate R k,m =log2(1+h k,m ), sink node D k,m According to the rate R k,m Determine the backoff time And continue to monitor the channel. If the channel remains idle during the backoff time, the destination node D k,m To the source node S k Send a CTS packet, where ∈ indicates belonging to the symbol, It means it does not belong to the symbol; 3) If there is a sink node that sends a CTS packet to the source node, the source node will RTS The channel occupancy can be monitored within the time, and the feedback time τ=minτ k,m , and accordingly calculate the achievable rate of the target destination node with the worst channel quality in the multicast group At this point, a channel competition process is completed, and the source node S k The node that wins the channel competition is called the winning source node. The source node that has not sent the RTS packet remains silent and then proceeds to step 4). If the source node is τ after sending the RTS packet RTS If the channel is not occupied within the time period, it is known that a channel conflict occurs, and all source nodes continue to compete in the next time slot, and return to step 1), where min represents the minimum value; 4) Winning source node S k According to the worst target destination node achievable rate R in its multicast group worst And the number of destinations corresponding to the current service |D|, calculate the equivalent transmission rate R of the multicast service that can be obtained by multicast transmission worst |D| and compare it with the threshold λ * For comparison, if R worst |D|≥λ * , then the winning source node S k At the rate R worst Perform multicast, and then go to step 5), otherwise if R worst |D|<λ * , winning source node S k Abandoning this transmission opportunity, all multicast group source nodes start a new round of channel competition and return to step 1). 5) After the transmission is completed, the target destination node with the worst channel quality feeds back an acknowledgment character, i.e., ACK, to the winning source node. The winning source node confirms the success of the multicast transmission by monitoring the occupancy of the channel, and a multicast transmission process is completed.

Images