A channel conflict avoidance method for large-scale self-organizing networks based on conflict graph

Abstract

The present invention relates to the field of field area network multi-hop communication technology, specifically, a large-scale self-organizing network channel conflict avoidance method based on conflict graph, which mainly adopts frequency hopping technology and coloring method to allocate channels and avoid channel conflicts. The main steps include: establishing network topology, establishing channel model, calculating interference and signal-to-interference-noise ratio; generating frequency hopping sequence to solve conflicts between nodes of different networks; establishing interference model, generating conflict graph, and establishing channel allocation problem; introducing frequency hopping mode, converting channel allocation problem into time slot allocation problem; establishing mathematical model to solve channel conflicts within the network; establishing optimization problem, converting time slot allocation problem into vertex coloring problem in graph theory; using graph coloring algorithm to solve the problem. This method can reduce wireless channel conflicts, improve data transmission efficiency, and reduce end-to-end transmission delay.

Description

Technical Field

[0001] The invention relates to the technical field of field area network multi-hop communication, and in particular to a large-scale self-organizing network channel conflict avoidance method based on a conflict graph. Background Art

[0002] The multi-mode communication network at the edge of the distribution station area is composed of a central master control node, a routing aggregation node and a terminal perception node. A central master control node and multiple routing aggregation nodes constitute a backbone network centered on the central master control node, and the terminal perception node is wirelessly connected to the backbone network: the terminal perception node device is a low-power device powered by a battery or inductive power supply, and the access point of the terminal perception node can be a central master control node or a routing aggregation node; the central master control node device and the routing aggregation node device are AC or DC powered node devices, and the central master control node and the routing aggregation node use power lines and wireless as transmission. transmission medium, forming a mesh network structure with the central master node as the center; the central master node and routing aggregation node can add corresponding communication modules to expand support for access to Q / GDW12020, Q / GDW12021 wireless sensor devices, as well as wireless sensor devices of standard communications such as BLE, WiFi, ZigBee, LoRa, etc.; the multi-mode communication network is a fusion of power line and wireless transmission media, supporting FHPLC, 470MHz-RF, 2.4GHz-RF and other communication modes, providing a communication network for data transmission and interaction between devices at the edge of the distribution station area.

[0003] The low-voltage power carrier network uses low-voltage power lines as the transmission medium and is composed of a large number of carrier nodes installed on the power lines. It is a multi-hop network system formed by power carrier communication. It supports automatic avoidance of the carrier working frequency band and realizes the equipment status information collection, control, event reporting, power user electricity consumption information aggregation, transmission, and interaction within the distribution station area. At present, communication technology is advancing by leaps and bounds, and communication systems are changing with each passing day. However, the corresponding network channel resources are becoming increasingly scarce. Using power network channels to carry new services is an effective measure to solve the shortage of communication frequency band resources.

[0004] Wireless radio frequency communication combined with power line carrier communication forms a multi-mode communication network, which is very helpful for optimizing network structure and improving network performance. However, when multiple networks are running in a field area network, due to the large number of nodes and complex network topology, many communication links will share the same channel during data transmission, resulting in increased channel conflicts, increased packet loss rate and delay, and reduced throughput. Therefore, it is necessary to design corresponding channel avoidance strategies to reasonably allocate limited channel resources to each link, reduce interference and conflicts between communication nodes, ensure the normal operation of the network, and improve the transmission performance of the network. Summary of the invention

[0005] The present invention aims to provide a large-scale self-organizing network channel conflict avoidance method based on a conflict graph. The method mainly solves the problem that when spectrum resources are limited and multiple networks coexist, channel conflicts and interference will occur during data transmission in the network. It is necessary to design a reasonable resource allocation scheme to avoid channel conflicts, improve data transmission efficiency, and reduce end-to-end transmission delay.

[0006] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is:

[0007] A large-scale self-organizing network channel conflict avoidance method based on conflict graph uses frequency hopping technology and coloring method to allocate channels and avoid channel conflicts. The specific steps are as follows:

[0008] Step S1, establishing a network topology;

[0009] Step S2: Establish a channel model, calculate interference and signal-to-interference-to-noise ratio, analyze the wireless channel model, and regard each routing aggregation node and central master control node that supports wireless transmission as a base station. Use a log-normal shadow path loss model, and the path loss is expressed as:

[0010]

[0011] Where d is the communication distance between two nodes, d 0 is the near-ground reference distance, n is the path loss coefficient, X δ is a random variable with a Gaussian distribution and a mean of 0 and a variance of δ caused by the shadow effect.

[0012] make is a constant, then PL(d)=K L d n .

[0013] Considering small-scale multipath fading as Rayleigh fading, let the transmitting and receiving nodes of a communication link be i and j respectively, and the transmitting power of node i be P i , then the power received by node j is This paper assumes that the antennas are all omnidirectional antennas, and the antenna gain of node i is Channel fading gain It obeys Rice distribution, with noise having mean 0 and variance σ 2 The interference received by the receiving node j in the communication link (i, j) is the cumulative interference I of all neighboring nodes except the sending node i. n for:

[0014]

[0015] The signal to interference plus noise ratio (SINR) received by node j is:

[0016]

[0017] Step S3, resolve conflicts between nodes of different networks. In the field area network, the frequency hopping sequences of other surrounding networks will be referenced during the networking stage, and a frequency hopping sequence orthogonal to the surrounding networks will be generated through a related algorithm. This can ensure that the channels used by different networks are orthogonal at the same time, avoiding channel conflicts and interference with surrounding networks. The specific method of designing the frequency hopping sequence can be adjusted according to the actual network situation. If a single network can support the data transmission of the network using a part of the channels, the available channels can be grouped and allocated to different networks, so that the channels used by the networks are not the same, and there will be no channel conflicts between the networks; if the scale of a single network is relatively large and all channels need to be used to support the data transmission of the network, the frequency hopping sequences between adjacent networks need to be orthogonal, and at the same time, it is necessary to ensure that the frequency sets used by adjacent networks have no intersection within a certain number of time slots, so as to prevent the adjacent network from jumping to the frequency to work after a node whose data transmission has not been completed in a certain network occupies a channel for too long, thereby causing interference between networks;

[0018] Step S4: Establish an interference model. After the network nodes are connected, the node positions will no longer move, and the transmission power of each node is fixed at P. t According to the transmission path given by the routing algorithm, a network topology graph G is established to represent the routing path of data transmission. Graph G is a connected graph. The points in the graph represent the RN nodes in the actual network, and the edges in the graph represent the routing path of data transmission to the next hop node. By analyzing the conflicts existing in the links, a conflict graph G′ is established to represent the conflicts between links. The points in the graph represent a link in the topology graph, and the edges in the graph represent that the links represented by the nodes at both ends of the edge will cause conflicts and need to be allocated to different channels.

[0019] Step S5: Introducing the frequency hopping mode can transform the channel allocation problem into a time slot allocation problem;

[0020] Step S6: Establish a mathematical model to solve channel conflicts within the network;

[0021] Process S6.1: Based on the routing path obtained by the routing algorithm, an adjacency matrix can be generated. The adjacency matrix E: E = {E i,j |E i,j ∈{0,1}} N×N , is a matrix of N rows and N columns, representing the routing relationship between node i and node j.i,j = 0 means there is no route from node i to node j, E i,j =1 means there is a route from node i to node j, where node i is the sending node and node j is the receiving node;

[0022] Process S6.2: Generate interference matrix I based on the conflict graph calculated by the established interference model: I = {I n , k ,|I n,k ,∈{0,1}} N×N , is a matrix with N rows and N columns, indicating the situation where links n and k will interfere when using the same channel at the same time. n,k =1 / 0, indicating that links n and k will / will not interfere when using the same channel at the same time;

[0023] Process S6.3, the result of the solution is the interference-free allocation matrix A: A = {a n,m |a n,m ∈{0,1}} N×M is a matrix with N rows and M columns, which represents a feasible time slot allocation scheme: if time slot m is allocated to link n, then a n,m =1, otherwise a n,m = 0. The interference-free allocation matrix must satisfy the interference constraint:

[0024] Step S7, establish an optimization problem, abstract the above time slot allocation into a vertex coloring problem in graph theory, for a graph G(V,E) and a color set c, where v is the vertex set of graph G, representing the links in the network, c represents the available color set of the vertex, each color represents a time slot, E is the edge set, determined by the interference constraint matrix I, if and only if I n,k = 1, there is an edge between two different vertices n,k∈V. At this time, the coloring condition corresponding to the effective time slot allocation that satisfies the interference constraint can be described as: when there is an edge between two vertices, they cannot be colored the same color at the same time. The coloring process is to assign a color c to each vertex v∈V v ∈C, ensuring that the color is different from the colors of its neighbors, mathematically expressed as: R(G,C)={<v,c> |(v∈V∩c∈C∩( <v i ,c i >∈R∩ <v,v i >∈E→c≠c i ))}. Define the color set C = [c 1 ,c 2 ,…,c n ], is the available color set, all initialized to 0, if it is a vertex v iAssign color c i , then it constitutes a mapping pair <v i ,c i >, at this time c i = 1. The optimization goal is to color all links in the current network using the least number of colors, that is: The minimum number of colors required is c tot And the corresponding coloring scheme;

[0025] Step S8, use the graph coloring algorithm to solve, and use the backtracking method to traverse and search to solve how to use K colors to complete the coloring process. If a solution can be found during the search and backtracking process, the problem is solved; if all situations are traversed, it can be considered that the problem has no solution. Introduce heuristic strategies, which are: in the current state, the vertex with the smallest value range is colored first, and the more colors of neighbor nodes that define a vertex, the smaller the value range of the vertex; when the value range size is the same, compare the degree of the vertex, and the vertex with the largest degree is colored first.

[0026] Beneficial effects of the present invention:

[0027] (1) After the coloring algorithm allocates channels, each link will not transmit data on the same channel as other surrounding links at the same time, which greatly reduces the interference signal in the channel, thereby significantly improving the signal-to-interference-to-noise ratio of the wireless link;

[0028] (2) After the coloring algorithm allocates channels, the throughput of the wireless link is improved and is close to the throughput of the wired link. From a theoretical analysis, this will also make the end-to-end throughput no longer limited by the wireless link, and the transmission performance of the entire network will be greatly improved;

[0029] (3) After coloring algorithm channel allocation, the end-to-end transmission delay is reduced;

[0030] (4) The more interference there is around a link, the higher the degree of interference reduction after the coloring algorithm allocates channels. Therefore, the coloring algorithm channel allocation scheme we proposed is suitable for networks with densely distributed nodes and can effectively reduce the interference during communication between nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a field area network node link relationship diagram provided by the present invention.

[0032] Figure 2 This is the conflict graph establishment process provided by the present invention.

[0033] Figure 3 This is a frequency hopping pattern diagram provided by the present invention.

[0034] Figure 4This is a frequency hopping sequence diagram provided by the present invention.

[0035] Figure 5 This is a schematic diagram of frequency hopping provided by the present invention. DETAILED DESCRIPTION

[0036] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. The embodiments are only used to explain the present invention and do not limit the protection scope of the present invention.

[0037] A large-scale self-organizing network channel conflict avoidance method based on conflict graph, the specific steps are as follows:

[0038] Step S1: Establish network topology. The field network node link relationship diagram of the present invention is as follows: Figure 1 As shown, the main research focuses on two types of nodes in the field area network, the central master node (MN) and the routing aggregation node (RN). The main research on the wireless radio frequency (RF) channel includes the following processes:

[0039] Process S1.1, the field network consists of multiple stations, Figure 1 The figure shows a substation in the field area network, which is a multi-hop network composed of a central master node (MN), several routing aggregation nodes (RN), and several terminal sensing nodes (EN). Each central master node collects data reported by subordinate terminal sensing nodes and transmits it to the central master node through the multi-hop network. The transmission link between RN nodes is divided into two modes: wireless radio frequency (RF) and high-speed power line carrier (HPLC). The HPLC link is divided into ABC three-phase link. This paper mainly studies the backbone network composed of RN nodes and the central master node (MN), as well as the channel conflict problem of wireless RF link.

[0040] Process S1.2: In order to control the cost of nodes, the RN nodes in the network adopt a half-duplex communication mode, that is, each node has only one wireless transceiver module, cannot receive and send at the same time, and can only communicate with one node on one channel in one time slot;

[0041] Step S2: Establish a channel model, calculate interference and signal-to-interference-to-noise ratio, analyze the wireless channel model, and regard each routing aggregation node and central master control node that supports wireless transmission as a base station. Use a log-normal shadow path loss model, and the path loss is expressed as:

[0042]

[0043] Where d is the communication distance between two nodes, d 0 is the near-ground reference distance, n is the path loss coefficient, X δIt is a random variable with a Gaussian distribution with a mean of 0 and a variance of δ caused by the shadow effect;

[0044] make is a constant, then PL(d)=K L d n ;

[0045] Considering small-scale multipath fading as Rayleigh fading, let the transmitting and receiving nodes of a communication link be i and j respectively, and the transmitting power of node i be P i , then the power received by node j is This paper assumes that the antennas are all omnidirectional antennas, and the antenna gain of node i is Channel fading gain It obeys Rice distribution, with noise having mean 0 and variance σ 2 The interference received by the receiving node j in the communication link (i, j) is the cumulative interference I of all neighboring nodes except the sending node i. n for:

[0046]

[0047] The signal to interference plus noise ratio (SINR) received by node j is:

[0048]

[0049] Step S3, generate a frequency hopping sequence to resolve conflicts between nodes of different networks. In the field area network, the frequency hopping sequences of other surrounding networks will be referenced during the networking stage, and a frequency hopping sequence orthogonal to the surrounding networks will be generated through a related algorithm. This can ensure that at the same time, the channels used by different networks are orthogonal, avoiding channel conflicts and interference with surrounding networks. The specific method of designing the frequency hopping sequence can be adjusted according to the actual network situation. If a single network can support the data transmission of the network using a part of the channels, the available channels can be grouped and allocated to different networks, so that the channels used by the networks are not the same, and there will be no channel conflict between the networks; if the scale of a single network is relatively large and all channels need to be used to support the data transmission of the network, the frequency hopping sequences between adjacent networks need to be orthogonal, and at the same time, it is necessary to ensure that within a certain number of time slots, the frequency sets used by adjacent networks have no intersection, so as to prevent the adjacent network from jumping to the frequency to work after a node whose data transmission has not been completed in a certain network occupies a channel for too long, thereby causing interference between networks;

[0050] Step S4: Establish an interference model. After the network nodes are connected, the node positions will no longer move, and the transmission power of each node is fixed at P. t According to the transmission path given by the routing algorithm, a network topology graph G is established to represent the routing path of data transmission. Graph G is a connected graph. The points in the graph represent RN nodes in the actual network, and the edges in the graph represent the routing path of data transmission to the next hop node. By analyzing the conflicts in the links, a conflict graph G′ is established to represent the conflicts between links. The points in the graph represent a link in the topology graph, and the edges in the graph represent that the links represented by the nodes at both ends of the edge will cause conflicts and need to be allocated to different channels.

[0051] The process of establishing the field network conflict graph of the present invention is as follows: Figure 2 As shown, the establishment process is as follows:

[0052] Process S4.1: For each link in G, establish a node in G′.

[0053] Process S4.2: For any link l in G, the transmitting node is t(l) and the receiving node is r(l). If there is another link l′, when the two links send data at the same time, the two links will conflict in the following three situations:

[0054] ① The SINR of the receiving end r(l) or r(l′) is lower than the threshold,

[0055] ② The receiving node of one link is the transmitting node of another link, that is, r(l′)=t(l) or r(l)=t(l′),

[0056] ③ The receiving node of the two links is the same node, that is, r(l) = r(l′),

[0057] Process S4.3: For the two conflicting links in the network topology graph G, establish a connection between the corresponding nodes in the conflict graph G′;

[0058] Step S5, introduce the frequency hopping mode. The field area network adopts the frequency hopping mode for wireless communication. In a time slot, the channels used by the nodes in the entire network are the same and unchanged. After the current time slot ends, the nodes in the entire network will synchronously jump to the channel corresponding to the next time slot. The channel hopping rule is determined by the frequency hopping sequence generated when the network is formed. However, there will be a situation where the data transmission of the receiving and transmitting nodes is not completed in a time slot of the current channel. The scheme adopted is as follows: Figure 3 As shown, the receiving and sending nodes of the data that have not completed transmission in the current time slot do not jump synchronously with other nodes in the network, but continue to occupy the current channel until the data transmission is completed. The frequency hopping diagram is shown in Figure 5 As shown, the nodes in the network are Figure 4The frequency hopping sequence in the network is used for frequency hopping, but the nodes in the network are not completely synchronized. The channels used by the idle nodes in the network will be synchronized according to the frequency hopping sequence. The nodes that are transmitting data will synchronize frequency hopping with the idle nodes in the network after the data transmission is completed. This is determined by the transmission rules set by the nodes. We stipulate that within a data reporting cycle, it is divided into multiple time slots. Each node can only start data transmission at the beginning of each time slot. The channel with synchronous frequency hopping in the current time slot network is used for transmission. If the data transmission is not completed at the end of the current time slot, the transmitting and receiving nodes will not synchronize frequency hopping with the network in the next time slot, but continue to use the channel for data transmission until the data transmission is completed. Based on this frequency hopping mode, the channel allocation problem can be converted into a time slot allocation problem;

[0059] Step S6: Establish a mathematical model to solve channel conflicts within the network;

[0060] Process S6.1: Based on the routing path obtained by the routing algorithm, an adjacency matrix can be generated. The adjacency matrix E: E = {E i,j |E i,j ∈{0,1}} N×N , is a matrix of N rows and N columns, representing the routing relationship between node i and node j. i,j = 0 means there is no route from node i to node j, E i,j =1 means there is a route from node i to node j, where node i is the sending node and node j is the receiving node;

[0061] Process S6.2: Generate interference matrix I based on the conflict graph calculated by the established interference model: I = {I n,k ,|I n,k ,∈{0,1}} N×N , is a matrix with N rows and N columns, indicating the situation where links n and k will interfere when using the same channel at the same time. n,k =1 / 0, indicating that links n and k will / will not interfere when using the same channel at the same time;

[0062] Process S6.3, the result of the solution is the interference-free allocation matrix A: A = {a n,m |a n,m ∈{0,1}} N×M is a matrix with N rows and M columns, which represents a feasible time slot allocation scheme: if time slot m is allocated to link n, then a n,m =1, otherwise a n,m = 0. The interference-free allocation matrix must satisfy the interference constraint:

[0063] Step S7, establish an optimization problem, abstract the above time slot allocation into a vertex coloring problem in graph theory, for a graph G(V,E) and a color set c, where v is the vertex set of graph G, representing the links in the network, c represents the available color set of the vertex, each color represents a time slot, E is the edge set, determined by the interference constraint matrix I, if and only if I n,k = 1, there is an edge between two different vertices n,k∈V. At this time, the coloring condition corresponding to the effective time slot allocation that satisfies the interference constraint can be described as: when there is an edge between two vertices, they cannot be colored the same color at the same time. The coloring process is to assign a color c to each vertex v∈V v ∈C, ensuring that the color is different from the colors of its neighbors, mathematically expressed as: R(G,C)={<v,c> |(v∈V∩c∈C∩( <v i ,c i >∈R∩ <v,v i >∈E→c≠c i ))}. Define the color set C = [c 1 ,c 2 ,…,c n ], is the available color set, all initialized to 0, if it is a vertex v i Assign color c i , then it constitutes a mapping pair <v i ,c i >, at this time c i = 1. The optimization goal is to color all links in the current network using the least number of colors, that is: The minimum number of colors required is c tot And the corresponding coloring scheme;

[0064] Step S8, use the graph coloring algorithm to solve, and use the backtracking method to traverse and search to solve how to use K colors to complete the coloring process. If a solution can be found during the search and backtracking process, the problem is solved; if all situations are traversed, it can be considered that the problem has no solution. Introduce heuristic strategies, which are: in the current state, the vertex with the smallest value range is colored first, and the more colors of neighbor nodes that define a vertex, the smaller the value range of the vertex; when the value range size is the same, compare the degree of the vertex, and the vertex with the largest degree is colored first.

[0065] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A large-scale self-organizing network channel conflict avoidance method based on conflict graph, characterized in that: The specific steps are as follows: Step S1, establishing a network topology; Step S2: Establish a channel model, calculate interference and signal-to-interference-noise ratio, and analyze the wireless channel model; Step S3: Generate a frequency hopping sequence to resolve conflicts between nodes in different networks; Step S4, establishing an interference model; Step S5, introducing a frequency hopping mode; Step S6: Establish a mathematical model to solve channel conflicts within the network; Step S7, establishing an optimization problem; Step S8, using a graph coloring algorithm to solve; In step S2, each routing aggregation node supporting wireless transmission and the central master control node are regarded as a base station, and a log-normal shadow path loss model is adopted, and the path loss is expressed as: Where d is the communication distance between two nodes, d0 is the near-ground reference distance, n is the path loss coefficient, and X δ is a random variable with a Gaussian distribution and a mean of 0 and a variance of δ caused by the shadow effect. make is a constant, then PL(d)=K L d n ; Assume that the transmitting and receiving nodes of a communication link are i and j respectively, and the transmission power of node i is P i , then the power received by node j is Assume that all antennas are omnidirectional antennas, and the antenna gain of node i is Channel fading gain It obeys Rice distribution, with noise having mean 0 and variance σ 2 The interference received by the receiving node j in the communication link (i, j) is the cumulative interference I of all neighboring nodes except the sending node i. n for: The signal-to-interference-to-noise ratio received by node j is: In the step S3, a frequency hopping sequence is generated to resolve conflicts between nodes of different networks, and adjustments are made according to the actual network conditions. If a single network supports data transmission of the network by using a part of the channels, the available channels are grouped and allocated to different networks; if a single network is relatively large and all channels are required to support data transmission of the network, the frequency hopping sequences between adjacent networks need to be orthogonal, and it is also necessary to ensure that within a certain number of time slots, there is no intersection between the frequency sets used by adjacent networks, so as to prevent a node in a network that has not completed data transmission from occupying a channel for too long, and the adjacent network just happens to jump to the frequency to work, thereby causing interference between networks; In step S4, an interference model is established. After the network nodes are connected, the node positions no longer move, and the transmission power of each node is fixed at P t According to the transmission path given by the routing algorithm, a network topology graph G is established to represent the routing path of data transmission. Graph G is a connected graph. The points in the graph represent the RN nodes in the actual network, and the edges in the graph represent the routing path of data transmission to the next hop node. By analyzing the conflicts existing in the links, a conflict graph G′ is established to represent the conflicts between links. The points in the graph represent a link in the topology graph, and the edges in the graph represent that the links represented by the nodes at both ends of the edge will cause conflicts and need to be allocated to different channels. In the step S5, a frequency hopping mode is introduced. The field area network uses the frequency hopping mode for wireless communication. In a time slot, the channels used by the nodes in the entire network are the same and unchanged. After the current time slot ends, the nodes in the entire network will synchronously jump to the channel corresponding to the next time slot. The channel hopping rule is determined by the frequency hopping sequence generated when the network is formed. However, there will be a situation where the data of the transceiver node is not transmitted in a time slot occupied by the current channel. The transceiver node of the data that has not been transmitted in the current time slot does not jump synchronously with other nodes in the network, but continues to occupy the current channel until the data transmission is completed. Based on the frequency hopping mode, the channel allocation problem is converted into a time slot allocation problem; The specific process of step S6 is as follows: Process S6.1: Based on the routing path obtained by the routing algorithm, an adjacency matrix can be generated. The adjacency matrix E: E = {E i,j |E i,j ∈{0,1}} N×N , is a matrix with N rows and N columns, representing the routing relationship between node i and node j, E i,j = 0 means there is no route from node i to node j, E i,j =1 means there is a route from node i to node j, where node i is the sending node and node j is the receiving node; Process S6.2: Generate interference matrix I based on the conflict graph calculated by the established interference model: I = {I n,k ,|I n,k ,∈{0,1}} N×N , is a matrix with N rows and N columns, indicating that when links n and k use the same channel at the same time, interference will occur. n,k =1 / 0, indicating that links n and k will / will not interfere when using the same channel at the same time; Process S6.3, the result of the solution is the interference-free allocation matrix A: A = {a n，m |a n，m ∈{0, 1}} N×M is a matrix with N rows and M columns, which represents a feasible time slot allocation scheme: if time slot m is allocated to link n, then a n，m =1, otherwise a n，m = 0, the interference-free allocation matrix must satisfy the interference constraint:

2. The large-scale self-organizing network channel conflict avoidance method based on conflict graph according to claim 1, characterized in that: The step S7, establishing an optimization problem, abstracting the above time slot allocation into a vertex coloring problem in graph theory, for a graph G (V, E) and a color set c, where v is the vertex set of the graph G, representing the links in the network, c represents the available color set of the vertex, each color represents a time slot, E is the edge set, determined by the interference constraint matrix, if and only if, n，k = 1, there is an edge between two different vertices n, k∈V. At this time, the coloring condition corresponding to the effective time slot allocation that satisfies the interference constraint can be described as: when there is an edge between two vertices, they cannot be colored with the same color at the same time. The coloring process is to assign a color c to each vertex v∈V. v ∈C, ensuring that the color is different from the colors of its neighbors, mathematically expressed as: R(G, C) = {<v,c> |(v∈V∩c∈C∩( <v i , c i >∈R∩ <v,v i >∈E→c≠c i ))}, define the color set C = [c1, c2, ..., c n ], is the available color set, all initialized to 0, if it is a vertex v i Assign color c i , then the mapping pair <v i , c i >, at this time c i =1, the optimization goal is to color all the links in the current network with the least number of colors, that is: The minimum number of colors required is c tot and the corresponding coloring scheme.

3. The large-scale self-organizing network channel conflict avoidance method based on conflict graph according to claim 2, characterized in that: The step S8 adopts a graph coloring algorithm to solve the problem, and uses a backtracking method to traverse the search: if a solution can be found during the search and backtracking process, the problem is solved; if all situations are traversed, it can be considered that the problem has no solution; a heuristic strategy is introduced, which is as follows: in the current state, the vertex with the smallest value range is colored first, and the more colors of neighbor nodes a vertex has, the smaller the value range of the vertex; when the value ranges are the same, the degrees of the vertices are compared, and the vertex with the largest degree is colored first.

Images