High-throughput time-slot scheduling method for ad hoc networks based on distributed MIS

Abstract

Based on the high-throughput time slot scheduling method of the distributed MIS self-organizing network, a network model diagram of the wireless self-organizing network is established, so that the interference between nodes only exists between links that can communicate normally, so that it can be directly judged through directed edges; each node interacts with its neighbors to obtain link information of nodes within its two-hop range, and constructs a local conflict graph of links connected to the node, and the links with conflicts constitute a link interference set; the distributed MIS algorithm is used to solve the maximum independent link set according to the local conflict graph; the number of MIS algorithm runs is the time slot scheduling scheme, that is, the number of time slots per frame in the end. The present invention can further improve the reuse of time slot resources and solve the problem of time slot resource waste generated when the prior art performs time slot scheduling based on the graph coloring algorithm in a distributed network.

Description

Technical Field

[0001] The invention belongs to the field of communication technology, and particularly relates to a time slot scheduling method for maximizing throughput of a self-organizing network based on distributed MIS, which is a high-throughput time slot scheduling method and can be used for time slot scheduling of a distributed self-organizing network. Background Art

[0002] In addition to realizing synchronization and network access, the most important thing for the MAC layer protocol of wireless ad hoc networks (which provides access control functions for shared media) is to complete data transmission between nodes. In a communication network based on time division multiple access, the time slot scheduling algorithm is a key issue that needs to be studied in depth.

[0003] In the existing time slot scheduling method, the time slot scheduling algorithm is combined with graph theory by introducing relevant interference models and graph theory knowledge, and the communication interference problem is transformed into a graph coloring problem. When the graph coloring algorithm is used for time slot scheduling in the existing method, it is ensured that the link is only scheduled once, which will cause the problem of time slot resource waste.

[0004] In his paper "Resource Scheduling of Multi-beam Wireless Ad Hoc Network System Based on Coloring" (Computer Simulation, 2019, 36(02): 462-468), Li Yi Ze applied graph coloring technology to the resource scheduling of multi-beam wireless ad hoc network system, transformed the resource scheduling problem into an edge coloring problem, and abstracted the high-load links in the network into multiple virtual links, assigning different colors to the conflicting links. However, the disadvantage is that when scheduling time slots, the same link in the network occupies multiple time slots in a frame to realize multiple calls of the link, which does not effectively improve the reuse rate of time slots.

[0005] That is, the application of the coloring method to the time slot scheduling problem will result in the problem of time slot waste, and compared with the centralized method, the information obtained by the nodes in the distributed method is more limited. Summary of the invention

[0006] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a high-throughput time slot scheduling method for self-organizing networks based on distributed MIS, to improve on the distributed MIS algorithm, to propose a distributed time slot scheduling algorithm, to further improve the multiplexing of time slot resources, so as to solve the problem of time slot resource waste caused by the prior art based on graph coloring algorithm for time slot scheduling in distributed networks.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] The high-throughput time slot scheduling method based on distributed MIS self-organizing network includes the following steps:

[0009] S1: Establish a network model diagram of the wireless ad hoc network so that the interference between nodes only exists between links that can communicate normally, so that it can be directly judged through directed edges;

[0010] S2: Each node interacts with its neighbors to obtain link information of nodes within its two-hop range, and constructs a local conflict graph of links connected to the node. Links with conflicts constitute a link interference set.

[0011] S3: Use the distributed MIS algorithm to solve the maximum independent link set based on the local conflict graph;

[0012] S4: After S3, the number of MIS algorithm runs is the time slot scheduling scheme, that is, the final number of time slots per frame.

[0013] Compared with the prior art, the present invention has the following beneficial effects:

[0014] The present invention compares the time slot reuse rate with the algorithm based on MG-MIS (Maximum Independent Set Algorithm proposed by Misra, Gandham). As the number of nodes increases, the reuse rate of time slot resources of various algorithms continues to increase. When based on the same type of antennas, compared with the MG-MIS-based algorithm in which the link can only be scheduled once, the improved distributed time slot scheduling algorithm of the present invention has a significant improvement in the reuse rate of time slot resources. When the algorithm type is the same, the use of directional antennas can also achieve better results in time slot reuse rate than the use of omnidirectional antennas. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the overall technical route of the present invention.

[0016] Figure 2 is the information scope for constructing the local conflict graph.

[0017] Figure 3 It is a flow chart of the distributed time slot scheduling algorithm of the present invention.

[0018] Figure 4 It is the network model diagram of the present invention.

[0019] Figure 5 It is a comparison chart of time slot resource reuse rate. DETAILED DESCRIPTION

[0020] The embodiments of the present invention are described in detail below with reference to the accompanying drawings and examples.

[0021] As mentioned above, the distributed implementation of time slot scheduling can use a distributed graph coloring algorithm, and the distributed graph coloring algorithm is often based on a distributed MIS algorithm. For the graph coloring algorithm, the result of the distributed implementation is also that the link is only colored once, which will cause the problem of waste of time slot resources. At the same time, how to achieve efficient and conflict-free communication has become one of the most important goals of the time slot scheduling algorithm, and its primary problem is to solve the communication conflict problem. The present invention introduces relevant interference models and graph theory knowledge, introduces how to combine the time slot scheduling algorithm with graph theory, transforms the communication interference problem into a graph coloring problem, and introduces the use of graph theory technology to solve the distributed time slot scheduling problem of larger network scales.

[0022] The main idea of ​​the present invention is: first, compare with all the scenarios that the centralized method can deal with, and analyze the differences in link information and scheduling issues from the centralized algorithm, that is, only local information can be obtained to build a local conflict graph, and at the same time, analyze how the links in adjacent areas negotiate to avoid conflicts. On this basis, the MIS algorithm is improved, and the point coloring algorithm is applied to the local conflict graph. Then, the link interference number is used as a method to initialize the link weight, which simplifies the processing flow when the weights are equal. By adding a repeatable state to the link and adding a repeated link selection link after each round of MIS calculation, multiple scheduling of the link in the same time slot in the distributed system is achieved.

[0023] Reference Figure 1 The specific operation steps of this embodiment are as follows:

[0024] Step 1: Establish the protocol interference model and create the network model diagram of the wireless self-organizing network.

[0025] The protocol interference model considers the communication situation within a limited range of nodes, ignores the interference superposition between non-adjacent nodes, and simplifies the network model. In the protocol interference model, both the communication range and the interference range are divided in the form of circles. Assume that the effective transmission radius is R c , the interference radius is R i , usually R i >R c , so node v i and node v j The conditions for successful transmission between ||v i -v j || <R c .

[0026] In the protocol interference model, it is mainly divided into primary interference and secondary interference. Primary interference includes three situations: a node cannot receive data from two other nodes at the same time, a node cannot send data to two targets at the same time, and a node cannot receive and send data at the same time. Secondary interference refers to the interference of nodes with a Euclidean distance less than R when transmitting data normally. iThe node of the primary interference causes interference, resulting in the node being unable to receive data correctly. Because both primary interference and secondary interference cannot communicate normally, they are not distinguished here and are both represented by solid undirected edges.

[0027] The interference model used in the present invention is a protocol interference model, and it is assumed that R i =R c At this time, there will be no mutual interference but the communication conditions will not be met. The interference between nodes only exists between links that can communicate normally, which can be directly judged through directed edges.

[0028] The network model graph is represented as a directed graph G, G = (V, E), where V is the vertex set, V = {v1, v2, ..., v N},v1,v2,…,v N is a node, N is the number of nodes, E is the edge set, E={e1,e2,…,e L},e1,e2,…,e L is a link, L is the number of links, when node v i The data sent can be sent by node v j Normal reception, then it is considered that v i With v j A communication link can be formed between them, recorded as (i, j); the link (i, j) is numbered by the function h(i, j) and corresponds to the edge in the edge set E, that is: (i, j) = e h(i,j) .

[0029] Through this network model diagram, the interference between nodes only exists between links that can communicate normally, so it can be directly judged through directed edges.

[0030] Step 2: Construct a local conflict graph.

[0031] Compared with the centralized type, the main problem of the distributed type is that the whole network topology cannot be obtained, and only local link information can be obtained. Therefore, when constructing the conflict graph, the information obtained is relatively limited. In order to build the conflict graph, at least the link status of the nodes within the two-hop range must be known. Therefore, before the scheduling algorithm of the present invention starts, each node needs to interact with its neighboring nodes for data, obtain the link information of the nodes within the two-hop range, that is, the link receiving and sending information of the nodes within 2 hops, and on this basis, construct a local conflict graph of the link connected to the node for receiving and sending, and the conflicting links constitute the link interference set. The distributed type cannot directly determine the receiving and sending of each link by the central control node like the centralized type. Each node can only decide whether the link connected to itself is sent, so it is necessary to constantly negotiate with the neighboring nodes.

[0032] Specifically, each node converts the network model graph into a local conflict graph G according to the information sent and received by the link within 2 hops.c =(V c ,E c ). Among them, V c is the vertex set of all links, each vertex represents a communicable link, which is obtained by mapping the edge set E in the network model graph G. Each vertex corresponds to a directed edge in the network model graph G. Then, according to the protocol interference model, the interference set of each link in the network model graph G is obtained after data interaction, corresponding to the local conflict graph G. c In the example, the relationship between links is through the edge set E c That is, if there is interference between links, undirected edges are used to connect them, and all undirected edges constitute the edge set E c , edge set E c Each edge in represents a conflict.

[0033] like Figure 2 , when node 4 wants to build a local conflict graph, it must know the link status of nodes 2, 3, 5, 6 and itself. For two adjacent nodes, it is assumed that both nodes know the link status between them, and the transmission and reception of the link should be jointly determined by them. Therefore, node 4 does not need to communicate with node 1, but after exchanging information with node 2, it knows the link e 12 and link e 21 Similarly, we also know that link e 67 and link e 76 Therefore, node 4 can completely construct link e 34 、e 43 、e 45 and e 54 Conflict diagram.

[0034] The distributed implementation of time slot scheduling can use the distributed graph coloring algorithm, which is often based on the distributed MIS algorithm. For the graph coloring algorithm, the distributed implementation result is also that the link is only colored once, which will also produce the same situation as the centralized one, that is, there is a waste of time slot resources. In the distributed system, global information cannot be obtained, and nodes can only construct local conflict graphs, such as Figure 2 In the example, node 4 can know that link e 34 With link e 32 Conflict, but don't know link e 32 Which links conflict?

[0035] The present invention Figure 2 In the embodiment shown, the local conflict graph G c Vertex Set V c For {e 12 ,e 23 ,e 43 ,e 45 ,e 65 ,e72 ,e 83 ,e 84}, the interfering links of each link are represented by their interference set, e.g. 23 The interference set can be seen in Table 1 as {e 12 , e 43 , e 72 , e 83}, the edge set E in the local conflict graph c Represents the interference relationship between links. The links with interference are connected by solid undirected edges. All solid undirected edges are the edge set E c .

[0036] Step 3: Use the distributed MIS algorithm to solve the maximum independent link set based on the local conflict graph.

[0037] The present invention adopts the SW-MIS algorithm. The nodes in the SW-MIS algorithm and the links in the network model graph G, namely the local conflict graph G c The neighbor nodes in the algorithm correspond to the conflicting links in the network model graph G. In the SW-MIS algorithm, when a link has been executed by the MIS algorithm and has been added to the maximum link independent set, the link is made in a repeated available state and is not deleted. When a new round of the MIS algorithm ends and the maximum link independent set is calculated, a repeated link selection link is added. The maximum link independent set obtained by the i-th round of the algorithm is defined as MIS. i , the set of repeated links obtained in round i is RS i .

[0038] like Figure 3 As shown, the specific steps of the SW-MIS algorithm of the present invention are:

[0039] (1) Initialize link status

[0040] The node initializes the links connected to itself, initializes the states of all links connected to itself to the competitive state, initializes the weight r of each link to the size of the link interference set to which it belongs, and initializes the round MIS-round of the MIS algorithm and the round RS-round of repeated link selection to 0, representing the period of time slot scheduling.

[0041] (2) Nodes exchange information to obtain the maximum independent link set.

[0042] (2.1) In the i-th round of the algorithm, each node interacts with the nodes within its two-hop range, compares the weight r of the link in the competitive state and the link in the interference link set, and the link with the larger weight r is added to the MIS first. i ; If the weights r are equal, compare the link numbers e h(ij), the smaller one joins MIS first i , joined MIS i The link state is transformed into the dominant state. After the i-th round of algorithm is completed, MIS i The link is converted into a repeatable selection state.

[0043] For example, the weight r of a link is represented by the interference set size of the link, for example, link e 12 The interference set is {e 23 , e 72}, so its weight r is 2, link e 23 The interference set is {e 12 , e 43 , e 72 , e 83}, so its weight r is 4. And when a link joins MIS i After that, in this round of MIS algorithm, this link will no longer compete with other links, so it will no longer appear in the interference set of other links. 23 After adding the first round of MIS1, in the subsequent MIS algorithm steps of this round, e 83 The interference set becomes {e 43 , e 84}, with a weight of 2.

[0044] (2.2) The link in the dominant state determines whether there are other links in the dominant state in its interference link set. If so, all the links in the dominant state in the interference link set are changed to the dominated state.

[0045] (2.3) Determine whether the link and the links in its interference link set are both in the dominant state or the dominated state. If not, jump to step (2); if so, end this round of the MIS algorithm, and MIS-round is increased by 1;

[0046] (3) Determine whether to proceed to the next round of the MIS algorithm based on the node link status. If so, go to (2). If not, the algorithm ends and the links in the same maximum link independent set are dyed the same color.

[0047] (3.1) Except for the first round of MIS algorithm, each subsequent round of algorithm will repeat the scheduling of the links that have been converted to the repeatable selection state at the end of the previous round of algorithm. The links in the repeatable selection state will be judged whether they are related to the MIS in this round. i If there is a conflict between the links in the interference set, if there is no conflict, the PR is compared with the links in the repeatable selection state in the interference set (add MIS i The smaller the value, the RS i , and update PR to the current MIS-round; if there is a conflict, give up joining the RS round i; RS-round plus 1; The PR value is that after each round of MIS, a round of interaction is required for the nodes that attempt to join repeatedly, and a new attribute is added to each node as a basis for judging whether to join MIS in this round. The attribute is the round-PR value of the node's last scheduling;

[0048] In the present invention, if a link in a repeatable selection state is i The interference set of all links in the MIS is calculated. If this link exists in these interference sets, there is a conflict. If this link does not exist, there is no conflict. In this embodiment of the present invention, the PR value is selected by adding MIS i The round of collection.

[0049] For example, Figure 4 In the example shown, when the fourth round of the MIS algorithm is performed, the link set RS4 can be repeatedly selected as {e 12 ,e 23 ,e 43 ,e 45 ,e 65 ,e 72 ,e 83}, the only link added to the MIS4 set in this round is e 84 , RS4 and e 84 The conflict-free link is {e 12 , e 23 , e 65 , e 72}; where e 23 It is MIS1 added in the first round of MIS algorithm, so its PR value is 1, which is the smallest PR value in RS4. 23 Prioritize adding MIS4, and then remove the same as e in RS4 23 Mutually interfering links 12 With e 72 , the remaining link is e 65 ; The remaining RS4 links 65 In the second round, MIS2 was added, with a PR value of 2, and 23 There is no conflict, so MIS4 is added, and finally MIS4 is {e 83 , e 23 , e 45}.

[0050] (3.2) Determine whether the status of all links in the node is the dominant state or the repeatable selection state. If not, the dominant state is converted to the repeatable selection state, and the dominated state is converted to the competition state, and then go to step (2) to perform the next round of MIS algorithm; if yes, the link of the node does not need to perform the next round of MIS algorithm, and the algorithm ends. The same maximum link independent set is dyed with the same color and divided into one time slot;

[0051] (3.3) The algorithm ends, and the final output MIS-round is equal to RS-round, which means the number of time slots per frame M. The total number of MIS sets i and RSi represent the allocated link combination, and the time slot allocation ends.

[0052] Figure 3 The parameters and other terms in the algorithm shown are explained as follows:

[0053] (1) competitor: initial state, indicating that each node is initially in a state of competing to join the MIS;

[0054] (2) dominator: dominant state, indicating that the node is in a dominant position and can join the MIS;

[0055] (3) dominated: the dominated state means that it is temporarily unable to join the MIS. The node in this state is usually a neighboring node in the dominant state.

[0056] (4) Ruler: indicates that the weights of the current node and its neighboring nodes are equal and in a balanced state;

[0057] (5) ruled: The neighboring node of the node in the ruler state.

[0058] (6) unordered_map <int,unordered_set <int>>I: This data structure is mainly used to record all corresponding link interference sets of the edges connected to the node;

[0059] (7)S l : Indicates the state of link l. Different states can be represented by different numbers;

[0060] (8)r l : represents the weight of link l;

[0061] (9)PR l : indicates the last scheduling round of link l;

[0062] (10)MIS-round, RS-round: represent the round of the link MIS algorithm and the round of the repeated link selection phase, respectively;

[0063] Step 4: Calculate the number of time slots and the time slot reuse rate

[0064] The number of time slot scheduling schemes obtained in the previous step is the MIS algorithm round MIS-round, which is the final number of time slots per frame. The maximum independent link set MIS i is the set of links transmitted in the i-th time slot. Assume that after the time slot allocation is completed, each frame is divided into M time slots. When analyzing the reuse rate of time slot resources, the number of links scheduled in each time slot can be calculated, that is, the reuse rate of a single time slot. The reuse rate of the M-th time slot is defined as:

[0065]

[0066] Where S ml For the conflict graph G c Edge coloring is performed, and each edge is colored multiple times, so that the link can communicate multiple times in one frame, that is:

[0067]

[0068] in Indicates link e l Whether to schedule in the Mth time slot, that is:

[0069]

[0070] The time slot resource reuse rate of the entire network is:

[0071]

[0072] Figure 4 This is an explanation of the distributed time slot scheduling algorithm. The figure shows a network model diagram based on directional antennas, in which there are 8 nodes in the figure, and the direction of the directed edge represents the propagation direction of the unicast service. First, each node exchanges information and builds its own interference set. The interference links corresponding to each link are shown in Table 1.

[0073] Table 1 Figure 4 Interference links of each link in

[0074] link Interference Link <![CDATA[e 12 ]]> <![CDATA[e 23 ,and 72 ]]> <![CDATA[e 23 ]]> <![CDATA[e 12 ,And 43 ,And 72 ,And 83 ]]> <![CDATA[e 43 ]]> <![CDATA[e 23 ,And 45 ,And 83 ,And 84 ]]> <![CDATA[e 45 ]]> <![CDATA[e 43 ,And 65 ,And 84 ]]> <![CDATA[e 65 ]]> <![CDATA[e 45 ]]> <![CDATA[e 72 ]]> <![CDATA[e 12 ,and 23 ]]> <![CDATA[e 83 ]]> <![CDATA[e 23 ,And 43 ,And 84 ]]> <![CDATA[e 84 ]]> <![CDATA[e 43 ,And 45 ,And 83 ]]>

[0075] Initially, each link is in competitor state, and the weight corresponding to each link is the number of interfering links. In the first round, the weight of link e23 is equal to that of the adjacent link, but the link number is smaller, so it is preferentially added to MIS1. Link e45 can also join MIS1 through the comparison in the second round of this round. At this time, the status of the two links joining MIS is changed to repeatable. In the second round, link e43, link e65 and link e12 join MIS2, and the links in repeatable state cannot join MIS2. In the third round, link e72 and link e83 join MIS3, and links e45 and e65 in repeatable state can join MIS3 of this round. In the fourth round, link e84 joins MIS4, and links e23 and link e45 in repeatable state can join MIS4 of this round. At this time, the status of all links is dominator or repeatable, and the algorithm terminates. The time slot reuse rate ρ obtained at this time is 2.75.

[0076] The improved distributed time slot scheduling algorithm of the present invention is compared with the algorithm based on MG-MIS in terms of time slot reuse rate. Figure 5 It can be seen that as the number of nodes increases, the reuse rate of time slot resources of various algorithms continues to increase. When based on the same type of antennas, compared with the MG-MIS-based algorithm where the link can only be scheduled once, the improved distributed time slot scheduling algorithm in this chapter has a significant improvement in the reuse rate of time slot resources. When the algorithm type is the same, using directional antennas can also achieve better results in time slot reuse rate than using omnidirectional antennas.< / int>

Claims

1. A high throughput time slot scheduling method based on a distributed MIS ad hoc network, characterized in that: The steps include: S1: Establish a network model diagram of the wireless ad hoc network so that the interference between nodes only exists between links that can communicate normally, so that it can be directly judged through directed edges; S2: Each node interacts with its neighbors to obtain link information of nodes within its two-hop range, and constructs a local conflict graph of links connected to the node. Links with conflicts constitute a link interference set. S3: Use the distributed MIS algorithm to solve the maximum independent link set based on the local conflict graph; S4: The number of MIS algorithm runs after S3 is the time slot scheduling scheme, that is, the final number of time slots per frame; S1 adopts the protocol interference model and sets its interference radius R i Equal to the effective transmission radius R c The network model graph is represented as a directed graph G, G = (V, E), where V is the vertex set, V = {v1, v2, ..., v N },v1,v2,…,v N is a node, N is the number of nodes, E is the edge set, E={e1,e2,…,e L },e1,e2,…,e L is a link, L is the number of links, when node v i The data sent can be sent by node v j Normal reception, then it is considered that v i With v j A communication link can be formed between them, denoted as (i, j); Link (i, j) is numbered by function h(i, j) and corresponds to the edge in edge set E, that is: (i, j) = e h(i,j) ; In S2, data is exchanged between nodes, each node obtains link transmission and reception information of nodes within 2 hops, and each node constructs a local conflict graph for all links that transmit and receive with itself, where conflicting links are represented by interference sets, i.e., link interference sets; Each node converts the network model graph into a local conflict graph G according to the information received and sent from the link within 2 hops. c =(V c ,E c ), where V c is the vertex set of all links, which is obtained by mapping the edge set E in the network model graph G. Each vertex corresponds to a directed edge in the network model graph G. Then, according to the protocol interference model, the interference set of each link in the network model graph G is obtained after data interaction, corresponding to the local conflict graph G. c In the example, the relationship between links is through the edge set E c That is, if there is interference between links, undirected edges are used to connect them, and all undirected edges constitute the edge set E c , edge set E c Each edge in represents a conflict; The distributed MIS algorithm in S3 adopts the SW-MIS algorithm. The nodes in the SW-MIS algorithm and the links in the network model graph G are the local conflict graph G. c The vertices correspond to the neighbor nodes in the algorithm, and the conflicting links in the network model graph G; After a link has executed the MIS algorithm and has been added to the maximum link independent set, the link is placed in a duplicate available state and is not deleted. When a new round of the MIS algorithm ends and the maximum link independent set is calculated, a duplicate link selection step is added. The maximum link independent set obtained by the i-th round of the algorithm is defined as MIS. i , the set of repeated links obtained in round i is RS i ; The specific steps of S3 are: (1) The node initializes the links connected to itself, initializes the states of all links connected to itself to the contention state, initializes the weight r of each link to the size of the link interference set to which it belongs, and initializes the round MIS-round of the MIS algorithm and the round RS-round of repeated link selection to 0, representing the period of time slot scheduling; (2) Nodes exchange information to obtain the maximum independent link set. The method is as follows: (2.1) In the i-th round of the algorithm, each node interacts with the nodes within its two-hop range, compares the weight r of the link in the competitive state and the link in the interference link set, and the link with the larger weight r is added to the MIS first. i ; If the weights r are equal, compare the link numbers e h(ij) , the smaller one joins MIS first i , joined MIS i The link state is transformed into the dominant state. After the i-th round of algorithm is completed, MIS i The link is transformed into a repeatable selection state; (2.2) The dominant link determines whether there are other dominant links in its interfering link set. If so, the dominant link in the interfering link set changes to the dominated state. (2.3) Determine whether the link and its interference set are both in the dominant state or the dominated state. If not, jump to step (2); if so, end this round of the MIS algorithm, and MIS-round increases by 1; (3) Determine whether to proceed to the next round of MIS algorithm based on the node link status. If so, go to (2). If not, the algorithm ends and the links in the same maximum link independent set are dyed the same color. The method is as follows: (3.1) Except for the first round of MIS algorithm, each subsequent round of algorithm will repeat the scheduling of the links that have been converted to the repeatable selection state at the end of the previous round of algorithm. The links in the repeatable selection state will be judged whether they are related to the MIS in this round. i If there is a conflict between the links in the interference set, if there is no conflict, the PR value is compared with the links in the repeatable selection state in the interference set, and the smaller one is added to RS i , and update PR to the current MIS-round; if there is a conflict, give up joining the RS round i ; RS-round plus 1; The PR value is that after each round of MIS, a round of interaction is required for the nodes that attempt to join repeatedly, and a new attribute is added to each node as a basis for judging whether to join MIS in this round. The attribute is the round-PR value of the node's last scheduling; (3.2) Determine whether the status of all links in the node is the dominant state or the repeatable selection state. If not, the dominant state is converted to the repeatable selection state, and the dominated state is converted to the competition state, and then go to step (2) to perform the next round of MIS algorithm; if yes, the link of the node does not need to perform the next round of MIS algorithm, and the algorithm ends. The same maximum link independent set is dyed with the same color and divided into one time slot; (3.3) The algorithm ends, and the final output MIS-round is equal to RS-round, which means the number of time slots per frame M. The total number of MIS sets i and RSi represent the allocated link combination, and the time slot allocation ends.

2. According to claim 1, the high-throughput time slot scheduling method based on distributed MIS ad hoc network is characterized in that: The time slot allocation scheme in S4 is as follows: the MIS algorithm round MIS-round is the number of time slots per frame, and the maximum independent link set MIS i is the set of links transmitted in the i-th time slot.

Images