Self-adaptive resource allocation method based on topology semitransparent
By using topological translucent algorithm to dynamically allocate control signaling resources in mobile ad hoc networks and using topological non-transparent algorithms to perform spatial reuse of data business resources, the problems of low resource utilization and poor adaptability to topological changes are solved, and high efficiency is achieved resource management and network capacity improvement.
Patent Information
- Application Number
- CN202411047825.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-08
AI Technical Summary
In the existing mobile ad hoc network, the utilization rate of static allocation of control signaling resources is low and cannot adapt to network topology changes. The dynamic allocation of data service resources cannot achieve the global optimal spatial multiplexing rate using local topology information.
Adaptive resource allocation method based on topology translucency is adopted to divide system resources from time domain and frequency domain into time frequency two-dimensional resources, dynamically allocate control signaling resources through topology transparent algorithm, and use topology non-transparent based on network topology information. The algorithm performs spatial multiplexing of data business resources.
It improves resource utilization, improves network capacity, can adapt to topological changes, reduces resource acquisition delay and control overhead, and avoids resource allocation conflicts.
Smart Images

Figure CN118921745A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of resource allocation in a mobile self-organizing network, and in particular to an adaptive resource allocation method based on topology semi-transparency. Background Art
[0002] The hierarchical structure diagram of the mobile ad hoc network (MANET) network protocol stack is as follows: Figure 1 As shown in the figure, the Media Access Control (MAC) sublayer is located in the data link layer of the Ad Hoc network protocol stack, which is at a lower level of the network protocol stack. Its main function is to control the network nodes to access the wireless channel at the right time. It is the direct controller of the message sending and receiving on the channel. Whether the channel access protocol can effectively utilize the limited bandwidth of the wireless channel and ensure the fair access of each node in the network to the wireless channel plays a very important role in the overall performance of the Ad Hoc network. The above capabilities are concentrated in the resource allocation algorithm of the multi-hop wireless network of the MAC layer of the MANET network.
[0003] According to different allocation methods, the resource allocation algorithm of the MAC protocol can be roughly divided into a contention-based resource allocation algorithm and a scheduling-based MAC resource allocation algorithm.
[0004] The basic idea of the contention-based MAC protocol is that nodes in the network actively acquire channels through competition. When a conflict occurs, they compete for channels again according to a backoff mechanism until the data packet is successfully sent or the transmission is abandoned due to delay overflow. In mobile ad hoc networks, typical representatives of contention-based MAC protocols are ALOHA and CSMA, which are more suitable for transmitting bursty data services. In this type of protocol, wireless nodes access the wireless channel through competition. When multiple nodes try to transmit signals at the same time, conflicts may occur. Therefore, each node needs to access the channel in turn according to the conflict resolution algorithm. For example, CSMA, CSMA / CD, MACA, FAMA, 802.11DFW, etc. are all typical contention-based resource allocation methods. This type of protocol is simple in design, and time slots can be divided or not divided. Nodes can compete for wireless network resources in an asynchronous manner.
[0005] The resource allocation algorithm based on competition has better performance when the network load is light. As the load increases, conflicts increase, packet delay increases, and throughput may drop to zero. Therefore, it is difficult to provide QOS guarantee. It is only suitable for low-load and low or no QOS requirements. Therefore, this type of protocol is not suitable for networks with high requirements for delay and throughput.
[0006] The scheduling-based MAC protocol allocates channel resources to each node in the network according to a certain scheduling algorithm. This type of protocol is suitable for services with high real-time requirements. The scheduling-based MAC protocol is mainly based on TDMA. Its basic idea is to divide time into frames, and then divide the frames into several time slots of equal length, and allocate time slots to nodes according to a certain time slot scheduling algorithm. Therefore, the scheduling-based resource allocation algorithm is also called dynamic TDMA. Although the above example uses time resources, it can also be extended to the use of frequency domain and space resources without affecting the essence of the algorithm. This type of algorithm can be divided into topology transparent algorithms and topology non-transparent algorithms according to the generation process of the resource allocation table and the dependency relationship of the topological structure.
[0007] The topology transparent algorithm refers to the time slot allocation method that is independent of the specific network topology structure. The movement of nodes will not affect the allocation of time slots. Even if the network topology changes, there is no need to recalculate the time slot allocation table. Therefore, the time slot allocation can be established in advance before the network is running, which can bring three benefits: first, the network performance is not affected by the overhead; second, the correct operation of the protocol can be guaranteed at any topology change speed, which can also be said that the protocol is mobile transparent; third, the time slot allocation calculation time is not a key factor. The typical representative of the MAC protocol based on the topology transparent algorithm is the fixed time slot allocation TDMA protocol (which can also be regarded as a special case of dynamic TDMA). Fixed TDMA allocates a permanent and unique transmission time slot to each node. Although TDMA is mobile transparent, its throughput is very low because there is no spatial multiplexing, that is, multiple nodes are not allowed to send at the same time, even if these nodes are far enough apart, there will be no conflict at all. The delay increases with the increase of the network scale, and it cannot meet the services with high delay requirements. However, when the network scale is not large, fixed TDMA has the advantage of minimum delay guarantee.
[0008] The time slot allocation table of the topology non-transparent algorithm is generated based on the topology information and can be implemented in a centralized or distributed manner. The centralized algorithm requires a central control node, which collects the node information of the entire network and centrally allocates time slots for the nodes in the entire network. This type of time slot allocation algorithm allocates time slots based on the topology information of the entire network, so it can achieve a higher channel utilization rate. However, when the central node fails, the entire network will be paralyzed. Therefore, the centralized time slot allocation method is an idealized method for distributed Ad Hoc networks and is not practical.
[0009] The general operation process of the distributed algorithm is: the nodes in the network first obtain the network topology information by interacting with neighboring nodes, usually local topology information, and then establish a time slot allocation table based on the obtained topology information or directly resolve conflicts in the neighborhood through the interaction of control packets to reserve time slots. These time slot allocation tables are independently generated by each node, so the protocol is fully distributed. In order to avoid conflicts in the time slot allocation table, an independent control channel is generally required in the process of sensing topology information, and the control channel is a conflict-free channel, which is generally limited by the number of network nodes. Therefore, the algorithm can make better use of channel resources with few or even no conflicts. Among them, the more representative ones are the Unifying Slot Assignment Protocol (USAP) and the Five-Phase Reservation Protocol (FPRP).
[0010] In mobile ad hoc networks, wireless resources are very limited. How to reasonably schedule wireless resources, improve the utilization of wireless resources, and ensure the fairness of access resources is a key issue. Reasonable resource scheduling algorithms can effectively solve the problem of multiple users sharing limited wireless channel resources in a balanced and efficient manner.
[0011] Based on the contention-based resource allocation algorithm, the protocol is simple in design and nodes can asynchronously compete for wireless network resources. However, there will inevitably be certain conflicts and it is only suitable for services with low load and low or no QOS requirements. It is not suitable for networks with high requirements for latency and throughput.
[0012] Resource allocation based on topology transparency algorithm plans the allocation method in advance before the network runs. It can ensure the correct operation of the protocol under any topology changes and there will be no conflicts between nodes. However, the delay will increase with the increase of network scale and the resource utilization rate is not high.
[0013] The distributed resource allocation algorithm based on topology opacity can schedule and coordinate resources on demand by obtaining local network topology information through signaling interactions between multiple nodes, thereby further reducing latency and spatial multiplexing to improve network capacity. However, relying solely on local network topology information will result in resource allocation effects that are not globally optimal.
[0014] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention
[0015] The purpose of the present invention is to provide an adaptive resource allocation method based on topology semi-transparency, which can solve the problem that static allocation of control signaling resources has low utilization rate and cannot adapt well to changes in network topology, and solve the problem that dynamic allocation of data service resources cannot achieve a globally optimal spatial multiplexing rate by using local topology information.
[0016] To achieve the above object, the present invention provides an adaptive resource allocation method based on topological semi-transparency, comprising the following steps:
[0017] S1: The system available resources are divided from the two dimensions of time domain and frequency domain. In the time domain, they are divided into multiple basic time slot resources of equal length, and in the frequency domain, they are divided into multiple transmission channels. The divided system resources are time-frequency two-dimensional resources, which are divided into control signaling resources and data service resources.
[0018] S2: A topology transparent algorithm is used, where the central node dynamically and uniformly allocates control signaling resources based on the number of nodes in the network, and a topology non-transparent algorithm is used to spatially multiplex data service resources based on network topology information.
[0019] In one embodiment of the present invention, in step S2, the topology transparency algorithm is adopted, and the specific steps of dynamically and uniformly allocating control signaling resources by the central node according to the number of nodes in the network are as follows:
[0020] S201: Allocate the resources of the members within the plan and the members outside the plan respectively, wherein the resources of the members within the plan are uniformly allocated by the central node according to the number of nodes currently on the network, and the control signaling resources are allocated cyclically in a frequency-time division manner according to the principle of MAC addresses from small to large; the resources of the members outside the plan are located by parsing the network parameter messages dynamically and periodically distributed by the central node to complete the occupation of the control signaling resources.
[0021] In one embodiment of the present invention, when an out-of-plan member fails to obtain its own network parameter information, it first selects the corresponding reserved time slot through the time-frequency selection algorithm to initiate a resource application to the central node. After receiving the information, the central node updates the network parameter message to carry the network parameter information of the out-of-plan member, and then parses the new network parameter message to locate its own MAC address information, thereby completing the occupation of control signaling resources.
[0022] In one embodiment of the present invention, after step S201, the following steps are further included:
[0023] S202: When the central node cannot sense the existence of a common node for a period of time, it considers that the common node has left the network, and the central node reclaims the control signaling resources occupied by the common node, updates the network parameter message to delete the relevant information of the common node, and periodically broadcasts the latest network parameter message.
[0024] In one embodiment of the present invention, the timing of sending the application for the resources of the out-of-plan member is determined by using a time-frequency selection algorithm. When there are multiple network members in the network that need to initiate resource applications, the time-frequency selection algorithm needs to be used to coordinate the sending resources of different members to avoid conflicts.
[0025] When transmission fails, the contention access window is expanded and a backoff operation is performed to avoid conflict. The specific process is as follows: if it is the nth transmission, a candidate transmission time-frequency resource is randomly selected from n frames from the current moment for transmission, where n<5.
[0026] In one embodiment of the present invention, in step S2, the specific steps of using a topology non-transparent algorithm to perform spatial multiplexing of data service resources according to network topology information are as follows:
[0027] S203: adding the one-hop and two-hop neighbor nodes of each node to its interference set to construct a node interference set;
[0028] S204: Obtain the entire network topology using the routing protocol, and calculate the number M of nearly optimal conflict-free sets;
[0029] S205: Maximize the spatial reuse rate.
[0030] In one embodiment of the present invention, before step S203, the following steps are further included: constraining the allocation of data service resources, where the constraint conditions are:
[0031] 1) Nodes that are one-hop neighbors cannot occupy the same time-frequency resource at the same time;
[0032] 2) Nodes that are two-hop neighbors cannot occupy the same time-frequency resource at the same time;
[0033] 3) All nodes can be allocated at least one time-frequency resource.
[0034] In one embodiment of the present invention, step S203 specifically includes: the nodes in the network use the routing protocol to obtain the one-hop and two-hop neighbor nodes of all nodes in the network, add the one-hop and two-hop neighbor nodes of each node to its interference set, and complete the construction of the node interference set.
[0035] In one embodiment of the present invention, in step S204, the set V arranged according to the interference degree decreasing rule and the interference set of each node are used as the input of the algorithm, and the solved conflict-free sets IS are used as the output of the algorithm, which specifically includes the following steps:
[0036] S2041: Initially, all nodes in the network cannot find available time and frequency resources;
[0037] S2042: Obtain the entire network topology structure according to the routing protocol and construct an interference set NH(i) of each node;
[0038] S2043: Sort the N nodes in the network according to the interference decreasing rule, and put them into the set V in order;
[0039] S2044: put the first node in the set V into the conflict-free set IS(k);
[0040] S2045: Calculate the nodes of the non-interference set NH(i) of the current node in the set V, and put them into the set V'(k); the set V'(k) represents the set of nodes of the non-interference set NH(i) of all nodes in the current conflict-free set IS(k);
[0041] S2046: Determine whether the set V'(k) is empty. If it is not empty, it means that there are nodes in the set V that can occupy the same time-frequency resources as the nodes in the current non-conflict set IS, and proceed to step S2047; if it is empty, it means that there are no nodes in the set V that can occupy the same time-frequency resources as the nodes in the current non-conflict set IS, and adding any node will cause a conflict, and proceed to step S2049;
[0042] S2047: If the set V'(k) is not empty, put the node with the largest interference (Indeg) in the set V'(k) into the conflict-free set IS(k);
[0043] S2048: Calculate the non-interference set NH(i) of all nodes in the current conflict-free set IS(k), calculate the intersection of these non-interference sets NH(i), update the nodes in the set V'(k) to the intersection, and return to step S2036;
[0044] S2049: If the set V'(k) is empty, output the conflict-free set IS(k), delete the nodes added to the conflict-free set IS(k) from the set V, and set k=k+1 to construct a new IS(k) set, which is initially empty;
[0045] S20410: Determine whether the set V is empty. If it is empty, the entire calculation process ends; if it is not empty, return to S2044 and repeat the above steps until the set V is empty.
[0046] In one embodiment of the present invention, the specific steps of step S205 are as follows:
[0047] S2051: Initialize the maximum conflict-free set MIS. Initially, the maximum conflict-free set MIS is equal to the conflict-free set IS(1) in step S204, that is, MIS(1)=IS(1);
[0048] S2052: sort the N nodes in the topology according to a certain rule (the nodes are arranged in ascending order according to the interference degree; if the interference degrees are the same, they are arranged in ascending order according to the node ID), and put them into the set V in order;
[0049] S2053: Each node traverses M conflict-free sets in sequence according to the conflict-free set number k=1, and searches for a conflict-free set that can be added again; wherein the node with sequence number i=1 in the set V in the initial state starts traversing from the conflict-free set number k=1.
[0050] S2054: Determine whether node i can be added to the maximal conflict-free set MIS(k), that is, determine whether node i is in the interference set of any node in the maximal conflict-free set MIS(k). If there is no interference, add it to the maximal conflict-free set MIS(k); if there is interference, do not add it to the maximal conflict-free set MIS(k);
[0051] S2055: Determine whether node i has traversed all M conflict-free sets. If yes, jump to the next step S2056; if no, set k=k+1, continue to determine whether node i can join the next conflict-free set, and jump to step S2054;
[0052] S2056: Determine whether all N nodes have traversed M conflict-free sets. If so, end the operation; if not, add 1 to the node number, set i=i+1, and go to step S2053.
[0053] Compared with the prior art, an adaptive resource allocation method based on topological semi-transparency according to the present invention adopts an adaptive resource allocation algorithm based on topological semi-transparency when performing dynamic resource allocation; the topological semi-transparent algorithm is a combination of topological transparent and topological non-transparent algorithms, and the algorithm fully utilizes the advantages of the topological transparent algorithm in terms of minimum delay guarantee, access fairness, topological independence, and robustness, and the advantages of the topological non-transparent algorithm in terms of spatial multiplexing. In addition, the present invention considers the resource allocation of control signaling and data services respectively, which not only ensures that the system can adapt to topological changes and make full use of time-frequency resources, but also improves network capacity. At the same time, the resource allocation algorithm does not require interactive signaling between nodes for resource coordination, so no additional control overhead is required, which avoids the resource acquisition delay caused by the interaction of control information and reduces overhead. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 It is a schematic diagram of a network protocol stack in the prior art;
[0055] Figure 2 is a flow chart of an adaptive resource allocation method based on topology semi-transparency according to an embodiment of the present invention;
[0056] Figure 3It is a schematic diagram of a divided frame structure of an adaptive resource allocation method based on topology semi-transparency according to an embodiment of the present invention;
[0057] Figure 4 This is a schematic diagram of control signaling resource allocation based on a topology semi-transparent adaptive resource allocation method according to an embodiment of the present invention;
[0058] Figure 5 It is a flowchart of the steps of calculating a conflict-free set of an adaptive resource allocation method based on topological semi-transparency according to an embodiment of the present invention;
[0059] Figure 6 This is a flowchart of a method for maximizing spatial reuse rate based on a topological semi-transparent adaptive resource allocation method according to an embodiment of the present invention;
[0060] Figure 7 is a network topology diagram of the example;
[0061] Figure 8 This is an example diagram for calculating the number of conflict-free sets;
[0062] Fig. 9 This is an example diagram of maximizing the spatial reuse rate. DETAILED DESCRIPTION
[0063] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings, but it should be understood that the protection scope of the present invention is not limited by the specific implementation modes.
[0064] Unless explicitly stated otherwise, throughout the specification and claims, the term “comprise” or variations such as “include” or “comprising”, etc., will be understood to include the stated elements or components but not to exclude other elements or components.
[0065] In order to solve the problem that static allocation of control signaling resources in multi-hop networks is prone to low resource utilization and cannot adapt well to rapid changes in network topology, and the problem that dynamic allocation of data service resources cannot achieve global optimal spatial reuse using local topology information, the present invention proposes an adaptive resource allocation method based on topology semi-transparency, which combines the topology transparent algorithm with the topology non-transparent algorithm and is implemented through the dynamic allocation strategy of control signaling resources and the dynamic allocation strategy of data service resources.
[0066] like Figures 1 to 9 As shown, a method for adaptive resource allocation based on topological semi-transparency according to a preferred embodiment of the present invention comprises the following steps:
[0067] S1: The available resources of the system are divided from two dimensions: time domain and frequency domain. In the time domain, they are divided into multiple basic time slot resources of equal length, and in the frequency domain, they are divided into multiple transmission channels. The divided time-frequency resources are as follows: Figure 3 As shown in the box.
[0068] The divided system resources are time-frequency two-dimensional resources, which are resources that can be allocated to nodes (i.e., network members). Time-frequency two-dimensional resources are divided into two categories: control signaling resources and data service resources. Among them, control signaling resources are used for resource application, network parameter distribution, and network perception (including network members' perception and collection of channel status, neighbor information, network topology information, etc.); data service resources are used for all nodes in the network (i.e., network members) to transmit data services.
[0069] like Figure 3 As shown, an example of dividing system resources in two dimensions, time domain and frequency domain, is shown, and the system resources are divided into M channels in the frequency domain. Figure 3 In the figure, the control signaling resource is the blank part in the figure, and a topological transparent algorithm is used to allocate a unique and conflict-free control signaling resource to the online node (i.e., the active node in the network); the data service resource is the shaded part in the figure, and a topological non-transparent algorithm is used in combination with the network topology information to perform distributed resource spatial multiplexing operations, and the globally optimal solution for spatial multiplexing performance is found by using a distributed maximal independent set method based on the graph vertex coloring theory.
[0070] S2: A topology transparent algorithm is used, where the central node dynamically and uniformly allocates control signaling resources based on the number of nodes in the network, and a topology non-transparent algorithm is used to spatially multiplex data service resources based on network topology information.
[0071] Specifically, for control signaling resources, it mainly includes two processes. The first is resource allocation. Control signaling resources are centrally and uniformly allocated by the central node, and ordinary nodes occupy corresponding time and frequency resources according to their own MAC addresses. The second is resource recovery. When the central node senses that an ordinary node has left the network, the control signaling resources of the ordinary node are recovered and reused by the central node.
[0072] Among them, the central node refers to: the time and frequency reference station of the system. Generally, the system has a central node, which can be obtained by preset or dynamic election. Ordinary nodes refer to: there can be multiple, and during the operation of the network, they participate in the dynamic management of the network together with the central node. Ordinary nodes can be divided into planned members and unplanned members. Planned members refer to: the central node carries the network parameter information of the member in the network parameter signaling, and the member is a planned member. Unplanned members refer to: the central node does not carry the network parameter information of the member in the network parameter signaling, and the member is an unplanned member.
[0073] In step S2, the topology transparency algorithm is used, and the specific steps for the central node to dynamically and uniformly allocate control signaling resources according to the number of nodes in the network are as follows:
[0074] S201: Allocate resources of members within the plan and members outside the plan respectively, wherein the allocation of resources of members within the plan is uniformly allocated by the central node according to the number of nodes currently on the network, and the control signaling resources are allocated in a frequency-time-division cycle according to the principle of MAC addresses from small to large; the allocation of resources of members outside the plan is to locate the MAC address information of the member by parsing the network parameter message dynamically distributed by the central node, so as to complete the occupation of the control signaling resources. The network parameter message includes at least the number of network members and the MAC addresses of the members.
[0075] When an out-of-plan member does not obtain its own network parameter information, it first selects the corresponding reserved time slot through the time-frequency selection algorithm to initiate a resource application to the central node. After receiving the information, the central node updates the network parameter message, carrying the network parameter information of the out-of-plan member, and then parses the new network parameter message and locates its own MAC address information to complete the occupation of control signaling resources.
[0076] The control signaling resources are divided into two parts: one part is the resources of the members in the plan, that is, fixed resources, which are dynamically occupied by the nodes in the network according to the MAC address; the other part is the resources of the members outside the plan, that is, reserved resources (a small amount), which are occupied by the members outside the plan using the two-dimensional CSMA method of time and frequency to send resource applications. Among them, the two-dimensional CSMA method of time and frequency is a resource usage strategy. Network members compete to seize available resources. The specific algorithm is as follows:
[0077] The resource allocation of members in the plan is that network members parse the network parameter messages dynamically distributed by the central node and obtain their own MAC address information to complete the occupation of control signaling resources. This process is a one-way network parameter distribution process. After the network members locate their own network parameter information, they can complete the resource occupation.
[0078] The control signaling resources adopt a dynamic allocation strategy that adapts to the number of network nodes and uses a topology transparency algorithm to flexibly allocate time and frequency resources according to the number of nodes in the network, avoiding the large resource waste that may be caused by static resource configuration, further improving resource utilization and being able to adapt to changes in network topology.
[0079] After dynamically and uniformly allocating the control signaling resources in step S201, the method further includes:
[0080] S202: Resource Recovery: When the central node fails to detect the presence of a certain ordinary node for a continuous period of time, it is considered that the ordinary node has withdrawn from the network. The central node recovers the control signaling resources occupied by the ordinary node, and at the same time updates the network parameter message to delete the relevant information of the ordinary node, and periodically broadcasts and sends the latest network parameter message to facilitate resource applications initiated by members outside the plan.
[0081] In step S202, for the resource application sending timing of members outside the plan, a time-frequency selection algorithm is used to determine. That is, when there are multiple network members in the network that need to initiate resource applications, the time-frequency selection algorithm needs to be used to coordinate the sending resources of different members to avoid conflicts.
[0082] Members outside the plan regard the reserved resources in the control signaling resources (i.e., idle resources that have not been allocated to network members) as candidate sending time-frequency resources, and adopt a time-frequency two-dimensional CSMA strategy. When the sending fails, the contention access window is expanded and a backoff operation is performed to avoid conflicts. The specific process is as follows:
[0083] 1. If it is the first transmission, within 1 time frame starting from the current moment, randomly select a candidate sending time-frequency resource for transmission;
[0084] 2. If it is the second transmission, within 2 time frames starting from the current moment, randomly select a candidate sending time-frequency resource for transmission;
[0085] 3. And so on. If it is the nth transmission, within n time frames from the current moment, randomly select a candidate sending time-frequency resource for transmission, where n < 5.
[0086] As Figure 4 shown, taking a network with N in-network nodes and the node MACs arranged from 1 to N as an example to illustrate the resource occupancy of the control signaling resources using a dynamic allocation method. Assume that there are M channels in the frequency domain. In this example, the relative relationship between the in-network nodes and the channels is M < N < 2M. The control signaling resources are occupied in a time-division and frequency-division cyclic manner in ascending order of the in-network node MAC. The specific resource allocation is as Figure 4 shown. In the first time frame, the control signaling resources are only allocated to nodes with serial numbers 1 to M - 1, and nodes with serial numbers M to N are not allocated resources. Node M starts to occupy the control signaling resources of the second time frame. After all N nodes are allocated time slots, nodes 1 to N start a new round of occupancy. Node 1 in the second round occupies the next time-frequency resource of node N in the first round, and so on, in a cyclic occupancy.
[0087] Specifically, for data service resources, a distributed maximal independent set approach is used based on the vertex coloring theory of graph theory to find the global optimal solution for spatial multiplexing performance. All nodes can obtain the network topology information of the entire network based on the routing protocol. Since two nodes that interfere with each other cannot transmit in the same time-frequency resource, each node in the network first constructs an interference set of all nodes; then, by running the frequency-frequency resource allocation algorithm, a nearly optimal conflict-free node set is solved. Each node in the set can be mapped to the time-frequency resource, and each node in the conflict-free set can occupy a time-frequency resource together.
[0088] For data service resources, a topology non-transparent algorithm is used to perform spatial multiplexing based on network topology information. Specifically, it is divided into the following three steps:
[0089] S203: adding the one-hop and two-hop neighbor nodes of each node to its interference set to construct a node interference set;
[0090] S204: using the routing protocol to obtain the topology of the entire network, and calculating the number M of nearly optimal conflict-free sets; the purpose is to calculate the minimum total number of time-frequency resources that need to be occupied in the topology, that is, using the routing protocol to obtain the topology of the entire network, and then solving the number M of the minimum conflict-free sets IS in the entire network;
[0091] S205: Maximizing the spatial reuse rate. Step S205 is mainly to maximize the spatial reuse rate based on step S204.
[0092] Step S204 allows each node to join a non-conflicting set (occupying one non-conflicting time-frequency resource), but since the above process uses a greedy algorithm, in some topological situations, some nodes can occupy more than one time-frequency resource among the M time-frequency resources without conflict. Step S205 is to enable some nodes to further occupy multiple time-frequency resources.
[0093] Before performing step S203, the allocation of data service resources needs to be constrained. For data service resources, the present invention adopts a topological non-transparent algorithm, combined with a dynamic allocation algorithm for space reuse in the whole network topology operation, allowing nodes separated by two hops to share the same time-frequency resource, thereby improving network capacity; wherein, all nodes in the network can generate a topological connection relationship diagram of the whole network locally through signaling interaction, and nodes occupy time-frequency resources based on the connection relationship diagram, and the constraints on the occupation of time-frequency resources are:
[0094] 1) Nodes that are one-hop neighbors cannot occupy the same time-frequency resource at the same time;
[0095] 2) Nodes that are two-hop neighbors cannot occupy the same time-frequency resource at the same time;
[0096] 3) All nodes can be allocated at least one time-frequency resource.
[0097] The above constraints are the prerequisites for calculating the number M of conflict-free sets.
[0098] All nodes run the resource allocation algorithm locally to occupy resources without conflict, and try to occupy the least time-frequency resources to obtain the maximum channel utilization. As long as the nodes within the two-hop range can be allocated different time-frequency resources, conflicts can be avoided.
[0099] In step S203, the principle of resource allocation is that two nodes that interfere with each other cannot use the same time-frequency resource. Regarding the construction of the node interference set, since the network model diagram can only intuitively show the positional relationship between each node and link in the network, but the interference relationship between nodes is not clear, each node needs to construct its own interference set before calculating the near-optimal number of conflict-free sets M.
[0100] Step S203 specifically includes: the nodes in the network use the routing protocol to obtain the one-hop and two-hop neighbor nodes of all nodes in the network, add the one-hop and two-hop neighbor nodes of each node to its interference set, and complete the construction of the node interference set.
[0101] like Figure 5 As shown, in step S204, the set V arranged according to the interference decreasing rule and the interference set of each node are used as the input of the algorithm, and the solved conflict-free sets IS are used as the output of the algorithm. The specific solving steps are as follows:
[0102] S2041: Initially, all nodes in the network have not found any available time and frequency resources, and the conflict-free set IS is empty;
[0103] S2042: Obtain the entire network topology structure according to the routing protocol and construct an interference set NH(i) of each node;
[0104] S2043: Sort the N nodes in the network according to the interference decreasing rule, and put them into the set V in order;
[0105] S2044: put the first node in the set V into the conflict-free set IS(k);
[0106] S2045: Calculate the nodes of the non-interference set NH(i) of the current node in the set V (i.e., the one-hop and two-hop neighbor nodes of the current node), and put them into the set V'(k); the set V'(k) represents the set of nodes of the non-interference set NH(i) of all nodes in the current conflict-free set IS(k);
[0107] S2046: Determine whether the set V'(k) is empty. If it is not empty, it means that there are nodes in the set V that can occupy the same time-frequency resources as the nodes in the current non-conflict set IS, and proceed to step S2047; if it is empty, it means that there are no nodes in the set V that can occupy the same time-frequency resources as the nodes in the current non-conflict set IS, and adding any node will cause a conflict, and proceed to step S2049;
[0108] S2047: If the set V'(k) is not empty, put the node with the largest interference (Indeg) in the set V'(k) into the conflict-free set IS(k);
[0109] S2048: Calculate the non-interference set NH(i) of all nodes in the current conflict-free set IS(k), calculate the intersection of these non-interference sets NH(i), update the nodes in the set V'(k) to the intersection, and return to step S2036;
[0110] S2049: If the set V'(k) is empty, output the conflict-free set IS(k), delete the nodes added to the conflict-free set IS(k) from the set V, and set k=k+1 to construct a new IS(k) set, which is initially empty;
[0111] S20410: Determine whether the set V is empty. If it is empty, the entire calculation process ends; if it is not empty, return to S2044 and repeat the above steps until the set V is empty.
[0112] Through the above steps, the number of time-frequency resources occupied by the nodes in the network can be calculated, and the nodes in each conflict-free set IS(k) can occupy the same time-frequency resource.
[0113] like Figure 6 As shown, the specific steps of step S205 are as follows:
[0114] S2051: Initialize the maximum conflict-free set MIS. Initially, the maximum conflict-free set MIS is equal to the conflict-free set IS(1) in step S204, that is, MIS(1)=IS(1);
[0115] S2052: sort the N nodes in the topology according to a certain rule (the nodes are arranged in ascending order according to the interference degree; if the interference degrees are the same, they are arranged in ascending order according to the node ID), and put them into the set V in order;
[0116] S2053: Each node traverses M conflict-free sets in sequence according to the conflict-free set number k=1, and searches for a conflict-free set that can be added again; wherein, the initial state is that the node with sequence number i=1 in set V starts traversing from the conflict-free set number k=1; the value of M is the number of IS(k) sets in the previous stage.
[0117] S2054: Determine whether node i can be added to the maximal conflict-free set MIS(k), that is, determine whether node i is in the interference set of any node in the maximal conflict-free set MIS(k). If there is no interference, add it to the maximal conflict-free set MIS(k); if there is interference, do not add it to the maximal conflict-free set MIS(k);
[0118] S2055: Determine whether node i has traversed all M conflict-free sets. If yes, jump to the next step S2056; if no, set k=k+1, continue to determine whether node i can join the next conflict-free set, and jump to step S2054;
[0119] S2056: Determine whether all N nodes have traversed M conflict-free sets. If so, end the operation; if not, add 1 to the node number, set i=i+1, and go to step S2053.
[0120] by Figure 7 Taking the network topology in as an example (the numbers in the circle represent the node ID, the two nodes connected by the solid line are one-hop neighbors, and the two nodes connected by the dotted line are two-hop neighbors), the interference set and interference degree of each node are shown in the following table.
[0121] Table 1 Interference sets and interference degrees of each node
[0122]
[0123]
[0124] The set V is sorted in descending order according to the interference degree. If the interference degrees of the nodes are the same, the node with the smaller ID is put into the set first, and the set V = {3, 7, 8, 2, 4, 5, 9, 1, 6}.
[0125] Figure 8 The operation diagram of step S204 of the present invention is shown in FIG. 1 , and the operation results are IS1={3, 6}, IS2={7}, IS3={8, 1}, IS4={2, 5}, and IS5={4, 9}. The calculation result of step S204 is that 9 nodes only need 5 time slots, which improves the spatial reuse rate compared with the common dynamic allocation of 9 nodes occupying 9 time slots.
[0126] Fig. 9The operation schematic diagram of step S205 of the present invention is shown in FIG. 1 , and the set V is sorted in ascending order according to the size of the interference degree. The interference degrees of the nodes are the same, and the nodes with smaller IDs are added to the set first, and the set V = {1, 6, 9, 5, 2, 4, 3, 7, 8}. The calculation results of step S204 are MIS1 = {3, 6}, MIS2 = {7, 6}, MIS3 = {8, 1}, MIS4 = {2, 5}, MIS5 = {4, 9, 1}. Node 6 can join both MIS1 and MIS2 sets at the same time, and node 1 can join both MIS3 and MIS5 sets at the same time.
[0127] Compared with step S204, node 6 can only join set IS1, and node 1 can only join set IS3. The result of step S205 is that 9 nodes only need 5 time slots, and nodes 1 and 6 can use 2 time slots; compared with the common dynamic allocation of 9 nodes occupying 9 time slots, each node can only use 1 time slot, which further improves the spatial reuse rate.
[0128] The following table illustrates the time slot occupancy effect after adopting the spatial reuse algorithm under the above 9 node topologies. The A marked in the table represents the time slot occupancy of each node in step S204, and the B marked in the table represents the time slot occupancy after further improving the spatial reuse rate in step S205.
[0129] Table 2 Service time slot allocation effect table
[0130]
[0131] The A marked in the table indicates the time slot occupancy of the 9 nodes in the network in step S204, and the IS(k) set occupies the kth time slot.
[0132] IS1 = {3, 6}, i.e. nodes 3 and 6 occupy time slot 1;
[0133] IS2 = {7}, i.e., node 7 occupies time slot 2;
[0134] IS3 = {1, 8}, i.e. nodes 1 and 8 occupy time slot 3;
[0135] IS4 = {2, 5}, i.e. nodes 2 and 5 occupy time slot 4;
[0136] IS5={4, 9}, that is, nodes 4 and 9 occupy time slot 5.
[0137] The B marked in the table indicates the occupancy of time slots further reused by the 9 nodes in the network in step S205.
[0138] ① In step S204, IS2={7}, and in step S205, MIS2={7, 6}, that is, slot 2 can only be allocated to node 7 for occupation in step S204, and can also be allocated to node 6 for occupation in step S205.
[0139] ② In step S204, IS5={4, 9}, and in step S205, MIS2={4, 9, 1}, that is, slot 5 can only be allocated to nodes 4 and 9 for occupation in step S204, and can also be allocated to node 1 for occupation in step S205.
[0140] Through the method in step S205, time slot 2 and time slot 5 can be allocated to more nodes for use, further improving the spatial reuse rate.
[0141] The present invention combines a topology transparent algorithm with a topology non-transparent algorithm and is implemented by controlling a dynamic unified allocation strategy of signaling resources and a dynamic allocation strategy of data service resources.
[0142] 1. The present invention proposes an adaptive allocation strategy for control signaling resources, which adopts a topology transparent algorithm, and the central node dynamically and uniformly allocates control resources according to the number of nodes in the network. This strategy can adapt to changes in network topology, and the nodes in the network divide the control signaling resources into frequency and time division cycles according to the principle of MAC addresses from small to large, ensuring that each node in the network can be allocated a unique and conflict-free control signaling resource. When the central node senses that a common node has left the network, the control signaling resources occupied by it are recycled and reused. The control signaling resource allocation strategy in the present invention adaptively allocates control signaling resources flexibly according to the number of nodes in the network, can adapt to changes in topology, make full use of time and frequency resources, and avoid the problem of large resource waste that may be caused by static resource configuration.
[0143] 2. The present invention proposes a dynamic allocation strategy for adaptive data service resources, which adopts a topological non-transparent algorithm and uses a distributed maximal independent set algorithm in combination with the topology information of the entire network to allocate and reuse data service resources. This strategy constructs a maximal conflict-free set, adds as many non-interfering nodes as possible (nodes separated by more than two hops) to the same conflict-free set, and allows multiple nodes in the same conflict-free set to occupy the same time-frequency resources. The data service resource allocation strategy in the present invention adaptively uses the topology information of the entire network to obtain a spatial reuse rate that is close to the global optimal, realizes the reuse of data service resources beyond two hops, improves resource utilization, increases network capacity, and can adapt well to changes in network topology while avoiding resource allocation conflicts that may be caused by competition.
[0144] The foregoing description of specific exemplary embodiments of the present invention is for the purpose of illustration and demonstration. These descriptions are not intended to limit the present invention to the precise form disclosed, and it is clear that many changes and variations can be made based on the above teachings. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the present invention and its practical application, so that those skilled in the art can realize and utilize various different exemplary embodiments of the present invention and various different selections and changes. The scope of the present invention is intended to be limited by the claims and their equivalents.
Claims
1. A method for adaptive resource allocation based on topological semi-transparency, characterized in that: The following steps are involved: S1: The system available resources are divided from the two dimensions of time domain and frequency domain. In the time domain, they are divided into multiple basic time slot resources of equal length, and in the frequency domain, they are divided into multiple transmission channels. The divided system resources are time-frequency two-dimensional resources, which are divided into control signaling resources and data service resources. S2: A topology transparent algorithm is used, where the central node dynamically and uniformly allocates control signaling resources based on the number of nodes in the network, and a topology non-transparent algorithm is used to spatially multiplex data service resources based on network topology information.
2. The method for adaptive resource allocation based on topological semi-transparency according to claim 1, characterized in that: In step S2, the topology transparency algorithm is adopted, and the central node dynamically and uniformly allocates the control signaling resources according to the number of nodes in the network. The specific steps are as follows: S201: Allocate the resources of the members within the plan and the members outside the plan respectively, wherein the resources of the members within the plan are uniformly allocated by the central node according to the number of nodes currently on the network, and the control signaling resources are allocated cyclically in a frequency-time division manner according to the principle of MAC addresses from small to large; the resources of the members outside the plan are located by parsing the network parameter messages dynamically and periodically distributed by the central node to complete the occupation of the control signaling resources.
3. The method for adaptive resource allocation based on topological semi-transparency according to claim 2, characterized in that: When an out-of-plan member does not obtain its own network parameter information, it first selects the corresponding reserved time slot through the time-frequency selection algorithm to initiate a resource application to the central node. After receiving the information, the central node updates the network parameter message, carrying the network parameter information of the out-of-plan member, and then parses the new network parameter message to locate its own MAC address information, thereby completing the occupation of control signaling resources.
4. The method for adaptive resource allocation based on topological semi-transparency according to claim 2, characterized in that: After step S201, the following steps are also included: S202: When the central node cannot sense the existence of a common node for a period of time, it considers that the common node has left the network, and the central node reclaims the control signaling resources occupied by the common node, updates the network parameter message to delete the relevant information of the common node, and periodically broadcasts the latest network parameter message.
5. The method for adaptive resource allocation based on topological semi-transparency according to claim 4, characterized in that: The timing of sending the application for resources of members outside the plan is determined by the time-frequency selection algorithm. When there are multiple network members that need to initiate resource applications in the network, the time-frequency selection algorithm is needed to coordinate the sending resources of different members to avoid conflicts. When transmission fails, the contention access window is expanded and a backoff operation is performed to avoid conflict. The specific process is as follows: if it is the nth transmission, a candidate transmission time-frequency resource is randomly selected from n frames from the current moment for transmission, where n<5.
6. The method for adaptive resource allocation based on topological semi-transparency according to claim 2, characterized in that: In step S2, the specific steps of using a topology non-transparent algorithm to perform spatial multiplexing of data service resources according to network topology information are as follows: S203: adding the one-hop and two-hop neighbor nodes of each node to its interference set to construct a node interference set; S204: Obtain the entire network topology using the routing protocol, and calculate the number M of nearly optimal conflict-free sets; S205: Maximize the spatial reuse rate.
7. The method for adaptive resource allocation based on topological semi-transparency according to claim 6, characterized in that: Before step S203, the following steps are also included: constraining the allocation of data service resources, and the constraint conditions are: 1) Nodes that are one-hop neighbors cannot occupy the same time-frequency resource at the same time; 2) Nodes that are two-hop neighbors cannot occupy the same time-frequency resource at the same time; 3) All nodes can be allocated at least one time-frequency resource.
8. The method for adaptive resource allocation based on topological semi-transparency according to claim 7, characterized in that: Step S203 specifically includes: the nodes in the network use the routing protocol to obtain the one-hop and two-hop neighbor nodes of all nodes in the network, add the one-hop and two-hop neighbor nodes of each node to its interference set, and complete the construction of the node interference set.
9. The method for adaptive resource allocation based on topological semi-transparency according to claim 8, characterized in that: In step S204, the set V arranged according to the interference degree decreasing rule and the interference set of each node are used as the input of the algorithm, and the solved conflict-free sets IS are used as the output of the algorithm, which specifically includes the following steps: S2041: Initially, all nodes in the network cannot find available time and frequency resources; S2042: Obtain the entire network topology structure according to the routing protocol and construct an interference set NH(i) of each node; S2043: Sort the N nodes in the network according to the interference decreasing rule, and put them into the set V in order; S2044: put the first node in the set V into the conflict-free set IS(k); S2045: Calculate the nodes of the non-interference set NH(i) of the current node in the set V, and put them into the set V'(k); the set V'(k) represents the set of nodes of the non-interference set NH(i) of all nodes in the current conflict-free set IS(k); S2046: Determine whether the set V'(k) is empty. If it is not empty, it means that there are nodes in the set V that can occupy the same time-frequency resources as the nodes in the current non-conflict set IS, and proceed to step S2047; if it is empty, it means that there are no nodes in the set V that can occupy the same time-frequency resources as the nodes in the current non-conflict set IS, and adding any node will cause a conflict, and proceed to step S2049; S2047: If the set V'(k) is not empty, put the node with the largest interference (Indeg) in the set V'(k) into the conflict-free set IS(k); S2048: Calculate the non-interference set NH(i) of all nodes in the current conflict-free set IS(k), calculate the intersection of these non-interference sets NH(i), update the nodes in the set V'(k) to the intersection, and return to step S2036; S2049: If the set V'(k) is empty, output the conflict-free set IS(k), delete the nodes added to the conflict-free set IS(k) from the set V, and set k=k+1 to construct a new IS(k) set, which is initially empty; S20410: Determine whether the set V is empty. If it is empty, the entire calculation process ends; if it is not empty, return to S2044 and repeat the above steps until the set V is empty.
10. The method for adaptive resource allocation based on topological semi-transparency according to claim 9, characterized in that: The specific steps of step S205 are as follows: S2051: Initialize the maximal conflict-free set MIS. Initially, the maximal conflict-free set MIS is equal to the conflict-free set IS(1) in step S204, that is, MIS(1)=IS(1); S2052: sort the N nodes in the topology according to a certain rule (the nodes are arranged in ascending order according to the interference degree; if the interference degrees are the same, they are arranged in ascending order according to the node ID), and put them into the set V in order; S2053: Each node traverses M conflict-free sets in sequence according to the conflict-free set number k=1, and searches for a conflict-free set that can be added again; wherein the node with sequence number i=1 in the set V in the initial state starts traversing from the conflict-free set number k=1. S2054: Determine whether node i can be added to the maximal conflict-free set MIS(k), that is, determine whether node i is in the interference set of any node in the maximal conflict-free set MIS(k). If there is no interference, add it to the maximal conflict-free set MIS(k); if there is interference, do not add it to the maximal conflict-free set MIS(k); S2055: Determine whether node i has traversed all M conflict-free sets. If yes, jump to the next step S2056; if no, set k=k+1, continue to determine whether node i can join the next conflict-free set, and jump to step S2054; S2056: Determine whether all N nodes have traversed M conflict-free sets. If so, end the operation; if not, add 1 to the node number, set i=i+1, and go to step S2053.
