Ad hoc network resource scheduling method based on selection cooperation

Abstract

The invention discloses an ad hoc network resource scheduling method based on selection cooperation, and belongs to the technical field of wireless ad hoc network communication networks. According to the invention, business information statistics and convergence are periodically carried out at a network node, and business applications are converged to a cluster head; the cluster head gathers whole-network service applications, calculates time slot allocation information according to service application information, then issues the whole-network time slot allocation information to the whole network in a unified manner through cooperative communication, and finally forms a time slot allocation result of the nodes through distributed calculation of the nodes of the whole network. According to the method, whether the cooperative communication mode is adopted for service and signaling transmission or not can be selected according to needs, energy consumption caused by node forwarding is reduced, and the problems that in cooperative communication, node energy consumption is high, signaling overhead is large, and the transmission rate cannot be self-adapted can be solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless ad hoc network (MANET) communication networks, and in particular relates to a MANET resource scheduling method based on selective collaboration, which can reduce the protocol overhead of MANET and improve resource allocation efficiency and transmission reliability. Background Art

[0002] Collaborative communication technology, a key research area in wireless networks, can significantly improve network coverage and transmission efficiency by enabling multiple nodes to work together to form a virtual MIMO (Multiple Input, Multiple Output) system. In recent years, this technology has demonstrated significant advantages in improving broadcast efficiency and enhancing network reliability.

[0003] In cooperative communication, multiple nodes can simultaneously send information to a target node, leveraging spatial diversity gain to expand coverage and reduce transmission latency. While cooperative communication offers performance advantages, efficient allocation of time slot resources in multi-point dynamic networking remains a challenge. Furthermore, because relay nodes in cooperative communication must continuously receive, process, and forward data, their power consumption becomes a key bottleneck restricting long-term applications. Consequently, existing solutions suffer from the following drawbacks:

[0004] (1) High node energy consumption: Relay nodes need to continuously receive, process and forward data, which will result in high energy consumption.

[0005] (2) Unable to achieve adaptive transmission: Collaborative transmission requires the nodes participating in the relay cooperation to use the same modulation and coding method, which makes it impossible to achieve point-to-point or end-to-end rate adaptive transmission.

[0006] (3) High signaling overhead: To ensure conflict-free transmission of service data, nodes need to negotiate time slot resources through multiple interactions, which occupies a large amount of bandwidth. Summary of the Invention

[0007] The present invention aims to overcome the shortcomings of the aforementioned background technologies and provide a method for resource scheduling in an ad hoc network based on selective collaboration. The present invention can achieve conflict-free data transmission between nodes, optimize node energy consumption, and balance signaling overhead with delivery efficiency, resulting in a highly efficient and low-overhead resource scheduling solution.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is:

[0009] A method for scheduling resources in an ad hoc network based on selective collaboration includes the following steps:

[0010] Step 1: Information is exchanged between nodes in the entire network. Nodes in each cluster periodically collect statistics on traffic volume and report the statistical results to the cluster head node.

[0011] Step 2: After the cluster head node summarizes the traffic statistics of the entire network, it calculates the number of time slots actually allocated to each node for each type of service based on the traffic type, and forms the time slot queue information as the input for calculating the time slot table;

[0012] Step 3: The cluster head node broadcasts the time slot queue information calculated in step 2 to the entire network through cooperative communication;

[0013] Step 4: Each node in the cluster arranges time slots according to the received time slot queue information, forms its own time slot table, and completes resource scheduling.

[0014] Furthermore, the specific method of step 1 is:

[0015] The nodes in the cluster count the number of packets to be sent to a certain destination node in their own cache and the number of packets they sent to the destination node in the previous cycle to obtain the traffic statistics for the destination node;

[0016] The nodes in the cluster report the traffic statistics results of each destination node to the cluster head node, which then summarizes the traffic information.

[0017] Furthermore, the specific method of step 2 is:

[0018] The cluster head node determines the type of traffic information;

[0019] If the received service is determined to be a unicast service, it is transmitted in a non-cooperative manner. Based on the source and destination node information of the service, the relay node information for the unicast service is calculated according to the current route. The number of time slots allocated to each relay node is determined based on the modulation and coding scheme used by each relay node in the service. Finally, the cluster head node stores the total number of time slots required for the unicast service, the number of time slots required for each participating node, the destination node information for the unicast service, and the priority information in a unicast time slot queue.

[0020] If it is determined that the received service is a broadcast service, the priority is further determined:

[0021] For high-priority broadcast services, collaborative transmission is used, employing the lowest modulation and coding scheme in the entire network. When allocating collaborative time slots, three time slots are used as a cycle, and the same time slot number is allocated to nodes three hops apart. Finally, the cluster head node stores the total number of time slots required for the collaborative broadcast service and the service source node information in the collaborative time slot queue.

[0022] If it is a low-priority broadcast service, the broadcast tree with the broadcast source node as the root is calculated, and data time slots are allocated to non-leaf nodes on the broadcast tree. The lowest modulation and coding mode in the link state set with the child nodes is selected for time slot resource allocation. Finally, the cluster head node fills the total number of time slots required for the broadcast service and the number of time slots required by each node participating in forwarding into the broadcast time slot queue.

[0023] Furthermore, the specific method of step 4 is:

[0024] Based on the time slot queue information, the total number of time slot requirements is calculated and compared with the number of allocatable time slots. If the total number of time slot requirements is not greater than the number of allocatable time slots, the time slots are arranged according to the total number of time slot requirements.

[0025] If the total number of time slot demands is greater than the number of allocable time slots, and the number of high-priority time slot demands is not greater than the number of allocable time slots, the high-priority time slots are allocated first, and the remaining low-priority time slots are allocated in proportion;

[0026] If the total number of time slot demands is greater than the number of allocable time slots, and the number of high priority time slot demands is greater than the number of allocable time slots, then p% of the allocable time slots will be allocated to high priority services, (100-p)% will be allocated to low priority services, and 60 <p<100;

[0027] After determining the allocation ratio, time slots are arranged. When arranging time slots, high-priority services are arranged first, and then low-priority services are arranged. The specific method is as follows:

[0028] Calculate the time slot arrangement interval required for each node for each service;

[0029] Occupy the time slots in the time slot table according to the arrangement interval;

[0030] After the time slot arrangement is completed, if the number of time slots currently applied for is greater than the number of time slots that can be allocated, time slot resources are reused according to the topology status of the entire network; if the number of time slots currently applied for is not greater than the number of time slots that can be allocated, the idle time slots are released to the nodes on the network, and time slot reuse is also performed.

[0031] The beneficial effects of the present invention are:

[0032] 1. The present invention proposes an information interaction mechanism for resource scheduling of integrated collaborative communication. Nodes within the network report resource application information (time slot requirements, priority) to the cluster head on demand. The cluster head collects service application information and decides whether the service adopts collaborative communication or non-collaborative communication based on the unicast and broadcast service volume and priority information, and generates time slot allocation information. The time slot allocation information is broadcast to the entire network through the collaboration of multiple relay nodes to ensure that the time slot resources of the entire network are quickly updated.

[0033] 2. The present invention adopts a hybrid transmission mechanism of "broadcast cooperation-unicast direct transmission". Network nodes do not need to participate in collaborative forwarding all the time. Instead, they switch between collaborative transmission and non-cooperative transmission sending strategies in real time according to network data and signaling requirements, reducing the energy consumption caused by node forwarding. It also integrates the advantages of efficient broadcasting of collaborative communication and adaptive transmission of non-cooperative communication, providing an efficient, reliable and low-overhead resource scheduling solution for dynamic networks.

[0034] 3. The present invention proposes a time slot arrangement algorithm based on greedy calculation, which optimizes the transmission delay jitter and improves the available bandwidth of the application service through the time slot release and multiplexing mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of an embodiment of the present invention.

[0036] Figure 2 It is a flowchart of time slot allocation based on selection collaboration.

[0037] Figure 3 It is a flow chart of adjusting the number of allocated time slots.

[0038] Figure 4 It is a time slot arrangement flow chart. DETAILED DESCRIPTION

[0039] The following is a clear and complete description of the technical solutions in the embodiments of the present invention with reference to the accompanying drawings and examples. However, the embodiments described herein are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0040] A resource scheduling method for self-organizing networks based on selective collaboration. The method periodically performs business information statistics and aggregation at network nodes, and aggregates business applications to cluster heads. The cluster heads summarize business applications from the entire network and calculate time slot allocation information based on the business application information. The time slot allocation information for the entire network is then uniformly distributed to the entire network through collaborative communication. Finally, all nodes in the entire network complete the time slot allocation results of their own nodes through distributed computing.

[0041] like Figure 1 As shown, the specific steps include:

[0042] (1) Business information statistics and aggregation:

[0043] Information is exchanged between nodes across the entire network. Nodes within each cluster periodically collect statistics on traffic volume and report the statistical results to the cluster head node.

[0044] (2) Calculation of time slot allocation information:

[0045] After the cluster head node collects the traffic statistics of the entire network, it calculates the number of time slots actually allocated to each node for each type of service based on the type of traffic information, and forms the time slot queue information as the input for calculating the time slot table;

[0046] (3) Collaborative distribution of time slot allocation information:

[0047] The cluster head node broadcasts the time slot queue information calculated in step 2 to the entire network through cooperative communication;

[0048] (4) Time slot allocation result formation:

[0049] Each node in the cluster arranges its own time slots according to the actual number of allocated time slots, forms its own time slot table, and completes resource scheduling.

[0050] Specifically, service information statistics and aggregation utilize a resource statistics strategy based on service requests, allowing the cluster head to aggregate network-wide resource request information. Each active node collects traffic statistics at the beginning of each scheduling cycle, characterizing its service demand by counting the number of packets to be sent in its cache and the number of packets it has sent in the previous cycle. Services sent by the node to different destinations are separately counted, and this statistical information is reported to the cluster head through inter-node information exchange. The cluster head aggregates resource request information from all active nodes across the network and forms a time slot request information table, which serves as input to the time slot calculation algorithm.

[0051] The time slot allocation information is calculated using a time slot allocation algorithm based on selection collaboration, and time slot allocation decisions are made based on the type of service application information collected. The time slot allocation information is represented by a time slot queue. Figure 2 As shown in the figure, first, if the received service is determined to be unicast, it is transmitted in a non-cooperative manner and requires allocation of non-cooperative time slots. The cluster head uses the service's source-destination information and current routing information to calculate the relay node information for the unicast service. It then determines the number of time slots allocated to each hop for relay forwarding based on the modulation and coding scheme used by each hop in the service. The specific calculation formula is as follows:

[0052]

[0053] Among them, N i,j represents the number of time slots required for the service sent from node i to node j, P tx is the number of service packets sent in the previous period, P buffer is the number of packets to be sent in the cache, M i,jis the maximum number of packets per time slot at the corresponding rate level of the current modulation and coding scheme, and ceil() rounds up. Finally, the cluster head stores the total number of time slots required for the unicast service, the number of time slots required by each participating forwarding node, the destination node information for the participating forwarding nodes, and their priority information in the unicast time slot queue.

[0054] If the received service is determined to be a broadcast service, a priority check is performed. If it is a high-priority service, collaborative transmission is used. Collaborative transmission uses the lowest modulation and coding scheme in the entire network to ensure transmission reliability. Since data transmission between nodes in an ad hoc network is considered non-interfering when the distance between nodes is three hops or more, collaborative time slot allocation uses three time slots as a cycle, assigning the same time slot number to nodes within three hops of each other to ensure that nodes do not interfere with each other. Since collaborative time slots are broadcast, and nodes with the same number of hops from the broadcast source node can transmit simultaneously, the cluster head stores the total number of time slots required for the collaborative broadcast service and the service source node information in the collaborative time slot queue.

[0055] If it is a low-priority broadcast service, the cluster head calculates a broadcast tree with the broadcast source node as the root and allocates data time slots to non-leaf nodes on the broadcast tree. When allocating data time slots to branch nodes, the link status between the branch node and its child nodes must be considered. The lowest modulation and coding mode in the link status set between the branch node and its child nodes is selected for time slot resource allocation:

[0056]

[0057] Among them, Nb i represents the number of time slots required for node i to send a broadcast, M min is the maximum number of packets per time slot at the rate level corresponding to the lowest modulation and coding scheme from node i to all its leaf nodes. Finally, the cluster head fills the total number of time slots required for the broadcast service and the number of time slots required by each participating forwarding node into the broadcast time slot queue.

[0058] Collaborative transmission of time slot allocation information means that the cluster head broadcasts the time slot allocation information to the entire network through collaborative communication. Collaborative communication can achieve hop-by-hop diffusion of broadcast information. Therefore, the cluster head only needs N time slots (where N is the maximum number of relay hops in the entire network) to transmit the time slot allocation result to the entire network through collaborative transmission.

[0059] The time slot allocation result is formed by taking the time slot allocation information sent by the cluster head as input, and finally generating the time slot allocation result of each node through the time slot arrangement algorithm based on greedy calculation. Figure 3As shown, the number of time slots used in the actual scheduling must first be determined based on the number of required time slots calculated by the time slot allocation algorithm. The total number of time slot requirements is calculated and compared with the final number of allocable time slots. If the total number of time slot requirements is less than the number of allocable time slots, the time slots are scheduled according to the number of time slot requirements. If the number of time slot requirements is greater than the number of allocable time slots, and the number of high-priority time slot requirements is less than the number of allocable time slots, the high-priority time slots are allocated first, and the remaining low-priority time slots are allocated proportionally. The specific strategy is as follows:

[0060]

[0061] Among them, free_Slot_num is the number of idle time slots after allocating high priority services, r i,j is the low priority time slot demand from node i to node j, S i The number of time slots actually allocated to node i. If the number of time slot demands is greater than the number of allocable time slots, if the total number of time slot demands is greater than the number of allocable time slots, and the number of high priority time slot demands is greater than the number of allocable time slots, then p% of the allocable time slots are allocated to high priority services, (100-p)% are allocated to low priority services, and 60 <p<100。

[0062] After determining the allocation ratio, specific time slot arrangement needs to be carried out. Figure 4 As shown in the figure, when arranging time slots, high priority services are arranged first, and low priority services are arranged second. When arranging, each service runs the time slot arrangement algorithm according to the node in turn, and the time slot arrangement interval required for each node is calculated. The calculation method of the time slot arrangement interval G is:

[0063]

[0064] Max_slot_num is the total number of time slots to be allocated, and Need_slot_num is the number of time slots required by the node. When a node allocates time slots, each interval G occupies one time slot. After the time slot table is allocated, if the number of currently requested time slots exceeds the number of available time slots, time slot resources are reused based on the network topology to increase throughput. If the number of currently requested time slots is less than the number of available time slots, idle time slots are released to active nodes, and time slot reuse is performed to increase the available bandwidth for the requested service.

[0065] In summary, the present invention dynamically collects service application information from the entire network and dynamically selects the transmission mode according to the service type (broadcast / unicast) and priority: high-priority broadcast services adopt multi-node collaborative transmission, low-priority broadcasts are allocated based on broadcast trees, and unicast services adaptively select modulation and coding for direct transmission based on the link status, supporting end-to-end rate optimization; time slot allocation information is broadcast through multi-relay node hop-by-hop collaborative broadcasting to achieve rapid distribution of allocation information, and combined with the greedy algorithm to evenly arrange time slots, reduce delay jitter, support time slot multiplexing and improve network throughput.

[0066] The present invention adopts a hybrid transmission mechanism of "broadcast cooperation-unicast direct transmission", which can choose whether to use cooperative communication for business and signaling transmission on demand, reducing the energy consumption caused by node forwarding, and integrating the advantages of efficient broadcasting of cooperative communication and adaptive transmission of non-cooperative communication, providing a dynamic network with an efficient, reliable and low-overhead resource scheduling solution, which can solve the problems of high node energy consumption, large signaling overhead and inability to adapt transmission rate in cooperative communication.

Claims

1. A resource scheduling method for an ad hoc network based on selective collaboration, characterized in that: The following steps are involved: Step 1: Information is exchanged between nodes in the entire network. Nodes in each cluster periodically collect statistics on traffic volume and report the statistical results to the cluster head node. Step 2: After the cluster head node summarizes the traffic statistics of the entire network, it calculates the number of time slots actually allocated to each node for each type of service based on the traffic type, and forms the time slot queue information as the input for calculating the time slot table; Step 3: The cluster head node broadcasts the time slot queue information calculated in step 2 to the entire network through cooperative communication; Step 4: Each node in the cluster arranges time slots according to the received time slot queue information, forms its own time slot table, and completes resource scheduling.

2. The method for scheduling resources in an ad hoc network based on selective collaboration according to claim 1, characterized in that: The specific method of step 1 is: The nodes in the cluster count the number of packets to be sent to a certain destination node in their own cache and the number of packets they sent to the destination node in the previous cycle to obtain the traffic statistics for the destination node; The nodes in the cluster report the traffic statistics results of each destination node to the cluster head node, which then summarizes the traffic information.

3. The method for resource scheduling in an ad hoc network based on selective collaboration according to claim 1, characterized in that: The specific method of step 2 is: The cluster head node determines the type of traffic information; If the received service is determined to be a unicast service, it is transmitted in a non-cooperative manner. Based on the source and destination node information of the service, the relay node information for the unicast service is calculated according to the current route. The number of time slots allocated to each relay node is determined based on the modulation and coding scheme used by each relay node in the service. Finally, the cluster head node stores the total number of time slots required for the unicast service, the number of time slots required for each participating node, the destination node information for the unicast service, and the priority information in a unicast time slot queue. If it is determined that the received service is a broadcast service, the priority is further determined: For high-priority broadcast services, collaborative transmission is used, employing the lowest modulation and coding scheme in the entire network. When allocating collaborative time slots, three time slots are used as a cycle, and the same time slot number is allocated to nodes three hops apart. Finally, the cluster head node stores the total number of time slots required for the collaborative broadcast service and the service source node information in the collaborative time slot queue. If it is a low-priority broadcast service, the broadcast tree with the broadcast source node as the root is calculated, and data time slots are allocated to non-leaf nodes on the broadcast tree. The lowest modulation and coding mode in the link state set with the child nodes is selected for time slot resource allocation. Finally, the cluster head node fills the total number of time slots required for the broadcast service and the number of time slots required by each node participating in forwarding into the broadcast time slot queue.

4. The method for resource scheduling in an ad hoc network based on selective collaboration according to claim 1, characterized in that: The specific method of step 4 is: Based on the time slot queue information, the total number of time slot requirements is calculated and compared with the number of allocatable time slots. If the total number of time slot requirements is not greater than the number of allocatable time slots, the time slots are arranged according to the total number of time slot requirements. If the total number of time slot demands is greater than the number of allocable time slots, and the number of high-priority time slot demands is not greater than the number of allocable time slots, the high-priority time slots are allocated first, and the remaining low-priority time slots are allocated in proportion; If the total number of time slot demands is greater than the number of allocable time slots, and the number of high priority time slot demands is greater than the number of allocable time slots, then p% of the allocable time slots will be allocated to high priority services, (100-p)% will be allocated to low priority services, and 60 <p<100; After determining the allocation ratio, time slots are arranged. When arranging time slots, high-priority services are arranged first, and then low-priority services are arranged. The specific method is as follows: Calculate the time slot arrangement interval required for each node for each service; Occupy the time slots in the time slot table according to the arrangement interval; After the time slot arrangement is completed, if the number of time slots currently applied for is greater than the number of time slots that can be allocated, time slot resources are reused according to the topology status of the entire network; if the number of time slots currently applied for is not greater than the number of time slots that can be allocated, the idle time slots are released to the nodes on the network, and time slot reuse is also performed.

Images