Wireless ad hoc network communication method for walkie-talkies

Abstract

The present invention proposes a wireless ad hoc network communication method for walkie-talkies, which relates to the technical field of wireless ad hoc networks, including: establishing an initial network topology model; optimizing the initial network topology model using a low-power ad hoc network protocol; dynamically adjusting node transmission power using an adaptive multi-hop power control algorithm, and optimizing the network topology model again based on the adjusted transmission power; using a predictive routing algorithm based on the re-optimized network topology model to calculate the optimal communication path; allocating the optimal communication path to each wireless communication node, and performing adaptive channel selection during data transmission, and dynamically selecting the optimal communication channel based on environmental interference and network load. The present invention can achieve high-performance communication of walkie-talkies in complex environments, while taking into account energy efficiency, network stability, communication quality and security, so as to meet the high-performance communication needs of walkie-talkies in various complex scenarios.

Description

Technical Field

[0001] The invention relates to the technical field of wireless ad hoc networks, and in particular to a wireless ad hoc network communication method for walkie-talkies. Background Art

[0002] As a flexible and reliable communication method, wireless ad hoc networking technology has been widely used in the field of walkie-talkies. Traditional walkie-talkie communication systems mainly rely on fixed infrastructure, which often makes it difficult to ensure the stability and reliability of communication in complex environments or emergency situations. With the development of technology, wireless ad hoc networking technology has been introduced into walkie-talkie systems, enabling devices to autonomously establish networks and achieve point-to-point communication, greatly improving the flexibility and adaptability of communication.

[0003] However, in practical applications, wireless ad hoc intercom systems still face many challenges. First, as portable devices, intercoms have limited battery life. How to improve energy efficiency while ensuring communication quality becomes a key issue. Second, in a complex and changing environment, the network topology often changes. How to quickly adapt to these changes and maintain a stable communication link is also an important issue. In addition, due to the openness of wireless channels, communication security and anti-interference capabilities also require special attention.

[0004] Existing wireless ad hoc intercom systems often use single or simple technical means to deal with these problems, which makes it difficult to fully solve the above challenges. For example, some systems only focus on energy efficiency but ignore communication quality; while other systems can adapt to changes in network topology but fail to fully consider the energy status and mobility of nodes. At the same time, existing systems also have shortcomings in dealing with network congestion, spectrum resource utilization, and communication security.

[0005] In summary, the existing wireless ad hoc network walkie-talkie communication system lacks a comprehensive solution, and it is difficult to take into account multiple aspects such as energy efficiency, network adaptability, communication quality and security at the same time, and cannot meet the high-performance communication needs of walkie-talkies in complex environments. Summary of the invention

[0006] In view of this, the present invention proposes a wireless ad hoc network communication method for walkie-talkies, which realizes high-performance communication of walkie-talkies in complex environments by integrating low-power ad hoc network protocols, adaptive multi-hop power control, predictive routing algorithms, adaptive channel selection and other methods, while taking into account energy efficiency, network stability, communication quality and security, so as to meet the high-performance communication needs of walkie-talkies in various complex scenarios.

[0007] The technical solution of the present invention is implemented as follows: The present invention provides a wireless ad hoc network communication method for walkie-talkies, comprising:

[0008] S1. Obtaining geographical location information, wireless signal strength information, and remaining energy information of wireless communication nodes, and establishing an initial network topology model based on the acquired information;

[0009] S2, using low-power self-organizing network protocol to optimize the initial network topology model, select the optimal cluster head, and balance network energy consumption;

[0010] S3, using an adaptive multi-hop power control algorithm to dynamically adjust the node transmission power, and based on the adjusted transmission power, optimizing the network topology model again;

[0011] S4. Based on the optimized network topology model, a predictive routing algorithm is used to calculate the optimal communication path that comprehensively considers network topology, node mobility and channel conditions in combination with real-time channel quality information;

[0012] S5. Allocate the optimal communication path to each wireless communication node. During the data transmission process, perform adaptive channel selection and dynamically select the optimal communication channel based on environmental interference and network load.

[0013] Based on the above technical solution, preferably, step S2 includes:

[0014] S21. Define the node energy state conversion function:

[0015] ,

[0016] In the formula, is the energy level of the node at time t, is the quantum-inspired state of the node at time t, is the energy attenuation coefficient, is the energy acquisition coefficient, M is the number of neighboring nodes, is a function related to the distance of the jth neighboring node, is the energy level of the jth neighboring node at time t, is the energy transfer efficiency coefficient;

[0017] S22. Initialize the quantum-inspired state of the node , the value range is [0,1];

[0018] S23. Define the quantum cost function C for cluster head selection:

[0019] ,

[0020] Where N is the total number of nodes, is the residual energy of node i, is the average distance from node i to other nodes in the cluster, is the mobility index of node i, , , To adjust the parameters;

[0021] S24. Use the quantum annealing algorithm to update the quantum-inspired state of the node. The state update probability p is:

[0022] ,

[0023] In the formula, is the cost function difference between the current state and the candidate state, is the quantum bit constant, T is the system temperature;

[0024] According to the updated quantum-inspired state , calculate the probability of a node becoming a cluster head:

[0025] P(cluster head)=sin 2 (πQ(t) / 2);

[0026] S25, determining whether a node becomes a cluster head by generating a random number and comparing it with P (cluster head), and selecting the best cluster head for non-cluster head nodes;

[0027] S26, updating node energy level;

[0028] S27. Calculate the entropy of energy distribution within the cluster:

[0029] ,

[0030] In the formula, K is the number of nodes in the cluster, is the energy of node i, is the total energy within the cluster;

[0031] Dynamically adjust the size and number of clusters according to the entropy value to balance network energy consumption;

[0032] S28. Periodically iterate and execute steps S22-S27 to adapt to the dynamic changes of network topology and energy distribution.

[0033] Based on the above technical solution, preferably, the adaptive multi-hop power control algorithm includes:

[0034] Calculate the channel comprehensive score :

[0035] ,

[0036] In the formula, is the comprehensive score of the channel between node i and node j, is the distance between node i and node j, is the path loss exponent, is the signal-to-noise ratio between node i and node j, is the remaining energy of the node, is the energy weight function, is the node mobility index, is the mobility weight function, , , , is the weight coefficient;

[0037] Calculate the transmission power between nodes :

[0038] ,

[0039] In the formula, is the transmission power when node i sends to node j, , are the minimum and maximum transmission powers, To adjust the parameters, control the steepness of the function, control the steepness of the function, is the comprehensive score of the channel between node i and node j, is the threshold parameter;

[0040] Dynamically adjust the node transmission power according to the calculation results.

[0041] On the basis of the above technical solution, preferably, step S4 includes:

[0042] S41. Based on historical data and current network status, a predictive routing model is constructed, and machine learning algorithms are used to predict future network topology changes and node mobility.

[0043] S42. Define comprehensive routing cost function :

[0044] ,

[0045] In the formula, is the distance between node i and node j predicted at time t, is the predicted moving speed of node j at time t, is the historical link stability between node i and node j, is the residual energy of node j at time t, , , , is the weight coefficient;

[0046] S43. Using the comprehensive routing cost function ,The optimal communication path is calculated by minimizing the cumulative routing cost.

[0047] Based on the above technical solution, preferably, the machine learning algorithm used in step S41 is a long short-term memory network or a spatiotemporal graph convolutional network to capture the spatiotemporal dependency of node mobility and link quality.

[0048] Based on the above technical solution, preferably, step S5 includes:

[0049] S51, real-time monitoring of environmental interference and network load, collecting signal-to-noise ratio (SNR) and channel occupancy data;

[0050] S52, define adaptive channel selection function :

[0051] ,

[0052] In the formula, is the channel score for channel k at time t, is the signal-to-noise ratio of channel k at time t, is the occupancy of channel k at time t, is the matching degree of channel k to meet the service quality requirements, , , is the weight coefficient;

[0053] S53, select the highest The channel with the best value is used as the optimal communication channel for data transmission;

[0054] S54. Implement a fast channel switching mechanism to reduce communication interruptions during channel switching.

[0055] Based on the above technical solution, preferably, the matching degree of service quality requirements Based on specific application requirements, including bandwidth, latency, and reliability, fuzzy logic methods are used to calculate:

[0056] W k (t) = Fuzzy(Bandwidth k (t),Delay k (t),Reliability k (t));

[0057] In the formula, Fuzzy() represents the fuzzy logic function, Bandwidth represents bandwidth, Delay represents delay, and Reliability represents reliability.

[0058] Based on the above technical solution, preferably, in step S5, before performing adaptive channel selection, it also includes using a collaborative spectrum sensing mechanism to perform preliminary channel screening, the steps are as follows:

[0059] Each node performs local spectrum sensing to detect available channels;

[0060] Nodes share spectrum sensing information to improve sensing accuracy;

[0061] Define cooperative spectrum sensing function :

[0062] ,

[0063] In the formula, is the collaborative perception score at time t, is the detection probability at time t, is the false alarm probability at time t, is the number of nodes participating in the collaboration at time t, , , is the weight coefficient;

[0064] Choose a higher The channel of the value is used for communication.

[0065] On the basis of the above technical solution, preferably, the method further comprises: implementing distributed network congestion control to balance the network load by adjusting the data transmission rate and routing strategy, the steps are as follows:

[0066] Each node independently monitors its queue length and packet transmission delay;

[0067] When congestion is detected, the node alleviates the congestion by reducing the data sending rate and adjusting the routing strategy;

[0068] Use a credit-based congestion control mechanism to allocate transmission credits based on the node's remaining energy and network contribution;

[0069] Congestion information is periodically exchanged in the network to achieve global congestion awareness and coordinated control.

[0070] On the basis of the above technical solution, preferably, the low-power self-organizing network protocol also includes a sleep scheduling mechanism to dynamically adjust the sleep cycle of the node according to the network traffic.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] (1) By integrating low-power self-organizing network protocols, adaptive multi-hop power control, predictive routing algorithms, and adaptive channel selection methods, high-performance communication of walkie-talkies in complex environments is achieved. This method takes into account multiple aspects such as energy efficiency, network adaptability, communication quality, and security, and significantly improves the overall performance of the walkie-talkie communication system. By dynamically optimizing network topology, intelligently adjusting transmission power, predicting network changes, and adaptively selecting communication channels, the present invention effectively solves the problems of uneven energy consumption, poor network adaptability, and unstable communication quality in the prior art, and provides a comprehensive solution to the high-performance communication needs of walkie-talkies in various complex scenarios;

[0073] (2) The present invention adopts quantum annealing algorithm to select cluster heads and optimize network topology. By dynamically adjusting the size and number of clusters, the network energy consumption is balanced and the life cycle of the entire network is extended. The method introduces node energy state conversion function and quantum cost function, taking into account multiple factors such as node residual energy, location and mobility, making cluster head selection more reasonable. By periodically iterating the cluster head selection process, the system can adapt to the dynamic changes of network topology and energy distribution. In addition, by calculating the entropy of energy distribution within the cluster to dynamically adjust the structure of the cluster, the network energy consumption is further optimized;

[0074] (3) By introducing an adaptive multi-hop power control algorithm, the present invention can dynamically adjust the node transmission power according to the comprehensive channel score, effectively improving the communication quality and energy utilization efficiency. The algorithm takes into account multiple factors such as the distance between nodes, signal-to-noise ratio, node residual energy and mobility, and calculates a comprehensive channel score. Based on this score, the system uses an adaptive function to dynamically adjust the node's transmission power, which not only ensures communication quality but also avoids unnecessary energy waste. This method can respond quickly to real-time changes in the network environment, maximizing energy efficiency while ensuring communication reliability, and is particularly suitable for portable devices such as walkie-talkies that have high requirements for energy management;

[0075] (4) The present invention uses a machine learning algorithm to predict network topology changes and node mobility, combined with a comprehensive routing cost function, to achieve more accurate and efficient routing selection. This predictive routing algorithm not only takes into account the current network status, but also predicts future network changes, thereby making more intelligent routing decisions. The comprehensive routing cost function takes into account multiple factors such as predicted node distance, mobile speed, historical link stability, and node residual energy, and calculates the optimal communication path by minimizing the cumulative routing cost. This method improves routing accuracy and efficiency, reduces unnecessary routing updates and data retransmissions, and thus improves the overall performance and reliability of the network;

[0076] (5) By real-time monitoring of environmental interference and network load, and adopting an adaptive channel selection function, the present invention can dynamically select the optimal communication channel, significantly improving the reliability and anti-interference ability of communication. This method not only takes into account the signal-to-noise ratio and channel occupancy rate, but also introduces the matching degree of service quality requirements, making channel selection more comprehensive and intelligent. By calculating the matching degree of service quality requirements by fuzzy logic method, the system can better meet the needs of specific applications. In addition, the implementation of a fast channel switching mechanism reduces communication interruptions during channel switching, further improving the continuity and stability of communication. This adaptive channel selection method enables the intercom system to maintain high-quality communication in a complex and changing wireless environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0078] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0079] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] like Figure 1 As shown, the present invention provides a wireless ad hoc network communication method for walkie-talkies, comprising:

[0081] S1. Obtaining geographical location information, wireless signal strength information, and remaining energy information of wireless communication nodes, and establishing an initial network topology model based on the acquired information;

[0082] S2, using low-power self-organizing network protocol to optimize the initial network topology model, select the optimal cluster head, and balance network energy consumption;

[0083] S3, using an adaptive multi-hop power control algorithm to dynamically adjust the node transmission power, and based on the adjusted transmission power, optimizing the network topology model again;

[0084] S4. Based on the optimized network topology model, a predictive routing algorithm is used to calculate the optimal communication path that comprehensively considers network topology, node mobility and channel conditions in combination with real-time channel quality information;

[0085] S5. Allocate the optimal communication path to each wireless communication node. During the data transmission process, perform adaptive channel selection and dynamically select the optimal communication channel based on environmental interference and network load.

[0086] Specifically, in one embodiment of the present invention, step S1 includes:

[0087] The geographical location information, wireless signal strength information and remaining energy information of the wireless communication nodes are obtained, wherein the geographical location information is the longitude and latitude coordinates of each walkie-talkie node, which is obtained through the built-in GPS module or triangulation technology; the wireless signal strength information is the signal strength value between each node and other directly communicable nodes, in dBm, which is measured by the wireless communication module of the walkie-talkie. The received signal strength indication RSSI; the remaining energy information is the percentage of the remaining battery power of each node, which is monitored and obtained by the battery management system of the walkie-talkie.

[0088] According to the acquired geographic location information, wireless signal strength information and remaining energy information, a graph model is established to represent the initial network topology structure, and the model contains both digital information and graphical representation.

[0089] The initial network topology model is represented as a weighted undirected graph G=(V,E), where V represents a set of nodes, each node represents a walkie-talkie; E represents a set of edges, each edge represents a communication link between nodes.

[0090] Each node v∈V contains the following attributes:

[0091] id: node unique identifier;

[0092] position: (latitude, longitude) geographic location coordinates;

[0093] energy: The remaining energy percentage.

[0094] Each edge e∈E has the following properties:

[0095] weight: link quality indicator calculated based on wireless signal strength;

[0096] distance: The geographical distance between two nodes.

[0097] Based on the acquired geographic location information and remaining energy information, a node is created for each walkie-talkie. Based on the wireless signal strength information, edges are established between pairs of nodes whose signal strength exceeds a predetermined threshold. The edge weight is calculated using the signal strength value, and the calculation method is: weight = 1 / (1+|RSSI|). The Haversine formula is used to calculate the actual distance based on the geographic coordinates of the node. After completing node creation, connection establishment, edge weight calculation, and distance calculation between nodes, the model is built.

[0098] After constructing the initial network topology model, the model can be expressed through adjacency matrix, adjacency list, visual graph, etc.

[0099] Specifically, in one embodiment of the present invention, after constructing the initial network topology model, a low-power self-organizing network protocol is used to optimize the model for the first time, with the main goal of selecting the optimal cluster head and balancing the network energy consumption. Traditional low-power self-organizing network protocols (such as LEACH) have limitations in cluster head selection and energy management, and cannot fully optimize the energy consumption distribution. This embodiment introduces a quantum-inspired method to solve the problems of cluster head optimization and energy allocation.

[0100] In this embodiment, step S2 includes:

[0101] S21. Define the node energy state conversion function to describe the change of node energy over time:

[0102] ,

[0103] In the formula, is the energy level of the node at time t, is the quantum-inspired state of the node at time t, is the energy attenuation coefficient, is the energy acquisition coefficient, M is the number of neighboring nodes, is a function related to the distance of the jth neighboring node, is the energy level of the jth neighboring node at time t, is the energy transfer efficiency coefficient; in order to more accurately describe the energy transfer between nodes, we further define function:

[0104] ,

[0105] In the formula, It is an adjustable parameter that controls the effect of distance on energy transfer.

[0106] S22. Initialize the quantum-inspired state of the node , the value range is [0,1]; initialization can be based on the initial energy level and position information of the node:

[0107] Q(0)=(E(0) / E max )×(1-d centroid / d max );

[0108] In the formula, is the initial energy of the node, E max is the maximum energy capacity of the node, d centroid is the distance from the node to the network centroid, d max is the distance between the two farthest nodes in the network.

[0109] S23. Define the quantum cost function C for cluster head selection:

[0110] ,

[0111] Where N is the total number of nodes, is the residual energy of node i, is the average distance from node i to other nodes in the cluster, is the mobility index of node i, , , To adjust the parameters.

[0112] In another embodiment, the quantum cost function C may also include other considerations, such as considering the connectivity index C i , then the quantum cost function C is expressed as:

[0113] ,

[0114] In the formula is the connectivity of node i, that is, the number of nodes directly connected to it, To adjust the parameters.

[0115] S24. Use the quantum annealing algorithm to update the quantum-inspired state of the node. The state update probability p is:

[0116] ,

[0117] In the formula, is the cost function difference between the current state and the candidate state, is the quantum bit constant, T is the system temperature; T decays with time: T(t)=T 0 ×e -t / τ , where T 0 is the initial temperature. τ is the temperature decay constant.

[0118] According to the updated quantum-inspired state , calculate the probability of a node becoming a cluster head:

[0119] P(cluster head)=sin 2 (πQ(t) / 2);

[0120] S25, by generating a random number and comparing it with P (cluster head), determine whether the node becomes a cluster head. For non-cluster head nodes, select the best cluster head; the selection of the best cluster head is based on the following comprehensive indicators:

[0121] I=w 1 ×(ES ch / ES max )+w 2 ×(1-d / d max )+w 3 ×(C ch / C max );

[0122] Among them: ES ch is the residual energy of the candidate cluster head, ES max is the maximum residual energy in the network, d is the distance from the non-cluster head node to the candidate cluster head, and d max is the communication range, C ch is the connectivity of candidate cluster heads, C max is the maximum connectivity in the network, w 1 ,w 2 ,w 3 is the weight coefficient.

[0123] S26, update the node energy level; considering the communication overhead and the additional energy consumption of the cluster head, for the cluster head: E(t+1)=E(t)-E ch-intra -E ch-inter ; For cluster members: E(t+1)=E(t)-E cm ; Among them, E ch-intra is the energy consumption of intra-cluster communication, E ch-inter is the energy consumption of inter-cluster communication, E cm is the energy consumption of cluster members sending data to the cluster head.

[0124] S27. Calculate the entropy of energy distribution within the cluster:

[0125] ,

[0126] In the formula, K is the number of nodes in the cluster, is the energy of node i, is the total energy within the cluster;

[0127] Dynamically adjust the size and number of clusters according to the entropy value to balance the network energy consumption; if H>H threshold, consider splitting the cluster, if H <H threshold If the cluster size is smaller than the preset threshold, consider merging the clusters.

[0128] S28. Periodically iterate and execute steps S22-S27 to adapt to the dynamic changes of network topology and energy distribution. The iteration interval is dynamically adjusted according to the speed of network changes: T interval =T base ×(1+δ×ΔE avg ), where T base is the basic iteration interval, δ is the adjustment coefficient, ΔE avg is the average energy change rate of the network in the previous iteration.

[0129] This embodiment introduces the idea of ​​quantum annealing heuristic algorithm to solve the problem of cluster head optimization, and uses quantum superposition state and tunneling effect to search for the optimal cluster head distribution in a global range to balance network energy consumption.

[0130] Specifically, this embodiment evaluates the optimization effect through four indicators: network life cycle, energy balance, cluster head distribution uniformity, and network throughput. If the indicators meet the requirements, the first optimization ends. Among them, the network life cycle is defined as the time when the first node energy is exhausted. Energy balance is defined as: EB=1-(σ E / μ E ), where σ E is the standard deviation of node energy, μ E is the average energy. The cluster head distribution uniformity is defined as: CHD=1-(σ d / μ d ), where σ d is the standard deviation of the distance between cluster heads, μ d is the average distance. Network throughput is defined as the amount of data successfully transmitted per unit time.

[0131] In one embodiment of the present invention, the low-power self-organizing network protocol also includes a sleep scheduling mechanism to dynamically adjust the sleep cycle of the node according to the network traffic. The mechanism dynamically adjusts the sleep cycle of the node according to the real-time network traffic, and maximizes the sleep time of the node while ensuring the communication quality. The specific implementation includes the steps of network traffic monitoring, sleep cycle calculation, dynamic adjustment mechanism, sleep synchronization, special processing of cluster head nodes, emergency wake-up mechanism and adaptive learning. Each node monitors the local network traffic, calculates the appropriate sleep cycle according to the preset algorithm, and coordinates the sleep time with other nodes through the time synchronization protocol. The cluster head node adopts a shorter sleep cycle because it undertakes more communication tasks. The system also includes an emergency wake-up mechanism to deal with emergencies, and uses a machine learning algorithm to continuously optimize the sleep strategy. This mechanism is seamlessly integrated with the existing low-power self-organizing network protocol, and the sleep factor is considered in the network topology optimization and cluster head selection process to ensure network connectivity. In this way, the method can significantly improve the energy efficiency and life cycle of the entire network while maintaining good communication performance.

[0132] Specifically, in one embodiment of the present invention, after completing the establishment of the initial network topology model in step S1 and the optimization of the low-power self-organizing network protocol in step S2, step S3 aims to dynamically adjust the node transmission power using an adaptive multi-hop power control algorithm to further optimize network performance.

[0133] In this embodiment, the adaptive multi-hop power control algorithm includes:

[0134] Calculate the channel comprehensive score :

[0135] ,

[0136] In the formula, is the comprehensive score of the channel between node i and node j, is the distance between node i and node j, is the path loss exponent, is the signal-to-noise ratio between node i and node j, is the remaining energy of the node, is the energy weight function, is the node mobility index, is the mobility weight function, , , , is the weight coefficient.

[0137] Define the energy weight function :

[0138] ,

[0139] This function ensures that nodes with lower energy levels get more consideration.

[0140] Defining the mobility weight function :

[0141] ,

[0142] This function allows node pairs with lower mobility to receive higher scores.

[0143] Calculate the transmission power between nodes based on the comprehensive channel score :

[0144] ,

[0145] In the formula, is the transmission power when node i sends to node j, , are the minimum and maximum transmission powers, To adjust the parameters and control the steepness of the function, is the comprehensive score of the channel between node i and node j, is the threshold parameter.

[0146] Dynamically adjust the node transmission power based on the calculation results. Periodically, such as every 5 seconds, recalculate the channel comprehensive score and transmission power. When the difference between the calculated new transmission power and the current power exceeds a predetermined threshold, such as 10%, perform power adjustment to achieve a smooth transition mechanism and avoid network instability caused by power mutation.

[0147] In multi-hop scenarios, the end-to-end transmission efficiency needs to be considered. Introducing the end-to-end transmission efficiency index :

[0148] ,

[0149] Where n is the number of hops, is the channel score between adjacent nodes, is the corresponding transmission power. By maximizing To optimize the power allocation on multi-hop paths.

[0150] Specifically, in one embodiment of the present invention, step S4 includes:

[0151] S41. Based on historical data and current network status, a predictive routing model is constructed, and machine learning algorithms are used to predict future network topology changes and node mobility.

[0152] Specifically, historical data is collected, including: node location changes, link quality changes, energy consumption, and network topology changes. The current network status is obtained, including the real-time location of the node, link quality, remaining energy, etc. The collected data is preprocessed, including data cleaning to remove outliers and noise; data standardization to normalize features of different scales; and time series processing to convert the data into a format suitable for time series prediction.

[0153] A long short-term memory network or a spatiotemporal graph convolutional network is selected to capture the spatiotemporal dependency of node mobility and link quality. This embodiment implements both algorithms simultaneously and selects the most appropriate one to construct a predictive routing model according to the specific situation.

[0154] The LSTM structure of the long short-term memory network includes: an input layer, which captures features such as node position, speed, direction, and link quality; an LSTM layer, which captures time series features; and a fully connected layer, which makes final predictions.

[0155] The structure of the spatiotemporal graph convolutional network STGCN includes: a spatial convolution layer to capture the spatial dependencies between nodes; a temporal convolution layer to capture time series features; and an output layer to generate the final prediction.

[0156] The model is trained and evaluated. The steps include: 1. Divide the training set and the test set; 2. Define the loss function as mean square error and the optimizer as Adam; 3. Train the model and monitor the training and validation losses; 4. Evaluate the model performance on the test set; 5. Use cross-validation to tune hyperparameters.

[0157] After the model is trained, the current network status, initial network topology, and optimized network topology are input to predict future network topology changes and node mobility. The specific prediction results include at least the distance between nodes at a certain point in the future and the node movement speed at a certain point in the future.

[0158] S42. Define comprehensive routing cost function :

[0159] ,

[0160] In the formula, is the distance between node i and node j predicted at time t, is the predicted moving speed of node j at time t, is the historical link stability between node i and node j, is the residual energy of node j at time t, , , , is the weight coefficient.

[0161] S43. Using the comprehensive routing cost function ,The optimal communication path is calculated by minimizing the cumulative routing cost.

[0162] In this embodiment, an improved Dijkstra algorithm is used to calculate the optimal communication path:

[0163] 1. Initialization: Set the distance of the starting node to 0 and other nodes to infinity; 2. Select the unvisited node with the smallest current distance; 3. Update the distance of adjacent nodes and use the comprehensive routing cost function as edge weight; 4. Repeat steps 2-3 until all nodes are visited; 5. Backtrack to get the optimal path.

[0164] To improve reliability and load balancing, multi-path routing is implemented: use the k-shortest path algorithm to calculate multiple candidate paths; evaluate the comprehensive performance of each path (considering total cost, number of hops, diversity gain, etc.); select the top-k paths as alternatives; implement an adaptive path selection mechanism to dynamically switch paths based on real-time network conditions.

[0165] Recalculate the optimal path regularly (e.g. every 5 minutes); trigger an immediate update when significant network changes are detected (e.g. node failure, new node joining); use routing table version control to ensure that routing information of all nodes is consistent.

[0166] Specifically, in one embodiment of the present invention, step S5 includes:

[0167] S51. Monitor environmental interference and network load in real time, and collect signal-to-noise ratio (SNR) and channel occupancy data.

[0168] Specifically, a real-time monitoring system can be established to continuously collect data, including signal-to-noise ratio (SNR), channel occupancy, interference source strength and type, network load, packet loss rate, and end-to-end delay.

[0169] Implement a distributed perception mechanism so that each node can participate in environmental monitoring: local perception, each node performs environmental monitoring independently; collaborative perception, nodes share perception information to improve perception accuracy; multi-dimensional perception, combining information from the physical layer, MAC layer, and network layer.

[0170] Establish a historical data management system: data storage, use a distributed database to store historical monitoring data; data compression, implement efficient data compression algorithms to reduce storage overhead; data analysis, periodically analyze historical data to identify long-term trends and patterns.

[0171] S52, define adaptive channel selection function ,

[0172] ,

[0173] In the formula, is the channel score for channel k at time t, is the signal-to-noise ratio of channel k at time t, is the occupancy of channel k at time t, is the matching degree of channel k to meet the service quality requirements, , , is the weight coefficient;

[0174] Matching degree of service quality requirements Based on specific application requirements, including bandwidth, latency, and reliability, fuzzy logic methods are used to calculate:

[0175] W k (t) = Fuzzy(Bandwidth k (t),Delay k (t),Reliability k (t));

[0176] In the formula, Fuzzy() represents the fuzzy logic function, Bandwidth represents bandwidth, Delay represents delay, and Reliability represents reliability.

[0177] S53, select the highest The channel with the best value is used as the optimal communication channel for data transmission.

[0178] Specifically, another channel selection strategy can also be selected, such as the ε-greedy strategy: select the best channel with a probability of 1-ε, and randomly explore with a probability of ε, or a learning-based strategy: use a reinforcement learning algorithm, such as Q-learning, to optimize long-term performance.

[0179] In multi-user scenarios, a graph coloring algorithm is used for channel allocation: an interference graph is constructed, where nodes represent users and edges represent interference relationships; a graph coloring algorithm, such as the DSATUR algorithm, is applied to minimize the number of channels required; conflict resolution uses game theory methods to resolve channel allocation conflicts.

[0180] S54. Implement a fast channel switching mechanism to reduce communication interruptions during channel switching.

[0181] Specifically, predictive switching, soft switching or collaborative switching can be used. Predictive switching: prepare backup channels in advance based on the prediction result of step S4; soft switching: maintain the connection between new and old channels simultaneously during the switching process to ensure seamless transition; collaborative switching: adjacent nodes coordinate channel switching to reduce network interruption.

[0182] Define the channel switching cost function:

[0183] ,

[0184] In the formula, is the switching time, is the switching energy consumption, To prevent data loss during the switching process, , , is the weight coefficient. Switching is performed only when the expected benefit is greater than the switching cost: .

[0185] Specifically, in one embodiment of the present invention, before performing adaptive channel selection, it also includes using a collaborative spectrum sensing mechanism to perform preliminary channel screening, the steps are as follows:

[0186] Each node performs local spectrum sensing to detect available channels.

[0187] The method for performing local spectrum sensing in this embodiment is as follows:

[0188] Each node uses energy detection to perform local spectrum sensing:

[0189] E(t)=∑|x(t)| 2 ;

[0190] Among them, E(t) is the detection energy and x(t) is the received signal.

[0191] Judgment criteria:

[0192] H={H0 (channel idle), if E(t)<λ; H1 (channel occupied), if E(t)≥λ}; where λ is a preset threshold.

[0193] Spectrum sensing information is shared between nodes to improve sensing accuracy.

[0194] Define standardized information sharing formats and develop sharing strategies, including:

[0195] Periodic sharing: broadcast the perception results every T seconds;

[0196] Event-triggered sharing: Share immediately when significant changes are detected;

[0197] Request-response mode: Nodes can actively request the perception results of other nodes.

[0198] Use compressed sensing technology to reduce the amount of shared data.

[0199] Define cooperative spectrum sensing function :

[0200] ,

[0201] In the formula, is the collaborative perception score at time t, is the detection probability at time t, is the false alarm probability at time t, is the number of nodes participating in the collaboration at time t, , , is the weight coefficient;

[0202] Choose a higher The channel of the value is used for communication.

[0203] Specifically, the collaborative spectrum sensing mechanism is an efficient channel screening method that improves the accuracy and reliability of spectrum sensing by allowing multiple nodes to participate in the sensing process.

[0204] The present invention introduces a collaborative spectrum sensing mechanism, which improves the accuracy of spectrum sensing by sharing spectrum sensing information between nodes, and realizes more efficient spectrum resource utilization. Each node first performs local spectrum sensing to detect available channels, and then shares this information between nodes. By defining a collaborative spectrum sensing function, the system comprehensively considers the detection probability, false alarm probability and the number of nodes participating in the collaboration, and selects channels with higher collaborative sensing scores for communication. This method not only improves the accuracy of spectrum sensing, but also makes better use of available spectrum resources, reduces channel interference and improves communication efficiency. In the case of tight spectrum resources, this collaborative spectrum sensing mechanism can significantly improve the spectrum utilization and communication performance of the system.

[0205] Specifically, in one embodiment of the present invention, the communication method further includes implementing distributed network congestion control to balance the network load by adjusting the data transmission rate and routing strategy, the steps are as follows:

[0206] Each node independently monitors its queue length and packet transmission delay;

[0207] Define the congestion index CI(t): ; Among them, w 1 ,w 2 ,w 3 is the weight coefficient, Q max and D max are the maximum allowed queue length and delay respectively.

[0208] When congestion is detected, the node alleviates the congestion by reducing the data sending rate and adjusting the routing strategy; specifically, when CI(t) exceeds the threshold θ, the sending rate R(t) is adjusted: ; Where β is the rate adjustment factor.

[0209] Use a credit-based congestion control mechanism to allocate transmission credits based on the node's remaining energy and network contribution;

[0210] Define the node credit value C(t): ; Where, E(t) is the remaining energy of the node, E max is the maximum energy, T(t) is the network contribution of the node, T max is the maximum contribution, and α is the weight coefficient.

[0211] Allocate transmission credit TC(t): Among them, C base is the basic credit and k is the proportional coefficient.

[0212] Using the AODV protocol, the congestion index CI(t) and node credit value C(t) are incorporated into the routing selection process: ; Among them, γ 1 ,γ 2 ,γ 3 is the weight coefficient, and HopCount is the number of hops.

[0213] Congestion information is periodically exchanged in the network to achieve global congestion awareness and coordinated control.

[0214] Periodically broadcast congestion status messages (CSM); implement a distributed congestion graph construction algorithm; use a consensus algorithm to make global congestion control decisions.

[0215] By implementing distributed network congestion control and credit-based congestion control mechanisms, the present invention effectively balances the network load and improves the overall throughput and stability of the network. Each node independently monitors its queue length and data packet transmission delay. When congestion is detected, congestion is alleviated by reducing the data transmission rate and adjusting the routing strategy. The credit-based congestion control mechanism allocates transmission credits according to the node's remaining energy and network contribution, ensuring fair distribution of network resources. Global congestion perception and coordinated control are achieved by periodically exchanging congestion information in the network. This distributed congestion control method not only improves the overall throughput of the network, but also enhances the robustness and scalability of the network, and is particularly suitable for distributed communication systems such as walkie-talkies.

[0216] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A wireless ad hoc network communication method for walkie-talkies, characterized in that: include: S1. Obtaining geographical location information, wireless signal strength information, and remaining energy information of wireless communication nodes, and establishing an initial network topology model based on the acquired information; S2, using low-power self-organizing network protocol to optimize the initial network topology model, select the optimal cluster head, and balance network energy consumption; Step S2 includes: S21. Define the node energy state conversion function: ； In the formula, is the energy level of the node at time t, is the quantum-inspired state of the node at time t, is the energy attenuation coefficient, is the energy acquisition coefficient, M is the number of neighboring nodes, is a function related to the distance of the jth neighboring node, is the energy level of the jth neighboring node at time t, is the energy transfer efficiency coefficient; S22. Initialize the quantum-inspired state of the node , the value range is [0,1]; S23. Define the quantum cost function C for cluster head selection: ； Where N is the total number of nodes, is the residual energy of node i, is the average distance from node i to other nodes in the cluster, is the mobility index of node i, , , To adjust the parameters; S24. Use the quantum annealing algorithm to update the quantum-inspired state of the node. The state update probability p is: ； In the formula, is the cost function difference between the current state and the candidate state, is the quantum bit constant, T is the system temperature; According to the updated quantum-inspired state , calculate the probability of a node becoming a cluster head: P(cluster head)=sin 2 (πQ(t) / 2); S25, determine whether the node becomes a cluster head by generating a random number and comparing it with P (cluster head), and select the best cluster head for non-cluster head nodes; S26, updating node energy level; S27. Calculate the entropy of energy distribution within the cluster: ； In the formula, K is the number of nodes in the cluster, is the energy of node i, is the total energy within the cluster; Dynamically adjust the size and number of clusters according to the entropy value to balance network energy consumption; S28, periodically iteratively executing steps S22-S27 to adapt to dynamic changes in network topology and energy distribution; S3, using an adaptive multi-hop power control algorithm to dynamically adjust the node transmission power, and based on the adjusted transmission power, optimizing the network topology model again; S4. Based on the optimized network topology model, a predictive routing algorithm is used to calculate the optimal communication path that comprehensively considers network topology, node mobility and channel conditions in combination with real-time channel quality information; S5. Allocate the optimal communication path to each wireless communication node. During the data transmission process, perform adaptive channel selection and dynamically select the optimal communication channel based on environmental interference and network load.

2. The wireless ad hoc network communication method for walkie-talkies according to claim 1, characterized in that: The adaptive multi-hop power control algorithm includes: Calculate the channel comprehensive score : ； In the formula, is the comprehensive score of the channel between node i and node j, is the distance between node i and node j, is the path loss exponent, is the signal-to-noise ratio between node i and node j, is the remaining energy of the node, is the energy weight function, is the node mobility index, is the mobility weight function, , , , is the weight coefficient; Calculate the transmission power between nodes : ； In the formula, is the transmission power when node i sends to node j, , are the minimum and maximum transmission powers, To adjust the parameters and control the steepness of the function, is the comprehensive score of the channel between node i and node j, is the threshold parameter; Dynamically adjust the node transmission power according to the calculation results.

3. The wireless ad hoc network communication method for walkie-talkies according to claim 1, characterized in that: Step S4 includes: S41. Based on historical data and current network status, a predictive routing model is constructed, and machine learning algorithms are used to predict future network topology changes and node mobility. S42. Define comprehensive routing cost function : ； In the formula, is the distance between node i and node j predicted at time t, is the predicted moving speed of node j at time t, is the historical link stability between node i and node j, is the residual energy of node j at time t, , , , is the weight coefficient; S43. Using the comprehensive routing cost function ,The optimal communication path is calculated by minimizing the cumulative routing cost.

4. The wireless ad hoc network communication method for walkie-talkies as claimed in claim 3, characterized in that: The machine learning algorithm used in step S41 is a long short-term memory network or a spatiotemporal graph convolutional network to capture the spatiotemporal dependency of node mobility and link quality.

5. The wireless ad hoc network communication method for walkie-talkies according to claim 1, characterized in that: Step S5 includes: S51, real-time monitoring of environmental interference and network load, collecting signal-to-noise ratio (SNR) and channel occupancy data; S52, define adaptive channel selection function : ； In the formula, is the channel score for channel k at time t, is the signal-to-noise ratio of channel k at time t, is the occupancy of channel k at time t, is the matching degree of channel k to meet the service quality requirements, , , is the weight coefficient; S53, select the highest The channel with the best value is used as the optimal communication channel for data transmission; S54. Implement a fast channel switching mechanism to reduce communication interruptions during channel switching.

6. The wireless ad hoc network communication method for walkie-talkies according to claim 5, characterized in that: Matching degree of service quality requirements Based on specific application requirements, including bandwidth, latency, and reliability, fuzzy logic methods are used to calculate: W k (t)=Fuzzy(Bandwidth k (t),Delay k (t),Reliability k (t)); In the formula, Fuzzy() represents the fuzzy logic function, Bandwidth represents bandwidth, Delay represents delay, and Reliability represents reliability.

7. The wireless ad hoc network communication method for walkie-talkies according to claim 5, characterized in that: In step S5, before performing adaptive channel selection, it also includes using a collaborative spectrum sensing mechanism to perform preliminary channel screening, the steps are as follows: Each node performs local spectrum sensing to detect available channels; Nodes share spectrum sensing information to improve sensing accuracy; Define cooperative spectrum sensing function : ； In the formula, is the collaborative perception score at time t, is the detection probability at time t, is the false alarm probability at time t, is the number of nodes participating in the collaboration at time t, , , is the weight coefficient; Choose a higher The channel of the value is used for communication.

8. The wireless ad hoc network communication method for walkie-talkies according to claim 1, characterized in that: The method further includes: implementing distributed network congestion control to balance network load by adjusting data transmission rate and routing strategy, the steps are as follows: Each node independently monitors its queue length and packet transmission delay; When congestion is detected, the node alleviates the congestion by reducing the data sending rate and adjusting the routing strategy; Use a credit-based congestion control mechanism to allocate transmission credits based on the node's remaining energy and network contribution; Congestion information is periodically exchanged in the network to achieve global congestion awareness and coordinated control.

9. The wireless ad hoc network communication method for walkie-talkies according to claim 1, characterized in that: The low-power ad hoc networking protocol also includes a sleep scheduling mechanism to dynamically adjust the sleep cycle of nodes according to network traffic.

Images