A method and system for recording device data transmission under wireless communication

Abstract

The present invention relates to the technical field of data transmission, and particularly to a method and system for recording instrument data transmission under wireless communication. The present invention proposes the following solutions: deploy a terminal cluster, construct a wireless communication network, where each terminal is configured with a recording instrument and selects whether to act as a data relay node according to its own communication capabilities; the recording instrument collects data and selects a channel through the wireless communication network for data transmission; update the wireless communication network according to environmental changes and optimize the data transmission order. Through environmental complexity assessment, dynamic selection of relay nodes and intelligent allocation of channel resources are realized, improving the stability and adaptability of data transmission. It can be widely applied to scenarios such as disaster rescue, industrial monitoring, and robot cooperative communication, effectively improving the efficiency and robustness of wireless communication networks in complex environments.

Description

Technical Field

[0001] The present invention relates to the technical field of data transmission, and particularly to a method and system for transmitting recorder data under wireless communication. Background Art

[0002] In a wireless communication network in a complex environment, traditional wireless communication methods usually rely on fixed topology structures or static channel allocation strategies, and it is difficult to adapt to the dynamic changes of the environment, such as problems like obstacle occlusion, wireless channel fading, changes in interference sources, and node energy consumption limitations. In practical applications, in scenarios such as disaster rescue, industrial monitoring, UAV swarm communication, and underground facility inspection, the instability of the communication link often leads to data loss, increased latency, and even communication interruption. In addition, existing relay node selection methods usually only rely on signal strength or the distance between nodes, and do not fully consider the impact of environmental complexity on signal propagation, resulting in unreasonable relay path selection and further reducing the reliability of the network. Traditional channel allocation methods also have limitations. They mainly rely on real-time measurement of channel states and lack prediction of future channel availability, resulting in frequent channel switching or unbalanced resource allocation. Therefore, there is an urgent need for a method that can adaptively adjust relay node and channel allocation strategies based on environmental complexity assessment to improve the stability of wireless communication and the reliability of data transmission in complex environments.

[0003] The existing technology has the problems raised in the background art: it is difficult to dynamically adapt to environmental changes, resulting in unreasonable relay node selection, inefficient channel allocation, unstable data transmission, and high-priority tasks cannot be guaranteed. To solve the above problems, the present application designs a method and system for transmitting recorder data under wireless communication. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide, in view of the deficiencies of the prior art, a method and system for transmitting recorder data under wireless communication, deploying a terminal cluster, constructing a wireless communication network, where each terminal is configured with a recorder and selects whether to be a data relay node according to its own communication capabilities; the recorder collects data and selects a channel through the wireless communication network for data transmission; the wireless communication network is updated according to environmental changes to optimize the data transmission order. Through environmental complexity assessment, dynamic selection of relay nodes and intelligent allocation of channel resources are realized, improving the stability and adaptability of data transmission. It can be widely applied to scenarios such as disaster rescue, industrial monitoring, and robot collaborative communication, effectively improving the efficiency and robustness of wireless communication networks in complex environments.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for transmitting recorder data under wireless communication, which is applied to terminal cluster cooperative communication in a complex environment. The method for transmitting recorder data includes:

[0007] Deploy a terminal cluster to build a wireless communication network, where each terminal is configured with a recorder and selects whether to be a data relay node according to its own communication ability;

[0008] The recorder collects data and selects a channel through the wireless communication network for data transmission;

[0009] Update the wireless communication network according to environmental changes and adjust the data transmission order.

[0010] The building of the wireless communication network includes:

[0011] Obtain the current environmental complexity and select relay nodes, where each terminal adjusts whether to be a data relay node according to the network topology and the current environmental complexity;

[0012] Build a relay link according to the selected relay nodes, where a data transmission path is built through relay link quality assessment, and the number of communication parameters between relay nodes is dynamically adjusted;

[0013] Perform channel allocation and interference avoidance on the relay link to build a wireless communication network.

[0014] The obtaining of the current environmental complexity and the selection of relay nodes include:

[0015] Obtain the spatial image of the area where the terminal is located through an image sensor;

[0016] Perform texture analysis on the spatial image to calculate an adjustment factor;

[0017] Calculate the obstacle entity features according to the adjustment factor;

[0018] Input the obstacle entity features into a preset obstacle recognition network, process them through the obstacle recognition network, and output the current environmental complexity;

[0019] Each terminal adjusts whether to be a data relay node according to the network topology and the current environmental complexity.

[0020] The performing of texture analysis on the spatial image to calculate an adjustment factor includes:

[0021] Quantify the texture complexity of the neighborhood of pixel points in the spatial image through a gray-level co-occurrence matrix;

[0022] Calculate the gray-level change rate of the neighborhood of each pixel point;

[0023] Calculate an adjustment factor according to the texture complexity and the gray-scale change rate, where the adjustment factor is used to measure the degree of physical change between pixel points.

[0024] Calculating the obstacle entity feature according to the adjustment factor includes:

[0025] Perform block processing on the spatial image to obtain the morphological information of each block;

[0026] Calculate the structure tensor matrix according to the morphological information, where the pixel gradient vector of each block is calculated, and the assignment weight of the structure tensor is calculated according to the adjustment factor;

[0027] Perform eigen-decomposition on the structure tensor matrix to extract the main direction information;

[0028] Perform boundary detection on the pixel points in the block area according to the main direction information, and encode the gradient direction, curvature information and neighborhood statistical features of the boundary point area according to the detection results to generate a feature description vector;

[0029] Aggregate the feature description vectors through max-pooling operations, abstract and aggregate the features according to sparse convolution, and obtain the obstacle entity feature through the RPN network.

[0030] Constructing the relay link according to the selected relay nodes includes:

[0031] Construct an initial relay link through link quality evaluation, where the relay nodes periodically detect the signal strength, link stability and data transmission rate of adjacent nodes, calculate the link quality score, and determine the initial relay link according to the link quality score;

[0032] Send a blank message in the initial relay link, and adjust the communication parameters of the relay nodes between the initial relay links according to the channel state;

[0033] Determine the number of hops in the relay path according to the communication parameters, and construct the relay link.

[0034] Select a channel for data transmission through the wireless communication network, including:

[0035] Construct a channel state database, where each relay node stores the detected channel occupancy, historical interference patterns and channel switching records in the channel state database;

[0036] Filter out idle channels according to the channel state database, and predict the future time slot availability of the idle channels through a preset network model;

[0037] Channel resource allocation is performed according to the future time slot availability, and a transmission path is obtained, where idle channel resources with different priorities are allocated according to the recorder data type.

[0038] The updating of the wireless communication network according to environmental changes includes:

[0039] Updating the wireless communication network topology according to the network state;

[0040] Detecting the current environmental complexity at time intervals and reselecting relay nodes;

[0041] Updating the wireless communication network nodes according to the reselected relay nodes.

[0042] A recorder data transmission system under wireless communication, the system includes a data acquisition module, a network construction module, a channel selection module and a network update module, where:

[0043] The data acquisition module is used to acquire the spatial image and recorder data of the area where the terminal is located;

[0044] The network construction module is used to select relay nodes and construct relay links based on the current environmental complexity and network topology;

[0045] The channel selection module is used to allocate channel resources with different priorities according to the data type;

[0046] The network update module is used to dynamically adjust the wireless communication network according to environmental changes.

[0047] The network construction module includes:

[0048] The relay node confirmation unit is used to select relay nodes according to the current environmental complexity;

[0049] The relay link construction unit is used to construct relay links according to the selected relay nodes;

[0050] The link optimization unit is used to perform channel allocation and interference avoidance on the relay link.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] By dynamically adjusting relay nodes, the present invention improves the stability of the communication link and avoids communication interruption caused by environmental changes; combines the channel state database and the network model to optimize channel selection, improves the anti-interference ability and data throughput of communication; adopts a task priority scheduling strategy to ensure low-latency and highly reliable transmission of high-priority data, and at the same time optimizes the reasonable allocation of bandwidth resources. Description of the Drawings

[0053] Other features, objectives, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments read in conjunction with the accompanying drawings:

[0054] Figure 1 It is a schematic flowchart of a data transmission method of a recorder under wireless communication in Embodiment 1 of the present invention;

[0055] Figure 2 It is a flowchart of constructing a wireless communication network in Embodiment 1 of the present invention;

[0056] Figure 3 It is a schematic flowchart of calculating the current environmental complexity in Embodiment 1 of the present invention;

[0057] Figure 4 It is a module diagram of a data transmission system of a recorder under wireless communication in Embodiment 2 of the present invention. Detailed implementation manners

[0058] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0059] Embodiment 1:

[0060] Please refer to Figure 1 , an embodiment provided by the present invention: A data transmission method of a recorder under wireless communication, which is applied to the collaborative communication of a terminal cluster in a complex environment, and is used to ensure the stable and efficient transmission of recorder data in a complex environment with multiple obstacles, high interference, and dynamic changes. The specific steps are as follows:

[0061] S1: Deploy a terminal cluster and construct a wireless communication network;

[0062] In this embodiment, first, a terminal cluster is deployed in a complex environment (such as underground pipe networks, dense forests, industrial facilities, disaster relief areas, etc.). Each terminal is equipped with a recorder and can autonomously decide whether to act as a data relay node. During the deployment process, the terminal obtains environmental information through an image sensor, and combines texture analysis to calculate an adjustment factor, thereby evaluating the environmental complexity, predicting the signal propagation situation, and determining a suitable data relay node. Subsequently, based on the current environmental complexity and network topology, the terminal uses a link quality evaluation method to establish an initial relay link and dynamically adjusts the communication parameters between relay nodes to ensure that data can still be efficiently transmitted in a complex environment. In this way, the present application realizes the construction of an adaptive wireless communication network, enabling the terminal cluster to adapt to environmental changes and improving the stability and reliability of data transmission.

[0063] S2: The recorder collects data and selects a channel through the wireless communication network for data transmission;

[0064] In this embodiment, the recorder of each terminal continuously collects data information of the surrounding environment, including video streams, sensor data, environmental monitoring data, etc., and classifies the priorities according to the type and urgency of the data. Subsequently, based on the constructed channel state database, the terminal filters idle channels, predicts the future time slot availability of the channels using a network model, and selects the optimal channel through channel hopping technology to ensure that high-priority data can be transmitted under the best channel quality. At the same time, for large-data-volume transmission tasks, this application adopts multi-channel parallel transmission technology to simultaneously distribute data on multiple low-interference channels, improving the throughput and reducing the risk of single-channel congestion. In this way, this application effectively reduces channel interference in complex environments, improves the data transmission rate and stability, and ensures the efficient transmission of critical data.

[0065] S3: Update the wireless communication network according to environmental changes and adjust the data transmission order;

[0066] In this embodiment, the wireless communication network will make adaptive adjustments according to environmental changes to ensure that the data transmission network is always in an optimal state. Specifically, the terminal detects the current environmental complexity at time intervals and reselects relay nodes based on the latest environmental information. When it detects changes in the terminal position, channel attenuation, obstacle blocking, or increased interference, the system will trigger the topology update mechanism of the wireless communication network to re-optimize the data transmission path. In addition, according to the urgency of the data, the data transmission order is dynamically adjusted. For example, emergency control signals and real-time monitoring data will be given priority in transmission, while low-priority data can be buffered or delayed. In the case of network congestion, edge caching technology is used to temporarily store data and prioritize data transmission after the network recovers, reducing data loss and improving transmission efficiency. In this way, this method ensures that in the face of complex environmental changes, the wireless communication network can quickly respond and optimize the data transmission strategy, improving the real-time performance and stability of data transmission.

[0067] In this embodiment, traditional wireless communication network construction methods usually rely on fixed network topologies or simple signal coverage strategies to ensure data transmission between terminals. However, in complex environments (such as disaster sites, underground tunnels, dense industrial areas, forest-covered areas, etc.), due to obstacle occlusion, wireless channel fading, and the existence of dynamic interference sources, it is difficult for traditional network construction methods to ensure communication stability, low latency, and high reliability. For example, in an industrial inspection scenario, due to the dense equipment and complex environmental structure, traditional network construction methods often rely on signal strength to select relay nodes. However, the signal strength is greatly affected by multipath effects and environmental interference, resulting in insufficient dynamic adjustment ability of the communication link, thus affecting the real-time nature of data transmission. In addition, in a robot cluster cooperation task, if the network topology is not updated in a timely manner, data interruption or a decrease in communication efficiency may occur due to the emergence of signal blind spots in certain areas, seriously affecting the execution effect of the cluster task.

[0068] Specifically, this application introduces an environmental complexity assessment mechanism and uses this mechanism to dynamically adjust the selection of relay nodes, the optimization of communication parameters, and the channel allocation strategy to ensure that the wireless communication network can adapt to environmental changes and continuously maintain high-efficient data transmission capabilities. Different from traditional wireless network optimization methods that rely on signal strength or node density, it not only considers the distribution density of obstacles but also comprehensively analyzes multiple factors such as wireless channel attenuation, interference intensity, node energy consumption level, and terminal mobility. For example, in an underground tunnel inspection task, due to the high reflection characteristics of the tunnel wall, signals may experience severe multipath fading in specific areas. However, through environmental complexity calculation, this method can predict high-attenuation areas in advance and actively optimize the layout of data relay nodes to reduce the data packet loss rate and improve communication stability.

[0069] Furthermore, the terminal cluster collects spatial images through image sensors and combines texture analysis and gray-level co-occurrence matrix calculations of obstacle features to construct a more accurate environmental complexity model. For example, in a forest monitoring scenario, traditional wireless communication networks are difficult to predict the impact of tree density on signals. However, this method can calculate the density and arrangement structure of trees through visual analysis and combine channel measurement data for complexity assessment, and finally optimize the distribution of relay nodes to make the wireless communication link have stronger anti-interference capabilities.

[0070] Exemplarily, at a disaster rescue site, the robot terminal cluster needs to transmit survivor detection data in real time. Due to the metal structures and high-density obstacles in the ruins, traditional wireless communication networks may experience communication interruptions or high latency due to multipath effects and severe channel fading. Through environmental complexity assessment, this application calculates the obstacle density in advance and optimizes the data transmission path, enabling the communication network to dynamically adapt to the on-site environment and ensuring the stability of data transmission. At the same time, based on task priority scheduling, the vital sign data of survivors can be preferentially transmitted to the rescue command center, while non-urgent data (such as temperature and humidity information) is transmitted with a delay, improving the response efficiency of rescue tasks.

[0071] Please refer to Figure 2 , the flowchart of constructing a wireless communication network according to an embodiment of the present invention. The specific steps of S1 are as follows:

[0072] S1.1: Obtain the current environmental complexity and select relay nodes. Each terminal adjusts whether it serves as a data relay node according to the network topology and the current environmental complexity.

[0073] In this embodiment, obtaining the environmental complexity is the core link for the self-adaptive adjustment of the wireless communication network, directly affecting the selection of relay nodes, the dynamic optimization of the network topology, and the stability of the data transmission path.

[0074] Specifically, the terminal obtains the surrounding space image through an image sensor and uses the gray-level co-occurrence matrix analysis to calculate the obstacle density, surface roughness, and spatial arrangement, thereby extracting the geometric features and complexity of the obstacles.

[0075] Furthermore, after obtaining the environmental complexity, the terminal adjusts whether it serves as a data relay node according to the network topology and the channel state. Specifically, each terminal calculates its network connectivity within the local area based on the K-nearest neighbor algorithm (KNN) and predicts its signal coverage ability when serving as a relay node in combination with the signal propagation model. Subsequently, the system optimizes the distribution of relay nodes through a reinforcement learning algorithm (such as Q-learning), enabling it to cover signal blind spots and avoid communication redundancy problems caused by excessive nodes participating in relaying. Finally, the adaptive network topology optimization algorithm will combine the calculation results of the environmental complexity and the node distribution to dynamically adjust the data relay strategy to optimize the overall data transmission quality, improve network efficiency, and reduce communication energy consumption.

[0076] Preferably, the present application can also establish a comprehensive environmental complexity calculation model through various means such as ultrasonic sensors, RSSI (Received Signal Strength Indication) measurement, CSI (Channel State Information) analysis, ambient temperature and humidity monitoring, and node energy consumption calculation. Subsequently, in combination with the ultrasonic sensor to measure the material reflectivity, distance, and contour structure of obstacles, the attenuation and reflection effects on wireless signals are analyzed. In addition, the terminal obtains the wireless signal propagation characteristics through RSSI measurement and CSI analysis, including parameters such as channel gain, phase offset, frequency selective fading, and multipath interference. These parameters can be used to calculate the signal attenuation degree and the location of interference sources, improving the accuracy of environmental complexity assessment.

[0077] S1.2: Construct a relay link based on the selected relay nodes, where a data transmission path is constructed through relay link quality assessment, and the number of communication parameters between relay nodes is dynamically adjusted;

[0078] In this embodiment, the construction of the relay link is based on the quality assessment of the selected relay nodes to ensure that the data transmission path can adapt to the dynamic changes of complex environments, improve the transmission efficiency, and reduce the data packet loss rate. To ensure the stability and communication quality of the link, this method uses an adaptive link quality assessment algorithm, combines key technologies such as channel capacity analysis, link stability calculation, dynamic hop count optimization, and energy consumption balancing strategy to construct an optimal relay path.

[0079] Specifically, the channel capacity analysis calculates the channel gain, bandwidth utilization, and interference intensity of the link based on CSI analysis, and combines a channel prediction model (such as LSTM) to calculate the availability trend of the channel in a short period of time; secondly, the link stability calculation measures the round-trip delay of data packets, packet loss rate, and time-varying channel characteristics, establishes a time series model, and predicts the future stability of the link; on this basis, the dynamic hop count optimization strategy continuously adjusts the number of relay nodes in the data transmission path through a reinforcement learning algorithm, reduces unnecessary hops, and thus reduces the data transmission delay.

[0080] Furthermore, this method also dynamically adjusts the communication parameters between relay nodes during the relay link establishment process, including data transmission rate, modulation and coding scheme (MCS), transmit power, and channel bandwidth. For example, in a high-interference environment, the system can reduce the data rate and increase the signal redundancy to ensure that data can be stably transmitted in a harsh environment; while in a low-interference environment, the transmit power can be increased to improve the data throughput, thereby enhancing the overall network performance. In addition, in the case of fewer relay nodes, to reduce the communication delay, this method can dynamically adjust the modulation and coding scheme to improve the data transmission efficiency and ensure that information can be quickly transmitted to the target node.

[0081] S1.3: Perform channel allocation and interference avoidance on the relay link to construct a wireless communication network.

[0082] In this embodiment, channel allocation and interference avoidance are key aspects to ensure the stable operation of a wireless communication network. Especially in the case of high-density terminal deployment and strong interference in complex environments, a reasonable channel allocation strategy can significantly reduce signal conflicts and improve data transmission efficiency.

[0083] Specifically, first, a distributed spectrum sensing technology is adopted. The spectrum sensing modules of each relay node scan for idle channels in the current wireless environment, and a machine learning prediction model (such as LSTM) is used to perform short-term prediction on the channel state and calculate the channel availability score. Combining with the historical data in the channel state database, the system will select channels with lower interference and higher channel stability based on the optimal channel allocation algorithm to ensure low-latency transmission of critical data. In addition, this method adopts a channel hopping mechanism. When it detects that the interference level of the current channel increases or the bandwidth utilization rate is too high, it automatically triggers a channel switching strategy to reduce the data packet loss rate and improve network throughput.

[0084] Furthermore, to further reduce co-channel interference, this method adopts an interference avoidance optimization algorithm based on a graph neural network (GNN). When constructing the topology of the wireless communication network, the system analyzes the spatial distribution relationship of each node in the terminal cluster and calculates the optimal channel allocation scheme, so that neighboring nodes try to use different channels, thereby reducing signal interference. In addition, in specific scenarios, such as an environment where multiple terminals communicate intensively, the system adaptively adjusts the transmission power and channel occupancy time to optimize network resource utilization and reduce unnecessary power consumption and channel competition.

[0085] Please refer to Figure 3 , the schematic diagram of the current environment complexity calculation process of the embodiment of the present invention. The specific steps of S1.1 are as follows:

[0086] S1.1.1: Obtain the spatial image of the area where the terminal is located through an image sensor;

[0087] In this embodiment, in order to accurately perceive the complexity of the environment where the wireless communication network is located, an image sensor is used to collect real-time spatial images of the area where the terminal is located. The image sensor can be an RGB camera, a depth camera, LiDAR (Light Detection and Ranging), or a multi-modal fusion sensor. Among them, the depth camera and LiDAR can provide more accurate three-dimensional environment data, while the RGB camera can provide basic visual information when the computing resources are limited. After the terminal is deployed, it first periodically or triggerly collects images of its surrounding environment. The specific collection frequency is dynamically adjusted according to the task requirements and computing power to ensure the real-time response ability to environmental changes. For example, in an industrial inspection scenario, if the terminal moves inside a closed pipeline, it will collect images at a high frequency to cope with the rapid changes in the pipeline structure, while in an open space, the sampling frequency can be reduced to reduce the computing burden. During the image collection process, in order to improve the stability and accuracy of the data, this method adopts an adaptive exposure control algorithm and an automatic white balance adjustment algorithm to ensure that high-quality image data can be obtained even in an environment with complex lighting conditions. In addition, in order to avoid the influence of dust, fog, etc. on the collection accuracy of the sensor, a multi-frame fusion denoising algorithm can be combined to enhance the acquired images to ensure that the acquired images can accurately reflect the real situation of the environment around the terminal.

[0088] S1.1.2: Perform texture analysis on the spatial image and calculate the adjustment factor;

[0089] In this embodiment, the acquired spatial image needs to be subjected to texture analysis to extract environmental structure features and calculate the adjustment factor. The core objective of texture analysis is to identify the density of obstacles, material types, boundary shapes, surface roughness, etc. These factors will all affect the propagation characteristics of wireless signals. Specifically, this method uses the Gray-Level Co-Occurrence Matrix (GLCM) and Local Binary Pattern (LBP) for texture feature extraction. Among them, GLCM can be used to evaluate features such as the contrast, energy, and uniformity of the image, while LBP is used to capture subtle texture information. In addition, in order to improve the calculation efficiency, an image pyramid downsampling strategy can be adopted to perform hierarchical processing on the image, so that key features can be extracted at different scales. When calculating the adjustment factor, the system will build a preliminary quantitative model of environmental complexity based on indicators such as texture complexity, gradient direction change, and pixel mean deviation. For example, in a forest environment, if there are large-scale high-frequency texture changes (such as tree branches and leaves) in the image, the system will consider that the attenuation of wireless signals in this area is relatively large and assign a higher adjustment factor value in subsequent calculations. The calculation of the adjustment factor not only depends on the image texture, but also combines the terminal historical data, channel measurement results, obstacle shape information, etc. for multi-dimensional optimization to make it more dynamically adaptable.

[0090] S1.1.3: Calculate the obstacle entity features according to the adjustment factor;

[0091] In this embodiment, the physical characteristics of the obstacle are further calculated based on the adjustment factor to accurately describe the impact of the environment on the propagation of wireless signals. The physical characteristics of the obstacle include three-dimensional shape, boundary contour, surface material, spatial distribution, etc. These information can help to more accurately select relay nodes and optimize the signal propagation path during the network construction process. To calculate the three-dimensional shape of the obstacle, this method uses multi-view stereo matching (MVS) or depth-map-based point cloud reconstruction algorithm to generate a three-dimensional model of the environment by fusing image data from multiple different perspectives. In terms of boundary contour recognition, methods such as Canny edge detection and Hough transform for fitting straight lines / curves are used to ensure that the boundaries of the obstacles are clearly distinguishable. In addition, for the surface material characteristics, this method introduces reflectivity modeling, that is, by analyzing the impact of light changes on pixels, the reflection characteristics of the obstacle surface are judged, and then its absorption or scattering ability of wireless signals is inferred. For example, in an underground tunnel environment, if it is calculated that the surface of the obstacle has a high reflectivity (such as a smooth metal surface), it means that the wireless signal may undergo strong multipath reflections, thus affecting the communication quality. Based on this, when constructing the subsequent wireless communication network, the low reflectivity area will be preferentially selected as the data transmission path to reduce the interference of the multipath effect on the signal quality.

[0092] S1.1.4: Input the physical characteristics of the obstacle into a preset obstacle recognition network, process it through the obstacle recognition network, and output the current environmental complexity;

[0093] In this embodiment, the calculated physical characteristics of the obstacle need to be further input into the obstacle recognition network for classification and complexity calculation. This network can adopt a deep neural network (DNN), a convolutional neural network (CNN) or a Transformer architecture, which can efficiently process large-scale environmental data and accurately identify the category of the obstacle, the influence range and the potential impact on the communication channel. For example, in an intelligent traffic management system, this network can distinguish static obstacles (such as buildings, trees) from dynamic obstacles (such as pedestrians, vehicles), and adjust the environmental complexity assessment strategy according to the type of the obstacle. In addition, this network will also perform feature fusion on the input data, that is, combine texture analysis results, depth information, historical data, etc., to finally calculate the environmental complexity. For example, in a disaster rescue scenario, if it is found that the types of obstacles in the environment include a large area of collapsed metal structures, the network will automatically increase the complexity score of this area and send a warning message to the wireless communication system to prompt the system to adjust the network topology to reduce the signal attenuation in this area.

[0094] Preferably, the obstacle recognition network includes an input layer, a feature extraction layer and a feature recognition layer;

[0095] The input layer establishes a feature sequence according to input parameters, and the feature sequence includes a suspected obstacle entity point cloud feature sequence;

[0096] The feature extraction layer is used to extract the feature vectors of the feature sequence. The feature extraction layer includes a geometric feature extraction module and a spatial feature extraction module, and the feature vectors include spatial features and geometric features;

[0097] The feature recognition layer is used to deeply fuse the spatial features and geometric features to generate obstacle fusion point cloud data, and output obstacle information according to the obstacle fusion point cloud data;

[0098] The input layer is used to perform accumulation according to input parameters, establish a grey differential equation, calculate the least square parameters of the grey differential equation, discretize the solution of the least square parameters, calculate the fitting value of the feature sequence, and obtain the suspected obstacle entity point cloud feature sequence;

[0099] The geometric feature extraction module is used to obtain the central point position of the suspected obstacle entity point cloud feature sequence through an attention mechanism, calculate the neighboring points of the central point according to the K-nearest neighbor algorithm, use the central point and the neighboring points as the sources of the Gaussian kernel function, output the local geometric features of the central point through the Gaussian kernel function, and finally map all local geometric features to a high-dimensional space through a residual connection network, perform a pooling operation, and output geometric features;

[0100] The spatial feature extraction module includes a multi-layer perceptron network. In each layer, first, a symmetric operation is performed on the suspected obstacle entity point cloud feature sequence through average pooling, then a spatial description is performed on the suspected obstacle entity point cloud feature sequence through one-dimensional convolution and normalization, and finally, activation is performed according to the ReLU activation function to obtain spatial features;

[0101] The feature recognition layer maps the spatial features and geometric features into a tuple sequence, performs a dot product on the corresponding input features in each candidate sequence in the tuple sequence, calculates the correlation score after the dot product of the candidate sequences through the Pearson correlation coefficient, compresses the correlation score according to the Sigmoid function, compares it with the correlation threshold. If it is greater than the correlation threshold, the candidate sequence is retained. If it is less than or equal to the correlation threshold, the candidate sequence is filtered. The local features of the candidate sequences retained in the tuple sequence are collected and enhanced, and the enhanced candidate sequences are concatenated along the head sequence of the tuple sequence. The local features are fused through a residual connection, and a pooling operation is performed on the feature vectors after the residual connection fusion to obtain the obstacle fusion point cloud data.

[0102] S1.1.5: Each terminal adjusts whether it acts as a data relay node according to the network topology and the current environmental complexity.

[0103] In this embodiment, after each terminal obtains the environmental complexity, it needs to decide whether to act as a data relay node and dynamically adjust its communication parameters. First, the terminal calculates the current channel conditions based on the Link Quality Assessment (LQA) algorithm, including parameters such as signal strength, bandwidth, delay, packet loss rate, etc., and performs a comprehensive calculation with the environmental complexity. If the environmental complexity is high and the terminal channel quality is poor, then this terminal may not be selected as a relay node to avoid affecting the stability of the overall communication link. On the contrary, if a certain terminal is in a low-complexity area and has good signal quality, then this terminal will actively join the relay link and appropriately increase its transmission power to ensure the stable transmission of data. In addition, this method also combines an energy consumption optimization strategy, that is, if a certain terminal has a low battery level, then even if it is in a high-quality signal area, it will not be selected as a relay node to extend the operating life of the entire network. For example, in a robot cluster task, if a certain robot is excluded as a relay node due to low battery power, the system will automatically find a robot with low environmental complexity and sufficient power as an alternative relay, thus ensuring the long-term stable operation of the communication network.

[0104] The specific steps of S1.1.2 are as follows:

[0105] S1.1.2.1: Quantify the texture complexity of the neighborhood of pixel points in the spatial image through a gray-level co-occurrence matrix;

[0106] In this embodiment, in order to accurately evaluate the complexity of the environment where the terminal is located, it is necessary to extract the texture information of the spatial image through an image analysis method. Among them, the gray-level co-occurrence matrix (GLCM) is an effective method for characterizing the texture features of an image. Specifically, the gray-level co-occurrence matrix can reflect the gray-level distribution relationship of pixel points within a certain neighborhood range, describe the spatial dependence between pixel points, and provide texture feature information of the image region. In practical applications, in order to calculate the gray-level co-occurrence matrix, it is first necessary to perform gray-level normalization on the spatial image, map the gray-level values of all pixels to a fixed gray-level range to eliminate the influence of lighting conditions on the image information. Then, set different pixel offset distances and directions (such as horizontal, vertical, diagonal directions) to calculate the co-occurrence matrix of pixel pairs, and form a complete gray-level co-occurrence matrix by statistically counting the frequencies of pixel pairs appearing in each direction. In the calculation process, considering the co-occurrence matrix information in multiple directions can effectively extract texture features at different angles, ensuring that the complexity information of the environment can be accurately obtained under different scenarios and obstacle distributions. The advantage of doing this is that the gray-level co-occurrence matrix can reflect the detailed information, structural changes, and texture complexity of the image region, providing a reliable data basis for subsequent environmental complexity calculations. Compared with traditional methods of evaluating the environment solely based on signal strength or color information, the method based on the gray-level co-occurrence matrix can more accurately identify the characteristics of obstacles. For example, in a forest, underground passage, or dense industrial environment, it can effectively distinguish different types of obstacles (such as trees, pipelines, building walls), improving the accuracy of environmental complexity assessment.

[0107] S1.1.2.2: Calculate the gray-level change rate of the neighborhood of each pixel point;

[0108] In this embodiment, in order to further optimize the evaluation of the environmental complexity, it is necessary to calculate the gray-scale change rate within the neighborhood of each pixel point to measure the brightness change of the local area of the image. The gray-scale change rate is mainly used to identify the gradient change within the image area, depict the edges, structural features of obstacles, and the impact of illumination on the image. In actual calculation, first determine the neighborhood range of the pixel, generally using a window of 3×3, 5×5 or 7×7, then calculate the maximum gray-scale value, minimum gray-scale value and average gray-scale value of the pixel points within this neighborhood, and then calculate the gray-scale change rate of this area. Specifically, the gray-scale change rate can be approximately estimated through the gradient information of the pixel points, that is, calculate the gray-scale difference between adjacent pixel points in the horizontal, vertical and diagonal directions, and take the maximum change value as the gray-scale change rate of the current pixel point. In addition, in order to improve the calculation accuracy, a bilateral filtering technique can be introduced to reduce noise interference when calculating the gray-scale gradient and avoid errors caused by uneven illumination. The purpose of calculating the gray-scale change rate is to extract the edge features and surface texture information of the object, thereby enhancing the ability to identify the boundaries of obstacles. For example, in a disaster rescue scenario, there may be different texture changes at the boundaries of rubble and buildings. By calculating the gray-scale change rate, the density of the rubble can be accurately judged, and then the construction method of the communication network can be optimized to avoid high-attenuation areas for the data transmission path and improve the communication quality. In addition, in the communication environment of a robot cluster, the calculation of the gray-scale change rate can be used to identify the object structure around the robot, ensure that the wireless communication network can adapt to the dynamic changes of the environment, and improve the stability and anti-interference ability of data transmission.

[0109] S1.1.2.3: Calculate an adjustment factor according to the texture complexity and the gray-scale change rate, where the adjustment factor is used to measure the degree of entity change between pixel points.

[0110] In this embodiment, the calculation of the adjustment factor is the core step of environmental complexity assessment. Its purpose is to measure the degree of entity change between pixel points, so as to classify different environmental regions and adjust weights. When calculating the adjustment factor, first use the gray-level co-occurrence matrix to obtain the global texture features of the image, and at the same time extract the local gradient information by combining the gray-level change rate. Then, through a weighted calculation method, combine the two to generate a comprehensive environmental complexity score. Specifically, the calculation of the adjustment factor considers multiple factors, including the texture fineness of the local area, the gray-level gradient change trend, the density of obstacle distribution, etc., among which the calculation parameters of different environments can be adaptively adjusted through weight factors. For example, in a smooth wall area, due to the low gray-level change rate and low texture complexity, the value of the adjustment factor will be small, while in an area with dense rubble, due to the high gray-level change rate and significant texture complexity, the value of the adjustment factor will be large, so as to give a higher weight in the subsequent optimization process of the wireless communication network. In addition, in the calculation of the adjustment factor, the local mean filtering method can also be combined to smooth the noise impact of the local area and make the calculated adjustment factor more stable. In terms of application, the adjustment factor can be widely used to optimize the topological structure of the wireless communication network, select the best data relay path, optimize the channel selection strategy, etc. For example, in the UAV relay communication in a forest environment, the adjustment factor can be used to identify areas with high tree density, so that the network can avoid high-interference areas, select the optimal signal propagation path, and improve the overall communication quality. At the same time, in the application of underground tunnel inspection, the adjustment factor can be used to identify the material characteristics of the tunnel wall and optimize the signal reflection area to reduce the impact of multipath effects on wireless communication. By calculating the adjustment factor, the wireless communication network can adaptively adjust network parameters according to different environments to ensure the reliability, stability and efficiency of communication.

[0111] The specific steps of S1.1.3 are as follows:

[0112] S1.1.3.1: Perform block processing on the spatial image to obtain the morphological information of each block;

[0113] In this embodiment, in order to analyze environmental information more accurately, first perform block processing on the obtained spatial image, adopt an adaptive segmentation strategy, so that the blocks can dynamically adjust their sizes according to environmental characteristics, rather than fixed division.

[0114] Specifically, adopt a method of adaptive adjustment based on image gradient change and texture complexity to perform local feature analysis on the entire image, and combine the adjustment factor to calculate the optimal block size, so that the blocks can more accurately capture the structural characteristics of different regions in a complex environment. This adaptive block method can use larger blocks in areas with uniform environments, and smaller blocks in areas with dense obstacles or complex signals to enhance the capture ability of key structural information.

[0115] Furthermore, during the block division process, an optimization strategy based on pixel connectivity analysis is adopted to ensure that the feature information between adjacent blocks does not break or get lost, thereby improving the accuracy of subsequent feature extraction. This processing method can not only improve the adaptability of block division, but also reduce computational redundancy, improve data processing efficiency, and ensure the accuracy of obstacle feature extraction, providing more accurate input data for subsequent steps.

[0116] S1.1.3.2: Calculate the structure tensor matrix according to the morphological information, where the pixel gradient vectors of each block are calculated, and the assignment weights of the structure tensor are calculated according to the adjustment factor;

[0117] In this embodiment, first, for the image area after block division, the gradient vector of each pixel point is calculated to obtain the local structure information of the image. The Sobel operator or Prewitt operator is used to calculate the gradient vector to extract the gradient information in the horizontal and vertical directions respectively, and two-way filtering is combined to enhance the edge gradient stability and reduce noise interference. After obtaining the gradient vector, a structure tensor matrix is further constructed to represent the direction change of the pixel gradient. The structure tensor matrix is calculated based on the local gradient covariance and can capture the gradient distribution characteristics of the area around each pixel point, thereby reflecting the obstacle morphology in the spatial image. To improve the adaptability of the structure tensor matrix to complex environments, an adjustment factor is introduced as a weighting parameter, and different weights are assigned to different regions during the calculation process to ensure more accurate feature extraction for key regions. For example, in areas with dense obstacles, the adjustment factor will be assigned a higher weight, enabling the tensor matrix to focus more on the detailed information of these regions, while a lower weight is assigned to the background area to reduce computational redundancy. In this way, the structure tensor matrix can more effectively capture the local morphological features of the spatial image and improve the accuracy of subsequent feature extraction.

[0118] S1.1.3.3: Perform eigen - decomposition on the structure tensor matrix to extract the main direction information;

[0119] In this embodiment, the structure tensor matrix is subjected to eigenvalue decomposition to calculate the eigenvalues and the corresponding eigenvectors, and the direction corresponding to the largest eigenvalue is selected as the main direction information. During the calculation process, the magnitude of the eigenvalue characterizes the change intensity of the local gradient, while the corresponding eigenvector represents the main direction distribution of the region. To improve the robustness of the calculation, the tensor matrix is regularized before eigenvalue decomposition to reduce the calculation deviation caused by noise or non-uniform illumination. In addition, to adapt to the directional features in different scenarios, an adaptive direction selection mechanism is introduced, that is, when calculating the main direction information, not only the features of the current block are considered, but also its neighborhood features are combined for global optimization to ensure the smoothness and continuity of the main direction information. This method is particularly important in scenarios where obstacles are densely arranged, such as inside buildings or forest-covered areas, where the arrangement direction of obstacles can be more accurately extracted, improving the accuracy of subsequent boundary detection and feature extraction.

[0120] S1.1.3.4: Perform boundary detection on the pixel points in the block region according to the main direction information, and encode the gradient direction, curvature information, and neighborhood statistical features of the boundary point region according to the detection results to generate a feature description vector;

[0121] In this embodiment, the Canny edge detection algorithm based on gradient direction voting is used for boundary detection. Combining the main direction information enables the edge detection to focus on regions with clear structures while suppressing noise interference. In addition, after boundary detection, deeper feature analysis is performed on the boundary point region to extract features including gradient direction, local curvature information, and neighborhood statistical features to form a more discriminative feature description vector. The gradient direction information is mainly used to characterize the direction change of the boundary points, while the curvature information is used to measure the morphological features of the boundary, such as whether it presents regular straight lines, curves, or complex polygons. The neighborhood statistical features are calculated by computing the distribution of pixel points within their local range, such as mean, variance, contrast, etc., to enhance the ability to identify the obstacle boundary region. This multi-level feature encoding method enables the system to effectively identify obstacles in different complex environments and provides accurate input data for subsequent obstacle modeling.

[0122] S1.1.3.5: Aggregate the feature description vectors through max-pooling operations, abstract and aggregate the features according to sparse convolution, and obtain the obstacle entity features through the RPN network.

[0123] In this embodiment, a max pooling operation is performed on the feature description vector to retain the most significant feature information, reduce redundant data, and improve computational efficiency. Max pooling can effectively enhance the robustness of features, making them more robust to scale changes and environmental noise. Next, sparse convolution is used for feature aggregation to further abstract the feature information of obstacles. The advantage of sparse convolution is that it can reduce the computational complexity while maintaining an efficient representation of obstacle features, especially suitable for complex environments with unevenly distributed obstacles. Finally, using the Region Proposal Network (RPN), the obstacle features are further optimized, and the obstacle entity features are generated. The RPN network can generate candidate regions of potential obstacles based on the extracted feature information, and further optimize the bounding boxes through classification and regression, making obstacle recognition more accurate. In this way, accurate detection and modeling of obstacles in complex environments can be achieved. Even when obstacles are stacked, densely arranged, or there are occlusions, the accuracy of recognition can be guaranteed, providing reliable data support for subsequent optimization of wireless communication networks.

[0124] The specific steps of S1.2 are as follows:

[0125] S1.2.1: Construct an initial relay link through link quality assessment, where the relay node periodically detects the signal strength, link stability, and data transmission rate of adjacent nodes, calculates the link quality score, and determines the initial relay link according to the link quality score;

[0126] In this embodiment, to ensure the stability and efficiency of the wireless communication network in complex environments, the system adopts a link quality assessment method based on multi-parameter fusion. By periodically monitoring the communication status between relay nodes, it quantifies the transmission capacity of the network link and selects the optimal relay path based on a comprehensive score. Specifically, the relay nodes regularly broadcast probe signals, exchange information with neighboring nodes around them, and perform operations such as signal strength detection, link stability calculation, and data transmission rate measurement on each possible link. Among them, the signal strength detection is mainly measured through RSSI (Received Signal Strength Indication) to evaluate the attenuation of the link; the link stability calculation is based on the packet loss rate, round-trip delay (RTT), and signal fluctuation range within a period of time for statistics to obtain the stability score of the link quality; the data transmission rate measurement is to analyze the maximum available bandwidth between nodes through throughput testing to evaluate the transmission capacity of the link. Subsequently, the system performs weighted summation on the above indicators based on the hierarchical weighted scoring algorithm to form a link quality score (LQS, Link Quality Score), and sorts all possible links according to the score, and selects several links with the highest scores as the initial relay links. The advantage of this method is that it not only considers the traditional signal strength but also introduces the stability and data transmission ability of the link, making the selected relay path more reliable, thus ensuring the continuity of data transmission in complex environments and avoiding link failure problems caused by signal fluctuations or short-term interference. In addition, this method can dynamically adapt to environmental changes. Once the quality score of a certain link drops to a certain threshold, the system will automatically trigger the network topology update mechanism to ensure the continuous optimization of the network.

[0127] S1.2.2: Send a blank message in the initial relay link and adjust the communication parameters of the relay nodes between the initial relay links according to the channel state;

[0128] In this embodiment, to further improve the adaptability of the relay link and the reliability of data transmission, after the initial relay link is established, the system dynamically detects the channel state using dummy packets and adjusts the communication parameters of the relay node according to the channel state. A dummy packet is a detection packet that does not contain actual service data. Its function is to test the transmission capacity of the link in the current environment and avoid the influence of channel interference on the transmission of real data. Specifically, the relay node sends dummy packets to adjacent nodes at a preset time interval and records key indicators such as the packet loss rate, delay, jitter, and channel occupancy rate of the packet in the link. Among them, the packet loss rate reflects the reliability of the channel, the delay and jitter can be used to determine whether there are bursty interferences in the link, and the channel occupancy rate is used to measure the available bandwidth of the channel. The system optimizes the communication parameters, including the data transmission rate, adaptive modulation and coding (AMC) method, transmit power adjustment, and retransmission mechanism, based on these measurement data. For example, when a large channel interference is detected, the system can reduce the data transmission rate and adopt a stronger forward error correction coding method to improve the anti-interference ability of the data; when sufficient channel resources are detected, the system can increase the transmission rate to improve the data throughput. In addition, if the packet loss rate of a certain link exceeds the set threshold, the system will trigger the adaptive channel hopping mechanism to switch the data transmission to a channel with less interference. The advantage of this method is that it can dynamically optimize the communication parameters based on the real-time channel conditions, thereby effectively reducing channel congestion and data loss, and improving the stability and reliability of data transmission, especially suitable for complex environments with high interference and dynamic changes. In addition, the use of dummy packets can also avoid the problem of affecting the system stability due to the failure of service data transmission in traditional testing methods, making the channel state evaluation more accurate.

[0129] S1.2.3: Determine the number of hops in the relay path according to the communication parameters and construct a relay link.

[0130] In this embodiment, in order to further optimize the data transmission path and improve the overall efficiency of the communication network, the system dynamically adjusts the number of hops in the relay path based on the communication parameter adaptive optimization algorithm according to the channel state and node communication capabilities in the relay link to construct an optimal relay link. Specifically, after the system completes the channel parameter adjustment, it calculates the maximum supportable number of hops for each node based on the feedback information of the blank message, that is, the maximum number of hops that the node can stably forward data under the current channel conditions. If the signal quality of a certain relay node is poor, the system will reduce the number of forwarding times of this node or even skip this node to reduce the bit error rate and delay of the overall link. On the other hand, if the signal quality of some relay nodes is high, the number of forwarding times of this node can be increased to make full use of its communication capabilities and improve the throughput of data transmission. In addition, the system will also comprehensively consider the time to live (TTL) of the data packet to ensure that the data will not be forwarded in the network too many times to prevent waste of network resources and data redundancy. In addition, the system also uses the path entropy optimization algorithm to comprehensively evaluate multiple candidate paths and selects the path with the lowest entropy value as the final relay link to ensure the stability and low latency of the data transmission path. The advantage of this method is that it can dynamically optimize the data transmission path based on the actual channel state, which can not only reduce unnecessary hop counts, improve the data transmission rate, but also ensure the stable transmission of data in a complex environment, especially suitable for highly dynamic wireless communication networks. In practical applications, for example, in the communication network of disaster rescue robots or the industrial equipment inspection network, due to many environmental interference factors, the traditional relay link with a fixed number of hops is difficult to adapt to the changing network requirements, while this method can significantly improve the reliability and flexibility of data transmission by adaptively adjusting the number of hops.

[0131] The specific steps of S2 are as follows:

[0132] S2.1: Construct a channel state database, where each relay node stores the detected channel occupancy, historical interference patterns, and channel switching records in the channel state database;

[0133] In this embodiment, the construction of the channel state database is a key link to ensure the stability of the wireless communication network. Since the channel state in a complex environment is affected by dynamic environmental changes, such as the movement of obstacles, the change of electromagnetic interference, the change of device power consumption, etc., the traditional fixed channel selection method is difficult to adapt to these real-time changes. Therefore, this application proposes a dynamic management method based on the channel state database to ensure the accuracy of channel allocation. Specifically, each relay node will monitor the usage of the current channel in real time during communication, including the bandwidth occupancy rate of the channel, the interference intensity of the channel, the historical availability of the channel, etc. These data are updated by means of periodic sampling to ensure the timeliness of the channel state database. In addition, the relay node will also store historical interference patterns. For example, the channels in certain time periods or specific locations are strongly interfered by the outside world. The system will analyze the change trend of interference through statistical methods to provide a basis for predicting the future availability of the channel. At the same time, when the task stage or data traffic demand changes, the relay node may switch channels, and these switching information will also be stored in the channel state database to form a complete channel usage record for subsequent optimization decisions. The advantage of this is that the channel state database can not only reflect the usage of the current channel, but also predict the future channel state based on historical data and interference patterns, avoiding transmission failures caused by real-time channel fluctuations, and improving the adaptability and stability of the wireless communication network.

[0134] S2.2: Screen idle channels according to the channel state database, and predict the future time slot availability of the idle channels through a preset network model;

[0135] In this embodiment, the screening of idle channels based on the channel state database is a key technical link for optimizing the channel selection strategy. Due to the dynamic change characteristics of channel interference and occupancy in complex environments, simply relying on the current channel state for selection will lead to short-term fluctuations affecting the transmission effect. Therefore, this application proposes a method for predicting the availability of future time slots based on the channel state database, so as to achieve more intelligent channel allocation. First, in the process of screening idle channels, the system comprehensively considers factors such as the bandwidth occupancy rate, interference intensity, data packet loss rate, and signal quality of the current channel, and eliminates channels that are not suitable for data transmission based on the set threshold, initially screening out a set of idle channels. Subsequently, the system calls a preset network model, which is based on machine learning or statistical analysis methods, and uses historical channel data to train the network model, so as to infer the availability of each channel in the future time slot. Specifically, the network model combines environmental complexity information, analyzes external factors that may affect the channel state, such as whether obstacles move, whether new interference sources appear, the usage of wireless devices, etc., and predicts the availability score of each channel in the future time slot based on multi-factor analysis. It can not only screen channels according to the current state, but also predict the usage of channels in advance, so as to preferentially select future stable channels in the case of tight channel resources, and avoid communication interruption or data loss caused by frequent channel switching.

[0136] S2.3: Perform channel resource allocation according to the availability of the future time slot to obtain a transmission path, where idle channel resources with different priorities are allocated according to the data type of the recorder;

[0137] In this embodiment, channel resource allocation and transmission path optimization are important links to ensure the efficient operation of a wireless communication network. Due to the diverse data transmission requirements in complex environments, for example, the recorder data may include emergency control signals, real-time monitoring data, low-priority storage data, etc. Therefore, simply selecting an idle channel based on the channel state is still insufficient. It is also necessary to perform priority allocation in combination with the data type to ensure that critical task data can be transmitted under the optimal channel quality. Specifically, when performing channel resource allocation, the system will first select a group of channels that are most suitable for the current network state according to the prediction results of future time slot availability, and allocate channel resources based on the priority of the recorder data. For example, the emergency control signal has the highest priority, and it is necessary to ensure its low-latency and high-reliability transmission. Therefore, the system will preferentially allocate channels with the least interference and optimal bandwidth for such data; for real-time monitoring data, such as video streams or environmental sensor data, the system will select channels with larger throughput to ensure the stability of data transmission; for low-priority storage data, such as historical record data, the remaining channels can be used for transmission, or even a delayed transmission strategy can be adopted to optimize the overall network resource utilization. In addition, in the case of tight channel resources, the system can also adopt dynamic bandwidth allocation technology, that is, share the available bandwidth among multiple data streams, so that high-priority tasks can obtain more bandwidth resources, while low-priority tasks are reasonably rate-limited to ensure the overall communication efficiency. In this way, the present application can reasonably allocate channel resources in a complex environment, not only ensuring the stability of data transmission, but also improving the transmission reliability of high-priority tasks, while maximizing the utilization of existing wireless channel resources and enhancing the efficiency and adaptability of the entire wireless communication network.

[0138] The specific steps of S3 are as follows:

[0139] S3.1: Update the wireless communication network topology according to the network state;

[0140] S3.2: Detect the current environmental complexity at time intervals and reselect relay nodes;

[0141] S3.3: Update the wireless communication network nodes according to the reselected relay nodes.

[0142] Embodiment 2:

[0143] Please refer to Figure 4 , the present invention provides an embodiment: A recorder data transmission system under wireless communication, the system includes a data acquisition module, a network construction module, a channel selection module, and a network update module, wherein:

[0144] The data acquisition module is used to obtain the spatial image and recorder data of the area where the terminal is located;

[0145] The network construction module is used to select relay nodes and construct relay links based on the current environmental complexity and network topology;

[0146] The channel selection module is used to allocate channel resources with different priorities according to the data type;

[0147] The network update module is used to dynamically adjust the wireless communication network according to environmental changes.

[0148] The network construction module includes:

[0149] The relay node confirmation unit is used to select relay nodes according to the current environmental complexity;

[0150] The relay link construction unit is used to construct relay links according to the selected relay nodes;

[0151] The link optimization unit is used to perform channel allocation and interference avoidance on the relay link.

[0152] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for transmitting recorder data under wireless communication, which is applied to terminal cluster collaborative communication, and is characterized in that, The data transmission method of the recorder includes: Deploy a terminal cluster, where each terminal is configured with a recorder and an image sensor. Collect the spatial image of the area where the terminal is located through the image sensor, and calculate an adjustment factor based on the texture features of the spatial image to determine the current environmental complexity; The terminal selects whether to be a data relay node according to the network topology and the current environmental complexity. If it is a data relay node, connect the relay link to build a wireless communication network; The recorder collects data and selects a channel through the wireless communication network for data transmission; Update the wireless communication network according to environmental changes and adjust the transmission order of data; The determination of the current environmental complexity includes: Perform texture analysis on the spatial image and calculate an adjustment factor; Calculate the obstacle entity features according to the adjustment factor; Input the obstacle entity features into a preset obstacle recognition network, process them through the obstacle recognition network, and output the current environmental complexity; The performing texture analysis on the spatial image and calculating an adjustment factor includes: Quantify the texture complexity of the pixel neighborhood in the spatial image through a gray-level co-occurrence matrix; Calculate the gray-level change rate of each pixel neighborhood; Calculate an adjustment factor according to the texture complexity and the gray-level change rate, where the adjustment factor is used to measure the entity change degree between pixel points; Selecting a channel for data transmission through the wireless communication network includes: Build a channel status database, where each relay node stores the detected channel occupancy, historical interference patterns, and channel switching records in the channel status database; Filter idle channels according to the channel status database and predict the future time slot availability of the idle channels through a preset network model; Perform channel resource allocation according to the future time slot availability to obtain a transmission path, where different priorities of idle channel resources are allocated according to the recorder data type.

2. The data transmission method of the recorder under wireless communication according to claim 1, wherein The building of the wireless communication network includes: Obtain the current environmental complexity and select relay nodes, where each terminal adjusts whether it is a data relay node according to the network topology and the current environmental complexity; Build a relay link according to the selected relay nodes, where a data transmission path is built through relay link quality evaluation, and the number of communication parameters between relay nodes is dynamically adjusted; Perform channel allocation and interference avoidance on the relay link to build a wireless communication network.

3. The data transmission method of the recorder under wireless communication according to claim 1, wherein, The calculating of the obstacle entity features according to the adjustment factor includes: Perform block processing on the spatial image to obtain the morphological information of each block; Calculate the structure tensor matrix according to the morphological information, where the pixel gradient vectors of each block are calculated, and the assignment weight of the structure tensor is calculated according to the adjustment factor; Perform eigen-decomposition on the structure tensor matrix to extract the main direction information; Perform boundary detection on the pixel points in the block area according to the main direction information, and encode the gradient direction, curvature information, and neighborhood statistical features of the boundary point area according to the detection results to generate a feature description vector; Aggregate the feature description vectors through max-pooling operations, abstract and aggregate features based on sparse convolution, and obtain obstacle entity features through the RPN network.

4. The method for transmitting recorder data under wireless communication according to claim 2, wherein The construction of the relay link according to the selected relay nodes includes: Construct an initial relay link through link quality evaluation, where the relay nodes periodically detect the signal strength, link stability, and data transmission rate of adjacent nodes, calculate the link quality score, and determine the initial relay link according to the link quality score; Send blank packets in the initial relay link and adjust the communication parameters of the relay nodes in the initial relay link according to the channel state; Determine the number of hops in the relay path according to the communication parameters and construct the relay link.

5. The data transmission method of the recorder under wireless communication according to claim 1, characterized in that, The update of the wireless communication network according to environmental changes includes: Update the topology of the wireless communication network according to the network state; Detect the current environmental complexity at time intervals and re-select relay nodes; Update the nodes of the wireless communication network according to the re-selected relay nodes.

6. A recorder data transmission system under wireless communication is used to implement a recorder data transmission method under wireless communication as described in any one of claims 1-5, characterized in that, The system includes a data acquisition module, a network construction module, a channel selection module, and a network update module, where: The data acquisition module is used to obtain the spatial image and recorder data of the area where the terminal is located, calculate the adjustment factor based on the texture features of the spatial image to determine the current environmental complexity. Among them, the determination of the current environmental complexity includes: Perform texture analysis on the spatial image and calculate the adjustment factor; Calculate the obstacle entity features according to the adjustment factor; Input the obstacle entity features into a preset obstacle recognition network, process them through the obstacle recognition network, and output the current environmental complexity; The performing texture analysis on the spatial image and calculating the adjustment factor includes: Quantify the texture complexity of the neighborhood of pixel points in the spatial image through a gray-level co-occurrence matrix; Calculate the gray-level change rate of the neighborhood of each pixel point; Calculate the adjustment factor according to the texture complexity and the gray-level change rate, where the adjustment factor is used to measure the entity change degree between pixel points; The network construction module is used to select relay nodes and construct relay links based on the current environmental complexity and network topology; The channel selection module is used to allocate channel resources with different priorities according to the data type, including: Construct a channel state database, where each relay node stores the detected channel occupancy, historical interference patterns, and channel switching records in the channel state database; Filter idle channels according to the channel state database and predict the future time slot availability of the idle channels through a preset network model; Perform channel resource allocation according to the future time slot availability to obtain a transmission path, where different priorities of idle channel resources are allocated according to the recorder data type; The network update module is used to dynamically adjust the wireless communication network according to environmental changes.

7. The data transmission system of a recorder under wireless communication according to claim 6, characterized in that, The network construction module includes: A relay node confirmation unit for selecting relay nodes according to the current environmental complexity; A relay link construction unit for constructing a relay link according to the selected relay nodes; A link optimization unit for performing channel allocation and interference avoidance on the relay link.

Images