Resource management method for distributed wireless sensor network

Abstract

The invention relates to a resource management method for a distributed wireless sensor network. The method comprises the following steps: acquiring channel information for communication between ad hoc network intelligent sensing devices in a power transmission line scene, and a resource conflict relationship between channels; the resource conflict relationship comprises a power interference relationship and a spectrum overlapping relationship; constructing a conflict hypergraph based on the channel information and the resource conflict relationship; and based on an adjacent matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in a preset time period, the number of residual available resources in the power transmission line scene and the resource conflict degree between the channels, obtaining resource allocation strategies corresponding to the plurality of channels through a resource allocation model which is trained in advance, the resource allocation strategy is used for carrying out resource allocation on a plurality of channels. By adopting the method, the problem of interference between sensor nodes can be solved.

Description

Technical Field

[0001] The present application relates to the technical field of wireless sensor networks, and in particular to a resource management method for distributed wireless sensor networks. Background Art

[0002] In recent years, my country has attached great importance to digital development and has started to vigorously promote the construction of "Digital China", which has put forward higher requirements for the application of digital technologies, including intelligent sensing, in the power industry. In June 2017, the grand vision of the next generation of smart grids, "transparent grids", was proposed. The future of the power grid is a transparent grid, which effectively integrates sensor technology, information technology, data communication technology, electronic control technology, artificial intelligence, etc. in the power system, and can realize the transparency of equipment status, operation status, and transaction status.

[0003] At present, the main online monitoring systems use many discrete and single sensors to monitor the operation status of the power system, which cannot achieve effective integration and information sharing at the sensor level, resulting in slow development of technologies such as multi-parameter information fusion sharing and comprehensive diagnosis, which cannot meet the development requirements of smart grids. The self-organizing network intelligent sensor device provides watt-level energy output through multiple induction power modules, ensuring that the self-organizing network communication device is online in real time, and adopts miniaturized integration, multi-physical quantity synchronous detection, full-scale measurement, and on-demand analog / digital format technical solutions to integrate sensitive components, signal conditioning circuits, microprocessors and communication interface modules, etc., with the intelligent trend of multiple environmental physical quantity perception and sensor information fusion. The intelligence and informatization of smart grids require the configuration of a large number of sensors and data acquisition modules. How to organically integrate sensors and corresponding data acquisition and transmission systems to form a multi-physical quantity integrated sensor array, and use multi-dimensional information comprehensive analysis technology to perform efficient status analysis and fault diagnosis on equipment is a hot research direction in the current digital grid construction.

[0004] At present, the geographical location and environmental conditions of transmission lines are special, with many tower points, long lines, and wide coverage. They are exposed in the wild all year round. In addition to being attacked by natural disasters such as strong winds, lightning strikes, and heavy rains, they are often affected by human factors. In recent years, with the rapid development of urban and rural economies, the total mileage of China's transmission lines has rapidly increased from 1.15 million kilometers to more than 1.59 million kilometers from 2014 to 2020. Long-distance overhead transmission lines in complex climates and environments are prone to abnormal faults such as wildfires, tree barriers, and external damage. Tripping incidents caused by external force damage occur frequently, seriously affecting the safe operation of transmission lines. The contradiction between the rapid growth of the mileage of power grid operation and maintenance lines and the relatively insufficient number of power grid operation and maintenance personnel has gradually emerged. In order to effectively prevent the safety hazards of external damage to transmission lines during construction, the transmission line operation and maintenance units have carried out special patrols for key lines and key time periods. This year, they have covered the key sections of the transmission line channels through video / image monitoring terminals, realizing remote online patrols of safety hazards in the transmission line channels. At present, many regions have achieved video surveillance coverage of most transmission lines, and many regions have achieved full line video coverage. The key sections of the 500kV and above lines in the network have also been basically fully covered. Through online video inspections, effective prevention of safety hazards such as important line construction has been achieved, and the efficiency of transmission line operation and maintenance has been improved. With the advancement of smart grid construction and the development of artificial intelligence technology, transmission line inspections have transitioned from traditional "human inspections" to "human inspections + drone inspections". However, the flight capabilities of drones are limited by weather and battery life. At the same time, there are also problems such as limited monitoring to images, high technical requirements for operators, short patrol distances, and the impact of no-fly zones. The monitoring needs of transmission lines cannot be fully and real-time covered; there is also the problem of how to achieve data fusion of multiple integrated sensors and long-distance transmission back to the server.

[0005] Therefore, in order to further improve the observability of power grid operation and enhance the information perception capability of transmission lines, it is necessary to study multi-sensor integration technology to realize all-round and multi-angle monitoring of transmission lines. By studying multi-physical quantity integrated sensor technology, massive deployment of integrated sensor elements can be realized; multi-physical quantity integrated sensor distributed self-organizing network data communication technology can be studied to complete the networking communication between multiple multi-physical quantity integrated sensor devices; multi-physical quantity data analysis, feature extraction technology and auxiliary decision-making technology of integrated sensors can be studied, multi-physical quantity data fusion application system can be developed, and dynamic capacity expansion monitoring software modules for transmission lines can be developed to promote the application of artificial intelligence technology in massive monitoring data of transmission lines, solve the actual needs of operation and maintenance of transmission lines, realize high-efficiency and low-latency real-time transmission of measurement data to the server for processing and real-time monitoring of the working status of transmission lines, so as to achieve the goal of automation, informatization and digitization of intelligent monitoring of transmission lines, and ultimately achieve the goal of transparent perception of power grids.

[0006] In recent years, with the rapid growth of the number of wireless smart terminals in smart grids and the increasing demand for mobile data traffic, wireless sensor networks (WSNs) have become an indispensable part of transmission line monitoring systems. In transmission line monitoring, the application of wireless multi-physics integrated sensors is not limited to the measurement of traditional electrical parameters such as current and voltage, but also extends to temperature, humidity, wind speed, vibration, and corrosion. Such comprehensive monitoring helps to timely detect potential fault points, such as insulator aging, wire loosening or overheating, so as to take preventive maintenance measures to avoid accidents. In addition, such sensors usually have low power consumption design, suitable for long-term deployment in field environments, reducing the need for frequent battery replacement. Due to the self-organizing characteristics of its distributed nodes, wireless sensor networks are widely used in remote areas with limited network resources. Among them, self-organizing network technology is one of the keys to achieving efficient wireless sensor networks. It allows sensor nodes to automatically establish connections and form networks without a central controller, and can quickly adapt and maintain stable service quality even when the network topology changes. This capability is crucial to improving the robustness and flexibility of the entire system, especially in the face of natural disasters or other emergencies that cause damage to part of the infrastructure. However, with the increasing density of sensor nodes and the continuous increase in communication frequency, spectrum resources are becoming increasingly scarce, and the interference problem caused by spectrum reuse is becoming increasingly serious. This poses a major challenge to wireless sensor networks in providing low-latency and high-reliability services. In order to solve the interference problem between sensor nodes in distributed wireless sensor networks, an effective resource management method is urgently needed. Summary of the invention

[0007] Based on this, it is necessary to provide a resource management method for a distributed wireless sensor network to solve the interference problem between sensor nodes in response to the above technical problems.

[0008] In a first aspect, the present application provides a resource management method for a distributed wireless sensor network, comprising:

[0009] Obtain channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as resource conflict relationships between channels; resource conflict relationships include power interference relationships and spectrum overlap relationships;

[0010] Construct a conflict hypergraph based on channel information and resource conflict relations;

[0011] Based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio of each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between channels, the resource allocation strategies corresponding to multiple channels are obtained through a pre-trained resource allocation model. The resource allocation strategies are used to allocate resources to multiple channels.

[0012] In one embodiment, the step of obtaining the resource conflict relationship between the channels includes:

[0013] For any pair of channels, a first power interference amount is determined based on the transmission power of the ad hoc network intelligent sensor device on the first channel on the first channel and the distance between the ad hoc network intelligent sensor devices on the first channel and the second channel; a second power interference amount is determined based on the transmission power of the ad hoc network intelligent sensor device on the second channel on the second channel and the distance; if at least one of the first power interference amount and the second power interference amount meets the power interference condition, it is determined that there is a power interference relationship between the first channel and the second channel;

[0014] Determine a spectrum overlap amount based on a center frequency and a bandwidth of the first channel and a center frequency and a bandwidth of the second channel; if the frequency overlap amount satisfies a spectrum overlap condition, determine that there is a spectrum overlap relationship between the first channel and the second channel;

[0015] At least one of the power interference relationship and the spectrum overlap relationship is used as the resource conflict relationship between the first channel and the second channel.

[0016] In one embodiment, the training process of the resource allocation model includes:

[0017] Obtain the sample adjacency matrix of the sample conflict hypergraph, the sample signal-to-noise ratio corresponding to each channel in a preset time period, the number of sample available resources, and the degree of sample resource conflict, and obtain the predicted resource allocation strategy corresponding to each of the multiple channels through the initial neural network model;

[0018] Calculate the model loss based on the empirical resource allocation strategy, the predicted resource allocation strategy, and the preset loss function;

[0019] The model parameters of the initial neural network model are adjusted based on the model loss until the preset stop condition is met to obtain a trained resource allocation model.

[0020] In one embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until a preset stop condition is met to obtain a trained resource allocation model, including:

[0021] Obtaining a reward value corresponding to the predicted resource allocation strategy based on a preset reward function;

[0022] The model parameters of the initial neural network model are adjusted based on the reward value and the model loss until the preset stop condition is met to obtain a trained resource allocation model.

[0023] In one embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until a preset stop condition is met to obtain a trained resource allocation model, including:

[0024] Obtaining parameter optimization conditions; the parameter optimization conditions include at least one of maximum resource throughput of each channel, conflict-free resource allocation, transmission power of each sensor device being within a preset power range, and at most one resource block being allocated to each channel;

[0025] Based on the model loss and parameter optimization conditions, the model parameters of the initial neural network model are adjusted until the preset stop conditions are met to obtain a trained resource allocation model.

[0026] In one embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until a preset stop condition is met to obtain a trained resource allocation model, including:

[0027] Obtaining value data corresponding to the predicted resource allocation strategy based on a preset action value function;

[0028] The model parameters of the initial neural network model are adjusted based on the value data and model loss until the preset stopping conditions are met to obtain a trained resource allocation model.

[0029] In one embodiment, constructing a conflict hypergraph based on channel information and resource conflict relations includes:

[0030] The channels indicated by each channel information are taken as vertices, and the resource conflict relationship between each channel is taken as an edge; a hyperedge is determined based on channels having a resource conflict relationship between each other, and a conflict hypergraph is constructed based on each vertex, edge and hyperedge.

[0031] In a second aspect, the present application also provides a resource management device for a distributed wireless sensor network, comprising:

[0032] An acquisition module is used to acquire channel information for communication between self-organizing network intelligent sensor devices in a transmission line scenario, as well as resource conflict relationships between channels; resource conflict relationships include power interference relationships and spectrum overlap relationships;

[0033] A construction module, used for constructing a conflict hypergraph based on channel information and resource conflict relations;

[0034] The allocation module is used to obtain the resource allocation strategies corresponding to multiple channels based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio of each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between channels through a pre-trained resource allocation model. The resource allocation strategy is used to allocate resources to multiple channels.

[0035] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0036] Obtain channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as resource conflict relationships between channels; resource conflict relationships include power interference relationships and spectrum overlap relationships;

[0037] Construct a conflict hypergraph based on channel information and resource conflict relations;

[0038] Based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio of each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between channels, the resource allocation strategies corresponding to multiple channels are obtained through a pre-trained resource allocation model. The resource allocation strategies are used to allocate resources to multiple channels.

[0039] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0040] Obtain channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as resource conflict relationships between channels; resource conflict relationships include power interference relationships and spectrum overlap relationships;

[0041] Construct a conflict hypergraph based on channel information and resource conflict relations;

[0042] Based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio of each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between channels, the resource allocation strategies corresponding to multiple channels are obtained through a pre-trained resource allocation model. The resource allocation strategies are used to allocate resources to multiple channels.

[0043] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0044] Obtain channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as resource conflict relationships between channels; resource conflict relationships include power interference relationships and spectrum overlap relationships;

[0045] Construct a conflict hypergraph based on channel information and resource conflict relations;

[0046] Based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio of each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between channels, the resource allocation strategies corresponding to multiple channels are obtained through a pre-trained resource allocation model. The resource allocation strategies are used to allocate resources to multiple channels.

[0047] The resource management method, device, computer equipment, computer-readable storage medium and computer program product for distributed wireless sensor networks obtain channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as the resource conflict relationship between each channel, and construct a conflict hypergraph based on the channel information and the resource conflict relationship. Since the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship, the adjacency matrix of the conflict hypergraph can be used to characterize the conflict situation between each channel, thereby converting the conflict problem between channels into the vertex coloring problem of the conflict hypergraph, which is beneficial to reducing the computational complexity; based on the adjacency matrix of the conflict hypergraph, the corresponding signal-to-noise ratio of each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel, a resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model. The resource allocation strategy is used to allocate resources to multiple channels, which is beneficial to reducing signal overlap interference in densely distributed sensor equipment scenarios and improving the reliability of signal transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 is an application environment diagram of a resource management method for a distributed wireless sensor network in one embodiment;

[0050] Figure 2 is a flow chart of a resource management method for a distributed wireless sensor network in one embodiment;

[0051] Figure 3A schematic diagram of a WSN network framework in one embodiment;

[0052] Figure 4 A schematic diagram of a communication network topology in one embodiment;

[0053] Figure 5 is a schematic diagram of a conflict hypergraph in one embodiment;

[0054] Figure 6 is a convergence diagram for different proportions of users in one embodiment;

[0055] Figure 7 is a convergence diagram under different numbers of convergence rounds in one embodiment;

[0056] Figure 8 is a schematic diagram of maximum network throughput in one embodiment;

[0057] Fig. 9 is a schematic diagram of average network throughput in one embodiment;

[0058] Fig.10 is a structural block diagram of a resource management device for a distributed wireless sensor network in one embodiment;

[0059] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0061] The resource management method for a distributed wireless sensor network provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 obtains the channel information for communication between the self-organizing network intelligent sensor devices in the transmission line scenario, and the resource conflict relationship between the channels; the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship; based on the channel information and the resource conflict relationship, a conflict hypergraph is constructed; based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in the preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel, the resource allocation strategy corresponding to each of the multiple channels is obtained through the pre-trained resource allocation model, and the resource allocation strategy is used to allocate resources to multiple channels. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device may be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0062] In an exemplary embodiment, Figure 2 As shown, a resource management method for a distributed wireless sensor network is provided, and the method is applied to Figure 1 The terminal 102 in the example is used as an example to illustrate, including the following steps 202 to 206. Among them:

[0063] Step 202, obtaining channel information for communication between self-organizing network intelligent sensor devices in a power transmission line scenario, and resource conflict relationships between channels; resource conflict relationships include power interference relationships and spectrum overlap relationships.

[0064] Among them, the self-organizing network intelligent sensor device refers to a sensor device that integrates various types of parameter measurement functions in the transmission line scenario. For example, it can integrate the measurement of traditional electrical parameters such as current and voltage, and also expand to temperature, humidity, wind speed, vibration, and corrosion degree. The self-organizing network intelligent sensor device adopts a technical solution of miniaturization integration, synchronous detection of multiple physical quantities, full-scale measurement, and on-demand analog / digital format. It integrates sensitive components, signal conditioning circuits, microprocessors, and communication interface modules. It has the intelligent trend of sensing multiple environmental physical quantities and sensor information fusion. Such comprehensive monitoring helps to timely discover potential fault points, such as aging of insulators, loose or overheated wires, so as to take preventive maintenance measures to avoid accidents. In addition, this type of sensor usually has a low-power design, which is suitable for long-term deployment in field environments, reducing the need for frequent battery replacement.

[0065] Channel information refers to information about a channel for communication between smart sensor devices in a self-organizing network. The channel information may indicate a channel used for communication.

[0066] like Figure 3 The figure shows a schematic diagram of a WSN network framework in an embodiment. For example, in a WSN network model in a dense urban scenario, multiple sensor nodes SN use S2S communication to establish a self-organizing network, and multiple sensor nodes SN use S2I communication to interact with a base station (BS) so as to send the collected information to a cloud server and accept the resource allocation strategy calculated by the cloud server.

[0067] The communication channel between sensor nodes is denoted as SL. The interference to the channel SL mainly comes from the background noise and the signal from the SN assigned to the same channel. The signal conforms to Rayleigh fading, with an average value of μ and a path loss factor of α. Therefore, the signal to interference plus noise ratio (SINR) value of the mth SL can be defined as:

[0068]

[0069] where Pm and Pn represent the transmission power of the mth and nth sensor devices, respectively, and σ 2 is the constant noise power, h m The power gain of the channel corresponding to the mth S2S pair, N m is the set of adjacent conflicting users of user m, and h n,m is the interference power gain from the nth sensor device, N S2S represents the number of S2S pairs. Therefore, the available rate of the mth S2S pair at time t can be calculated as:

[0070]

[0071] Where B is the channel bandwidth.

[0072] Based on directed graph theory, a directed graph is constructed according to the communication relationship between SNs in the WSN network, and the directed graph is represented as an adjacency matrix, such as Figure 4 FIG. 1 is a communication network topology diagram in an embodiment. Each node in the communication network topology diagram represents a SN, and each edge represents a SL. In the figure, there are 11 SNs and 14 SLs in the communication network topology diagram, where SNs are represented as {S1, S2, ..., S 11}, and SL is represented as {SL1,SL2,…,SL 14}. The adjacency matrix of the communication network topology graph can be expressed as:

[0073]

[0074] The communication network topology can be represented as a graph in is a collection of nodes, and is a collection of SL. Existence n ={(v i ,v j ):v i ,v j ∈e n}, where the sequence pair (v i ,v j ) is determined by the order of the element, indicating that from v i to v j For clarity, and Respectively represented as set e j The first and second elements of .

[0075] for Figure 4 In the communication network topology diagram, the communication relationship between SNs and the transmission direction of SL can be represented by the adjacency matrix G T express.

[0076]

[0077] in

[0078]

[0079] Resource conflict relationship refers to the conflict of resource blocks allocated between channels. Due to the limited communication resources densely deployed in WSN networks, the close proximity of a large number of S2S pairs may cause interference caused by different power levels between pairs sharing the same RB within a specific range. Therefore, resource conflict relationship includes power interference relationship, which is used to characterize the channel resource conflict relationship caused by the interference caused by the transmission power of the device. When certain power interference conditions are met, it is determined that there is a power interference relationship between channels. The power interference relationship can be determined by comparing the signal power with the path loss model.

[0080] At the same time, when spectrum resources are limited, spectrum overlap may occur between different channels, resulting in conflicts. Therefore, the resource conflict relationship also includes the spectrum overlap relationship, which is used to characterize the spectrum overlap between channels. This spectrum overlap interference may aggravate resource conflicts and have an adverse effect on the overall performance of the network. Therefore, it is crucial to analyze and model the relationship between SLs in the WSN network and allocate different RBs to the S2S links that cause interference. When certain spectrum overlap conditions are met, it is determined that there is a spectrum overlap relationship between channels.

[0081] Step 204: construct a conflict hypergraph based on the channel information and resource conflict relations.

[0082] Among them, the embodiment of the present application proposes to construct a conflict hypergraph based on two types of SLs, which is used to accurately describe the channel conflict relationship in the WSN network. In order to reduce the complexity of analyzing the conflict hypergraph, the maximum clique theory can be used to simplify the hyperedges of the conflict hypergraph, thereby reducing the dimension of the association matrix representing the conflict hypergraph. Finally, the resource allocation problem is solved by solving the vertex coloring problem of the conflict hypergraph. In some embodiments, the channels indicated by the channel information can be used as vertices of the conflict hypergraph, the resource conflict relationship between each channel can be used as the edge of the hypergraph, and the channels that have resource conflict relationships between each other can determine the hyperedge.

[0083] Step 206, based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between the channels, a resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model, and the resource allocation strategy is used to allocate resources to the multiple channels.

[0084] Among them, the adjacency matrix refers to the matrix composed of the association relationship between each node in the oil conflict hypergraph. Figure 5 FIG. 1 is a schematic diagram of a conflict hypergraph in one embodiment. The adjacency matrix of the conflict hypergraph can be expressed as follows:

[0085]

[0086] The signal-to-noise ratio refers to the signal-to-interference-plus-noise ratio (SINR) of each channel SL. The number of remaining available resources refers to the number of all available resource blocks currently remaining in the transmission line scenario. The resource conflict level refers to the conflict level between the densely deployed channels in the entire WSN network.

[0087] The resource allocation model refers to a pre-trained neural network model. In some embodiments, it can be an Agent based on a DDQN (Double Deep Q-Network) algorithm.

[0088] Exemplarily, the terminal uses the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel within a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel as inputs to the resource allocation model, and obtains resource allocation strategies corresponding to multiple channels output by the resource allocation model.

[0089] The resource allocation strategy is used to allocate resource blocks to multiple channels to avoid interference problems between sensor nodes.

[0090] In the above-mentioned resource management method for distributed wireless sensor networks, by acquiring channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario and the resource conflict relationship between each channel, a conflict hypergraph is constructed based on the channel information and the resource conflict relationship. Since the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship, the adjacency matrix of the conflict hypergraph can be used to characterize the conflict situation between each channel, thereby converting the conflict problem between channels into the vertex coloring problem of the conflict hypergraph, which is beneficial to reducing the computational complexity; based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel, the resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model. The resource allocation strategy is used to allocate resources to multiple channels, which is beneficial to reducing signal overlap interference in the densely distributed sensor equipment scenario and improving the reliability of signal transmission.

[0091] In an exemplary embodiment, the step of acquiring the resource conflict relationship between each channel includes: for any pair of channels, based on the transmission power of the self-organizing network intelligent sensor device on the first channel on the first channel, and the distance between the self-organizing network intelligent sensor devices on the first channel and the second channel, determining the first power interference amount; based on the transmission power of the self-organizing network intelligent sensor device on the second channel on the second channel, and the distance, determining the second power interference amount; if at least one of the first power interference amount and the second power interference amount meets the power interference condition, it is determined that there is a power interference relationship between the first channel and the second channel; based on the center frequency and bandwidth of the first channel, and the center frequency and bandwidth of the second channel, determining the spectrum overlap amount; if the frequency overlap amount meets the spectrum overlap condition, it is determined that there is a spectrum overlap relationship between the first channel and the second channel; and at least one of the power interference relationship and the spectrum overlap relationship is used as the resource conflict relationship between the first channel and the second channel.

[0092] The resource conflict relationship between the channels can be described by taking the resource conflict relationship between any pair of channels as an example. Wherein, any pair of channels includes a first channel and a second channel.

[0093] The first power interference amount is used to characterize the power interference amount caused by the transmission power of the self-organizing network intelligent sensor device on the first channel. In some embodiments, the first power interference amount can be expressed as follows: Among them, the self-organizing network intelligent sensor device SN i In Channel SL i The transmission power on i , Self-organizing network intelligent sensor device SN i and SN j The distance between them is d ij , α is the path loss exponent.

[0094] The second power interference amount is used to characterize the power interference amount caused by the transmission power of the self-organizing network intelligent sensor device on the second channel. In some embodiments, the second power interference amount can be expressed as follows: Among them, the self-organizing network intelligent sensor device SN j In Channel SL j The transmission power on j .

[0095] The power interference condition may be a condition satisfied by the power interference amount. If at least one of the first power interference amount and the second power interference amount satisfies the power interference condition, it is determined that there is a power interference relationship between the first channel and the second channel. In some embodiments, the power interference condition may be expressed as follows:

[0096]

[0097] or,

[0098]

[0099] The above formula indicates that if the power interference level between channels exceeds a preset SINR threshold, it is determined that a power interference relationship exists between the channels.

[0100] The frequency overlap is used to characterize the overlap of the spectrum between channels. In some embodiments, the spectrum overlap can be expressed as: max(0, min(f i +B i / 2,f j +B j / 2)-max(f i -B i / 2,f j -B j / 2)), where channel SL i The center frequency is f i , bandwidth is B i , channel SL j The center frequency is f j , bandwidth is B j .

[0101] In some embodiments, the spectrum overlap condition can be expressed as:

[0102] max(0,min(f i +B i / 2,f j +B j / 2)-max(f i -B i / 2,f j -B j / 2))>OL

[0103] Wherein, OL represents the spectrum overlap threshold.

[0104] If the spectrum overlap between channels is greater than 0, it means that there is an overlap between channels, otherwise, it means that there is no overlap. The above formula indicates that if the spectrum overlap between channels is greater than the spectrum overlap threshold, it means that there is a spectrum overlap relationship between channels, otherwise, there is no spectrum overlap relationship.

[0105] If the channels have any one of a power interference relationship and a spectrum overlap relationship, it is determined that the channels have a resource conflict relationship. Therefore, at least one of the power interference relationship and the spectrum overlap relationship can be used as the resource conflict relationship between the first channel and the second channel.

[0106] In this implementation, for any pair of channels, based on the transmission power of the self-organizing network intelligent sensor device and the distance between the self-organizing network intelligent sensor devices on the channel pair, it is determined whether there is a power interference relationship between the channels, which is conducive to reflecting the impact of the power interference level on the resource conflict between the channels; based on the center frequency and bandwidth of the channel, it is determined whether there is a spectrum overlapping relationship between the channels, which is conducive to reflecting the impact of the spectrum overlapping situation on the resource conflict between the channels. The resource conflict relationship includes at least one of the power interference relationship and the spectrum overlapping relationship, which is conducive to accurately reflecting the resource conflict situation between the channels.

[0107] In an exemplary embodiment, the training process of the resource allocation model includes: obtaining a sample adjacency matrix of a sample conflict hypergraph, a sample signal-to-noise ratio corresponding to each channel within a preset time period, a number of sample available resources, and a degree of sample resource conflict, and obtaining a predicted resource allocation strategy corresponding to each of the multiple channels through an initial neural network model; calculating the model loss based on the empirical resource allocation strategy, the predicted resource allocation strategy, and a preset loss function; and adjusting the model parameters of the initial neural network model based on the model loss until the preset stop condition is met to obtain a trained resource allocation model.

[0108] Among them, in dense WSN network deployment, resource allocation may bring instability to the DDQN training process. This instability has an adverse effect on the speed and accuracy of the SL agent, mainly due to the large number of decisions required for resource management of sensor devices. To alleviate this situation, FedAvg (Federated Averaging) can be integrated with DDQN to form the FDDQN framework, that is, the initial neural network model can adopt the FDDQN framework. FDDQN combines the advantages of centralized and distributed learning to improve the efficiency and performance of the model. In this framework, the central server coordinates a group of clients, denoted as M. The FDDQN training process will iterate multiple communication rounds to obtain a trained resource allocation model.

[0109] The terminal uses the sample adjacency matrix, the sample signal-to-noise ratio corresponding to each channel in the preset time period, the number of sample available resources, and the degree of sample resource conflict as the input of the initial neural network model, and obtains the predicted resource allocation strategy corresponding to each of the multiple channels output by the initial neural network model. Among them, the empirical resource allocation strategy is used as a training label to calculate the model loss with the predicted resource allocation strategy through the preset loss function. The initial neural network will adjust the model parameters based on the model loss until the preset stop condition is met to obtain a trained resource allocation model.

[0110] Exemplarily, the FDDQN training process iterates the following steps over communication rounds from 1 to T:

[0111] 1. Global model propagation: The model training process requires a lot of data interaction. In order to minimize communication overhead and improve training efficiency, after initializing the global model, a sample subset S is randomly selected in each round based on multiple rounds of experiments, where the number of selected samples is S·M. The initialized global model is then distributed to these selected subsets.

[0112] 2. Mode: Each selected subset receives an initialized global model and uses the DDQN algorithm based on the MDP model to determine the training parameters for the mth local model. Exemplarily, the goal is to minimize the loss function by gradient descent, which is defined as:

[0113]

[0114] 3. Model aggregation: After t rounds of training, each agent uploads its trained local model to the server. The server then performs weighted average aggregation, which is expressed as:

[0115]

[0116] The above steps are repeated until the FDDQN global model converges.

[0117] The details of the process are summarized in the following FDDQN-based conflict-free resource allocation algorithm.

[0118] Requirements: learning rate α, discount factor γ, experience replay memory capacity.

[0119] Ensure: Main Q network parameters, target Q network parameters.

[0120] 1. Random initialization: The server randomly initializes the online network parameters and sets the parameter w1 = w2. This step is to ensure the consistency of the initial state.

[0121] 2. Client selection: In each iteration, Rc clients are randomly selected from all clients to participate. Here, it is assumed that there are SM clients in total.

[0122] 3. Client local update: For each selected client m belonging to [Rc], perform the following operations to calculate the new parameter value

[0123] 4. Local update end mark: After all clients complete local updates, they continue to the next round of iteration.

[0124] 5. Initialize the preprocessing model: This step is to prepare for subsequent learning tasks.

[0125] 6. Reset state: For each S2S agent m, reset the vehicle state information s0 and the network state. This step is to ensure that the state is consistent at the beginning of each training.

[0126] 7. Multi-step training: For each time step t, use the ε-greedy strategy to select the action and obtain the reward function R and the next state s'. The ε-greedy strategy can balance the relationship between exploration and utilization.

[0127] 8. Store experience: Store the experience of each interaction in the experience player, represented as {s,a,r,s'}.

[0128] 9. Sampling experience: Randomly extract a small batch of experience data from the experience player. These experience data are used to calculate the loss function

[0129] 10. Update the main network parameters: Update the parameters wmt of the main network according to the calculated loss function.

[0130] 11. Update target network parameters: Periodically update the parameters wt of the target network.

[0131] 12. End mark: When the specified number of training times is reached or other stopping conditions are met, the entire training process ends.

[0132] Repeat the above steps until the Fedavg-DDQN global model converges. To consider the computational complexity of this process, two aspects need to be considered: the model training of the DDQN algorithm used by a single S2S agent and the weighted average aggregation of parameters of the Fedavg server. The complexity of a single DDQN network depends on its complexity TDDQN and the number of local training K lt Therefore, the local model complexity of DDQN with local updates can be expressed as O((T D3QN )×K lt ). In addition, the complexity of model aggregation grows linearly with the number of agents P, which is O(P). Therefore, the total complexity of the FedAvg-DDQN algorithm can be expressed as:

[0133]

[0134] In this embodiment, by taking the sample adjacency matrix, sample signal-to-noise ratio, sample available resource quantity, and sample resource conflict degree as inputs of the initial neural network model, and taking the empirical resource allocation strategy as a training label, the initial neural network is trained to obtain a trained resource allocation model. The resource allocation model can be used to allocate resources for channels, which is beneficial to avoiding conflict interference problems between sensor nodes.

[0135] In an exemplary embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until a preset stop condition is met to obtain a trained resource allocation model, including: obtaining a reward value corresponding to the predicted resource allocation strategy based on a preset reward function; adjusting the model parameters of the initial neural network model based on the reward value and the model loss until the preset stop condition is met to obtain a trained resource allocation model.

[0136] The preset reward function is a function for calculating the reward value, and the reward value is used to reward the predicted resource allocation strategy. The effectiveness of the model training can be determined by maximizing the reward value, so that conflict-free and efficient resource allocation can be performed for dense deployment in the WSN network.

[0137] The setting of the preset reward function needs to consider the maximization of throughput and the minimization of constraint conflicts. In some embodiments, the preset reward function is expressed as follows:

[0138]

[0139] Among them, λ1,λ2,λ3∈(0,1) represents the weight of each reward in the total reward, k represents the total number of resources, Indicates the number of RBs used at a specific time, C f represents the conflict factor, which introduces a penalty into the total reward function.

[0140] After each output prediction of the resource allocation strategy, the reward value is calculated, so that the model is trained based on the reward value and model loss, which is conducive to ensuring the effectiveness of the model.

[0141] Exemplarily, in order to provide a clear, consistent, and computable framework for the model, it is necessary to construct the corresponding Markov decision process (MDP), including the state space, action space, reward function, and strategy.

[0142] State Space: In a resource allocation environment that supports dense deployment within a WSN network, each S2S link is considered as an agent. At each time period t, the model learns to make the best choice through interaction with the environment and trial and error. In particular, the agent observes the current environment and chooses the best action. Based on this choice, the model receives corresponding reward feedback, and then the environment changes again and a new round of cycles begins. Assume that the observation results of the nth S2S link are as follows:

[0143]

[0144] in, Indicates the number of remaining available resources in the WSN network, G H represents the adjacency matrix of the conflict hypergraph, represents the set of SINRs of all SLs at time k, It indicates the conflict degree of dense deployment in the whole WSN network.

[0145] Action space: The action taken by the model at time step t based on the state space analysis environment is expressed as:

[0146]

[0147] in, represents the selection of channel nodes, and Indicates the selection of a resource block.

[0148] Reward function: Once in state r t If you select action A(t) in the game, you will get a reward r according to the reward function t It is worth noting that one of the advantages of rewards is the ability to determine the effectiveness of model training by maximizing the reward value. This allows for conflict-free and efficient resource allocation for dense deployments within WSN networks.

[0149] In this embodiment, after each output of the predicted resource allocation strategy, the reward value is calculated by using a preset reward function, so that the model is trained based on the reward value and the model loss, which is conducive to ensuring the effectiveness of the model.

[0150] In an exemplary embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until the preset stop condition is met to obtain a trained resource allocation model, including: obtaining parameter optimization conditions; the parameter optimization conditions include at least one of maximum resource throughput of each channel, conflict-free resource allocation, the transmission power of each sensor device being within a preset power range, and at most one resource block being allocated to each channel; the model parameters of the initial neural network model are adjusted based on the model loss and the parameter optimization conditions until the preset stop condition is met to obtain a trained resource allocation model.

[0151] In order to ensure efficient communication in dense deployment within the WSN network, the throughput of the SL link needs to be as large as possible. The embodiment of the present application achieves this by allocating resources in dense deployment within the WSN network to the SL as a hypergraph node coloring problem. In some embodiments, the optimization problem is defined as the following four parameter optimization conditions:

[0152]

[0153] in, Where k indicates that the kth resource block is allocated to the mth S2S link, stC f =0 indicates the condition for achieving conflict-free resource allocation. Indicates that at most one resource block can be selected for allocation to the SL, i.e. each channel is allocated at most one resource block. represents maximizing the resource throughput of each channel, Indicates that the transmission power of each sensor device is within the preset power range.

[0154] The terminal will adjust the model parameters of the initial neural network model based on the model loss and parameter optimization conditions until the preset stop conditions are met to obtain a trained resource allocation model.

[0155] In this embodiment, by determining the parameter optimization conditions, model training is performed based on the model loss and the parameter optimization conditions, which is conducive to ensuring that the channel throughput is as large as possible, thereby ensuring efficient communication.

[0156] In an exemplary embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until a preset stop condition is met to obtain a trained resource allocation model, including: obtaining value data corresponding to the predicted resource allocation strategy based on a preset action-value function; adjusting the model parameters of the initial neural network model based on the value data and the model loss until the preset stop condition is met to obtain a trained resource allocation model.

[0157] Among them, after converting the proposed conflict-free resource allocation problem for dense deployment within the WSN network into an MDP, reinforcement learning (RL) can be used to solve the MDP. In this case, SL is set as an agent. Q-learning is a model-free algorithm that uses the Bellman equation to iteratively update the value function. This is achieved by continuously selecting actions A and updating the action-value function Q through the state space S. Finally, the optimal action-value function Q is learned, thereby achieving the optimal strategy. In the dense deployment within the WSN network, SL resource allocation is performed according to the allocation strategy π, which can be expressed as a preset action-value function Q(s,a) according to the Bellman equation, which is defined as:

[0158]

[0159] Among them, γ t represents the discount factor, and the initial state and action are represented by S0 and A0 respectively.

[0160] The terminal can calculate the value data corresponding to the predicted resource allocation strategy based on the preset action value function, and then perform model training based on the value data and model loss.

[0161] In some embodiments, the terminal can also continuously select the optimal action according to the optimal strategy, using π * Indicates that the optimal action value function Q can be obtained *, which represents the optimal long-term return. It is expressed as follows:

[0162]

[0163] Since determining the optimal action-value function requires evaluating many actions for each state recorded in the Q-table, the Bellman equation can be used to simplify this process by performing an iterative calculation. * It is defined by the Bellman equation as follows:

[0164]

[0165] In dense deployment scenarios within WSN networks, model-free algorithms (such as Q-learning) are often used, which makes it difficult to determine the transition probabilities. To address this problem, the time difference (TD) method is used to solve the Q * Value estimation problem. In some embodiments, the update process for each time step is defined as follows:

[0166] Q(S t ,A t )←Q(S t ,A t )+α[R t+1 +Q(S t+1 ,A t+1 )-Q(S t ,A t )]

[0167] Where α is the learning rate. However, in dense WSN network deployment, the large-scale state-action space makes it difficult to store the Q table, and a deep neural network (DNN) is introduced to iteratively update the weight w through training. m To approximate Q * In some embodiments, this approximation in Deep QNetworks (DQN) is mathematically represented as:

[0168] y t =r t +γq(s t+1 ,argmax a q(s t+1 ,a;w m );w m )

[0169] In DQN, the action value function Q updates its estimate with the currently estimated action value. Due to the correlation of the above phenomena, the dense deployment of agents within the WSN network makes them prone to overestimating bootstrapping during the learning process. This leads to problems of instability and inefficient learning. In order to avoid bootstrapping and implement the DDQN algorithm, a target network can be incorporated. In some embodiments, the algorithm is defined as follows:

[0170] y t =r t +γq(s t+1 ,argmax a q(s t+1 ,a;w m );w t )

[0171] Among them, w t Represents the training network parameters. Both the DDQN training network and the target network exist. The former selects actions, and the latter estimates the selected actions based on the current state.

[0172] In order to further reduce the correlation of agent training in the dense deployment of resource allocation in WSN networks and achieve better training results, an experience replay mechanism can be introduced. The SL agent reads the current state and calculates the optimal Q value Q * Choose the corresponding optimal action a * In addition, in order to avoid falling into the local optimal solution, each SL agent adopts the ε greedy mechanism to select the optimal action a with a probability of 1-ε * and obtain the corresponding reward R. This then enables the transition to the next state st+1, completing a round and transferring the experience E=(s,a,r,s t+1 ) is stored in the experience replay memory and marked as This process is called experience replay [X]. Once the experience replay memory is full, a small batch of size The samples are used for training. The loss function is then calculated by continuously adjusting the estimate of the Q-value function through the TD method, with the goal of approximating the optimal Q-value function, which in some embodiments can be defined as:

[0173]

[0174] The continuous evaluation of the loss function relative to the value function allows the model to adapt to the environment more effectively, thereby increasing the probability of achieving higher rewards.

[0175] In this embodiment, the value data corresponding to the predicted resource allocation strategy is obtained by presetting the action value function, so that the model is trained based on the value data and model loss. The model learns the optimal action value formula, which is conducive to realizing the optimal resource allocation strategy.

[0176] In an exemplary embodiment, a conflict hypergraph is constructed based on channel information and resource conflict relationships, including: taking channels indicated by each channel information as vertices, and taking resource conflict relationships between channels as edges; determining hyperedges based on channels that have resource conflict relationships between each other, and constructing a conflict hypergraph based on each vertex, edge, and hyperedge.

[0177] In order to manage resource conflicts in WSN networks, it is necessary to model SL for further analysis. Since the conflict relationship between SLs is a complex many-to-many relationship, a hypergraph is introduced as a modeling tool to construct a conflict hypergraph with SL as the vertex. refer to Figure 5 The conflict hypergraph model shows two types of conflict relationships between SLs, where yes The set of SLs in is represented as It is a set of edges representing the conflict relationship of SL.

[0178] The conflict hypergraph can be represented by the incidence matrix H,

[0179]

[0180] where h ij Indicates vertex v i Is it a hyperedge? j :

[0181]

[0182] The incidence matrix H accurately describes the relationships between vertices and hyperedges in the conflict hypergraph.

[0183] Based on the two resource conflict relationships between SLs introduced in step 202, it can be quantified Figure 4 The conflict relationship between SLs in the communication topology diagram shown in Figure 1 is shown in Figure 2. Taking SL as the vertex, the conflicting vertices are connected with edges, and the hyperedges are determined based on the channels that have resource conflict relationships between each other, thus forming a conflict hypergraph. This method not only accurately represents the many-to-many conflict relationship between SLs, but also provides a mathematical solution based on graph theory for the subsequent resource allocation problem.

[0184] refer to Figure 5 The conflict hypergraph constructed in , in which some SLs are omitted, is used to clearly show the topological relationship between SLs in the conflict hypergraph for subsequent analysis.

[0185] Since all SLs within the same hyperedge are fully connected, that is, their conflict relations are the same. Therefore, when considering the conflict avoidance problem, it is only necessary to perform vertex coloring on each hyperedge separately, which means that the number of hyperedges to be calculated greatly affects the computational complexity. This transformation reduces the dimension of the association matrix of the hypergraph structure while retaining all conflict information, thereby avoiding unnecessary computational overhead for processing redundant information. Finally, using hypergraph theory as a mathematical model, the spectrum resource allocation problem is transformed into a more manageable hypergraph vertex coloring problem.

[0186] In this embodiment, the channels indicated by the channel information are taken as vertices, and the resource conflict relationships between the channels are taken as edges; hyperedges are determined based on the channels that have resource conflict relationships between each other, and a conflict hypergraph is constructed based on the vertices, edges and hyperedges. This method not only accurately represents the many-to-many conflict relationships between channels, but also provides a mathematical solution based on graph theory for subsequent resource allocation problems.

[0187] To explain in detail the resource management method and effect of the distributed wireless sensor network in this solution, a most detailed embodiment is described below:

[0188] For the resource management scenario of self-organizing network intelligent sensor devices in the transmission line scenario, the terminal obtains the channel information for communication between the self-organizing network intelligent sensor devices in the transmission line scenario, as well as the resource conflict relationship between the channels; the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship; based on the channel information and the resource conflict relationship, a conflict hypergraph is constructed; based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel, the resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model, and the resource allocation strategy is used to allocate resources to the multiple channels.

[0189] Among them, the effectiveness of the proposed algorithm is verified by simulation experiments. Numerical experiments are carried out through the example of classification problems of the UCI database. The results of numerical experiments illustrate the effectiveness of the proposed algorithm. The simulation results verify the performance of the proposed strategy; the FedAvg-DDQN algorithm is compared with the greedy network resource allocation algorithm, the randomized network resource allocation algorithm and the maximum node degree (MND)-based resource allocation algorithm. The model parameters of the proposed FedAvg-DDQN algorithm and the simulation parameters of the WSN network scenario are shown in Table 1.

[0190] Table 1 Simulation parameters

[0191]

[0192] like Figure 6It is a convergence diagram under different proportions of users in an embodiment. It can be seen from the figure that the cumulative reward changes with the number of training rounds under different discount rates (γ). It can be seen that all discount rate strategies have achieved convergence, among which the discount rate is 0.9, which performs best, converges in about 150 rounds, and the final cumulative reward is the highest, close to 360. In contrast, although the curves with discount rates of 0.95 and 0.85 are relatively stable, the final cumulative reward value is slightly lower than the former, and when the discount rate is 0.99, the model converges the slowest, and the final cumulative reward only reaches about 340. This shows that a discount rate of 0.9 can achieve the best balance in the short term in the current scenario, which not only ensures the convergence speed, but also obtains a higher cumulative reward.

[0193] like Figure 7 It is a convergence diagram under different convergence rounds in an embodiment. It can be seen from the figure that the cumulative reward changes with the number of training rounds under different learning rates (Lr). It can be seen that all strategies under all learning rates have achieved convergence, but the choice of learning rate has a significant impact on the convergence speed and cumulative reward of the model. Among them, the learning rate is 0.0003 when the performance is the best, the convergence speed is fast, and a stable state is reached in about 300 rounds, and the final cumulative reward is close to 350. The performance of the learning rate of 0.0005 is also good, and the cumulative reward is slightly lower than 0.0003. The convergence speed of the learning rate of 0.0007 is fast, but the stability is poor, and the final cumulative reward is slightly lower. The convergence speed of the learning rate of 0.0001 is the slowest, and the cumulative reward is the lowest, about 300. This shows that 0.0003 is the best learning rate in the current scenario.

[0194] like Figure 8 Figure 1 is a schematic diagram of the maximum network throughput in one embodiment. The figure shows the maximum network throughput of different numbers of SLs in a dense WSN network scenario using different algorithms. As the number of SLs increases, the number of available communication channels in the system also increases, resulting in an overall upward trend in the maximum network throughput of the four algorithms. In a dense WSN network scenario with 20 SLs, FedavgDDQN achieves a maximum network throughput of more than 80Mb / s, while all other compared algorithms are below 70Mb / s. With 50 SLs, Fedavg-DDQN achieves a maximum network throughput of 180Mb / s, while the MDQN algorithm ranges from 100Mb / s to 140Mb / s. In summary, according to Figure 8 ,FedAvg-DDQN algorithm outperforms MDQN, FDQN and SDQN algorithms.FedAvg-DDQN algorithm provides significant advantages for enhancing network capacity in dense WSN network scenarios.

[0195] like Fig. 9is a schematic diagram of the average network throughput in one embodiment. The figure shows the average network throughput of different numbers of SLs in a dense WSN network scenario using different algorithms. As the number of S2S agents increases, the network throughput of all four algorithms shows a significant upward trend. When there are 20 SLs in a dense WSN network scenario, the DDQN algorithm can reach close to 80Mb / s, while the FDQN and SDQN algorithms remain relatively constant. The MDQN algorithm is achieved at around 60Mb / s. When there are 40 SLs in a dense WSN network scenario, the Fedavg-DDQN algorithm achieves an average throughput of close to 160Mb / s, while the other three algorithms have an average throughput of less than 100Mb / s. In summary, Fig. 9 It shows that the FedAvg-DDQN algorithm outperforms the other three algorithms and has significant advantages in average network throughput in dense WSN network scenarios.

[0196] The above-mentioned resource management method for distributed wireless sensor networks obtains channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as the resource conflict relationship between each channel, and constructs a conflict hypergraph based on the channel information and the resource conflict relationship. Since the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship, the adjacency matrix of the conflict hypergraph can be used to characterize the conflict situation between each channel, thereby converting the conflict problem between channels into the vertex coloring problem of the conflict hypergraph, which is beneficial to reducing the computational complexity; based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel, the resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model. The resource allocation strategy is used to allocate resources to multiple channels, which is beneficial to reducing signal overlap interference in the densely distributed sensor equipment scenario and improving the reliability of signal transmission. In addition, the conflict hypergraph is used to model the communication channels in densely distributed wireless sensor network scenarios. By utilizing the characteristics of the hypergraph, different types of conflict channels are integrated into a hypergraph, which reduces redundant information. The resource allocation problem is solved by solving the vertex coloring problem of the hypergraph, which reduces the computational complexity. The hyperedges of the hypergraph are simplified by using the clique theory, and the redundancy of conflict information in the hyperedges is reduced by solving the maximum clique. This method reduces the dimension of the adjacency matrix representing the hypergraph and improves the computational efficiency of resource allocation. In order to solve the node coloring problem in the conflict hypergraph, the problem is transformed into a Markov decision process (MDP) model. The introduced MDP model is solved by the DDQN algorithm, which reduces the amount of calculation and speeds up the training. For the synchronously updated F-DDQN algorithm, the simulation results show that the proposed algorithm has a faster convergence speed and higher network throughput.

[0197] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0198] Based on the same inventive concept, the embodiment of the present application also provides a resource management device for a distributed wireless sensor network for implementing the resource management method for a distributed wireless sensor network involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the resource management device for a distributed wireless sensor network provided below can refer to the limitations of the resource management method for a distributed wireless sensor network in the above text, and will not be repeated here.

[0199] In an exemplary embodiment, Fig.10 As shown, a resource management device 100 for a distributed wireless sensor network is provided, comprising: an acquisition module 120, a construction module 140 and an allocation module 160, wherein:

[0200] The acquisition module 120 is used to obtain channel information for communication between the self-organizing network intelligent sensor devices in the power transmission line scenario, and the resource conflict relationship between the channels; the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship;

[0201] A construction module 140, configured to construct a conflict hypergraph based on channel information and resource conflict relations;

[0202] The allocation module 160 is used to obtain resource allocation strategies corresponding to multiple channels based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel within a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel through a pre-trained resource allocation model. The resource allocation strategy is used to allocate resources to multiple channels.

[0203] The above-mentioned resource management device for distributed wireless sensor networks obtains channel information for communication between self-organizing network intelligent sensor devices in the transmission line scenario, as well as the resource conflict relationship between each channel, and constructs a conflict hypergraph based on the channel information and the resource conflict relationship. Since the resource conflict relationship includes the power interference relationship and the spectrum overlap relationship, the adjacency matrix of the conflict hypergraph can be used to characterize the conflict situation between each channel, thereby converting the conflict problem between channels into the vertex coloring problem of the conflict hypergraph, which is beneficial to reducing the computational complexity; based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel in a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between each channel, the resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model. The resource allocation strategy is used to allocate resources to multiple channels, which is beneficial to reducing signal overlap interference in the densely distributed sensor equipment scenario and improving the reliability of signal transmission.

[0204] In one embodiment, during the training process of the resource allocation model, the allocation module 160 is also used to: obtain the sample adjacency matrix of the sample conflict hypergraph, the sample signal-to-noise ratio corresponding to each channel within a preset time period, the number of sample available resources, and the degree of sample resource conflict, and obtain the predicted resource allocation strategy corresponding to each of the multiple channels through the initial neural network model; calculate the model loss based on the empirical resource allocation strategy, the predicted resource allocation strategy, and the preset loss function; adjust the model parameters of the initial neural network model based on the model loss until the preset stop condition is met, and obtain a trained resource allocation model.

[0205] In one embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until the preset stop condition is met, and a trained resource allocation model is obtained. The allocation module 160 is also used to: obtain a reward value corresponding to the predicted resource allocation strategy based on a preset reward function; adjust the model parameters of the initial neural network model based on the reward value and the model loss until the preset stop condition is met, and a trained resource allocation model is obtained.

[0206] In one embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until the preset stop condition is met, and a trained resource allocation model is obtained. The allocation module 160 is also used to: obtain parameter optimization conditions; the parameter optimization conditions include at least one of the following: maximum resource throughput of each channel, conflict-free resource allocation, the transmission power of each sensor device is within a preset power range, and each channel is allocated at most one resource block; based on the model loss and the parameter optimization conditions, the model parameters of the initial neural network model are adjusted until the preset stop condition is met, and a trained resource allocation model is obtained.

[0207] In one embodiment, the model parameters of the initial neural network model are adjusted based on the model loss until the preset stop condition is met, thereby obtaining a trained resource allocation model. The allocation module 160 is also used to: obtain value data corresponding to the predicted resource allocation strategy based on a preset action value function; adjust the model parameters of the initial neural network model based on the value data and the model loss until the preset stop condition is met, thereby obtaining a trained resource allocation model.

[0208] In one embodiment, a conflict hypergraph is constructed based on channel information and resource conflict relationships, and the construction module 140 is also used to: use the channels indicated by each channel information as vertices, and use the resource conflict relationships between channels as edges; determine hyperedges based on channels that have resource conflict relationships between each other, and construct a conflict hypergraph based on each vertex, edge, and hyperedge.

[0209] Each module in the resource management device for distributed wireless sensor networks can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0210] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.11As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a resource management method for a distributed wireless sensor network is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0211] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0212] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0213] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0214] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0215] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0216] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0217] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0218] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A resource management method for a distributed wireless sensor network, characterized in that: The method comprises: Acquire channel information for communication between self-organizing network intelligent sensor devices in a power transmission line scenario, and resource conflict relationships between channels; the resource conflict relationships include power interference relationships and spectrum overlap relationships; Constructing a conflict hypergraph based on the channel information and the resource conflict relationship; Based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel within a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between channels, a resource allocation strategy corresponding to each of the multiple channels is obtained through a pre-trained resource allocation model, and the resource allocation strategy is used to allocate resources to the multiple channels.

2. The method according to claim 1, characterized in that The step of acquiring the resource conflict relationship between the channels includes: For any pair of channels, a first power interference amount is determined based on the transmission power of the ad hoc network intelligent sensor device on the first channel on the first channel and the distance between the ad hoc network intelligent sensor devices on the first channel and the second channel; a second power interference amount is determined based on the transmission power of the ad hoc network intelligent sensor device on the second channel on the second channel and the distance; if at least one of the first power interference amount and the second power interference amount meets the power interference condition, it is determined that there is a power interference relationship between the first channel and the second channel; Determine a spectrum overlap amount based on a center frequency and a bandwidth of the first channel and a center frequency and a bandwidth of the second channel; if the frequency overlap amount satisfies a spectrum overlap condition, determine that there is a spectrum overlap relationship between the first channel and the second channel; At least one of the power interference relationship and the spectrum overlap relationship is used as a resource conflict relationship between the first channel and the second channel.

3. The method according to claim 1, characterized in that The training process of the resource allocation model includes: Obtain the sample adjacency matrix of the sample conflict hypergraph, the sample signal-to-noise ratio corresponding to each channel in a preset time period, the number of sample available resources, and the degree of sample resource conflict, and obtain the predicted resource allocation strategy corresponding to each of the multiple channels through the initial neural network model; Calculate the model loss based on the empirical resource allocation strategy, the predicted resource allocation strategy, and the preset loss function; The model parameters of the initial neural network model are adjusted based on the model loss until a preset stop condition is met, thereby obtaining a trained resource allocation model.

4. The method according to claim 3, characterized in that The adjusting the model parameters of the initial neural network model based on the model loss until a preset stop condition is met to obtain a trained resource allocation model includes: Obtaining a reward value corresponding to the predicted resource allocation strategy based on a preset reward function; The model parameters of the initial neural network model are adjusted based on the reward value and the model loss until a preset stop condition is met, thereby obtaining a trained resource allocation model.

5. The method according to claim 3, characterized in that: The adjusting the model parameters of the initial neural network model based on the model loss until a preset stop condition is met to obtain a trained resource allocation model includes: Obtaining parameter optimization conditions; the parameter optimization conditions include at least one of maximum resource throughput of each channel, conflict-free resource allocation, transmission power of each sensor device being within a preset power range, and at most one resource block being allocated to each channel; The model parameters of the initial neural network model are adjusted based on the model loss and the parameter optimization conditions until a preset stop condition is met, thereby obtaining a trained resource allocation model.

6. The method according to claim 3, characterized in that The adjusting the model parameters of the initial neural network model based on the model loss until a preset stop condition is met to obtain a trained resource allocation model includes: Obtaining value data corresponding to the predicted resource allocation strategy based on a preset action value function; The model parameters of the initial neural network model are adjusted based on the value data and the model loss until a preset stop condition is met, thereby obtaining a trained resource allocation model.

7. The method according to claim 1, characterized in that The constructing a conflict hypergraph based on the channel information and the resource conflict relationship includes: The channels indicated by each channel information are taken as vertices, and the resource conflict relationship between each channel is taken as an edge; a hyperedge is determined based on channels having a resource conflict relationship between each other, and a conflict hypergraph is constructed based on each vertex, edge and hyperedge.

8. A resource management device for a distributed wireless sensor network, characterized in that: The device comprises: An acquisition module is used to acquire channel information for communication between self-organizing network intelligent sensor devices in a transmission line scenario, as well as resource conflict relationships between channels; the resource conflict relationships include power interference relationships and spectrum overlap relationships; A construction module, configured to construct a conflict hypergraph based on the channel information and the resource conflict relationship; An allocation module is used to obtain resource allocation strategies corresponding to multiple channels based on the adjacency matrix of the conflict hypergraph, the signal-to-noise ratio corresponding to each channel within a preset time period, the number of remaining available resources in the transmission line scenario, and the degree of resource conflict between the channels through a pre-trained resource allocation model, and the resource allocation strategy is used to allocate resources to the multiple channels.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images