Network resource allocation method for relay backhaul network of electric power intelligent sensor

Abstract

The invention relates to a network resource allocation method for an electric power intelligent sensor relay backhaul network. The method comprises the following steps: for a sensor group formed by an S2S pair sharing the same channel, according to communication performance parameters of each sensor device in the sensor group and noise signal power of the channel, determining a signal interference noise ratio of the sensor group, and under the condition that the type of interference suffered by an interfered sensor in the sensor group is accumulated interference, determining a signal interference noise ratio of the interfered sensor in the sensor group; determining interference weights corresponding to the interfered sensors according to the signal interference noise ratios, constructing an interference hypergraph model based on the interference weights corresponding to all the interfered sensors, determining the maximum interference degree of the sensor network according to the interference hypergraph model, and determining a resource allocation scheme based on the maximum interference degree. Therefore, the network resource distribution of the power sensor is realized. By adopting the method, the optimal resource allocation scheme can be adaptively selected, and the resource allocation is effectively realized.

Description

Technical Field

[0001] The present application relates to the technical field of power transmission line sensor networks, and in particular to a network resource allocation method for a power intelligent sensor relay backhaul network. Background Art

[0002] In power sensor networks, sensor devices often need to transmit information in the same frequency band, which leads to overlapping communication radius and tight spectrum resources. This overlap not only exacerbates the communication interference problem, but also significantly reduces the communication quality, posing a potential threat to the stable operation of the power system. Among traditional methods, reinforcement learning is widely used in such optimization problems. Its core mechanism is to continuously iterate and optimize decision-making strategies through continuous interaction with the environment in order to find the optimal solution in complex or uncertain scenarios. However, in the actual application of power sensor networks, this method faces many limitations.

[0003] Specifically, power sensor networks are usually deployed in complex and changeable power environments, such as high-voltage transmission lines and substations. These environments not only have limited spectrum resources, but also serious electromagnetic interference, making it difficult for reinforcement learning algorithms to make optimal decisions in real time in a dynamically changing environment. In addition, in terms of energy extraction, photovoltaic energy extraction is difficult to meet the long-term stable operation requirements of sensor nodes due to its large size, strong volatility and high maintenance cost. Electromagnetic field and vibration energy extraction have the characteristics of good stability and sustainability, but the working range of magnetic field energy extraction is limited, and the power of electric field and vibration energy extraction is relatively low, which cannot fully meet the energy supply requirements of sensors and self-organizing network nodes. In terms of networking, self-organizing network technology based on orthogonal frequency division multiplexing (OFDM) technology is currently mainly used. Although significant results have been achieved in network synchronization accuracy, key performance indicators such as multi-hop transmission rate and latency still need to be further improved.

[0004] In particular, the problem of power sensor data transmission is particularly prominent in communication blind areas such as mountainous areas and deserts. Due to the complex geographical environment and weak communication infrastructure in these areas, it is difficult for sensor data to be effectively transmitted back to the control center. Therefore, it is urgent to develop a multi-hop networking backhaul technology based on the self-energy collection of integrated sensors for power transmission multi-physical quantities and the fusion of spacer bars, which can effectively solve the communication problems in these areas and has broad application prospects and huge market potential. Summary of the invention

[0005] Based on this, it is necessary to provide a network resource allocation method, device, computer equipment, medium and product for an electric power intelligent sensor relay backhaul network that can effectively implement resource allocation in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for allocating network resources of a power intelligent sensor relay backhaul network, comprising:

[0007] For a sensor group consisting of an S2S pair sharing the same channel, a signal-to-interference-noise ratio of the sensor group is determined according to the communication performance parameters of each sensor device in the sensor group and the noise signal power of the channel;

[0008] When the type of interference to the interfered sensor in the sensor group is cumulative interference, determining an interference weight corresponding to the interfered sensor according to a signal interference noise ratio;

[0009] Based on the interference weights corresponding to all interfered sensors, an interference hypergraph model is constructed; the vertices of the interference hypergraph model represent sensor devices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other;

[0010] The maximum interference degree of the sensor network is determined according to the interference hypergraph model, and a resource allocation scheme is determined based on the maximum interference degree to realize network resource allocation of power sensors.

[0011] In one embodiment, the process of determining the type of interference to which the interfered sensor in the sensor group is subjected includes:

[0012] When the signal interference noise ratio of the sensor group is less than the maximum interference threshold, determining that the type of interference suffered by the interfered sensor in the sensor group is strong interference;

[0013] When the signal interference noise ratio of the sensor group is greater than or equal to the maximum interference threshold, determining that the type of interference to which the interfered sensor is subjected is weak interference;

[0014] In the case that the interfered sensor is subjected to strong interference and weak interference at the same time, it is determined that the type of interference subjected to the interfered sensor is cumulative interference.

[0015] In one embodiment, the step of determining the interference weight corresponding to the interfered sensor in the sensor group according to the signal interference noise ratio includes:

[0016] According to the relationship expression between the signal interference noise ratio and the maximum interference threshold, the upper limit of the total interference of the interfered sensors in the sensor group is determined;

[0017] Obtaining the power of the interference signal received by the interfered sensor from the interfering sensor;

[0018] According to the upper limit and power of the total interference, the interference weight corresponding to the interfered sensor is determined.

[0019] In one embodiment, the step of constructing an interference hypergraph model based on the interference weights corresponding to all interfered sensors includes:

[0020] According to the interference weights corresponding to all interfered sensors, an interference weight matrix is ​​constructed; each row of the interference weight matrix represents a sensor device, and each column of the interference weight matrix represents another sensor device that interferes with the sensor device;

[0021] According to the interference weight matrix, an interference hypergraph model is constructed; the nodes of the hypergraph in the interference hypergraph model are the sensor devices in the interference weight matrix, the hyperedges are the connections between the sensor devices and other sensor devices that interfere with the sensor devices, and the weight of the hyperedges is the sum of the interference weights between the sensor devices connected by the hyperedges.

[0022] In one embodiment, the step of determining the maximum interference degree of the sensor network according to the interference hypergraph model includes:

[0023] Determine a resource allocation matrix according to the interference hypergraph model; the resource allocation matrix is ​​used to represent resource allocation for each S2S pair;

[0024] The maximum interference degree of the sensor network is obtained by solving the resource allocation matrix.

[0025] In one embodiment, the step of determining a resource allocation scheme based on the maximum interference degree includes:

[0026] Construct the state space according to the communication state parameters of all S2S pairs;

[0027] Determine the objective function according to the resource allocation goal, and establish an optimization allocation model based on the objective function;

[0028] The optimization allocation model is trained based on the state space and the objective function to obtain a target allocation model;

[0029] A resource allocation plan is generated based on the target allocation model, and the resource allocation plan is updated according to the maximum interference degree.

[0030] In a second aspect, the present application also provides a network resource allocation device for a power intelligent sensor relay backhaul network, comprising:

[0031] A noise ratio determination module, for determining a signal-to-noise ratio of a sensor group consisting of an S2S pair sharing the same channel according to communication performance parameters of each sensor device in the sensor group and a noise signal power of the channel;

[0032] A weight determination module, used for determining an interference weight corresponding to the interfered sensor according to a signal interference noise ratio when the type of interference to the interfered sensor in the sensor group is cumulative interference;

[0033] A hypergraph construction module is used to construct an interference hypergraph model based on the interference weights corresponding to all interfered sensors; the vertices of the interference hypergraph model represent sensor devices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other;

[0034] The resource allocation module is used to determine the maximum interference degree of the sensor network according to the interference hypergraph model, and determine the resource allocation scheme based on the maximum interference degree to realize the network resource allocation of the power sensor.

[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 method steps of any one of the first aspects are implemented.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements any one of the method steps in the first aspect when the computer program is executed by a processor.

[0037] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements any one of the method steps in the first aspect when executed by a processor.

[0038] The network resource allocation method, device, computer equipment, medium and product of the above-mentioned power intelligent sensor relay backhaul network accurately calculates the signal interference to noise ratio of a sensor group composed of an S2S pair sharing the same channel. When the type of interference to the interfered sensor in the sensor group is cumulative interference, the interference weight corresponding to the interfered sensor is determined according to the signal interference to noise ratio, and then an interference hypergraph model is constructed. The maximum interference degree of the sensor network is determined based on the interference hypergraph model, and the resource allocation plan is determined based on the maximum interference degree. The optimal resource allocation plan can be adaptively selected, which can effectively avoid interference, improve the signal to noise ratio, and ensure the transmission rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] 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.

[0040] Figure 1 is a structural diagram of a sensor network in one embodiment;

[0041] Figure 2 A schematic diagram of a flow chart of a method for allocating network resources of a power intelligent sensor relay backhaul network in one embodiment;

[0042] FIG3( a ) is a schematic diagram of the interference relationship between S2S and communication in one embodiment;

[0043] FIG3( b ) is a schematic diagram of an interference weight matrix in one embodiment;

[0044] Figure 4 is a schematic diagram of an interference hypergraph model in one embodiment;

[0045] Figure 5 is a convergence diagram of a network resource management algorithm in one embodiment;

[0046] Figure 6 is a convergence diagram of a network resource management algorithm in another embodiment;

[0047] Figure 7 is a comparison diagram of network throughput in one embodiment;

[0048] Figure 8 is a graph showing the relationship between the number of S2S links and SINR in one embodiment;

[0049] Fig. 9 A schematic diagram of a flow chart of a method for allocating network resources of a power intelligent sensor relay backhaul network in another embodiment;

[0050] Fig.10 It is a structural block diagram of a network resource allocation device of a power intelligent sensor relay backhaul network in one embodiment;

[0051] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0052] 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.

[0053] The network resource allocation method of the power intelligent sensor relay backhaul network provided in the embodiment of the present application can be applied to Figure 1 The sensor network shown in Figure 1. The sensor network (Dense Distributed Power Sensor Network, DD-PSN) includes Sensor devices and Base stations provide spectrum resources to sensor devices through device-to-infrastructure (D2I) communication to achieve data transmission. Sensor devices exchange data through server-to-server (S2S) communication. Assuming that the spectrum resources can be shared between sensor devices, if each sensor device uses an independent channel, it will lead to a waste of spectrum resources. By sharing spectrum resources, multiple sensor devices can use the same sub-channel set, thereby improving spectrum utilization. That is, sensor devices use a sub-channel set. Shared spectrum resources.

[0054] In an exemplary embodiment, Figure 2 As shown, a network resource allocation method for a power intelligent sensor relay backhaul network is provided, and the method is applied to Figure 1 The sensor network in FIG. 1 is taken as an example to illustrate the method, which includes the following steps 202 to 208. Among them:

[0055] S202: For a sensor group consisting of an S2S pair sharing the same channel, determine a signal-to-interference-noise ratio of the sensor group according to communication performance parameters of each sensor device in the sensor group and noise signal power of the channel.

[0056] Optionally, Signal to Interference plus Noise Ratio (SINR) is an important indicator for measuring the performance of a communication system. SINR refers to the ratio of the received useful signal power to the sum of the received interference signal power (including interference from other signal sources and noise within the system). The higher the SINR value, the stronger the received useful signal is relative to the interference and noise, and the better the communication quality. Conversely, a low SINR value means that the communication quality may be greatly affected.

[0057] For example, in Figure 1 In the sensor network shown, for a sensor group consisting of the mth S2S pair sharing the same channel r (including the nth sensor device and the mth sensor device), the SINR of the nth sensor device can be expressed as:

[0058]

[0059] Among them, σ 2 represents the noise signal power at the base station on subchannel r, represents the transmission power of the nth sensor device when communicating on the rth subchannel, h nrepresents the channel gain of the nth sensor device, represents the transmission power of the mth S2S pair on the rth subchannel, h m Represents the interference channel gain.

[0060] Similarly, the SINR of the m′th S2S pair in the rth subchannel can be calculated as:

[0061]

[0062] in, represents the transmission power of the m′th S2S pair in the rth subchannel, h m′ represents the channel gain of the m′th S2S pair.

[0063] S204: When the type of interference suffered by the interfered sensor in the sensor group is cumulative interference, determine an interference weight corresponding to the interfered sensor according to the signal-to-interference-to-noise ratio.

[0064] Optionally, in order to solve the problem of composite overlapping interference in the power sensor network, it is necessary to classify it so as to solve it in a targeted manner, wherein cumulative interference refers to the simultaneous existence of interference sources of different strengths in the sensor system, which interact and gradually interfere with each other, and eventually have a negative impact on the performance of the system. For the case of cumulative interference, the corresponding interference weight is determined according to the signal interference noise ratio of each interfered sensor device, and the interference weight represents the interference intensity of the interfered sensor.

[0065] S206: constructing an interference hypergraph model based on the interference weights corresponding to all interfered sensors; the vertices of the interference hypergraph model represent sensor devices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other.

[0066] Optionally, in order to intuitively represent the interference relationship in the sensor network, a hypergraph model can be used to represent it. The interference hypergraph model uses sensor devices as vertices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other. This model is constructed by analyzing and quantifying the interference relationship between sensor devices in the sensor network. For example, if multiple sensor devices communicate on the same frequency band or are in close physical locations, there may be interference between them, and these devices can be connected by a hyperedge.

[0067] S208: Determine the maximum interference degree of the sensor network according to the interference hypergraph model, and determine a resource allocation scheme based on the maximum interference degree to implement network resource allocation of the power sensor.

[0068] Optionally, based on the interference hypergraph model, the maximum interference degree of the sensor network can be calculated, where the maximum interference degree is used to quantify the overall network interference level. When the maximum interference degree is zero, it means that there is no interference conflict in the network, and when the maximum interference degree is not zero, it means that there is interference. The maximum interference degree reflects the severity of interference in the sensor network, and the rationality of resource allocation can directly affect the magnitude of interference. For example, unreasonable allocation of resources such as channel resources and transmission power may lead to increased interference between sensor devices. By analyzing the maximum interference degree, it is possible to understand which areas or sensor devices are more seriously interfered with, so as to make targeted resource adjustments.

[0069] Furthermore, according to the maximum interference degree, a corresponding resource allocation plan is formulated, for example, adjusting the channel allocation of sensor devices, allocating devices with greater interference to different channels to reduce mutual interference, adjusting the transmission power, reducing the transmission power of the interference source to reduce interference to other devices, or optimizing the network topology to increase the distance between sensor devices and reduce the propagation of interference. By adjusting resource allocation, interference in the sensor network can be reduced, communication quality and network performance can be improved, and power sensors can be ensured to transmit data accurately and stably.

[0070] In the network resource allocation method of the above-mentioned power intelligent sensor relay backhaul network, the signal interference to noise ratio of the sensor group composed of an S2S pair sharing the same channel is accurately calculated. When the type of interference to the interfered sensor in the sensor group is cumulative interference, the interference weight corresponding to the interfered sensor is determined according to the signal interference to noise ratio, and then an interference hypergraph model is constructed. The maximum interference degree of the sensor network is determined based on the interference hypergraph model, and then the resource allocation scheme is determined based on the maximum interference degree. The optimal resource allocation scheme can be adaptively selected, interference can be effectively avoided, the signal to noise ratio can be improved, and the transmission rate can be guaranteed.

[0071] In an exemplary embodiment, the process of determining the type of interference suffered by an interfered sensor in a sensor group includes: when a signal-interference-to-noise ratio of the sensor group is less than a maximum interference threshold, determining that the type of interference suffered by the interfered sensor in the sensor group is strong interference; when a signal-interference-to-noise ratio of the sensor group is greater than or equal to the maximum interference threshold, determining that the type of interference suffered by the interfered sensor is weak interference; when the interfered sensor is subjected to both strong interference and weak interference, determining that the type of interference suffered by the interfered sensor is cumulative interference.

[0072] Optionally, the interference to the sensor device can be divided into three categories: strong interference, weak interference and cumulative interference, where the cumulative interference is composed of multiple strong interferences or weak interferences. Different interference types can be represented by the relationship between SINR and the maximum interference threshold. For example, when Exceeding the maximum interference threshold γ th The first S2S pair can communicate normally. In this case, strong interference can be caused by It means that weak interference can be represented by express.

[0073] Specifically, when multiple sensor devices operate on the same spectrum, there is a risk of severe interference between them. For example, if multiple sensor devices transmit on similar frequencies, they may interfere with each other, resulting in deterioration of signal quality or even complete communication interruption. At this time, the interference to the sensor devices is strong interference. Weak interference refers to the interference level in the sensor system is relatively light, which may have a certain impact on the communication quality, but will not cause interruption or seriously affect the effectiveness of communication. Cumulative interference refers to the simultaneous existence of interference sources of different strengths in the sensor system, which interact and gradually interfere with each other, and eventually have a negative impact on the performance of the system. This type of interference consists of strong interference and weak interference, which involves the normal operation of the DD-PSN network. Among them, S2S pairs with strong interference cannot share resources, while S2S pairs with weak interference have the least impact on each other, allowing normal communication and sharing of the same resources. Therefore, in order to determine the operating status of the DD-PSN network, it is necessary to accurately evaluate the cumulative interference to determine whether the cumulative interference has reached a level that seriously interferes with normal communication, thereby preventing resource waste.

[0074] In this embodiment, by classifying the interference to the sensor device into three types: strong interference, weak interference and cumulative interference, different management strategies can be adopted for different types of interference in a targeted manner. Through precise interference management, the impact of interference on sensor performance can be reduced and the transmission rate can be guaranteed.

[0075] In an exemplary embodiment, the step of determining the interference weight corresponding to the interfered sensor in the sensor group according to the signal interference-noise ratio includes: determining the upper limit of the total interference of the interfered sensors in the sensor group according to the relationship expression between the signal interference-noise ratio and the maximum interference threshold; obtaining the power of the interference signal received by the interfered sensor from the interfering sensor; and determining the interference weight corresponding to the interfered sensor according to the upper limit of the total interference and the power.

[0076] Optionally, first, according to the relationship expression between SINR and the maximum interference threshold, a threshold can be set, which represents the maximum interference level that each sensor in the sensor group can withstand. Through this threshold, the upper limit of the total interference of the interfered sensor in the sensor group can be calculated, that is, all interference signals added together cannot exceed this upper limit to ensure the normal operation of the sensor. After that, it is necessary to obtain the power of the interference signal received by the interfered sensor from the interfering sensor, and then calculate the interference weight corresponding to the interfered sensor according to the upper limit of the total interference and the power of each interference signal. The interference weight reflects the degree of influence of each interference signal on the interfered sensor.

[0077] Exemplarily, when the mth S2S pair can communicate normally, the relationship between its SINR and the maximum interference threshold is expressed as:

[0078]

[0079] in, represents the transmission power of the m′th S2S pair in the rth subchannel, h m′ represents the channel gain of the m′th S2S pair, p r i represents the interference signal power from other sensor devices i, h i represents the channel gain σ of sensor device i 2 Represents the noise power.

[0080] Furthermore, it can be deduced that:

[0081]

[0082] Therefore, the interference weight It is obtained based on the ratio of the power received by the interfered sensor device from the interfering sensor device to the upper limit of the total interference, and its expression is:

[0083]

[0084] Among them, P s An expression that represents the power received by the interfered sensor device from the interfering sensor device, with the denominator being the upper limit of the total interference.

[0085] In this embodiment, by determining the upper limit of the total interference of the interfered sensors in the sensor group based on the relationship expression between the signal interference-noise ratio and the maximum interference threshold, obtaining the power of the interference signal received by the interfered sensor from the interfering sensor, and determining the interference weight corresponding to the interfered sensor based on the upper limit of the total interference and the power, the interference situation between each sensor in the sensor network can be understood more accurately, thereby optimizing the network layout and parameter settings, and improving the overall performance of the network.

[0086] In an exemplary embodiment, the step of constructing an interference hypergraph model based on the interference weights corresponding to all interfered sensors includes: constructing an interference weight matrix according to the interference weights corresponding to all interfered sensors; each row of the interference weight matrix represents a sensor device, and each column of the interference weight matrix represents another sensor device that interferes with the sensor device; constructing an interference hypergraph model according to the interference weight matrix; the nodes of the hypergraph in the interference hypergraph model are the sensor devices in the interference weight matrix, the hyperedges are the connections between the sensor devices and other sensor devices that interfere with the sensor devices, and the weight of the hyperedges is the sum of the interference weights between the sensor devices connected by the hyperedges.

[0087] Optionally, an interference weight matrix is ​​constructed based on the interference weights corresponding to all interfered sensors, where each row of the matrix represents a sensor device, and each column represents another sensor device that interferes with the sensor device. The elements in the matrix (i.e., interference weights) reflect the degree of interference between sensor devices. Afterwards, an interference hypergraph model is constructed based on the interference weight matrix, in which the nodes of the hypergraph correspond to the sensor devices in the interference weight matrix, and the hyperedges represent the connections between the sensor devices and other sensor devices that interfere with the sensor devices. The weight of the hyperedge is the sum of the interference weights between the sensor devices connected by the hyperedge, which reflects the strength of the interference relationship represented by this hyperedge.

[0088] For example, as shown in FIG3 , FIG3 is a schematic diagram of an example S2S pair communication topology and interference weight matrix between sensor devices in a DD-PSN network, wherein FIG3(a) is a schematic diagram of interference relationship of S2S pair communication, and the arrow lines depicted in FIG3(a) indicate potential interference caused by communication with the indicated sensor device. FIG3(b) is an interference weight matrix recorded according to the interference relationship diagram of FIG3(a), wherein each row represents a different sensor device experiencing interference, and each column in the row represents interference caused by other sensor devices, for example, rows S3, S4, S6, S7, S8, S9, S10, S11, S12, S13, S14, S15, S16, S17, S18, S19, S20, S21, S22, S23, S24, S25, S26, S27, S28, S29, S30, S31, S32, S33, S34, S35, S36, S37, S38, S39, S40, S41, S42, S43, S44, S45, S46, S47, S48, S49, S50, S51, S52, S5 10 and S 11 The interference weights for S3 are 0.2, 0.3, 0.4, and 0.4 respectively. The 16th row in the matrix represents the interference weights from S 15 The independent interference exceeds S 16 In the matrix, each row is marked by a color, indicating different conditions, e.g., green indicates that the interference remains within the maximum interference threshold, thus eliminating the need to allocate additional resources to the corresponding sensor device, red indicates the requirement for resource allocation, and yellow indicates sensor devices that cause interference.

[0089] For example, the hypergraph can be represented as Where ε=(e1,e2,...,e J ) is the edge set of the hypergraph, is the set of vertices in the hypergraph. The relationship between vertices and hyperedges in the hypergraph can be expressed by the J×K incidence matrix M. J×K Represented, where the elements in the matrix can be expressed as:

[0090]

[0091] Among them, h (j，k) Represents the value of the j-th row and k-th column in the matrix M.

[0092] Furthermore, when only cumulative interference is analyzed, the cumulative interference can be represented by an interference hypergraph model. Based on the cumulative interference represented in FIG3(b), the receiving end is regarded as a hyperedge in the hypergraph, and the transmitting end interfering with the receiving end is regarded as a point in the hypergraph, thereby establishing an interference hypergraph model.

[0093] For example, the interference hypergraph model is Figure 4 As shown, in Figure 4 There are six vertices (S1, S4, S5, S8, S 11 , S 12 ) and five hyperedges (R2, R3, R5, R7, R 10 ), representing six different interference sources and five cumulative interferences, respectively. The hyperedges connecting the vertices represent the interference relationship between these sensor devices, that is, the sensor devices in the same hyperedge interfere with each other. The same sensor device may be in multiple hyperedges, resulting in multiple cumulative interferences. Figure 4 The interference hypergraph model in can intuitively understand the interference relationship between S2S pairs, where Figure 4 The incidence matrix in can be expressed as:

[0094]

[0095] Furthermore, when the cumulative interference weight exceeds 1, it means a significant impact on communication, and different resource allocation strategies are needed to mitigate interference. When the sum of the weights is less than 1, the same resources can continue to be shared. In order to solve the interference avoidance problem in resource allocation, for example, in row R2 of the association matrix A, S6, S 10 and S 12 The cumulative interference weight of any two sensor devices in R2 exceeds 1. Therefore, different resources must be allocated to these three sensor devices to prevent interference. Based on this, the interference avoidance problem in the interference hypergraph can be transformed into a point coloring problem to solve.

[0096] In this embodiment, by constructing an interference weight matrix according to the interference weights corresponding to all interfered sensors, and constructing an interference hypergraph model according to the interference weight matrix, it is possible to more accurately analyze the interference situation in the network, find out potential interference sources and interference paths, and thus select the best resource allocation plan in a targeted manner to effectively avoid interference.

[0097] In an exemplary embodiment, the step of determining the maximum interference degree of the sensor network according to the interference hypergraph model includes: determining a resource allocation matrix according to the interference hypergraph model; the resource allocation matrix is ​​used to characterize the resource allocation for each S2S pair; and solving the resource allocation matrix to obtain the maximum interference degree of the sensor network.

[0098] Optionally, based on the interference hypergraph model that has been constructed, a resource allocation matrix is ​​determined. The resource allocation matrix is ​​used to characterize the resource allocation between each S2S pair in the sensor network. The elements of the resource allocation matrix represent the amount of resources or resource identifiers allocated to each S2S pair. Through reasonable resource allocation, the interference between sensors can be reduced and the communication efficiency of the network can be improved. Afterwards, the maximum interference degree of the sensor network is solved according to the resource allocation matrix. The maximum interference degree is an indicator to measure the interference intensity in the network. It reflects the most serious interference situation in the network under a given resource allocation strategy.

[0099] For example, in order to know the interference between S2S pairs in DD-PSN, the maximum interference degree θ of the correlation matrix based on the interference hypergraph model is calculated. M To quantify the overall interference of DD-PSN. Among them, the maximum interference degree Θ M It can be expressed as:

[0100] Θ M =MAX[max(M J×K H K×C ,1]-1,

[0101] Among them, H K×C represents the resource allocation matrix, which specifies how to allocate C resources to K S2S pairs, max(·) represents the maximum value in the matrix, and MAX[·,1] means comparing each element with 1 and selecting the larger of the two values. M =0 indicates that there is no interference conflict in the network, the maximum interference level Θ M The overall network interference level can be quantified, while Θ M ≠0 means there is interference.

[0102] In this embodiment, the resource allocation matrix is ​​determined according to the interference hypergraph model. The resource allocation matrix is ​​used to characterize the resource allocation for each S2S pair. The maximum interference degree of the sensor network is obtained by solving the resource allocation matrix. The optimal resource allocation scheme can be adaptively selected, which can effectively avoid interference, improve the signal-to-interference-noise ratio, and ensure the transmission rate.

[0103] In an exemplary embodiment, the step of determining a resource allocation scheme based on the maximum interference degree includes: constructing a state space based on the communication state parameters of all S2S pairs; determining an objective function based on the resource allocation goal, and establishing an optimization allocation model based on the objective function; training the optimization allocation model based on the state space and the objective function to obtain a target allocation model; generating a resource allocation scheme based on the target allocation model, and updating the resource allocation scheme based on the maximum interference degree.

[0104] Optionally, by obtaining the communication state parameters of all S2S pairs in the sensor network, a state space is constructed, wherein the state space is a multidimensional space, each dimension represents a communication state parameter, and each point in the state space represents a specific communication state. The purpose of constructing the state space is to provide input data for the subsequent optimization allocation model and help the model understand the communication state changes in the network. Afterwards, according to the goal of resource allocation, such as maximizing network throughput, minimizing transmission delay, or maximizing resource utilization, an objective function is determined, which will be used to measure the quality of the resource allocation scheme. Then, based on the objective function and the state space, an optimization allocation model is established. The optimization allocation model is used to find the optimal resource allocation scheme, that is, to make the objective function reach the optimal value under the premise of satisfying the constraints.

[0105] Furthermore, the collected communication state parameter data is used to train the optimization allocation model so that the model can learn the relationship between the state space and the objective function, thereby being able to predict the objective function value under different resource allocation schemes. During the training process, the optimal resource allocation scheme is found through the optimization algorithm, and the resource allocation scheme is updated according to the maximum interference degree. For example, the interference situation under the current scheme is evaluated through the maximum interference degree, and the scheme is adjusted according to the evaluation results to reduce interference and improve network performance.

[0106] For example, for resource allocation in DD-PSN scenarios, the joint channel and power resource allocation problem is modeled as a Markov decision process (MDP) model. Since the dynamics of the environment are usually unknown in wireless networks, deep reinforcement learning (DRL) is used to learn the best strategy for a large number of trainings on the current state, which may be transferred to other states with incompletely unknown transition probabilities. Among them, the MDP model is mainly composed of state space, action space, reward function and value function, as shown below:

[0107] 1) State space: At time slot t, for the mth link, the DRL agent observes the communication and collects the following parameters that constitute the link state:

[0108] C m : represents the set of channels selected for all S2S links in dense DD-PSN;

[0109] C s : represents the remaining available channel set in dense DD-PSN;

[0110] make represents the state space, so the system state at time slot t is Defined as:

[0111] 2) Action space: The most suitable action space A contains all possible channel allocation decisions and power allocation decisions Therefore, the action of time slot t can be expressed as:

[0112]

[0113] Among them, N c Indicates the number of available channels. In the MDP model, the power resources are designed to be divided into N p parts, so the dimension of the action space is N c ×N p , each action corresponds to a specific combination of spectrum subband and power selection.

[0114] 3) Reward function: used to evaluate the quality of the learned decision-making strategy.

[0115] 4) Value Function: The value function is used to evaluate the long-term benefits of taking an action in a given state.

[0116] Furthermore, when solving the MDP model, it can be implemented through the multi-agent deep double Q network (MADDQN) algorithm, which uses deep neural networks to approximate the Q value function and finds the optimal resource allocation strategy through the collaboration of multiple agents. When constructing the DDQN-based power sensor network training framework, each S2S link acts as an intelligent agent, the agent observes the current state, selects actions to interact with the environment, the environment updates the state according to the actions of multiple agents, and provides feedback (such as updated state and reward), and uses two independent neural networks (training network and target network) to select actions and calculate Q values ​​respectively, reduce estimation bias, update the parameters of the training network according to the loss function, and use a soft update strategy to update the parameters of the target network. A distributed decision-making framework is adopted to enable each sensor device to autonomously learn and adjust decisions based on local information, thereby enhancing the scalability, adaptability and overall robustness of the network.

[0117] For example, the action selection is performed using the decay ε-greedy strategy, and the action selection is expressed as:

[0118]

[0119] ε t =ε min +(ε max -ε min )·e -βt ,

[0120] Among them, ε t represents the probability of randomly selecting an action, ε max represents the initial exploration probability, ε min represents the minimum exploration probability, and β represents the decay exponent. t The greedy strategy can introduce randomness to better explore the environment and learn more useful information. Initially, when ε t When ε is larger, the agent learns less about the environment, which can help the agent to be more likely to randomly choose exploration actions. Over time, as the agent learns more about the environment, ε t Decreasing, allowing the agent to favor high-value actions of its own choice.

[0121] The loss function of DDQN can be expressed as:

[0122]

[0123] in, represents the target value of the target network output, Q(s t , a t ; θ) represents the estimated value obtained by training the network, θ, θ -are the parameters of the training network and the target network, respectively. γ represents the discount factor, which is used to balance the current reward and future reward. a′ represents the action in the action set that can obtain the maximum Q value. The training of the neural network can be completed by updating the loss function through stochastic gradient descent to make the loss function converge.

[0124] The soft update strategy is expressed as: θ - ←ηθ+(1-η)θ - , where θ represents the soft update parameter. The larger θ is, the higher the synchronization rate of the target network is.

[0125] Since a single DDQN network model is trained using random samples from its local experience replay region, the complexity of the local update is O(T s ×N lr ), which depends on the target Q network complexity T s and the number of local training N lr The complexity of model aggregation is O(k) because it increases linearly with the number of agents. The total complexity of the MADDQN algorithm is Therefore, the larger the number of agents, the faster the MADDQN algorithm can be trained.

[0126] In this embodiment, a state space is constructed according to the communication state parameters of all S2S pairs, an objective function is determined according to the resource allocation target, an optimization allocation model is established according to the objective function, the optimization allocation model is trained based on the state space and the objective function to obtain a target allocation model, a resource allocation plan is generated based on the target allocation model, and the resource allocation plan is updated according to the maximum interference degree. The best resource allocation plan can be adaptively selected, interference can be effectively avoided, the signal-to-interference-noise ratio can be improved, and the transmission rate can be guaranteed.

[0127] In an exemplary embodiment, in order to verify the effectiveness of the MADDQN algorithm, a simulation comparison is performed with the SADQN algorithm, the SADDQN algorithm and the BASELINE algorithm. The simulation parameter settings are shown in Table 1:

[0128] Table 1

[0129]

[0130] For example, Figure 5 As shown, Figure 5The convergence behavior of the FedAvg-DQN-based network resource management algorithm at various learning rates. Throughout the training phase, when the number of S2S links is fixed to 20, the reward value stabilizes as the training progresses, and when the learning rate is set to 0.001, it significantly shows improved convergence performance. Too high a learning rate leads to unstable parameter updates, hindering stable convergence. On the contrary, too low a learning rate may cause the algorithm to fall into a local optimum. Therefore, for subsequent experiments, a learning rate of 0.001 is used.

[0131] Further, if Figure 6 As shown, Figure 6 The figure shows the convergence behavior of the MADDQN-based network resource management algorithm at various discount factors. During the entire training phase, when the number of S2S links is fixed to 20 and the learning rate is 0.0001, the reward value stabilizes with training. When the discount factor is set to 0.001, the improved convergence performance is significantly demonstrated. A discount factor that is too high will lead to unstable parameter updates and hinder stable convergence. Conversely, a discount factor that is too low may cause the algorithm to fall into a local optimum. Therefore, for subsequent experiments, a discount factor of 0.95 is used. Figure 6 and Figure 5 It can be seen that the MADQN algorithm can achieve network resource management more effectively than the FedAvg-DQN algorithm.

[0132] Furthermore, when evaluating the performance of the power sensor network, the evaluation indicators involved mainly include network throughput and SINR.

[0133] For example, Figure 7 As shown, Figure 7 The network throughput achieved using various algorithms in a power sensor network environment shows that as the number of D2D connections increases, the network throughput also increases. This can be attributed to the enhanced transmission links, which enable a larger amount of data to be transmitted simultaneously, thereby improving the overall network throughput. Figure 7 It can be seen that the MADDQN algorithm is also significantly better than other algorithms.

[0134] For example, Figure 8 As shown, Figure 8 Figure 2 is the relationship between the number of S2S links and SINR. A higher SINR value means that the network has lower interference. Compared with the other three algorithms, the MADDQN algorithm has a higher signal to interference noise ratio.

[0135] In an exemplary embodiment, Fig. 9 As shown, a method for allocating network resources of a power intelligent sensor relay backhaul network is provided, the method comprising the following steps:

[0136] S902: For a sensor group consisting of an S2S pair sharing the same channel, determine a signal-to-interference-noise ratio of the sensor group according to communication performance parameters of each sensor device in the sensor group and noise signal power of the channel.

[0137] S904: When the signal interference-noise ratio of the sensor group is less than the maximum interference threshold, it is determined that the type of interference suffered by the interfered sensor in the sensor group is strong interference; when the signal interference-noise ratio of the sensor group is greater than or equal to the maximum interference threshold, it is determined that the type of interference suffered by the interfered sensor is weak interference; when the interfered sensor is subjected to both strong interference and weak interference, it is determined that the type of interference suffered by the interfered sensor is cumulative interference.

[0138] S906: When the type of interference suffered by the interfered sensor in the sensor group is cumulative interference, determine the upper limit of the total interference of the interfered sensors in the sensor group according to the relationship expression between the signal interference-noise ratio and the maximum interference threshold; obtain the power of the interference signal received by the interfered sensor from the interfering sensor; determine the interference weight corresponding to the interfered sensor according to the upper limit of the total interference and the power.

[0139] S908: Construct an interference weight matrix based on the interference weights corresponding to all interfered sensors; each row of the interference weight matrix represents a sensor device, and each column of the interference weight matrix represents another sensor device that interferes with the sensor device; construct an interference hypergraph model based on the interference weight matrix; the nodes of the hypergraph in the interference hypergraph model are the sensor devices in the interference weight matrix, the hyperedges are the connections between the sensor devices and other sensor devices that interfere with the sensor devices, and the weight of the hyperedges is the sum of the interference weights between the sensor devices connected by the hyperedges.

[0140] S910: Determine a resource allocation matrix according to the interference hypergraph model; the resource allocation matrix is ​​used to represent resource allocation for each S2S pair; and obtain the maximum interference degree of the sensor network by solving the resource allocation matrix.

[0141] S912: construct a state space according to the communication state parameters of all S2S pairs; determine the objective function according to the resource allocation target, and establish an optimization allocation model according to the objective function; train the optimization allocation model based on the state space and the objective function to obtain a target allocation model; generate a resource allocation plan based on the target allocation model, and update the resource allocation plan according to the maximum interference degree.

[0142] In this embodiment, by accurately calculating the signal interference to noise ratio of a sensor group composed of an S2S pair sharing the same channel, when the type of interference to the interfered sensor in the sensor group is cumulative interference, the interference weight corresponding to the interfered sensor is determined according to the signal interference to noise ratio, and then an interference hypergraph model is constructed. The maximum interference degree of the sensor network is determined based on the interference hypergraph model, and then the resource allocation scheme is determined based on the maximum interference degree. The optimal resource allocation scheme can be adaptively selected, interference can be effectively avoided, the signal interference to noise ratio can be improved, and the transmission rate can be guaranteed.

[0143] It should be understood that, although the steps in the flowcharts involved in the above 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 is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may 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 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.

[0144] Based on the same inventive concept, the embodiment of the present application also provides a network resource allocation device for a power intelligent sensor relay backhaul network for implementing the network resource allocation method for the power intelligent sensor relay backhaul 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 the embodiments of the network resource allocation device for one or more power intelligent sensor relay backhaul networks provided below can refer to the limitations of the network resource allocation method for the power intelligent sensor relay backhaul network above, and will not be repeated here.

[0145] In an exemplary embodiment, Fig.10 As shown, a network resource allocation device for a power intelligent sensor relay backhaul network is provided, comprising: a noise ratio determination module 10, a weight determination module 20, a hypergraph construction module 30 and a resource allocation module 40, wherein:

[0146] The signal-to-noise ratio determination module 10 is used to determine the signal-to-interference-noise ratio of the sensor group composed of a S2S pair sharing the same channel according to the communication performance parameters of each sensor device in the sensor group and the noise signal power of the channel.

[0147] The weight determination module 20 is used to determine the interference weight corresponding to the interfered sensor according to the signal interference noise ratio when the type of interference suffered by the interfered sensor in the sensor group is cumulative interference.

[0148] The hypergraph construction module 30 is used to construct an interference hypergraph model based on the interference weights corresponding to all interfered sensors; the vertices of the interference hypergraph model represent sensor devices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other.

[0149] The resource allocation module 40 is used to determine the maximum interference degree of the sensor network according to the interference hypergraph model, and determine the resource allocation scheme based on the maximum interference degree to implement network resource allocation of the power sensor.

[0150] In an exemplary embodiment, the weight determination module 20 is also used to determine that the type of interference suffered by the interfered sensor in the sensor group is strong interference when the signal interference-noise ratio of the sensor group is less than the maximum interference threshold; to determine that the type of interference suffered by the interfered sensor is weak interference when the signal interference-noise ratio of the sensor group is greater than or equal to the maximum interference threshold; and to determine that the type of interference suffered by the interfered sensor is cumulative interference when the interfered sensor is subjected to both strong interference and weak interference.

[0151] In an exemplary embodiment, the weight determination module 20 is also used to determine the upper limit of the total interference of the interfered sensors in the sensor group according to the relationship expression between the signal interference-noise ratio and the maximum interference threshold; obtain the power of the interference signal received by the interfered sensor from the interfering sensor; and determine the interference weight corresponding to the interfered sensor according to the upper limit of the total interference and the power.

[0152] In an exemplary embodiment, the hypergraph construction module 30 is also used to construct an interference weight matrix based on the interference weights corresponding to all interfered sensors; each row of the interference weight matrix represents a sensor device, and each column of the interference weight matrix represents other sensor devices that interfere with the sensor device; an interference hypergraph model is constructed based on the interference weight matrix; the nodes of the hypergraph in the interference hypergraph model are the sensor devices in the interference weight matrix, the hyperedges are the connections between the sensor devices and other sensor devices that interfere with the sensor devices, and the weight of the hyperedges is the sum of the interference weights between the sensor devices connected by the hyperedges.

[0153] In an exemplary embodiment, the resource allocation module 40 is further used to determine a resource allocation matrix according to the interference hypergraph model; the resource allocation matrix is ​​used to characterize resource allocation for each S2S pair; and the maximum interference degree of the sensor network is obtained by solving the resource allocation matrix.

[0154] In an exemplary embodiment, the resource allocation module 40 is also used to construct a state space based on the communication state parameters of all S2S pairs; determine the objective function based on the resource allocation target, and establish an optimization allocation model based on the objective function; train the optimization allocation model based on the state space and the objective function to obtain a target allocation model; generate a resource allocation plan based on the target allocation model, and update the resource allocation plan based on the maximum interference degree.

[0155] Each module in the network resource allocation device of the above-mentioned power intelligent sensor relay backhaul network can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0156] 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.11 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, 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 the 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 (Near Field Communication, NFC) or other technologies. When the computer program is executed by the processor, a network resource allocation method for a power intelligent sensor relay backhaul 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.

[0157] Those skilled in the art will understand that Fig.11The 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.

[0158] 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.

[0159] 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.

[0160] 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.

[0161] 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.

[0162] 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.

[0163] 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 understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations 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 network resource allocation method for a power intelligent sensor relay backhaul network, characterized in that: Applicable to a sensor network composed of multiple sensors using S2S communication; the method comprises: For a sensor group consisting of an S2S pair sharing the same channel, determining a signal-to-interference-noise ratio of the sensor group according to a communication performance parameter of each sensor device in the sensor group and a noise signal power of the channel; When the type of interference suffered by the interfered sensor in the sensor group is cumulative interference, determining an interference weight corresponding to the interfered sensor according to the signal interference noise ratio; Based on the interference weights corresponding to all interfered sensors, an interference hypergraph model is constructed; the vertices of the interference hypergraph model represent sensor devices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other; The maximum interference degree of the sensor network is determined according to the interference hypergraph model, and a resource allocation scheme is determined based on the maximum interference degree to implement network resource allocation of power sensors.

2. The method according to claim 1, characterized in that The process of determining the type of interference to which the interfered sensor in the sensor group is subjected comprises: When the signal interference noise ratio of the sensor group is less than the maximum interference threshold, determining that the type of interference suffered by the interfered sensor in the sensor group is strong interference; When the signal interference noise ratio of the sensor group is greater than or equal to the maximum interference threshold, determining that the type of interference to which the interfered sensor is subjected is weak interference; In the case that the interfered sensor is subjected to strong interference and weak interference at the same time, it is determined that the type of interference subjected to the interfered sensor is cumulative interference.

3. The method according to claim 2, characterized in that The determining, according to the signal-to-interference-noise ratio, an interference weight corresponding to the interfered sensor in the sensor group includes: Determining the upper limit of the total interference of the interfered sensors in the sensor group according to the relationship expression between the signal interference noise ratio and the maximum interference threshold; Acquire the power of the interference signal received by the interfered sensor from the interfering sensor; An interference weight corresponding to the interfered sensor is determined according to the interference sum upper limit and the power.

4. The method according to claim 1, characterized in that The construction of an interference hypergraph model based on the interference weights corresponding to all interfered sensors includes: Constructing an interference weight matrix according to the interference weights corresponding to all interfered sensors; each row of the interference weight matrix represents a sensor device, and each column of the interference weight matrix represents another sensor device that interferes with the sensor device; According to the interference weight matrix, an interference hypergraph model is constructed; the nodes of the hypergraph in the interference hypergraph model are the sensor devices in the interference weight matrix, the hyperedges are the connections between the sensor devices and other sensor devices that interfere with the sensor devices, and the weight of the hyperedges is the sum of the interference weights between the sensor devices connected by the hyperedges.

5. The method according to claim 1, characterized in that The determining the maximum interference degree of the sensor network according to the interference hypergraph model comprises: Determine a resource allocation matrix according to the interference hypergraph model; the resource allocation matrix is ​​used to characterize resource allocation for each of the S2S pairs; The maximum interference degree of the sensor network is obtained by solving the resource allocation matrix.

6. The method according to claim 1, characterized in that The determining of a resource allocation scheme based on the maximum interference degree comprises: Construct the state space according to the communication state parameters of all S2S pairs; Determine an objective function according to the resource allocation target, and establish an optimization allocation model according to the objective function; Training the optimization allocation model based on the state space and the objective function to obtain a target allocation model; A resource allocation scheme is generated based on the target allocation model, and the resource allocation scheme is updated according to the maximum interference degree.

7. A network resource allocation device for a power intelligent sensor relay backhaul network, characterized in that: The device comprises: A noise ratio determination module, for determining, for a sensor group consisting of an S2S pair sharing the same channel, a signal-to-interference-noise ratio of the sensor group according to a communication performance parameter of each sensor device in the sensor group and a noise signal power of the channel; a weight determination module, configured to determine an interference weight corresponding to the interfered sensor according to the signal interference noise ratio when the type of interference to which the interfered sensor in the sensor group is subjected is cumulative interference; A hypergraph construction module, used to construct an interference hypergraph model based on the interference weights corresponding to all interfered sensors; the vertices of the interference hypergraph model represent sensor devices, and the hyperedges connecting the vertices represent a set of sensor devices that interfere with each other; The resource allocation module is used to determine the maximum interference degree of the sensor network according to the interference hypergraph model, and determine a resource allocation scheme based on the maximum interference degree to implement network resource allocation of the power sensor.

8. 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 6 are implemented.

9. 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 6 are implemented.

10. A computer program product, comprising a computer program, 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 6 are implemented.

Images