Intelligent reflecting surface coverage enhancement method and device for electric power inspection

Abstract

The embodiment of the invention provides an intelligent reflecting surface coverage enhancement method and device for electric power inspection, and the method comprises the steps: building an intelligent reflecting surface and inspection equipment unified scheduling model, and building a unified scheduling model for a multi-intelligent reflecting surface networking network; deducing a signal interference distribution analytic expression, solving a space measure function of a point process and determining heterogeneous station density based on an interference level; constructing a multi-intelligent reflector networking correlation model, and analyzing a link coupling factor; based on modeling of a link coupling factor, the link performance under the collaborative networking is studied; a cross-domain reconfigurable capability boundary is explored, and a spatial statistical average analysis method for coverage, capacity and mobility performance is innovated; an adjustable intelligent reflecting surface distribution dependence model is provided, and a semi-closed theoretical analysis expression of the user coverage rate is given based on a stochastic geometric tool.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a collaborative networking solution for intelligent reflective surfaces in the context of existing unmanned aerial vehicle power inspections. Background Art

[0002] With the continuous advancement of smart grid construction, the way of power line inspection is undergoing profound changes and moving rapidly towards digitalization. Drone inspection has gradually become an important means of power line inspection due to its high efficiency and flexibility. However, in actual application scenarios, especially in complex geographical environments such as mountainous and forested areas, drone inspection faces many severe challenges.

[0003] These areas often have no signal or weak signal. In order to ensure stable communication between the drone and the data center and achieve high-speed transmission of inspection data, it is necessary to increase the transmission power of the drone to enhance the signal strength. However, drones are strictly limited by their own load and power and cannot be equipped with high-power RF modules. This limitation directly leads to the degradation of drone communication performance, making it difficult to meet the urgent need for high-speed data transmission during power line inspection.

[0004] There are currently two mainstream solutions to this problem in the industry. The first solution is to use relay drones as the communication transmission medium between drones and data centers. Relay drones can effectively expand the communication range and ensure the continuity of data transmission to a certain extent. However, it also has obvious shortcomings. Relay drones themselves also face energy supply problems and cannot obtain sufficient power supply in time, resulting in limited endurance. In actual inspection tasks, insufficient endurance may cause the relay drone to fail to work normally at critical moments, thus affecting the smooth progress of the entire inspection process.

[0005] The second solution is to install Mesh networking equipment on the pole tower. In this way, when the drone is within the effective coverage of the equipment, it can continue to communicate. However, this solution also faces many difficulties in implementation. The pole tower is usually located at a high position and the environment is complex. It is difficult to install Mesh networking equipment, and it requires the collaboration of multiple people to climb the pole tower for deployment. This not only increases the labor cost and time cost, but is also affected by factors such as weather and terrain, making equipment deployment more difficult.

[0006] In this context, smart reflective surface technology came into being and has become an important candidate technology for power line inspection with its unique advantages. Smart reflective surface is composed of multiple independent passive reflective units, which can accurately control the amplitude and phase of electromagnetic signals, so as to produce an ideal multipath effect. It has the ability to reconstruct the reflection angle and can flexibly adjust the propagation direction of the signal according to actual needs to adapt to different terrain and environmental conditions. In addition, smart reflective surface can also provide beyond line of sight links, effectively overcoming the limitations of complex environments such as mountainous forests on signal propagation, and providing new possibilities and solutions for the high-speed and stable transmission of power line inspection data.

[0007] Power line inspections cover a wide range and long distances, while the service capabilities of a single smart reflective surface are limited by cascade path loss and large-angle non-ideal characteristics. The collaborative networking of smart reflective surfaces can break through the limitations of the service capabilities of a single smart reflective surface, alleviate the obstruction effects of mountains and rivers, and ensure the consistency of drone services. Multiple smart reflective surfaces can expand the service scope through collaborative networking and provide communication guarantees for drones in different locations. Summary of the invention

[0008] The purpose of the present invention is to provide a method and device for enhancing the coverage of intelligent reflective surfaces for power inspection, so as to solve the problems of communication signal shielding, attenuation and interference faced by UAVs during power inspection, improve the communication quality and reliability, and ensure the smooth progress of UAV power inspection. The specific technical solution is as follows:

[0009] Step 200, establish a unified scheduling model for intelligent reflective surfaces and inspection equipment. For a multi-intelligent reflective surface network, establish a unified scheduling model, model node capabilities (time, frequency, and space resources) and channel status (LOS / NLOS, propagation delay) as a unified scheduling scale, and perform differential mapping according to the inspection data transmission requirements (rate, delay). Extract the characteristics of the law of change of performance indicators with large-scale spatiotemporal constraints, cross-domain multi-dimensional resources (time, frequency, and space), and spatial propagation environment parameters, so as to explore the boundaries of network elastic adaptability and guide the deployment of network spatial characteristic parameters. According to the geographical environment and topological structure of the power line, the intelligent reflective surfaces are arranged in different positions in a hierarchical deployment manner. In areas with severe signal obstruction such as mountainous areas, the intelligent reflective surfaces are deployed on the top of the mountain or high places to ensure that the signal can pass through obstacles to the drone. In urban high-rise areas, the intelligent reflective surfaces are installed on the outer walls or roofs of buildings, and the height advantage of the buildings is used to guide signal propagation to avoid signals being blocked by high-rise buildings. Smart reflective surfaces can also be appropriately installed on the poles of power lines to enhance signal transmission between poles and towers and improve communication continuity. At the same time, the direction and angle of the smart reflective surface can be reasonably set so that it can dynamically adjust the reflection direction according to the flight trajectory of the drone, always providing the best signal reflection service for the drone.

[0010] Step 210, derive the analytical expression of signal interference distribution. In the multi-intelligent reflective surface network, the high dynamic characteristics of the UAV make the spatial structure of the interference coordination area more irregular, and it is difficult to obtain a definite mathematical analysis of the joint distance distribution function and the number of sites of the group interference site set. Based on the beam model and the LoS / NLoS probability model and path loss of the UAV air-to-ground link, the distribution of the interference station is modeled as a sorting process about the site's own service capability and path loss, and the spatial measurement function of the point process and the heterogeneous site density under the determined interference level are solved to obtain the interference site distribution function reflecting the interference link fluctuation and the probability density function of the number of scheduling sites.

[0011] Step 220, construct a multi-intelligent reflective surface networking correlation model. The spatial correlation factor will mine the change law of site correlation with the group construction principle (site distance, link status, backhaul, load conditions, etc.). After quantitatively modeling the correlation factor, analyze the change characteristics of the parameters that directly affect the signal-to-interference-to-noise ratio index (effective load, covariance size of interference to users in different time slots, etc.) with the correlation factor, and analyze the impact of the link coupling factor on the integrated sky-ground group interference management. Use efficient modulation methods (orthogonal frequency division multiplexing modulation) and advanced coding technology (low-density parity check code) to improve signal transmission efficiency and reliability. In the frame structure design, add fields for control information and feedback information to achieve interaction and coordination between drones, intelligent reflective surfaces, and ground control stations.

[0012] Step 230, explore the boundary of cross-domain reconfigurable capabilities, innovate the spatial statistical average analysis method of coverage, capacity, and mobility performance, based on the modeling of link coupling factors, study the probability of the base station being in different states, and the link performance in this state, and finally obtain the coverage and capacity spatial statistical average performance reflecting the overall level of users by deconditioning the link performance in different states of the site. Through real-time collection and analysis of information such as the position, flight speed, and signal strength of the drone, the deep learning model is used to predict the optimal phase and amplitude configuration of the reflective element. First, a deep neural network model is established, with the relevant state information of the drone as input and the reflection parameters of the intelligent reflective surface as output. Then, the model is trained with a large amount of training data so that it can learn the optimal reflection strategy under different circumstances. In the actual operation process, according to the real-time input drone state information, the model can quickly output the corresponding reflection parameters to realize real-time intelligent control of the intelligent reflective surface.

[0013] Step 240, modeling the intelligent reflective surface dependent distribution deployment, starting from the unique multiplicative fading characteristics of the intelligent reflective surface, considering the position dependence of the reflective surface on the base station, an adjustable intelligent reflective surface distribution dependence model is proposed, and a semi-closed theoretical analysis expression of user coverage is given based on random geometry tools. Specifically, the intelligent reflective surface is deployed in a circle with an adjustable radius centered on the base station. The multiplicative path loss is controlled by reducing the distance from the base station to the intelligent reflective surface. This deployment method can maximize the signal enhancement potential of the intelligent reflective surface and has high feasibility. Design a cooperative communication algorithm between the UAV and the intelligent reflective surface. When the UAV detects that the signal quality has decreased, it will send a request signal to the ground control station. The ground control station calculates the optimal adjustment parameters of the intelligent reflective surface according to the position of the UAV and the current communication environment, and sends a control instruction to the intelligent reflective surface. After receiving the instruction, the intelligent reflective surface immediately adjusts the reflection parameters to enhance the received signal strength of the UAV. At the same time, the UAV will also adjust its own flight attitude and communication frequency and other parameters according to the received signal quality feedback information, and work with the intelligent reflective surface to achieve the best communication effect.

[0014] Beneficial Effects

[0015] As can be seen from the above content, the intelligent reflective surface coverage enhancement method and device for power inspection provided by the present invention breaks through the limitation of the service capacity of a single intelligent reflective surface and ensures the consistency of drone service. By establishing a unified scheduling model of intelligent reflective surface and inspection equipment, the node capacity and channel state are modeled as a unified scheduling scale, and differential mapping is performed according to the inspection data transmission requirements to improve resource utilization, and an innovative performance space statistical average analysis method is used to obtain coverage and capacity space statistical average performance, which reflects the overall level of users. Based on random geometry tools, a semi-closed theoretical analysis expression of user coverage is given, and the multiplicative path loss is reduced by controlling the distance from the base station to the intelligent reflective surface, and the signal enhancement potential is excavated, which has high feasibility. The reasonable deployment of intelligent reflective surfaces can effectively overcome the signal shielding problem and guide the communication signal to the area where the drone is located, thereby expanding the communication coverage range and enabling the drone to perform power inspection in a wider area. Compared with traditional communication relay equipment, intelligent reflective surfaces have the advantage of low power consumption. The application of intelligent reflective surface technology in drone power inspection can reduce the transmission power of the drone communication module, thereby reducing the energy consumption of the drone and extending its flight time. Stable and reliable communication guarantees enable drones to conduct power inspections more efficiently. It can transmit inspection data back to the ground control station in real time for timely analysis and processing, while also reducing inspection interruptions and duplication of work caused by communication problems, thus improving inspection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1This is a model diagram of the intelligent reflective surface collaborative networking system suitable for drone inspection of the present invention;

[0017] Figure 2 It is a flowchart of the implementation of the scheme of the present invention; DETAILED DESCRIPTION

[0018] In order to clarify the purpose, implementation scheme and technical advantages of the present invention, the present invention is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0019] Figure 1 The practical application scenario of the intelligent reflective surface coverage enhancement method and device for power inspection of the present invention is: during the power inspection process, the UAV transmits inspection data to the data center, and the intelligent reflective surfaces at different locations are coordinated and networked to provide coverage of the UAV in a wider area. The present invention aims to assist the communication between the power inspection UAV and the data center.

[0020] According to the actual situation of the power line, select the appropriate drone model, and install inspection equipment such as high-definition cameras, infrared thermal imagers, and high-performance communication modules on the drone. The communication module should have strong anti-interference ability and long transmission distance to meet the communication needs of power inspection. Install the intelligent reflective surface system at key locations along the power line. For mountainous areas, choose to install the intelligent reflective panel on the top of the mountain or at a high place with a wide field of view, and ensure that it is firmly installed and can withstand the harsh natural environment. In urban high-rise areas, install the intelligent reflective surface on the exterior wall or roof of a suitable building, and adjust its direction and angle. On the pole tower of the power line, use a special mounting bracket to fix the intelligent reflective surface on the pole tower, and pay attention to avoid affecting the normal operation of the power line. Finally, establish a ground control station. The ground control station should be equipped with high-performance computer servers, communication equipment, and monitoring display screens. Connect the ground control station to the intelligent reflective surface system by wire or wireless means, and perform corresponding parameter configuration and initialization work.

[0021] After the system is built, comprehensive debugging is carried out. First, the flight performance of the drone, the working status of the inspection equipment, and the connection of the communication module are tested to ensure that the drone can fly and collect data normally, and communicate stably with the ground control station. Debug the intelligent reflector system. Send control instructions to the intelligent reflector through the ground control station to check whether the phase and amplitude adjustment of the reflective element are normal, and whether the reflection effect of the intelligent reflector on the signal meets expectations. During the debugging process, the deployment position, orientation, and angle of the intelligent reflector are fine-tuned according to the actual signal propagation situation to achieve the best signal reflection effect. Optimize the communication protocol and algorithm. In the actual test environment, collect the flight data, signal strength and other information of the drone, and further train and optimize the intelligent reflector control algorithm based on deep learning and the cooperative communication algorithm between the drone and the intelligent reflector, so that it can better adapt to the actual power inspection scene. By continuously adjusting the algorithm parameters and model structure, the accuracy and real-time performance of the algorithm are improved, ensuring that the intelligent reflector can adjust the reflection parameters in real time according to the dynamic changes of the drone to achieve the best communication effect.

[0022] When the system debugging and optimization is completed, the UAV power inspection task can be started. The ground control station sends task instructions to the UAV according to the inspection plan of the power line, including inspection route, flight altitude, shooting requirements and other information. After receiving the instructions, the UAV takes off according to the predetermined route and conducts power inspection. During the inspection process, the UAV collects images and data information of the power line in real time, and transmits the data back to the ground control station through the communication module. At the same time, the UAV continuously monitors the quality of the communication signal. When the signal quality is found to be reduced, it immediately sends a request signal to the ground control station according to the collaborative communication algorithm. After receiving the request, the ground control station calculates the optimal adjustment parameters of the intelligent reflective surface according to the current communication environment and the position of the UAV, and sends a control instruction to the intelligent reflective surface. After receiving the instruction, the intelligent reflective surface quickly adjusts the reflection parameters to enhance the receiving signal strength of the UAV. The ground control station receives the inspection data transmitted back by the UAV in real time, and analyzes and processes it. At the same time, the ground control station can also further optimize and adjust the control algorithm and deployment strategy of the intelligent reflective surface according to the inspection data to improve the overall performance and adaptability of the system.

[0023] Figure 2 A schematic diagram of a process of cooperative networking of intelligent reflective surfaces provided by the present invention includes:

[0024] Step 200, modeling the collaborative networking of multiple intelligent reflective surfaces and solving the conditional coverage probability. First, the communication space is divided into different grids. These grids are sufficiently small areas in the network to ensure that the performance index values ​​of the drones are the same when they are located at different positions in the same grid. Through this fine grid division, the propagation characteristics of the signal in space and the distribution of network performance can be more accurately described. According to factors such as signal attenuation, interference distribution, and topography, the space is reasonably divided into square or hexagonal grids with appropriate side lengths. Each grid is used as an independent analysis unit, which provides a basis for subsequent analysis and calculation. The base stations at each layer of the intelligent reflective surface network are divided into different states to further study the impact of different states on network performance. This state division can be based on multiple factors such as the working mode, load condition, and signal transmission power of the base station. The base station state is divided into a high-load operating state, a low-load operating state, a normal working state, etc., and each state corresponds to different network parameters and performance performance.

[0025] Analyze the probability of base stations at each layer of the intelligent reflector network being in different states. By collecting and statistically analyzing a large amount of network operation data, using probability theory and mathematical statistics methods, calculate the probability distribution of base stations at each layer in each state. Understand the frequency of base stations under different working conditions to provide a basis for network optimization design and resource allocation. By monitoring the load changes of base stations for a long time, count the time proportions of base stations in high load, low load and normal load states in different time periods, and obtain the corresponding probability values.

[0026] According to the performance of the samples in the network, the grids can be merged. When the performance index values ​​of adjacent grids are similar or meet certain similarity conditions, these grids are merged into a larger grid. This can simplify the complexity of the model and improve the calculation efficiency while ensuring the accuracy of the analysis. If the difference in performance indicators such as signal coverage probability and transmission rate of two adjacent grids is within a certain threshold range, they can be merged into one grid for subsequent analysis and processing. By continuously adjusting the division and merging of grids, the model can better adapt to different network scenarios and performance analysis requirements. The space is divided into different grids, and the base stations at each layer of the intelligent reflection surface network are divided into different states. The probability of the base stations at each layer of the intelligent reflection surface network being in different states and the performance index values ​​of the base stations at each layer of the intelligent reflection surface network when they are in different states are analyzed separately. The statistical averaging method in the multi-intelligent reflection surface group network is based on the large-scale spatial averaging of network performance, and divides the space into grids Φ i (i=0,1,2,...), so that A grid is a sufficiently small area in the network that the performance index value is the same when the drone is located at different positions in the same grid. Grids can be merged according to the performance of samples in the network. The spatial average coverage performance index is defined as

[0027]

[0028] Where a is the area of ​​the grid, f(a) is the probability density function of the area, and ρ th Indicates the load threshold; I i is the user set of the smart reflective surface, |h i,j | is the intelligent reflector-user link signal; I is the interference signal power, and W is the noise power. At this time, the grids belonging to the same area are merged, and the coverage rate is calculated in different areas. Calculate separately.

[0029] Step 210, decondition the state and traverse different grids to obtain the statistical average performance of the entire network. Further study the performance indicator values ​​of the base stations at each layer of the intelligent reflective surface network when they are in different states. For different base station states, analyze in detail their impact on network performance indicators, such as signal coverage, transmission rate, bit error rate, etc. By establishing mathematical models and simulation experiments, quantitatively evaluate the performance indicator values ​​under each state. When the base station is in a high-load state, due to intensified resource competition, the signal coverage may be reduced, the transmission rate may be reduced, and the bit error rate may be increased; while in a low-load state, network resources are relatively abundant and performance indicators may be improved. Among them, the statistical average performance can be expressed as

[0030]

[0031] Where PI is the de-conditioned network performance index value in the multi-intelligent reflector collaborative networking, and pi (state / grid = i) is the value of the state / grid in the grid Φ i The network performance index value when the multi-intelligent reflector network base station is in a specific state. P(gird=i) is the network performance index value when the multi-intelligent reflector network base station is in a specific state. i The probability of, and area(i) represents the grid Φ i Area represents the total area.

[0032] Step 220, based on the link model unique to the smart reflector, a unified smart reflector group joint signal processing model is constructed. A statistical averaging method based on the large-scale spatial average of network performance is adopted. The purpose of dividing the space into grids is to combine local and overall analysis of network performance. By statistically averaging the performance data in each grid, a representative performance index value of the grid can be obtained. Then, the performance indicators of all grids are comprehensively averaged in a large-scale space to obtain the performance evaluation results of the entire network. This method can more comprehensively reflect the distribution and changes of network performance in space. Considering the widely used linear signal processing technology, the signal processing gain brought by the service site is modeled as

[0033]

[0034] Where W represents the user combining matrix and F represents the transmitter precoding matrix. W and F with different mathematical structures correspond to different linear signal processing technologies, which in turn affect the distribution of signal processing gain.

[0035] In step 230, the coverage of the smart reflective surface is derived based on the general group signal processing model. A general group signal processing model suitable for the smart reflective surface system is constructed. The model comprehensively considers the working principle of the smart reflective surface, the signal propagation characteristics and the interaction with the surrounding environment. In the model, the signal emission source (base station or other communication equipment), the parameters of the smart reflective surface (including the number, size, reflection coefficient, etc. of the reflective surface) and the relevant characteristics of the receiving terminal (drone or mobile user equipment) are defined in detail. Through in-depth analysis of the path loss, fading characteristics and interference of the signal during propagation, a mathematical relationship between signal strength and parameters such as distance and angle is established. Taking into account the phase change and amplitude attenuation of the signal after reflection by the smart reflective surface, as well as the signal superposition effect between different paths, a suitable mathematical model is used to describe these complex physical processes.

[0036] Based on the above general group signal processing model, a strict definition of the coverage of the intelligent reflective surface is given. Coverage is usually defined as the ratio of the area or volume that can receive signals that meet certain quality requirements (signal strength threshold, signal-to-noise ratio threshold, etc.) to the total area or volume in a specific area. According to the signal processing model, the expression of the signal strength received at any location is calculated. Then, by comparing with the set quality threshold, it is determined whether the location is within the coverage range. Finally, the entire target area is traversed or integrated to obtain the size of the coverage area, thereby calculating the coverage.

[0037] Considering that the signal may go through multiple reflections and propagation paths from the transmitting source to the receiving terminal, the ray tracing method or other appropriate methods are used to calculate the total signal strength. For each possible propagation path, the attenuation and phase change of the signal on the path are calculated based on the path length, the reflection characteristics of the smart reflective surface, and the loss characteristics of the propagation medium. Then, the signals of all paths are superimposed to obtain the total signal strength expression at the receiving terminal.

[0038] The coverage is defined as:

[0039]

[0040] where g obeys Γ(M,θ).

[0041] Solve the log-Laplace transform L(s) of the interference term and the nth-order derivative after the transform. Due to the unique path loss structure of the multi-intelligent reflector network, it is necessary to convert the existing path loss structure into an elementary analytical form through progressive processing, that is,

[0042]

[0043] Considering the distance r between the intelligent reflector and the drone and the deployment height H, the complex road loss structure is asymptotically equivalent to an elementary analytical form within a specific interval, and the Laplace transform can be solved. According to the Leibniz formula, the linear relationship between the n-order Laplace transform derivatives is constructed, that is,

[0044]

[0045] The above relationship can be abstracted into a Toeplitz matrix through matrix representation. The solution of the n-th order derivative can be completed by only solving the coefficients of the Toeplitz matrix.

[0046] In step 240, consider a smart reflector-assisted cellular network with data center location dependence, where the data center set Φ B The distribution follows density λ B The black dotted line represents the Voronoi cell boundary. In order to reduce the “multiplicative fading” of the smart reflector cascade channel, the smart reflector is considered to be deployed near the data center. A clustered point process (MCP) is used to model the distribution of the smart reflector. B As the parent point process, the child point process is generated as the position of the smart reflective surface. Specifically, the smart reflective surface is distributed in a circular area centered on the data center in "clusters". In the circular area, the child points are evenly distributed. Therefore, in each cluster, the distance from the typical child point smart reflective surface to the parent point is distributed as follows:

[0047]

[0048] Where R is the cluster radius. The user distribution follows the density λ u According to Slivnyaks theorem, the properties of the Poisson point process Φ observed at any point are the same, so the performance of the typical user can represent the performance of the entire network. Without loss of generality, the signal transmission of typical users is considered. Each data center serves at most one user in each time slot, that is, intra-cell time division multiple access (TDMA) is adopted, so there is no intra-cell interference.

[0049] All data centers send signals with the same power, and users select the data center with the largest average received power as the service data center, that is, users connect to the data center closest to them, and all other data centers act as interference. The data center closest to a typical user is B0, and the distance between the typical user and B0 is l0. Based on the characteristics of the Poisson point process, it can be inferred that l0 follows the Rayleigh distribution, and its probability density distribution is:

[0050]

[0051] In addition, the intelligent reflector is scheduled in a time division multiplexing manner, that is, only one reflector assists B0 in sending signals to typical users in each time slot. In order to control at least one of the two distances of the intelligent reflector cascade channel to be shorter, two different intelligent reflector connection mechanisms are used for users inside and outside the intelligent reflector distribution cluster: for users in a circle with a radius of R centered on the data center, the auxiliary intelligent reflector is the intelligent reflector closest to the user; and for users outside the cluster, the auxiliary intelligent reflector is the intelligent reflector closest to the data center. Due to the passive characteristics of the intelligent reflector, the intelligent reflector needs to perform coordinated signal processing with the data center to form a data center-intelligent reflector-user cascade channel. The data center sends a signal, and the intelligent reflector applies an additional phase to the cascade channel through a large number of independently controlled reflection units, so that the phases of each independent reflection signal are aligned, thereby achieving directional signal enhancement.

[0052] The data center and inspection equipment are equipped with an antenna, and each smart reflector has N independent reflective elements. Data Center B i With reflective surface IRS j The distance between IRS i The distance from the typical user, and Respectively represent the data from data center B i To the intelligent reflective surface IRS j and intelligent reflective surface IRS j The channel matrix to the typical inspection equipment UE, where represents a complex matrix of dimension N×1, [·] T Represents matrix transpose. Data center B i With reflective surface IRS j The nth reflection unit Channels between Modeled as:

[0053]

[0054] in, represents the average channel fading coefficient, β=(4πf c / c) -2 represents the average signal power gain when the reference distance is 1m, f c represents the system carrier frequency, c represents the speed of light, c≈3×10 8 m / s, α≥2 is the path loss exponent. Data Center B i The channel between the intelligent reflector and the reflector unit and the channel from the intelligent reflector to the typical inspection equipment are subject to Rayleigh fading, θ d,i represents the phase of the incident channel. Similarly, from the IRS j The nth reflection unit Channel to typical inspection equipment UE Modeled as:

[0055]

[0056] Considering Rayleigh small-scale fading, ζ ij,n ~CN(0,1),ζ jU,n ~CN(0,1), indicating that ζ ij,n and jU,n It obeys the circularly symmetric complex Gaussian (CSCG) distribution with zero mean and variance 1, so its modulus |ζ ij,n | and ζ jU,n |Subject to shape parameters Rayleigh distribution, θ jU,n Represents the phase of the reflection channel.

[0057] use represents the reflection characteristic of the jth smart reflective surface, and the reflection coefficient of the nth reflective unit is where β n and θ j,n They represent amplitude response and phase shift respectively. For simplicity, only phase-controlled smart reflectors are considered, and amplitude-controlled smart reflectors are not considered. That is, in a passive scenario, β1=β2=…=β N = 1. That is, the cascade channel is established in the form of multiplicative path loss between the data center-intelligent reflector channel and the intelligent reflector-inspection equipment channel as follows:

[0058]

[0059] in,

[0060]

[0061] θ j,nis the additional phase of the reflector. If the reflector under consideration is an auxiliary reflector, the phase can be adjusted by applying a control signal to the reflective element on the auxiliary reflector. The smart reflector adjusts the phase of the cascaded channel formed by its independent reflection to zero, that is, the phase θ of each independent reflector unit j,n is θ j,n =-(θ ij,n +θ jU,n ), so that the phases of the independent reflection cascade channels are aligned to achieve the effect of passive beamforming, then:

[0062]

[0063] Under Rayleigh small-scale fading, the cascade channel coefficient |ζ ij,n ||ζ jU,n | is the product of two Rayleigh random variables, called a "double-Rayleigh" random variable. For a double-Rayleigh distribution random variable, its mean and variance are as follows:

[0064] E{|ζ ij,n ||ζ jU,n |}=π / 4

[0065]

[0066] The introduction of cascaded links changes the mathematical structure of the original beamforming algorithm, and requires joint beamforming and phase unification for optimization design. First, the cascaded channel is modeled using the Rayleigh channel model. For a single intelligent reflector service system, the intelligent reflector has M reflective units, and a unified channel model is modeled. The cascade channel from the tth antenna of the BS to the kth user through the mth reflective unit can be modeled as follows, considering LOS and NLOS multipath components. and

[0067]

[0068] Among them, the LOS channel part can be expressed as

[0069]

[0070] Among them, ψ m is the phase of the mth unit.

[0071] Then model the incident K A (m) and the normalized power radiation model of the reflected signal,

[0072] K A (m)=|cos 3 θ A (m)|,θ A(m)∈(0,π)

[0073]

[0074] Refers to the reflected signal, Refers to the transmission signal. The reflection / transmission mode of the smart reflection surface is uniformly modeled by setting the reflection and transmission signal power ratio ε.

[0075] In summary, the present invention has made innovations and in-depth research in the reconfigurable feature modeling, reconfigurable capability boundary analysis, and location deployment of multi-intelligent reflective surface collaborative networking for assisting UAV power inspection, providing a comprehensive and efficient solution for application scenarios such as UAV power inspection. These technological innovations will strongly promote the widespread application and development of multi-intelligent reflective surface collaborative networking technology in the field of wireless communications, and make important contributions to improving communication quality and system performance.

Claims

1. A method and device for enhancing intelligent reflective surface coverage for power inspection, characterized in that: include: For the intelligent reflective surface collaborative networking system, a unified scheduling model for intelligent reflective surfaces and inspection drones is established; Derive analytical expressions for signal interference distribution, solve for spatial measure functions of point processes and heterogeneous site density based on determined interference levels; According to the unique link model of the smart reflector, a unified smart reflector group joint signal processing gain model is constructed; based on the general group signal processing model, the smart reflector coverage is derived; Construct a multi-intelligent reflective surface networking correlation model, analyze the link coupling factor, and study the link performance under collaborative networking based on the modeling of the link coupling factor; explore the cross-domain reconfigurable capability boundaries, and innovate the spatial statistical average analysis method of coverage, capacity, and mobility performance; propose an adjustable intelligent reflective surface distribution dependency model, and give a semi-closed theoretical analysis expression for user coverage based on random geometry tools.

2. The method according to claim 1, characterized in that The smart reflective surface collaborative networking system establishes a unified scheduling model for the smart reflective surface and the inspection drone, specifically including: The statistical averaging method in the intelligent reflector group network is based on the large-scale spatial average of network performance, dividing the space into grids Φ i (i=0,1,2,...), so that A grid is a sufficiently small area in the network. The performance index value is the same when the drone is located at different positions in the same grid. According to the performance of the sample in the network, the grid can be merged and the spatial average coverage performance index is defined as Where a is the area of ​​the grid, f(a) is the probability density function of the area, and ρ th Indicates the load threshold; I i is the user set of the smart reflective surface, |h i,j | is the intelligent reflector-user link signal; I is the interference signal power, W is the noise power. At this time, the grids belonging to the same area are merged, and the coverage rate is calculated in different areas. Calculate separately.

3. The method according to claim 1, characterized in that The derivation of the signal interference distribution analytical expression, solving the spatial measurement function of the point process and the heterogeneous site density based on the determined interference level specifically includes: The state is deconditioned and different grids are traversed to obtain the statistical average performance of the entire network. The statistical average performance can be expressed as Where PI is the de-conditioned network performance index value in the multi-intelligent reflector collaborative networking, and pi (state / grid = i) is the value of the state / grid in the grid Φ i The network performance index value when the multi-intelligent reflector network base station is in a specific state, P(gird=i) is the network performance index value when the multi-intelligent reflector network base station is located in the grid Φ i The probability of, and area(i) represents the grid Φ i Area represents the total area.

4. The method according to claim 1, characterized in that: The method of constructing a unified smart reflector group joint signal processing gain model according to the link model unique to the smart reflector specifically includes: A multi-intelligent reflector network correlation model is constructed to analyze the link coupling factor. The performance indicators of all grids are comprehensively averaged in a large-scale space to obtain the performance evaluation results of the entire network. The signal processing gain brought by the service site is modeled as Wherein, W represents the user combining matrix, F represents the transmitting end precoding matrix, and W and F with different mathematical structures correspond to different linear signal processing technologies, thereby affecting the distribution of signal processing gain.

5. The method according to claim 1, characterized in that The derivation of the intelligent reflective surface coverage rate based on the general group signal processing model specifically includes: According to the path length, the reflection characteristics of the smart reflective surface and the loss characteristics of the propagation medium, the attenuation and phase change of the signal on the path are calculated, and then the signals of all paths are superimposed to obtain the total signal strength expression at the receiving terminal, and the coverage rate of the smart reflective surface can be obtained: where g obeys Γ(M,θ).

6. The method according to claim 5, characterized in that Solve the coverage rate of the smart reflective surface, including: The interference term log-Laplace transform L(s) and the transformed n-order derivative are solved, and the existing path loss structure is equivalent to the elementary analytical form through asymptotic processing, that is, Considering the distance r between the intelligent reflector and the drone and the deployment height H, the complex road loss structure is asymptotically equivalent to an elementary analytical form within a specific interval, and the Laplace transform can be solved. According to the Leibniz formula, the linear relationship between the n-order Laplace transform derivatives is constructed, that is, The above relationship can be abstracted into a Toeplitz matrix through matrix representation. The solution of the n-th order derivative can be completed by only solving the coefficients of the Toeplitz matrix.

7. The method according to claim 1, characterized in that The exploration of cross-domain reconfigurable capability boundaries and the innovative spatial statistical average analysis methods of coverage, capacity, and mobility performance specifically include: The data center and the inspection drone are equipped with an antenna, and each smart reflective surface has N independent reflective elements. Indicates data center B i With reflective surface IRS j The distance between IRS i The distance from the typical user, and Respectively represent the data from data center B i To the intelligent reflective surface IRS j and intelligent reflective surface IRS j The channel matrix to the typical inspection equipment UE, where represents a complex matrix of dimension N×1, [·] T Represents matrix transpose, data center B i With reflective surface IRS j The nth reflection unit Channels between Modeled as: in, represents the average channel fading coefficient, β=(4πf c / c) -2 represents the average signal power gain when the reference distance is 1m, f c represents the system carrier frequency, c represents the speed of light, c≈3×10 8 m / s, α≥2 is the path loss index, data center B i The channel between the intelligent reflector and the reflector unit and the channel from the intelligent reflector to the typical inspection equipment are subject to Rayleigh fading, θ d,i Represents the phase of the incident channel.

8. The method according to claim 1, characterized in that The intelligent reflective surface distribution dependence model and the semi-closed theoretical analysis expression of the user coverage rate based on the random geometry tool specifically include: Consider deploying smart reflective surfaces near data centers and use a clustering point process (MCP) to model the distribution of smart reflective surfaces. B As the parent point process, the child point process is used as the position of the smart reflection surface. Specifically, the smart reflection surface is distributed in a circular area centered on the data center in "clusters". In the circular area, the child points are evenly distributed. Therefore, in each cluster, the distance distribution from the typical child point smart reflection surface to the parent point is as follows: Where R is the cluster radius.

Images