Communication optimization method and device of wireless communication system and wireless communication system

Abstract

The invention discloses a communication optimization method and device of a wireless communication system and the wireless communication system, and belongs to the field of communication optimizing.The method is characterized in that the wireless communication system comprises a plurality of intelligent reflecting surfaces, and the method comprises the steps that a preset evaluation model is called to obtain target parameters of the intelligent reflecting surfaces; and calculating an adjustment parameter of each intelligent reflecting surface by using the target parameter based on an adaptive algorithm, and performing communication optimization processing on each intelligent reflecting surface by using the adjustment parameter. According to the method, the target parameters required by optimization of the intelligent reflecting surface in the current environment can be predicted through the evaluation model trained by the dynamic characteristic parameters, and then optimization adjustment can be performed according to the target parameters, so that actual optimization requirements are met, and the optimization precision and accuracy are improved; and moreover, the number of equipment does not need to be additionally increased, and the technical problems of poor optimization effect and high optimization cost in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the field of communication optimization, and particularly to a communication optimization method, device and wireless communication system for a wireless communication system. Background Art

[0002] Intelligent Reflecting Surface (IRS) is a new type of wireless communication technology, also known as Reconfigurable Intelligent Surface (RIS) or Intelligent Metasurface. IRS is an artificial material or device with adjustable reflection characteristics. In a wireless communication system, a large number of adjustable intelligent reflecting surfaces are generally deployed, and through cooperation with base stations and network architectures, signal transmission is carried out.

[0003] With the popularization of the Internet of Things and intelligent devices, the number of access devices in the network will increase significantly, and effective resource management and scheduling strategies are required to ensure the stability and efficiency of the network. Therefore, signal optimization of the wireless communication system is needed. Currently, the commonly used optimization method is to perform channel modeling according to the structure of the wireless communication system, and then adjust the base station and intelligent reflecting surface (including the number and communication parameters) in combination with the measured signal propagation path to increase the transmission power and achieve signal optimization.

[0004] However, the currently commonly used optimization method has the following technical problems: The base station and intelligent reflecting surface are generally set in an outdoor environment. In a complex wireless propagation environment, due to effects such as multipath propagation, reflection, and scattering, the measured signal propagation path deviates greatly from the actual situation, making it difficult for the subsequent optimization results to meet the actual requirements, with poor optimization effects. Moreover, increasing the number of base stations and intelligent reflecting surfaces requires a large amount of infrastructure construction and maintenance, which also increases the overall cost of the network. Summary of the Invention

[0005] The present invention provides a communication optimization method, device and wireless communication system for a wireless communication system, which can solve the technical problems of poor optimization effect and high optimization cost in the prior art.

[0006] The communication optimization method for the wireless communication system of the present invention, the wireless communication system includes a plurality of intelligent reflecting surfaces, and the method includes:

[0007] Invoking a preset evaluation model to obtain the target parameters of the intelligent reflecting surface, wherein the preset evaluation model is a model trained using dynamic characteristic parameters of the environment where the wireless communication system is located, and the dynamic characteristic parameters are channel state data collected at the base station node, user equipment node, and intelligent reflecting surface node of the wireless communication system;

[0008] Calculate the adjustment parameters of each intelligent reflecting surface using the target parameters based on an adaptive algorithm. The adjustment parameters include: reflection phase and amplitude, and use the adjustment parameters to perform communication optimization processing on each intelligent reflecting surface.

[0009] In one implementation, the evaluation model trained by the dynamic characteristic parameters of the present invention can predict the target parameters required for optimizing the intelligent reflecting surface in the current environment, and then can perform optimization adjustment according to the target parameters, thereby fitting the actual optimization requirements to improve the optimization accuracy and accuracy; and there is no need to additionally increase the number of devices, which can reduce the optimization cost and the overall cost of the communication network.

[0010] Further, the preset evaluation model includes a combination of a deep neural network and a convolutional neural network. The update operation of the preset evaluation model includes:

[0011] Obtain real-time parameters and prediction parameters, where the prediction parameters are the channel data predicted by the deep neural network based on the preprocessed channel state data;

[0012] Call the convolutional neural network to calculate the model weights according to the real-time parameters and the prediction parameters;

[0013] Update the weights of the deep neural network using the model weights.

[0014] Further, the training operation of the preset evaluation model includes:

[0015] Use a preset environment model to collect the dynamic change parameters of the environment where the wireless communication system is located in real time;

[0016] Preprocess the dynamic change parameters to obtain processed parameters. The preprocessing includes: denoising, normalization, and feature extraction;

[0017] Use the processed parameters to perform model training on the deep learning model to obtain a preset evaluation model;

[0018] Among them, the preset evaluation model is shown as follows:

[0019]

[0020] Among them, is the predicted target parameter, y i is the actually measured parameter, N is the number of samples, and L(θ) is the minimized loss function.

[0021] Further, the using the adjustment parameters to perform communication optimization processing on each intelligent reflecting surface includes:

[0022] Obtain real-time communication parameters, where the real-time communication parameters are signal parameters measured in real time;

[0023] Determine the feedback signal of each intelligent reflecting surface according to the real-time communication parameters and the adjustment parameters according to a low-latency feedback control mechanism;

[0024] Use the feedback signal of each intelligent reflecting surface to perform communication optimization processing on each intelligent reflecting surface;

[0025] Wherein, the feedback signal is shown in the following formula:

[0026] F=(Y actual -Y predicted )

[0027] Wherein, F is the feedback signal, Y actual is the real-time communication parameter, and Y predicted is the adjustment parameter.

[0028] Further, the using the feedback signal of each intelligent reflecting surface to perform communication optimization processing on each intelligent reflecting surface includes:

[0029] Construct the feedback signal of each intelligent reflecting surface into a control task based on a distributed control method;

[0030] Assign the control task to each corresponding intelligent reflecting surface for optimizing the adjustment of the communication configuration of each intelligent reflecting surface.

[0031] Further, after the step of calculating the adjustment parameter of each intelligent reflecting surface by using the target parameter based on the adaptive algorithm, the method further includes:

[0032] Obtain channel gain parameters, where the channel gain parameters are the channel gains between each user equipment of the wireless communication system and the base station;

[0033] Adjust the transmission rate of each user equipment by using the channel gain parameters and the adjustment parameters.

[0034] Further, after the step of adjusting the transmission rate of each user equipment by using the channel gain parameters and the adjustment parameters, the method further includes:

[0035] Obtain the base station communication information of the base station, where the base station communication information is the parameters of the channel quality and interference level of the user equipment;

[0036] Optimize the adjustment parameter by using the base station communication information.

[0037] Another embodiment of the present invention further provides a communication optimization device for a wireless communication system, the wireless communication system includes a plurality of intelligent reflecting surfaces, and the device includes:

[0038] An acquisition module, configured to call a preset evaluation model to obtain target parameters of an intelligent reflecting surface, where the preset evaluation model is a model trained using dynamic characteristic parameters of the environment where the wireless communication system is located, and the dynamic characteristic parameters are channel state data collected at a base station node, a user equipment node, and an intelligent reflecting surface node of the wireless communication system;

[0039] An optimization module, configured to calculate adjustment parameters for each intelligent reflecting surface based on an adaptive algorithm using the target parameters, where the adjustment parameters include: reflection phase and amplitude, and perform communication optimization processing on each intelligent reflecting surface using the adjustment parameters.

[0040] Further, the preset evaluation model includes a combination of a deep neural network and a convolutional neural network, and the update operation of the preset evaluation model includes:

[0041] Obtain real-time parameters and prediction parameters, where the prediction parameters are channel data predicted by the deep neural network based on the preprocessed channel state data;

[0042] Call the convolutional neural network to calculate model weights based on the real-time parameters and the prediction parameters;

[0043] Update the weights of the deep neural network using the model weights.

[0044] Further, the training operation of the preset evaluation model includes:

[0045] Use a preset environment model to collect dynamic change parameters of the environment where the wireless communication system is located in real time;

[0046] Preprocess the dynamic change parameters to obtain processed parameters, and the preprocessing includes: denoising, normalization, and feature extraction;

[0047] Use the processed parameters to perform model training on a deep learning model to obtain a preset evaluation model;

[0048] Wherein, the preset evaluation model is shown as the following formula:

[0049]

[0050] Wherein, is the predicted target parameter, y i is the actually measured parameter, N is the number of samples, and L(θ) is the minimized loss function.

[0051] Further, the performing communication optimization processing on each intelligent reflecting surface using the adjustment parameters includes:

[0052] Obtain real-time communication parameters, where the real-time communication parameters are signal parameters measured in real time;

[0053] Determine the feedback signal of each intelligent reflecting surface according to the real-time communication parameters and the adjustment parameters according to a low-latency feedback control mechanism;

[0054] Perform communication optimization processing on each intelligent reflecting surface by using the feedback signal of each intelligent reflecting surface;

[0055] Wherein, the feedback signal is shown in the following formula:

[0056] F=(Y actual -Y predicted )

[0057] Wherein, F is the feedback signal, Y actual is the real-time communication parameter, and Y predicted is the adjustment parameter.

[0058] Further, the performing communication optimization processing on each intelligent reflecting surface by using the feedback signal of each intelligent reflecting surface includes:

[0059] Construct the feedback signal of each intelligent reflecting surface into a control task based on a distributed control method;

[0060] Allocate the control task to each corresponding intelligent reflecting surface for optimizing and adjusting the communication configuration of each intelligent reflecting surface.

[0061] Further, after the step of calculating the adjustment parameter of each intelligent reflecting surface by using the target parameter based on the adaptive algorithm, the method further includes:

[0062] Obtain channel gain parameters, where the channel gain parameters are the channel gains between each user equipment of the wireless communication system and the base station;

[0063] Adjust the transmission rate of each user equipment by using the channel gain parameters and the adjustment parameters.

[0064] Further, after the step of adjusting the transmission rate of each user equipment by using the channel gain parameters and the adjustment parameters, the method further includes:

[0065] Obtain the base station communication information of the base station, where the base station communication information is the parameters of the channel quality and interference level of the user equipment;

[0066] Optimize the adjustment parameter by using the base station communication information.

[0067] Another embodiment of the present invention further provides a communication optimization system for a wireless communication system, which is applicable to the communication optimization method of the wireless communication system as described above. The wireless communication system includes: a control platform, a base station, and user equipment. The control platform controls multiple intelligent reflecting surfaces, and the base station, the user equipment, and the multiple intelligent reflecting surfaces are communicatively connected to each other.

[0068] Further, the intelligent reflecting surface is constructed using a conductive polymer or a ceramic material.

[0069] Another embodiment of the present invention further provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps of the communication optimization method of the wireless communication system of the present invention are implemented.

[0070] Another embodiment of the present invention further provides a computer-readable storage medium item, including: a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the steps of the communication optimization method of the wireless communication system of the present invention.

[0071] By implementing the present invention, the following beneficial effects are achieved:

[0072] The present invention can call a preset evaluation model to obtain the target parameters of the intelligent reflecting surface, where the preset evaluation model is a model trained using the dynamic characteristic parameters of the environment where the wireless communication system is located; based on an adaptive algorithm, the adjustment parameters of each intelligent reflecting surface are calculated using the target parameters, and the communication optimization process is performed on each intelligent reflecting surface using the adjustment parameters. The evaluation model trained using the dynamic characteristic parameters can predict the target parameters required for optimizing the intelligent reflecting surface in the current environment, and then the optimization adjustment can be made according to the target parameters, thereby fitting the actual optimization requirements to improve the optimization accuracy and accuracy; and there is no need to additionally increase the number of devices, which can reduce the optimization cost and the overall cost of the communication network. Description of the Drawings

[0073] To more clearly illustrate the technical solutions of the present application, the drawings required for implementation will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0074] Figure 1 It is a flowchart of a communication optimization method for a wireless communication system provided by an embodiment of the present invention;

[0075] Figure 2It is a schematic structural diagram of a communication optimization device for a wireless communication system provided by an embodiment of the present invention;

[0076] Figure 3 It is a schematic structural diagram of a wireless communication system provided by an embodiment of the present invention;

[0077] Figure 4 It is a schematic circuit diagram of an intelligent reflecting surface provided by an embodiment of the present invention. Detailed implementation manners

[0078] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Apparently, the described embodiments are some but not all of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in this application belong to the scope of protection of this application.

[0079] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above drawings are intended to cover non-exclusive inclusion.

[0080] In the description of the embodiments of this application, technical terms such as "first" and "second" are only used to distinguish different objects and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order, or primary-secondary relationship of the indicated technical features. In the description of the embodiments of this application, "a plurality of" means two or more unless otherwise specifically defined.

[0081] Referring to "embodiments" herein means that specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of this application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0082] In the description of the embodiments of this application, the term " / and" is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0083] In the description of the embodiments of the present application, the term "a plurality of" means two or more (including two). Similarly, "a plurality of groups" means two or more groups (including two groups), and "a plurality of pieces" means two or more pieces (including two pieces).

[0084] In the description of the embodiments of the present application, unless otherwise clearly defined and limited, technical terms such as "installation", "connection", "connection", "fixation", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can also be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication between two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific situations.

[0085] An Intelligent Reflecting Surface (IRS) is a new type of wireless communication technology, also known as a Reconfigurable Intelligent Surface (RIS) or an Intelligent Metasurface. IRS is an artificial material or device with adjustable reflection characteristics. In a wireless communication system, a large number of adjustable intelligent reflecting surfaces are generally deployed, and through cooperation with the base station and the network architecture, signal transmission is carried out.

[0086] With the popularization of the Internet of Things and intelligent devices, the number of access devices in the network will increase significantly, and effective resource management and scheduling strategies are required to ensure the stability and efficiency of the network. Therefore, signal optimization of the wireless communication system is required. Currently, the commonly used optimization method is to model the channel according to the structure of the wireless communication system, and then adjust the base station and the intelligent reflecting surface (including the number and communication parameters) in combination with the measured signal propagation path to increase the transmission power and achieve signal optimization.

[0087] However, the commonly used optimization methods at present have the following technical problems: The base station and the intelligent reflecting surface are generally set in an outdoor environment. In a complex wireless propagation environment, due to effects such as multipath propagation, reflection, and scattering, the measured signal propagation path deviates greatly from the actual situation, making it difficult for the subsequent optimization results to meet the actual requirements, resulting in poor optimization effects. Moreover, increasing the number of base stations and intelligent reflecting surfaces requires a large amount of infrastructure construction and maintenance, which also increases the overall cost of the network.

[0088] To solve the technical problems of poor communication optimization effect and high optimization cost in the prior art, refer to Figure 1 , which shows a schematic flowchart of a communication optimization method for a wireless communication system provided by an embodiment of the present invention.

[0089] In one embodiment, the method is applicable to a base station or a router of a wireless communication system, specifically, it can be a base station, a router, etc. corresponding to a network node close to a user equipment. In specific operations, the present invention can adopt edge computing technology. Edge computing is to transfer the computing tasks from the central server to the network nodes (such as base stations, routers, etc.) close to the user for processing, which can reduce the computing pressure on the central server and shorten the data transmission delay, thereby improving the response speed of the system and enhancing the optimized processing efficiency.

[0090] Among them, as an example, the communication optimization method of the wireless communication system may include the following sub-steps:

[0091] S11. Invoke a preset evaluation model to obtain the target parameters of the intelligent reflecting surface. Among them, the preset evaluation model is a model trained using the dynamic characteristic parameters of the environment where the wireless communication system is located, and the dynamic characteristic parameters are channel state data collected at the base station node, user equipment node, and intelligent reflecting surface node of the wireless communication system.

[0092] In one embodiment, the preset evaluation model can be a deep learning model. The deep learning model can learn the rules of signal optimization from a large amount of historical data, thereby improving the adaptive ability of the reflecting unit. The present invention uses a machine learning model to deeply analyze the channel state, reflecting unit configuration, and environmental changes to obtain the target parameters of the intelligent reflecting surface.

[0093] For example, a deep neural network (DNN) or a recurrent neural network (RNN) can be trained using historical channel quality data and network state data, and the future channel quality and network load can be predicted based on these models to obtain the target parameters of the intelligent reflecting surface. Based on these prediction results, the configuration of the intelligent reflecting surface can be adjusted to optimize the network performance. This method can greatly improve the intelligence level of the wireless communication system, reduce manual intervention, and improve the automation degree of the system.

[0094] Among them, the historical channel quality data and network state data can be data from multiple sensors, including the location information of the user equipment, base station feedback, signal quality measurement, etc.

[0095] To improve the adaptability of the wireless communication system, the present invention can collect the corresponding parameters of various dynamic change factors to use the above parameters and data for model training to obtain the preset evaluation model.

[0096] Among them, as an example, the training operation of the preset evaluation model may include the following sub-steps:

[0097] S21. Use a preset environment model to collect the dynamic change parameters of the environment where the wireless communication system is located in real time.

[0098] S22. Preprocess the dynamically varying parameters to obtain processed parameters, where the preprocessing includes: denoising, normalization, and feature extraction;

[0099] S23. Use the processed parameters to train a deep learning model to obtain a preset evaluation model;

[0100] Among them, the preset evaluation model is shown as follows:

[0101]

[0102] Among them, is the predicted target parameter, y i is the actually measured parameter, N is the number of samples, and L(θ) is the minimized loss function.

[0103] In specific operations, the present invention can monitor and analyze various dynamically varying factors in real time through an environment perception algorithm, including user location, signal strength, user movement state, building changes, weather conditions, etc., to obtain dynamically varying parameters. These dynamically varying parameters have a significant impact on the propagation characteristics of wireless signals. Therefore, it is necessary to accurately obtain and analyze these dynamically varying parameters in order to dynamically adjust the configuration of the reflection unit.

[0104] The environment perception algorithm can combine data from multiple sensors, such as the location information of user equipment, base station feedback, signal quality measurement, etc., to construct an environment model in real time. Through the deep learning model, the current environment can be evaluated at any time, and based on these evaluation results, the reflection phase, amplitude, and direction of the intelligent reflecting surface can be adjusted to optimize the signal propagation path and communication quality to the greatest extent.

[0105] Specifically, real-time signal data and environment information are collected from nodes such as user equipment, base stations, and sensors to obtain dynamically varying parameters. The collected dynamically varying parameters (including data such as signal strength, location information, and weather changes) are preprocessed, such as denoising, normalization, feature extraction, etc. The deep learning model is used to model the environmental characteristics and predict the signal change trend, thereby obtaining the target parameter. Subsequently, the parameters of the intelligent reflecting surface can be adjusted according to the predicted target parameter to optimize the signal propagation path and network performance of the wireless communication system.

[0106] Through this intelligent method, high-quality communication services can be maintained under the influence of factors such as user movement, weather changes, and building occlusion, and the changes and interferences of wireless signals can be effectively addressed.

[0107] In one operating mode, channel state data can be collected by nodes such as base stations, user equipment, and reflection units, including factors such as signal strength, user location, building obstruction, and weather changes. These channel state data will be preprocessed through time-domain and frequency-domain analysis and standardized into a format suitable for input to a deep learning model. The input data X contains the spatial and temporal characteristics of the signal, as follows:

[0108] X = [x1, x2, …, x n ;

[0109] where x1 is the feature of the i-th signal sample, including location, signal strength, interference, etc.

[0110] During the training process, a deep neural network (DNN) is used in combination with a convolutional neural network (CNN). The goal of the model is to predict the channel state information (CSI) through the training data X to obtain the target parameters.

[0111] Among them, the CNN is used to extract signal features from the time domain and frequency domain, while the DNN further learns the non-linear relationship between the signal and the channel state. The weights and biases of the model are optimized by minimizing the loss function, and the mean squared error (MSE) is used as the loss function:

[0112]

[0113] where is the predicted CSI, that is, the target parameter, y i is the actual CSI, that is, the real-time parameter, and N is the number of training data. The model optimizes the adjustment strategy of the reflection unit by minimizing the loss function L(θ).

[0114] In the deep learning model, channel state estimation is obtained by non-linearly transforming the input data X through a multi-layer neural network to obtain the predicted channel state

[0115]

[0116] where f(.) represents the output function of the deep learning model, and W and b are the weights and biases of the model respectively.

[0117] After the model training is completed, the system can predict the current channel state information (CSI) according to the real-time environmental data (such as user location, signal strength, etc.). Assuming that the current input data is X t , then through the deep learning model f, the real-time channel state estimation can be obtained

[0118]

[0119] Here, is the channel state information predicted at time t. Based on this CSI, the intelligent reflecting surface will perform corresponding configuration adjustments.

[0120] To improve the estimation accuracy of the model and adapt to environmental changes. Among them, as an example, the preset evaluation model includes the combination of a deep neural network and a convolutional neural network. The update operation of the preset evaluation model may include the following sub-steps:

[0121] S31. Obtain real-time parameters and prediction parameters. Among them, the prediction parameters are the channel data predicted by the deep neural network based on the preprocessed channel state data.

[0122] S32. Call the convolutional neural network to calculate the model weights according to the real-time parameters and the prediction parameters.

[0123] S33. Update the weights of the deep neural network with the model weights.

[0124] In one embodiment, the present invention can adopt a real-time feedback mechanism. Each time the channel estimation and the measurement of the actual signal are fed back to the deep learning model for online update. Assuming the feedback signal is F, the model updates the weights according to the feedback signal F:

[0125]

[0126] where η is the learning rate, is the gradient of the loss function with respect to the weights. This process allows the model to adjust the parameters according to the feedback of the actual environment, thereby further improving the accuracy and real-time performance of the channel estimation.

[0127] Based on the accurate target parameter (CSI), the adjustment strategy of each intelligent reflecting surface can be optimized. The adjustment of the intelligent reflecting surface depends on the target parameter (CSI). Assuming the reflection coefficient of the i-th intelligent reflecting surface is θ i , then its optimization goal is to maximize the signal strength:

[0128]

[0129] where A(θ) is the adjustment matrix of the intelligent reflecting surface, and h i is the gain of the i-th channel.

[0130] Through the deep learning model, the prediction of the channel state information and the adjustment of the reflection unit can be mathematically represented by the following process:

[0131] Channel estimation:

[0132]

[0133] Feedback Update:

[0134]

[0135] Adjustment of Intelligent Reflecting Surface:

[0136]

[0137] In this process, the channel estimation is continuously optimized through a deep learning model, and the intelligent reflecting surface is dynamically adjusted according to real-time feedback, significantly improving the accuracy of channel estimation and the configuration efficiency of reflection units.

[0138] Through the channel estimation method based on deep learning, combined with a real-time feedback mechanism, the following technical effects can be achieved:

[0139] First, dynamic channel adaptation: Through the deep learning model, the channel estimation can be adjusted and adapted to environmental changes in real time, avoiding the limitations of traditional static models.

[0140] Second, high-precision channel estimation: Deep learning can capture complex signal changes and improve the accuracy of channel estimation, especially in complex environments such as multipath propagation, reflection, and occlusion.

[0141] Third, reduced estimation delay: The combination of edge computing and the feedback mechanism reduces the delay in the channel estimation process, ensuring that the reflection units can respond quickly and adjust the configuration.

[0142] Fourth, enhanced adaptability: The real-time feedback and online learning mechanism enable the system to continuously optimize the channel estimation and adapt to the dynamic environment, further improving the performance of the wireless communication system.

[0143] Through these innovative methods, the present invention significantly improves the performance of the wireless communication system in complex environments and provides strong support for the next-generation wireless communication network.

[0144] S12. Calculate the adjustment parameters of each intelligent reflecting surface based on the adaptive algorithm using the target parameters. The adjustment parameters include: reflection phase and amplitude, and use the adjustment parameters to perform communication optimization processing on each intelligent reflecting surface.

[0145] In one embodiment, the adjustment parameters of each intelligent reflecting surface can be calculated based on the adaptive algorithm using the target parameters. The adjustment parameters include: reflection phase and amplitude, and then use the adjustment parameters to perform communication optimization processing on each intelligent reflecting surface. By precisely controlling the reflection phase and amplitude of the intelligent reflecting surface, precise control of electromagnetic waves can be achieved, thereby improving the signal propagation path, reducing signal interference, and increasing signal strength.

[0146] The optimization process of the present invention can be achieved by controlling the reflection phase θ of the signal iand amplitude A i to optimize the signal propagation. After receiving the signal, the intelligent reflecting surface dynamically adjusts these parameters according to various factors such as the current network environment, channel state information (CSI), signal strength, interference level, etc. The goal is to maximize the gain of the reflected signal and reduce interference, while optimizing the energy efficiency of the wireless communication system.

[0147] To achieve precise adjustment of the intelligent reflecting surface, the present invention can adopt an adaptive control algorithm. This algorithm can dynamically adjust the parameters of the reflecting unit according to real-time channel information and network status, thereby optimizing signal propagation. The core of the adaptive control algorithm is to continuously adjust the parameters of the reflecting unit through online learning and real-time feedback to cope with the changes in the dynamic environment. Through the adaptive control algorithm, the phase and amplitude of the intelligent reflecting surface can be precisely adjusted, thus optimizing the signal propagation path and improving network performance.

[0148] Specifically, the adaptive control algorithm dynamically calculates the optimal reflection phase and amplitude of each intelligent reflecting surface based on the target parameter (channel state information CSI) and other network parameters (such as signal strength, interference, delay, etc.) to obtain the adjustment parameters.

[0149] Among them, the algorithm model for calculating the adjustment parameters can be represented by the following optimization problem:

[0150]

[0151] where h i is the gain of the i-th channel, A(θ i , A i ) is the adjustment matrix of the i-th intelligent reflecting surface, and θ i and A i represent the reflection phase and amplitude of the intelligent reflecting surface, respectively.

[0152] The above optimization objective function aims to maximize the gain of the reflected signal and optimize the network signal quality by continuously adjusting the reflection phase and amplitude. To make this process real-time and effective, the adjustment parameters can be calculated according to the real-time monitored signal quality, and the configuration of the intelligent reflecting surface can be adjusted using the adjustment parameters, thereby ensuring the continuity and accuracy of signal optimization.

[0153] To improve the response speed of the intelligent reflecting surface, among them, as an example, the communication optimization process for each intelligent reflecting surface using the adjustment parameters may include the following sub-steps:

[0154] S121. Obtain real-time communication parameters, where the real-time communication parameters are signal parameters measured in real time.

[0155] S122. Determine the feedback signal of each intelligent reflecting surface according to the real-time communication parameters and the adjustment parameters based on a low-latency feedback control mechanism.

[0156] S123. Perform communication optimization processing on each intelligent reflecting surface by using the feedback signal of each intelligent reflecting surface.

[0157] In a practical operation mode, the present invention adopts a low-latency feedback control mechanism for optimization regulation. The low-latency feedback control mechanism can collect multi-dimensional information such as signal strength, interference level, and transmission quality in real time, and input this information into the control algorithm. Through low-latency feedback, it can quickly respond to changes in the channel environment and timely adjust the configuration of the reflection unit to avoid signal loss or performance degradation.

[0158] In a specific implementation, the feedback control mechanism adopts an online estimation method to obtain real-time communication parameters, including parameters such as signal strength, bit error rate (BER), and transmission quality, so as to evaluate the current signal quality. Among them, the above parameters can also be optimized channel state information (CSI) and other network parameters.

[0159] Based on real-time data, the system dynamically adjusts the phase and amplitude of each reflection unit, so that the reflection unit can adapt to changes in the network environment in the shortest time. For example, when the user equipment changes its position or the signal is blocked, the feedback control system can quickly adjust the configuration of the reflection unit to restore the stability and quality of the signal.

[0160] Among them, the feedback signal is shown as the following formula:

[0161] F=(y actual - Y predicted )

[0162] Among them, F is the feedback signal, Y actual is the real-time communication parameter, and Y predicted is the adjustment parameter. Through this feedback signal, the control algorithm can automatically adjust the configuration of the intelligent reflecting surface to optimize the network performance. The present invention adopts a low-latency feedback control mechanism, which can quickly respond to environmental changes, timely adjust the configuration of the reflection unit, and reduce the degradation of network performance.

[0163] In order to reduce the computational complexity and improve the processing efficiency, among them, as an example, the performing communication optimization processing on each intelligent reflecting surface by using the feedback signal of each intelligent reflecting surface may include the following sub-steps:

[0164] S1231. Construct the feedback signal of each intelligent reflecting surface into a control task based on a distributed control method.

[0165] S1232. Assign the control task to each corresponding intelligent reflecting surface for optimizing and adjusting the communication configuration of each intelligent reflecting surface.

[0166] In one embodiment, the present invention may use a distributed control method. The distributed control method assigns the task of optimizing and adjusting to multiple intelligent reflecting surfaces, and improves the performance of the overall system through the combination of local optimization and global optimization.

[0167] In actual operation, based on the distributed control method, the feedback signals of each intelligent reflecting surface can be constructed into control tasks, and then the control tasks are assigned to each corresponding intelligent reflecting surface for optimizing and adjusting the communication configuration. In the distributed control architecture, each intelligent reflecting surface can self-regulate according to local channel information and network status. The optimization strategy of each intelligent reflecting surface is locally optimized, but the configuration and adjustment strategies of all intelligent reflecting surfaces will be coordinated as a whole to ensure the maximization of global signal optimization. This distributed method effectively avoids the possible computational bottlenecks brought by centralized control and improves the processing efficiency of the system.

[0168] The distributed control method realizes the combination of local and global optimization in the following form:

[0169]

[0170] where w ij is the weight coefficient between intelligent reflecting surfaces, indicating the influence of the i-th intelligent reflecting surface on the j-th intelligent reflecting surface. In the above formula, the parameters of each intelligent reflecting surface can be adjusted through local optimization, and the configuration of all intelligent reflecting surfaces can be ensured to be optimal through the global coordination strategy.

[0171] Through the distributed control method, the computational complexity can be reduced, the efficiency and reliability of the adjustment of the reflecting unit can be improved. Moreover, it can continuously optimize the configuration of the reflecting unit in a dynamic environment to ensure efficient signal optimization.

[0172] To further optimize the network resources of communication to ensure the effective cooperation between the intelligent reflecting surface and the traditional base station and the base station cooperation system (such as MIMO, multi-antenna technology), for example, after the step of calculating the adjustment parameters of each intelligent reflecting surface using the target parameters based on the adaptive algorithm, the method may further include the following sub-steps:

[0173] S31. Obtain the channel gain parameter, where the channel gain parameter is the channel gain between each user equipment and the base station in the wireless communication system.

[0174] S32. Adjust the transmission rate of each user equipment using the channel gain parameter and the adjustment parameter.

[0175] In one embodiment, the present invention can also optimize network performance by maximizing the resource utilization rate of the network, so as to ensure the effective cooperation between the intelligent reflecting surface and traditional base stations and base station cooperation systems (such as MIMO, multi-antenna technology).

[0176] Specifically, assume that there are N user equipment and M base stations in the network of a wireless communication system, where the channel gain between each user equipment and the base station is h ij , and the adjustment parameter of the intelligent reflecting surface is θ i (reflection phase and amplitude). The optimization goal this time is to maximize the transmission rate of users while ensuring the optimal reflection path of the intelligent reflecting surface.

[0177] The optimization problem can be shown as the following formula:

[0178]

[0179] where, N0 is the noise power, h ij is the channel gain between the i-th user equipment and the j-th base station, A(θ i ) is the adjustment matrix of the intelligent reflecting surface, and the reflection coefficient θ i is the adjustment parameter of the reflection unit.

[0180] By obtaining the channel gain between each user equipment and the base station, and then using the channel gain parameter and the adjustment parameter to adjust the transmission rate of each user equipment.

[0181] In addition, the present invention can perform channel and resource scheduling. Specifically, based on the network load and real-time feedback data, it can dynamically adjust the reflection path at different times and positions to ensure the maximization of signal enhancement and reasonable resource allocation. The constraint conditions of the scheduling include the transmission power P j of each base station and the channel capacity C j , and its model is:

[0182]

[0183] where, P max is the maximum power of the base station, and C min is the minimum channel capacity of each base station.

[0184] In another optional embodiment, in order to further adjust the parameters according to the feedback of the base station and the user equipment and improve the optimization effect. Among them, as an example, after the step of using the channel gain parameter and the adjustment parameter to adjust the transmission rate of each user equipment, the method may further include the following sub-steps: (real-time feedback adjustment)

[0185] S41. Obtain the base station communication information of the base station, and the base station communication information is the parameters of the channel quality and interference level of the user equipment.

[0186] S42. Optimize the adjustment parameters by using the base station communication information.

[0187] In one implementation, the base station and the user equipment help the intelligent reflecting surface to perform dynamic adjustment through real-time channel quality feedback.

[0188] Suppose F j is the feedback information sent by the j-th base station, and the feedback content includes parameters such as channel quality and interference level. The intelligent reflecting surface can optimize the configuration of its reflecting units according to this feedback information:

[0189] θ i = θ i + ɑ·F j ;

[0190] where ɑ is the feedback adjustment coefficient, which controls the adjustment speed of the reflecting units.

[0191] In this embodiment, the present invention provides a communication optimization method for a wireless communication system. The beneficial effects are as follows: The present invention can call a preset evaluation model to obtain the target parameters of the intelligent reflecting surface, where the preset evaluation model is a model trained by using the dynamic characteristic parameters of the environment where the wireless communication system is located; Based on an adaptive algorithm, use the target parameters to calculate the adjustment parameters of each intelligent reflecting surface, and use the adjustment parameters to perform communication optimization processing on each intelligent reflecting surface. The evaluation model trained by the dynamic characteristic parameters can predict the target parameters required for the intelligent reflecting surface to be optimized in the current environment, and then can perform optimization adjustment according to the target parameters, so as to meet the actual optimization requirements, improve the optimization accuracy and accuracy; And there is no need to additionally increase the number of devices, which can reduce the optimization cost and the overall cost of the communication network.

[0192] As Figure 2 shown, based on the above method item embodiment, a corresponding device item embodiment is provided;

[0193] An embodiment of the present invention provides a communication optimization device for a wireless communication system. The wireless communication system includes multiple intelligent reflecting surfaces. The device includes:

[0194] An acquisition module, configured to call a preset evaluation model to obtain the target parameters of the intelligent reflecting surface, where the preset evaluation model is a model trained by using the dynamic characteristic parameters of the environment where the wireless communication system is located, and the dynamic characteristic parameters are channel state data collected at the base station node, user equipment node, and intelligent reflecting surface node of the wireless communication system;

[0195] An optimization module, configured to calculate adjustment parameters for each intelligent reflecting surface based on an adaptive algorithm using the target parameters, where the adjustment parameters include: reflection phase and amplitude, and perform communication optimization processing on each intelligent reflecting surface using the adjustment parameters.

[0196] Further, the preset evaluation model includes a combination of a deep neural network and a convolutional neural network, and the update operation of the preset evaluation model includes:

[0197] Obtain real-time parameters and prediction parameters, where the prediction parameters are channel data predicted by the deep neural network based on the preprocessed channel state data;

[0198] Invoke the convolutional neural network to calculate model weights according to the real-time parameters and the prediction parameters;

[0199] Update the weights of the deep neural network using the model weights.

[0200] Further, the training operation of the preset evaluation model includes:

[0201] Use a preset environment model to collect dynamic change parameters of the environment where the wireless communication system is located in real time;

[0202] Preprocess the dynamic change parameters to obtain processed parameters, where the preprocessing includes: denoising, normalization, and feature extraction;

[0203] Use the processed parameters to train a deep learning model to obtain a preset evaluation model;

[0204] Among them, the preset evaluation model is shown as follows:

[0205]

[0206] Among them, is the predicted target parameter, y i is the actually measured parameter, N is the number of samples, and L(θ) is the minimized loss function.

[0207] Further, the performing communication optimization processing on each intelligent reflecting surface using the adjustment parameters includes:

[0208] Obtain real-time communication parameters, where the real-time communication parameters are signal parameters measured in real time;

[0209] Determine the feedback signal of each intelligent reflecting surface according to the real-time communication parameters and the adjustment parameters according to a low-latency feedback control mechanism;

[0210] Use the feedback signal of each intelligent reflecting surface to perform communication optimization processing on each intelligent reflecting surface;

[0211] Among them, the feedback signal is shown as follows:

[0212] F = (Y actual - Y predicted );

[0213] Among them, F is the feedback signal, Y actual is the real-time communication parameter, and Y predicted is the adjustment parameter.

[0214] Furthermore, using the feedback signal of each intelligent reflecting surface to perform communication optimization processing on each intelligent reflecting surface includes:

[0215] Constructing the feedback signal of each intelligent reflecting surface into a control task based on a distributed control method;

[0216] Assigning the control task to each corresponding intelligent reflecting surface for optimizing and adjusting the communication configuration of each intelligent reflecting surface.

[0217] Furthermore, after the step of calculating the adjustment parameter of each intelligent reflecting surface using the target parameter based on the adaptive algorithm, the method further includes:

[0218] Obtaining a channel gain parameter, where the channel gain parameter is the channel gain between each user equipment of the wireless communication system and the base station;

[0219] Adjusting the transmission rate of each user equipment using the channel gain parameter and the adjustment parameter.

[0220] Furthermore, after the step of adjusting the transmission rate of each user equipment using the channel gain parameter and the adjustment parameter, the method further includes:

[0221] Obtaining base station communication information of the base station, where the base station communication information is a parameter of the channel quality and interference level of the user equipment;

[0222] Optimizing the adjustment parameter using the base station communication information.

[0223] As Figure 3 shown, based on the above method item embodiments, corresponding system item embodiments are provided;

[0224] An embodiment of the present invention provides a wireless communication system, and the wireless communication system is applicable to the communication optimization method of the wireless communication system as described in the above embodiments,

[0225] The wireless communication system includes: a control platform, a base station, and user equipment. The control platform controls multiple intelligent reflecting surfaces, and the base station, the user equipment, and the multiple intelligent reflecting surfaces are communicatively connected to each other.

[0226] In one embodiment, to address the feasibility and cost issues in existing hardware designs, the present invention optimizes the design of the intelligent reflecting surface, uses materials with lower manufacturing costs but capable of achieving high-precision adjustment, and through modular design, enables the hardware device to be flexibly configured according to different network requirements. By adopting the above method, not only the cost of the intelligent reflecting surface is reduced, but also the overall efficiency and stability of the wireless communication system are improved.

[0227] Referring to Figure 4 , the intelligent reflecting surface includes a Dielectric layer, a Patch unit layer, a Dielectric substrate / underlayer, and a metal reflection layer / ground layer, which are arranged in sequence from bottom to top.

[0228] Among them, the metal reflection layer / ground layer is located at the top and is used to reflect and shield electromagnetic waves. The Dielectric substrate / underlayer serves as a support layer to ensure the stability of the upper metal structure.

[0229] The Patch unit layer is composed of periodically arranged metal patches, and each patch is precisely designed to achieve the required phase control.

[0230] The Dielectric layer is located at the bottom layer and is used to adjust the dielectric constant, protect the metal structure, and improve the radiation performance.

[0231] Each layer of the intelligent reflecting surface can be constructed using conductive polymers or ceramic materials. Optionally, low-cost materials such as polymer composites and thin-film metal materials can also be used.

[0232] Specifically, conductive polymers can be a class of polymer materials with conductive properties, which are usually used in application scenarios that require good conductivity but also require light weight and low cost. Compared with traditional metal materials, conductive polymers have the following advantages:

[0233] First, low cost: Compared with metal materials, the raw materials of conductive polymers are cheaper and the production process is simpler.

[0234] Second, flexibility and tunability: The conductivity and other physical properties of conductive polymers can be adjusted by adjusting their chemical structure, making them suitable for different reflection requirements.

[0235] Third, strong environmental adaptability: This material has a strong adaptability to environmental changes, with high corrosion resistance and good thermal stability.

[0236] Ceramic materials, especially those with high dielectric constants, can play an effective regulatory role in the process of radio wave propagation and reflection. The selection of ceramic materials is mainly based on their excellent electromagnetic properties and thermal stability, and is suitable for the adjustment part of the reflection unit, especially having advantages in high-frequency applications:

[0237] First, high dielectric constant: Ceramic materials have a relatively high dielectric constant, enabling them to provide more efficient signal regulation under high-frequency conditions.

[0238] Second, high stability: Ceramics have good heat resistance and corrosion resistance, and can operate stably under various environmental conditions.

[0239] Third, lower cost: Compared with other high-performance materials, certain ceramic materials (such as bauxite ceramics) have lower costs and are feasible for large-scale production.

[0240] Through the application of these low-cost materials, not only can the production cost of the intelligent reflecting surface be reduced, but also its good electromagnetic properties and stability can be ensured, thus meeting the requirements of large-scale deployment.

[0241] To address the challenges in different network requirements and application scenarios, the concept of modular design can be adopted, enabling the intelligent reflecting surface to be flexibly configured according to specific application scenarios. Modular design allows for the flexible adjustment of the number, layout, and adjustment range of the intelligent reflecting surface according to the scale of the network, environmental characteristics, and deployment requirements, thereby reducing the manufacturing and deployment costs of the device without sacrificing performance.

[0242] For example, the intelligent reflecting surface can be expanded or reduced as needed. By combining different numbers of reflection units, systems with different configurations and performances can be formed. In addition, modular design makes the repair, upgrade, and replacement of hardware devices more convenient, reduces the maintenance cost, and improves the sustainability of the system.

[0243] In the hardware design, the anti-environmental interference ability and durability of the intelligent reflecting surface are also improved by using durable materials and improved packaging technologies. For example, the outer shell of the reflection unit is waterproof and dustproof, enabling it to adapt to various harsh climate conditions, thereby extending the service life of the device.

[0244] In addition, low-power design is also adopted to reduce the energy consumption of the hardware device, keep its heat output low during long-term operation, and thus further improve the stability and long-term reliability of the system.

[0245] Through the above technological innovations, the present invention can provide an efficient and low-cost solution in large-scale network deployments. The modular design and the selection of low-cost materials make the reflection units highly cost-effective during the manufacturing and deployment processes. Moreover, by optimizing the production process, rapid and low-cost mass production and deployment can be achieved. This enables each hardware device to be sustainable and highly cost-effective in large-scale applications.

[0246] Meanwhile, the high reliability and long lifespan of the hardware further reduce the long-term operation and maintenance costs, making the present invention applicable not only to current network requirements but also to potential future expansion needs, ensuring that the system operates efficiently and stably throughout its entire lifecycle.

[0247] To enable the intelligent reflecting surface to be compatible with existing network architectures, the wireless communication system of the present invention can adopt a network integration solution that combines software and hardware.

[0248] Specifically, to solve the problem of the coordinated operation of the intelligent reflecting surface with traditional base stations and base station cooperation systems (such as MIMO, multi-antenna technologies), an intelligent scheduling algorithm can be used. This algorithm can intelligently adjust the working parameters of the intelligent reflecting surface according to the current network state to ensure coordinated operation with traditional network devices. The core task of the algorithm is to dynamically adjust the reflection path of the intelligent reflecting surface based on the real-time state of the network (such as signal strength, load conditions, etc.) to achieve optimal signal enhancement and resource allocation. The specific operation method can refer to the process of steps S31 - S32 in the above-mentioned embodiment.

[0249] Under the software-hardware cooperation mechanism, the intelligent reflecting surface obtains network status information in real time through communication with base stations, user equipment, and other network nodes. The base stations and user equipment send feedback data to the intelligent reflecting surface, such as channel quality, data transmission rate, signal interference, etc. The intelligent reflecting surface can dynamically adjust the configuration of the reflection units based on this data, thereby improving the network throughput, reducing latency, and decreasing interference.

[0250] Meanwhile, based on the network load conditions, which can be monitored in real time through network traffic, user demands, and channel states, the system can automatically select the best reflection path and resource allocation strategy at different times and locations, thus ensuring seamless cooperation between network devices and the IRS.

[0251] Through the above network integration solution, intelligent reflecting surfaces can be smoothly incorporated into the existing network architecture, especially in base stations and MIMO systems. After optimizing each intelligent reflecting surface, they will indeed be subject to cooperative control processing. After parameter optimization at the unit level, the entire wireless communication system achieves global coordination through a software-hardware cooperation mechanism, ensuring that the configuration of each intelligent reflecting surface matches the overall network state, thereby optimizing the resource scheduling and signal propagation of the entire network.

[0252] This solution not only ensures the effective cooperation between intelligent reflecting surfaces and traditional base stations and base station cooperation systems (such as MIMO and multi-antenna technologies), but also ensures seamless cooperation with traditional network devices through dynamic adjustment and intelligent scheduling. Through the cooperation of software and hardware, network performance can be effectively improved while simplifying the complexity of device replacement and upgrade.

[0253] To further improve the compatibility of the wireless communication system and enable the access of different types of base stations and devices, an expandable control platform can be added to the wireless communication system. This control platform has an open interface that can support the access of multiple base station types, user devices, and other network nodes, ensuring that intelligent reflecting surfaces can be seamlessly integrated into the existing network architecture. The architecture of the control platform includes the following parts:

[0254] The control platform provides standardized API interfaces, allowing various base stations and devices to exchange data with the wireless communication system according to different requirements. Through these open interfaces, new devices can easily access the existing network system, reducing the complexity of device replacement and upgrade.

[0255] The control platform can monitor the status of the entire network in real time, including information such as network load, channel quality, and interference level, and provide real-time feedback through a centralized management system. The platform dynamically adjusts the configuration of intelligent reflecting surfaces according to the network status data to ensure optimal network performance.

[0256] The design of the control platform enables the system to be flexibly adjusted as the network scale expands and supports the access of different types of network devices. Whether it is a traditional base station or a future new base station or device, it can cooperate with intelligent reflecting surfaces through the platform's interface.

[0257] In terms of the integration of the wireless communication system, a cloud computing and distributed control integration solution can be adopted, aiming to improve network performance, optimize resource scheduling, and ensure that intelligent reflecting surfaces (IRS devices) can be seamlessly integrated into the existing network management system. This solution combines the centralized management advantages of cloud computing and the flexibility of distributed control, making network management more efficient, scalable, and adaptable.

[0258] In one embodiment, the above control platform can be a cloud platform, which uniformly manages all intelligent reflecting surfaces to achieve efficient scheduling and management of network resources. As the core control center, the cloud platform can monitor the network status in real time and remotely control and optimize the configuration of intelligent reflecting surfaces (IRS devices) based on real-time data. The cloud computing platform has the following key advantages:

[0259] First, resource scheduling and management: By centrally managing all intelligent reflecting surfaces (IRS devices), the cloud platform can uniformly schedule network resources to achieve network load balance. With precise algorithms, the system can dynamically adjust the configuration of intelligent reflecting surfaces (IRS devices) according to the network load, user requirements, and channel status in different regions to optimize signal propagation and network throughput.

[0260] Second, intelligent decision support: The cloud platform uses deep learning and machine learning algorithms to analyze historical data and real-time feedback, predict network trends and device requirements, and thus achieve intelligent scheduling decisions. Based on this data, the reflection units of intelligent reflecting surfaces (IRS devices) can be automatically adjusted according to the specific needs of the network to improve network performance and efficiency.

[0261] Third, seamless integration into the existing network architecture: Through precise algorithms and system design, the management, scheduling, and channel optimization of intelligent reflecting surfaces (IRS devices) can be seamlessly integrated into the existing network management system to coordinate with traditional base stations and MIMO systems. The cloud platform enables intelligent reflecting surfaces (IRS devices) to be compatible with different types of network architectures through standardized APIs and open interfaces, avoiding the complexity of device replacement and upgrade.

[0262] To further improve the efficiency and reliability of the wireless communication system, a network management strategy of distributed signal processing can be used. This strategy allows each network node (such as base stations, user equipment, and intelligent reflecting surfaces (IRS devices)) to independently perform local optimization according to the local signal conditions and resource requirements. Each node considers its own status and requirements when performing local optimization and shares information with other nodes through a coordination mechanism to ultimately optimize the global network performance.

[0263] Among them, local optimization can be that each network node (such as intelligent reflecting surfaces (IRS devices) and base stations) analyzes the channel status and user requirements in real time through local signal processing technologies and adjusts its parameters according to the analysis results. For example, intelligent reflecting surfaces (IRS devices) can optimize the configuration of reflection units according to real-time feedback to maximize signal coverage and data throughput. The base station can adjust the resource allocation strategy according to the channel quality and load conditions to ensure network stability and efficiency.

[0264] The coordination mechanism can achieve global optimization, and the local optimization process needs to exchange information with other nodes through the coordination mechanism. For example, an intelligent reflecting surface (IRS device) can feedback information such as channel quality and reflection unit configuration to the cloud platform or nearby base stations, and the base station can optimize resource allocation based on this feedback data. During this process, the coordination mechanism can ensure that the optimization decisions of each node can work together to avoid conflicts and resource waste.

[0265] Global network performance optimization can be achieved through distributed signal processing and coordination mechanism. The system can efficiently allocate network resources in different regions or scenarios, improve the network throughput, reduce interference, and lower latency. Compared with traditional centralized management methods, the distributed management method can more flexibly respond to the dynamic changes of the network environment, improving the robustness of the system and its ability to handle complex network environments.

[0266] To ensure that the distributed signal processing strategy can achieve efficient resource allocation and optimization, the following model can be used to describe the optimization process of the system:

[0267] Determine the resource allocation and optimization goals. For example, let R i be the resource allocation of the i-th node (such as an IRS device or a base station), and the goal is to maximize the total throughput T of the network, that is:

[0268]

[0269] where, P i is the power or resource allocation of the i-th node, N0 is the noise power, and N is the total number of nodes in the network.

[0270] In distributed control, each node i optimizes its configuration according to local channel state information h i and other relevant parameters. The coordination mechanism can be expressed by the following formula:

[0271]

[0272] where, w ij represents the coordination weight between the i-th node and the j-th node, and A(θ j ) is the adjustment matrix of the reflection unit.

[0273] For local optimization and global coordination, each node considers its own channel state information and feedback from other nodes during local optimization, shares information with other nodes through the coordination mechanism, and achieves the optimization of global network performance.

[0274] Through the above technical means and methods, the present invention successfully solves the technical problems of low optimization accuracy of intelligent reflecting surface (IRS) in network performance optimization due to the complexity of the distance between the reflecting surface and the signal source, environmental factors, etc., the technical problem of hardware incompatibility, and the high optimization cost.

[0275] After adopting the wireless communication system of the present invention, its specific performance is as follows:

[0276] First, the accuracy of channel modeling and estimation: The channel estimation method based on deep learning proposed by the present invention significantly improves the estimation accuracy of channel state information (CSI), can dynamically adapt to the changes of the wireless environment, and thus optimizes the configuration of reflection units. The deep learning algorithm and the online feedback mechanism make the channel estimation more accurate, reduce the delay, and improve the response speed and stability of the intelligent reflecting surface (IRS) in complex environments.

[0277] Second, the high precision and low latency of reflection unit adjustment: The reflection unit adjustment technology based on the adaptive control algorithm, combined with the low-latency feedback mechanism and the distributed control method, enables the reflection unit to adjust the reflection phase and amplitude in real time in a dynamic environment, ensuring the stability of network performance and the optimization effect. Through the modular design and efficient adjustment method, the system can provide an efficient and low-cost solution in large-scale networks.

[0278] Third, the optimization of hardware implementation and low-cost deployment: By selecting low-cost materials and improving the production process, this case effectively reduces the hardware manufacturing cost while maintaining the high performance and reliability of the hardware. The modular design enables the hardware to be flexibly configured according to different application scenarios and reduces the economic pressure during large-scale deployment.

[0279] Fourth, the compatibility with the existing network architecture: Through the intelligent scheduling algorithm and the scalable control platform, the intelligent reflecting surface (IRS) can cooperate seamlessly with existing base stations, MIMO systems and other network devices, ensuring the efficient utilization of network resources and the efficient cooperation between devices. The open interface and standardized protocol simplify the process of device replacement and system upgrade, enhancing the compatibility and flexibility of the system.

[0280] Fifth, the adaptability and intelligence in dynamic environments: By introducing the environmental perception algorithm and deep learning technology, the wireless communication system can adjust the reflection unit in real time in complex and dynamic environments to achieve the best signal optimization effect. The edge computing technology further improves the real-time performance of the algorithm, reduces the load of the central server, and improves the system response speed and overall efficiency.

[0281] Sixth, system integration and network management optimization: The network integration solution based on cloud computing and distributed control enables efficient management and scheduling of IRS devices, ensuring the stability and scalability of the system. Distributed signal processing and local optimization strategies allow each network node to optimize based on local information and share information with other nodes through a coordination mechanism to optimize the global network performance and simplify the complexity of traditional network management.

[0282] It can be understood that the above device item embodiments correspond to the method item embodiments of the present invention, and can implement the communication optimization method of the wireless communication system provided by any one of the above method item embodiments of the present invention.

[0283] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement without creative efforts.

[0284] Based on the above embodiments of the communication optimization method of the wireless communication system, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the communication optimization method of the wireless communication system according to any embodiment of the present invention.

[0285] Exemplarily, in this embodiment, the computer program can be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more module elements can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the terminal device.

[0286] The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0287] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the terminal device and connects various parts of the entire terminal device through various interfaces and lines.

[0288] Based on the above method item embodiments, another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the communication optimization method of the wireless communication system described in any one of the above method item embodiments of the present invention.

[0289] Among them, if the modules / units integrated in the device / terminal device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0290] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A communication optimization method for a wireless communication system, characterized in that: The wireless communication system includes a plurality of smart reflective surfaces, and the method includes: Calling a preset evaluation model to obtain target parameters of the smart reflective surface, wherein the preset evaluation model is a model trained using dynamic characteristic parameters of an environment in which a wireless communication system is located, and the dynamic characteristic parameters are channel state data collected from base station nodes, user equipment nodes, and smart reflective surface nodes of the wireless communication system; The target parameters are used to calculate adjustment parameters of each smart reflective surface based on an adaptive algorithm. The adjustment parameters include reflection phase and amplitude. The communication optimization processing is performed on each smart reflective surface using the adjustment parameters.

2. The communication optimization method of a wireless communication system according to claim 1, characterized in that: The preset evaluation model includes a combination of a deep neural network and a convolutional neural network, and the updating operation of the preset evaluation model includes: Acquire real-time parameters and prediction parameters, wherein the prediction parameters are channel data predicted by the deep neural network based on the preprocessed channel state data; Calling the convolutional neural network to calculate model weights according to the real-time parameters and the prediction parameters; The model weights are used to update the weights of the deep neural network.

3. The communication optimization method of a wireless communication system according to claim 1, characterized in that: The training operation of the preset evaluation model includes: Using the preset environment model to collect the dynamic changing parameters of the environment where the wireless communication system is located in real time; Preprocessing the dynamically changing parameters to obtain processing parameters, wherein the preprocessing includes: denoising, normalization, and feature extraction; The processing parameters are used to perform model training on the deep learning model to obtain a preset evaluation model; The preset evaluation model is as follows: in, is the target parameter for prediction, y i is the actual measured parameter, N is the number of samples, and L(θ) is the minimization loss function.

4. The communication optimization method of a wireless communication system according to claim 1, wherein: The adopting the adjustment parameters to perform communication optimization processing on each smart reflective surface includes: Acquiring real-time communication parameters, wherein the real-time communication parameters are signal parameters measured in real time; Determining a feedback signal of each smart reflective surface according to the real-time communication parameter and the adjustment parameter according to a low-latency feedback control mechanism; Using the feedback signal of each smart reflective surface to perform communication optimization processing on each smart reflective surface; The feedback signal is shown in the following formula: F=(Y actual -Y predicted ) Among them, F is the feedback signal, Y actual is the real-time communication parameter, Y predicted To adjust the parameters.

5. The communication optimization method of the wireless communication system according to claim 4, characterized in that: The method of performing communication optimization processing on each smart reflective surface by using the feedback signal of each smart reflective surface comprises: The feedback signal of each intelligent reflective surface is constructed into a control task based on a distributed control method; The control task is assigned to each corresponding intelligent reflecting surface so that each intelligent reflecting surface can optimize and adjust the communication configuration.

6. The communication optimization method of a wireless communication system according to any one of claims 1 to 5, characterized in that: After the step of calculating the adjustment parameters of each smart reflective surface using the target parameters based on the adaptive algorithm, the method further includes: Acquire a channel gain parameter, where the channel gain parameter is a channel gain between each user equipment and a base station in a wireless communication system; The transmission rate of each user equipment is adjusted using the channel gain parameter and the adjustment parameter.

7. The communication optimization method of a wireless communication system according to claim 6, characterized in that: After the step of adjusting the transmission rate of each user equipment by using the channel gain parameter and the adjustment parameter, the method further includes: Acquire base station communication information of the base station, wherein the base station communication information is a parameter of a channel quality and an interference level of a user equipment; The base station communication information is used to optimize the adjustment parameters.

8. A communication optimization device for a wireless communication system, characterized in that: The wireless communication system comprises a plurality of intelligent reflective surfaces, and the device comprises: An acquisition module, used to call a preset evaluation model to acquire target parameters of the smart reflective surface, wherein the preset evaluation model is a model trained using dynamic characteristic parameters of the environment in which the wireless communication system is located, and the dynamic characteristic parameters are channel state data collected from base station nodes, user equipment nodes, and smart reflective surface nodes in the wireless communication system; The optimization module is used to calculate the adjustment parameters of each smart reflective surface based on the adaptive algorithm using the target parameters, the adjustment parameters including: reflection phase and amplitude, and use the adjustment parameters to perform communication optimization processing on each smart reflective surface.

9. A wireless communication system, characterized in that: The wireless communication system is applicable to the communication optimization method of the wireless communication system as described in any one of 1-7, and the wireless communication system includes: a control platform, a base station, and a user device. The control platform controls multiple intelligent reflecting surfaces, and the base station, the user device and the multiple intelligent reflecting surfaces are communicatively connected to each other.

10. The wireless communication system according to claim 9, wherein: The smart reflective surface is constructed by using conductive polymer or ceramic material.

Images