Interference-adaptive RFID multi-band antenna signal integration system and interference-adaptive RFID multi-band antenna signal integration method

Abstract

The invention discloses an interference-adaptive RFID multi-band antenna signal integration system and method, and belongs to the field of wireless communication networks, and the system comprises an array construction module which is used for constructing a multi-band antenna array and completing the initialization of a decision center; the fixed interference construction module is used for establishing a communication path according to the position structure information and the position of the signal source, performing obstacle shielding interference and multipath interference fitting, and generating fixed interference data; the dynamic interference monitoring module is used for establishing a first electromagnetic interference result and a second electromagnetic interference result; the feedback optimization module is used for establishing an interference adaptive compensation result; and the adjusting module is used for performing parameter compensation optimization on the multi-band antenna array. The technical problems that an existing RFID system is insufficient in anti-interference capability and poor in reliability and adaptability in a complex environment are solved, and the technical effect of improving the anti-interference capability and the environment adaptability of the RFID multi-band antenna in different application scenes is achieved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication networks, and in particular to an interference-adaptive RFID multi-band antenna signal integration system and method. Background Art

[0002] Existing RFID systems often face many challenges in complex application environments. Factors such as obstacles and electromagnetic interference in the external environment will have a serious impact on the signal quality of the RFID antenna, resulting in a decrease in recognition rate and an increase in bit error rate. At the same time, different application scenarios also have different requirements for performance indicators such as the antenna's operating frequency band and recognition distance. Therefore, existing RFID systems generally have technical problems such as insufficient anti-interference ability, poor reliability and adaptability in complex environments, making it difficult to achieve flexible and stable recognition and data transmission, and unable to meet the high-quality requirements of IoT applications. Summary of the invention

[0003] The present application provides an interference-adaptive RFID multi-band antenna signal integration system and method, aiming to solve the technical problems of insufficient anti-interference ability, poor reliability and adaptability of existing RFID systems in complex environments.

[0004] In view of the above problems, the present application provides an interference-adaptive RFID multi-band antenna signal integration system and method.

[0005] The first aspect disclosed in the present application provides an interference-adaptive RFID multi-band antenna signal integration system, which includes: an array construction module, which is used to construct a multi-band antenna array and establish digital communication between the multi-band antenna array and a decision center, and complete the initialization of the decision center based on the array information and attribute information of the multi-band antenna array, wherein the array information includes position structure information, and the attribute information includes parameter information; a fixed interference construction module, which is used to obtain the application scenario of the multi-band antenna array, obtain the signal source position, receive the position structure information of the decision center, establish a communication path with the position structure information and the signal source position, and perform obstacle occlusion interference and multipath interference fitting based on the communication path to generate a fixed interference digital signal. a dynamic interference monitoring module, used to obtain the preset communication frequency band of the multi-band antenna array, perform inter-band interference evaluation based on the preset communication frequency band, establish a first electromagnetic interference result, control the receiver to perform environmental electromagnetic noise monitoring, and establish a second electromagnetic interference result based on the environmental electromagnetic noise monitoring result; a feedback optimization module, used to synchronize the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, call the reinforcement reward network of the decision center to perform interference adaptive compensation, and establish interference adaptive compensation results. The reinforcement reward network is built with array information and attribute information as source matching data; an adjustment module, used to optimize parameter compensation of the multi-band antenna array based on the interference adaptive compensation result.

[0006] Another aspect disclosed in the present application provides an interference-adaptive RFID multi-band antenna signal integration method, the method comprising: constructing a multi-band antenna array, establishing digital communication between the multi-band antenna array and a decision center, initializing the decision center based on array information and attribute information of the multi-band antenna array, the array information including position structure information, and the attribute information including parameter information; obtaining an application scenario of the multi-band antenna array, and obtaining a signal source position, receiving the position structure information of the decision center, establishing a communication path with the position structure information and the signal source position, and performing obstacle blocking interference and multipath interference fitting based on the communication path to generate fixed interference data; obtaining a preset communication band of the multi-band antenna array, performing inter-band interference evaluation based on the preset communication band, establishing a first electromagnetic interference result, controlling the receiver to perform environmental electromagnetic noise monitoring, and establishing a second electromagnetic interference result with the environmental electromagnetic noise monitoring result; synchronizing the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, calling the enhanced reward network of the decision center to perform interference adaptive compensation, and establishing an interference adaptive compensation result, the enhanced reward network being constructed with the array information and attribute information as source matching data; and optimizing parameter compensation of the multi-band antenna array based on the interference adaptive compensation result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The multi-band antenna array is constructed through the array construction module, and digital communication with the decision center is established to complete the initialization of the decision center to meet the needs of different application scenarios; the application scenario and signal source position of the antenna array are obtained through the fixed interference construction module, the communication path is established, and the obstacle blocking interference and multipath interference are fitted to generate fixed interference data, which is helpful to fully understand the fixed interference factors in the actual environment; the preset communication frequency band of the antenna array is obtained through the dynamic interference monitoring module, the interference between the frequency bands is evaluated, and the first electromagnetic interference result is obtained. At the same time, the environmental electromagnetic noise is monitored through the receiver to obtain the second electromagnetic interference result, so as to dynamically know the electromagnetic interference situation in the environment; the feedback optimization module is used to The fixed interference data and the first and second electromagnetic interference results are sent to the decision center, and the reinforcement reward network is called to perform interference adaptive compensation, establish the compensation result, and obtain the optimal antenna parameter compensation scheme according to the current interference situation and the antenna array state; the adjustment module is used to optimize the parameters of the multi-band antenna array according to the interference adaptive compensation result. After optimization and adjustment, the antenna array can flexibly and stably complete the technical solution of RFID tag identification and data transmission tasks in different scenarios, which solves the technical problems of insufficient anti-interference ability, poor reliability and adaptability of the existing RFID system in complex environments, and achieves the technical effect of improving the anti-interference ability and environmental adaptability of RFID multi-band antennas in different application scenarios.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A structural schematic diagram of an interference-adaptive RFID multi-band antenna signal integration system is provided for an embodiment of the present application.

[0010] Figure 2 A flow chart of an interference-adaptive RFID multi-band antenna signal integration method is provided for an embodiment of the present application; Explanation of the reference numerals: array construction module 11 , fixed interference construction module 12 , dynamic interference monitoring module 13 , feedback optimization module 14 , adjustment module 15 . DETAILED DESCRIPTION

[0011] The overall idea of ​​the technical solution provided by this application is as follows: The embodiment of the present application provides an interference-adaptive RFID multi-band antenna signal integration system and method. First, a multi-band antenna array is constructed through an array construction module, and communication is established with a decision center. Then, the interference information in the environment is fully acquired through a fixed interference construction module and a dynamic interference monitoring module, including obstacle obstruction, multipath effect, inter-band interference, and electromagnetic noise. On this basis, the feedback optimization module sends the interference data to the decision center and calls the reinforcement learning algorithm to obtain the optimal antenna parameter compensation solution. Afterwards, the adjustment module optimizes the antenna array according to the compensation results, thereby improving the anti-interference ability and environmental adaptability of the RFID system, and achieving flexible and stable identification and data transmission in different application scenarios.

[0012] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0013] Embodiment 1 like Figure 1 As shown, the embodiment of the present application provides an interference adaptive RFID multi-band antenna signal integration system, the system comprising: The array construction module 11 is used to construct a multi-band antenna array and establish digital communication between the multi-band antenna array and the decision center, and complete the initialization of the decision center based on the array information and attribute information of the multi-band antenna array, wherein the array information includes position structure information, and the attribute information includes parameter information.

[0014] Specifically, first, the array construction module 11 constructs a multi-band antenna array. The multi-band antenna array is composed of a plurality of antenna array elements that can work in different frequency bands, and each array element can independently configure its operating frequency and related parameters. By designing the spatial position and arrangement structure of the array element, directional transmission and reception of RFID signals in different frequency bands can be achieved. While completing the construction of the multi-band antenna array, the array construction module 11 establishes digital communication between the multi-band antenna array and the decision center. Through this digital communication, the constructed multi-band antenna array can transmit its own array information and attribute information to the decision center for it to complete the initialization configuration. Among them, the array information mainly includes the spatial position coordinates of the array element, the arrangement structure of the array and other position structure information, and the attribute information covers the operating frequency, gain, beam pattern and other parameter information of each array element. The array information and attribute information of the multi-band antenna array are completed by the corresponding sensors and measurement circuits, and are transmitted to the decision center in a specified data format through the digital communication interface. After the decision center obtains the array information and attribute information from the array construction module 11, it can establish an overall understanding of the multi-band antenna array and grasp its specific characteristics, providing the necessary prerequisites for the decision center to carry out subsequent tasks such as interference environment perception, optimized frequency selection, and dynamic parameter adjustment.

[0015] By constructing a multi-band antenna array through the array construction module 11 and establishing digital communication between it and the decision-making center, the initialization configuration of the decision-making center is completed, so that the decision-making center has a comprehensive grasp of the multi-band antenna array, thus laying the foundation for intelligent and adaptive anti-interference optimization.

[0016] The fixed interference construction module 12 is used to obtain the application scenario of the multi-band antenna array, obtain the signal source position, receive the position structure information of the decision center, establish a communication path based on the position structure information and the signal source position, and perform obstacle blocking interference and multipath interference fitting based on the communication path to generate fixed interference data.

[0017] Specifically, the interference construction module 12 is used to identify and model fixed interference sources in the application scenario where the multi-band antenna array is located, providing a basis for the decision-making center to optimize the anti-interference strategy.

[0018] First, the fixed interference construction module 12 obtains the actual application scenario information of the multi-band antenna array, such as the topography and building layout of the deployment environment; at the same time, the spatial position of the target signal source such as the RFID tag is obtained to form the signal source position. The acquisition of this information can be achieved through various methods such as manual entry, drawing import, and field survey. After receiving the decision center position structure information transmitted by the array construction module 11, the fixed interference construction module 12 combines the acquired signal source position to construct the communication path between the multi-band antenna array and the tag. In actual applications, these communication paths are often affected by factors such as obstacle obstruction and signal multipath propagation, resulting in impaired RFID system performance.

[0019] In order to accurately evaluate the impact of the above fixed interference sources, the fixed interference construction module 12 performs obstacle occlusion interference and multipath interference fitting on the established communication path. Among them, obstacle occlusion interference estimates the blocking attenuation of the signal by obstacles through algorithms such as ray tracing, combined with the communication path and the three-dimensional scene model; multipath interference fitting estimates the parameters such as delay, amplitude and phase of signals on different paths by analyzing the reflection, diffraction and other mechanisms of the signal propagation path. Subsequently, the fixed interference construction module 12 integrates the analysis results of obstacle occlusion and multipath interference to generate fixed interference data corresponding to each communication path. On the one hand, it describes the basic attenuation and distortion of the communication link between the multi-band antenna array and the tag, and on the other hand, it also reveals the influence of fixed interference sources on the performance of the RFID system. The law and trend.

[0020] The fixed interference construction module 12 establishes a communication path based on the position structure information and the signal source position, and performs obstacle blocking interference and multipath interference fitting based on the communication path to generate data that comprehensively describes the impact of fixed interference sources, providing a basis for the decision-making center to optimize the anti-interference strategy.

[0021] The dynamic interference monitoring module 13 is used to obtain the preset communication frequency band of the multi-band antenna array, perform inter-band interference evaluation based on the preset communication frequency band, establish a first electromagnetic interference result, control the receiver to perform environmental electromagnetic noise monitoring, and establish a second electromagnetic interference result based on the environmental electromagnetic noise monitoring result.

[0022] Specifically, the dynamic interference monitoring module 13 is responsible for real-time monitoring and evaluation of the dynamic interference status of the electromagnetic environment in which the multi-band antenna array is located, and provides a basis for the decision-making center to dynamically optimize antenna parameters.

[0023] First, the dynamic interference monitoring module 13 obtains the preset communication frequency band. The preset communication frequency band refers to the working frequency parameters of each antenna array element pre-planned and configured in the design and deployment stage, which determines at what frequency each array element transmits and receives electromagnetic waves to achieve communication with the tag. Then, based on the known preset communication frequency band, the dynamic interference monitoring module 13 carries out the evaluation of inter-band interference. Inter-band interference mainly includes intermodulation interference, harmonic interference, adjacent channel interference, etc., which is caused by the nonlinear interaction between signals in different frequency bands. By analyzing the transceiver signals of the multi-band antenna array and using algorithms such as spectrum analysis and power detection, the dynamic interference monitoring module 13 evaluates the severity of interference between various frequency bands and forms a first electromagnetic interference result, which reflects the signal quality and interference level of the multi-band RFID system itself. In addition to the inter-band interference within the multi-band antenna array, the RFID system is often interfered by electromagnetic noise from the external environment. In order to comprehensively evaluate the interference condition, the dynamic interference monitoring module 13 controls the receiver to perform environmental electromagnetic noise monitoring. By collecting electromagnetic signals from the environment around the antenna array in real time and using energy detection, feature recognition and other algorithms, we analyze the characteristic parameters of the environmental electromagnetic noise, such as spectral composition, power level, modulation method, etc., to determine its interference impact on the RFID system and generate a second electromagnetic interference result, which reflects the degree of influence of external uncertain interference sources on the RFID system.

[0024] The dynamic interference monitoring module monitors and evaluates the internal inter-band interference and external environmental noise interference of the RFID system in real time, forming an interference evaluation result that reflects the current electromagnetic environment, providing a basis for the decision-making center to optimize the antenna working parameters in a timely manner, thereby dynamically improving the anti-interference performance and enhancing the environmental adaptability of multi-band antennas.

[0025] The feedback optimization module 14 is used to synchronize the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, call the reinforcement reward network of the decision center to perform interference adaptive compensation, and establish the interference adaptive compensation result. The reinforcement reward network is built with array information and attribute information as source matching data.

[0026] Specifically, the feedback optimization module 14 feeds back the acquired interference information to the decision center, and calls the intelligent algorithm of the decision center to adaptively compensate for the interference, so as to optimize the working parameters of the antenna array and improve the anti-interference capability of the system.

[0027] First, the feedback optimization module 14 synchronously transmits the fixed interference data generated by the fixed interference construction module 12, the first electromagnetic interference result and the second electromagnetic interference result obtained by the dynamic interference monitoring module 13 to the decision center as the current interference situation. Then, the feedback optimization module 14 will further call the built-in reinforcement reward network of the decision center to execute the adaptive compensation algorithm for the current interference situation. The so-called reinforcement reward network is a neural network structure designed based on reinforcement learning theory, which can autonomously learn and optimize the system control strategy according to the feedback reward signal. Among them, the reinforcement reward network is trained and built with the array information and attribute information provided by the array construction module 11 as source matching data, so that it can learn and generate the optimal array control scheme for different array deployment schemes and interference conditions. The feedback optimization module 14 inputs the various interference data obtained synchronously into the trained reinforcement reward network, and the network can evaluate the impact of the current interference on the array performance, and obtain the best array compensation scheme to deal with the current interference situation through reward function calculation, strategy gradient descent and other methods, and form an interference adaptive compensation result, including dynamic adjustment strategies for multiple parameters such as antenna array element frequency, transmission power, and phase weights. The interference adaptive compensation results can guide the real-time reconstruction of the multi-band antenna array, realize the adaptive optimization of array parameters according to the interference feedback, dynamically improve the anti-interference performance, and enhance the adaptability to the electromagnetic environment.

[0028] By synchronizing various types of interference information through the feedback optimization module 14 and calling the reinforcement reward network to perform adaptive compensation, the intelligent optimization of the antenna array parameters based on interference feedback is realized, so that the RFID system can dynamically adjust its own parameters according to the real-time changes of the electromagnetic environment, thereby maintaining a stable and reliable working state under complex interference conditions.

[0029] The adjustment module 15 is used to perform parameter compensation optimization of the multi-band antenna array based on the interference adaptive compensation result.

[0030] Specifically, the adjustment module 15 is responsible for receiving the interference adaptive compensation result output by the feedback optimization module 14, and adjusting and optimizing the relevant parameters of the multi-band antenna array in real time accordingly, so as to implement the compensation scheme in actual operation.

[0031] First, the interference adaptive compensation result generated by the feedback optimization module 14 is obtained through the adjustment module 15, which includes a series of array parameter adjustment strategies optimized and generated for the interference situation in the current electromagnetic environment, such as the dynamic configuration scheme of the operating frequency, transmission power, phase weight, etc. of each array element. Then, the adjustment module 15 parses the various parameter optimization strategies in the compensation result and makes corresponding parameter adjustments to the multi-band antenna array. Taking frequency adjustment as an example, the adjustment module 15 adjusts the operating frequency of a specific array element to the frequency point specified by the compensation scheme to achieve spectrum avoidance for specific interference. Similarly, for parameters such as transmission power and phase weight, the adjustment module 15 performs corresponding parameter configuration according to the requirements of the compensation result to optimize the electromagnetic radiation characteristics of the antenna array and suppress the influence of interference.

[0032] Due to the dynamic changes in the electromagnetic environment, the interference adaptive compensation results will change according to the real-time update of the feedback information. In order to adapt to the dynamics, the adjustment module 15 maintains real-time communication with the feedback optimization module 14, and updates and adjusts the array parameters in time after obtaining the latest interference adaptive compensation results to ensure the timeliness and continuity of the compensation optimization.

[0033] The adjustment module 15 adjusts the antenna array parameters in real time based on the interference adaptive compensation result to achieve dynamic configuration of the multi-band antenna array, so that the multi-band antenna array can dynamically respond to complex interference, improve communication quality, and enhance environmental adaptability in practical applications.

[0034] Furthermore, the embodiment of the present application also includes: Establish the state representation, action representation, and reward function of the reinforcement reward network as follows: ; ; ; in, for The state at the moment indicates, for The frequency band of the moment, for The gain of time, for The direction of time, To fix the interference data, is electromagnetic interference data established according to the first electromagnetic interference result and the second electromagnetic interference result, for The action of the moment indicates, For frequency band adjustment, For gain adjustment, To adjust the direction, for The reward function at the moment, is the signal power, is the interference power, is the noise power, and is the weight coefficient; Establishing a training database using array information and attribute information as source matching data; The policy gradient and the value function are configured with the state representation, the action representation, and the reward function, the policy gradient and the value function are updated and iterated through the training database, and interference adaptive compensation is performed based on the updated and iterated results.

[0035] In one possible implementation, first, define the reinforcement reward network at each moment Status indication , which contains the working frequency band of the antenna array at the current moment , Gain ,direction , and fixed interference in the environment and electromagnetic interference Among them, the fixed interference Modeling results from Fixed Interference Building Block 12, Electromagnetic Interference The state representation is then established based on the first electromagnetic interference result and the second electromagnetic interference result of the dynamic interference monitoring module 13. The expression is: , which describes the internal and external interference conditions and the working status of the current moment. Next, the reinforcement reward network is defined at each moment. Action representation , which includes the adjustment of three key parameters of the multi-band antenna array, namely, the frequency band adjustment , Gain Adjustment and direction adjustment Therefore, the action represents The expression is: , which specifies the optimal operation space of the reinforcement reward network for the system state.

[0036] Afterwards, construct a Reward function for system performance , the function consists of two parts: the first part , characterizes the system communication quality, where , , are the signal power, interference power and noise power received at the current moment, respectively. is the weight coefficient; the second part Characterizes the level of fixed interference and electromagnetic interference to the system, is the weight coefficient. Therefore, the reward function The expression is: ,The system’s anti-interference communication performance is quantitatively evaluated by the signal-to-noise ratio and ,interference strength. The larger the value is, the better the system performance is.

[0037] In order to enable the reinforcement reward network to learn and adapt to different array deployment schemes, a training database containing various typical array configurations is established. The array information and attribute information provided by the array construction module 11 are the source matching data of the training database. By applying different degrees of fixed interference and electromagnetic interference on the basis of the source matching data, and collecting the corresponding system performance indicators, a reinforcement learning training database that fully covers different array deployments and interference conditions can be built.

[0038] Then, using the established state representation , Action Representation And the reward function , configure the basic architecture of the reinforcement reward network. State representation As the input of the network, it describes the internal and external interference conditions and its own working status faced by the system at each moment; the action represents The output of the corresponding network specifies the network's optimal operation space for the system state; the reward function This provides the network with an evaluation index that quantifies the performance of the system and guides the optimization direction of the network parameters. By setting the network's hyperparameters such as the number of layers, the number of neurons, the activation function, and constructing a loss function based on the state representation, action representation, and reward function, a reinforcement reward network for adaptive interference compensation can be initially built. Next, in order to enable the reinforcement reward network to learn and adapt to the optimal compensation strategy under different interference conditions, the network is trained and optimized using the established training database. Specifically, various typical array deployment schemes and interference conditions in the training database are input into the reinforcement reward network, and the array parameters are virtually adjusted according to the actions output by the network, the adjusted system performance is evaluated, and the corresponding reward value is calculated. By repeating this process continuously and using reinforcement learning algorithms such as policy gradients and value functions, the network parameters are recursively updated and iteratively optimized, so that the network can gradually master the array compensation strategy for various complex interference conditions. After training, the reinforcement reward network can be deployed in the actual RFID system to guide the real-time adaptive interference compensation of the multi-band antenna array. When the system detects a change in the electromagnetic environment, the dynamic interference monitoring module 13 feeds back the real-time updated interference information to the decision center to generate a state representation at the current moment. The adjustment module 15 represents the state Input into the trained reinforcement reward network, the network can output the action representation for the current interference situation based on the learned strategy , including the adjustment of frequency band, gain and direction. The adjustment module 15 performs real-time parameter optimization on the multi-band antenna array based on this, completes interference adaptive compensation, and realizes dynamic optimization of system performance.

[0039] By configuring the strategy gradient and value function with state representation, action representation, and reward function, and performing iterative optimization based on the training database, the RFID multi-band antenna integration system can autonomously learn and adjust the array operating parameters according to the dynamic changes of environmental interference, achieve adaptive optimization of performance, effectively overcome the interference challenges in complex electromagnetic environments, and improve the intelligence level and environmental adaptability of the system.

[0040] Furthermore, the policy gradient is as follows: ; ; ; in, Characterize the policy function, that is, given the state Next, action The probability of control, Representation in state Take Action The value of represents all possible actions, is the gradient of the strategy, which means relative to the parameter The gradient of the expected reward of Indicated in strategy The expected value under For time The strategy parameters, For time The strategy parameters, is the learning rate, is the policy gradient, which means relative to the parameter The objective function gradient.

[0041] In a preferred embodiment, in the reinforcement reward network, the policy function Represents a given state Next, the network selects an action representation The probability of . is the parameter of the policy function, which controls the preference of action selection. The expression of the policy function is as follows: .in, is the value function, representing the state Take action The long-term cumulative reward expectation of This means that for all possible actions The exponential summation of the value function is performed. The larger the value function, the higher the probability of the action being selected, but at the same time it also retains a certain degree of exploratory nature to avoid premature convergence to the local optimum.

[0042] In the training phase, to optimize the policy function parameters , so that it can maximize the long-term cumulative reward and solve the gradient of the following objective function: .in, is the objective function, which means that The expected cumulative reward under . According to the policy gradient theorem, the gradient of the objective function It can be expressed as the state-action log probability and instant rewards The product of the strategy Intuitively, this gradient helps actions with high rewards and hurts actions with high rewards, thereby optimizing the policy function parameters toward maximizing the expected reward.

[0043] Based on the gradient of the objective function, the policy function parameters are adjusted using the stochastic gradient ascent algorithm. Perform update iteration, the iteration formula is: . Among them, among them, and Respectively Second and The parameter vector for the iteration, is the learning rate, which controls the parameter adjustment amplitude of each iteration. is the gradient of the objective function at the tth iteration.

[0044] Through sampling estimation of policy gradient and stochastic gradient ascent optimization, the policy function can adaptively adjust array parameters in complex and changeable electromagnetic environments, maximize the system communication performance and anti-interference capability, and provide strong support for intelligent and adaptive RFID antenna interference compensation.

[0045] Furthermore, the value function is as follows: ; Among them, the above formula represents the value function Value update formula, For the status Take Action Current value, is the learning rate, , is the discount factor, , Representation in the next state Take the best action Maximum value.

[0046] In a preferred embodiment, in the reinforcement reward network, the value function Represents the state Take action The long-term cumulative reward expectation. Since the state-action space of the actual system is often very complex, it is difficult to directly calculate the exact value of the value function. Therefore, the value function is approximated by recursive updating. Specifically, at each moment , according to the system's instant rewards and the maximum value at the next moment Value estimation, for the current state action pair of The value is updated, and the update formula is as follows: .

[0047] in, State-action pair Current Value estimation, is the learning rate, and its range is , controlling the adjustment amplitude of each update, is the discount factor, and its range is , represents the degree of attenuation of future rewards, Indicates the next state Take the best action The maximum that can be obtained Therefore, based on the system in state S Take action The actual instant reward , and the next state Optimal The value of the state-action pair of The target value is corrected and updated. The value consists of two parts: immediate reward Reflects the action taken The immediate return after It means that from the next state The discounted value of the optimal long-term cumulative reward at the beginning. By minimizing the estimated Values ​​and Goals The difference in value, Gradual improvement and convergence of value estimates.

[0048] The dynamic approximation estimation of the value function is achieved by recursive updating, which effectively overcomes the computational difficulties caused by the high complexity of the state-action space and makes full use of the immediate reward and the optimal next state. The timely feedback information provided by the value can achieve strong self-learning and online optimization capabilities, providing an important reference for accurately evaluating and optimizing the dynamic decision-making of multi-band RFID antenna arrays.

[0049] Furthermore, the embodiment of the present application also includes: Performing interference ratio evaluation on the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to establish an interference ratio evaluation result; A frequency band optimization decision is made based on the interference ratio evaluation result, and the frequency band optimization decision result is used as a pre-update result to re-perform interference adaptive compensation based on the reinforcement reward network.

[0050] In a feasible implementation, first, the intensity and influence of various types of interference data synchronized to the decision center by the feedback optimization module 14 are quantitatively analyzed. Specifically, for the fixed interference data generated by the fixed interference construction module 12, by analyzing its spectral components, power spectrum density and other parameters, the influence weights of fixed interference sources (such as surrounding buildings, topography, etc.) on the performance of different frequency bands of the RFID system are evaluated. For the first electromagnetic interference result (inter-band interference) and the second electromagnetic interference result (environmental noise interference) generated by the dynamic interference monitoring module 13, the power proportion of different types of interference is calculated to evaluate their proportion in the total interference. Through the above analysis, an interference ratio evaluation result reflecting the proportion of various types of interference data is obtained, which quantitatively characterizes the relative influence of fixed interference, inter-band interference and environmental noise interference on the RFID system, and provides an important reference for subsequent interference adaptive decision-making.

[0051] Then, based on the interference ratio evaluation results, the decision center comprehensively considers the fixed interference intensity of each frequency band, the interference level between frequency bands, and the degree of environmental noise pollution, runs the frequency band optimization algorithm, and obtains the frequency band optimization decision results, which include two aspects: one is the retention, elimination or adjustment of the original working frequency band; the other is the introduction and switching of alternative frequency bands. Its core goal is to select the optimal frequency band combination that is least affected by fixed interference and environmental noise, and has the weakest interference between frequency bands. The frequency band optimization decision results are used as the pre-update results, and the enhanced reward network is re-run to carry out interference adaptive compensation, further optimize the gain and directional parameters of the antenna array, and realize the coordinated optimization of frequency band selection and array configuration.

[0052] By introducing interference ratio evaluation and frequency band optimization decision, building a hierarchical and progressive interference adaptive compensation architecture, and pre-optimizing the frequency band selection according to the interference ratio evaluation results, the impact of fixed interference and environmental noise can be reduced at the source, creating more favorable electromagnetic environment conditions for array compensation, and enabling the RFID multi-band antenna system to respond to interference in complex electromagnetic environments more comprehensively, efficiently and flexibly from two dimensions: frequency band and array, thereby improving system performance and environmental adaptability.

[0053] Furthermore, the embodiment of the present application also includes: The verification and reinforcement module is used to establish a response data set, the response data set has a mapping relationship with the interference adaptive compensation result, evaluate the response delay and response accuracy through the response data set, establish the reinforcement result, and update the reinforcement reward network through the reinforcement result.

[0054] In a preferred embodiment, the interference adaptive RFID multi-band antenna signal integration system also includes a verification reinforcement module for evaluating and improving the interference adaptive compensation strategy of the reinforcement reward network, and achieving self-correction and self-optimization of the compensation strategy by introducing a response evaluation feedback mechanism.

[0055] First, the verification and enhancement module establishes a response data set, and the response data set has a mapping relationship with the interference adaptive compensation result performed by the adjustment module 15. The response data set records various performance indicators of the multi-band antenna array under different compensation strategies, such as received signal strength, signal-to-noise ratio, tag response delay and reading accuracy. By collecting the compensation results and their corresponding performance response data, the verification and enhancement module can build a response data set that fully reflects the effect of interference adaptive compensation. Secondly, the verification and enhancement module uses the established response data set to evaluate the response delay and response accuracy. Among them, the response delay evaluation determines the impact of the compensation scheme on the real-time performance of the system by calculating the mean and variance of the delay under different compensation strategies; the response accuracy evaluation evaluates the contribution of the compensation scheme to the system reliability by comparing the tag reading error rate and missed reading rate under different compensation strategies. The evaluation results can be expressed in numerical performance change percentages or grade scores, objectively reflecting the pros and cons of each compensation strategy. Afterwards, the verification and enhancement module forms an enhanced result for the current interference adaptive compensation strategy based on the evaluation results of the response delay and accuracy. If the enhancement result shows that the current compensation strategy not only improves the anti-interference ability but also meets the real-time and reliability requirements, the strategy is considered to be of high quality. The verification and reinforcement module will use the enhancement result as positive feedback to strengthen the current compensation strategy through reward signals. On the contrary, if the enhancement result shows that the current compensation strategy optimizes the anti-interference performance but at the expense of response delay or reading accuracy, the enhancement result will be used as negative feedback to prompt the reinforcement reward network to adjust the optimization direction through penalty signals.

[0056] The verification and reinforcement module realizes real-time verification and dynamic reinforcement of the interference adaptive compensation strategy. The closed-loop feedback mechanism gives the reinforcement reward network the ability to evaluate and improve, enabling it to continuously learn and evolve during actual application, and continuously provide high-quality interference adaptive compensation solutions.

[0057] Furthermore, the embodiment of the present application also includes: The continuous warning module is used to determine whether the enhancement result is a negative enhancement result for a preset number of consecutive times. If the judgment result is yes, the optimization direction penalty is performed according to the interference adaptive compensation result corresponding to the enhancement result, an abnormal warning is reported according to the penalty result, and the next round optimization strategy of the reinforcement reward network is adjusted.

[0058] In a preferred embodiment, in order to further enhance the system's recognition and early warning capabilities for low-quality interference adaptive compensation strategies and improve the reliability and stability of the enhanced reward network optimization strategy, a continuous early warning module can be added on the basis of the verification and enhancement module.

[0059] The continuous warning module works in conjunction with the verification and reinforcement module to monitor and warn the optimization process of the reinforcement reward network by analyzing the reinforcement results generated by the verification and reinforcement module. Specifically, the continuous warning module receives the reinforcement results output by the verification and reinforcement module in real time to determine whether it is a negative reinforcement result for a preset number of consecutive times. The negative reinforcement result means that through the response delay and response accuracy evaluation, the verification and reinforcement module believes that although the current interference adaptive compensation strategy improves the system's anti-interference performance, it fails to comprehensively and evenly improve the system performance at the expense of response real-time or reliability. If the interference adaptive compensation strategy generated by the reinforcement reward network continues to be negatively reinforced within the preset continuous optimization rounds (such as 5 rounds, 10 rounds, etc.), that is, the corresponding reinforcement results are negative for many consecutive times, it indicates that the optimization direction of the current reinforcement reward network may be biased or unstable, and needs to be adjusted and corrected in time. When the continuous warning module detects the above-mentioned continuous negative reinforcement, the warning and adjustment mechanism will be triggered. First, the continuous warning module marks the interference adaptive compensation result corresponding to the negative reinforcement result as a low-quality compensation strategy, and punishes the optimization direction during strategy evaluation and selection, that is, reduces its probability or priority of being selected. Through the penalty mechanism, the reinforcement reward network can actively avoid and abandon the compensation strategy that causes negative reinforcement for many times in a row, so as to avoid the continuous deviation of the optimization direction. Secondly, the continuous warning module reports an abnormal warning based on the penalty result, indicating that the current interference adaptive compensation process may have abnormal situations such as continuous negative reinforcement, which requires diagnosis and intervention, thereby arousing the attention of manual or automated supervision, so that it can intervene and analyze the causes of the abnormality in time, take corresponding error correction measures, and ensure the normal progress of the optimization process of the reinforcement reward network. After that, the continuous warning module uses the penalty result to adjust the next round of optimization strategy of the reinforcement reward network. Specifically, by modifying the loss function of the policy gradient function, on the basis of the original task goal, the penalty term of the number of continuous negative reinforcement is introduced, so that when the network updates the policy parameters, it not only optimizes the interference adaptive compensation performance, but also minimizes the probability of continuous negative reinforcement. At the same time, by dynamically adjusting the balance coefficient of exploration and utilization, the exploration intensity is increased when continuous negative reinforcement occurs, jumping out of the current strategy space that may be the local optimal, and seeking high-quality compensation strategies in a larger range. Through these strategy adjustments, the learning direction of the reinforcement reward network can be effectively improved and optimized, and its stability and robustness can be improved.

[0060] Through the continuous early warning module, the strategy quality monitoring and optimization error correction mechanism are realized, and abnormal situations such as directional deviation and oscillation that may occur in the process of strengthening the reward network optimization are discovered. Through strategy punishment and dynamic adjustment, the network is prompted to correct the optimization path in time, avoid the continuous generation of low-quality compensation strategies, improve the overall quality and reliability of the interference adaptive compensation strategy, and ensure the continuous and stable optimization of the performance of the RFID multi-band antenna system.

[0061] In summary, the interference adaptive RFID multi-band antenna signal integration system provided in the embodiments of the present application has the following technical effects: The array construction module is used to construct a multi-band antenna array and establish digital communication between the multi-band antenna array and the decision center. The decision center is initialized based on the array information and attribute information of the multi-band antenna array. The array information includes position structure information, and the attribute information includes parameter information, which provides the necessary hardware foundation for subsequent interference perception, optimization compensation and other functions. The fixed interference construction module is used to obtain the application scenario of the multi-band antenna array, obtain the signal source position, receive the position structure information of the decision center, establish a communication path based on the position structure information and the signal source position, and perform obstacle blocking interference and multipath interference fitting based on the communication path to generate fixed interference data, comprehensively obtain fixed interference information in the environment, and provide input for interference compensation. The dynamic interference monitoring module is used to obtain the preset communication band of the multi-band antenna array, perform inter-band interference evaluation based on the preset communication band, establish a first electromagnetic interference result, control the receiver to perform environmental electromagnetic noise monitoring, establish a second electromagnetic interference result based on the environmental electromagnetic noise monitoring result, obtain dynamic electromagnetic interference data in real time, and combine with fixed interference data to comprehensively reflect the interference status. The feedback optimization module is used to synchronize the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, call the enhanced reward network of the decision center to perform interference adaptive compensation, and establish the interference adaptive compensation result. The enhanced reward network is built with array information and attribute information as source matching data. By establishing the interference adaptive compensation result, the antenna optimization is guided. The adjustment module is used to optimize the parameter compensation of the multi-band antenna array based on the interference adaptive compensation result, improve the anti-interference ability and environmental adaptability of the RFID multi-band antenna in different application scenarios, and realize flexible and stable identification and data transmission.

[0062] Embodiment 2 Based on the same inventive concept as the interference adaptive RFID multi-band antenna signal integration system in the aforementioned embodiment, Figure 2 As shown, the embodiment of the present application provides an interference adaptive RFID multi-band antenna signal integration method, the method comprising: Constructing a multi-band antenna array, and establishing digital communication between the multi-band antenna array and a decision center, and completing the initialization of the decision center based on array information and attribute information of the multi-band antenna array, wherein the array information includes position structure information, and the attribute information includes parameter information; Obtain an application scenario of a multi-band antenna array, obtain a signal source position, receive position structure information of a decision center, establish a communication path based on the position structure information and the signal source position, and perform obstacle blocking interference and multipath interference fitting based on the communication path to generate fixed interference data; Obtaining a preset communication frequency band of the multi-band antenna array, performing inter-band interference evaluation based on the preset communication frequency band, establishing a first electromagnetic interference result, controlling the receiver to perform environmental electromagnetic noise monitoring, and establishing a second electromagnetic interference result based on the environmental electromagnetic noise monitoring result; Synchronize the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, call the reinforcement reward network of the decision center to perform interference adaptive compensation, and establish the interference adaptive compensation result, wherein the reinforcement reward network is built with array information and attribute information as source matching data; Parameter compensation optimization of the multi-band antenna array is performed based on the interference adaptive compensation result.

[0063] Furthermore, the reinforcement reward network of the decision center is called to perform interference adaptive compensation, which also includes: Establish the state representation, action representation, and reward function of the reinforcement reward network as follows: ; ; ; in, for The state at the moment indicates, for The frequency band of the moment, for The gain of time, for The direction of time, To fix the interference data, is electromagnetic interference data established according to the first electromagnetic interference result and the second electromagnetic interference result, for The action of the moment indicates, For frequency band adjustment, For gain adjustment, To adjust the direction, for The reward function at the moment, is the signal power, is the interference power, is the noise power, and is the weight coefficient; Establishing a training database using array information and attribute information as source matching data; The policy gradient and the value function are configured with the state representation, the action representation, and the reward function, the policy gradient and the value function are updated and iterated through the training database, and interference adaptive compensation is performed based on the updated and iterated results.

[0064] Furthermore, the policy gradient is as follows: ; ; ; in, Characterize the policy function, that is, given the state Next, action The probability of control, Representation in state Take Action The value of represents all possible actions, is the gradient of the strategy, which means relative to the parameter The gradient of the expected reward of Indicated in strategy The expected value under For time The strategy parameters, For time The strategy parameters, is the learning rate, is the policy gradient, which means relative to the parameter The objective function gradient.

[0065] Furthermore, the value function is as follows: ; Among them, the above formula represents the value function Value update formula, For the status Take Action Current value, is the learning rate, , is the discount factor, , Representation in the next state Take the best action Maximum value.

[0066] Furthermore, the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result are synchronized to the decision center, and the reinforcement reward network of the decision center is called to perform interference adaptive compensation and establish the interference adaptive compensation result, which also includes: Performing interference ratio evaluation on the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to establish an interference ratio evaluation result; A frequency band optimization decision is made based on the interference ratio evaluation result, and the frequency band optimization decision result is used as a pre-update result to re-perform interference adaptive compensation based on the reinforcement reward network.

[0067] Furthermore, the embodiment of the present application also includes: The verification and reinforcement module is used to establish a response data set, the response data set has a mapping relationship with the interference adaptive compensation result, evaluate the response delay and response accuracy through the response data set, establish the reinforcement result, and update the reinforcement reward network through the reinforcement result.

[0068] Furthermore, the embodiment of the present application also includes: The continuous warning module is used to determine whether the enhancement result is a negative enhancement result for a preset number of consecutive times. If the judgment result is yes, the optimization direction penalty is performed according to the interference adaptive compensation result corresponding to the enhancement result, an abnormal warning is reported according to the penalty result, and the next round optimization strategy of the reinforcement reward network is adjusted.

[0069] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0070] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these changes and variations.

Claims

1. Interference adaptive RFID multi-band antenna signal integration system, characterized in that: The system comprises: An array construction module is used to construct a multi-band antenna array and establish digital communication between the multi-band antenna array and a decision center, and complete the initialization of the decision center based on array information and attribute information of the multi-band antenna array, wherein the array information includes position structure information and the attribute information includes parameter information; A fixed interference construction module is used to obtain the application scenario of the multi-band antenna array, obtain the signal source position, receive the position structure information of the decision center, establish a communication path based on the position structure information and the signal source position, and perform obstacle blocking interference and multipath interference fitting based on the communication path to generate fixed interference data; A dynamic interference monitoring module is used to obtain a preset communication frequency band of a multi-band antenna array, perform inter-band interference evaluation based on the preset communication frequency band, establish a first electromagnetic interference result, control the receiver to perform environmental electromagnetic noise monitoring, and establish a second electromagnetic interference result based on the environmental electromagnetic noise monitoring result; A feedback optimization module is used to synchronize the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, call the reinforcement reward network of the decision center to perform interference adaptive compensation, and establish the interference adaptive compensation result, wherein the reinforcement reward network is built with array information and attribute information as source matching data; An adjustment module is used to perform parameter compensation optimization of the multi-band antenna array based on the interference adaptive compensation result.

2. The interference adaptive RFID multi-band antenna signal integration system according to claim 1, characterized in that: The reinforcement reward network calling the decision center performs interference adaptive compensation, and also includes: Establish the state representation, action representation, and reward function of the reinforcement reward network as follows: ； ； ； in, for The state at the moment indicates, for The frequency band of the moment, for The gain of time, for The direction of time, To fix the interference data, is electromagnetic interference data established according to the first electromagnetic interference result and the second electromagnetic interference result, for The action of the moment indicates, For frequency band adjustment, For gain adjustment, To adjust the direction, for The reward function at the moment, is the signal power, is the interference power, is the noise power, and is the weight coefficient; Establishing a training database using array information and attribute information as source matching data; The policy gradient and the value function are configured with the state representation, the action representation, and the reward function, the policy gradient and the value function are updated and iterated through the training database, and interference adaptive compensation is performed based on the updated and iterated results.

3. The interference adaptive RFID multi-band antenna signal integration system according to claim 2, characterized in that: The policy gradient is as follows: ； ； ； in, Characterize the policy function, that is, given the state Next, action The probability of control, Representation in state Take Action The value of represents all possible actions, is the gradient of the strategy, which means relative to the parameter The gradient of the expected reward of Indicated in strategy The expected value under For time The strategy parameters, For time The strategy parameters, is the learning rate, is the policy gradient, which means relative to the parameter The objective function gradient.

4. The interference adaptive RFID multi-band antenna signal integration system as claimed in claim 3, characterized in that: The value function is as follows: ； Among them, the above formula represents the value function Value update formula, For the status Take Action Current value, is the learning rate, , is the discount factor, , Representation in the next state Take the best action Maximum value.

5. The interference adaptive RFID multi-band antenna signal integration system according to claim 1, characterized in that: The method further includes: synchronizing the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, calling the reinforcement reward network of the decision center to perform interference adaptive compensation, and establishing the interference adaptive compensation result. Performing interference ratio evaluation on the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to establish an interference ratio evaluation result; A frequency band optimization decision is made based on the interference ratio evaluation result, and the frequency band optimization decision result is used as a pre-update result to re-perform interference adaptive compensation based on the reinforcement reward network.

6. The interference adaptive RFID multi-band antenna signal integration system according to claim 1, characterized in that: The system further comprises: The verification and reinforcement module is used to establish a response data set, the response data set has a mapping relationship with the interference adaptive compensation result, evaluate the response delay and response accuracy through the response data set, establish the reinforcement result, and update the reinforcement reward network through the reinforcement result.

7. The interference adaptive RFID multi-band antenna signal integration system according to claim 6, characterized in that: The system further comprises: The continuous warning module is used to determine whether the enhancement result is a negative enhancement result for a preset number of consecutive times. If the judgment result is yes, the optimization direction penalty is performed according to the interference adaptive compensation result corresponding to the enhancement result, an abnormal warning is reported according to the penalty result, and the next round optimization strategy of the reinforcement reward network is adjusted.

8. An interference-adaptive RFID multi-band antenna signal integration method, characterized in that: The method comprises: Constructing a multi-band antenna array, and establishing digital communication between the multi-band antenna array and a decision center, and completing the initialization of the decision center based on array information and attribute information of the multi-band antenna array, wherein the array information includes position structure information, and the attribute information includes parameter information; Obtain an application scenario of a multi-band antenna array, obtain a signal source position, receive position structure information of a decision center, establish a communication path based on the position structure information and the signal source position, and perform obstacle blocking interference and multipath interference fitting based on the communication path to generate fixed interference data; Obtaining a preset communication frequency band of the multi-band antenna array, performing inter-band interference evaluation based on the preset communication frequency band, establishing a first electromagnetic interference result, controlling the receiver to perform environmental electromagnetic noise monitoring, and establishing a second electromagnetic interference result based on the environmental electromagnetic noise monitoring result; Synchronize the fixed interference data, the first electromagnetic interference result, and the second electromagnetic interference result to the decision center, call the reinforcement reward network of the decision center to perform interference adaptive compensation, and establish the interference adaptive compensation result, wherein the reinforcement reward network is built with array information and attribute information as source matching data; Parameter compensation optimization of the multi-band antenna array is performed based on the interference adaptive compensation result.

Images