Product identification method and device based on RFID technology, equipment and medium

Abstract

The invention relates to a product identification method and device based on an RFID technology, equipment and a medium. The method comprises the following steps: extracting a time sequence feature of a first RFID signal to obtain a first RFID signal feature; according to the first RFID signal features, matching acquisition regulation and control parameters from a preset acquisition regulation and control parameter library; acquiring a second RFID signal, wherein the second RFID signal is acquired by the RFID equipment of which the acquisition parameters are adjusted by the acquisition regulation and control parameters; evaluating the second RFID signal by adopting a preset acquisition index to obtain an evaluation result; and according to the evaluation result, the acquisition regulation and control parameters are optimized to obtain optimized regulation and control parameters, and the optimized regulation and control parameters are used for updating the acquisition parameters of the RFID equipment. By adopting the method, the accuracy, the stability and the environmental adaptability of RFID product identification can be effectively improved by dynamically adjusting the acquisition parameters.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things automatic identification technology, and in particular relates to product identification methods, devices, equipment and media based on RFID technology. Background Technology

[0002] With the rapid development of IoT automatic identification technology, RFID (Radio Frequency Identification) technology has emerged. This technology features non-contact data transmission, flexible identification distance, the ability to read multiple tags in batches, and strong resistance to environmental pollution. It has been widely used in product identification scenarios in various fields such as logistics warehousing, product traceability, intelligent manufacturing, and retail management. In traditional technology, RFID devices typically use pre-set fixed acquisition parameters (including signal transmission power, signal sampling frequency, signal frame interval, etc.) to collect signals when identifying products. During the identification process, regardless of whether the environmental conditions of the application scenario (such as electromagnetic interference intensity, distribution of obstructions) change, whether the product tag is in motion, or whether there is mutual interference from multiple tags communicating simultaneously, the acquisition parameters remain unchanged. Only simple filtering and parsing processing of the acquired raw RFID signals is performed to extract tag information to complete product identification.

[0003] The current RFID product identification method based on fixed acquisition parameters has significant drawbacks: real-world application scenarios are complex and dynamic. Environmental interference (such as electromagnetic interference and occlusion loss), changes in tag movement status, and mutual interference among multiple tags can directly lead to problems such as overlap distortion, noise pollution, and intensity fluctuations in the acquired RFID signals. Fixed acquisition parameters cannot adapt to these dynamic changes, making it difficult to guarantee signal quality. Traditional methods lack a multi-dimensional evaluation mechanism for acquired signals and have not established a feedback optimization channel for acquisition parameters based on signal quality. They cannot adjust acquisition parameters according to the actual signal state, resulting in low product identification accuracy, poor stability of the identification process, and insufficient adaptability to complex application environments, making it difficult to meet the requirements for high-precision and high-reliability product identification. Summary of the Invention

[0004] Therefore, it is necessary to provide a product identification method, device, equipment, and medium based on RFID technology that can solve the above problems.

[0005] In a first aspect, this application provides a product identification method based on RFID technology, including:

[0006] Extract the temporal features of the first RFID signal to obtain the features of the first RFID signal;

[0007] Based on the characteristics of the first RFID signal, the acquisition and control parameters are matched from the preset acquisition and control parameter library;

[0008] Acquire a second RFID signal, which is obtained by an RFID device whose acquisition parameters have been adjusted by acquisition control parameters;

[0009] The second RFID signal is evaluated using preset collection indicators to obtain evaluation results;

[0010] Based on the evaluation results, the acquisition and control parameters are optimized to obtain optimized control parameters, which are then used to update the acquisition parameters of the RFID device.

[0011] In one embodiment, the timing features of the first RFID signal are extracted to obtain the first RFID signal features, including:

[0012] The overlapping signals in the first RFID signal are separated to obtain the separated signal;

[0013] Extract the intensity distribution of the separated signal;

[0014] Based on the intensity value distribution, the noise components are determined according to the preset noise model;

[0015] Based on the noise components, an adaptive filtering algorithm is used to denoise the separated signal to obtain a purified signal.

[0016] The timing characteristics of the purification signal are acquired and used as the first RFID signal characteristics.

[0017] In one embodiment, based on the first RFID signal characteristics, acquisition and control parameters are matched from a preset acquisition and control parameter library, including:

[0018] An improved ALOHA algorithm is used to analyze the features of the first RFID signal and predict the tag's adjusted frame length.

[0019] Obtain the denoising parameters of the adaptive filtering algorithm;

[0020] Integrate the tag adjustment frame length and denoising parameters to generate initial adjustment parameters;

[0021] Based on the initial adjustment parameters, feature matching is performed in the preset acquisition and control parameter library to determine the acquisition and control parameters, which include signal transmission power, signal sampling frequency, and signal frame interval.

[0022] In one embodiment, the preset acquisition indicators include signal integrity indicators, signal stability indicators, and signal identification indicators;

[0023] The second RFID signal is evaluated using preset acquisition indicators to obtain evaluation results, including:

[0024] Based on signal integrity index, signal stability index, and signal identification index, a hierarchical identification index evaluation model is constructed. The identification index evaluation model includes an environmental interference loss sub-layer corresponding to the signal integrity index, a tag motion stability sub-layer corresponding to the signal stability index, and a tag mutual interference sub-layer corresponding to the signal identification index.

[0025] The second RFID signal is input into the hierarchical identification index evaluation model. The signal deviation coefficient is calculated through the environmental interference loss sub-layer, the signal adaptation deviation is calculated through the tag motion stability sub-layer, and the signal mutual interference index is calculated through the tag mutual interference sub-layer.

[0026] Based on preset scenario weights, the signal deviation coefficient, signal adaptation deviation, and signal interference index are weighted and calculated to obtain a comprehensive defect value.

[0027] The signal confidence level is determined by comparing the overall defect value with a preset confidence assessment threshold.

[0028] The signal confidence level, signal deviation coefficient, signal adaptation deviation, and signal interference index are integrated as the evaluation result.

[0029] In one embodiment, the collected control parameters are optimized based on the evaluation results to obtain optimized control parameters, which are achieved using the following formula:

[0030]

[0031] in, For signal transmission power, The signal sampling frequency, For signal frame interval, To optimize signal transmission power, To optimize the signal sampling frequency, To optimize the signal frame interval, Let C be a 3x3 identity matrix, D be the adjustment scaling factor, D be the signal deviation factor, A be the signal adaptation deviation, and I be the signal cross-interference index. The scene weights are the signal deviation coefficients. Scenario weights for signal adaptation deviations The scene weights for the signal interference index, The confidence threshold for the signal deviation coefficient. The confidence threshold for signal adaptation deviation. The confidence threshold for the signal interference index is [value]. This is a matrix of positive coefficients for multiple maintenance functions.

[0032] In one embodiment, an improved ALOHA algorithm is used to analyze the features of the first RFID signal and predict the tag adjustment frame length, including:

[0033] Frame boundaries for identifying the first RFID signal feature;

[0034] Based on frame boundaries, the frame conflict timing parameters of the first RFID signal feature are extracted using signal level transition detection technology.

[0035] An improved ALOHA algorithm is used to calculate the collision probability of frame collision timing parameters to obtain the collision probability.

[0036] The frame length is adjusted based on the predicted label determined by the collision probability.

[0037] In one embodiment, the initial adjustment parameters are generated by integrating the label-adjusted frame length with the denoising parameters of the adaptive filtering algorithm, including:

[0038] Extract the labels, adjust the frame length, and use the denoising parameters of the adaptive filtering algorithm to construct a two-parameter dataset;

[0039] The original two-parameter dataset is time-aligned to obtain an aligned two-parameter dataset;

[0040] The interval mapping method is used to map the frame length adjustment value of each label in the aligned two-parameter dataset to the effective range of the corresponding denoising parameters of the adaptive filtering algorithm;

[0041] Extract parameter values ​​that match the frame length temporal features within the effective interval to form initial adjustment parameters.

[0042] Secondly, this application also provides a product identification device based on RFID technology, comprising:

[0043] The first signal feature extraction module is used to extract the temporal features of the first RFID signal to obtain the first RFID signal features;

[0044] The control parameter matching module is used to match the acquisition control parameters from the preset acquisition control parameter library based on the characteristics of the first RFID signal;

[0045] The second signal acquisition module is used to acquire the second RFID signal, which is obtained by the RFID device after the acquisition parameters are adjusted by the acquisition control parameters.

[0046] The second signal evaluation module is used to evaluate the second RFID signal using preset acquisition indicators and obtain the evaluation results.

[0047] The control parameter optimization module is used to optimize the collected control parameters based on the evaluation results, and obtain optimized control parameters. The optimized control parameters are used to update the collection parameters of the RFID device.

[0048] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described product identification method based on RFID technology.

[0049] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described product identification method based on RFID technology.

[0050] The aforementioned product identification method, apparatus, equipment, and medium based on RFID technology extracts the temporal characteristics of a first RFID signal to obtain signal features, matches acquisition and control parameters from a preset acquisition and control parameter library, and uses the adjusted RFID equipment to acquire a second RFID signal. The signal is then evaluated using preset acquisition indicators to obtain evaluation results, and the acquisition and control parameters are optimized based on the evaluation results to update the equipment's acquisition parameters. By dynamically adjusting the acquisition parameters, the shortcomings of traditional fixed-parameter methods in complex environments are overcome: parameter matching based on signal features ensures the adaptability of acquisition parameters to real-time signal states, reducing signal distortion caused by environmental changes; and a closed-loop feedback mechanism is formed through evaluation and optimization, continuously improving signal quality and enhancing the accuracy, stability, and environmental adaptability of product identification. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart of the product identification method based on RFID technology of the present invention;

[0053] Figure 2 This is a structural diagram of the product identification device based on RFID technology of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] In one embodiment, such as Figure 1As shown, a product identification method based on RFID technology is provided. This embodiment illustrates the method by applying it to an RFID reader, product tag, terminal, and server. It is understood that this method can also be applied independently to intelligent RFID devices integrating the above functions, or to IoT systems with multiple devices working together, achieved through interaction between various hardware components. In product identification scenarios requiring dynamic adaptation to complex environments, such as logistics warehousing, product traceability, and intelligent manufacturing, when faced with environmental interference, tag movement, or interference between multiple tags leading to signal distortion and insufficient identification accuracy, the RFID reader collects the RFID signal from the product tag and transmits it to a computer device. The terminal or server executes relevant algorithms to extract signal features, matches and adjusts collection parameters, and then feeds the adjusted parameters back to the RFID reader to optimize the collection configuration. Subsequently, through a closed-loop interaction of signal evaluation and parameter iterative optimization, the stability and accuracy of product identification are ensured. In this embodiment, the method includes the following steps:

[0056] S01, extract the timing features of the first RFID signal to obtain the features of the first RFID signal.

[0057] The first RFID signal is the original radio frequency signal initially collected by the RFID device. It may contain noise and distortion introduced by environmental interference, tag movement, or multi-tag interference. The timing features refer to the attributes of the signal in the time dimension, such as signal strength distribution, frame boundary timing, and level transition patterns, which are used to characterize the dynamic behavior of the signal. The first RFID signal can be separated by overlapping signals to eliminate multi-tag conflicts. Then, based on the strength value distribution and a preset noise model, noise components are identified. An adaptive filtering algorithm is used to perform adaptive denoising to obtain a purified signal. The timing features are extracted from the purified signal as the features of the first RFID signal, providing a data foundation.

[0058] S02, based on the characteristics of the first RFID signal, match the acquisition and control parameters from the preset acquisition and control parameter library.

[0059] The preset acquisition and control parameter library is a database that pre-stores various acquisition parameter configurations, including adjustable parameters such as signal transmission power, signal sampling frequency, and signal frame interval. Matching acquisition and control parameters refers to the process of dynamically selecting the optimal parameters based on signal characteristics. In implementation, the first RFID signal characteristics can be processed using a feature analysis algorithm (such as the improved ALOHA algorithm) to predict the tag adjustment frame length, and denoising parameters can be integrated to generate initial adjustment parameters. Feature matching is then performed in the parameter library to determine the appropriate acquisition and control parameters, achieving dynamic optimization of the correspondence between parameters and real-time signal status.

[0060] S03, acquire the second RFID signal, which is acquired by the RFID device after the acquisition parameters have been adjusted by the acquisition control parameters.

[0061] The second RFID signal is a radio frequency signal acquired by the RFID device after the acquisition parameters are dynamically adjusted by the acquisition control parameters. In practice, the matching acquisition control parameters can be applied to the RFID device hardware, its acquisition settings can be updated, and the signal can be reacquired to obtain the second RFID signal, which provides input for subsequent evaluation of the control effect.

[0062] S04, the second RFID signal is evaluated using preset acquisition indicators to obtain the evaluation results.

[0063] The preset acquisition indicators are pre-established multi-dimensional signal quality assessment standards, which may include core indicators such as signal integrity, stability, and recognizability, used to quantify the reliability of RFID signals. The assessment result is a comprehensive judgment output based on the analysis of these indicators, which may include parameters such as signal defect value and confidence level. In implementation, an assessment model (such as a hierarchical identification indicator assessment system) can be constructed to perform multi-dimensional analysis on the second RFID signal. For example, by calculating parameters such as signal deviation coefficient, adaptation deviation, and mutual interference index, and combining them with preset scenario weights, a weighted fusion is performed to generate a comprehensive assessment value. The signal quality level is determined by threshold comparison, thereby achieving an objective quantitative assessment of the signal status.

[0064] S05. Based on the evaluation results, the acquisition and control parameters are optimized to obtain optimized control parameters, which are then used to update the acquisition parameters of the RFID device.

[0065] Among them, the optimized control parameters are the final parameters used to iteratively update the equipment configuration after optimization calculation. When implementing this, a parameter optimization model can be established (such as a mathematical optimization algorithm that combines a multidimensional weight matrix with a confidence threshold). Based on the degree of deviation between each index value in the evaluation results and the preset threshold, the parameter correction amount is dynamically calculated, and the collected control parameters are weighted and adjusted to generate optimized control parameters. The collected parameters are then updated in real time through the equipment interface.

[0066] The aforementioned product identification method based on RFID technology extracts the temporal characteristics of a first RFID signal to obtain its features. Based on these features, it matches acquisition and control parameters from a pre-set acquisition and control parameter library. Then, it uses the adjusted RFID device to acquire a second RFID signal, evaluates this signal using pre-set acquisition indicators, and optimizes the acquisition and control parameters based on the evaluation results to update the device's acquisition parameters. By dynamically adjusting the acquisition parameters, it overcomes the shortcomings of traditional fixed-parameter methods, which are susceptible to interference in complex environments, thus enhancing the accuracy, stability, and environmental adaptability of product identification.

[0067] In one embodiment, the timing features of the first RFID signal are extracted to obtain the first RFID signal features, including:

[0068] S11, Separate the overlapping signals in the first RFID signal to obtain the separated signal;

[0069] S12, extract the intensity value distribution of the separated signal;

[0070] S13, Based on the intensity value distribution, determine the noise components according to the preset noise model;

[0071] Based on the noise components, an adaptive filtering algorithm is used to denoise the separated signal to obtain a purified signal.

[0072] The timing characteristics of the purification signal are acquired and used as the first RFID signal characteristics.

[0073] For example, an independent component analysis algorithm based on blind source separation can be used to process the first RFID signal. By identifying the correlation and independence characteristics of the signal, overlapping signals caused by simultaneous communication of multiple tags are separated to obtain a separated signal without signal overlap. The separated signal is converted into a digital signal by the analog-to-digital conversion module of the signal acquisition device. Statistical analysis methods are used to extract the signal strength distribution of the digital signal within a preset time window, which may include key distribution parameters such as maximum, minimum, average, and variance of the strength. Based on the extracted strength distribution, combined with a preset noise model (such as thermal noise, electromagnetic interference noise, etc.) established by collecting RFID signal noise data under different environments through experiments, the strength distribution is compared and matched with the noise features in the preset noise model to identify the type and intensity ratio of noise components contained in the separated signal. According to the determined noise components, a minimum mean square error adaptive filtering algorithm can be selected to selectively filter out the noise components in the separated signal by dynamically adjusting the filtering coefficients, resulting in a purified signal free of noise interference. Key features of the purified signal in the time dimension, such as the signal frame boundary timing, level transition time, signal duration, and intensity change trend, can be extracted using time-series analysis tools. These are the characteristics of the first RFID signal.

[0074] In one embodiment, based on the first RFID signal characteristics, acquisition and control parameters are matched from a preset acquisition and control parameter library, including:

[0075] S21, the improved ALOHA algorithm is used to analyze the features of the first RFID signal and predict the tag adjustment frame length;

[0076] S22, obtain the denoising parameters of the adaptive filtering algorithm;

[0077] S23, integrate the tag adjustment frame length and denoising parameters to generate initial adjustment parameters;

[0078] S24. Based on the initial adjustment parameters, feature matching is performed in the preset acquisition and control parameter library to determine the acquisition and control parameters, which include signal transmission power, signal sampling frequency and signal frame interval.

[0079] Specifically, an improved ALOHA algorithm can be used to analyze the first RFID signal features to predict tag adjustment frame length. Frame boundaries in the first RFID signal features are determined using signal edge detection technology. Based on these frame boundaries, signal level transition detection technology is used to extract frame collision timing parameters. These parameters are then input into the improved ALOHA algorithm, which uses a built-in Poisson distribution-based collision probability calculation model to calculate the frame collision probability during multi-tag communication. If the collision probability is higher than a preset collision threshold (e.g., 5%), the frame length is increased by a preset step size (e.g., 5ms per step). If the probability is lower than the preset threshold... If the value is 2%, the frame length is reduced by a preset step size (e.g., 3ms per step) until the collision probability is within a preset reasonable range (2%-5%), thus determining the predicted label adjustment frame length. The denoising parameters of the adaptive filtering algorithm are obtained. These parameters are key parameters generated and stored in the parameter cache module during the denoising process of the separated signal using the adaptive filtering algorithm, including filter coefficients, iteration step size, and filter window size. The label adjustment frame length and denoising parameters are integrated to generate initial adjustment parameters. The predicted label adjustment frame length and the aforementioned denoising parameters are extracted to construct a two-parameter dataset. Linear interpolation can be used to analyze this two-parameter dataset. The dataset undergoes time-series alignment to obtain an aligned two-parameter dataset. An interval mapping method can be used to map the frame length adjustment value of each label in the aligned two-parameter dataset to the effective range of the corresponding adaptive filtering algorithm's denoising parameters. The denoising parameter values ​​within this effective range that match the frame length temporal features are extracted to form the initial adjustment parameters. Based on the initial adjustment parameters, feature matching is performed in a preset acquisition and control parameter library to determine the acquisition and control parameters. This preset acquisition and control parameter library is a pre-constructed structured relational database that stores several sets of preset initial adjustment parameters and their corresponding signal transmission powers (unit: dBm, range: 15-). The mapping relationship between the initial adjustment parameters (30dBm), signal sampling frequency (MHz, range 2-8MHz), and signal frame interval (ms, range 15-40ms) is established. During feature matching, the Euclidean distance formula is used to calculate the feature similarity between the currently generated initial adjustment parameters and each set of preset initial adjustment parameters in the library (the label adjustment frame length, filter coefficient, iteration step size, and filter window size in the initial adjustment parameters are used as feature vectors). The signal transmission power, signal sampling frequency, and signal frame interval corresponding to the preset initial adjustment parameters with the smallest Euclidean distance (i.e. the highest similarity) are selected as the determined acquisition and control parameters.

[0080] In one embodiment, the preset acquisition indicators include signal integrity indicators, signal stability indicators, and signal identification indicators;

[0081] The second RFID signal is evaluated using preset acquisition indicators to obtain evaluation results, including:

[0082] S31. Based on the signal integrity index, signal stability index, and signal identification index, a hierarchical identification index evaluation model is constructed. The identification index evaluation model includes an environmental interference loss sub-layer corresponding to the signal integrity index, a tag motion stability sub-layer corresponding to the signal stability index, and a tag mutual interference sub-layer corresponding to the signal identification index.

[0083] S32, input the second RFID signal into the hierarchical identification index evaluation model, calculate the signal deviation coefficient through the environmental interference loss sub-layer, calculate the signal adaptation deviation through the tag motion stability sub-layer, and calculate the signal mutual interference index through the tag mutual interference sub-layer.

[0084] S33, based on preset scene weights, performs weighted calculations on the signal deviation coefficient, signal adaptation deviation, and signal interference index to obtain a comprehensive defect value;

[0085] S34, determine the signal confidence level by comparing the comprehensive defect value with a preset confidence assessment threshold;

[0086] S35 integrates signal confidence level, signal deviation coefficient, signal adaptation deviation, and signal interference index as the evaluation result.

[0087] For example, the preset acquisition indicators include signal integrity indicators (used to characterize the degree of conformity between the RFID signal and the standard signal after environmental interference during transmission), signal stability indicators (used to reflect the fluctuation of parameters such as signal strength and phase under tag motion), and signal distinguishability indicators (used to measure the distinguishability of each tag signal when multiple tags communicate simultaneously). A hierarchical identification indicator evaluation model can be constructed based on the above three types of indicators. This model has a modular architecture. The environmental interference loss sublayer corresponding to the signal integrity indicator has a built-in electromagnetic interference loss calculation unit (based on a preset mapping relationship table between electromagnetic interference intensity and signal attenuation) and an obstruction loss calculation unit (based on experimental data models of obstruction material, thickness, and signal loss rate). The tag motion stability sublayer corresponding to the signal stability indicator integrates a tag motion speed detection unit (through signal Doppler). The model includes a frequency shift calculation speed unit and a signal fluctuation threshold matching unit (different speeds correspond to different allowable signal fluctuation thresholds). The tag mutual interference sublayer corresponding to the signal identification index includes a multi-tag signal separation unit (using wavelet transform to separate overlapping signals) and a signal cross-correlation analysis unit (calculating the cross-correlation coefficients of different tag signals). The second RFID signal is input into this hierarchical identification index evaluation model. The signal deviation coefficient is calculated through the environmental interference loss sublayer. The electromagnetic interference loss calculation unit obtains the current environmental electromagnetic interference intensity (collected by the interference detection module built into the RFID device, unit dBμV / m) and matches the corresponding signal attenuation. Simultaneously, the obstruction loss calculation unit calculates the signal loss caused by obstruction based on the information of the obstruction between the RFID device and the tag (collected by external sensors of the device, including the material and thickness of the obstruction). The total signal loss is then substituted into the " (The standard signal strength is the signal strength at the same distance under interference-free and unobstructed conditions, which is pre-stored in the model database), to obtain the signal deviation coefficient; the signal adaptation deviation is calculated through the tag motion stabilization sublayer, and the tag motion speed detection unit calculates the tag motion speed based on the Doppler frequency shift of the second RFID signal (frequency shift = actual signal frequency - transmitted signal frequency). The signal fluctuation threshold matching unit can call the preset "speed-fluctuation threshold" lookup table based on this speed (e.g., speed 0-0.5m / s corresponds to fluctuation threshold ±5%, 0.5-1m / s corresponds to ±8%) to calculate the difference between the actual fluctuation amplitude of the second RFID signal (the ratio of the difference between the maximum and minimum signal strength values ​​over a period of time to the average value) and the corresponding fluctuation threshold. This difference is the signal adaptation deviation; the signal mutual interference index is calculated through the tag mutual interference sublayer, and the multi-tag signal separation unit can use this index to analyze the second RFID signal. The overlapping signals in the tag are separated to obtain the independent signals of each individual tag. The signal cross-correlation analysis unit calculates the cross-correlation coefficient (ranging from -1 to 1) between any two independent signals. If the absolute value of the cross-correlation coefficient is ≥0.6, cross-interference is determined to exist. The proportion of cross-interfering signal pairs to the total number of signal pairs is counted, and this proportion is the signal cross-interference index. The above three parameters are weighted according to the preset scenario weights to obtain the comprehensive defect value. The preset scenario weights can be experimentally calibrated according to the actual application scenario. For example, in the logistics and warehousing scenario (strong environmental interference, slow tag movement), the signal deviation coefficient weight is set to 0.4, the signal adaptation deviation weight is set to 0.3, and the signal cross-interference index weight is set to 0.3. In the retail settlement scenario (dense multiple tags, weak interference), the signal cross-interference index weight is set to 0.4, the signal deviation coefficient weight is set to 0.2, and the signal adaptation deviation weight is set to 0.4. The weighting formula is " (The weights D, A, and I are the preset scenario weights for the three parameters); The signal confidence level is determined by comparing the comprehensive defect value with the preset confidence assessment thresholds (determined in advance through a large number of experiments, divided into four levels: threshold 1=5%, threshold 2=15%, threshold 3=30%). If the comprehensive defect value < threshold 1, it is level A (high confidence); if threshold 1 ≤ comprehensive defect value < threshold 2, it is level B (medium confidence); if threshold 2 ≤ comprehensive defect value < threshold 3, it is level C (low confidence); if the comprehensive defect value ≥ threshold 3, it is level D (very low confidence); The signal confidence level, signal deviation coefficient (retaining two decimal places), signal adaptation deviation (retaining two decimal places), and signal interference index are integrated into structured data as the evaluation result.

[0088] In one embodiment, S41, the collected control parameters are optimized based on the evaluation results to obtain optimized control parameters, which are achieved using the following formula:

[0089]

[0090] in, For signal transmission power, The signal sampling frequency, For signal frame interval, To optimize signal transmission power, To optimize the signal sampling frequency, To optimize the signal frame interval, Let C be a 3x3 identity matrix, D be the adjustment scaling factor, D be the signal deviation factor, A be the signal adaptation deviation, and I be the signal cross-interference index. The scene weights are the signal deviation coefficients. Scenario weights for signal adaptation deviations The scene weights for the signal interference index, The confidence threshold for the signal deviation coefficient. The confidence threshold for signal adaptation deviation. The confidence threshold for the signal interference index is [value]. This is a matrix of positive coefficients for multiple maintenance functions.

[0091] Specifically, the initial signal transmission power Signal sampling frequency Signal frame interval The parameters are the acquisition and control parameters previously matched from the preset acquisition and control parameter library; the signal deviation coefficient D, signal adaptation deviation A, and signal interference index I are the core results obtained after evaluating the second RFID signal through the hierarchical identification index evaluation model; the scene weight of the signal deviation coefficient is... Scene weights for signal adaptation deviation Scene weights of the signal interference index The maximum permissible defect value (i.e., confidence assessment threshold) corresponding to these three types of parameters. , , ), and a multi-maintenance positive coefficient matrix consisting of the adjustment ratio C and multiple correction coefficients. Pre-calibrated fixed data can be collected from a large number of RFID signal acquisition experiments in different application scenarios (such as logistics warehousing and retail settlement). Adjustment coefficients can be determined based on the signal confidence level (A to D), with smaller values ​​(e.g., 0.1) for high-confidence A-level signals and larger values ​​(e.g., 0.4) for low-confidence D-level signals. The adjustment range of the parameters is calculated by combining the scene weight, correction coefficient matrix, and the differences between the signal deviation coefficient, adaptation deviation, mutual interference index, and their respective maximum allowable defect values. This allows for the calculation of the required adjustment ratios for the initial transmit power, sampling frequency, and frame interval (e.g., ...). The closer the signal deviation coefficient is to the maximum allowable defect value, the larger the adjustment ratio of the transmission power. The initial parameters are multiplied by (1 + adjustment ratio coefficient C) to obtain the optimized signal transmission power, signal sampling frequency, and signal frame interval. These optimized control parameters can be written into the parameter register of the RFID device through the parameter configuration interface to update the acquisition parameters of the RFID device and complete the optimization iteration of the acquisition parameters.

[0092] In one embodiment, an improved ALOHA algorithm is used to analyze the features of the first RFID signal and predict the tag adjustment frame length, including:

[0093] S51, identify the frame boundary of the first RFID signal feature;

[0094] S52, based on frame boundaries, extracts frame conflict timing parameters of the first RFID signal features through signal level transition detection technology;

[0095] S53 uses the improved ALOHA algorithm to calculate the collision probability of frame collision timing parameters and obtain the collision probability;

[0096] S54, determine the predicted label and adjust the frame length based on the collision probability.

[0097] For example, a signal edge detection technique based on the first derivative threshold method can be used to process the time-series data of the first RFID signal feature. The first derivative of this time-series data at each sampling point (reflecting the rate of signal level change) is calculated. The preset derivative threshold is 0.5V / ms (calibrated through experiments with 100 sets of standard RFID signal frames to ensure accurate differentiation between normal level fluctuations and frame boundary abrupt changes). When the absolute value of the derivative exceeds 0.5V / ms, if the derivative changes from negative to positive, it is determined as the frame start boundary; if the derivative changes from positive to negative, it is determined as the frame end boundary, thereby determining the specific time range of each signal frame (e.g., from frame start time t1 to frame end time t2). Extraction... Frame collision timing parameters, based on identified frame boundaries, can employ real-time signal level change detection technology. The preset normal intra-frame level fluctuation range is ±0.3V (set according to the standard level range of the RFID device's output signal). When a signal level change within a frame boundary exceeds ±0.3V within 1ms and the duration of this abnormal change exceeds 200μs, a multi-tag communication collision is determined for that frame. Simultaneously, frame collision timing parameters are extracted, including the collision occurrence time (time difference relative to the frame start boundary, e.g., 12ms from the frame start), the collision duration (duration of the abnormal change, e.g., 8ms), and the signal at the time of the collision. The maximum level transition amplitude (e.g., 1.2V) is used to calculate the collision probability. The extracted frame collision timing parameters are input into the improved ALOHA algorithm. This algorithm constructs a collision probability calculation model based on the Poisson distribution. The model is pre-set with a collision probability calculation formula obtained by fitting experimental data from 1000 sets of different collision scenarios (covering different numbers of tags and communication distances) (e.g., collision probability P = 0.02 × collision duration + 0.1 × maximum level transition amplitude, where the collision duration is in ms and the maximum level transition amplitude is in V). Substituting the parameters, the frame collision probability of the current multi-tag communication can be calculated (e.g., substituting the collision duration of 8ms and the maximum level transition amplitude). With a jump amplitude of 1.2V, we obtain P = 0.02 × 8 + 0.1 × 1.2 = 0.28, which is 28%. The frame length for predicting the label is then determined, with a preset reasonable range for the collision probability of 2%-5% (verified through testing to achieve a balance between recognition efficiency and collision rate). If the calculated collision probability is higher than 5% (e.g., 28%), the current frame length is increased by a fixed step of 5ms (e.g., if the current frame length is 50ms, it is adjusted to 55ms). If the collision probability is lower than 2%, the current frame length is decreased by a fixed step of 3ms. If the collision probability is within the reasonable range of 2%-5%, the current frame length is the final frame length for predicting the label, thus completing the frame length prediction.

[0098] In one embodiment, the initial adjustment parameters are generated by integrating the label-adjusted frame length with the denoising parameters of the adaptive filtering algorithm, including:

[0099] S61, extract the denoising parameters of the label adjustment frame length and adaptive filtering algorithm, and construct a two-parameter dataset;

[0100] S62, perform time-series alignment on the original two-parameter dataset to obtain an aligned two-parameter dataset;

[0101] S63 uses the interval mapping method to map the frame length adjustment value of each label in the aligned dual-parameter dataset to the effective range of the corresponding adaptive filtering algorithm denoising parameters;

[0102] S64, extract parameter values ​​that match the frame length temporal features within the effective interval to form initial adjustment parameters.

[0103] Specifically, a two-parameter dataset is constructed, and the predicted label adjustment frame length (e.g., time-series data within the range of 50ms-120ms, including the acquisition timestamp corresponding to each data point) is extracted. Simultaneously, the denoising parameters generated during the denoising process by the adaptive filtering algorithm are extracted (including filter coefficients, ranging from 0.05 to 0.3; iteration step size, ranging from 0.01 to 0.1; and filter window size, ranging from 5 to 20 sampling points, with each denoising parameter accompanied by the acquisition timestamp corresponding to the label adjustment frame length). The aforementioned label adjustment frame length and the three types of denoising parameters are then formatted according to the field "timestamp - label adjustment frame length - filter coefficient - iteration step size - filter window size". The data is organized into a structured data table to form a two-parameter dataset. Timing alignment is then performed, using the standard sampling timestamp (preset sampling interval is 100ms, e.g., 10:00:00.000, 10:00:00.100, 10:00:00.200…) when the RFID device collects the first RFID signal as a benchmark. The dataset is checked to ensure that the timestamps of each data point match the standard sampling timestamps perfectly. If a timestamp discrepancy exists (e.g., the timestamp of a tag's frame length adjustment data point is 10:00:00.050, falling between two standard sampling timestamps), linear interpolation can be used to calculate the denoising parameter value corresponding to that discrepancy time point (e.g., using 1…). Using the filter coefficients of 0.1 at 0:00:00.000 and 0.15 at 10:00:00.100 as a baseline, the filter coefficient at 10:00:00.050 is interpolated to 0.125. If there is missing data (e.g., no corresponding denoising parameter for a certain standard sampling timestamp), the data is also completed by linear interpolation of two adjacent valid data points to obtain an aligned dual-parameter dataset with perfectly aligned timestamps. Then, interval mapping is performed by calling the preset "Label Adjustment Frame Length - Valid Range of Denoising Parameters" mapping table in the system (this table is calibrated using 500 sets of experimental data with different frame lengths and denoising effects, for example, label adjustment frame length 50-80m). When the frame length is s, the effective range of the filter coefficient is 0.1-0.2, the effective range of the iteration step size is 0.03-0.06, and the effective range of the filter window size is 8-12 sampling points; when the frame length is 80-120ms, the effective range of the filter coefficient is 0.2-0.3, the effective range of the iteration step size is 0.06-0.1, and the effective range of the filter window size is 12-18 sampling points. The frame length value of each label in the aligned dual-parameter dataset is read, and the corresponding effective range of the three types of denoising parameters is matched in the mapping table according to its numerical range. For example, if a frame length value is 95ms (belonging to the 80-120ms range), then a filter coefficient of 0.2-0.3 and an iteration step size of 0.06-0.06 are matched.1. The effective range of 12-18 sampling points in the filtering window; extract the matching parameter values, first analyze the temporal characteristics of the frame length adjustment of the labels in the aligned dual-parameter dataset (e.g., if the frame length increases from 85ms to 95ms within 5 consecutive standard sampling timestamps, it is determined as "frame length increasing" feature; if the frame length fluctuates within 88-92ms, it is determined as "frame length stable" feature). For the effective range of the denoising parameters corresponding to each frame length value, select the appropriate parameter value in combination with the temporal characteristics—e.g., when "frame length increasing", within the effective range of the filtering coefficients. The filter window size is adjusted by selecting a value slightly above the upper limit (e.g., 0.27 for the 0.2-0.3 range) and an iteration step size slightly above the upper limit (e.g., 0.09 for the 0.06-0.1 range). When the frame length is stable, the middle value of the range is selected (e.g., 0.25 for the filter coefficient). The selected filter coefficient, iteration step size, and filter window size are then combined with the corresponding label to adjust the frame length, forming structured data containing "timestamp - label adjusted frame length - optimized filter coefficient - optimized iteration step size - optimized filter window size," which serves as the initial adjustment parameters.

[0104] The aforementioned product identification method based on RFID technology extracts the temporal characteristics of the first RFID signal to obtain the first RFID signal features. This process can initially separate invalid components in the signal caused by environmental interference, tag movement, or mutual interference among multiple tags, providing a signal basis for subsequent parameter adaptation. Based on the first RFID signal features, acquisition and control parameters are matched from a preset acquisition and control parameter library, overcoming the limitations of traditional fixed parameters and enabling the acquisition parameters to initially adapt to the real-time signal state, reducing signal distortion caused by parameters being out of sync with the scene. The second RFID signal is acquired through the RFID device with the acquisition parameters adjusted by the aforementioned control parameters. At this point, the device's acquisition parameters are aligned with the current signal characteristics, effectively reducing the impact of environmental interference, tag movement, and other factors on signal quality. The second RFID signal is evaluated using preset acquisition indicators to obtain evaluation results, quantifying key indicators such as signal integrity and stability, and identifying deficiencies in the current parameters. Based on the evaluation results, the acquisition and control parameters are optimized and the RFID device's acquisition parameters are updated, forming a closed loop of "signal analysis - parameter matching - signal acquisition - evaluation and optimization". By dynamically adjusting the acquisition parameters, the signal quality is continuously improved, effectively solving the problems of poor signal quality, low recognition accuracy, insufficient stability and weak environmental adaptability in traditional fixed parameter recognition, thereby improving the accuracy, stability and environmental adaptability of RFID product recognition.

[0105] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0106] Based on the same inventive concept, this application also provides an RFID-based product identification device for implementing the above-mentioned RFID-based product identification method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more RFID-based product identification device embodiments provided below can be found in the limitations of the RFID-based product identification method described above, and will not be repeated here.

[0107] In one exemplary embodiment, such as Figure 2 As shown, a product identification device based on RFID technology is provided, comprising:

[0108] The first signal feature extraction module 101 is used to extract the temporal features of the first RFID signal to obtain the first RFID signal features;

[0109] The control parameter matching module 102 is used to match the acquisition control parameters from the preset acquisition control parameter library according to the characteristics of the first RFID signal;

[0110] The second signal acquisition module 103 is used to acquire the second RFID signal, which is acquired by the RFID device after the acquisition parameters are adjusted by the acquisition control parameters.

[0111] The second signal evaluation module 104 is used to evaluate the second RFID signal using preset acquisition indicators and obtain evaluation results.

[0112] The control parameter optimization module 105 is used to optimize the collected control parameters based on the evaluation results to obtain optimized control parameters, which are then used to update the collection parameters of the RFID device.

[0113] In one embodiment, the first signal feature extraction module 101 is further configured to:

[0114] The overlapping signals in the first RFID signal are separated to obtain the separated signal;

[0115] Extract the intensity distribution of the separated signal;

[0116] Based on the intensity value distribution, the noise components are determined according to the preset noise model;

[0117] Based on the noise components, an adaptive filtering algorithm is used to denoise the separated signal to obtain a purified signal.

[0118] The timing characteristics of the purification signal are acquired and used as the first RFID signal characteristics.

[0119] In one embodiment, the control parameter matching module 102 is further configured to:

[0120] An improved ALOHA algorithm is used to analyze the features of the first RFID signal and predict the tag's adjusted frame length.

[0121] Obtain the denoising parameters of the adaptive filtering algorithm;

[0122] Integrate the tag adjustment frame length and denoising parameters to generate initial adjustment parameters;

[0123] Based on the initial adjustment parameters, feature matching is performed in the preset acquisition and control parameter library to determine the acquisition and control parameters, which include signal transmission power, signal sampling frequency, and signal frame interval.

[0124] In one embodiment, the preset acquisition indicators in the second signal evaluation module 104 include signal integrity indicators, signal stability indicators, and signal identification indicators.

[0125] The second RFID signal is evaluated using preset acquisition indicators to obtain evaluation results, including:

[0126] Based on signal integrity index, signal stability index, and signal identification index, a hierarchical identification index evaluation model is constructed. The identification index evaluation model includes an environmental interference loss sub-layer corresponding to the signal integrity index, a tag motion stability sub-layer corresponding to the signal stability index, and a tag mutual interference sub-layer corresponding to the signal identification index.

[0127] The second RFID signal is input into the hierarchical identification index evaluation model. The signal deviation coefficient is calculated through the environmental interference loss sub-layer, the signal adaptation deviation is calculated through the tag motion stability sub-layer, and the signal mutual interference index is calculated through the tag mutual interference sub-layer.

[0128] Based on preset scenario weights, the signal deviation coefficient, signal adaptation deviation, and signal interference index are weighted and calculated to obtain a comprehensive defect value.

[0129] The signal confidence level is determined by comparing the overall defect value with a preset confidence assessment threshold.

[0130] The signal confidence level, signal deviation coefficient, signal adaptation deviation, and signal interference index are integrated as the evaluation result.

[0131] In one embodiment, the control parameter optimization module 105 uses the following formula to optimize the collected control parameters based on the evaluation results, thereby obtaining optimized control parameters:

[0132]

[0133] in, For signal transmission power, The signal sampling frequency, For signal frame interval, To optimize signal transmission power, To optimize the signal sampling frequency, To optimize the signal frame interval, Let C be a 3x3 identity matrix, D be the adjustment scaling factor, D be the signal deviation factor, A be the signal adaptation deviation, and I be the signal cross-interference index. The scene weights are the signal deviation coefficients. Scenario weights for signal adaptation deviations The scene weights for the signal interference index, The confidence threshold for the signal deviation coefficient. The confidence threshold for signal adaptation deviation. The confidence threshold for the signal interference index is [value]. This is a matrix of positive coefficients for multiple maintenance functions.

[0134] In one embodiment, the control parameter matching module 102 is further configured to:

[0135] Frame boundaries for identifying the first RFID signal feature;

[0136] Based on frame boundaries, the frame conflict timing parameters of the first RFID signal feature are extracted using signal level transition detection technology.

[0137] An improved ALOHA algorithm is used to calculate the collision probability of frame collision timing parameters to obtain the collision probability.

[0138] The frame length is adjusted based on the predicted label determined by the collision probability.

[0139] In one embodiment, the control parameter matching module 102 is further configured to:

[0140] Extract the labels, adjust the frame length, and use the denoising parameters of the adaptive filtering algorithm to construct a two-parameter dataset;

[0141] The original two-parameter dataset is time-aligned to obtain an aligned two-parameter dataset;

[0142] The interval mapping method is used to map the frame length adjustment value of each label in the aligned two-parameter dataset to the effective range of the corresponding denoising parameters of the adaptive filtering algorithm;

[0143] Extract parameter values ​​that match the frame length temporal features within the effective interval to form initial adjustment parameters.

[0144] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the product identification method based on RFID technology as described above.

[0145] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0146] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0147] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A product identification method based on RFID technology, characterized in that, The method includes: Extract the temporal features of the first RFID signal to obtain the features of the first RFID signal; Based on the characteristics of the first RFID signal, the acquisition and control parameters are matched from the preset acquisition and control parameter library; Acquire a second RFID signal, which is obtained by an RFID device whose acquisition parameters have been adjusted by the acquisition control parameters; The second RFID signal is evaluated using preset acquisition indicators to obtain evaluation results; Based on the evaluation results, the acquisition control parameters are optimized to obtain optimized control parameters, which are used to update the acquisition parameters of the RFID device.

2. The method according to claim 1, characterized in that, The extraction of the temporal features of the first RFID signal to obtain the first RFID signal features includes: The overlapping signals in the first RFID signal are separated to obtain the separated signal; Extract the intensity value distribution of the separated signal; Based on the intensity value distribution, the noise components are determined according to a preset noise model; Based on the noise components, an adaptive filtering algorithm is used to denoise the separated signal to obtain a purified signal; The timing characteristics of the purification signal are obtained and used as the first RFID signal characteristics.

3. The method according to claim 2, characterized in that, The step of matching acquisition and control parameters from a preset acquisition and control parameter library based on the characteristics of the first RFID signal includes: An improved ALOHA algorithm is used to analyze the features of the first RFID signal and predict the tag's adjusted frame length. Obtain the denoising parameters of the adaptive filtering algorithm; Integrate the label adjustment frame length with the denoising parameters to generate initial adjustment parameters; Based on the initial adjustment parameters, feature matching is performed in a preset acquisition and control parameter library to determine the acquisition and control parameters, which include signal transmission power, signal sampling frequency, and signal frame interval.

4. The method according to claim 3, characterized in that, The preset acquisition indicators include signal integrity indicators, signal stability indicators, and signal identification indicators; The evaluation of the second RFID signal using preset acquisition indicators to obtain evaluation results includes: Based on the signal integrity index, the signal stability index, and the signal identification index, a hierarchical identification index evaluation model is constructed. The identification index evaluation model includes an environmental interference loss sub-layer corresponding to the signal integrity index, a tag motion stability sub-layer corresponding to the signal stability index, and a tag mutual interference sub-layer corresponding to the signal identification index. The second RFID signal is input into the hierarchical identification index evaluation model. The signal deviation coefficient is calculated through the environmental interference loss sub-layer, the signal adaptation deviation is calculated through the tag motion stability sub-layer, and the signal mutual interference index is calculated through the tag mutual interference sub-layer. Based on preset scenario weights, the signal deviation coefficient, the signal adaptation deviation, and the signal mutual interference index are weighted and calculated to obtain a comprehensive defect value. The signal confidence level is determined by comparing the comprehensive defect value with a preset confidence assessment threshold. The signal confidence level, the signal deviation coefficient, the signal adaptation deviation, and the signal interference index are integrated as the evaluation result.

5. The method according to claim 4, characterized in that, Based on the evaluation results, the acquisition and control parameters are optimized to obtain optimized control parameters, which are achieved using the following formula: in, For signal transmission power, The signal sampling frequency, For signal frame interval, To optimize signal transmission power, To optimize the signal sampling frequency, To optimize the signal frame interval, Let C be a 3x3 identity matrix, D be the adjustment scaling factor, D be the signal deviation factor, A be the signal adaptation deviation, and I be the signal cross-interference index. The scene weights are the signal deviation coefficients. Scenario weights for signal adaptation deviations The scene weights for the signal interference index, The confidence threshold for the signal deviation coefficient. The confidence threshold for signal adaptation deviation. The confidence threshold for the signal interference index is [value]. This is a matrix of positive coefficients for multiple maintenance functions.

6. The method according to claim 3, characterized in that, The step of analyzing the features of the first RFID signal using the improved ALOHA algorithm to predict the tag's adjusted frame length includes: Identify the frame boundary of the first RFID signal feature; Based on the frame boundary, the frame conflict timing parameters of the first RFID signal feature are extracted using signal level jump detection technology; The improved ALOHA algorithm is used to calculate the collision probability of the frame collision timing parameters to obtain the collision probability; The frame length of the predicted label is adjusted based on the collision probability.

7. The method according to claim 3, characterized in that, The process of integrating the tag-adjusted frame length with the denoising parameters of the adaptive filtering algorithm to generate initial adjustment parameters includes: Extract the label adjustment frame length and the denoising parameters of the adaptive filtering algorithm to construct a two-parameter dataset; The original two-parameter dataset is subjected to time-series alignment to obtain an aligned two-parameter dataset; The interval mapping method is used to map the frame length value of each label in the aligned dual-parameter dataset to the effective range of the corresponding adaptive filtering algorithm denoising parameters; Extract the parameter values ​​that match the frame length temporal features within the effective interval to form the initial adjustment parameters.

8. A product identification device based on RFID technology, characterized in that, The device includes: The first signal feature extraction module is used to extract the temporal features of the first RFID signal to obtain the first RFID signal features; The control parameter matching module is used to match the acquisition control parameters from the preset acquisition control parameter library according to the characteristics of the first RFID signal; The second signal acquisition module is used to acquire the second RFID signal, which is acquired by the RFID device after the acquisition parameters are adjusted by the acquisition control parameters. The second signal evaluation module is used to evaluate the second RFID signal using preset acquisition indicators and obtain evaluation results. The control parameter optimization module is used to optimize the acquisition control parameters based on the evaluation results to obtain optimized control parameters, which are used to update the acquisition parameters of the RFID device.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

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

Images