Ultrahigh frequency RFID system data transmission method, apparatus and device, and storage medium

Abstract

The invention relates to the technical field of data transmission, and discloses an ultrahigh frequency RFID system data transmission method, device and equipment and a storage medium, and the method comprises the steps: carrying out the band-pass filtering and orthogonal demodulation of a multi-tag backscattering signal received by an RFID reader, and obtaining a tag feature data sequence; performing multi-dimensional feature analysis and time sequence correlation calculation to obtain an independent label signal frame; boundary identification is carried out, and a label data frame starting point is determined; carrying out continuous sampling analysis and label signal conflict identification, and outputting a conflict position sequence; performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-conflict data packet; the label access scheduling queue is constructed based on the transmission time sequence and the priority attribute of the anti-collision data packet, the multi-label concurrent transmission time sequence is output, and the data transmission performance of the ultrahigh-frequency RFID system is improved through multi-stage signal processing and an intelligent scheduling mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of data transmission, and in particular to an ultra-high frequency RFID system data transmission method, device, equipment and storage medium. Background Art

[0002] Ultra-high frequency RFID technology has been widely used in logistics tracking, asset management, industrial production and other fields due to its advantages of non-contact, multi-target and long-distance identification. However, in actual application scenarios, when multiple tags respond to the reader's query at the same time, signal interference and collision problems between tags significantly reduce the recognition efficiency and reliability of the system. Especially in dense deployment environments, traditional anti-collision algorithms are difficult to meet the needs of efficient data transmission.

[0003] With the rapid development of Internet of Things technology, RFID systems are facing higher data throughput and lower transmission delay requirements. Existing data transmission methods mainly rely on simple random time slot allocation mechanisms, lack of in-depth analysis of tag signal characteristics and optimized scheduling of transmission resources, resulting in a sharp decline in system performance in multi-tag concurrent communication scenarios, and unable to guarantee the real-time and reliability of data transmission. In addition, traditional RFID data transmission schemes often adopt fixed frame structures and synchronization mechanisms, which are prone to frame synchronization failures and data loss when channel conditions are poor or tags are densely distributed. At the same time, the existing retransmission mechanism lacks effective error control and resource scheduling strategies, and exhibits large performance fluctuations when dealing with sudden conflicts, making it difficult to meet the needs of high-reliability application scenarios. Summary of the invention

[0004] The present invention provides an ultra-high frequency RFID system data transmission method, device, equipment and storage medium. The present invention improves the performance of ultra-high frequency RFID system data transmission through multi-level signal processing and intelligent scheduling mechanism.

[0005] In a first aspect, the present invention provides a method for transmitting data in an ultra-high frequency RFID system, the method comprising: Bandpass filtering and orthogonal demodulation are performed on the multi-tag backscattered signals received by the RFID reader to obtain the tag characteristic data sequence; Performing multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame; Perform boundary identification on the independent tag signal frame to determine the starting point of the tag data frame; Performing continuous sampling analysis and tag signal conflict identification on the signal after the starting point of the tag data frame, and outputting a conflict position sequence; Performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet; A tag access scheduling queue is constructed based on the transmission timing and priority attributes of the anti-collision data packet, and a multi-tag concurrent transmission timing is output.

[0006] In a second aspect, the present invention provides an ultra-high frequency RFID system data transmission device, the ultra-high frequency RFID system data transmission device comprising: A filtering module is used to perform bandpass filtering and orthogonal demodulation on the multi-tag backscattering signals received by the RFID reader to obtain a tag feature data sequence; A calculation module, used for performing multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame; An identification module, used to identify the boundary of the independent tag signal frame and determine the starting point of the tag data frame; An analysis module, used for performing continuous sampling analysis and tag signal conflict identification on the signal after the starting point of the tag data frame, and outputting a conflict position sequence; A rearrangement module, used for performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet; The output module is used to construct a tag access scheduling queue based on the transmission timing and priority attributes of the anti-collision data packet, and output the multi-tag concurrent transmission timing.

[0007] The third aspect of the present invention provides an ultra-high frequency RFID system data transmission device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the ultra-high frequency RFID system data transmission device executes the above-mentioned ultra-high frequency RFID system data transmission method.

[0008] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned UHF RFID system data transmission method.

[0009] In the technical solution provided by the present invention, 1. the signal feature analysis and multi-dimensional feature mapping technology separated in time and space are adopted, combined with multi-level feature extraction and correlation analysis, which effectively improves the recognition accuracy in a multi-tag environment; 2. the adaptive interleaving coding and multiple staggered transmission mechanisms are innovatively introduced to enhance the anti-interference ability and data transmission reliability of the system in a high-density tag environment; 3. a priority-based multi-level scheduling algorithm is designed, combined with a dynamic time slot allocation strategy, which reduces tag response conflicts and reduces transmission delays; 4. through hierarchical staggered control and performance adaptive optimization, the system resource utilization efficiency is optimized and the data transmission throughput is improved; 5. multiple anti-collision coding and intelligent retransmission control are adopted to enhance the transmission stability of the system in complex environments.

[0010] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0011] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of an embodiment of a method for transmitting data in an ultra-high frequency RFID system according to an embodiment of the present invention; Figure 2 A schematic diagram of an embodiment of a UHF RFID system data transmission device in an embodiment of the present invention; Figure 3 It is a schematic diagram of an embodiment of an ultra-high frequency RFID system data transmission device in an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0015] To facilitate understanding of this embodiment, a method for transmitting data in an ultra-high frequency RFID system disclosed in an embodiment of the present invention is first introduced in detail. Figure 1 As shown, the method comprises the following steps: 101. Perform bandpass filtering and orthogonal demodulation on the multi-tag backscatter signals received by the RFID reader to obtain a tag feature data sequence; It is understandable that the execution subject of the present invention may be an ultra-high frequency RFID system data transmission device, or a terminal or a server, which is not specifically limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0016] Specifically, the backscattered signals emitted by multiple tags are received by the RFID reader, and then these signals are filtered through a bandpass filter with a center frequency of 915MHz to obtain a primary filtered signal. The bandpass filter can filter out effective signals in a specific frequency band from the received broadband signal, effectively remove environmental noise and irrelevant interference signals, retain the useful components of the backscattered signal, and obtain a primary filtered signal. The primary filtered signal is sampled at four times the carrier frequency, and the signal is frequency converted by a digital down-conversion circuit to obtain an intermediate frequency signal. Four times the carrier frequency sampling helps to improve the resolution of the signal and better retain the details of the signal. The digital down-conversion circuit converts the signal into an intermediate frequency range by adjusting the frequency of the signal, making the signal easier to process later. The intermediate frequency signal is input into the orthogonal demodulator and multiplied with the in-phase branch and the orthogonal branch of the local carrier respectively to obtain the in-phase component I and the orthogonal component Q. The in-phase component I and the orthogonal component Q represent the projection of the signal in the in-phase and orthogonal directions respectively. The orthogonal demodulation process enables the system to effectively separate and process the different frequency components of the signal and extract more information. The in-phase component I and the orthogonal component Q are band-limited by a low-pass filter to obtain baseband I-channel signal and baseband Q-channel signal. The function of the low-pass filter is to remove the high-frequency component in the intermediate frequency signal, retain the baseband part, and obtain the baseband I-channel and Q-channel signals, which contain low-frequency features directly related to the tag information. The baseband I-channel signal and the baseband Q-channel signal are digitally sampled by a 16-bit analog-to-digital converter to obtain a digitized I-channel data stream and a digitized Q-channel data stream. The digitized I-channel data stream and the digitized Q-channel data stream are timed and recovered, and a sampling clock sequence is established through the timing recovery process to avoid signal distortion or information loss caused by the offset of the sampling time point. According to the sampling clock sequence, the digitized I-channel data stream and the digitized Q-channel data stream are divided according to the 16-bit RN16 length to obtain an I-channel data block and a Q-channel data block. The RN16 length is 16 bits, which is the basic length unit of the tag response frame. The I-channel data block and the Q-channel data block are combined into a complex tag feature data sequence, while retaining the amplitude and phase information of the signal.

[0017] 102. Perform multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame; Specifically, complex operations are performed on the tag feature data sequence, and the tag signal strength sequence and phase sequence are obtained by calculating the square of the amplitude and the phase angle of each data point. The tag signal strength sequence and phase sequence represent the amplitude and phase characteristics of each tag signal on the complex plane, which are important information for identifying different tags. By sorting the tag signal strength sequence according to the signal energy, a valid signal point set is obtained, which represents the most energetic part of the received tag signal, and can effectively remove interference signals and noise. The arrival time information is extracted for each data point in the valid signal point set to capture the specific time characteristics of the tag response. In order to improve the accuracy of the time information, a weighted average method based on signal strength is used to weight the arrival time of each valid signal point to obtain a tag response time sequence. The weighted average method makes the data points with higher signal strength have a greater impact on the response time, effectively reduces the interference of noise on the response time calculation, and improves the accuracy of the tag response time sequence. The tag response time sequence is input into the feature classifier, and the time intervals are grouped based on the K-means clustering algorithm to obtain a tag response period set. The K-means clustering algorithm groups the response moments and classifies similar response moments into one category, effectively identifying the response cycle of each tag and obtaining a cycle set containing the response moments of each tag. Based on the tag response cycle set, an A×B-dimensional feature matrix is ​​constructed, where A represents the number of tags and B represents the feature dimension, which contains the amplitude, phase and time information of each tag, and obtains the feature space mapping matrix. The Pearson correlation coefficient with the adjacent data frame is calculated for each row of the feature space mapping matrix. The Pearson correlation coefficient is used to measure the similarity between two data frames. The correlation between each tag signal frame is quantified to determine whether it belongs to the same tag response. In the process of calculating the Pearson correlation coefficient, the correlation threshold is set to 0.8, and only data frames with a correlation coefficient higher than 0.8 are considered to have a high correlation. This process can effectively eliminate noise and irrelevant signals, retain highly correlated data frames for subsequent processing, and obtain a tag correlation matrix. Based on the tag correlation matrix, connected component analysis is performed, and data frames with correlation coefficients higher than the set threshold are combined to obtain an independent tag candidate set. Connected component analysis is an algorithm in graph theory that can merge highly correlated data frames into a connected component to identify and distinguish signals with different labels. By combining highly correlated data frames, a set of multiple candidate label signals is obtained, namely the independent label candidate set. The independent label candidate set is subjected to time consistency verification to check whether each data frame meets the fixed response period, and eliminate the data frames that do not meet the fixed period characteristics to ensure that the output signal frame is an independent label signal frame with good time consistency.

[0018] 103. Perform boundary identification on the independent tag signal frame to determine the starting point of the tag data frame; Specifically, a 16-bit cyclic shift synchronization codeword is constructed for the independent tag signal frame to generate a standard synchronization codeword with a maximum and minimum Hamming distance to improve the anti-interference ability of the synchronization process. The maximum and minimum Hamming distances are selected to ensure that there is sufficient discrimination between the synchronization codewords, so that the starting boundary of the tag can still be effectively identified in a noisy environment. The standard synchronization codeword is subjected to a full-cycle cyclic shift operation to generate a set of 16 shifted versions of the synchronization codeword. The flexibility of matching is increased by generating multiple shifted versions, so that no matter what phase the signal appears in, the corresponding matching version can be found in the synchronization codeword set. The independent tag signal frame is segmented according to the length of the synchronization codeword, and each signal sequence is normalized to obtain the data sequence to be matched. The normalization process can eliminate the influence caused by the intensity difference between different signal segments, so that the correlation operation result only depends on the relative shape of the signal without being interfered by the absolute amplitude. The data sequence to be matched is correlated with each shifted version in the synchronization codeword set, and the correlation coefficient matrix is ​​obtained by calculating the correlation value sequence of 256 points. Perform peak detection on each column of the correlation coefficient matrix, extract the location information and amplitude information of the local maximum point, and obtain a set of candidate synchronization points. According to the amplitude distribution of the candidate synchronization point set, calculate the optimal decision threshold and obtain a valid synchronization position sequence. Through the analysis of the amplitude distribution, a decision threshold is adaptively set to exclude pseudo synchronization points caused by noise, and only synchronization points with sufficiently high amplitude and high credibility are retained. Then, a majority vote is performed on the synchronization points of three adjacent data frames in the valid synchronization position sequence to enhance the stability and robustness of the synchronization points. The most likely synchronization position is selected from multiple candidate positions, that is, the position with the highest frequency of occurrence is selected as the final frame synchronization position. The data frame is segmented and reorganized based on the frame synchronization position, and the independent label signal frame is segmented into data units of fixed length according to the synchronization position, and the starting point of the label data frame is output.

[0019] 104. Perform continuous sampling analysis and tag signal conflict identification on the signal after the starting point of the tag data frame, and output a conflict position sequence; Specifically, the signal after the starting point of the tag data frame is segmented into sliding windows, the length of the window is set to the length of RN16, and the sliding step is set to one code element to obtain a continuous signal window sequence. The sliding window segmentation method helps to gradually extract the features in the signal and capture small changes in the signal sequence to achieve more accurate conflict detection. Complex envelope detection is performed on each window data in the continuous signal window sequence, and the amplitude trajectory and phase trajectory of the signal are extracted from it to generate a trajectory sequence containing signal features. The purpose of complex envelope detection is to separate the amplitude and phase characteristics of the signal, which helps to analyze the change pattern of the signal and identify conflicts. The signal feature trajectory sequence is subjected to differential operation, and the amplitude change rate and phase jump amount between adjacent sampling points are calculated to obtain the signal change feature sequence. The differential operation reveals the change characteristics of the signal on the time axis. The amplitude change rate reflects the sudden change of the signal strength, while the phase jump amount reveals the discontinuous change of the signal phase. The signal change feature sequence is input into the dual threshold detector, and the threshold of the amplitude change rate is set to 20%, and the threshold of the phase jump is set to 45 degrees. By comparing with these thresholds, the primary collision judgment result is obtained. The signal vector diagram analysis is performed on the primary collision judgment results, and the existence of the conflict area is judged by counting the number of cluster centers in the signal space. In the signal vector space, if the number of cluster centers in a certain area is greater than 1, it means that there are multiple groups of signal features in the area, which is caused by the overlap of signals of multiple labels. These areas are marked as potential collision points, and a collision point candidate set is obtained. Based on the collision point candidate set, a collision feature vector containing multiple feature quantities is constructed, including three feature quantities: amplitude mutation, phase jump, and cluster discreteness. These feature quantities are used to describe the change characteristics of the signal in the collision area and generate a collision feature sequence. The amplitude mutation reflects the drastic change of the signal intensity at the collision point, the phase jump describes the discontinuous change of the phase, and the cluster discreteness is used to evaluate the concentration of the signal cluster. These features comprehensively describe the behavior of the signal in the collision area. The collision feature sequence is input into the support vector machine classifier to identify the collision event and obtain the collision event label sequence. As a machine learning classification algorithm, the support vector machine can use the input feature vector to determine whether the signal segment belongs to a collision event, thereby classifying each signal segment and outputting a collision label sequence. Use machine learning algorithms to accurately classify conflict events, avoid misjudgments or missed judgments caused by simple threshold judgments, and improve the accuracy of overall conflict identification. Perform time domain correlation analysis on the collision event marker sequence, merge adjacent collision event markers, and generate the final conflict position sequence.

[0020] 105. Perform matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet; Specifically, the data marked by the conflict position sequence is divided according to a fixed length of 8 bits to construct a data unit sequence and obtain a set of data units to be retransmitted. The set of data units to be retransmitted is filled into an M×N interleaving matrix in row priority order, where M represents the length of the data unit and N represents the collision frequency value currently detected. The data is filled into each row of the matrix row by row priority order to obtain an initial interleaving matrix. The construction of the interleaving matrix aims to improve the anti-collision capability by rearranging the data. In particular, when there is a conflict of multiple labels, this data scattering process effectively reduces the probability of repeated conflicts in certain data units and ensures the stability and reliability of data transmission in the system. The initial interleaving matrix is ​​subjected to a column cyclic shift operation, and the shift amount is linearly related to the column number to generate an interleaved data matrix with position scattering, and obtain a rearranged data matrix. The column cyclic shift operation is to cyclically shift the data in each column according to a specific shift amount, disrupt the position of the data, and enable the data units originally at the same position to be scattered throughout the matrix. The position scattering strategy reduces the possibility of conflict by increasing the randomness of the data and improves the anti-interference capability of the transmitted data. The rearranged data matrix is ​​read out in column priority order, and the read data is grouped, each data group is 4 bits long, and the data group to be encoded is obtained. The Hamming coding operation is performed on the data group to be encoded, and a 3-bit check bit is generated for each data group, and the check bit is added to the original data to obtain the encoded data sequence. The introduction of Hamming coding is to provide error correction capability. By adding redundant check bits, errors that occur during data transmission can be detected and corrected to improve the reliability of data transmission. The header of the data packet is constructed based on the encoded data sequence. The data packet header contains interleaving depth information, the length of the data packet, and the original position identifier, which provides necessary auxiliary information for the decoding and recovery of subsequent data. The interleaving depth information is used to indicate the degree of data interleaving, the data packet length helps the receiving end determine the integrity of the data, and the original position identifier provides a basis for the restoration of the data during decoding. The data packet header and the encoded data sequence are combined together, encapsulated into a complete data packet according to a fixed format, and a CRC (cyclic redundancy check) check code is added to the data packet to generate a 32-bit cyclic redundancy check code. The role of the CRC check is to verify the content of the entire data packet to detect whether an error occurred during the transmission process. CRC is an error detection mechanism used in communication systems. It generates a checksum by performing specific mathematical operations on the data and attaches it to the data packet. At the receiving end, the received data is checked using the same operation method to determine whether there is a transmission error and take appropriate error correction measures. Through the above steps, an anti-collision data packet is finally obtained.

[0021] 106. Based on the transmission timing and priority attributes of the anti-collision data packet, a tag access scheduling queue is constructed to output the multi-tag concurrent transmission timing.

[0022] Specifically, the first-level priority calculation is performed on the anti-collision data packet, in which the label data volume is assigned a weight coefficient of 0.4, the number of retransmissions is assigned a weight coefficient of 0.3, and the waiting time is assigned a weight coefficient of 0.3. The first priority sequence of each label is obtained by calculating the weighted sum function. The second-level priority calculation is performed on the first priority sequence, and exponential weighting is performed based on the transmission delay requirement to obtain a new label priority sequence. The introduction of exponential weighting makes the priority of delay-sensitive labels amplified, and these time-sensitive label data are transmitted preferentially to ensure the response speed and overall efficiency of the system. The label priority sequence is input into the dual threshold selector, the first threshold is set to 80% of the system processing capacity, and the second threshold is set to the mean of the priority. Only labels that meet both threshold conditions are selected into the first transmission queue; labels that meet only one of the threshold conditions are selected into the second transmission queue, and a hierarchical set of labels to be scheduled is obtained. Through the dual threshold selection method, the labels are screened more carefully, the system resources are allocated reasonably, the system overload is avoided, and the overall transmission efficiency and fairness are improved. For the hierarchical tag set to be scheduled, a two-level clustering algorithm is performed on it to roughly group the tags, and then each rough group is finely divided. After two levels of clustering, the time slot protection factor β=0.1 is set between groups to ensure that the probability of conflict between different groups is minimized, and a multi-level transmission group structure is obtained. The multi-level grouping method can effectively manage the transmission conflicts between a large number of tags, and achieve the optimal scheduling of system resources by reasonably planning the time slot allocation within and between tag groups. Based on the multi-level transmission group structure, a time slot resource allocation tree is constructed. The first-level nodes of the time slot resource allocation tree are used to allocate basic time slot resources, the second-level nodes are responsible for allocating staggered intervals for transmission within the group, and the third-level nodes specifically allocate the transmission time of each tag, thereby forming a hierarchical time slot allocation scheme. The hierarchical time slot allocation scheme is optimized by triple peak shifting. The peak shifting between groups is optimized. By setting reasonable peak shifting between groups, the conflict caused by simultaneous transmission of different groups is reduced. Then the peak shifting within the group is optimized to ensure that the transmission between the tags within the group is avoided as much as possible. Finally, the peak shifting between nodes is optimized to ensure that the transmission of each specific tag is carried out at the most appropriate time. Through the optimization of triple peak shifting, a multi-dimensional peak shifting control sequence that minimizes the probability of conflict is generated. According to the multi-dimensional peak shifting control sequence, a three-level scheduling queue is established: the first level is the emergency transmission queue, which is used to transmit tag data with high delay requirements; the second level is the standard transmission queue, which is used for general priority tags; and the third level is the low priority transmission queue, which is used to transmit data with no strict delay requirements. At the same time, each queue is assigned a differentiated scheduling period to obtain a hierarchical scheduling control sequence.Based on the hierarchical scheduling control sequence, a performance monitoring matrix is ​​set, which includes three main monitoring points: the first monitoring point is set at the queue switching position to monitor whether the switching between queues of different priorities is smooth; the second monitoring point is set at the boundary of the time slot to ensure that the use of each time slot meets expectations and there is no conflict; the third monitoring point is set at the position where the transmission is completed to evaluate the success rate of data transmission and the overall delay. Through the joint analysis of the three monitoring points, multi-dimensional performance evaluation results are obtained. According to the results of the multi-dimensional performance evaluation, a three-level dynamic adjustment mechanism is triggered: the first level adjusts the division of the transmission group to adapt to the changing system load by re-dividing the label group; the second level adjusts the allocation of time slots to optimize the use of time slots according to the feedback of transmission performance; the third level adjusts the peak shifting parameters to reduce the occurrence of conflicts by adjusting the peak shifting strategy. All adjustments use the exponential backoff algorithm to control the adjustment cycle. In this way, flexible scheduling optimization is dynamically performed according to the actual operation of the system, and a multi-label concurrent transmission sequence is output to achieve efficient and stable data transmission.

[0023] In the embodiments of the present invention, 1. the time-space separated signal feature analysis and multi-dimensional feature mapping technology are adopted, combined with multi-level feature extraction and correlation analysis, to effectively improve the recognition accuracy in a multi-tag environment; 2. the innovative introduction of adaptive interleaving coding and multiple staggered transmission mechanisms enhances the system's anti-interference capability and data transmission reliability in a high-density tag environment; 3. a priority-based multi-level scheduling algorithm is designed, combined with a dynamic time slot allocation strategy, to reduce tag response conflicts and reduce transmission delays; 4. through hierarchical staggered control and performance adaptive optimization, the system resource utilization efficiency is optimized and the data transmission throughput is improved; 5. multiple anti-collision coding and intelligent retransmission control are adopted to enhance the transmission stability of the system in complex environments.

[0024] In a specific embodiment, the process of executing step 101 may specifically include the following steps: The multi-tag backscattered signals are received by an RFID reader, and the multi-tag backscattered signals are filtered by a bandpass filter with a center frequency of 915 MHz to obtain a primary filtered signal; The primary filter signal is sampled at four times the carrier frequency, and the frequency is converted through a digital down-conversion circuit to obtain an intermediate frequency signal, which is input into an orthogonal demodulator and multiplied with the in-phase branch and the orthogonal branch of the local carrier to obtain an in-phase component I and an orthogonal component Q; The in-phase component I and the quadrature component Q are subjected to band-limiting processing by a low-pass filter to obtain a baseband I-channel signal and a baseband Q-channel signal, and the baseband I-channel signal and the baseband Q-channel signal are subjected to digital sampling by a 16-bit analog-to-digital converter to obtain a digital I-channel data stream and a digital Q-channel data stream; Performing timing recovery on the digitized I-channel data stream and the digitized Q-channel data stream, establishing a sampling clock sequence, and dividing the digitized I-channel data stream and the digitized Q-channel data stream according to the 16-bit RN16 length according to the sampling clock sequence to obtain an I-channel data block and a Q-channel data block; The I-way data blocks and the Q-way data blocks are combined into a label feature data sequence in a plural form.

[0025] Specifically, when the RFID reader receives multi-tag signals, it will receive various clutter and interference signals. In order to effectively extract useful backscatter information, the center frequency of the bandpass filter is set to 915MHz, which is the operating frequency of the ultra-high frequency RFID system. The received original signal is processed by the bandpass filter to filter out noise and interference that are not in the working frequency band, and only the effective frequency band signal containing the tag information is retained. The primary filter signal is sampled at four times the carrier frequency. Assume that the carrier frequency is , its value is 915MHz, and the four times sampling frequency is . High-frequency sampling helps to retain the detailed characteristics of the signal and better capture the information in the tag backscattered signal. The sampling process is described by the Nyquist theorem, which requires that the sampling frequency be at least twice the highest frequency of the signal in order to fully restore the original signal. Therefore, four times the carrier frequency is used for sampling to ensure that no information is lost. The sampled signal is input into the digital down-conversion circuit, and the intermediate frequency signal is obtained through frequency conversion. Suppose the original received signal is , its expression is expressed as: ; in, represents the amplitude change of the signal, represents the phase change of the signal, The digital down-conversion process is to mix the signal with the local oscillator signal to reduce the frequency and convert the signal to an intermediate frequency range. Assume that the local oscillator signal is , the intermediate frequency signal obtained after down-conversion is: ; in, The intermediate frequency signal is input into the orthogonal demodulator and multiplied with the in-phase branch and the orthogonal branch of the local carrier respectively to obtain the in-phase component. and the orthogonal components The function of the orthogonal demodulator is to decompose the intermediate frequency signal into two independent components. These two components describe the amplitude and phase changes of the signal on the complex plane. The demodulated signal of the in-phase branch is , the demodulated signal of the orthogonal branch is After quadrature demodulation, the in-phase component and the quadrature component are obtained: ; Through these two components and , describing the amplitude and phase characteristics of the signal. The in-phase component is filtered by a low-pass filter. and the orthogonal components Band-limiting processing is performed to obtain the baseband Road signals and The purpose of low-pass filtering is to remove the high-frequency components in the signal and retain the baseband part related to the tag information. The signals after low-pass filtering are recorded as and , contains the main information of the signal. Use a 16-bit analog-to-digital converter to convert the baseband Road signals and The signal is digitally sampled to obtain a digital Data flow and Suppose the sampling frequency is , the sampled digital signals are: ; ; in, is the sampling period, is the serial number of the sampling point. Through 16-bit analog-to-digital conversion, high quantization accuracy is ensured, signal details are retained, and the accuracy of signal processing is improved. Data flow and The timing of the data stream is recovered to establish a sampling clock sequence. The appropriate sampling time is extracted from the received signal to ensure that the sampling point of the data stream can be aligned with the modulation characteristics of the original signal to minimize the sampling error. Suppose the recovered sampling clock is , the corresponding digital data is expressed as: ; The sampling points obtained by timing recovery ensure stable data sampling, thereby accurately extracting the characteristics of the signal. Data flow and The data stream is divided into 16-bit RN16 lengths to obtain Road data block and RN16 is a basic response length unit in the RFID system, representing 16 bits of data. The data is divided according to the length of RN16 to ensure that each data block contains a complete tag response information. Road data block and The data blocks are combined into a complex tag feature data sequence. Each data block is represented as a complex number, where is the real part, is the imaginary part: ; in, is an imaginary unit, representing the imaginary part of a complex number. This complex number representation method retains both the amplitude and phase information of the signal, making it easier to perform multidimensional analysis of label features. For example, the modulus and phase of a complex number are represented as: ; ; in, represents the amplitude of the signal, Indicates the phase angle of the signal. This complex number tag feature data sequence provides complete amplitude and phase information for subsequent tag decoding and feature analysis.

[0026] In a specific embodiment, the process of executing step 102 may specifically include the following steps: Perform complex operations on the tag feature data sequence, calculate the square of the amplitude and the phase angle of each data point, obtain the tag signal strength sequence and phase sequence, and sort the tag signal strength sequence according to the signal energy to obtain the effective signal point set; The arrival time information is extracted for each data point in the valid signal point set, and the time information is weighted averaged according to the signal strength to obtain the tag response time sequence; The tag response time series is input into the feature classifier, and the time intervals are grouped based on the K-means clustering algorithm to obtain the tag response period set; According to the tag response cycle set, an A×B dimensional feature matrix is ​​constructed, where A is the number of tags and B is the feature dimension, including amplitude, phase and time information, to obtain the feature space mapping matrix; For each row of the feature space mapping matrix, the Pearson correlation coefficient with the adjacent data frame is calculated, and the correlation threshold is set to 0.8 to obtain the label correlation matrix; Connected component analysis is performed based on the label correlation matrix, and data frames with correlation coefficients higher than the threshold are combined to obtain an independent label candidate set. Temporal consistency verification is performed on the independent label candidate set, and data frames that do not meet the fixed response period are eliminated, and independent label signal frames are output.

[0027] Specifically, the square of the amplitude and the phase angle of each data point are calculated to obtain the intensity sequence and phase sequence of the tag signal. Assume that the complex number representation of each data point is ,in and are the in-phase and quadrature components respectively, is an imaginary unit. The square of the amplitude (i.e. the strength of the signal) is calculated by the following formula: ; in, Represents the square of the signal amplitude, that is, the strength of the signal. and are the real and imaginary parts of the data point respectively. By calculating the square of the amplitude of each data point, the tag signal strength sequence is obtained. The phase angle is calculated by the following formula: ; in, Represents the phase angle of the data point. Through these two formulas, we can get the signal strength and phase information respectively. Sort the tag signal strength sequence according to the signal energy to get the valid signal point set. Suppose the signal strength sequence is By sorting these signal intensities from large to small, a new sequence is obtained ,in ( ). By sorting, the signals with lower energy are filtered out, and only the valid signal points with higher signal energy are retained to form a valid signal point set. For each data point in the valid signal point set, its arrival time information is extracted. Let the first data point in the valid signal point set be The arrival time of the data point is , and its corresponding signal strength is In order to improve the accuracy of the signal arrival time, the signal strength is used to perform a weighted average on the arrival time. The calculation formula is: ; in, is the weighted average arrival time, is the number of valid signal points, For the The strength of the signal point, is the corresponding arrival time. Through weighted averaging, a more stable and accurate tag response time sequence is obtained, reducing the impact of noise and unstable signals on the response time calculation. The tag response time sequence is input into the feature classifier, and the time intervals are grouped based on the K-means clustering algorithm to obtain the tag response period set. The tag response time is divided into several groups, each group represents a tag characteristic response period. Suppose the tag response time sequence is , dividing these response times into clusters, the objective function of clustering is expressed as: ; in, is the objective function, For the Clusters, For the By minimizing the objective function , dividing the response time into clusters, each cluster represents the response period of a tag. According to the tag response period set, construct a dimensional feature matrix, where is the number of labels, is the dimension of the feature, including amplitude, phase and time information. Let the feature matrix be , then Row, No. The elements of a column are represented as: ; in, Indicates of tags eigenvalues, including amplitude , Phase and time information In this way, the constructed feature space mapping matrix can fully describe the multi-dimensional features of each label. The Pearson correlation coefficient with the adjacent data frame is calculated for each row of the feature space mapping matrix. The Pearson correlation coefficient is used to measure the correlation between two vectors. Let the first row of the feature space mapping matrix be Behavior , No. Behavior , then the Pearson correlation coefficient is expressed as: ; in, For the Row and Pearson correlation coefficient between rows, and Respectively Row and By calculating the Pearson correlation coefficient, we get the label correlation matrix , the elements of this matrix Indicates Tags and Based on the label correlation matrix, a connected component analysis is performed to combine data frames with correlation coefficients higher than a threshold (set to 0.8) to obtain an independent label candidate set. , then it is considered that Tags and The labels belong to the same connected component and are merged into an independent label candidate set. The independent label candidate set is verified for temporal consistency and the data frames that do not meet the fixed response period are eliminated to obtain the final independent label signal frame.

[0028] In a specific embodiment, the process of executing step 103 may specifically include the following steps: Construct a 16-bit cyclic shift synchronization codeword for the independent tag signal frame to generate a standard synchronization codeword with the maximum and minimum Hamming distance, and perform a full-cycle cyclic shift operation on the standard synchronization codeword to generate a set of 16 shifted versions of the synchronization codeword; The independent tag signal frame is segmented according to the synchronization codeword length, and each segment of the signal sequence is normalized to obtain a data sequence to be matched, and the data sequence to be matched is correlated with each shifted version in the synchronization codeword set, and a 256-point correlation value sequence is calculated to obtain a correlation coefficient matrix; Perform peak detection on each column of the correlation coefficient matrix, extract the position information and amplitude information of the local maximum point, obtain a set of candidate synchronization points, and calculate the optimal decision threshold according to the amplitude distribution of the candidate synchronization point set to obtain a valid synchronization position sequence; A majority vote is performed on the synchronization points of three adjacent data frames in the valid synchronization position sequence, and the position with the highest frequency of occurrence is selected to obtain the frame synchronization position. The data frame is segmented and reorganized based on the frame synchronization position, and the independent label signal frame is segmented into data units of fixed length according to the synchronization position, and the starting point of the label data frame is output.

[0029] Specifically, a 16-bit cyclic shift synchronization codeword is constructed for the independent tag signal frame to generate a standard synchronization codeword with the maximum and minimum Hamming distance. The Hamming distance is an indicator to measure the degree of difference between two binary sequences, which is defined as the number of different bits in two binary sequences of the same length. In order to ensure the robustness of the synchronization codeword in different signal scenarios, the maximum and minimum Hamming distance is selected to construct the synchronization codeword, which effectively improves the reliability of frame synchronization in a noisy environment. Assume that the synchronization codeword is ,in Represents each bit of the codeword. Through the optimization algorithm, a binary sequence with a length of 16 is found to maximize the minimum Hamming distance between it and other possible codewords, thereby improving the signal's anti-interference ability. Perform a full-cycle cyclic shift operation to generate a set of 16 shifted versions of the synchronization codeword. , shift all its bits to the right by one bit each time, and move the last bit to the first bit, repeat this operation 16 times in total, and get 16 shifted versions. The codeword after the shift is , get a synchronization codeword set , which is used for subsequent matching operations. The independent tag signal frame is segmented according to the length of the synchronization codeword, and the signal frame is divided into multiple segments with a length of 16 bits. Assume that the independent tag signal frame is , it is divided into several segments of length 16, each segment signal is represented as ,in is the sequence number of the segment. For each signal sequence Normalization is performed to eliminate the influence of amplitude and make the amplitude of the signal comparable at the same scale. The normalized signal is expressed as: ; in, Signal sequence The Euclidean norm of is defined as: ; Normalized signal sequence The amplitude of is adjusted to 1 to facilitate correlation with the standard synchronization codeword. The normalized data sequence to be matched is correlated with each shifted version in the synchronization codeword set, and a 256-point correlation value sequence is calculated to form a correlation coefficient matrix. Suppose the normalized data sequence to be matched is , the first in the synchronization codeword set The shifted version is , then the result of the relevant operation is: ; in, Indicates The signal sequence of the first The correlation between the synchronization codeword versions, and Respectively represent the signal to be matched and the synchronization codeword By calculating the synchronization codewords for all segments and all shifted versions, a correlation coefficient matrix is ​​obtained , where each element Indicates Segment signal and The correlation between the synchronization codewords. Perform peak detection on each column of the correlation coefficient matrix, extract the location information and amplitude information of the local maximum point, and obtain a set of candidate synchronization points. Peak detection finds the signal position with the maximum correlation with the synchronization codeword, which is most likely to be the starting point of the tag data frame. The sequence of related values ​​for the columns is By detecting the local maximum among these values, the position and amplitude information of the candidate synchronization point are extracted to obtain the candidate synchronization point set ,in Indicates The location of candidate synchronization points, Represents the corresponding correlation amplitude. According to the amplitude distribution of the candidate synchronization point set, the optimal decision threshold is calculated to obtain a valid synchronization position sequence. Suppose the amplitude of the candidate synchronization point set is , by analyzing the distribution of these amplitudes, a decision threshold is set , only values ​​above The synchronization point is considered to be a valid synchronization point, and a valid synchronization position sequence is obtained. . Perform majority voting on the synchronization points of three adjacent data frames in the valid synchronization position sequence to select the position with the highest frequency as the final frame synchronization position. Split and reorganize the data frame based on the frame synchronization position, split the independent label signal frame into fixed-length data units according to the synchronization position, and finally output the starting point of the label data frame.

[0030] In a specific embodiment, the process of executing step 104 may specifically include the following steps: The signal after the starting point of the tag data frame is segmented by a sliding window, the window length is set to RN16 length, and the sliding step length is 1 code element to obtain a continuous signal window sequence; Perform complex envelope detection on each window data in the continuous signal window sequence, extract the signal amplitude trajectory and phase trajectory, obtain the signal feature trajectory sequence, perform differential operation on the signal feature trajectory sequence, calculate the amplitude change rate and phase jump amount between adjacent sampling points, and obtain the signal change feature sequence; The signal change feature sequence is input into the dual threshold detector, the amplitude change rate threshold is set to 20%, and the phase jump threshold is set to 45 degrees, and the primary collision judgment result is obtained; Perform signal vector diagram analysis on the primary collision judgment result, count the number of cluster centers in the signal space, mark the area where the number of cluster centers is greater than 1 as a potential collision point, and obtain a candidate set of collision points; Based on the collision point candidate set, a collision feature vector is constructed, which includes three feature quantities: amplitude mutation degree, phase jump degree and clustering discreteness, and a collision feature sequence is obtained; The collision feature sequence is input into the support vector machine classifier for collision event recognition to obtain a collision event marker sequence. The collision event marker sequence is then subjected to time domain correlation analysis, and adjacent collision event markers are merged to generate a conflict position sequence.

[0031] Specifically, the signal after the starting point of the tag data frame is segmented into sliding windows, the window length is set to RN16 length, the sliding step is one code element, and a continuous signal window sequence is obtained. Assuming that the RN16 length is 16 code elements, the input signal sequence is , by sliding the window with a sliding step of 1 symbol, we get a signal window of length 16 ,in is the starting position of the sliding window, ranging from In this way, a continuous signal window sequence containing the complete characteristics of the signal is obtained. Complex envelope detection is performed on each window in the continuous signal window sequence to map the signal from the real domain to the complex domain. Suppose the data of each signal window is , through the complex envelope operation, each signal point is expressed as a complex form: ; in, Indicates In the window The complex representation of the signal points is and are the in-phase and quadrature components of the signal, respectively. is an imaginary unit. By calculating the complex envelope, we can obtain the amplitude trajectory and phase trajectory of the signal, where the amplitude and counter They are expressed by the following formulas respectively: ; ; By performing complex envelope detection on each window, the amplitude trajectory sequence of the signal is obtained. and phase trajectory sequence These features describe the change of the signal in the time domain. Perform differential operations on the signal feature trajectory sequence to calculate the amplitude change rate and phase jump between adjacent sampling points. Suppose the signal amplitude trajectory is And the phase trajectory is , amplitude change rate and phase jump variable They are calculated by the following formulas: ; ; in, Indicates In the window and The amplitude change rate between sampling points, Represents the amount of phase change. By performing differential operations between each sampling point, the signal change feature sequence is obtained to reflect the signal mutation. Especially in a multi-label environment, these features are used to identify signal conflicts and collisions. Input the signal change feature sequence into the dual threshold detector, set the amplitude change rate threshold to 20%, and the phase jump threshold to 45 degrees. The dual threshold detector determines whether the signal has changed dramatically by using the preset amplitude and phase change thresholds, thereby preliminarily determining whether there is a collision of tag signals. Set the amplitude change rate threshold to , the phase jump threshold is If the amplitude change rate of a signal point Or phase jump variable , then the point is considered to be a potential collision point, and the primary collision judgment result is obtained. For the primary collision judgment result, perform signal vector diagram analysis, and determine the collision area by counting the number of cluster centers in the signal space. Plot the signal change feature sequence as points in the feature space, analyze the distribution of these points, and use a clustering algorithm (such as K-means clustering) to count the number of cluster centers of the signal change features. If the number of cluster centers in a certain area is greater than 1, it is considered that there is a signal collision in the area, and it is marked as a potential collision point to obtain a candidate set of collision points. Based on the candidate set of collision points, a collision feature vector is constructed, which includes three feature quantities: amplitude mutation degree, phase jump degree, and clustering discreteness, to obtain a collision feature sequence. Suppose the amplitude mutation degree of a certain collision point is , the phase jump degree is , the cluster dispersion is These characteristics reflect the amplitude change degree, phase change degree and signal clustering concentration degree of the collision point. Through the comprehensive analysis of these characteristics, a complete collision feature vector is formed: ; The collision feature sequence is input into the support vector machine (SVM) classifier for collision event recognition to obtain the label sequence of the collision event. SVM is a machine learning classifier suitable for binary and multi-classification problems. Determine whether a collision event occurs in the current signal, and output a sequence of markers containing whether each time window is a collision. ,in , 0 means no collision, 1 means collision. The time domain correlation analysis of the collision event marker sequence is performed to merge adjacent collision event markers and generate a conflict position sequence. Suppose the collision event marker sequence is If several adjacent marks are all 1, these positions are merged into a single conflict event, and finally the conflict position sequence is obtained. , where each Indicates the starting position of a collision event.

[0032] In a specific embodiment, the process of executing step 105 may specifically include the following steps: The data marked by the conflict position sequence are divided according to a fixed length of 8 bits to construct a data unit sequence and obtain a set of data units to be retransmitted; Fill the set of data units to be retransmitted into an M×N interleaving matrix in row priority order, where M is the data unit length and N is the currently detected collision frequency value, to obtain an initial interleaving matrix; Performing a column cyclic shift operation on the initial interleaving matrix, the shift amount is linearly related to the column number, generating an interleaving data matrix with scattered positions, obtaining a rearranged data matrix, reading the rearranged data matrix according to column priority order, and grouping the read data, each group of data is 4 bits long, to obtain a data group to be encoded; Perform Hamming coding operation on the coded data group to generate a 3-bit check bit, and add the check bit to the original data to obtain a coded data sequence, and construct a data packet header based on the coded data sequence, including interleaving depth information, data packet length and original position identifier, to obtain a data packet header; Combine the data packet header with the coded data sequence, encapsulate it into a data packet according to a fixed format, add a CRC check code, generate a 32-bit cyclic redundancy check code, and output an anti-collision data packet.

[0033] Specifically, the data marked by the conflict position sequence is divided into 8-bit fixed lengths to construct a data unit sequence and obtain a set of data units to be retransmitted. ,in Indicates each bit. The data units are divided into groups of 8 bits each and are expressed as follows: ,in is the index of the data unit, ranging from By this segmentation method, the original conflicting data is divided into multiple 8-bit data units, ensuring that the subsequent data processing is performed at a fixed length, thereby simplifying the encoding and interleaving operations. The data units to be retransmitted are filled into one according to the row priority order. in an interwoven matrix. Indicates the length of the data unit, that is, the number of bits contained in each data unit, and its value is 8; Indicates the currently detected collision frequency value, reflecting the frequency of tag collision during data transmission. The initial interleaving matrix is ​​obtained by filling the data units into the interleaving matrix row by row in row priority order. , each element of which Indicates Row, No. The bit value in the data unit of the column. The padding operation is expressed as: ; in, , In this way, data is filled into the matrix in row order to form an initial interleaving matrix. The initial interleaving matrix is ​​subjected to column cyclic shift operation. The shift amount is linearly related to the column number, which means that the first The column shift amount is ,in is a constant, usually less than The value of is to ensure that the column data after the shift is not lost. The column shift amount is , the shifted matrix elements are expressed as: ; in, represents the rearranged matrix elements, Represents the new row index after circular shift. Through this column circular shift, the data order in the original matrix is ​​disrupted, and an interleaved data matrix with scattered positions is generated to obtain a rearranged data matrix. The position scattering operation aims to disperse the data in the original order to reduce the impact of continuous collisions and improve the success rate of data retransmission. The rearranged data matrix is ​​read out in column priority order. That is, starting from the first column, each element is read column by column, and a new data sequence is obtained after all the data are read out. Suppose the read data is , these data are then grouped, each group is 4 bits long, to obtain the data group to be encoded ,in Represents the index of the data group. For each data group to be encoded, perform Hamming coding. Hamming coding is a coding method that can detect and correct single-bit errors. By adding redundant check bits, the receiving end can detect and correct single-bit errors that occur during data transmission. Suppose the data group to be encoded is The Hamming encoding process is to generate 3 check bits , these check digits are calculated by the following linear equations: ; ; ; in, Represents a bitwise XOR operation. Add to the original data group to get a 7-bit coded data sequence The data sequence after Hamming coding can effectively improve the reliability of data transmission, ensuring that even if a single-bit error occurs during transmission, it can be corrected by the check bit. The data packet header is constructed based on the coded data sequence, and the data packet header contains information such as interleaving depth information, data packet length, and original position identifier. Assume that the interleaving depth is , the length of the data packet is , the original position is marked as , then the packet header is expressed as . Interweaving Depth Indicates the level at which data is interleaved, and the length of the data packet Indicates the number of bits in the entire data packet, while the original position identifier It is used to indicate the location of the data in the original conflicting data so that the receiving end can restore the data to the correct location. With the encoded data sequence Combined together, encapsulated into a complete data packet according to a fixed format, and a cyclic redundancy check code (CRC) is added to the end of the data packet. CRC is an error detection method used in digital communications. It generates a check code by performing a specific polynomial operation on the data packet and appends it to the end of the data packet so that the receiving end can verify the integrity of the data. Suppose the data packet is , the generation of CRC check code is expressed as: ; in, represents the polynomial corresponding to the data packet, is a predefined generating polynomial, The generated 32-bit CRC checksum is added to the end of the data packet to form the final anti-collision data packet. .

[0034] In a specific embodiment, the process of executing step 106 may specifically include the following steps: Perform the first-level priority calculation on the anti-collision data packets, assign a weight coefficient of 0.4 to the label data volume, a weight coefficient of 0.3 to the number of retransmissions, and a weight coefficient of 0.3 to the waiting time, and obtain the first priority sequence according to the weighted sum function; perform the second-level priority calculation on the first priority sequence, perform exponential weighting based on the transmission delay requirement, and obtain the label priority sequence; The tag priority sequence is input into the dual threshold selector, the first threshold is set to 80% of the system processing capacity, the second threshold is set to the priority average, and tags that meet the dual threshold conditions are selected into the first transmission queue, and tags that only meet the single threshold conditions are selected into the second transmission queue, thereby obtaining a hierarchical set of tags to be scheduled; A two-level clustering algorithm is performed on the hierarchical set of labels to be scheduled. The labels are roughly grouped in the first level, and each rough group is finely divided in the second level. The time slot protection factor β=0.1 is set between groups to obtain a multi-level transmission group structure. A time slot resource allocation tree is constructed based on a multi-level transmission group structure. The first-level nodes allocate basic time slot resources, the second-level nodes allocate peak-shifting intervals within the group, and the third-level nodes allocate specific transmission times, thereby obtaining a hierarchical time slot allocation scheme. The hierarchical time slot allocation scheme is optimized by three levels: the first level optimizes the peak shift between groups, the second level optimizes the peak shift within groups, and the third level optimizes the peak shift at nodes, generating a multi-dimensional peak shift control sequence that minimizes the probability of conflict. According to the multi-dimensional peak-shifting control sequence, a three-level scheduling queue is established: the first level is the emergency transmission queue, the second level is the standard transmission queue, and the third level is the low-priority transmission queue. Differentiated scheduling cycles are assigned to each queue to obtain a hierarchical scheduling control sequence. The performance monitoring point matrix is ​​set based on the hierarchical scheduling control sequence, including: the first monitoring point is set at the queue switching, the second monitoring point is set at the time slot boundary, and the third monitoring point is set at the transmission completion point. The multi-dimensional performance evaluation result is obtained through the joint analysis of the three points; A three-level dynamic adjustment mechanism is triggered based on the results of multi-dimensional performance evaluation: the first level adjusts the transmission group division, the second level adjusts the time slot allocation, and the third level adjusts the peak shifting parameters. An exponential backoff algorithm is used to control the adjustment cycle and output the multi-tag concurrent transmission timing.

[0035] Specifically, the data volume, number of retransmissions, and waiting time of each tag are comprehensively evaluated, and different weights are assigned to these factors to obtain the first-level priority sequence. Suppose the parameters of a tag are the data volume , Retransmission times , Waiting time , set the weight coefficient of data volume to 0.4, the weight coefficient of retransmission times to 0.3, and the weight coefficient of waiting time to 0.3, then the first-level priority value Calculated by weighted sum function, the formula is: ; in, Indicates The priority values ​​of the labels are given by assigning different weights to each factor, so that labels with higher data volume, more frequent retransmissions and longer waiting time have higher priority, thus being given more priority in transmission scheduling. By performing corresponding calculations on all labels, the first priority sequence is obtained: ,in Represents the total number of labels. Perform the second-level priority calculation on the first priority sequence and perform exponential weighting based on the transmission delay requirement to refine the priority differences of labels. Suppose the transmission delay requirement is , an exponential weighting function is applied to the first-level priority value to obtain a new label priority sequence , the formula is: ; in, is a weighting factor that determines the degree of influence of transmission delay on the priority value. Through exponential weighting, the priority of tags that are not demanding on transmission delay can be significantly reduced, while the priority of tags that are more sensitive to transmission delay can be increased, so that the system can better adapt to application scenarios with strong delay constraints. The tag priority sequence is input into the dual threshold selector for screening. The first threshold of the dual threshold selector is set to 80% of the system processing capacity, and the second threshold is set to the mean of the current tag priority sequence. Assume that the first threshold is ,in Indicates the system processing capacity, the second threshold is If the priority value of a tag satisfies both and , then the tag is selected into the first transmission queue, otherwise if only a single threshold is met, it is selected into the second transmission queue. Through this step, the tags are graded and screened to obtain a hierarchical set of tags to be scheduled. For the hierarchical set of tags to be scheduled, a secondary clustering algorithm is executed to build a multi-level transmission group structure. In the secondary clustering algorithm, the tags are roughly grouped and divided into several groups according to priority, each group contains tags with similar priority; then each coarse group is finely divided, and the tags in each group are further subdivided into smaller subgroups. Set a time slot protection factor between groups , to ensure that the transmission time slots between different groups do not interfere with each other, and obtain a multi-level transmission group structure. Based on the multi-level transmission group structure, a time slot resource allocation tree is constructed. The first-level nodes of the time slot resource allocation tree are used to allocate basic time slot resources, the second-level nodes are responsible for allocating staggered intervals for transmission within the group, and the third-level nodes specifically allocate the transmission time of each label. Assume that the basic time slot resource is , then the allocation of the first layer nodes is expressed as: ; in, is the number of groups; the second-layer nodes allocate staggered intervals for transmission within the group, and the staggered interval within the group is , the allocated time slots are: ; in, is the index of the group; the third-layer node assigns a specific transmission time to each label to obtain a hierarchical time slot allocation scheme. Through hierarchical allocation, it is ensured that the transmission of different groups and different labels within the group can be carried out in an orderly manner, thereby reducing conflicts. A triple peak-shifting optimization is performed on the hierarchical time slot allocation scheme to reduce the probability of conflicts. The first optimization is inter-group peak-shifting optimization, which makes the transmission between different groups more staggered by adjusting the time slot difference between groups; the second optimization is intra-group peak-shifting optimization, which adjusts the transmission time of the label within the group to reduce the mutual interference of the transmission within the group; the third optimization is node peak-shifting optimization, which makes the transmission time more dispersed by further staggering the specific nodes. Through these three optimizations, a multi-dimensional peak-shifting control sequence that minimizes the probability of conflicts is generated, thereby effectively improving the success rate of data transmission. According to the multi-dimensional peak-shifting control sequence, three-level scheduling queues are established, namely, the emergency transmission queue, the standard transmission queue, and the low-priority transmission queue. For each queue, a differentiated scheduling period is assigned according to its urgency. Suppose the scheduling period of the emergency transmission queue is , the scheduling period of the standard transmission queue is , the scheduling period of the low priority transmission queue is ,and . Through the hierarchical scheduling method, the transmission requirements of emergency tags are met first, while the transmission of standard and low-priority tags is taken into account, and a hierarchical scheduling control sequence is obtained. Based on the hierarchical scheduling control sequence, a performance monitoring matrix is ​​set to monitor the performance of the system in real time. The performance monitoring matrix contains three main monitoring points: the first monitoring point is set at the queue switching point to monitor whether the switching between queues of different priorities is smooth; the second monitoring point is set at the time slot boundary to ensure that the use of each time slot meets expectations and there is no time slot overlap; the third monitoring point is set at the transmission completion point to evaluate the success rate and delay of the transmission. Through the joint analysis of the three monitoring points, a multi-dimensional performance evaluation result is obtained to ensure the effectiveness of the entire transmission scheduling process. According to the multi-dimensional performance evaluation results, a three-level dynamic adjustment mechanism is triggered to adapt to the changing conditions during the system operation. The first level of dynamic adjustment is to adjust the division of the transmission group, and adapt to the changing system load by re-dividing the tag group; the second level of adjustment is to adjust the allocation of time slots, and optimize the use of time slots according to the real-time performance evaluation results; the third level of adjustment is to adjust the peak shifting parameters, and further reduce conflicts by changing the peak shifting parameters between and within groups. All adjustments use the exponential backoff algorithm to control the adjustment period. Suppose the adjustment period is , dynamically adjust the period through exponential backoff to adapt to the actual situation of the system: ; in, is the initial adjustment period, is the number of adjustments. In this way, the adjustment frequency changes dynamically according to the stability of the system, thereby outputting a concurrent transmission sequence of multiple tags.

[0036] The above describes the UHF RFID system data transmission method in the embodiment of the present invention. The following describes the UHF RFID system data transmission device in the embodiment of the present invention. Figure 2 , an embodiment of the ultra-high frequency RFID system data transmission device in the embodiment of the present invention includes: The filtering module 201 is used to perform bandpass filtering and orthogonal demodulation on the multi-tag backscattering signals received by the RFID reader to obtain a tag feature data sequence; The calculation module 202 is used to perform multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame; The identification module 203 is used to identify the boundary of the independent tag signal frame and determine the starting point of the tag data frame; The analysis module 204 is used to perform continuous sampling analysis and tag signal conflict identification on the signal after the starting point of the tag data frame, and output a conflict position sequence; A rearrangement module 205, for performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet; The output module 206 is used to construct a tag access scheduling queue based on the transmission timing and priority attributes of the anti-collision data packet, and output the multi-tag concurrent transmission timing.

[0037] Through the coordinated cooperation of the above components, 1. The time-space separated signal feature analysis and multi-dimensional feature mapping technology, combined with multi-level feature extraction and correlation analysis, effectively improve the recognition accuracy in multi-tag environment; 2. The innovative introduction of adaptive interleaving coding and multiple staggered transmission mechanisms enhances the system's anti-interference ability and data transmission reliability in high-density tag environments; 3. A priority-based multi-level scheduling algorithm is designed, combined with a dynamic time slot allocation strategy, to reduce tag response conflicts and reduce transmission delays; 4. Through hierarchical staggered control and performance adaptive optimization, the system resource utilization efficiency is optimized and the data transmission throughput is improved; 5. The use of multiple anti-collision coding and intelligent retransmission control enhances the system's transmission stability in complex environments.

[0038] above Figure 2 The medium and ultra-high frequency RFID system data transmission device in the embodiment of the present invention is described in detail from the perspective of modular functional entities, and the ultra-high frequency RFID system data transmission device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0039] Figure 33 is a schematic diagram of the structure of an ultra-high frequency RFID system data transmission device provided by an embodiment of the present invention. The ultra-high frequency RFID system data transmission device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 can be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the ultra-high frequency RFID system data transmission device 300. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the ultra-high frequency RFID system data transmission device 300 to implement the steps of the above-mentioned ultra-high frequency RFID system data transmission method.

[0040] The UHF RFID system data transmission device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Figure 3 The illustrated structure of the UHF RFID system data transmission device does not constitute a limitation on the UHF RFID system data transmission device provided by the present invention, and may include more or fewer components than illustrated, or a combination of certain components, or a different arrangement of components.

[0041] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the ultra-high frequency RFID system data transmission method.

[0042] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0043] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0044] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for transmitting data in an ultra-high frequency RFID system, characterized in that: The method comprises: Bandpass filtering and orthogonal demodulation are performed on the multi-tag backscatter signals received by the RFID reader to obtain the tag characteristic data sequence; Performing multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame; Perform boundary identification on the independent tag signal frame to determine the starting point of the tag data frame; Performing continuous sampling analysis and tag signal conflict identification on the signal after the starting point of the tag data frame, and outputting a conflict position sequence; Performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet; A tag access scheduling queue is constructed based on the transmission timing and priority attributes of the anti-collision data packet, and a multi-tag concurrent transmission timing is output.

2. The UHF RFID system data transmission method according to claim 1, characterized in that: The method of performing bandpass filtering and orthogonal demodulation on the multi-tag backscatter signals received by the RFID reader to obtain a tag feature data sequence includes: Receiving multi-tag backscatter signals through an RFID reader, and filtering the multi-tag backscatter signals through a bandpass filter with a center frequency of 915 MHz to obtain a primary filtered signal; The primary filter signal is sampled at four times the carrier frequency, and frequency converted by a digital down-conversion circuit to obtain an intermediate frequency signal, and the intermediate frequency signal is input into an orthogonal demodulator, and multiplied with the in-phase branch and the orthogonal branch of the local carrier respectively to obtain an in-phase component I and an orthogonal component Q; The in-phase component I and the quadrature component Q are subjected to band-limiting processing by a low-pass filter to obtain a baseband I-channel signal and a baseband Q-channel signal, and the baseband I-channel signal and the baseband Q-channel signal are subjected to digital sampling by a 16-bit analog-to-digital converter to obtain a digital I-channel data stream and a digital Q-channel data stream; Performing timing recovery on the digitized I data stream and the digitized Q data stream, establishing a sampling clock sequence, and dividing the digitized I data stream and the digitized Q data stream according to the sampling clock sequence according to a 16-bit RN16 length to obtain an I data block and a Q data block; The I-path data blocks and the Q-path data blocks are combined into a label feature data sequence in a plural form.

3. The UHF RFID system data transmission method according to claim 2, characterized in that: The performing multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame includes: Performing complex operations on the tag feature data sequence, calculating the square of the amplitude and the phase angle of each data point, obtaining a tag signal strength sequence and a phase sequence, and sorting the tag signal strength sequence according to the signal energy to obtain a valid signal point set; Extracting arrival time information for each data point in the effective signal point set, and performing weighted averaging on the time information according to the signal strength to obtain a tag response time sequence; Inputting the tag response time sequence into a feature classifier, grouping the time intervals based on a K-means clustering algorithm, and obtaining a tag response period set; According to the tag response period set, an A×B dimensional feature matrix is ​​constructed, where A is the number of tags and B is the feature dimension, including amplitude, phase and time information, to obtain a feature space mapping matrix; Calculate the Pearson correlation coefficient of each row of the feature space mapping matrix with the adjacent data frame, and set the correlation threshold to 0.8 to obtain a label correlation matrix; Based on the label correlation matrix, connected component analysis is performed, data frames with correlation coefficients higher than a threshold are combined to obtain an independent label candidate set, and timing consistency verification is performed on the independent label candidate set to eliminate data frames that do not meet the fixed response period and output independent label signal frames.

4. The UHF RFID system data transmission method according to claim 3, characterized in that: The step of identifying the boundary of the independent tag signal frame and determining the starting point of the tag data frame includes: Constructing a 16-bit cyclic shift synchronization codeword for the independent tag signal frame to generate a standard synchronization codeword with a maximum and minimum Hamming distance, and performing a full-cycle cyclic shift operation on the standard synchronization codeword to generate a set of 16 shifted versions of the synchronization codeword; The independent tag signal frame is segmented according to the synchronization codeword length, each segment of the signal sequence is normalized to obtain a data sequence to be matched, and the data sequence to be matched is correlated with each shifted version in the synchronization codeword set to calculate a 256-point correlation value sequence to obtain a correlation coefficient matrix; Performing peak detection on each column of the correlation coefficient matrix, extracting the position information and amplitude information of the local maximum point, obtaining a candidate synchronization point set, and calculating the optimal decision threshold according to the amplitude distribution of the candidate synchronization point set to obtain a valid synchronization position sequence; A majority vote is performed on the synchronization points of three adjacent data frames in the valid synchronization position sequence, and the position with the highest frequency of occurrence is selected to obtain the frame synchronization position. The data frame is divided and reorganized based on the frame synchronization position, and the independent label signal frame is divided into data units of fixed length according to the synchronization position, and the starting point of the label data frame is output.

5. The UHF RFID system data transmission method according to claim 4, characterized in that: The continuous sampling analysis and tag signal conflict identification of the signal after the starting point of the tag data frame and outputting the conflict position sequence include: The signal after the starting point of the tag data frame is segmented by a sliding window, the window length is set to RN16 length, the sliding step length is 1 code element, and a continuous signal window sequence is obtained; Performing complex envelope detection on each window data in the continuous signal window sequence, extracting the signal amplitude trajectory and phase trajectory to obtain a signal feature trajectory sequence, performing differential operation on the signal feature trajectory sequence, calculating the amplitude change rate and phase jump amount between adjacent sampling points, and obtaining a signal change feature sequence; Input the signal change characteristic sequence into a dual threshold detector, set the amplitude change rate threshold to 20%, and the phase jump threshold to 45 degrees, to obtain a primary collision judgment result; Performing signal vector diagram analysis on the primary collision judgment result, counting the number of cluster centers in the signal space, marking the area where the number of cluster centers is greater than 1 as a potential collision point, and obtaining a collision point candidate set; Based on the candidate set of collision points, a collision feature vector is constructed, which includes three feature quantities: amplitude mutation degree, phase jump degree and clustering discreteness, to obtain a collision feature sequence; The collision feature sequence is input into a support vector machine classifier for collision event recognition to obtain a collision event marker sequence, and a time domain correlation analysis is performed on the collision event marker sequence to merge adjacent collision event markers to generate a conflict position sequence.

6. The UHF RFID system data transmission method according to claim 5, characterized in that: The step of performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet includes: The data marked by the conflict position sequence is divided according to a fixed length of 8 bits to construct a data unit sequence to obtain a set of data units to be retransmitted; Fill the set of data units to be retransmitted into an M×N interleaving matrix in a row priority order, where M is the data unit length and N is the currently detected collision frequency value, to obtain an initial interleaving matrix; Performing a column cyclic shift operation on the initial interleaving matrix, wherein the shift amount is linearly related to the column number, generating an interleaving data matrix with positions scattered, obtaining a rearranged data matrix, reading the rearranged data matrix in column priority order, and grouping the read data, wherein each group of data has a length of 4 bits, to obtain a data group to be encoded; Performing a Hamming coding operation on the data group to be encoded, generating a 3-bit check bit, and adding the check bit to the original data to obtain a coded data sequence, constructing a data packet header based on the coded data sequence, including interleaving depth information, data packet length and original position identifier, to obtain a data packet header; The data packet header is combined with the coded data sequence, encapsulated into a data packet according to a fixed format, and a CRC check code is added to generate a 32-bit cyclic redundancy check code, and an anti-collision data packet is output.

7. The UHF RFID system data transmission method according to claim 6, characterized in that: The step of constructing a tag access scheduling queue based on the transmission timing and priority attributes of the anti-collision data packet and outputting a multi-tag concurrent transmission timing includes: Perform a first-level priority calculation on the anti-collision data packet, assign a weight coefficient of 0.4 to the label data volume, a weight coefficient of 0.3 to the number of retransmissions, and a weight coefficient of 0.3 to the waiting time, and obtain a first priority sequence according to a weighted sum function; perform a second-level priority calculation on the first priority sequence, perform exponential weighting based on the transmission delay requirement, and obtain a label priority sequence; The tag priority sequence is input into a dual-threshold selector, the first threshold is set to 80% of the system processing capacity, the second threshold is set to the priority average, and tags that meet the dual-threshold conditions are selected into the first transmission queue, and tags that only meet the single threshold are selected into the second transmission queue, so as to obtain a hierarchical set of tags to be scheduled; A two-level clustering algorithm is performed on the hierarchical set of labels to be scheduled. The labels are roughly grouped in the first level, and each rough group is finely divided in the second level. A time slot protection factor β=0.1 is set between groups to obtain a multi-level transmission group structure. Based on the multi-level transmission group structure, a time slot resource allocation tree is constructed, where the first-layer nodes allocate basic time slot resources, the second-layer nodes allocate peak-shifting intervals within the group, and the third-layer nodes allocate specific transmission times, to obtain a hierarchical time slot allocation scheme; Performing triple peak-shifting optimization on the hierarchical time slot allocation scheme: firstly optimizing inter-group peak-shifting, secondly optimizing intra-group peak-shifting, and thirdly optimizing node peak-shifting, to generate a multi-dimensional peak-shifting control sequence that minimizes the probability of conflict; According to the multi-dimensional peak-shifting control sequence, a three-level scheduling queue is established: the first level is an emergency transmission queue, the second level is a standard transmission queue, and the third level is a low-priority transmission queue, and a differentiated scheduling period is assigned to each queue to obtain a hierarchical scheduling control sequence; The performance monitoring point matrix is ​​set based on the hierarchical scheduling control sequence, including: a first monitoring point is set at a queue switching location, a second monitoring point is set at a time slot boundary, and a third monitoring point is set at a transmission completion location, and a multi-dimensional performance evaluation result is obtained by joint analysis of the three points; A three-level dynamic adjustment mechanism is triggered according to the multi-dimensional performance evaluation results: the first level adjusts the transmission group division, the second level adjusts the time slot allocation, and the third level adjusts the peak shifting parameters, and uses an exponential backoff algorithm to control the adjustment cycle and output the multi-label concurrent transmission timing.

8. An ultra-high frequency RFID system data transmission device, characterized in that: The device is used to execute the UHF RFID system data transmission method according to any one of claims 1 to 7, the device comprising: A filtering module is used to perform bandpass filtering and orthogonal demodulation on the multi-tag backscattering signals received by the RFID reader to obtain a tag feature data sequence; A calculation module, used for performing multi-dimensional feature analysis and time series correlation calculation on the tag feature data sequence to obtain an independent tag signal frame; An identification module, used to identify the boundary of the independent tag signal frame and determine the starting point of the tag data frame; An analysis module, used for performing continuous sampling analysis and tag signal conflict identification on the signal after the starting point of the tag data frame, and outputting a conflict position sequence; A rearrangement module, used for performing matrix interleaving transformation and position rearrangement on the data marked by the conflict position sequence to generate an anti-collision data packet; The output module is used to construct a tag access scheduling queue based on the transmission timing and priority attributes of the anti-collision data packet, and output the multi-tag concurrent transmission timing.

9. An ultra-high frequency RFID system data transmission device, characterized in that: The UHF RFID system data transmission device comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the UHF RFID system data transmission device to execute the UHF RFID system data transmission method according to any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the UHF RFID system data transmission method according to any one of claims 1 to 7 is implemented.

Images