Bicycle anti-loss tracking method and system based on radio frequency identification

Abstract

The invention relates to the technical field of radio frequency identification, in particular to a bicycle anti-loss tracking method and system based on radio frequency identification, and the method comprises the following steps: analyzing the passing frequency and the staying frequency of an identification region, generating scene communication configuration, obtaining a channel response sequence, generating compensation scheduling configuration, and building an abnormal behavior identification record; and detecting the border crossing behavior and evaluating the credibility, sending a GPS starting and locking instruction, and outputting a behavior control response result. According to the method, high matching of communication configuration and environment characteristics and dynamic identification of channel conflict degree are realized by using traffic data analysis, the stability of channel access response is optimized, and continuous control of identification compensation and analysis of integrity of a label identification chain are enhanced by combining a node state change trend and a spatial position relationship; the accuracy of abnormal behavior detection is improved, the boundary misjudgment rate is reduced in combination with signal fluctuation and response stability analysis, and accurate recognition of the boundary crossing state and control response linkage are achieved.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency identification technology, and in particular to a bicycle anti-loss tracking method and system based on radio frequency identification. Background Art

[0002] The field of radio frequency identification technology includes a key technology system that uses radio frequency signals to achieve contactless data communication and target identification. The core content is built around radio frequency tags, reading and writing devices, antenna components, and data analysis systems. It completes the capture and interaction of the target object's identity information through the electromagnetic field coupling of the radio wave frequency band. Radio frequency identification works based on the backscatter modulation principle and involves multiple underlying technologies such as carrier frequency generation, signal modulation and demodulation, encoding and decoding protocols, and is widely used in logistics tracking, inventory management, electronic payment and other scenarios.

[0003] Among them, the bicycle anti-loss tracking method based on radio frequency identification refers to an anti-theft technology formed by integrating radio frequency identification components with vehicle positioning architecture. By deploying radio frequency tags with unique codes on the bicycle body, periodic signal scanning is performed based on a fixed reader network within a preset geographical range. When the vehicle moves, the reader node receives the change in the signal strength value fed back by the tag, and combines the multi-base station signal arrival time difference calculation technology to realize dynamic monitoring of the vehicle position. By continuously comparing the vehicle's authorized use area with the real-time location data, the abnormal movement status judgment mechanism is triggered. When the vehicle exceeds the authorized area, the wireless communication unit of the base station node is triggered, and a coordinate offset alarm is sent to the management terminal.

[0004] Traditional bicycle anti-lost tracking technology uses base stations to receive changes in tag signal strength and combines arrival time difference to determine displacement behavior. The recognition basis relies on the response characteristics of a single tag during periodic reading. Dynamic channel adjustment is difficult to achieve in high-density environments, resulting in an increased frequency band conflict rate and a high recognition loss rate. The recognition path lacks a continuity integrity verification mechanism, resulting in the tag's jumping behavior under short-term occlusion or interference fluctuations not being accurately identified, causing location record jumps or behavioral anomalies to be misjudged. The out-of-bounds determination process mainly relies on a simple comparison of tag coordinates with the boundaries of the authorized area. There is a lack of analysis of the signal characteristics of the out-of-bounds point and the trend of environmental interference. This leads to frequent false alarms in areas with strong boundary signal reflections, a large number of false out-of-bounds triggers at the edges of densely deployed areas, and the effective scheduling and execution reliability of the interference control response system. There is a lack of analysis and feedback methods for spatial behavior continuity, real-time channel load conditions, and tag response credibility, which limits the recognition accuracy and positioning stability in complex dynamic scenarios. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a bicycle anti-lost tracking method and system based on radio frequency identification.

[0006] To achieve the above-mentioned object, the present invention adopts the following technical solution: a bicycle anti-lost tracking method based on radio frequency identification, comprising the following steps:

[0007] S1: Through tag recognition records, analyze the frequency of passage and stay in the identified area within multiple time periods, classify the identified area into types and output scene identifications. Combined with the matching combination between the area density level and the communication stability level, the scene communication configuration is generated;

[0008] S2: Call the communication configuration of the scenario, analyze the number of active tags and the fluctuation amplitude of the communication channel identification signal, compare the activation tag density and signal fluctuation trend of each channel, calculate the matching ratio of tag distribution density and channel signal-to-noise ratio change, identify the degree of signal conflict, and obtain the channel response sequence;

[0009] S3: Analyze the tag recognition failure of each identification node in consecutive cycles according to the channel response sequence, match the status identifier of the identification node, calculate the compensation recognition priority of the node in combination with the spatial distance of the adjacent nodes, and generate a compensation scheduling configuration;

[0010] S4: Call the compensation scheduling configuration, build a label identification path according to the spatial distribution of the identification nodes, extract the interruption position of the identification path by analyzing the number of consecutive records of the label between each pair of nodes in the path sequence, and detect abnormal jumping behavior in combination with the identification integrity of the path, and establish an abnormal behavior identification record.

[0011] As a further solution of the present invention, the scenario communication configuration includes an area classification label, a frequency band selection group, and an identification channel identifier; the channel response sequence is specifically a label distribution sorting group, a channel interference sorting list, and a response priority mapping table; the compensation scheduling configuration includes a node failure identifier, a spatial position chain group, and a compensation priority table; the abnormal behavior identification record is specifically an identification broken link position set, a jump event annotation table, and a path stability mapping set.

[0012] As a further solution of the present invention, the step of obtaining the scene communication configuration is specifically as follows:

[0013] S111: Obtain the frequency of bicycle traffic and the frequency of bicycle stops in multiple time periods through tag identification records, analyze the distribution of the number of bicycles passing through and the length of stay in the target area in multiple time periods, and establish a regional behavior feature set;

[0014] S112: Analyze the activity level and stay concentration of the target area in multiple time periods based on the set of regional behavior features, identify regional usage features, classify and label the regional types, and output scene identification information;

[0015] S113: Call the scene identification information, match the communication frequency band group corresponding to the tag according to the corresponding area density level and communication stability level, and generate a scene communication configuration.

[0016] As a further solution of the present invention, the step of obtaining the channel response sequence is specifically as follows:

[0017] S211: Calling the scene communication configuration, analyzing the tag identification records and signal strength records of each communication channel in the current identification area, obtaining the number of active tags and the signal strength change amplitude of each channel within a unit period, and generating a tag and signal fluctuation indicator group;

[0018] S212: Comparing the label quantity density and signal fluctuation amplitude change trend of each communication channel based on the label and signal fluctuation index group, calculating the matching relationship between the label distribution density and the channel signal-to-noise ratio change, and calculating the communication interference matching coefficient;

[0019] S213: Identify the signal conflict degree of each communication channel according to the communication interference matching coefficient, calculate the priority of each communication channel, and establish a channel response sorting sequence.

[0020] As a further solution of the present invention, the step of obtaining the compensation scheduling configuration is specifically as follows:

[0021] S311: Collect identification records of each identification node in multiple consecutive cycles according to the channel response sorting sequence, analyze the number of tag identification failures recorded by each node in the consecutive cycles, and analyze the direction of failure change in the consecutive cycles to generate a node status identification set;

[0022] S312: Calling the node status identification set, combining the geometric coordinate data between the nodes, calculating the spatial distance between each identified node and the adjacent node, determining the jump between consecutive identified nodes in the identification chain structure, and establishing path structure continuity information;

[0023] S313: Based on the path structure continuity information, extract the recognition failure behavior, failure trend, spatial jump characteristics and structural fracture risk of each node, analyze the recognition behavior strength and structural impact, calculate the priority score value of each node during compensation scheduling, establish a compensation priority queue, and generate a compensation scheduling configuration.

[0024] As a further solution of the present invention, the steps for obtaining abnormal behavior identification records are specifically as follows:

[0025] S411: calling the compensation scheduling configuration, constructing a label identification path according to the spatial distribution of the identification nodes, and establishing a label path sequence index set;

[0026] S412: Based on the label path sequence index set, by analyzing the number of consecutive records of labels between each pair of nodes in the path sequence, the interruption position in the path is identified and marked as a jump segment. The jump start and end node indexes, index span, response loss time and node spatial distance are recorded to obtain a jump segment feature set;

[0027] S413: Call the jump segment feature set, normalize the response missing time and spatial distance, and calculate the jump behavior abnormality score value based on the path span. Combined with the recognition coverage completeness of the label in the path, detect abnormal jump behavior and establish an abnormal behavior identification record.

[0028] As a further embodiment of the present invention, the method further comprises:

[0029] S5: Calling the abnormal behavior recognition record, using the boundary node to detect bicycle crossing the boundary, combining the signal fluctuation change characteristics of the recognition node corresponding to the crossing boundary coordinate and the response stability performance of the tag during the recognition cycle, analyzing the interference risk of the target location signal environment, outputting the cross-border state recognition credibility, detecting and eliminating drift recognition events, and sending abnormal behavior warning instructions to the bicycle corresponding to the abnormal cross-border state, including starting GPS and locking the bicycle, and outputting the behavior control response result;

[0030] The behavior control response result includes positioning start status, abnormal locking command, and out-of-bounds alarm information.

[0031] As a further solution of the present invention, the steps of obtaining the behavior control response result are specifically as follows:

[0032] S511: Calling the abnormal behavior identification record, obtaining the authorized area coordinate set consisting of boundary nodes, detecting bicycle crossing the boundary based on the spatial relative relationship between the real-time position of each tag and the boundary coordinates, and obtaining a crossing state identifier set;

[0033] S512: Based on the out-of-bounds status identifier set, extract the channel signal strength sequence and fluctuation amplitude characteristics of the corresponding identification node, calculate the response stability performance of the tag in the current identification cycle, analyze the interference risk of the target location signal environment, output the out-of-bounds status identification credibility, detect and eliminate drift identification events, and obtain a valid out-of-bounds behavior set;

[0034] S513: According to the valid out-of-bounds behavior set, identify the target tag number corresponding to the abnormal out-of-bounds state, send an abnormal behavior warning instruction and activate the remote control module, including starting GPS and locking the car instructions, and establish a behavior control response result.

[0035] A bicycle anti-lost tracking system based on radio frequency identification, wherein the bicycle anti-lost tracking system based on radio frequency identification is used to execute the above-mentioned bicycle anti-lost tracking method based on radio frequency identification, and the system comprises:

[0036] The regional scene calibration module analyzes the frequency of passage and dwell time in the identified area within multiple time periods based on tag identification records, identifies regional features and obtains scene identification. It then combines the matching between the regional density level and the communication stability level to obtain the scene communication configuration.

[0037] The channel dynamic optimization module analyzes the number of active tags in the target identification area and the fluctuation amplitude of the communication channel identification signal based on the communication configuration of the scenario, compares the activation tag density and signal fluctuation trend of each channel, calculates the matching ratio of tag distribution density and channel signal-to-noise ratio change, identifies the degree of signal conflict, sorts the response priority of the channel, and generates a channel response sequence;

[0038] The node status classification module extracts the number of tag recognition failures of each identification node in consecutive cycles based on the channel response sequence, analyzes the stability status of the node according to the recognition failure trend, calculates the priority of the node compensation demand according to the spatial distance relationship between adjacent nodes, and establishes a compensation scheduling configuration;

[0039] The path continuity detection module constructs a path sequence to identify the spatial position of nodes and performs continuity analysis of label identification based on the compensation scheduling configuration, detects interruption positions of node records in the label identification path, detects abnormal jump behaviors based on the identification integrity of the labels in the path, and establishes abnormal behavior identification records;

[0040] The out-of-bounds credibility assessment module is based on the abnormal behavior identification records and uses boundary nodes to detect bicycle out-of-bounds behavior. By analyzing the signal fluctuation characteristics of the node corresponding to the out-of-bounds location and the response stability performance within the tag identification cycle, it evaluates the interference risk level of the out-of-bounds signal environment, obtains the credibility of the out-of-bounds status identification, identifies abnormal out-of-bounds status and sends abnormal behavior warning instructions, including starting GPS and locking the bicycle instructions, and establishes behavior control response results.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are:

[0042] In the present invention, through the analysis of traffic behavior, a high degree of matching between communication configuration and environmental characteristics is achieved, the degree of channel conflict is dynamically identified, the stability of channel access response is optimized, the node state change trend and spatial position relationship are combined to enhance the continuity control of identification compensation, the analysis of the integrity of the tag identification chain, and the accuracy of abnormal behavior detection are improved. Combined with the analysis of signal fluctuation and response stability, the boundary misjudgment rate is reduced, and accurate identification of out-of-bounds status and control response linkage are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0044] Figure 2 Acquisition flow chart for scenario communication configuration of the present invention;

[0045] Figure 3 Obtaining a flow chart for the channel response sequence of the present invention;

[0046] Figure 4 Obtaining a flow chart for the compensation scheduling configuration of the present invention;

[0047] Figure 5 A flowchart for obtaining abnormal behavior identification records of the present invention;

[0048] Figure 6 The present invention provides a flow chart for obtaining behavior control response results. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0051] See also Figure 1 The present invention provides a technical solution: a bicycle anti-lost tracking method based on radio frequency identification, comprising the following steps:

[0052] S1: Through tag recognition records, analyze the frequency of passage and stay in the identified area within multiple time periods, classify the identified area into types and output scene identifications. Combined with the matching combination between the area density level and the communication stability level, the scene communication configuration is generated;

[0053] S2: Call the scene communication configuration, analyze the number of active tags and the fluctuation amplitude of the communication channel identification signal, compare the activation tag density and signal fluctuation trend of each channel, calculate the matching ratio of tag distribution density and channel signal-to-noise ratio change, identify the degree of signal conflict, and obtain the channel response sequence;

[0054] S3: Based on the channel response sequence, analyze the tag recognition failures of each identification node in consecutive cycles, match the status identifier of the identification node, calculate the node's compensation recognition priority based on the spatial distance of adjacent nodes, and generate a compensation scheduling configuration;

[0055] S4: Call the compensation scheduling configuration to build a label identification path based on the spatial distribution of the identification nodes. By analyzing the number of consecutive records of labels between each pair of nodes in the path sequence, the interruption position of the identification path is extracted. Combined with the identification integrity of the path, abnormal jumping behavior is detected and an abnormal behavior identification record is established.

[0056] S5: Call abnormal behavior recognition records, use boundary nodes to detect bicycle out-of-bounds behavior, combine the signal fluctuation change characteristics of the recognition node corresponding to the out-of-bounds coordinates and the response stability performance of the tag during the recognition cycle, analyze the interference risk of the target location signal environment, output the out-of-bounds state recognition credibility, detect drift recognition events and eliminate them, and send abnormal behavior warning instructions to the bicycle corresponding to the abnormal out-of-bounds state, including starting GPS and locking the bicycle, and output the behavior control response result.

[0057] The scenario communication configuration includes area classification labels, frequency band selection groups, and identification channel identifiers. The channel response sequence is specifically a label distribution sorting group, a channel interference sorting list, and a response priority mapping table. The compensation scheduling configuration includes a node failure identifier, a spatial position chain group, and a compensation priority table. The abnormal behavior identification record is specifically an identification broken link position set, a jump event annotation table, and a path stability mapping set. The behavior control response results include positioning start status, abnormal locking command, and out-of-bounds alarm information.

[0058] See also Figure 2 , the specific steps for obtaining scene communication configuration are:

[0059] S111: Obtain the frequency of bicycle traffic and the frequency of bicycle stops in multiple time periods through tag identification records, analyze the distribution of the number of bicycles passing through and the length of stay in the target area in multiple time periods, and establish a regional behavior feature set;

[0060] The response of bicycle tags is monitored by radio frequency readers installed in bicycle parking areas. Data on the frequency of traffic and the frequency of dwelling are collected for the morning, afternoon, and evening time periods. The number of identification responses and the dwell period of each response are obtained for each tag. By calling the number of responses and dwell period of each tag in the corresponding time period, the frequency of traffic and the frequency of dwelling in the corresponding time period are calculated, and the behavioral feature set of the area is set. For example, the response data of the tag numbered RF001 in the three time periods are recorded as shown in the following table:

[0061] Table 1 Label response monitoring data table

[0062] Tag number Time Number of responses (times) Single dwell period (s) RF001 morning rush hour 5 60 RF001 noon 2 240 RF001 Evening rush hour 4 120

[0063] As shown in Table 1, the data of the RF001 tag during the morning peak period is called, in which the pass frequency is 5 times and the stop frequency is 60 seconds each time. The data of the noon period is called, in which the pass frequency is 2 times and the stop frequency is 240 seconds. The data of the evening peak period is called, in which the pass frequency is 4 times and the stop frequency is 120 seconds. All the time period data of each tag are called and calculated item by item to obtain the time period distribution data of each tag. By further calling the data of all tags in the area for statistical superposition, a set of regional behavior characteristics is generated.

[0064] S112: Analyze the activity level and stay concentration of the target area in multiple time periods based on the set of regional behavioral characteristics, identify regional usage characteristics, classify and label the regional types, and output scene identification information;

[0065] Based on the passage frequency and stay period data of each tag in the set, the activity level and stay concentration of each time period are calculated separately; when analyzing the activity level, the number of responses per hour is called to determine the activity level, where the number of responses per hour in the range of 0-2 times is judged as low activity, the number of responses in the range of 3-5 times is judged as medium activity, and the number of responses above 6 times is judged as high activity; when analyzing the stay concentration, the single stay period is called, and the stay period less than 100 seconds is low concentration, 100-200 seconds is medium concentration, and greater than 200 seconds is high concentration; for example, through actual data calculation, the average number of responses per hour in area A during the morning rush hour is 5 times, so the activity level is medium; the average single stay period is 90 seconds, so the concentration is low. This method is used to classify the areas and mark the area types as passage type or stay type, and finally output the scene identification.

[0066] S113: Calling the scene identification information, matching the communication frequency band group corresponding to the tag according to the corresponding area density level and communication stability level, and generating the scene communication configuration;

[0067] The population density level and communication link quality level corresponding to the area type are matched. By calling the population density level of the transit area and the communication link quality level of the resident area in the area type, the population density level and link quality level benchmark values ​​are set respectively. The population density level benchmark value is set by the number of tag responses per square meter. The number of tags per square meter is determined to be high density, 4-7 is determined to be medium density, and 0-3 is determined to be low density. The communication link quality level is set by calling the signal-to-noise ratio (SNR) of the wireless channel. An SNR greater than 30dB is determined to be high quality, 20-30dB is determined to be medium quality, and less than 20dB is determined to be low quality. For example, in area B, by monitoring the number of tags per square meter and calling and calculating the measured SNR value of 25dB, the population density level is determined to be medium density and the link quality level is determined to be medium quality. Then, based on the matching combination of the population density level and the link quality level, the corresponding communication frequency band is matched and set to the UHF band (860-960MHz). In this way, the scene communication configuration is generated.

[0068] See also Figure 3 , the specific steps for obtaining the channel response sequence are:

[0069] S211: Call the scene communication configuration, analyze the tag identification records and signal strength records of each communication channel in the current identification area, obtain the number of active tags and the signal strength change amplitude of each channel within a unit period, and generate a tag and signal fluctuation indicator group;

[0070] The reader / writer installed in the identification area collects tag identification records and signal strength records. The number of active tags and the signal strength variation within a unit cycle are recorded for each communication channel. The number of active tags is obtained by directly counting the number of tag responses within a unit cycle. For example, in communication channels C1 and C2 in a certain area, the number of active tags collected is 12 and 18 times, respectively. The signal strength variation is calculated based on the difference between the highest and lowest received signal intensity (RSSI) values ​​of the channel within the cycle. For example, the highest RSSI in C1 is -55dBm and the lowest is -75dBm, and the RSSI variation is calculated to be |-55-(-75)|=20dB. The highest RSSI in C2 is -50dBm and the lowest is -65dBm, and the RSSI variation is calculated to be |-50-(-65)|=15dB. Based on the above example, data is collected for all communication channels in the identification area, and specific monitoring data on the number of active tags and signal strength variation of each channel are obtained, as shown in Table 2. Finally, a tag and signal fluctuation indicator group is generated.

[0071] Table 2 Communication channel monitoring data table

[0072]

[0073] As shown in Table 2, the number of active tags and signal strength change amplitude data of each communication channel constitute the tag and signal fluctuation indicator group.

[0074] S212: Based on the tag and signal fluctuation index group, compare the tag quantity density and signal fluctuation amplitude change trend of each communication channel, and calculate the matching relationship between the tag distribution density and the channel signal-to-noise ratio change using the formula:

[0075]

[0076] Calculate communication interference matching coefficient;

[0077] Among them, R k is the communication interference matching coefficient, ρ i is the label activity density of the i-th channel, is the normalized value of the signal fluctuation amplitude of the i-th channel, Δ i is the number of cycles of fluctuation in the i-th channel, is the normalized value of the average signal amplitude of the i-th channel, is the normalized value of the interruption frequency of the i-th channel, is the normalized value of the noise sampling range of the i-th channel, n is the total number of communication channels, and i is the index number of the communication channel;

[0078] According to the tag and signal fluctuation index group, the matching relationship between tag distribution density and channel signal-to-noise ratio change is calculated in detail. The tag distribution density is determined by the number of tag responses within the unit recognition area. For example, the recognition area is 50m 2 , the number of tags in communication channel C1 is 12, then ρ1=12 / 50=0.24 tags / m 2 , the number of C2 tags is 18, then ρ2=18 / 50=0.36 / m 2 The channel signal-to-noise ratio (SNR) change is determined by the maximum and minimum differences of the SNR values ​​measured multiple times within a cycle. For example, the maximum SNR of C1 is 32dB, the minimum is 25dB, and the variation is 32-25=7dB. The maximum SNR of C2 is 28dB, the minimum is 22dB, and the variation is 28-22=6dB. The above parameter data is used to calculate the matching ratio. The specific formula is:

[0079]

[0080] In the formula, R k is the communication interference matching coefficient, which represents the matching relationship between channel label density and signal fluctuation; ρ i is the label distribution density of the i-th channel; is the normalized value of the signal strength change of the i-th channel, such as 20dB of C1 and 15dB of C2, which are calculated to be 0.8 and 0.6 after normalization; Δ i is the number of cycles of fluctuation in the i-th channel, with the measured values ​​of Δ1=4 and Δ2=3; is the normalized value of the average signal amplitude, the measured value 0.7; is the normalized value of the interruption frequency, the measured value is the normalized value of the noise sampling range, the measured value Calculate based on the above data:

[0081]

[0082] Detailed calculation process:

[0083]

[0084] Among them, the communication interference matching coefficient is used to reflect the comprehensive conflict index of the communication channel under the combined effects of different signal fluctuation intensities, tag number density and channel instability factors during the tag identification process. It is a normalized dimensionless indicator. The larger the value, the stronger the interference and the higher the recognition instability when the unit channel carries the tag identification task in a signal disturbance environment. There is a risk of potential recognition failure, conflict or response conflict. It is used in scenarios such as sorting multi-channel priorities, judging the severity of congestion, and dynamic scheduling and screening of auxiliary channels. It is a key intermediate calculation indicator for realizing the coupled judgment of communication load and signal quality. The communication interference matching coefficient provides a quantitative basis for channel response sorting, and effectively supports subsequent actions such as node activation selection, tag access diversion and identification frequency band configuration. The result shows that the communication interference matching coefficient is 0.8116, and the communication interference matching coefficient is obtained.

[0085] S213: Identify the signal conflict degree of each communication channel based on the communication interference matching coefficient, calculate the priority of each communication channel, and establish a channel response sorting sequence;

[0086] The degree of signal conflict is determined by calling the communication interference matching coefficient value calculated for each communication channel. The specific process is: when the communication interference matching coefficient is in the range of 0.8 to 1.0, it is determined that the signal conflict degree is high; in the range of 0.5 to 0.79, it is determined that the signal conflict degree is medium; below 0.5, it is determined that the signal conflict degree is low; according to the communication interference matching coefficients calculated for the aforementioned C1 and C2 channels, C1 is 0.78, which is determined to be a medium degree of conflict, and C2 is 0.81, which is determined to be a high degree of conflict; when calculating the priority of the communication channel, the communication interference matching coefficient is called, and the channels are sorted according to the principle that the lower the value, the higher the priority. For example, according to the aforementioned calculation results, C1 has a higher priority than C2, and the channel response sorting sequence is C1 and C2, and the channel response sorting sequence is obtained.

[0087] See also Figure 4 , the steps to obtain the compensation scheduling configuration are as follows:

[0088] S311: Collect identification records of each identification node in multiple consecutive cycles according to the channel response sorting sequence, analyze the number of tag identification failures recorded by each node in the consecutive cycles, and analyze the direction of failure change in the consecutive cycles to generate a node status identification set;

[0089] According to the channel response ranking of each identification node in the identification area, the label identification records of each node are collected in continuous cycles, and the node record data is specifically called; the number of label identification failures is directly calculated by statistically analyzing the label events that failed to successfully return a response in the node cycle. For example, the number of label identification failures collected by identification node A in continuous cycles T1, T2, and T3 are 3, 4, and 6 times respectively. After calling the data, the set of identification failures of node A in continuous cycles is {3, 4, 6}; the direction of failure change is obtained by calculating the difference of the number of failures in each cycle. For example, for node A, by calling the set data and calculating the difference of the number of failures between cycles, the difference set is {T2-T1=1, T3-T2=2}, and it is judged that the direction of change of the number of identification failures of node A in continuous cycles is increasing; other nodes in the area are calculated one by one in the same way.

[0090] Table 3 Node continuous cycle identification failure record table

[0091] Node number Number of failures in cycle T1 Number of failures in cycle T2 Number of failures in cycle T3 Direction of change A 3 4 6 Increment B 5 4 2 Decreasing C 2 2 3 Stablize

[0092] As shown in Table 3, the node status identification sets are node A identified as "continuous failure increase", node B identified as "continuous failure decrease", and node C identified as "stable failure state". The node status identification set is obtained by processing the data records of all identified nodes in the area one by one.

[0093] S312: Calling the node state identification set, combining the geometric coordinate data between the nodes, calculating the spatial distance between each identified node and the adjacent node, determining the jump between consecutive identified nodes in the identification chain structure, and establishing path structure continuity information;

[0094] By calling all the nodes identified in the set, the geometric coordinate data of the nodes in the area are further called, and the spatial position is represented by a two-dimensional rectangular coordinate system. For example, nodes A, B, and C are located at coordinates (5, 3), (9, 7), and (12, 11) respectively. When calling and calculating the spatial distance, the Euclidean distance between the coordinates of two adjacent nodes is obtained; for example, the distance calculation formula between adjacent nodes A and B is: Similarly, the spatial distance between node B and node C is calculated as: The spatial distance between all adjacent nodes is calculated in the above manner, the distance value is compared with the node status identifier, the distance value between nodes is called, and the baseline value for determining the distance is set to 5 meters; when the calculated distance between nodes exceeds the baseline value, it is determined that there is a node jump in the identification chain structure; taking node A and node B as an example, the distance between them is 5.66 meters, which is greater than 5 meters, so it is determined that a node jump occurs at this location; after calculating the distances of all nodes and determining the jump situation, the path structure continuity information is formed.

[0095] S313: Based on the path structure continuity information, extract each node's recognition failure behavior, failure trend, spatial jump characteristics, and structural fracture risk, and analyze the recognition behavior intensity and structural impact using the formula:

[0096]

[0097] Calculate the priority score of each node during compensation scheduling, establish a compensation priority queue, and generate a compensation scheduling configuration;

[0098] Among them, P c is the priority score value of the node compensation scheduling, f j is the number of identification failures of the jth identification node in the current cycle, The average number of identification failures for all participating nodes, d j is the number of consecutive cycles in which the jth identified node is in a failed state, g j is the normalized spatial distance between the jth identified node and its adjacent structural fracture node, l j is the number of jump intervals between the interruption points in the identification chain path corresponding to the jth identification node, m is the total number of identification nodes involved in the calculation, and j is the position number of the identification node in the index sequence;

[0099] Call each node for detailed analysis of its recognition failure behavior, trend changes, spatial jumps, and structural fracture risks; calculate the node compensation scheduling priority score using the formula:

[0100]

[0101] In the formula, f j The number of identification failures for node j in the current cycle, is the average number of failures of all participating nodes. For example, if node A has 6 identification failures, B has 2, and C has 3, the average value is:

[0102]

[0103] d j is the number of consecutive node failure cycles. For example, node A is 3 cycles, node B is 2 cycles, and node C is 1 cycle. j is the normalized value of spatial distance, which is obtained by calculating the ratio of the actual distance between nodes to the maximum distance between nodes. If the maximum distance between nodes is 10 meters, the distance between node A is 5.66 meters, and its normalized value is g A =5.66 / 10=0.566; Node B is 5 meters, the normalized value is 0.5; Node C is 0 if no jump occurs; l j To identify the number of jump intervals for a chain interruption, there is one jump between nodes A and B, and l A =1, there is one jump between nodes B and C, l B =1, node C has no jump, then l C =0; taking node A as an example, the detailed calculation is as follows:

[0104]

[0105] Take node B as an example for detailed calculation:

[0106]

[0107] Take node C as an example:

[0108]

[0109] Table 4 Node compensation scheduling priority score table

[0110] Node number Number of failures Continuous cycle Distance normalization value Number of jumps Priority Scoring A 6 3 0.566 1 3.25 B 2 2 0.500 1 1.58 C 3 1 0.000 0 0.67

[0111] As shown in Table 4, a compensation priority score table is formed based on the above calculation results. The score values ​​are then used to sort and generate the compensation scheduling configuration. The results indicate that node A has the highest priority, followed by node B, and node C has the lowest. The priority score for node compensation scheduling is an important quantitative indicator of whether a node should be prioritized for activation or replacement to maintain the integrity of the identification chain under the current system operating state. Nodes with significant anomalies, prolonged failure duration, or located at critical locations where the spatial identification chain is broken are considered targets for priority activation or monitoring in the compensation mechanism. The actual effect of the score is reflected in the node scheduling order: nodes with higher scores are prioritized for inclusion in operations such as compensation path reconstruction, signal coverage adjustment, and identification chain closure. These nodes serve as a key input to support the subsequent execution logic of the system's compensation strategy. The priority score directly determines the order of nodes in the compensation scheduling queue.

[0112] See also Figure 5 ,The specific steps for obtaining abnormal behavior identification records are:

[0113] S411: Call the compensation scheduling configuration, build a label identification path based on the spatial distribution of the identification nodes, and establish a label path sequence index set;

[0114] According to the two-dimensional coordinate distribution of the identified nodes in the actual parking lot space, nodes with high compensation scheduling priority are called one by one as the starting point of the path, and the path is constructed by calculating the geometric space position distance between adjacent nodes. Taking the parking lot identification nodes as an example, if the coordinates of node P1 are (2, 3), the coordinates of node P2 are (5, 8), and the coordinates of node P3 are (10, 12), the node coordinates are called one by one, and the distance between adjacent nodes is calculated one by one using the Euclidean distance formula. For example, the distance between node P1 and node P2 is calculated as:

[0115] The distance between nodes P2 and P3 is calculated as: After calling and calculating the spatial distance data between all nodes, the label identification path formed by the serial connection of each node is determined to be path 1: P1→P2→P3. Then the same operation is performed on all nodes to finally establish a label path sequence index set.

[0116] S412: Based on the label path sequence index set, by analyzing the number of consecutive records of labels between each pair of nodes in the path sequence, the interruption position in the path is identified and marked as a jump segment. The jump start and end node indexes, index span, response loss time and node spatial distance are recorded to obtain the jump segment feature set;

[0117] Each pair of adjacent nodes in the path sequence index is called one by one, and the number of consecutive records of label recognition between each node is calculated and analyzed. The number of consecutive records is obtained by counting the number of times the node successfully recognizes the label within the recognition cycle. For example, the number of consecutive successful label recognitions in the cycle between nodes P1 and P2 is 10 times, and the number of successful label recognitions in the cycle between nodes P2 and P3 is 2 times. The baseline value of the number of consecutive successful label recognitions is set to 5 times. The number of consecutive successful recognitions between nodes is called to determine whether it is lower than the baseline value. For example, the number of recognitions from nodes P2 to P3 is only 2 times, which is lower than the baseline value. The path is determined to be interrupted and recorded as a jump segment. The spatial coordinates of the nodes corresponding to the jump segment are called, and their spatial distance is calculated. The response loss time, that is, the total length of time the label is not recognized, is called and recorded. Through the above operations, a jump segment feature set is obtained, which records feature data such as the jump start and end node indexes (such as P2→P3), index span (1), response loss time (such as 30s), and spatial distance (such as 6.40m).

[0118] S413: Call the jump segment feature set, normalize the response missing time and spatial distance, and combine it with the path span, using the formula:

[0119]

[0120] Calculate the jump behavior anomaly score, combine it with the label's recognition coverage completeness in the path, detect abnormal jump behavior, and establish abnormal behavior identification records;

[0121] Among them, A is the abnormal jumping behavior score, is the normalized response missing time of the k-th jump segment, is the normalized spatial distance, s k is the path index span, q is the number of jump segments, and k is the jump segment number;

[0122] Normalization is performed by calling the response missing time and spatial distance of each jump segment. The normalization of the response missing time is determined by calculating the ratio of the missing time to the preset missing time upper limit, and the spatial distance normalization is determined by the ratio of the actual distance to the maximum inter-node distance. For example, if the preset upper limit of the response missing time is 60 seconds and the maximum spatial distance is 10 meters, taking the P2→P3 jump segment as an example, the missing time is called for 30 seconds, then the normalized response missing time is calculated as:

[0123]

[0124] The calling space distance is 6.40 meters, and the calculated space distance is normalized to:

[0125]

[0126] At the same time, the path index span s of the jump segment is called k , the span of the P2→P3 segment is 1; call the normalization parameter and calculate the jumping behavior abnormality score value A using the following formula:

[0127]

[0128] In the formula, A is the abnormal jumping behavior score, which represents the degree of jumping behavior. is the normalized response missing time of the k-th jump segment, is the normalized spatial distance, s k is the path index span, and q is the number of jump segments. A detailed calculation is performed using the P2→P3 jump segment as an example:

[0129]

[0130] Among them, the jump behavior anomaly score is an indicator to measure the severity of the non-continuous recognition behavior of the tag along the entire path. It aggregates the impact score of each jump segment and is a quantitative expression of the stability and integrity of the recognition path. The larger the score, the more jumps the tag has in the recognition path, and these jumps have abnormal characteristics in time, space and structure. It is used as an abnormality judgment threshold input in the subsequent recognition process, such as setting the behavior anomaly recognition threshold to screen whether it is necessary to trigger alarms, lock the car, locate and other control operations, and make quantitative comparisons of recognition behaviors under different tags and different paths to provide support for behavior classification and security strategies. There are multiple jump segments in the set area at the same time, as shown in Table 5:

[0131] Table 5 Jump segment characteristics and abnormality score table

[0132]

[0133] As shown in Table 5, the anomaly scores of all jump segments are calculated one by one using the above formula. The jump behavior anomaly score value is called to set the anomaly judgment benchmark value to 0.15. If the jump behavior anomaly score value is higher than the benchmark value of 0.15, the jump behavior is judged to be abnormal. Taking the P2→P3 segment as an example, the calculated anomaly score value is 0.20, which is higher than the benchmark value of 0.15, and the P2→P3 jump behavior is judged to be abnormal. Through the above operations, anomaly judgment analysis is performed on all jump segments, and the label path recognition coverage completeness is combined with the jump behavior judgment situation, and further sorted to establish an abnormal behavior recognition record. The results show that the anomaly score value can clearly judge the abnormality of the recognition path.

[0134] See also Figure 6 , the steps for obtaining the behavior control response result are as follows:

[0135] S511: Invoke abnormal behavior identification records to obtain a set of authorized area coordinates consisting of boundary nodes. Detect bicycle out-of-bounds behavior based on the spatial relative relationship between the real-time position of each tag and the boundary coordinates, and obtain a set of out-of-bounds status identifiers.

[0136] The authorized area coordinate set is constructed by calling the position coordinates of the nodes deployed on the parking area boundary, and the boundary nodes of the authorized area are represented in the form of two-dimensional plane coordinates. For example, the parking lot boundary consists of nodes A (0, 0), B (0, 50), C (50, 50), and D (50, 0). The above boundary coordinates are called and a closed area is formed by a closed boundary line to construct the authorized area coordinate set. The real-time position coordinates of each tag are further called. For example, the tag X coordinate is (52, 10) and the tag Y coordinate is (30, 45). The position relationship between the point and the boundary polygon is used to determine whether it has crossed the boundary. The specific process is as follows: Taking tag X as an example, the real-time coordinates of the tag (52, 10) are called through the point-polygon position relationship calculation method. It is determined that the coordinates are outside the authorized area boundary line and the tag X is marked as out of bounds. The tag Y coordinate (30, 45) is within the boundary and has not crossed the boundary. The real-time position determination of all tags is completed in the same way. This method is used to determine the real-time position of all tags one by one and obtain the out-of-bounds status identification set.

[0137] S512: Based on the out-of-bounds status identifier set, the channel signal strength sequence and fluctuation amplitude characteristics of the corresponding identification node are extracted, the response stability performance of the tag in the current identification cycle is calculated, the interference risk of the target location signal environment is analyzed, the out-of-bounds status identification reliability is output, the drift identification events are detected and eliminated, and the valid out-of-bounds behavior set is obtained;

[0138] The channel signal strength sequence of the identification node corresponding to the tag marked as out of bounds is called, and the signal strength fluctuation amplitude is analyzed through the actual channel RSSI data within the period. The specific calculation method is: the difference between the maximum and minimum signal strength values ​​at each moment in the signal strength sequence is called to calculate the fluctuation amplitude. For example, the RSSI data collected by node K in period T is {-60dBm, -63dBm, -58dBm, -65dBm}, and the calculated signal fluctuation amplitude is |-58-(-65)|=7dB; the number of successful responses of the tag in the identification period is called to calculate the response stability performance. The response stability is specifically calculated as the ratio of the number of successful responses to the total number of identifications in the period, such as the total number of identifications in the period. 10 times, 8 successes, then the stability performance is 8 / 10=0.8; further analyze the signal environment interference risk of the target location, and judge the interference risk to be high by the signal fluctuation amplitude greater than 5dB, and judge the interference risk to be low by less than or equal to 5dB; the fluctuation amplitude of the above-mentioned node K is 7dB, which is greater than 5dB, and the interference risk is high; call the tag response stability performance, set the credibility threshold to 0.7, and judge the response stability to be high when it is higher than the threshold 0.7, otherwise it is judged to be low credibility; the above-mentioned tag stability 0.8 is higher than 0.7, then the out-of-bounds state recognition credibility is high, and it is not judged as a drift event; use the same method to call the tag data analysis one by one; as shown in Table 6, the calculation data of the actual example are listed.

[0139] Table 6 Cross-border behavior signal analysis data table

[0140]

[0141] As shown in Table 6, the label drift is judged one by one by the above method to obtain the valid cross-border behavior set.

[0142] S513: Based on the valid out-of-bounds behavior set, identify the target tag number corresponding to the abnormal out-of-bounds state, send an abnormal behavior warning instruction and activate the remote control module, including starting GPS and locking the vehicle instructions, and establish a behavior control response result;

[0143] By calling a tag number marked as valid out-of-bounds within the collection, such as tag X, the vehicle information corresponding to the tag number is retrieved and matched with the vehicle information in the vehicle management database in combination with the tag's real-time location. The remote control unit is called to send an abnormal behavior warning instruction to the vehicle, which is sent via the data communication network to the control device installed on the vehicle. The remote control unit includes a GPS activation instruction and a vehicle lock instruction. The specific sending process is as follows: calling the vehicle control unit communication interface corresponding to tag X, sending a GPS start instruction via wireless communication, enabling the GPS device to activate real-time positioning, and then further sending a vehicle lock instruction after successful communication; confirming the execution of the instruction through data feedback. If confirmation feedback is received from the vehicle, the successful transmission status of the communication instruction is recorded; and finally, the transmission status is recorded. The transmission status of each instruction of the remote control unit constitutes the behavior control response result.

[0144] A bicycle anti-lost tracking system based on radio frequency identification is used to implement the above-mentioned bicycle anti-lost tracking method based on radio frequency identification. The system includes:

[0145] The regional scene calibration module analyzes the frequency of passage and dwell time in the identified area within multiple time periods based on tag identification records, identifies regional features and obtains scene identification. It then combines the matching between the regional density level and the communication stability level to obtain the scene communication configuration.

[0146] The channel dynamic optimization module analyzes the number of active tags in the target identification area and the fluctuation amplitude of the communication channel identification signal based on the scene communication configuration. It compares the activation tag density and signal fluctuation trend of each channel, calculates the matching ratio of tag distribution density and channel signal-to-noise ratio change, identifies the degree of signal conflict, sorts the response priority of the channel, and generates a channel response sequence.

[0147] The node status classification module extracts the number of tag recognition failures of each identification node in consecutive cycles based on the channel response sequence. It analyzes the stability status of the node based on the recognition failure trend, calculates the priority of the node compensation demand based on the spatial distance relationship between adjacent nodes, and establishes the compensation scheduling configuration.

[0148] The path continuity detection module constructs a path sequence to identify the spatial location of nodes and analyzes the continuity of label identification based on the compensation scheduling configuration. It detects the interruption position of node records in the label identification path, combines the identification integrity of the label in the path, detects abnormal jumping behavior, and establishes abnormal behavior identification records.

[0149] The out-of-bounds trust assessment module is based on abnormal behavior recognition records and uses boundary nodes to detect bicycle out-of-bounds behavior. By analyzing the signal fluctuation characteristics of the node corresponding to the out-of-bounds location and the response stability performance within the tag recognition cycle, it evaluates the interference risk level of the out-of-bounds signal environment, obtains the credibility of out-of-bounds status recognition, identifies abnormal out-of-bounds status and sends abnormal behavior warning instructions, including starting GPS and locking the bicycle, and establishes behavior control response results.

[0150] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A bicycle anti-lost tracking method based on radio frequency identification, characterized in that: The following steps are involved: S1: Through tag recognition records, analyze the frequency of passage and stay in the identified area within multiple time periods, classify the identified area into types and output scene identifications. Combined with the matching combination between the area density level and the communication stability level, the scene communication configuration is generated; S2: Call the communication configuration of the scenario, analyze the number of active tags and the fluctuation amplitude of the communication channel identification signal, compare the activation tag density and signal fluctuation trend of each channel, calculate the matching ratio of tag distribution density and channel signal-to-noise ratio change, identify the degree of signal conflict, and obtain the channel response sequence; S3: Analyze the tag recognition failure of each identification node in consecutive cycles according to the channel response sequence, match the state identifier of the identification node, calculate the compensation recognition priority of the node in combination with the spatial distance of the adjacent nodes, and generate a compensation scheduling configuration; S4: Call the compensation scheduling configuration, build a label identification path according to the spatial distribution of the identification nodes, extract the interruption position of the identification path by analyzing the number of consecutive records of the label between each pair of nodes in the path sequence, and detect abnormal jumping behavior in combination with the identification integrity of the path, and establish an abnormal behavior identification record.

2. The bicycle anti-lost tracking method based on radio frequency identification according to claim 1 is characterized in that: The scenario communication configuration includes an area classification label, a frequency band selection group, and an identification channel identifier. The channel response sequence is specifically a label distribution sorting group, a channel interference sorting list, and a response priority mapping table. The compensation scheduling configuration includes a node failure identifier, a spatial position chain group, and a compensation priority table. The abnormal behavior identification record is specifically an identification broken link position set, a jump event annotation table, and a path stability mapping set.

3. The bicycle anti-lost tracking method based on radio frequency identification according to claim 2, characterized in that: The steps for obtaining the scene communication configuration are specifically as follows: S111: Obtain the frequency of bicycle traffic and the frequency of bicycle stops in multiple time periods through tag identification records, analyze the distribution of the number of bicycles passing through and the length of stay in the target area in multiple time periods, and establish a regional behavior feature set; S112: Analyze the activity level and stay concentration of the target area in multiple time periods based on the set of regional behavior features, identify regional usage features, classify and label the regional types, and output scene identification information; S113: Call the scene identification information, match the communication frequency band group corresponding to the tag according to the corresponding area density level and communication stability level, and generate a scene communication configuration.

4. The bicycle anti-lost tracking method based on radio frequency identification according to claim 3 is characterized in that: The steps for obtaining the channel response sequence are specifically as follows: S211: Calling the scene communication configuration, analyzing the tag identification records and signal strength records of each communication channel in the current identification area, obtaining the number of active tags and the signal strength change amplitude of each channel within a unit period, and generating a tag and signal fluctuation indicator group; S212: Comparing the label quantity density and signal fluctuation amplitude change trend of each communication channel based on the label and signal fluctuation index group, calculating the matching relationship between the label distribution density and the channel signal-to-noise ratio change, and calculating the communication interference matching coefficient; S213: Identify the signal conflict degree of each communication channel according to the communication interference matching coefficient, calculate the priority of each communication channel, and establish a channel response sorting sequence.

5. The bicycle anti-lost tracking method based on radio frequency identification according to claim 4 is characterized in that: The steps for obtaining the compensation scheduling configuration are specifically as follows: S311: Collect identification records of each identification node in multiple consecutive cycles according to the channel response sorting sequence, analyze the number of tag identification failures recorded by each node in the consecutive cycles, and analyze the direction of failure change in the consecutive cycles to generate a node status identification set; S312: Calling the node status identification set, combining the geometric coordinate data between the nodes, calculating the spatial distance between each identified node and the adjacent node, determining the jump between consecutive identified nodes in the identification chain structure, and establishing path structure continuity information; S313: Based on the path structure continuity information, extract the recognition failure behavior, failure trend, spatial jump characteristics and structural fracture risk of each node, analyze the recognition behavior strength and structural impact, calculate the priority score value of each node during compensation scheduling, establish a compensation priority queue, and generate a compensation scheduling configuration.

6. The bicycle anti-lost tracking method based on radio frequency identification according to claim 5, characterized in that: The steps for obtaining the abnormal behavior identification record are specifically as follows: S411: calling the compensation scheduling configuration, constructing a label identification path according to the spatial distribution of the identification nodes, and establishing a label path sequence index set; S412: Based on the label path sequence index set, by analyzing the number of consecutive records of labels between each pair of nodes in the path sequence, the interruption position in the path is identified and marked as a jump segment. The jump start and end node indexes, index span, response loss time and node spatial distance are recorded to obtain a jump segment feature set; S413: Call the jump segment feature set, normalize the response missing time and spatial distance, and calculate the jump behavior abnormality score value based on the path span. Combined with the recognition coverage completeness of the label in the path, detect abnormal jump behavior and establish an abnormal behavior identification record.

7. The bicycle anti-lost tracking method based on radio frequency identification according to claim 6, characterized in that: The method further comprises: S5: Calling the abnormal behavior recognition record, using the boundary node to detect bicycle crossing the boundary, combining the signal fluctuation change characteristics of the recognition node corresponding to the crossing boundary coordinate and the response stability performance of the tag during the recognition cycle, analyzing the interference risk of the target location signal environment, outputting the cross-border state recognition credibility, detecting and eliminating drift recognition events, and sending abnormal behavior warning instructions to the bicycle corresponding to the abnormal cross-border state, including starting GPS and locking the bicycle, and outputting the behavior control response result; The behavior control response result includes positioning start status, abnormal locking command, and out-of-bounds alarm information.

8. The bicycle anti-lost tracking method based on radio frequency identification according to claim 7, characterized in that: The steps for obtaining the behavior control response result are specifically as follows: S511: Calling the abnormal behavior identification record, obtaining the authorized area coordinate set consisting of boundary nodes, detecting bicycle crossing the boundary based on the spatial relative relationship between the real-time position of each tag and the boundary coordinates, and obtaining a crossing state identifier set; S512: Based on the out-of-bounds status identifier set, extract the channel signal strength sequence and fluctuation amplitude characteristics of the corresponding identification node, calculate the response stability performance of the tag in the current identification cycle, analyze the interference risk of the target location signal environment, output the out-of-bounds status identification credibility, detect and eliminate drift identification events, and obtain a valid out-of-bounds behavior set; S513: According to the valid out-of-bounds behavior set, identify the target tag number corresponding to the abnormal out-of-bounds state, send an abnormal behavior warning instruction and activate the remote control module, including starting GPS and locking the car instructions, and establish a behavior control response result.

9. A bicycle anti-lost tracking system based on radio frequency identification, characterized in that: The system is used to implement the bicycle anti-loss tracking method based on radio frequency identification according to any one of claims 1 to 8, and the system includes: The regional scene calibration module analyzes the frequency of passage and dwell time in the identified area within multiple time periods based on tag identification records, identifies regional features and obtains scene identification. It then combines the matching between the regional density level and the communication stability level to obtain the scene communication configuration. The channel dynamic optimization module analyzes the number of active tags in the target identification area and the fluctuation amplitude of the communication channel identification signal based on the communication configuration of the scenario, compares the activation tag density and signal fluctuation trend of each channel, calculates the matching ratio of tag distribution density and channel signal-to-noise ratio change, identifies the degree of signal conflict, sorts the response priority of the channel, and generates a channel response sequence; The node status classification module extracts the number of tag recognition failures of each identification node in consecutive cycles based on the channel response sequence, analyzes the stability status of the node according to the recognition failure trend, calculates the priority of the node compensation demand according to the spatial distance relationship between adjacent nodes, and establishes a compensation scheduling configuration; The path continuity detection module constructs a path sequence to identify the spatial position of nodes and performs continuity analysis of label identification based on the compensation scheduling configuration, detects interruption positions of node records in the label identification path, detects abnormal jump behaviors based on the identification integrity of the labels in the path, and establishes abnormal behavior identification records; The out-of-bounds credibility assessment module is based on the abnormal behavior identification records and uses boundary nodes to detect bicycle out-of-bounds behavior. By analyzing the signal fluctuation characteristics of the node corresponding to the out-of-bounds location and the response stability performance within the tag identification cycle, it evaluates the interference risk level of the out-of-bounds signal environment, obtains the credibility of the out-of-bounds status identification, identifies abnormal out-of-bounds status and sends abnormal behavior warning instructions, including starting GPS and locking the bicycle instructions, and establishes behavior control response results.

Images