A method for detecting the liveness status of multiple users based on wearable RFID tags

Abstract

This invention discloses a multi-user liveness detection method based on wearable RFID tags, belonging to the field of detection technology. The key technical solution includes the following steps: collecting raw signal data of RFID tags worn by users within an area using an RFID reader; the raw signal data includes Received Signal Strength Indication (RSSI) and phase; preprocessing the raw signal data to obtain processed data; filtering the RSSI differential value and the processed phase to suppress noise; calculating a comprehensive feature value for each RFID tag based on the processed data to characterize signal fluctuations; the comprehensive feature value consists of a first feature component and a second feature component; using the maximum comprehensive feature value indicating no liveness as the detection threshold for determining the user's liveness; the effect is to quickly obtain the user's liveness status for the current period.

Description

Technical Field

[0001] This invention relates to the field of detection technology, and more specifically, to a method for detecting the survival status of multiple users based on wearable RFID tags. Background Technology

[0002] Human presence detection is crucial in many scenarios. For example, in smart home and smart healthcare applications, accurately detecting a user's presence in a room can provide convenient personalized services and offer real-time monitoring support for elderly health management. In disaster relief scenarios, emergency responders can quickly detect the presence of trapped individuals, buying valuable time for rescue efforts. Acoustic signal-based methods, however, are easily affected by environmental noise, exhibiting lower reliability, especially in complex environments. Furthermore, many studies have attempted to monitor user presence using radar or Wi-Fi devices, but these non-contact methods are difficult to extend to multi-user scenarios in practice. Most RFID-based systems are primarily used for deviceless behavior sensing or device-oriented target localization, with limited exploration of simultaneous multi-user presence detection. Traditional RFID signal analysis often relies solely on a single RSSI (Received Signal Strength Indication) parameter, but RSSI is susceptible to distance and environmental interference, making it difficult to accurately capture subtle signal changes caused by minute user movements (such as breathing or slight limb movements), resulting in low accuracy in determining the survival status of stationary users. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide a method for detecting the liveness status of multiple users based on wearable RFID tags.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A method for detecting the liveness status of multiple users based on wearable RFID tags, the method comprising the following steps: The raw signal data of RFID tags worn by users in the area is collected by an RFID reader. The raw signal data includes Received Signal Strength Indicator (RSSI) and phase. The original signal data is preprocessed to obtain processed data; the preprocessing includes calculating the RSSI difference value of adjacent time samples, performing modulo operation on the phase to limit it within a predetermined interval to obtain the processed phase; and filtering the RSSI difference value and the processed phase to suppress noise. Based on the processed data, a comprehensive feature value is calculated for each RFID tag to characterize signal fluctuations. The comprehensive feature value consists of a first feature component and a second feature component. The first feature component is the sum of the variance of the RSSI difference value and the variance of the processed phase within the detection period, and the second feature component is the fluctuation range of the processed phase within the detection period. The maximum comprehensive feature value of the non-live state is used as the detection threshold for determining the user's liveness status; The comprehensive characteristic value of the real-time monitoring process is compared with the detection threshold to determine whether the user is alive in the current period.

[0005] Preferably, the method of comparing the comprehensive feature value of the real-time monitoring process with the detection threshold to determine whether the user is alive in the current period specifically includes the following steps: The comprehensive feature value is compared with the detection threshold; If the comprehensive feature value is greater than the detection threshold, it is determined that the user corresponding to the RFID tag is alive in the current period, and the user corresponding to the node with suspicious survival status is verified. If the comprehensive feature value is less than or equal to the detection threshold, it is determined that the user corresponding to the RFID tag is in an inactive state in the current period.

[0006] Preferably, the user corresponding to the node with a suspicious liveness status is verified, specifically including the following steps: Obtain device parameters and user distribution data of wearable RFID tags, and construct an RFID signal monitoring architecture based on the device parameters and user distribution data. The RFID signal monitoring architecture contains several signal acquisition nodes, and each signal acquisition node corresponds to a user to be detected. The static signal data and dynamic response data of the RFID tag of the user to be tested are acquired through the signal acquisition node; The static signal data of the RFID tag is analyzed to obtain the static signal anomaly coefficient and static anomaly node of the user to be detected; the dynamic response data of the RFID tag is analyzed to obtain the dynamic response anomaly coefficient and dynamic anomaly node of the user to be detected. Static and dynamic abnormal nodes are used to construct nodes with suspected survival status; the survival suspicion of these nodes is obtained based on the static signal anomaly coefficient and the dynamic response anomaly coefficient. After analyzing the nodes with suspicious survival status and their survival suspicion level, the detection response priority is assigned. The signal processing terminal then verifies the users corresponding to the nodes with suspicious survival status based on the detection response priority and the status detection signal.

[0007] Preferably, the detection response is prioritized after analyzing the nodes with suspicious survival status and their survival suspicion level, specifically including the following steps: Based on the survival suspicion level, generate the signal warning strength of the survival suspicious node, and based on the signal warning strength, generate the status detection signal of the survival suspicious node; The signal propagation distance is obtained based on the surviving suspicious nodes and the signal convergence processing terminal. The detection response priority of the surviving suspicious nodes is obtained based on the signal warning strength and the signal propagation distance.

[0008] Preferably, the device parameters include the signal transmission power and signal coverage radius of the RFID tag; The user distribution data includes the number of users and the spatial distribution density of users.

[0009] Preferably, the static signal data of the RFID tag is analyzed to obtain the static signal anomaly coefficient and static anomaly nodes of the user to be detected, specifically including the following steps: RFID tag static signal data includes stable values ​​of tag signal strength and stable values ​​of tag signal frequency; The intensity anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal strength; the frequency anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal frequency. The static signal anomaly coefficient of the user to be detected is obtained based on the intensity anomaly coefficient and the frequency anomaly coefficient. The signal acquisition nodes corresponding to the static signal anomaly coefficients are marked as static anomaly nodes.

[0010] Preferably, the intensity anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal intensity, specifically including the following steps: The stable value of the tag signal strength includes the real-time signal strength value of each RFID tag collected by the signal acquisition node, and the signal strength stability curve corresponding to the user to be tested is generated based on the real-time signal strength value; If the real-time signal strength value of the signal strength stability curve exceeds the preset strength stability range threshold, the first strength anomaly coefficient of the user to be detected is obtained based on the difference between the real-time signal strength value and the strength stability range threshold. The slope of the intensity fluctuation of the signal intensity stability curve is obtained. If the slope of the intensity fluctuation is greater than the preset intensity fluctuation threshold, the second intensity anomaly coefficient of the user to be detected is obtained based on the difference between the slope of the intensity fluctuation and the intensity fluctuation threshold. The intensity anomaly coefficient of the user to be tested is obtained based on the first intensity anomaly coefficient and the second intensity anomaly coefficient.

[0011] Preferably, the frequency anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal frequency, specifically including the following steps: The stable value of the tag signal frequency includes the real-time signal frequency value of each RFID tag collected by the signal acquisition node, and the signal frequency stability curve corresponding to the user to be tested is generated based on the real-time signal frequency value. If the real-time signal frequency value in the signal frequency stability curve exceeds the preset frequency stability range threshold, the first frequency anomaly coefficient of the user to be detected is obtained based on the difference between the real-time signal frequency value and the frequency stability range threshold. The frequency fluctuation slope of the signal frequency stability curve is obtained. If the frequency fluctuation slope is greater than the preset frequency fluctuation threshold, the second frequency anomaly coefficient of the user to be detected is obtained based on the difference between the frequency fluctuation slope and the frequency fluctuation threshold. The frequency anomaly coefficient of the user to be detected is obtained based on the first frequency anomaly coefficient and the second frequency anomaly coefficient.

[0012] Preferably, the dynamic response anomaly coefficient and dynamic anomaly node of the user to be detected are obtained by analyzing the dynamic response data of the RFID tag, specifically including the following steps: The RFID tag dynamic response data includes the tag's response delay time to the excitation signal and the response signal attenuation rate. Based on the response delay time and the response signal attenuation rate, a dynamic response monitoring curve for the user to be detected is generated, and the slope of the response change of the dynamic response monitoring curve is obtained. If the slope of the response change is greater than the preset response change threshold, the dynamic response anomaly coefficient of the user to be detected is obtained based on the difference between the slope of the response change and the response change threshold, and the signal acquisition node corresponding to the user to be detected is marked as a dynamic anomaly node.

[0013] Preferably, the signal propagation distance is obtained based on the suspected surviving nodes and the signal convergence processing terminal, and the detection response priority of the suspected surviving nodes is obtained based on the signal warning strength and the signal propagation distance. Specifically, this includes the following steps: Obtain the transmission time of the status detection signal sent by the suspicious node with a live status, and obtain the reception time of the status detection signal received by the signal convergence and processing terminal; obtain the signal propagation duration based on the transmission and reception times; The signal propagation medium parameters of the monitoring environment are obtained, and the actual propagation speed of the status detection signal is obtained based on the signal propagation medium parameters. The signal propagation distance between the suspected live nodes corresponding to the status detection signal and the signal convergence and processing terminal is obtained based on the signal propagation duration and the actual propagation speed. The warning coefficient for a suspected node in surviving status is obtained based on the signal warning strength and its weight; the distance weight coefficient for a suspected node in surviving status is obtained based on the signal propagation distance and its weight. The priority evaluation coefficient of the suspicious nodes in the survival status is obtained based on the warning coefficient and the distance weight coefficient, and the detection response priority of the suspicious nodes in the survival status is set according to the priority evaluation coefficient.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This invention leverages the ability of RFID readers to collect user tag signals within a region, enabling the simultaneous acquisition of raw signal data from multiple users. This eliminates the need for individual user monitoring, adapting to multi-user scenarios and significantly improving detection efficiency and coverage. By calculating RSSI difference values ​​to highlight signal changes and performing modulo operations on the phase to standardize the data range, combined with filtering to suppress noise, it eliminates the impact of environmental interference on the raw signal while allowing signal characteristics to reflect user status, avoiding misjudgments caused by signal instability. The first feature component, combining RSSI difference and phase variance, comprehensively reflects the overall degree of signal fluctuation; the second feature component focuses on the phase fluctuation range, capturing subtle signal changes caused by minute movements when the user is stationary. This covers signal fluctuations from dynamic users as well as weak features from static users, making the determination of survival status more comprehensive and accurate. Using the maximum comprehensive feature value of the non-survival state as a threshold avoids misjudging user status due to signal interference from the environment itself, making the threshold setting more aligned with actual scenarios and improving detection reliability. By directly comparing the comprehensive feature value calculated in real time with the threshold, the user's current survival status can be obtained quickly, which reduces the cost of technology implementation and makes the status feedback faster, making it easier to detect anomalies and take measures in a timely manner. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating a multi-user liveness detection method based on wearable RFID tags proposed in this invention; Figure 2 This is a schematic diagram illustrating the steps of user verification in a multi-user liveness detection method based on wearable RFID tags proposed in this invention. Figure 3 This is a schematic diagram illustrating the steps for obtaining the detection response priority in a multi-user liveness detection method based on wearable RFID tags proposed in this invention. Detailed Implementation

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0018] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0019] Reference Figures 1-3 As shown.

[0020] The embodiments further illustrate the multi-user liveness detection method based on wearable RFID tags proposed in this invention.

[0021] A method for detecting the liveness status of multiple users based on wearable RFID tags, the method comprising the following steps: The raw signal data of RFID tags worn by users in the area is collected by an RFID reader. The raw signal data includes Received Signal Strength Indicator (RSSI) and phase. The original signal data is preprocessed to obtain processed data; the preprocessing includes calculating the RSSI difference value of adjacent time samples, performing modulo operation on the phase to limit it within a predetermined interval to obtain the processed phase; and filtering the RSSI difference value and the processed phase to suppress noise. Based on the processed data, a comprehensive feature value is calculated for each RFID tag to characterize signal fluctuations. The comprehensive feature value consists of a first feature component and a second feature component. The first feature component is the sum of the variance of the RSSI difference value and the variance of the processed phase within the detection period, and the second feature component is the fluctuation range of the processed phase within the detection period. The maximum comprehensive feature value of the non-live state is used as the detection threshold for determining the user's liveness status; The comprehensive characteristic value of the real-time monitoring process is compared with the detection threshold to determine whether the user is alive in the current period.

[0022] The RFID reader continuously collects raw signal data emitted by all users wearing wearable RFID tags within the monitoring area. The raw signal data includes Received Signal Strength Indicator (RSSI) and phase. The reader captures the real-time RSSI value and corresponding phase information transmitted from each user tag.

[0023] Preprocessing of the raw signal data involves calculating the RSSI difference between adjacent time samples, subtracting the RSSI value of the previous time from the RSSI value of the next time step to highlight the changes in signal strength rather than simply the magnitude of the strength. Phase modulus operation restricts phase values ​​that might otherwise exceed the range to a predetermined interval, such as the common 0 to 2π interval. The obtained RSSI difference value and the processed phase are then filtered to suppress noise interference in the environment.

[0024] To characterize signal fluctuations using quantitative metrics, each RFID tag has a corresponding comprehensive feature value, which is composed of a first feature component and a second feature component. The first feature component is the sum of the variance of the RSSI difference value and the variance of the processed phase within the detection period. The variance reflects the dispersion of the data, thus the first feature component reflects the overall fluctuation range of the RSSI difference and phase within the period. The second feature component is the fluctuation range of the processed phase within the detection period, obtained by subtracting the minimum value from the maximum value of the processed phase within the detection period. It captures the range of phase variation within the period. The comprehensive feature value is a combination of the first and second feature components, comprehensively reflecting the signal's fluctuation characteristics.

[0025] When the monitoring area is in a state of no activity, that is, when there is no user activity in the area, signal data is collected and the corresponding comprehensive feature value is calculated. Then, the maximum value in the comprehensive feature value is set as the detection threshold for determining the user's liveness status. The purpose of this step is to clarify the upper limit of signal fluctuation in a state of no activity.

[0026] During actual monitoring, the comprehensive feature value corresponding to each RFID tag is continuously calculated, and then the real-time comprehensive feature value is compared with a previously determined detection threshold. If the real-time comprehensive feature value is greater than the detection threshold, the user corresponding to that RFID tag is determined to be alive in the current period; if the real-time comprehensive feature value is less than or equal to the detection threshold, the user is determined to be in a non-living state in the current period. For example, when a user is stationary but breathing, the phase of their tag will fluctuate slightly, and the corresponding comprehensive feature value will exceed the threshold for the non-living state, thus being determined to be alive.

[0027] The system compares the comprehensive feature values ​​of the real-time monitoring process with the detection threshold to determine whether the user is alive in the current period. This process includes the following steps: The comprehensive feature value is compared with the detection threshold; If the comprehensive feature value is greater than the detection threshold, it is determined that the user corresponding to the RFID tag is alive in the current period, and the user corresponding to the node with suspicious survival status is verified. Suspicious survival nodes are signal acquisition nodes that exhibit abnormal signals after static signal and dynamic response analysis in the RFID multi-user survival detection process.

[0028] If the comprehensive feature value is less than or equal to the detection threshold, it is determined that the user corresponding to the RFID tag is in an inactive state in the current period.

[0029] Furthermore, users corresponding to nodes with suspicious liveness status are verified, specifically including the following steps: Obtain device parameters and user distribution data of wearable RFID tags, and construct an RFID signal monitoring architecture based on the device parameters and user distribution data. The RFID signal monitoring architecture contains several signal acquisition nodes, and each signal acquisition node corresponds to a user to be detected. The static signal data and dynamic response data of the RFID tag of the user to be tested are acquired through the signal acquisition node; After the monitoring architecture is built, static signal data and dynamic response data of the RFID tags of the corresponding users to be monitored are acquired through each signal acquisition node. The static signal data of the RFID tags reflects the characteristics of the tag signal in a stable state, such as the stable value of the signal strength and the stable value of the signal frequency. The dynamic response data of the RFID tags reflects the feedback of the tags to external excitation signals, such as the response delay time after the tag receives the excitation signal and the attenuation rate of the response signal. When a user is in the monitoring area, the RFID tag he wears will continuously send signals. The corresponding acquisition node will record the stable value of the tag signal strength and the stable value of the signal frequency in real time. At the same time, when an excitation command is sent to the tag, the node will record the time interval from the tag receiving the command to responding and the rate of change of the response signal from being sent to weakening.

[0030] The static signal data of the RFID tag is analyzed to obtain the static signal anomaly coefficient and static anomaly node of the user to be detected; the dynamic response data of the RFID tag is analyzed to obtain the dynamic response anomaly coefficient and dynamic anomaly node of the user to be detected. Static and dynamic abnormal nodes are used to construct nodes with suspected survival status; the survival suspicion of these nodes is obtained based on the static signal anomaly coefficient and the dynamic response anomaly coefficient. After analyzing the nodes with suspicious survival status and their survival suspicion level, the detection response priority is assigned. The signal processing terminal then verifies the users corresponding to the nodes with suspicious survival status based on the detection response priority and the status detection signal.

[0031] After analyzing the suspicious nodes' survival status and their survival suspicion level, the detection response is prioritized, specifically including the following steps: Based on the survival suspicion level, generate the signal warning strength of the survival suspicious node, and based on the signal warning strength, generate the status detection signal of the survival suspicious node; The signal propagation distance is obtained based on the surviving suspicious nodes and the signal convergence processing terminal. The detection response priority of the surviving suspicious nodes is obtained based on the signal warning strength and the signal propagation distance.

[0032] The signal alert strength for a node is determined based on its survival suspicion level. Survival suspicion level is a quantitative indicator obtained from the static signal anomaly coefficient and the dynamic response anomaly coefficient. A higher suspicion level indicates a greater risk of user status anomalies associated with that node, and consequently, a higher signal alert strength. For example, if the survival suspicion level of a certain node is significantly higher than that of other nodes, its signal alert strength is set to a high level. After determining the signal alert strength, a corresponding status detection signal is generated based on this strength. This status detection signal carries the node's anomaly information, providing a clear detection basis for subsequent verification processes.

[0033] The system obtains the transmission time of the status detection signal sent by a suspicious node and the reception time of the signal at the signal aggregation processing terminal. The difference between these two times is used to obtain the signal propagation duration. Simultaneously, the actual propagation speed of the status detection signal in the current environment is determined by combining the signal propagation medium parameters of the monitoring environment. The signal propagation distance = signal propagation duration × actual propagation speed, thus calculating the signal propagation distance between the suspicious node and the signal aggregation processing terminal. The detection response priority is determined by combining the signal warning strength and the signal propagation distance: nodes with higher signal warning strength require priority verification; while nodes with shorter signal propagation distances have lower signal transmission delays and relatively higher data accuracy, and are therefore assigned higher priority weights. All suspicious nodes are sorted to obtain the detection response priority, allowing the signal aggregation processing terminal to efficiently verify the status of users corresponding to suspicious nodes according to priority order.

[0034] Equipment parameters include the signal transmission power and signal coverage radius of the RFID tag; User distribution data includes the number of users and the spatial distribution density of users.

[0035] The analysis of static signal data from RFID tags yields the static signal anomaly coefficients and static anomaly nodes for the user under test. This process includes the following steps: RFID tag static signal data includes stable values ​​of tag signal strength and stable values ​​of tag signal frequency; The intensity anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal strength; the frequency anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal frequency. The static signal anomaly coefficient of the user to be detected is obtained based on the intensity anomaly coefficient and the frequency anomaly coefficient. The signal acquisition nodes corresponding to the static signal anomaly coefficients are marked as static anomaly nodes.

[0036] The static signal data of the RFID tag includes stable values ​​of the tag signal strength and stable values ​​of the tag signal frequency. Real-time signal strength values ​​of the RFID tags of the user under test are collected, and a corresponding signal strength stability curve is generated based on these real-time signal strength values. Then, a preset strength stability range threshold is set. If the real-time signal strength value of the curve exceeds the strength stability range threshold, a first strength anomaly coefficient is calculated based on the difference between the real-time value and the threshold. The slope of the signal strength stability curve is calculated; if this slope is greater than a preset strength fluctuation threshold, a second strength anomaly coefficient is obtained based on the difference between the slope and the threshold. The strength anomaly coefficient = first strength anomaly coefficient + second strength anomaly coefficient.

[0037] The system collects the real-time signal frequency values ​​of the tags to generate a signal frequency stability curve. A preset frequency stability range threshold is set. If the real-time frequency value exceeds the threshold, the first frequency anomaly coefficient is obtained based on the difference. The frequency fluctuation slope of the frequency stability curve is calculated. If the slope is greater than the preset frequency fluctuation threshold, the second frequency anomaly coefficient is obtained based on the difference. Finally, the frequency anomaly coefficient is obtained by summing the values. The frequency anomaly coefficient = the first frequency anomaly coefficient + the second frequency anomaly coefficient.

[0038] By combining the intensity anomaly coefficient and the frequency anomaly coefficient, the static signal anomaly coefficient = intensity anomaly coefficient + frequency anomaly coefficient. The static signal anomaly coefficient can quantify the overall degree of anomaly in the static signal of the RFID tag of the user to be tested.

[0039] Identify the signal acquisition node corresponding to the static signal anomaly coefficient and mark that signal acquisition node as a static anomaly node. For example, if the intensity anomaly coefficient and frequency anomaly coefficient of a user to be detected are both high, and the corresponding static signal anomaly coefficient exceeds the set standard, then the signal acquisition node corresponding to that user will be marked as a static anomaly node.

[0040] The intensity anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal strength. The specific steps include: The stable value of the tag signal strength includes the real-time signal strength value of each RFID tag collected by the signal acquisition node, and the signal strength stability curve corresponding to the user to be tested is generated based on the real-time signal strength value; If the real-time signal strength value of the signal strength stability curve exceeds the preset strength stability range threshold, the first strength anomaly coefficient of the user to be detected is obtained based on the difference between the real-time signal strength value and the strength stability range threshold. The slope of the intensity fluctuation of the signal intensity stability curve is obtained. If the slope of the intensity fluctuation is greater than the preset intensity fluctuation threshold, the second intensity anomaly coefficient of the user to be detected is obtained based on the difference between the slope of the intensity fluctuation and the intensity fluctuation threshold. The intensity anomaly coefficient of the user to be tested is obtained based on the first intensity anomaly coefficient and the second intensity anomaly coefficient.

[0041] The stable signal strength value of a tag is the real-time signal strength value of each RFID tag collected by the signal acquisition node. Based on the real-time signal strength value, a stable signal strength curve for the user under test is generated. This curve can intuitively show the trend of the user's tag signal strength over a period of time. For example, the real-time signal strength value of a user's tag will be continuously recorded within a monitoring period, thus forming a continuous strength curve.

[0042] The signal strength stability curve is used to determine whether the signal strength exceeds the stable range, and a first strength anomaly coefficient is calculated. A preset strength stability range threshold is used, which is determined based on the fluctuation range of the tag signal strength under normal conditions. If the real-time signal strength value of the signal strength stability curve exceeds the strength stability range threshold, the first strength anomaly coefficient of the user under test is obtained by subtracting the strength stability range threshold from the real-time signal strength value.

[0043] The slope of the intensity fluctuation of the signal strength stability curve is obtained, which reflects the rate of change of signal strength. A preset intensity fluctuation threshold is set. If the calculated intensity fluctuation slope is greater than the intensity fluctuation threshold, the intensity fluctuation slope is subtracted from the intensity fluctuation threshold to obtain the second intensity anomaly coefficient.

[0044] The intensity anomaly coefficient of the user to be detected is obtained by combining the first intensity anomaly coefficient and the second intensity anomaly coefficient. Intensity anomaly coefficient = First intensity anomaly coefficient + Second intensity anomaly coefficient. The intensity anomaly coefficient quantifies the degree of anomaly in the signal strength of the user's tag.

[0045] The frequency anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal frequency, specifically including the following steps: The stable value of the tag signal frequency includes the real-time signal frequency value of each RFID tag collected by the signal acquisition node, and the signal frequency stability curve corresponding to the user to be tested is generated based on the real-time signal frequency value. If the real-time signal frequency value in the signal frequency stability curve exceeds the preset frequency stability range threshold, the first frequency anomaly coefficient of the user to be detected is obtained based on the difference between the real-time signal frequency value and the frequency stability range threshold. The frequency fluctuation slope of the signal frequency stability curve is obtained. If the frequency fluctuation slope is greater than the preset frequency fluctuation threshold, the second frequency anomaly coefficient of the user to be detected is obtained based on the difference between the frequency fluctuation slope and the frequency fluctuation threshold. The frequency anomaly coefficient of the user to be detected is obtained based on the first frequency anomaly coefficient and the second frequency anomaly coefficient.

[0046] The stable value of the tag signal frequency is the real-time signal frequency value of each RFID tag collected by the signal acquisition node. Based on the real-time signal frequency value, a signal frequency stability curve for the user to be monitored is generated. The signal frequency stability curve shows the trend of the user's tag signal frequency change within the monitoring period. For example, the real-time frequency value of a user's tag is continuously recorded, thus forming a continuous frequency change curve.

[0047] A preset frequency stability range threshold is established, which represents the reasonable fluctuation range of the tag signal frequency under normal conditions. If the real-time signal frequency value of the signal frequency stability curve exceeds the frequency stability range threshold, the first frequency anomaly coefficient is obtained by subtracting the frequency stability range threshold from the real-time frequency value.

[0048] The slope of the signal frequency stability curve is determined, and the second frequency anomaly coefficient is calculated. The slope of the curve's frequency fluctuation is obtained, reflecting the rate of frequency change. A preset frequency fluctuation threshold is also set. If the slope exceeds the preset threshold, the second frequency anomaly coefficient is obtained by subtracting the threshold from the slope. For example, if the preset threshold is 3, and the slope of a certain segment of the curve is 6, then the second frequency anomaly coefficient is 3.

[0049] The frequency anomaly coefficient of the user under test is obtained by combining the first and second frequency anomaly coefficients. The frequency anomaly coefficient = first frequency anomaly coefficient + second frequency anomaly coefficient. The frequency anomaly coefficient can quantify the degree of anomaly in the frequency of the user's tag signal.

[0050] The analysis of RFID tag dynamic response data yields the dynamic response anomaly coefficient and dynamic anomaly nodes for the user to be detected. This process includes the following steps: The RFID tag dynamic response data includes the tag's response delay time to the excitation signal and the response signal attenuation rate. Based on the response delay time and the response signal attenuation rate, a dynamic response monitoring curve for the user to be detected is generated, and the slope of the response change of the dynamic response monitoring curve is obtained. If the slope of the response change is greater than the preset response change threshold, the dynamic response anomaly coefficient of the user to be detected is obtained based on the difference between the slope of the response change and the response change threshold, and the signal acquisition node corresponding to the user to be detected is marked as a dynamic anomaly node.

[0051] RFID tag dynamic response data includes the tag's response delay time to the excitation signal and the response signal decay rate. The response delay time refers to the time interval between the tag receiving the excitation signal from the system and issuing a signal response; the response signal decay rate is how quickly the strength of the tag's response signal weakens over time.

[0052] The response delay time and response signal attenuation rate of the RFID tags of the user under test are continuously recorded, and these data are correlated in chronological order to form a curve that reflects the dynamic response change trend of the tags. For example, within a monitoring period, the response delay time of the tag after receiving an excitation signal each time and the corresponding attenuation rate of the response signal are recorded sequentially and plotted into a continuous dynamic response monitoring curve.

[0053] Extract the slope of the dynamic response monitoring curve. The slope of the response change is a quantitative description of the trend of the curve, which can reflect the speed of change of the tag's dynamic response index. The larger the slope, the more drastic the fluctuation of the tag's dynamic response in a short period of time, and the higher the risk of anomaly.

[0054] The response change slope is compared with a preset response change threshold, and the dynamic response anomaly coefficient is calculated. The response change threshold is a reasonable critical value set based on the fluctuation range of the tag's dynamic response under normal conditions. If the extracted response change slope is greater than the response change threshold, the dynamic response anomaly coefficient of the user to be detected is obtained by subtracting the response change threshold from the response change slope: Dynamic response anomaly coefficient = Response change slope - Response change threshold. For example, if the preset response change threshold is 4, and the response change slope of a tag's dynamic response monitoring curve is 7, then the dynamic response anomaly coefficient of that user is 3. The signal acquisition node corresponding to this user is marked as a dynamic anomaly node by the system.

[0055] Based on the signal propagation distance obtained from the suspected surviving nodes and the signal convergence and processing terminal, and based on the signal warning strength and signal propagation distance, the detection response priority for the suspected surviving nodes is determined, specifically including the following steps: Obtain the transmission time of the status detection signal sent by the suspicious node with a live status, and obtain the reception time of the status detection signal received by the signal convergence and processing terminal; obtain the signal propagation duration based on the transmission and reception times; The signal propagation medium parameters of the monitoring environment are obtained, and the actual propagation speed of the status detection signal is obtained based on the signal propagation medium parameters. The signal propagation distance between the suspected live nodes corresponding to the status detection signal and the signal convergence and processing terminal is obtained based on the signal propagation duration and the actual propagation speed. The warning coefficient for a suspected node in surviving status is obtained based on the signal warning strength and its weight; the distance weight coefficient for a suspected node in surviving status is obtained based on the signal propagation distance and its weight. The priority evaluation coefficient of the suspicious nodes in the survival status is obtained based on the warning coefficient and the distance weight coefficient, and the detection response priority of the suspicious nodes in the survival status is set according to the priority evaluation coefficient.

[0056] The system obtains the transmission time of the status detection signal sent by a suspicious node in its live state, and simultaneously records the reception time of the signal at the signal aggregation processing terminal. The signal propagation time is obtained by subtracting the transmission time from the reception time. For example, if a suspicious node sends a signal at 10:00:00 and the terminal receives the signal at 10:00:02, the signal propagation time is 2 seconds.

[0057] The signal propagation medium parameters of the monitoring environment are obtained. These parameters include the type of medium in the environment, such as the distribution of air and walls. Different media affect the signal propagation speed, which determines the actual propagation speed of the status detection signal in the current environment. The signal propagation distance between the suspected node and the signal aggregation and processing terminal is obtained by multiplying the signal propagation duration by the actual propagation speed. Signal propagation distance = signal propagation duration × actual propagation speed. For example, if the signal propagation duration is 2 seconds and the actual propagation speed is 300 meters per second, then the signal propagation distance is 600 meters.

[0058] A warning intensity weight is set, which is a fixed value determined based on the importance of warning priority in the actual scenario. The warning coefficient for suspicious nodes is obtained by multiplying the signal warning intensity by the warning intensity weight: Warning Coefficient = Signal Warning Intensity × Warning Intensity Weight. Simultaneously, a propagation distance weight is set, generally with higher weight for closer distances. The distance weight coefficient is calculated using the weight coefficient corresponding to the signal propagation distance: Distance Weight Coefficient = Weight Conversion Value of Signal Propagation Distance × Propagation Distance Weight. The warning coefficient and the distance weight coefficient are added to obtain the priority evaluation coefficient: Priority Evaluation Coefficient = Warning Coefficient + Distance Weight Coefficient. All suspicious nodes are sorted according to their priority evaluation coefficients; nodes with higher coefficients have higher detection and response priority, and the terminal prioritizes verifying the status of users corresponding to these nodes.

[0059] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0060] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the liveness status of multiple users based on wearable RFID tags, characterized in that, The method includes the following steps: The raw signal data of RFID tags worn by users in the area is collected by an RFID reader. The raw signal data includes Received Signal Strength Indicator (RSSI) and phase. The original signal data is preprocessed to obtain processed data; the preprocessing includes calculating the RSSI difference value of adjacent time samples, performing modulo operation on the phase to limit it within a predetermined interval to obtain the processed phase; and filtering the RSSI difference value and the processed phase to suppress noise. Based on the processed data, a comprehensive feature value is calculated for each RFID tag to characterize signal fluctuations. The comprehensive feature value consists of a first feature component and a second feature component. The first feature component is the sum of the variance of the RSSI difference value and the variance of the processed phase within the detection period, and the second feature component is the fluctuation range of the processed phase within the detection period. The maximum comprehensive feature value of the non-live state is used as the detection threshold for determining the user's liveness status; The comprehensive characteristic value of the real-time monitoring process is compared with the detection threshold to determine whether the user is alive in the current period.

2. The method for detecting the liveness status of multiple users based on wearable RFID tags according to claim 1, characterized in that, The system compares the comprehensive feature values ​​of the real-time monitoring process with the detection threshold to determine whether the user is alive in the current period. This process includes the following steps: The comprehensive feature value is compared with the detection threshold; If the comprehensive feature value is greater than the detection threshold, it is determined that the user corresponding to the RFID tag is alive in the current period, and the user corresponding to the node with suspicious survival status is verified. If the comprehensive feature value is less than or equal to the detection threshold, it is determined that the user corresponding to the RFID tag is in an inactive state in the current period.

3. The method for detecting the survival status of multiple users based on wearable RFID tags according to claim 2, characterized in that, Furthermore, users corresponding to nodes with suspicious liveness status are verified, specifically including the following steps: Obtain device parameters and user distribution data of wearable RFID tags, and construct an RFID signal monitoring architecture based on the device parameters and user distribution data. The RFID signal monitoring architecture contains several signal acquisition nodes, and each signal acquisition node corresponds to a user to be detected. The static signal data and dynamic response data of the RFID tag of the user to be tested are acquired through the signal acquisition node; The static signal data of the RFID tag is analyzed to obtain the static signal anomaly coefficient and static anomaly node of the user to be detected; the dynamic response data of the RFID tag is analyzed to obtain the dynamic response anomaly coefficient and dynamic anomaly node of the user to be detected. Static and dynamic abnormal nodes are used to construct nodes with suspected survival status; the survival suspicion of these nodes is obtained based on the static signal anomaly coefficient and the dynamic response anomaly coefficient. After analyzing the nodes with suspicious survival status and their survival suspicion level, the detection response priority is assigned. The signal processing terminal then verifies the users corresponding to the nodes with suspicious survival status based on the detection response priority and the status detection signal.

4. The method for detecting the survival status of multiple users based on wearable RFID tags according to claim 3, characterized in that, After analyzing the suspicious nodes' survival status and their survival suspicion level, the detection response is prioritized, specifically including the following steps: Based on the survival suspicion level, generate the signal warning strength of the survival suspicious node, and based on the signal warning strength, generate the status detection signal of the survival suspicious node; The signal propagation distance is obtained based on the surviving suspicious nodes and the signal convergence processing terminal. The detection response priority of the surviving suspicious nodes is obtained based on the signal warning strength and the signal propagation distance.

5. The method for detecting the survival status of multiple users based on wearable RFID tags according to claim 3, characterized in that, The device parameters include the signal transmission power and signal coverage radius of the RFID tag; The user distribution data includes the number of users and the spatial distribution density of users.

6. The method for detecting the survival status of multiple users based on wearable RFID tags according to claim 5, characterized in that, The analysis of static signal data from RFID tags yields the static signal anomaly coefficients and static anomaly nodes for the user under test. This process includes the following steps: RFID tag static signal data includes stable values ​​of tag signal strength and stable values ​​of tag signal frequency; The intensity anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal strength; the frequency anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal frequency. The static signal anomaly coefficient of the user to be detected is obtained based on the intensity anomaly coefficient and the frequency anomaly coefficient. The signal acquisition nodes corresponding to the static signal anomaly coefficients are marked as static anomaly nodes.

7. The method for detecting the liveness status of multiple users based on wearable RFID tags according to claim 6, characterized in that, The intensity anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal strength. The specific steps include: The stable value of the tag signal strength includes the real-time signal strength value of each RFID tag collected by the signal acquisition node, and the signal strength stability curve corresponding to the user to be tested is generated based on the real-time signal strength value; If the real-time signal strength value of the signal strength stability curve exceeds the preset strength stability range threshold, the first strength anomaly coefficient of the user to be detected is obtained based on the difference between the real-time signal strength value and the strength stability range threshold. The slope of the intensity fluctuation of the signal intensity stability curve is obtained. If the slope of the intensity fluctuation is greater than the preset intensity fluctuation threshold, the second intensity anomaly coefficient of the user to be detected is obtained based on the difference between the slope of the intensity fluctuation and the intensity fluctuation threshold. The intensity anomaly coefficient of the user to be tested is obtained based on the first intensity anomaly coefficient and the second intensity anomaly coefficient.

8. The method for detecting the liveness status of multiple users based on wearable RFID tags according to claim 6, characterized in that, The frequency anomaly coefficient of the user to be detected is obtained by analyzing the stable value of the tag signal frequency, specifically including the following steps: The stable value of the tag signal frequency includes the real-time signal frequency value of each RFID tag collected by the signal acquisition node, and the signal frequency stability curve corresponding to the user to be tested is generated based on the real-time signal frequency value. If the real-time signal frequency value in the signal frequency stability curve exceeds the preset frequency stability range threshold, the first frequency anomaly coefficient of the user to be detected is obtained based on the difference between the real-time signal frequency value and the frequency stability range threshold. The frequency fluctuation slope of the signal frequency stability curve is obtained. If the frequency fluctuation slope is greater than the preset frequency fluctuation threshold, the second frequency anomaly coefficient of the user to be detected is obtained based on the difference between the frequency fluctuation slope and the frequency fluctuation threshold. The frequency anomaly coefficient of the user to be detected is obtained based on the first frequency anomaly coefficient and the second frequency anomaly coefficient.

9. A method for detecting the liveness status of multiple users based on wearable RFID tags according to claim 8, characterized in that, The analysis of RFID tag dynamic response data yields the dynamic response anomaly coefficient and dynamic anomaly nodes for the user to be detected. This process includes the following steps: The dynamic response data of RFID tags includes the tag's response delay time to the excitation signal and the response signal attenuation rate. Based on the response delay time and the response signal attenuation rate, a dynamic response monitoring curve of the user to be detected is generated, and the slope of the response change of the dynamic response monitoring curve is obtained. If the slope of the response change is greater than the preset response change threshold, the dynamic response anomaly coefficient of the user to be detected is obtained based on the difference between the slope of the response change and the response change threshold, and the signal acquisition node corresponding to the user to be detected is marked as a dynamic anomaly node.

10. A method for detecting the liveness status of multiple users based on wearable RFID tags according to claim 9, characterized in that, Based on the signal propagation distance obtained from the suspected surviving nodes and the signal convergence and processing terminal, and based on the signal warning strength and signal propagation distance, the detection response priority for the suspected surviving nodes is determined, specifically including the following steps: Obtain the transmission time of the status detection signal sent by the suspicious node with a live status, and obtain the reception time of the status detection signal received by the signal convergence processing terminal; obtain the signal propagation duration based on the transmission time and reception time; The signal propagation medium parameters of the monitoring environment are obtained, and the actual propagation speed of the status detection signal is obtained based on the signal propagation medium parameters. The signal propagation distance between the suspected live nodes corresponding to the status detection signal and the signal convergence and processing terminal is obtained based on the signal propagation duration and the actual propagation speed. The warning coefficient for a suspected node in surviving status is obtained based on the signal warning strength and its weight; the distance weight coefficient for a suspected node in surviving status is obtained based on the signal propagation distance and its weight. The priority evaluation coefficient of the suspicious nodes in the survival status is obtained based on the warning coefficient and the distance weight coefficient, and the detection response priority of the suspicious nodes in the survival status is set according to the priority evaluation coefficient.

Images