A zigbee human presence sensor automatic test system and method based on millimeter wave micro-motion simulation and multi-dimensional interference injection

Abstract

The application relates to the technical field of human body sensor detection, and discloses a Zigbee human body existence sensor automation test system and method based on millimeter wave micro-motion simulation and multi-dimensional interference injection, which comprises the following steps: step 1, initialization and Mesh networking, step 2, micro-motion sensitivity test, step 3, interference suppression performance test, step 4, Mesh networking collaborative test, step 5, 3D mapping and report output; through high-precision micro-motion simulation and multi-dimensional interference injection, the core performance test dimensions of the human body existence sensor can be covered, not only can the micro-motion detection be upgraded from artificial qualitative judgment to automatic quantitative evaluation, but also can various standard interference sources such as air flow disturbance, heat convection, electromagnetic interference and pet passing be included, and the anti-interference ability of the sensor in different composite scenes can be verified in depth, so that the test result can more accurately reflect the real performance of the product in the actual deployment environment, and a scientific basis can be provided for product research and development and mass production calibration.

Description

Technical Field

[0001] This invention relates to the field of human body sensor detection technology, and more specifically discloses an automated testing system and method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection. Background Technology

[0002] With the rapid development of smart home and building automation technologies, human presence sensors based on millimeter-wave radar and ToF (Time-of-Flight) technology have gradually replaced traditional PIR sensors and become core sensing devices in scenarios such as smart lighting, HVAC (heating, ventilation and air conditioning) control, and security monitoring, thanks to their excellent detection accuracy and environmental adaptability.

[0003] In the prior art, document CN121878668A discloses a "robust human presence perception method and system based on millimeter-wave radar under complex interference environment". This system uses millimeter-wave radar as the core sensor and, through signal processing, including adaptive background modeling and coherent subtraction based on complex range spectrum, cleanliness index constructed by extracting multi-dimensional features from micro-Doppler spectrum, and dual-channel parallel detection mechanism for moving and stationary targets, effectively overcomes the core challenges such as difficulty in static presence detection, sensitivity to environmental interference, and insufficient system adaptability.

[0004] Document CN119199822B discloses "A method, system, device, and storage medium for detecting minute displacements using radar," with the following steps: The radar transmits multiple sets of linear frequency modulated (LFM) signals with different carrier frequencies for detection; each set of signals includes multiple pulses. The received echoes from each pulse undergo FFT transformation, and the results of the FFT processing of all pulse sets are accumulated to obtain the range Doppler matrix of any pulse set. The range Doppler matrix of the first pulse set is used for target detection to obtain the target position. The radar then transmits the signal repeatedly, and the phase difference between the detected target position and the phase difference between the detected target position and the detected target position in subsequent pulse sets is calculated to obtain the coarse displacement. Based on the coarse displacement, the fuzzy displacement is calculated to obtain the true displacement of the target.

[0005] In summary, while existing technologies have improved perception robustness in complex environments through signal processing and multi-index fusion mechanisms, and enhanced the signal-to-noise ratio and stability of target detection by using multi-carrier frequency signals to solve fuzzy displacement, they can optimize static detection and anti-interference capabilities to some extent. However, existing technologies still rely on manual posture simulation and subjective judgment, making it impossible to precisely control the amplitude and frequency of micro-movements. Environmental interference depends on random events, making it difficult to standardize and reproduce complex scenarios. Furthermore, they lack Zigbee Mesh networking performance verification, and coverage mapping relies on manual point-by-point recording, resulting in low step accuracy and a lack of height dimension data. This leads to inaccurate blind zone positioning and fails to meet the full-dimensional quantitative testing requirements of millimeter-wave human presence sensors. Summary of the Invention

[0006] This invention provides an automated testing system and method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection, which can solve the problems mentioned in the background art.

[0007] To address the aforementioned technical problems, this invention provides the following technical solution, more specifically, an automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection, comprising:

[0008] S1. Fix the Zigbee human presence sensor under test at the test station, compile test cases and configure corresponding parameters in the host computer, control each hardware to perform zero self-test, build a Zigbee Mesh network with a specified topology according to the test case configuration, and confirm that all nodes are connected to the network and the link status is normal.

[0009] S2. Control the cylinder to drive the micro-motion actuator to move to the preset detection distance and mechanically lock it. Load the standard breathing curve algorithm, drive the piezoelectric ceramic to output the corresponding amplitude and frequency of breathing micro-motion, collect the sensor's detection response data, adjust the breathing amplitude parameters and repeat the test to obtain the sensor's detection sensitivity data and minimum detectable threshold.

[0010] S3. Start the corresponding standardized interference source according to the test case, configure the corresponding operating parameters of the interference, and collect the sensor's false alarm count, missed alarm count and response time data in the stationary state of the micro-moving target and the standard breathing state respectively.

[0011] S4. Trigger standard breathing micro-motion at the corresponding position of the designated sensor node, collect the local detection timestamp and gateway reception timestamp of each node, calculate the data reporting delay and packet loss rate; apply millimeter wave interference to the designated sensor node to trigger Mesh network route reselection, record the rerouting time and packet loss rate after rerouting, and verify the correctness of the gateway-side state aggregation logic.

[0012] S5. Control the two-dimensional electric moving stage and the Z-axis lifting stage to drive the standard micro-movement target to move point by point along the preset grid path. At each point, the standard breathing is triggered and the response data of the sensor is collected. The data is fused to generate a 3D detection heat map and the blind zone coordinates and beam angle parameters are marked. All test data are summarized to generate a quantitative test report and output it.

[0013] Furthermore, in step S1, it supports custom test case orchestration, configuration of test items, running parameters and pass / fail judgment rules, and execution of the entire test process according to the selected test cases, and finally outputs quantitative test results and summary reports.

[0014] Furthermore, in step S2, a dual-drive structure of cylinder and piezoelectric ceramic is used to perform micro-motion simulation. The cylinder drives the micro-motion actuator to complete the rapid positioning and mechanical locking of the target detection distance. The piezoelectric ceramic outputs periodic reciprocating micro-displacement to simulate the fluctuations of human breathing. The breathing amplitude and breathing frequency can be adjusted and configured.

[0015] Furthermore, in step S3, the standardized interference sources include four categories: airflow disturbance, moving target passage, thermal convection, and millimeter-wave electromagnetic interference. The operating parameters of each type of interference source can be independently adjusted, supporting the injection of a single type of interference or the coordinated injection of multiple types of interference according to a preset scenario template.

[0016] Furthermore, in step S4, each sensor node and gateway node deploys a high-precision timer to calculate the end-to-end data reporting delay by measuring the difference between the local trigger timestamp of the detected event and the gateway's received timestamp.

[0017] Furthermore, in step S4, millimeter-wave interference is applied directionally to the designated sensor node to trigger the Mesh network to automatically perform route reselection, and the duration of the rerouting process and packet loss data are recorded; the gateway side can be configured with a variety of state aggregation strategies, and the correctness of the aggregation logic can be verified by triggering detection through multiple nodes respectively.

[0018] Furthermore, in step S5, the two-dimensional electric moving stage, in conjunction with the height lifting mechanism, drives the standard micro-motion target to move point by point in the test space according to the preset grid path. Each point triggers the standard breathing micro-motion and collects sensor response data. Based on the fusion of all point data, a three-dimensional detection heat map is generated, and the detection blind area is automatically identified and marked.

[0019] According to another aspect of the present invention, an automated testing system for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection is provided. The system is based on the above-mentioned automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection, and specifically includes: a millimeter-wave micro-motion simulation execution module, a standardized multi-dimensional interference injection module, a Zigbee Mesh networking testing module, a 3D detection heat map mapping module, and a control and quantitative output module.

[0020] The millimeter-wave micro-motion simulation execution module adopts a dual-drive structure of cylinder and piezoelectric ceramic to bear simulated human body load and output periodic respiratory micro-motions with adjustable amplitude and frequency.

[0021] The standardized multidimensional interference injection module includes four types of interference sources: airflow disturbance, pet trajectory, thermal convection, and millimeter-wave electromagnetic interference. The operating parameters of each interference source can be independently adjusted, and it supports single-type interference injection and multi-type interference injection according to preset scenario templates.

[0022] The Zigbee Mesh networking test module supports the construction of Zigbee Mesh networks with various topologies, can collect data on the data reporting latency and packet loss rate of each node, supports triggering network route reselection tests, and can verify the correctness of various state aggregation strategies on the gateway side.

[0023] The 3D detection heat map mapping module uses a two-dimensional electric moving stage and a height lifting mechanism to drive the standard micro-movement target to move point by point along a preset grid path, collect sensor response data at each point, generate a three-dimensional detection heat map and mark the detection blind zone and beam angle parameters.

[0024] The control and quantitative output module is used to orchestrate test cases, control the automated execution of the entire test process, summarize test data, and output quantitative test reports.

[0025] The beneficial effects of this invention, an automated testing system and method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection, are as follows: Through high-precision micro-motion simulation and multi-dimensional interference injection, it can cover the core performance testing dimensions of human presence sensors. It not only realizes the leap from manual qualitative judgment to automated quantitative evaluation of micro-motion detection, but also incorporates multiple standard interference sources such as airflow disturbance, thermal convection, electromagnetic interference, and pet passage. Furthermore, it deeply verifies the sensor's anti-interference capability under different composite scenarios, making the test results more accurately reflect the product's real performance in the actual deployment environment, and providing a scientific basis for product development and mass production calibration.

[0026] In addition, by using Zigbee Mesh collaborative testing to build a multi-node network topology, the gaps in traditional single-node testing in terms of route reselection, state aggregation, and reporting latency verification are effectively filled.

[0027] Meanwhile, through 3D heat map mapping and fully automated execution, it provides a high-precision means of coverage visualization, which can clearly show the detection blind zone and beam angle boundary, and realize quality grading and trend warning based on quantitative analysis, which greatly improves testing efficiency and product quality control level. Attached Figure Description

[0028] The present invention will now be described in further detail with reference to the accompanying drawings and specific implementation methods.

[0029] Figure 1 This is a flowchart illustrating the method.

[0030] Figure 2 This is a schematic diagram of the system framework;

[0031] Figure 3 This is a schematic diagram of the structure of a millimeter-wave micro-motion simulation execution engine;

[0032] Figure 4 A schematic diagram of four types of interference sources and composite scenario templates for a standardized interference injection system;

[0033] Figure 5 This is a schematic diagram of the Zigbee Mesh collaborative testing engine architecture;

[0034] Figure 6 This is a schematic diagram of the architecture of a 3D detection heat map mapping engine. Detailed Implementation

[0035] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0036] According to one aspect of the invention, such as Figures 1-6 As shown, an automated testing system and method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection are provided, including:

[0037] Step 1: Initialize and set up the Mesh network

[0038] The Zigbee human presence sensor under test is fixed at the test station. Test cases are arranged and corresponding parameters are configured in the host computer. Each hardware component is controlled to perform a zero-reset self-test. A Zigbee Mesh network with a specified topology is constructed according to the test case configuration. All nodes are confirmed to be connected to the network and the link status is normal.

[0039] Specifically, it supports custom test case orchestration, configuration of test items, running parameters and pass / fail rules, execution of the entire test process according to the selected test cases, and finally output of quantitative test results and summary reports.

[0040] First, the system reads the pre-programmed test case configuration file from the host computer, parses the test items, parameter thresholds, and pass / fail rules, and then sends a global reset command to each lower-level execution unit via the RS-485 industrial control bus. After receiving the reset command, the millimeter-wave micro-motion simulation execution module drives the cylinder to push the micro-motion execution stage back to the mechanical zero point along the X-axis guide rail. At the same time, it controls the piezoelectric ceramic actuator to reset to the mid-position zero displacement state. It reads the feedback value of the built-in laser displacement sensor in real time, and after confirming that the positioning deviation is controlled within ±0.05mm, it returns a self-test pass status code to the central control unit. The standardized multi-dimensional interference injection module synchronously performs the reset operation, resetting the fan module, pet trajectory motion platform, heating plate, and millimeter-wave signal source to the off state. It reads the real-time data of the wind speed sensor, platinum resistance temperature sensor, and RF power meter respectively, confirming that all physical quantities are at the zero reference, and completes the self-test verification of its own hardware path.

[0041] The Zigbee Mesh networking test module, after all hardware units pass self-testing, broadcasts networking configuration frames to all sensor nodes and routing nodes under test through the gateway node according to the topology type and number of nodes specified in the test case. It configures the network role, working channel, and parent node binding rules for each node in turn. After the node completes the parameter writing, it automatically initiates a network access request. The gateway records the short address and received signal strength indication value of each node entering the network. After all nodes successfully enter the network, it issues routing table configuration instructions according to the preset routing rules of star, tree, or mesh topologies to build a Mesh network with the specified structure. After the networking is completed, the networking test module sends link probe frames to each node at 100ms intervals, continuously counts the return packets of 100 frames, calculates the single-hop latency and packet loss rate of each node, and after confirming that the link status of all nodes is stable and the packet loss rate is less than 1%, it feeds back the networking success status information to the control and quantification output module.

[0042] Finally, the self-test results of all execution units and the networking status data of the Mesh network are summarized to generate an initialization status verification table. The consistency between the actual running parameters and the test case configuration parameters is compared item by item. After confirming that there are no abnormal deviations, the initial state of each unit is locked, waiting for the triggering instructions of subsequent test steps. At the same time, the 3D detection heat map mapping module performs coordinate system calibration operation, spatially mapping and aligning the mechanical coordinate origin of the two-dimensional electric moving stage and the Z-axis lifting stage with the installation position of the sensor under test, establishing a unified three-dimensional test coordinate system, and establishing an accurate spatial reference for subsequent point-by-point mapping operations to ensure that the coordinates of the test points correspond one-to-one with the actual physical positions.

[0043] Step 2, Micro-motion sensitivity test

[0044] The control cylinder drives the micro-motion actuator to move to the preset detection distance and mechanically locks it. The standard breathing curve algorithm is loaded, and the piezoelectric ceramic is driven to output breathing micro-motions with corresponding amplitude and frequency. The sensor's detection response data is collected, the breathing amplitude parameters are adjusted and the test is repeated to obtain the sensor's detection sensitivity data and minimum detectable threshold.

[0045] Specifically, a dual-drive structure using cylinders and piezoelectric ceramics is employed to perform micro-motion simulation (e.g. Figure 3 As shown, the cylinder drives the micro-motion actuator to quickly locate and mechanically lock the target detection distance. The piezoelectric ceramic outputs periodic reciprocating micro-displacement to simulate the fluctuations of human breathing. The breathing amplitude and breathing frequency are adjustable.

[0046] First, the detection distance, initial breathing amplitude, breathing frequency, and number of test cycles corresponding to the micro-motion sensitivity test are extracted from the loaded test cases and sent to the millimeter-wave micro-motion simulation execution module via the industrial control bus. After receiving the position command, the cylinder drive unit of this module pushes the micro-motion execution stage carrying the silicone doll load along the X-axis guide rail to move towards the target position at a rated speed of 50mm / s. During the movement, the current displacement value is fed back in real time by the grating ruler built into the guide rail. When the deviation between the actual displacement and the target position is reduced to within ±0.1mm, the cylinder decelerates and stops and triggers the end mechanical locking mechanism to complete rigid positioning. After locking, a confirmation signal is returned to the control and quantization output module to lock the test reference distance to eliminate position drift in the subsequent micro-motion process.

[0047] Then load the standard respiratory curve algorithm, the specific algorithm is as follows:

[0048]

[0049] In the formula, The range of chest cavity movement (adjustable from 0.1 to 5 mm). This refers to the respiratory rate (0.1~0.5Hz, corresponding to 6~30 breaths / minute). The respiratory cycle (adjustable from 2 to 10 seconds) is generated by substituting preset respiratory amplitude and frequency parameters into the algorithm to produce a continuous time-domain displacement control sequence. After being converted from digital to analog, the sequence is output to the piezoelectric ceramic actuator. The piezoelectric ceramic performs reciprocating micro-displacement modulation with a response speed of less than 1ms. At the same time, it performs real-time closed-loop feedback correction through a built-in capacitive displacement sensor, controlling the actual output respiratory amplitude error within ±0.02mm. This causes the silicone puppet's chest cavity to produce periodic fluctuations consistent with the breathing reflex characteristics of a real human body, which can stably reproduce different levels of micro-motion states such as standard breathing, shallow breathing, and micro-breathing.

[0050] Finally, the status output signal of the Zigbee human presence sensor under test is collected. The number of sensor trigger responses is recorded with a single complete respiratory cycle as the statistical unit. The detection pass rate at the current amplitude is calculated. After the current amplitude test is completed, the respiratory amplitude parameter is automatically adjusted down by a preset step size. The complete process of positioning verification, micro-motion output, and data acquisition is repeated. Multiple sets of detection performance data at different amplitudes are obtained through step-by-step testing. A respiratory amplitude-detection response curve is generated by fitting. Finally, the minimum detectable threshold of the sensor is determined. The pass / fail judgment of this test item is completed by comparing it with the preset four-level detection sensitivity matrix. All raw data and judgment results are stored in the test result database.

[0051] Step 3: Interference Suppression Performance Test

[0052] According to the test cases, start the corresponding standardized interference source, configure the corresponding operating parameters of the interference, and collect the sensor's false alarm count, false alarm count, and response time data in the stationary state of the micro-moving target and the standard breathing state, respectively.

[0053] Specifically, standardized interference sources include four categories: airflow disturbance, moving target passage, thermal convection, and millimeter-wave electromagnetic interference (such as...). Figure 4 As shown, the operating parameters of various interference sources can be independently adjusted, supporting the injection of a single type of interference or the collaborative injection of multiple types of interference according to a preset scenario template.

[0054] The control and quantification output module parses the interference test item parameters in the test cases and then sends interference type, operating parameters, and test mode instructions to the standardized multi-dimensional interference injection module. This module, for single interference test modes, independently drives the corresponding interference execution unit and completes parameter closed-loop calibration. During airflow disturbance testing, the output power of the DC fan is adjusted via a PWM speed control circuit, and the wind speed is stabilized at the set value using real-time sampling data from a hot-film anemometer. Simultaneously, the gimbal servo mechanism precisely controls the airflow direction, oscillation angle, and oscillation frequency, ensuring that the parameter deviation within the wind speed range of 0~10m / s and the direction range of 0~360° does not exceed ±5%. During pet-walking testing, a stepper motor is driven along a miniature guide rail to move a silicone pet model along a set trajectory, and the encoder reflects the movement. The calibration movement speed supports multiple paths such as arcs, polygons, and circles, with a speed range of 0.1~2m / s. The model height and dwell time can be set as needed. During thermal convection testing, a PID temperature control algorithm drives the heating plate, and combined with multi-point platinum resistance temperature sampling, the temperature of the test area is stabilized in the range of 20~50℃ with a temperature control accuracy of ±0.5℃. With the help of a low-noise fan, a stable thermal convection field is formed. During millimeter-wave electromagnetic interference testing, a programmable signal source is controlled to generate a specified type of interference signal, which is radiated to the test area through a directional antenna. The calibration output strength is transmitted back in real time through an RF power meter, achieving accurate and controllable output of interference strength in the range of -90~0dBm. During single interference testing, all other interference sources are kept in a closed or zero state throughout the process to ensure the accuracy of single variable testing.

[0055] The standardized multi-dimensional interference injection module is designed for composite scenario testing modes. It calls the built-in multi-scenario composite templates and uses time axis synchronous scheduling logic to achieve parameter coordinated control of four types of interference sources. The system pre-stores typical composite scenario templates such as evening peak hours in high-rise residential buildings, weekends in commercial venues, and late nights in harsh environments. Each template contains the start-up sequence, operating parameters, dynamic change curves, and duration of each interference source. During testing, the module synchronously issues operating instructions to each unit of fan, pet trajectory, thermal convection, and millimeter wave interference according to the time axis configuration of the template. This ensures that the start time and parameter change rhythm of multiple types of interference are strictly aligned, which can reproduce the superposition effect of multiple interferences in real scenarios. During the test, the millimeter wave micro-motion simulation execution module is linked synchronously. The micro-motion target is set to a stationary state and a standard breathing state, respectively, to test the false alarm performance and false alarm performance of the sensor, ensuring the consistency and reproducibility of test conditions in composite scenarios.

[0056] This also includes real-time acquisition of the sensor's status output signal at a millisecond sampling frequency throughout the entire interference injection process. During the test period when the micro-moving target is stationary, the number of times the sensor falsely triggers the "personnel" state within a unit of time is counted, and the false alarm rate under the corresponding interference conditions is calculated. During the test period when the micro-moving target maintains standard breathing, the number of missed detections and the response time of each detection are counted, and the missed detection rate and average response time are calculated. After completing the current parameter test, the system automatically adjusts the interference intensity parameter according to a preset step size, repeats the process of interference injection, status acquisition, and index calculation, and fits and generates the corresponding relationship curves of interference intensity-false alarm rate and interference intensity-missed detection rate to locate the upper limit of interference tolerance of the sensor to meet the qualified threshold. Finally, the pass / fail judgment is completed according to the judgment rules of the test cases, and all raw data, performance curves and judgment results are uniformly stored in the test result database.

[0057] Step 4: Mesh Network Collaborative Testing

[0058] A standard breathing micro-motion is triggered at the corresponding location of a designated sensor node. The local detection timestamp and gateway reception timestamp of each node are collected, and the data reporting delay and packet loss rate are calculated. Millimeter wave interference is applied to the designated sensor node to trigger Mesh network route reselection. The rerouting time and packet loss rate after rerouting are recorded to verify the correctness of the gateway-side state aggregation logic.

[0059] Specifically, high-precision timers are deployed on each sensor node and gateway node. The end-to-end data reporting delay is calculated by the difference between the local trigger timestamp of the detected event and the gateway's received timestamp. Millimeter-wave interference is applied to designated sensor nodes to trigger the Mesh network to automatically perform route reselection, and the duration of the rerouting process and packet loss data are recorded. Multiple state aggregation strategies can be configured on the gateway side, and the correctness of the aggregation logic is verified by triggering detection on multiple nodes.

[0060] First, based on the test node location specified in the test case, a standard breathing trigger command is sent to the millimeter-wave micro-motion simulation execution module, and a delay test start signal is simultaneously sent to the Zigbee Mesh networking test module (e.g., Figure 5As shown), the Zigbee Mesh networking test module activates the 10µs high-precision timers built into each sensor node and gateway node. When the radar of a designated node detects a breathing micro-movement and triggers the "person" state, the node immediately generates a detection event timestamp locally and encapsulates it into a Zigbee data frame, which is then forwarded to the gateway through the Mesh network. When the gateway receives the data frames reported by each node, it synchronously adds the gateway's reception timestamp and sends it back to the networking test module. The module automatically calculates the end-to-end delay of each node's local detection and gateway reception, and simultaneously counts the number of successfully reported frames and the total number of sent frames within the specified test period. It calculates the packet loss rate in single-hop and multi-hop scenarios. Under normal conditions, the target packet loss rate is less than 1%, and under interference conditions, it is less than 5%.

[0061] Then, during the route reselection test under interference, the control and quantization output module sends a directional interference command to the standardized multi-dimensional interference injection module, controlling the millimeter-wave interference unit to aim the directional antenna at the designated sensor node, and gradually increasing the signal strength of the co-channel interference according to the preset step size until the uplink received signal strength of the node drops below the route switching threshold, triggering the Mesh network to automatically initiate route reselection. The Zigbee Mesh networking test module monitors the network topology changes and link status in real time, records the time interval from when the interference strength reaches the threshold to when the new route link stabilizes and data resumes normal transmission, and counts the number of data frames lost and the total number of frames sent during the rerouting process, calculates the packet loss rate during the rerouting phase, and verifies whether the rerouting time is less than 500ms and the packet loss rate is less than 5%.

[0062] Finally, during the state aggregation logic verification, the Zigbee Mesh networking test module sends the specified state aggregation policy configuration to the gateway node, including various rules such as triggering upon detection of any node, triggering only upon detection of all nodes, triggering upon detection of more than half of the nodes, and weighted threshold aggregation. Subsequently, standard breathing micro-motion is triggered sequentially or simultaneously at the corresponding positions of multiple sensor nodes, and the final aggregation state output by the gateway is collected. The actual aggregation result is compared with the expected result of the preset strategy one by one to verify the correctness of the aggregation logic. After completing all networking test items, the module summarizes and reports all data such as latency distribution, packet loss rate, rerouting time, and aggregation logic verification results. It performs a Pass / Fail judgment against the pass / fail threshold of the test cases, and uploads the raw data and judgment results to the control and quantitative output module to store them in the test database, providing data support for subsequent report generation.

[0063] Step 5: 3D Mapping and Report Output

[0064] The two-dimensional electric moving stage and the Z-axis lifting stage are controlled to move the standard micro-movement target point by point along the preset grid path. At each point, the standard breathing is triggered and the response data of the sensor is collected. The data are fused to generate a 3D detection heat map and the blind zone coordinates and beam angle parameters are marked. All test data are summarized to generate a quantitative test report and output it.

[0065] Specifically, a two-dimensional electric moving stage, in conjunction with a height lifting mechanism, drives a standard micro-movement target to move point by point within a preset grid path in the test space. Each point triggers a standard breathing micro-movement and collects sensor response data. Based on the fusion of all point data, a three-dimensional detection heat map is generated, which automatically identifies and marks the detection blind zone.

[0066] The 3D thermal mapping module receives the mapping range, grid step, and height level parameters from the control and quantization output module, and then performs a secondary coordinate system calibration (e.g., ...). Figure 6 As shown in the figure, the spatial mapping relationship between the mechanical zero point of the X / Y axis of the two-dimensional electric moving stage, the reference height of the Z-axis lifting stage, and the installation position of the sensor under test is confirmed to ensure that the test coordinates correspond one-to-one with the physical position. Then, according to the preset 10mm grid step and S-shaped line-by-line scanning path, the X-axis and Y-axis electric moving stages are driven to move the standard breathing target point by point. The height of the Z-axis lifting stage is adjusted synchronously to cover the full height test range of 0.3~3m. During the movement, the displacement data is fed back in real time through the built-in grating ruler of the guide rail, and the repeatability accuracy is controlled within ±0.05mm. When each coordinate point is reached, the module immediately sends a synchronous trigger command to the millimeter-wave micro-motion simulation execution module to start the standard breathing output with an amplitude of 4mm and a frequency of 0.2Hz. At the same time, the sensor's detection response time and sensitivity score in the 0~1 interval are collected at a millisecond sampling frequency. The collected data is bound and stored with the current spatial coordinates (x,y,z). After completing the single-point test, it automatically moves to the next point until all preset test grids are traversed.

[0067] The control and quantitative output module, after the full-grid data acquisition is completed, calls a three-dimensional spatial interpolation algorithm to continuously fit all the coordinate-marked sensitivity data, generating a complete 3D detection heatmap. A red-blue scale mapping is used to correspond to the high and low distribution of detection sensitivity. The system automatically identifies continuous areas with sensitivity scores below 0.2 as detection blind zones, calculates and marks the center coordinates and three-dimensional boundary range of each blind zone, and simultaneously calculates the radar's horizontal and vertical beam angle parameters based on the sensitivity attenuation distribution of the heatmap, outputting a quantitative score for coverage integrity. Finally, it summarizes the raw data and Pass / Fail judgment results of all test items, including micro-motion sensitivity, interference suppression, Mesh networking, and 3D mapping, generating a complete quantitative test report containing the breathing amplitude response curve, interference false alarm rate curve, reporting delay histogram, 3D heatmap, beam angle parameters, and a list of blind zone coordinates. It also performs batch trend analysis, comparing the performance dispersion of products in the same batch with the data drift of historical batches. When key indicators deviate from the standard value by more than ±15%, an automatic warning is triggered, supporting product quality grading and full lifecycle quality traceability.

[0068] To illustrate the effects achieved by using this invention, we conducted the following experiment:

[0069] Experiment Example 1: Sensitivity Test for Shallow Breathing Detection

[0070] The Zigbee human presence sensor to be tested was installed on the top of the test chamber at a height of 2.5m.

[0071] The two-dimensional electric stage is controlled to perform a zeroing operation, moving the micro-motion stage carrying the chest cavity of the silicone doll to a position 2m directly below the distance sensor;

[0072] Configure the breathing curve parameters in the host computer, set the breathing amplitude to 0.5mm and the breathing frequency to 0.2Hz, control the cylinder to quickly push the micro-motion actuator to position it to a detection distance of 2m and then lock it mechanically, and then drive the piezoelectric ceramic to perform fine modulation to stably control the actual breathing amplitude within the range of 0.5mm±0.02mm.

[0073] The micro-motion actuator was triggered to run continuously for 100 complete respiratory cycles, and the number of trigger responses of the sensor was collected synchronously. A total of 97 responses were measured in the actual test. The pass rate of the current amplitude was calculated to be 97%, which meets the qualification standard of ≥95%.

[0074] By gradually reducing the breathing amplitude to 0.2 mm using a preset step size and repeating the test procedure, the minimum detectable breathing threshold of the sensor was finally measured to be 0.18 mm.

[0075] The entire test took 5 minutes, and the accuracy of respiratory amplitude quantification reached 0.1mm.

[0076] Experiment Example 2: Fan Disturbance False Alarm Rate Test

[0077] Configure the operating parameters of the fan disturbance simulation module, set the wind speed to 3m / s, the air outlet direction to 45°, and the oscillation frequency to 10Hz, and place the fan module at the corresponding position 1.5m away from the sensor;

[0078] The millimeter-wave micro-motion simulation execution module is controlled to keep the micro-motion target in a stationary state, that is, the breathing amplitude is set to 0mm, to simulate an indoor test scenario without people.

[0079] The fan was started and ran continuously for 30 minutes. The status output signal of the sensor was collected in real time throughout the process, and the number of times the "personnel" status was falsely triggered was recorded. A total of 3 false alarms occurred in the actual test, and the false alarm rate was calculated to be 0.1 times / minute.

[0080] The wind speed is gradually increased to 5 m / s and 8 m / s according to the preset gradient. The above test procedure is repeated, and a wind speed-false alarm rate curve is established based on multiple sets of test data.

[0081] Based on the qualified threshold of false alarm rate <0.5%, the upper limit of wind speed that the sensor can withstand is determined to be 6.5 m / s.

[0082] The entire test process took 45 minutes and can quantitatively output a complete wind speed-false alarm rate performance curve.

[0083] Experiment Example 3: Reporting Latency Test on Zigbee Mesh Network

[0084] The Zigbee Mesh networking test module automatically built a 5-node mesh topology, which included 1 gateway node, 2 router nodes and 5 sensor nodes to be tested. It was confirmed that all nodes successfully joined the network and the link status was stable.

[0085] A standard respiratory micro-motion is triggered at the location corresponding to sensor node number 3, i.e., a respiratory amplitude of 4mm and a frequency of 0.2Hz, which simultaneously activates the high-precision timers of each node and the gateway.

[0086] The data reporting latency of each node was collected and statistically analyzed. The measured latency was 87ms for node 1, 103ms for node 2, 142ms for node 3 (including local detection and processing time), 156ms for node 4, and 189ms for node 5. All of these latency met the requirement of multi-hop latency <200ms.

[0087] By injecting millimeter-wave interference into node 1, co-frequency interference is applied in a directional manner, and the interference intensity is gradually increased to trigger the Mesh network to perform route reselection. The measured rerouting process takes 312ms, and the data packet loss rate after rerouting is stable is 3.2%, which meets the qualified standards of rerouting time < 500ms and packet loss rate < 5% in the interference environment.

[0088] The entire test process took 20 minutes and can quantify the output of node-reported latency distribution histogram and packet loss rate data.

[0089] Through the above experiments, we can see that the present invention can achieve high-precision micro-motion detection and quantitative testing, can standardize the reproduction of environmental interference and verify anti-interference performance, and at the same time complete the automated verification of Zigbee networking performance. The testing efficiency and quantification degree are significantly improved, which can fully meet the full-dimensional testing needs of human presence sensors.

[0090] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention are also within the protection scope of the present invention.

Claims

1. An automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection, characterized in that, The method includes: S1. Fix the Zigbee human presence sensor under test at the test station, compile test cases and configure corresponding parameters in the host computer, control each hardware to perform zero self-test, build a Zigbee Mesh network with a specified topology according to the test case configuration, and confirm that all nodes are connected to the network and the link status is normal. S2. Control the cylinder to drive the micro-motion actuator to move to the preset detection distance and mechanically lock it. Load the standard breathing curve algorithm, drive the piezoelectric ceramic to output the corresponding amplitude and frequency of breathing micro-motion, collect the sensor's detection response data, adjust the breathing amplitude parameters and repeat the test to obtain the sensor's detection sensitivity data and minimum detectable threshold. S3. Start the corresponding standardized interference source according to the test case, configure the corresponding operating parameters of the interference, and collect the sensor's false alarm count, missed alarm count and response time data in the stationary state of the micro-moving target and the standard breathing state respectively. S4. Trigger standard breathing micro-motion at the corresponding position of the designated sensor node, collect the local detection timestamp and gateway reception timestamp of each node, calculate the data reporting delay and packet loss rate; apply millimeter wave interference to the designated sensor node to trigger Mesh network route reselection, record the rerouting time and packet loss rate after rerouting, and verify the correctness of the gateway-side state aggregation logic. S5. Control the two-dimensional electric moving stage and the Z-axis lifting stage to drive the standard micro-movement target to move point by point along the preset grid path. At each point, the standard breathing is triggered and the response data of the sensor is collected. The data is fused to generate a 3D detection heat map and the blind zone coordinates and beam angle parameters are marked. All test data are summarized to generate a quantitative test report and output it.

2. The automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection as described in claim 1, characterized in that: In step S1, it is possible to customize test cases, configure test items, running parameters and pass / fail rules, execute the entire test process according to the selected test cases, and finally output quantitative test results and summary reports.

3. The automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection as described in claim 1, characterized in that: In step S2, a dual-drive structure of cylinder and piezoelectric ceramic is used to perform micro-motion simulation. The cylinder drives the micro-motion actuator to complete the rapid positioning and mechanical locking of the target detection distance. The piezoelectric ceramic outputs periodic reciprocating micro-displacement to simulate the fluctuations of human breathing. The breathing amplitude and breathing frequency can be adjusted and configured.

4. The automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection as described in claim 1, characterized in that: In step S3, the standardized interference sources include four categories: airflow disturbance, moving target passage, thermal convection, and millimeter-wave electromagnetic interference. The operating parameters of each type of interference source can be adjusted independently, supporting the injection of a single type of interference or the coordinated injection of multiple types of interference according to a preset scenario template.

5. The automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection according to claim 1, characterized in that: In step S4, each sensor node and gateway node deploys a high-precision timer to calculate the end-to-end data reporting delay by using the difference between the local trigger timestamp of the detected event and the gateway's received timestamp.

6. The automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection according to claim 1, characterized in that: In step S4, millimeter-wave interference is applied directionally to the designated sensor node to trigger the Mesh network to automatically perform route reselection, and the duration of the rerouting process and packet loss data are recorded. The gateway can be configured with various state aggregation strategies, and multiple nodes can be triggered to detect and verify the correctness of the aggregation logic.

7. The automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection as described in claim 1, characterized in that: In step S5, the two-dimensional electric moving stage and the height lifting mechanism work together to drive the standard micro-motion target to move point by point in the test space according to the preset grid path. Each point triggers the standard breathing micro-motion and collects sensor response data. Based on the fusion of all point data, a three-dimensional detection heat map is generated, and the detection blind area is automatically identified and marked.

8. An automated testing system for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection, characterized in that, The system is based on an automated testing method for Zigbee human presence sensors based on millimeter-wave micro-motion simulation and multi-dimensional interference injection as described in any one of claims 1-7, specifically including: a millimeter-wave micro-motion simulation execution module, a standardized multi-dimensional interference injection module, a Zigbee Mesh networking testing module, a 3D detection heat map mapping module, and a control and quantitative output module; The millimeter-wave micro-motion simulation execution module adopts a dual-drive structure of cylinder and piezoelectric ceramic to bear simulated human body load and output periodic respiratory micro-motions with adjustable amplitude and frequency. The standardized multidimensional interference injection module includes four types of interference sources: airflow disturbance, pet trajectory, thermal convection, and millimeter-wave electromagnetic interference. The operating parameters of each interference source can be independently adjusted, and it supports single-type interference injection and multi-type interference injection according to preset scenario templates. The Zigbee Mesh networking test module supports the construction of Zigbee Mesh networks with various topologies, can collect data on the data reporting latency and packet loss rate of each node, supports triggering network route reselection tests, and can verify the correctness of various state aggregation strategies on the gateway side. The 3D detection heat map mapping module uses a two-dimensional electric moving stage and a height lifting mechanism to drive the standard micro-movement target to move point by point along a preset grid path, collect sensor response data at each point, generate a three-dimensional detection heat map and mark the detection blind zone and beam angle parameters; The control and quantitative output module is used to orchestrate test cases, control the automated execution of the entire test process, summarize test data, and output quantitative test reports.

Images