Automatic test method and system for eye protection reminding of intelligent equipment

Abstract

The invention discloses an automatic test method for eye protection reminding of intelligent equipment, which is realized on the basis of WiFi channel state perception and specifically comprises the following steps: initializing a test environment, constructing a standardized and repeatable test scene, calibrating equipment coordinates and synchronizing a clock; controlling the mechanical arm to move according to a preset track to simulate a human body approaching intelligent equipment behavior; collecting CSI data and associating spatio-temporal information; performing sliding window segmentation and phase difference modeling on the CSI data; classifying behaviors through a deep learning model and outputting behavior labels and confidence coefficients; triggering eye protection reminding of the intelligent equipment and verifying the effectiveness of acousto-optic feedback; generating a test report and recording abnormal nodes; the invention further discloses an automatic test system for eye protection reminding of the intelligent equipment. The method and the system are used for verifying the sensing response reliability of the intelligent equipment to human body behaviors.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication testing, in particular to an intelligent device eye protection reminding automation testing method and system. BACKGROUND

[0002] With the continuous development of intelligent technology, modern smart TVs have developed innovative interaction functions based on Wi-Fi signal changes - when detecting changes in user distance, devices such as TVs can automatically trigger voice warnings, light prompts, and other humanized feedback. However, current industry testing still generally uses manual simulation schemes: testers need to repeatedly move their limbs to interfere with Wi-Fi signals to verify the TV response mechanism. This traditional method has significant limitations: manual operation cannot achieve high-frequency repetition (standardized movements of 10,000 times require a large amount of labor), and there are natural deviations in limb movements, resulting in a lack of consistency in test results; it is also impossible to conduct 7x24 hour continuous testing. In the rapid product iteration cycle, this inefficient and subjective testing method cannot meet the large-scale reliability verification needs, and this limitation makes it difficult for existing methods to meet the needs of smart TVs in rapid product iteration and large-scale performance verification.

[0003] As can be seen, the traditional testing scheme requires manual repeated movement to interfere with the Wi-Fi channel, which has the defects of low efficiency, long testing period, highly subjective test results, and difficulty in establishing unified evaluation standards, especially the inability to adapt to the needs of large-scale, high-intensity testing in the process of rapid iteration of smart TVs. SUMMARY

[0004] To solve the problems in the prior art, the purpose of the present application is to provide an intelligent device eye protection reminding automation testing method and system, which is used to verify the reliability of the perception response of intelligent devices to human behavior.

[0005] To achieve the above-mentioned purpose, the technical solution adopted by the present application is: an intelligent device eye protection reminding automation testing method based on WiFi channel state perception, specifically comprising the following steps:

[0006] Step 1, initialize the test environment, build a standardized and repeatable test scene, and calibrate the device coordinates and synchronize the clock;

[0007] Step 2, control the robot arm to move according to the preset trajectory to simulate the behavior of approaching the intelligent device by the human body;

[0008] Step 3, collect CSI data and associate the space-time information;

[0009] Step 4, perform sliding window segmentation and phase difference modeling on the CSI data;

[0010] Step 5, classify the behavior through a deep learning model and output the behavior label and confidence.

[0011] Step 6, triggering the smart device eye protection reminder and verifying the effectiveness of the sound and light feedback;

[0012] Step 7, generating a test report and recording abnormal nodes.

[0013] As a further improvement of the application, the step 1 specifically comprises the following steps:

[0014] Step 1.1, setting up WiFi routers, collecting array and interference source position coordinates, and eliminating environmental variable interference;

[0015] Step 1.2, building a ROS2 bus, completing device power-on and state verification, and ensuring the collaborative work of each module;

[0016] Step 1.3, calibrating the clock synchronization signal to control the data collection time error within the preset time range.

[0017] As a further improvement of the application, the step 3 specifically comprises the following steps:

[0018] Step 3.1, collecting CSI data and obtaining multi-dimensional signal features;

[0019] Step 3.2, using data labeling method: recording timestamp, frequency point and spatial coordinates, establishing time-space correlation, then temporarily storing original CSI data or uploading original CSI data to processing module through Ethernet and recording timestamp.

[0020] As a further improvement of the application, the step 4 is specifically as follows:

[0021] The CSI data is analyzed by Python tool for sliding window analysis, the continuous CSI data is segmented, the local features of signal change over time are captured, and phase difference modeling is performed: the phase difference between adjacent antennas is calculated by Python tool, the environmental noise is eliminated and the signal disturbance features caused by human movement are extracted.

[0022] As a further improvement of the application, in step 5, the behavior label includes approaching the smart device and moving away from the smart device, and the threshold of the confidence degree is ≥ 90%.

[0023] As a further improvement of the application, in step 6, the smart device sends sound and light prompts, pop-ups or notifications according to the classification results.

[0024] As a further improvement of the application, verifying the effectiveness of the sound and light feedback specifically includes:

[0025] Image recognition: matching the preset template of screen pop-up or light flashing through OpenCV algorithm to confirm whether the reminder is displayed;

[0026] Voice recognition: use MFCC algorithm to extract voice features, verify the consistency of prompt content and expected instructions.

[0027] As a further improvement of the application, the intelligent device is a smart TV.

[0028] The application also discloses an intelligent device eye protection reminding automatic test system, which comprises:

[0029] A test environment initialization module is used to initialize the test environment, build a standardized and repeatable test scene, calibrate the device coordinates and synchronize the clock.

[0030] A dynamic human behavior simulation module is used to control the mechanical arm to move according to the preset trajectory to simulate the human body approaching the intelligent device behavior.

[0031] A CSI data acquisition module is used to acquire CSI data and associate the space-time information.

[0032] A signal analysis processing module is used to perform sliding window segmentation and phase difference modeling on the CSI data.

[0033] A behavior classification module is used to classify behaviors through a deep learning model and output behavior labels and confidence.

[0034] A device interaction feedback verification module is used to trigger the eye protection reminder of the intelligent device and verify the effectiveness of the sound and light feedback.

[0035] A test report generation module is used to generate a test report and record abnormal nodes.

[0036] The application has the following beneficial effects:

[0037] 1. The application proposes an innovative solution for the test requirements of the eye protection warning and distance monitoring functions of the smart TV realized by WiFi signal fluctuation, designs an automatic test method and system, simulates the human body moving track through an intelligent mechanical device, controls the WiFi signal interference intensity and frequency, and realizes the full-process automation from signal simulation, response monitoring to effect evaluation.

[0038] 2. The application can complete tens of thousands of standardized test cycles, significantly improve the test efficiency and reduce the labor cost, and provides a scientific test means for the reliability verification of the interactive function of the smart TV. DETAILED DESCRIPTION

[0039] Figure 1 The step flowchart of the embodiment of the application is shown in the figure.

[0040] The embodiments of the application will be described in detail below with reference to the accompanying drawings.​

[0041] Embodiment 1

[0042] As Figure 1 shown, this embodiment takes a smart TV as an example for illustration. A smart TV eye protection reminding automatic test method based on WiFi channel state perception includes the following steps:

[0043] Step 1, test environment initialization: build a standardized and repeatable test scene to ensure the accuracy of data collection and behavior simulation:

[0044] Set the coordinates of the WiFi router, array, and interference source positions to eliminate environmental variable interference (such as multipath effect);

[0045] Establish a ROS2 bus to complete the device power-on and state verification, ensuring the collaborative work of each module;

[0046] Calibrate the clock synchronization signal to control the data collection time error within 10 nanoseconds, avoiding analysis deviation caused by timing disorder.

[0047] Step 2, dynamic human behavior simulation: simulate real human movement through high-precision mechanical arms to cover multiple behavior patterns to verify system sensitivity:

[0048] Control the JR-601 six-axis mechanical arm to move in a 4-meter radius spherical space according to the preset path, with a repeat positioning accuracy of ±0.3mm, ensuring accurate path reproduction;

[0049] Mechanical arm working range: 4m radius spherical space, repeat positioning accuracy ±0.3mm;

[0050] Reflector size: 1.2m (high) x 0.4m (wide) x 0.25m (thick), simulating adult body shape, close to actual use scenarios;

[0051] Motion pattern library: preset straight line, circular arc, polyline, etc. Motion trajectory, speed coverage 0.3-1.5m / s (walking to sprint range), test system response ability to different behaviors.

[0052] Step 3, CSI data collection and synchronization: real-time capture of WiFi channel state information (CSI) to provide raw data for behavior recognition:

[0053] Start Intel AX200 chipset to collect CSI data at 100MHz bandwidth, obtain multi-dimensional signal features (such as amplitude, phase);

[0054] Data labeling method: Record timestamps, frequency points, and spatial coordinates, establish spatiotemporal correlation (e.g., mapping of human body position and signal changes), then use a ring buffer to temporarily store raw CSI data (capacity > 1 GB) or upload raw CSI data to the processing module through Ethernet at high speed while recording timestamps.

[0055] Step 4, perform sliding window analysis on the CSI data of step 3 using Python tools, segment the continuous CSI data (e.g., 500 milliseconds per window), capture local features of signal changes over time; at the same time, perform phase difference modeling (use Python tools to calculate the phase difference between adjacent antennas, eliminate environmental noise, and extract signal disturbance features caused by human movement).

[0056] Step 5, identify user behavior through a deep learning model to determine whether to trigger an eye protection reminder:

[0057] Input the phase difference feature sequence (time dimension data) extracted in step 4 into the model, then input these features into the pre-trained LSTM model, output behavior labels (such as "approaching TV" and "moving away from TV") and confidence (e.g., 95%), ensuring decision reliability.

[0058] Step 6, verify whether the system can correctly trigger the eye protection reminder and record the response effect:

[0059] Feedback mechanism: control the TV to issue sound and light prompts, pop-ups, or notifications (such as voice broadcast "please maintain distance") based on classification results;

[0060] Response verification:

[0061] Image recognition: match the preset template of screen pop-up or light flashing through OpenCV algorithm to confirm whether the reminder is displayed;

[0062] Voice recognition: use MFCC algorithm to extract voice features to verify the consistency of the prompt content and the expected instructions (such as the keyword "vision protection").

[0063] Step 7, summarize the test data throughout the process to form a structured report to evaluate system performance:

[0064] Integrate the above content, integrate environmental parameters, robot trajectory, CSI data samples, model output results, and device response logs, and output a test report.

[0065] Example 2

[0066] A WiFi channel state sensing-based intelligent TV eye protection reminder automation testing system, comprising:

[0067] The test environment initialization module, the dynamic human behavior simulation module, the CSI data collection module, the signal analysis processing module, the behavior classification module, the device interaction feedback verification module, and the test report generation module.

[0068] The system triggers WiFi channel changes by simulating human motion through a mechanical arm, controls the TV to issue an eye protection reminder after identifying the behavior through an AI model, and automatically verifies the effectiveness of the function.

[0069] The test environment initialization module comprises:

[0070] The ROS2 communication bus, the clock synchronization unit, and the device coordinate calibration unit, wherein the clock synchronization error is ≤10 ns, and the device coordinate positioning accuracy is ±1 cm.

[0071] The dynamic human behavior simulation module comprises:

[0072] The JR-601 six-axis mechanical arm, the adult body-shaped reflector (1.2 m x 0.4 m x 0.25 m), and the motion trajectory library, wherein the mechanical arm moves at a speed of 0.3-1.5 m / s within a spherical space with a radius of 4 m, and the repeated positioning accuracy is ±0.3 mm.

[0073] The CSI data collection module comprises:

[0074] The Intel AX200 chip set, the ring buffer, and the Ethernet transmission unit are configured to collect CSI data at a bandwidth of 100 MHz, and associate spatial coordinates and frequency point information through timestamps.

[0075] The signal analysis processing module comprises:

[0076] The sliding window segmentation unit and the phase difference modeling unit are configured to segment the CSI data in 500 ms windows and calculate the phase difference between multiple antennas to eliminate environmental noise.

[0077] The behavior classification module is a pre-trained LSTM model, the input is a phase difference feature sequence, and the output is a behavior label and a confidence level; the behavior label includes "approaching the TV" and "moving away from the TV", and the confidence level threshold is ≥90%.

[0078] The device interaction feedback verification module comprises:

[0079] The acousto-optic prompt unit (TV screen pop-up window, LED flashing), the voice instruction unit, and the feedback verification unit.

[0080] The feedback verification unit identifies the screen pop-up window through the OpenCV image matching algorithm, or extracts the voice spectrum features through the MFCC algorithm to verify the instruction content.

[0081] The test report generation module comprises:

[0082] The data aggregation unit, the performance index statistics unit and the exception marking unit are configured to output a test report containing visualizations of response delay, false positive rate and false negative rate.

[0083] The above-described embodiments only express specific implementation manners of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the patent scope of the present application. It should be noted that, for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the protection scope of the present application.

Claims

1. A method for automatically testing eye protection reminders for smart devices, characterized in that: Based on WiFi channel status awareness, the implementation specifically includes the following steps: Step 1: Initialize the test environment, build a standardized and repeatable test scenario, calibrate the device coordinates, and synchronize the clock; Step 2: Control the robotic arm to move along a preset trajectory to simulate the behavior of a human body approaching the smart device; Step 3: Collect CSI data and associate spatiotemporal information; Step 4: Perform sliding window segmentation and phase difference modeling on the CSI data; Step 5: Use the deep learning model to classify the behavior and output the behavior label and confidence level; Step 6: Trigger the eye protection reminder on the smart device and verify the effectiveness of the sound and light feedback; Step 7: Generate a test report and record abnormal nodes.

2. The automated testing method for eye protection reminder of smart devices according to claim 1, characterized in that: The step 1 specifically includes the following steps: Step 1.1: Set up the WiFi router, acquisition array, and interference source location coordinates to eliminate environmental variable interference. Step 1.2: Build the ROS2 bus, complete device power-on and status verification, and ensure that all modules work together; Step 1.3: Calibrate the clock synchronization signal to control the data acquisition time error within the preset time range.

3. The automated testing method for eye protection reminder of smart devices according to claim 1, characterized in that: The step 3 specifically includes the following steps: Step 3.1, collect CSI data and obtain multi-dimensional signal features; Step 3.2: Use the data annotation method: record the timestamp, frequency point, and spatial coordinates to establish spatiotemporal correlation, then temporarily store the original CSI data or upload the original CSI data to the processing module via Ethernet at high speed while recording the timestamp.

4. The automated testing method for eye protection reminder of smart devices according to claim 3, characterized in that: The step 4 is specifically as follows: The CSI data is analyzed using a Python tool through a sliding window. The continuous CSI data is segmented to capture the local characteristics of the signal over time. Phase difference modeling is also performed: Python tools are used to calculate the phase difference between adjacent antennas, eliminate environmental noise, and extract signal disturbance characteristics caused by human movement.

5. The automated testing method for eye protection reminder of smart devices according to claim 1, characterized in that: In step 5, the behavior tags include approaching the smart device and moving away from the smart device, and the confidence threshold is ≥90%.

6. The automated testing method for eye protection reminder of smart devices according to claim 1, characterized in that: In step 6, the smart device is controlled to emit an audio and visual prompt, a pop-up window, or a notification based on the classification result.

7. The automated testing method for eye protection reminder of smart devices according to claim 6, characterized in that: Verifying the effectiveness of acousto-optic feedback specifically includes: Image recognition: Use OpenCV algorithms to match preset templates of screen pop-ups or flashing lights to confirm whether reminders are displayed; Speech recognition: MFCC algorithm is used to extract speech features and verify the consistency between prompt content and expected instructions.

8. The automated testing method for eye protection reminder of smart devices according to any one of claims 1 to 7, characterized in that: The smart device is a smart TV.

9. An automated testing system for eye protection reminders for smart devices, characterized in that: include: Test environment initialization module: used to initialize the test environment, build standardized and repeatable test scenarios, calibrate device coordinates, and synchronize clocks; Dynamic human behavior simulation module: used to control the robot arm to move along a preset trajectory to simulate the behavior of a human body approaching an intelligent device; CSI data acquisition module: used to collect CSI data and associate spatiotemporal information; Signal analysis and processing module: used to perform sliding window segmentation and phase difference modeling on CSI data; Behavior classification module: used to classify behaviors through deep learning models and output behavior labels and confidence levels; Device interaction feedback verification module: used to trigger eye protection reminders on smart devices and verify the effectiveness of audio and visual feedback; Test report generation module: used to generate test reports and record abnormal nodes.

Images