WiFi router product test method

Abstract

The invention relates to the field of product testing, and particularly discloses a WiFi router product testing method which comprises the following steps: after a to-be-tested router starts a WiFi signal of a specified frequency band and channel, firstly recording a stable signal strength indication value of the router under a standard distance, and then applying standardized micro vibration excitation to the router through a micro vibration motor; during the period, a fluctuation data stream of the signal intensity is acquired at high frequency, and then dynamic excitation response analysis is carried out on the fluctuation data stream of the high-frequency signal intensity and the stable signal intensity value of the router to be tested by introducing a deep learning algorithm; and based on the steady-state signal intensity level of the router and the signal intensity fluctuation mode of the router under the micro vibration disturbance, judging whether the router has a hidden hardware fault or not and the fault type. According to the method, hidden hardware faults caused by assembly tolerance, mechanical stress and the like can be effectively excited and accurately recognized, and powerful support is provided for fault detection and quality control of router products.

Description

Technical Field

[0001] The present application relates to the field of product testing, and more specifically, to a method for testing a WiFi router product. Background Art

[0002] With the widespread adoption and development of wireless communication technology, WiFi routers have become core network access devices for homes, businesses, and public places. Their performance stability and signal reliability are directly related to the user's network experience. Therefore, establishing an efficient quality inspection system in the production and manufacturing of WiFi routers is crucial to ensuring product quality, reducing after-sales repair rates, and maintaining brand reputation.

[0003] Currently, conventional testing methods for WiFi routers in the industry primarily focus on functional verification and measuring static performance metrics. For example, automated scripts are used to test the normal operation of basic router functions such as startup, configuration, and data forwarding. Key performance parameters such as stable signal strength and throughput at specific channels and power levels are measured in a shielded environment. However, this traditional static testing approach primarily focuses on overt faults that can lead to complete device failure or significant performance deviations from design specifications. However, in practice, hidden hardware faults caused by assembly tolerances and mechanical stress (such as loose antenna connectors and poor solder joints on PCBs) are often difficult to detect through conventional testing. However, during actual user use, these faults can be triggered by fluctuations in ambient temperature and humidity, minor vibrations, or daily disturbances such as plugging and unplugging cables. These can lead to a sharp drop in signal quality and frequent network disconnections, severely impacting the user experience and significantly complicating troubleshooting.

[0004] Therefore, an optimized WiFi router product testing method is expected. Summary of the Invention

[0005] In order to solve the above technical problems, the present application is proposed. An embodiment of the present application provides a WiFi router product testing method. After the router to be tested turns on the WiFi signal of the specified frequency band and channel, it first records its stable signal strength indication value at a standard distance, then applies standardized micro-vibration excitation to the router through a micro-vibration motor, and during this period, collects the high-frequency signal strength fluctuation data stream. Then, by introducing a deep learning algorithm, a dynamic excitation response analysis is performed on the high-frequency signal strength fluctuation data stream and stable signal strength value of the router to be tested. Based on the steady-state signal strength level of the router and its signal strength fluctuation pattern under micro-vibration disturbance, it is determined whether the router has a hidden hardware fault and the fault type. This method can effectively stimulate and accurately identify hidden hardware faults caused by assembly tolerances, mechanical stress, etc., and provide strong support for fault detection and quality control of router products.

[0006] According to one aspect of the present application, a WiFi router product testing method is provided, which includes: Establish communication with the router under test through the preset serial port, drive the router under test to turn on the WiFi signal in the 2.4GHz or 5GHz band, and specify the test channel and maximum transmission power; Use a test antenna at a preset standard distance to receive the WiFi signal transmitted by the router to be tested, and record the stable signal strength indicator value; Applying standardized micro-vibration to the router under test through a micro-vibration motor, and collecting high-frequency signal intensity fluctuation data stream during the vibration; A deep learning-based dynamic stimulus response analysis is performed on the high-frequency signal strength fluctuation data stream and the stable signal strength indicator value to obtain a fault diagnosis result of the router to be tested.

[0007] Compared to existing technologies, the WiFi router product testing method provided in this application first records the stable signal strength indicator value at a standard distance after the router under test turns on the WiFi signal of a specified frequency band and channel. A micro-vibration motor then applies standardized micro-vibration excitation to the router, and during this period, a high-frequency data stream of signal strength fluctuations is collected. Furthermore, a deep learning algorithm is introduced to perform dynamic excitation response analysis on the high-frequency signal strength fluctuation data stream and stable signal strength value of the router under test. Based on the router's steady-state signal strength level and its signal strength fluctuation pattern under micro-vibration perturbations, the presence of hidden hardware faults in the router and the type of fault are determined. This method can effectively stimulate and accurately identify hidden hardware faults caused by assembly tolerances, mechanical stress, and other factors, providing strong support for fault detection and quality control of router products. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 Flowchart of a WiFi router product testing method according to an embodiment of the present application.

[0010] Figure 2 4 is a flowchart of sub-step S4 of the WiFi router product testing method according to an embodiment of the present application.

[0011] Figure 3Schematic diagram of data flow of sub-step S4 of the WiFi router product testing method according to an embodiment of the present application.

[0012] Figure 4 4 is a flowchart of sub-step S41 of the WiFi router product testing method according to an embodiment of the present application.

[0013] Figure 5 Flowchart of sub-step S42 of the WiFi router product testing method according to an embodiment of the present application.

[0014] Figure 6 4 is a flowchart of sub-step S421 of the WiFi router product testing method according to an embodiment of the present application. DETAILED DESCRIPTION

[0015] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0016] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0017] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0018] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0019] It is worth noting that in this application, all actions to obtain data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0020] In response to the technical problems described in the above background technology, this application proposes a WiFi router product testing method. After the router under test turns on the WiFi signal of a specified frequency band and channel, the method first records its stable signal strength indicator value at a standard distance. Then, a standardized small vibration excitation is applied to the router via a micro-vibration motor. During this period, a high-frequency data stream of signal strength fluctuations is collected. Then, by introducing a deep learning algorithm, a dynamic excitation response analysis is performed on the high-frequency signal strength fluctuation data stream and stable signal strength value of the router under test. Based on the steady-state signal strength level of the router and its signal strength fluctuation pattern under small vibration perturbations, the presence of hidden hardware faults in the router and the fault type are determined. This method can effectively stimulate and accurately identify hidden hardware faults caused by assembly tolerances, mechanical stress, etc., providing strong support for fault detection and quality control of router products.

[0021] Figure 1 FIG. 1 is a flow chart of a WiFi router product testing method according to an embodiment of the present application. Figure 1 As shown, the WiFi router product testing method includes the following steps: S1, establishing communication with the router to be tested through a preset serial port, driving the router to be tested to turn on the WiFi signal in the 2.4GHz or 5GHz frequency band, and specifying the test channel and maximum transmission power; S2, using a test antenna at a preset standard distance to receive the WiFi signal transmitted by the router to be tested, and recording the stable signal strength indicator value; S3, applying standardized micro-vibration to the router to be tested through a micro-vibration motor, and collecting a high-frequency signal strength fluctuation data stream during the vibration; S4, performing a deep learning-based dynamic excitation response analysis on the high-frequency signal strength fluctuation data stream and the stable signal strength indicator value to obtain a fault diagnosis result of the router to be tested.

[0022] In the aforementioned WiFi router product testing method, step S1 establishes communication with the router under test via a preset serial port, driving the router to activate WiFi signals in the 2.4GHz or 5GHz band, while also specifying the test channel and maximum transmit power. It should be understood that the WiFi router's internal hardware and software must be activated in a controlled manner to simulate real-world operating conditions, and the test requires precise control of core parameters such as frequency band, channel, and power to eliminate variable interference. Therefore, in order to build a standardized test environment and ensure that the device under test enters the target working mode, this application is based on the router's underlying firmware communication protocol. It sends an AT command set to the router under test (DUT) through a preset serial port (such as UART or USB to serial port), driving it to turn on the WiFi radio module of the specified frequency band (2.4GHz / 5GHz), and force the router radio module to operate in the specified test channel (such as channel 6) and maximum transmission power state (such as 20dBm). For example, sending the standardized control command sequence "AT+RF_ON=2.4G,CH6,PWR20" can achieve accurate reproduction of test conditions and programmable control of the device working state, ensuring the standardization of the test environment and laying the foundation for subsequent signal acquisition.

[0023] Specifically, AT commands are a command language widely used in communications devices. Their concise syntax and efficient execution make them particularly suitable for debugging and controlling embedded systems. By parsing these commands, the router's internal firmware can identify the user's intent and adjust the operating parameters of the hardware modules accordingly. In this solution, testers leverage this mechanism to send a standardized command such as "AT+RF_ON=2.4G,CH6,PWR20" to the router under test, forcing the activation of the wireless radio frequency module in a specific frequency band. "2.4G" represents the 2.4 GHz band, and "CH6" indicates channel 6; these two parameters together determine the frequency at which the WiFi signal is transmitted. "PWR20" sets the maximum transmit power to 20dBm, ensuring the expected signal strength. The standardized format of this command facilitates automated script invocation, improving the repeatability and consistency of the testing process.

[0024] During implementation, the test environment must be constructed with full consideration for signal interference, thus maintaining strict control over test conditions. For example, in a lab environment with multiple devices, other wireless access points may operate on the same channel, causing signal overlap and affecting test results. To avoid this issue, testers should use a spectrum analyzer to scan the surrounding wireless environment in advance and select the channel with the least interference as the target test channel. Furthermore, non-essential wireless devices should be turned off to minimize the impact of external noise on test data. Furthermore, AT commands sent via the serial port ensure that the router under test always operates on the specified channel, unaffected by automatic channel switching. This control method not only improves test accuracy but also provides a stable foundation for subsequent data collection and analysis.

[0025] To further improve testing efficiency, an automated testing framework is often employed to manage the sending and response of commands. This framework consists of a host computer, a serial communication module, and the router under test. The host computer is responsible for generating and sending commands, while also receiving feedback from the router. Test scripts can flexibly configure parameters such as frequency band, channel, and transmit power level as needed, enabling rapid switching between different scenarios. During execution, the script continuously monitors the serial port output to verify that the router has successfully entered the target operating state. If an erroneous response is detected, it is automatically logged and an alarm is triggered, allowing testers to intervene promptly. This approach significantly reduces the need for manual intervention and improves the stability and scalability of the testing process.

[0026] It's important to note that before sending AT commands, you need to ensure that the router is in a communicative state. Some devices may undergo a brief initialization process during startup, during which they are unable to respond to external commands. Therefore, test scripts should include waiting mechanisms, such as inserting an appropriate delay before sending commands or listening for specific startup completion identifiers, to ensure that commands are received correctly. In addition, routers from different manufacturers may use slightly different AT command sets, so during implementation, it is necessary to refer to the technical documentation of the specific device to confirm the validity of the command format and parameter range. For devices that support custom firmware, it is also possible to expand the functionality of the original command set by modifying the underlying code to make it more suitable for testing requirements.

[0027] When a router receives an AT command, its internal firmware interprets the command content and calls the appropriate driver to configure the wireless module. This process involves setting multiple hardware registers, including key parameters such as frequency band selection, channel number, and power amplifier gain. Because these configurations directly impact the operating state of the RF front-end, it is crucial to ensure that all modifications remain within safe limits. For example, certain devices may have settable transmit power levels limited by national regulations. Exceeding these limits may result in device violations or even damage. Therefore, testers should carefully verify parameter limits before sending commands. If necessary, validation logic can be incorporated into scripts to prevent invalid values from being written to hardware registers. Furthermore, some high-end routers feature dynamic power adjustment, which may automatically adjust transmit power based on factors such as temperature and voltage, potentially affecting test consistency. To avoid these issues, test commands should explicitly disable the automatic adjustment mechanism to ensure that the power value remains at the set level.

[0028] The baud rate setting of the serial port is also crucial throughout the communication process. The baud rate determines the speed of data transmission. If set incorrectly, data loss or misinterpretation may occur. Typically, routers ship with a default baud rate of 115200bps, but some devices use 9600bps or other values. Before connecting, testers should consult the device manual to confirm the correct baud rate setting and match it in the test script. Additionally, pay attention to parameters such as data bits, stop bits, and parity bits to ensure that the data format is consistent between the master and the device under test. If communication fails due to a baud rate mismatch, try reconnecting multiple times or manually adjusting the parameters until a stable connection is established.

[0029] To further enhance testing flexibility, some advanced test platforms support remote control. By integrating network protocols (such as Telnet or SSH) with a local serial port server, testers can remotely control the router under test, enabling remote debugging and parameter adjustments. This approach is particularly suitable for large-scale deployment testing or cross-regional collaborative projects, helping to reduce on-site maintenance costs and improve testing efficiency. However, remote control mechanisms also introduce new security risks, necessitating appropriate measures to ensure the security of communication links. For example, enabling encrypted channels, setting access permissions, and regularly renewing authentication credentials can prevent unauthorized access from interfering with testing.

[0030] In the aforementioned WiFi router product testing method, step S2 involves using a test antenna at a preset standard distance to receive the WiFi signal transmitted by the router under test and recording the stable signal strength indicator (RSSI). It should be understood that the signal strength level of the WiFi signal transmitted by the router is the most intuitive indicator of hidden hardware faults. Therefore, to quantify the router's initial performance level in an interference-free state and establish a comparison baseline for subsequent dynamic analysis, this application, based on an electromagnetic wave propagation model, uses a calibrated test antenna (such as a biconical antenna) at a preset standard distance (e.g., 1 meter) to receive the WiFi signal transmitted by the device under test (DUT). The stable signal strength indicator (RSSI) is recorded using a spectrum analyzer or a dedicated wireless tester. In a specific implementation, the relative position of the test antenna and the router under test is fixed in an electromagnetically shielded chamber. RSSI data is continuously collected at a sampling rate of 1kHz. Once the signal strength stabilizes (e.g., the RSSI value fluctuates within ±1dBm for 5 consecutive seconds), the average RSSI value over that period is recorded as the stable signal strength indicator. This approach eliminates the effects of test distance and environmental variables on signal measurement, resulting in a reproducible steady-state performance benchmark.

[0031] In the above-mentioned WiFi router product testing method, step S3 applies standardized micro-vibrations to the router under test via a micro-vibration motor, and during the vibration, collects a high-frequency signal strength fluctuation data stream. It should be understood that this application takes into account that during the production and assembly process of the router, the antenna is typically connected to the RF connector (such as the IPEX interface) on the motherboard via a thin coaxial cable. The physical nature of hidden structural defects such as loose antenna connectors and poor solder joints on PCB boards is poor mechanical contact. Their electrical characteristics may appear normal in a static environment, but when subjected to external physical disturbances (such as transportation bumps or daily user operations), the contact state undergoes instantaneous changes at the micron level, causing degradation of signal transmission path characteristics. Therefore, in order to actively and controllably stimulate and expose potential faults that cannot be detected in conventional static testing during production testing, this application further applies standardized external excitation to the router under test, observes and analyzes its response signal, and thereby infers the internal structural state of the router under test. Specifically, first, a micro-vibration motor is fixed to a key location on the router, such as the area of the outer casing near the antenna interface. The test system controls the motor's start and stop, vibration frequency, and amplitude, applying brief, standardized, minute vibrations to the router. During this excitation, the signal analyzer continuously records the instantaneous RSSI value at a high frequency (e.g., a 10kHz sampling rate), generating time-series data streams of high-frequency signal strength fluctuations. The unique fluctuation patterns inherent in this high-frequency signal strength fluctuation data stream strongly correlate with potential fault types (e.g., intermittent contact due to loose connectors, or regular changes in contact resistance of weak solder joints under vibration), providing direct data evidence for subsequent fault diagnosis.

[0032] In the above-mentioned WiFi router product testing method, the step S4 performs a deep learning-based dynamic excitation response analysis on the high-frequency signal strength fluctuation data stream and the stable signal strength indicator value to obtain the fault diagnosis result of the router to be tested. It should be understood that the stable signal strength indicator value represents the baseline performance of the router in an interference-free state, reflecting the signal transmission capability of its hardware components under ideal conditions. A lower stable signal strength indicator value usually means that the RF components of the router have some degree of performance degradation or hardware defects. The high-frequency signal strength fluctuation data stream records the instantaneous fluctuation of the signal strength of the router when it is subjected to external micro-vibration excitation, revealing the response characteristics of the hardware components inside the router under physical disturbances. Usually, signal strength data with large fluctuations means that there are hidden dangers of poor mechanical contact or unstable component connection inside the router. Based on this, in order to achieve efficient diagnosis of hidden hardware faults of the router to be tested, the present application further introduces deep learning technology, extracts the timing feature patterns in the high-frequency signal strength fluctuation data stream through automatic learning, and performs correlation analysis with the stable signal strength indicator value to construct a comprehensive description of the router fault characteristics, thereby achieving accurate identification of different fault types. Among them, Figure 2 4 is a flowchart of sub-step S4 of the WiFi router product testing method according to an embodiment of the present application. Figure 3 FIG. 4 is a data flow diagram of sub-step S4 of the WiFi router product testing method according to an embodiment of the present application. Figure 2 and Figure 3 As shown, the step S4 includes the steps of: S41, performing local time domain fluctuation analysis on the high-frequency signal strength fluctuation data stream to obtain a time series of signal strength local time domain fluctuation feature coding vectors; S42, performing local-global double-layer context association perception coding on the time series of the signal strength local time domain fluctuation feature coding vectors to obtain a signal strength time series fluctuation feature global time domain context coding vector; S43, inputting the stable signal strength indication value and the signal strength time series fluctuation feature global time domain context coding vector into a router fault diagnosis model based on a multi-layer perceptron to obtain the fault diagnosis result.

[0033] Figure 4 FIG. 4 is a flowchart of sub-step S41 of the WiFi router product testing method according to an embodiment of the present application. Figure 4As shown, the step S41 includes the steps of: S411, time-series segmenting the high-frequency signal strength fluctuation data stream based on a preset time window to obtain a time series of local time domain segments of the signal strength fluctuation data; S412, performing signal strength local fluctuation feature extraction based on one-dimensional convolution coding on each local time domain segment of the signal strength fluctuation data in the time series of the local time domain segments of the signal strength fluctuation data to obtain a time series of the signal strength local time domain fluctuation feature coding vector.

[0034] More specifically, the step S411 performs time-series segmentation on the high-frequency signal strength fluctuation data stream based on a preset time window to obtain a time series of local time domain segments of the signal strength fluctuation data. Specifically, since the signal strength fluctuation caused by vibration has non-stationary characteristics, its fault characteristic mode is often manifested as a millisecond-level transient event (such as a periodic pulse caused by a cold solder joint), and directly processing the long time series of the high-frequency signal strength fluctuation data stream may dilute the local key information. Therefore, in order to capture the local sensitive pattern in the high-frequency signal strength fluctuation data stream more finely, the present application first slices the continuous high-frequency signal strength fluctuation data stream based on a fixed time window, for example, setting 200ms as the time window, and slidingly segmenting the original high-frequency signal strength fluctuation data stream with an overlap rate of 30%, generating a series of local time domain segments, and forming a time series of local time domain segments of the signal strength fluctuation data, thereby effectively retaining the microscopic time domain structure of the fault response and improving the resolution of fault feature extraction.

[0035] More specifically, step S412 performs one-dimensional convolutional coding-based signal strength local fluctuation feature extraction on each local time domain segment of the signal strength fluctuation data in the time series of local time domain segments to obtain a time series of signal strength local time domain fluctuation feature encoding vectors. Specifically, the present application takes into account that different fault types may exhibit different waveform patterns in local segments (e.g., random spikes caused by a loose antenna, damped oscillations caused by a poor solder joint on a PCB), while the original signal strength fluctuation data contains a large amount of noise interference. Therefore, in order to effectively extract local time domain features that are strongly correlated with faults and suppress noise interference, this application adopts a one-dimensional convolutional neural network (1D-CNN) to perform feature encoding on each local time domain segment of signal strength fluctuation data respectively, so as to utilize the local perception characteristics of the one-dimensional convolution kernel and extract the key signal strength fluctuation patterns in the local time domain segment of each signal strength fluctuation data through a sliding convolution operation along the time dimension, such as edges, peaks and valleys, falling edges, oscillation patterns of specific frequencies, etc., and output a fixed-dimensional one-dimensional vector representation through a pooling operation to obtain a time series of signal strength local time domain fluctuation feature encoding vectors, thereby achieving data dimensionality reduction while retaining the fault fingerprint, and providing a solid information foundation for identifying microscopic fluctuation patterns unique to different fault types.

[0036] Specifically, the step S42 performs local-global double-layer context association perception coding on the time series of the signal strength local time domain fluctuation feature coding vector to obtain a global time domain context coding vector of the signal strength time series fluctuation feature. It should be understood that the present application takes into account that isolated signal strength fluctuation events may be just random noise, and the real fault response has a time series evolution characteristic. For example, under vibration excitation, the periodic change of the contact resistance of the PCB solder joint will cause the signal strength to present a regular attenuation oscillation mode, and the contact of the solder joint will deteriorate as the vibration time increases, causing the oscillation amplitude to gradually increase. The looseness of the antenna connector manifests itself as a random spike signal during the vibration process, and its frequency and amplitude increase with the increase of the vibration intensity. Therefore, in order to further capture the temporal evolution relationship and global distribution law of the local signal strength fluctuation events of the router, this application proposes a local-global double-layer context association perception coding method, which performs double-layer context association coding on the time series of the signal strength local time domain fluctuation feature coding vector. While capturing the global fluctuation evolution law of the signal strength, it enhances the semantic correlation between local fluctuation events, thereby obtaining a signal strength time series fluctuation feature global time domain context coding vector that takes into account both the local fine description of the signal strength fluctuation event and the macroscopic understanding of the global time series evolution. Among them, Figure 5 FIG. 4 is a flowchart of sub-step S42 of the WiFi router product testing method according to an embodiment of the present application. Figure 5 As shown, the step S42 includes the steps of: S421, performing local context enhancement perception based on adaptive neighborhood window anchoring on each signal strength local time domain fluctuation feature coding vector in the time series of the signal strength local time domain fluctuation feature coding vector to obtain a time series of signal strength local time domain fluctuation feature context enhancement coding vectors; S422, performing global time domain transfer coding on the time series of the signal strength local time domain fluctuation feature context enhancement coding vectors to obtain the signal strength time series fluctuation feature global time domain context coding vector.

[0037] More specifically, in step S421, each signal strength local time domain fluctuation feature coding vector in the time series of the signal strength local time domain fluctuation feature coding vector is subjected to local context enhancement perception based on adaptive neighborhood window anchoring to obtain a time series of signal strength local time domain fluctuation feature context enhancement coding vectors. Figure 6 FIG. 4 is a flowchart of sub-step S421 of the WiFi router product testing method according to an embodiment of the present application. Figure 6As shown, the step S421 includes the steps of: S4211, determining the window size of the local context enhanced perception window of each signal strength local time domain fluctuation feature coding vector based on the characteristic distribution of each signal strength local time domain fluctuation feature coding vector in the time series of the signal strength local time domain fluctuation feature coding vector; S4212, performing local neighborhood context enhanced perception on each signal strength local time domain fluctuation feature coding vector based on all the signal strength local time domain fluctuation feature coding vectors in the local context enhanced perception window of each signal strength local time domain fluctuation feature coding vector to obtain the time series of the signal strength local time domain fluctuation feature context enhanced coding vector.

[0038] In a specific example of the present application, step S4211 is expressed as follows: in, Indicates The exponential function operation with base , and They represent the first and The signal strength local time domain fluctuation feature encoding vector, Indicates the calculation of the Euclidean norm, represents the smoothing coefficient, express and The feature correlation between is the preset neighborhood, represents the main damping factor, represents the auxiliary damping factor, represents the index of the vector, Represents a preset neighborhood Neidi The signal strength local time domain fluctuation feature encoding vector is relative to The correlation weight coefficient, represents the logarithmic function with base 2, Represents a vector The neighbor feature distribution entropy of Indicates the preset maximum window size. express The window size of the local context enhancement perception window corresponding to the signal strength local time domain fluctuation feature encoding vector.

[0039] That is, by adaptively defining the local context range of each signal strength local time domain fluctuation feature encoding vector through feature distribution, the local context enhanced perception window can go beyond the simple temporal proximity relationship and anchor on the local environment that is most semantically relevant to the current feature vector, thereby achieving refined extraction of local time domain fluctuation features in high-frequency signal strength fluctuation data streams, and providing feature input with more semantic representation capabilities for subsequent fault diagnosis.

[0040] In a specific example of the present application, step S4212 includes: inputting all signal strength local time domain fluctuation feature encoding vectors in the local context enhanced perception window of the signal strength local time domain fluctuation feature encoding vector into the local neighborhood context enhanced fusion network based on the attention mechanism to obtain the signal strength local time domain fluctuation feature context enhanced encoding vector, which is expressed as follows: in, represents the normalized exponential function, 、 and Represents the different weight parameter matrices in the local neighborhood context enhancement fusion network, represents the transpose of a vector, express The characteristic dimension of express The corresponding attention weight factor, express activation function, express The corresponding signal strength local time domain fluctuation feature context enhanced coding vector.

[0041] That is, through the local neighborhood context enhancement fusion network, the signal strength local time domain fluctuation feature coding vector within the local context enhancement perception window is weightedly fused to strengthen the feature components related to the fault and suppress noise interference, thereby upgrading the signal strength local time domain fluctuation feature coding to a signal strength local time domain fluctuation feature context enhancement coding vector containing rich fault semantic information, providing a more discriminative local context feature representation for subsequent global time domain feature analysis, thereby improving the recognition accuracy of the deep learning model for hidden hardware faults and their types.

[0042] In particular, here, since the smoothing coefficient is introduced in the determination of the window size of the local semantic enhancement perception window and the main damping factor and auxiliary damping factor , so for local semantic embedding enhancement 、 and , synchronously integrate the deformation feature encoding mechanism to improve the contextual semantic embedding effect under the predetermined local semantic enhancement perception window. Based on this, in a preferred example of the present application, the step S4212 includes: performing local perception field adaptive weight adjustment optimization on the weight parameter matrix of the local neighborhood context enhancement fusion network to obtain an updated weight parameter matrix; then, based on the updated weight parameter matrix, performing context enhancement fusion based on the attention mechanism on all the signal strength local time domain fluctuation feature encoding vectors in the local context enhancement perception window of the signal strength local time domain fluctuation feature encoding vector to obtain the signal strength local time domain fluctuation feature context enhancement encoding vector.

[0043] Specifically, first for the deformation correlation vector , linear interpolation is performed to align the characteristic dimension of the signal intensity local time domain fluctuation feature encoding vector, and then the curvature feature transformation is realized by one-dimensional convolution to obtain the deformation connection feature vector , and respectively enhance the weight parameter matrix of the local neighborhood context fusion network 、 and As the mapping target, the deformation curvature associated eigenvector is obtained: in, represents the deformation connection eigenvector, 、 and Represent the weight parameter matrix 、 and The corresponding deformation curvature associated eigenvector.

[0044] Then, considering the direct correlation with the deformation curvature, the curvature-related dominance of the weighted strain can be simplified, that is, the high-order nonlinear strain under the non-intrinsic term can be ignored. Therefore, the above deformation curvature-related eigenvectors are respectively correlated with the eigenvectors of the weight parameter matrix, that is, 、 and To achieve coupling correction of the weight parameter matrix, it is associated with: in, 、 and Respectively 、 and The eigenvectors of represents the dot product operation, 、 and Respectively 、 and The corresponding updated weight parameter matrix.

[0045] In this way, the weight parameter matrix 、 and The local semantic embedding enhancement can take into account the deformation feature effect term introduced in the process of determining the window size of the local semantic enhancement perception window, thereby improving the robustness of the context representation of the local time domain fluctuation feature context enhancement coding vector of the signal strength based on the predetermined local semantic enhancement perception window.

[0046] More specifically, in a specific example of the present application, step S422 includes: inputting the time series of the signal strength local time domain fluctuation feature context enhancement coding vector into the global transfer coding network based on the Transformer architecture to obtain the signal strength time series fluctuation feature global time domain context coding vector, which is expressed as follows: in, Represents the time series of context-enhanced coding vectors of local temporal fluctuation characteristics of signal intensity, 、 and Respectively The first, second and The context-enhanced coding vector of the local temporal fluctuation feature of signal strength, represents the Transformer architecture, The global time-domain context coding vector representing the temporal fluctuation characteristics of signal strength.

[0047] That is, by leveraging the Transformer architecture's self-attention mechanism, the context-enhanced encoding vector of the local temporal fluctuation characteristics of signal strength at each time step can dynamically integrate the information of all other local temporal fluctuation characteristics of signal strength in the entire time series, thereby constructing a feature representation that includes global temporal context associations. This fusion of local fine features with global fluctuation patterns provides fault diagnosis with feature input that combines local details with a global perspective. The resulting global temporal context encoding vector of signal strength temporal fluctuation characteristics can effectively capture the complex dependencies and dynamic interaction patterns of signal strength fluctuation data streams across the entire time domain, such as the periodicity, burstiness, or persistence of signal fluctuations corresponding to different fault types in the global time domain, thereby improving the ability to identify complex fault scenarios such as multiple concurrent faults.

[0048] Specifically, in step S43, the stable signal strength indicator value and the global time-domain context encoding vector of the signal strength time-series fluctuation feature are input into a router fault diagnosis model based on a multi-layer perceptron to obtain the fault diagnosis result. It should be understood that the stable signal strength indicator value reflects the basic performance of the signal link element of the router, while the global time-domain context encoding vector of the signal strength time-series fluctuation feature reveals the connection structure defects of the router, and the two are diagnostically complementary. Therefore, in order to construct a multi-dimensional fault judgment criterion and output an interpretable classification result, this application designs a multi-layer perceptron (MLP) classifier based on the principle of joint decision-making of heterogeneous features to jointly learn and classify the stable signal strength indicator value and the global time-domain context encoding vector of the signal strength time-series fluctuation feature, so as to fully utilize the complementarity between the two and achieve comprehensive coverage and accurate identification of router hardware fault types. Specifically, the stable signal strength indicator is first normalized to the interval [0, 1] and concatenated with the global temporal context encoding vector of the signal strength temporal fluctuation feature. This is then fed into a three-layer fully connected network, each layer activated with Reluctant Units (ReLUs) and batch normalized. Finally, a softmax function is used to output a probability distribution for five fault categories, representing the likelihood that the router under test has no faults, a loose antenna connector, a poorly soldered PCB, a faulty signal link component, or multiple concurrent faults. Ultimately, by selecting the category with the highest probability as the fault diagnosis result, the system achieves efficient and automated diagnosis of hidden hardware faults in the router under test.

[0049] In summary, a WiFi router product testing method based on an embodiment of the present application is described. After the router under test turns on the WiFi signal of a specified frequency band and channel, it first records its stable signal strength indicator value at a standard distance. Then, a standardized micro-vibration excitation is applied to the router via a micro-vibration motor. During this period, a high-frequency data stream of signal strength fluctuations is collected. Furthermore, by introducing a deep learning algorithm, a dynamic excitation response analysis is performed on the high-frequency signal strength fluctuation data stream and stable signal strength value of the router under test. Based on the router's steady-state signal strength level and its signal strength fluctuation pattern under micro-vibration perturbations, the presence of a hidden hardware fault in the router and the type of fault are determined. This method can effectively stimulate and accurately identify hidden hardware faults caused by assembly tolerances, mechanical stress, etc., providing strong support for fault detection and quality control of router products.

[0050] The basic principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in the present invention are merely illustrative and non-limiting, and should not be construed as necessarily possessed by each embodiment of the present invention. Furthermore, the specific details of the above embodiments are provided for illustrative purposes and to facilitate understanding, and are not intended to be limiting. These details do not necessarily limit the present invention to being implemented using these specific details.

[0051] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0052] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be encompassed therein. Any reference to a figure in a claim should not be construed as limiting the claim to which it relates.

[0053] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0054] Finally, it should be noted that the above description has been provided for purposes of illustration and description. Furthermore, the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to be limiting. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art will appreciate that the technical solutions of the present invention may be modified or replaced with equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A WiFi router product testing method, characterized in that: include: Establish communication with the router under test through the preset serial port, drive the router under test to turn on the WiFi signal in the 2.4GHz or 5GHz band, and specify the test channel and maximum transmission power; Use a test antenna at a preset standard distance to receive the WiFi signal transmitted by the router to be tested, and record the stable signal strength indicator value; Applying standardized micro-vibration to the router under test through a micro-vibration motor, and collecting high-frequency signal intensity fluctuation data stream during the vibration; A deep learning-based dynamic stimulus response analysis is performed on the high-frequency signal strength fluctuation data stream and the stable signal strength indicator value to obtain a fault diagnosis result of the router to be tested.

2. The WiFi router product testing method according to claim 1, characterized in that: The fault diagnosis results are no fault, loose antenna connector, poor solder joint on PCB board, faulty signal link component or multiple concurrent faults.

3. The WiFi router product testing method according to claim 2, characterized in that: Performing a deep learning-based dynamic stimulus response analysis on the high-frequency signal strength fluctuation data stream and the stable signal strength indicator value to obtain a fault diagnosis result of the router to be tested, including: Performing local time-domain fluctuation analysis on the high-frequency signal strength fluctuation data stream to obtain a time series of signal strength local time-domain fluctuation feature encoding vectors; Performing local-global double-layer context association perceptual coding on the time series of the signal strength local time domain fluctuation feature coding vector to obtain a signal strength time series fluctuation feature global time domain context coding vector; The stable signal strength indicator value and the global time domain context coding vector of the signal strength time series fluctuation feature are input into a router fault diagnosis model based on a multi-layer perceptron to obtain the fault diagnosis result.

4. The WiFi router product testing method according to claim 3, characterized in that: Performing local time-domain fluctuation analysis on the high-frequency signal strength fluctuation data stream to obtain a time series of signal strength local time-domain fluctuation feature encoding vectors, including: Performing time-series segmentation on the high-frequency signal strength fluctuation data stream based on a preset time window to obtain a time series of local time domain segments of the signal strength fluctuation data; The signal strength local fluctuation feature extraction based on one-dimensional convolution coding is performed on each local time domain segment of the signal strength fluctuation data in the time series of the local time domain segments of the signal strength fluctuation data to obtain the time series of the signal strength local time domain fluctuation feature coding vector.

5. The WiFi router product testing method according to claim 4, characterized in that: Performing local-global dual-layer context association perceptual coding on the time series of the signal strength local time domain fluctuation feature coding vector to obtain a signal strength time series fluctuation feature global time domain context coding vector, including: Performing local context enhancement perception based on adaptive neighborhood window anchoring on each signal strength local time domain fluctuation feature coding vector in the time series of the signal strength local time domain fluctuation feature coding vector to obtain a time series of signal strength local time domain fluctuation feature context enhancement coding vectors; Global time domain transfer coding is performed on the time series of the signal strength local time domain fluctuation feature context enhancement coding vector to obtain the signal strength time series fluctuation feature global time domain context coding vector.

6. The WiFi router product testing method according to claim 5, characterized in that: Performing local context enhancement perception based on adaptive neighborhood window anchoring on each signal strength local time domain fluctuation feature coding vector in the time series of the signal strength local time domain fluctuation feature coding vector to obtain a time series of signal strength local time domain fluctuation feature context enhancement coding vectors, including: Determining a window size of a local context enhancement perception window of each signal strength local time domain fluctuation feature coding vector based on a feature distribution of each signal strength local time domain fluctuation feature coding vector in a time series of the signal strength local time domain fluctuation feature coding vector; Based on all the signal strength local time domain fluctuation feature coding vectors in the local context enhanced perception window of each signal strength local time domain fluctuation feature coding vector, local neighborhood context enhanced perception is performed on each signal strength local time domain fluctuation feature coding vector to obtain a time series of the signal strength local time domain fluctuation feature context enhanced coding vector.

7. The WiFi router product testing method according to claim 6, characterized in that: Based on all the signal strength local time domain fluctuation feature coding vectors in the local context enhanced perception window of each signal strength local time domain fluctuation feature coding vector, performing local neighborhood context enhanced perception on each signal strength local time domain fluctuation feature coding vector to obtain a time series of the signal strength local time domain fluctuation feature context enhanced coding vector, including: All the signal strength local time domain fluctuation feature coding vectors in the local context enhanced perception window of the signal strength local time domain fluctuation feature coding vector are input into the local neighborhood context enhanced fusion network based on the attention mechanism to obtain the signal strength local time domain fluctuation feature context enhanced coding vector.

8. The WiFi router product testing method according to claim 7, characterized in that: Inputting all signal strength local time domain fluctuation feature encoding vectors in the local context enhancement perception window of the signal strength local time domain fluctuation feature encoding vector into a local neighborhood context enhancement fusion network based on an attention mechanism to obtain the signal strength local time domain fluctuation feature context enhancement encoding vector, including: Performing local perception field adaptive weight adjustment optimization on the weight parameter matrix of the local neighborhood context enhancement fusion network to obtain an updated weight parameter matrix; Based on the updated weight parameter matrix, all the signal strength local time domain fluctuation feature coding vectors in the local context enhanced perception window of the signal strength local time domain fluctuation feature coding vector are subjected to context enhanced fusion based on the attention mechanism to obtain the signal strength local time domain fluctuation feature context enhanced coding vector.

9. The WiFi router product testing method according to claim 5, characterized in that: Performing global time-domain transfer coding on the time series of the signal strength local time-domain fluctuation feature context enhancement coding vector to obtain the signal strength time series fluctuation feature global time-domain context coding vector, including: The time series of the context enhancement coding vector of the local time domain fluctuation feature of the signal strength is input into the global transfer coding network based on the Transformer architecture to obtain the global time domain context coding vector of the signal strength temporal series fluctuation feature.

Images