Intelligent wearable control method and device resistant to strong magnetic interference, equipment and storage medium

Abstract

This application relates to the technical field of anti-interference for smart wearables, and particularly to a control method, device, equipment, and storage medium for smart wearables resisting strong magnetic interference. It includes: acquiring raw data sets from multiple sensors; determining whether the target smart wearable is under strong magnetic interference based on the raw magnetic field data; if the target smart wearable is under strong magnetic interference, mapping the raw magnetic field data to a preset level range to match the current interference level; determining a filtering processing mode matching the current interference level; applying the filtering processing mode to the raw data set to obtain a filtered data set; adjusting the operating mode of the target smart wearable based on the filtered data set and the current interference level; and integrating the adjusted operating mode with the current interference level into feedback information and outputting it. This application can dynamically adjust the filtering processing mode and the smart wearable's operating mode to cope with strong magnetic interference, ensuring data accuracy and system reliability.

Description

Technical Field

[0001] This application relates to the technical field of anti-interference for smart wearables, and in particular to a control method, device, equipment and storage medium for smart wearables that resist strong magnetic interference. Background Technology

[0002] Smartwatches, as commonly used wearable devices, integrate various sensors such as magnetometers and accelerometers to achieve functions such as health monitoring and navigation. However, in everyday use, smartwatches often encounter strong magnetic field interference from mobile phones, household appliances, or industrial equipment. This interference directly affects sensor accuracy, causing problems such as compass malfunction and deviations in movement trajectory, reducing device reliability and user experience.

[0003] To address the problem of strong magnetic interference, there are currently two main anti-interference methods: one is to install a metal shielding layer on the internal circuitry of the device to physically isolate the external magnetic field; the other is to use software algorithms to correct the collected abnormal data. However, once a physical shielding layer is installed, its protective strength cannot be dynamically adjusted according to the actual interference level, resulting in limited effectiveness in complex and variable strong magnetic environments. Furthermore, continuous use increases the device's power consumption and size. Software correction methods, on the other hand, typically lack real-time assessment of interference intensity, making it difficult to match the most appropriate correction strength when interference occurs, potentially leading to untimely or over-processing corrections.

[0004] Yes, the core problem with these two existing technical solutions is the lack of dynamic adaptability. Static processing modes are difficult to adapt to the dynamic changes of strong magnetic interference in real environments, and cannot optimize system resource allocation while ensuring data accuracy. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, this application provides a smart wearable control method, device, equipment, and storage medium that resists strong magnetic interference, so as to dynamically adjust the filtering processing mode and the working mode of the smart wearable device, thereby effectively dealing with strong magnetic interference, ensuring the accuracy of data and the reliability of the system, while optimizing resource allocation and improving user experience.

[0006] The technical solution adopted by this application to solve its technical problem is: In a first aspect, this application provides a smart wearable control method resistant to strong magnetic interference, applied to a target smart wearable equipped with multiple sensors, the method comprising: Acquire raw data sets from multiple sensors, including raw magnetic field data, and determine whether the target smart wearable is under strong magnetic interference based on the raw magnetic field data; If the target smart wearable is under strong magnetic interference, the original magnetic field data is mapped to a preset level range to match and obtain the current interference level; Determine the filtering mode that matches the current interference level, and apply the filtering mode to the original data group to obtain the filtered data group; The working mode of the target smart wearable is adjusted by combining the filtered data set with the current interference level, and the adjusted working mode and the current interference level are integrated into feedback information and output.

[0007] Optionally, the step of acquiring raw data sets from multiple sensors, including raw magnetic field data, and determining whether the target smart wearable is under strong magnetic interference based on the raw magnetic field data includes: The original magnetic field data for the current period is obtained based on a preset sampling frequency; Perform analog-to-digital conversion on the original magnetic field data to obtain the current magnetic field strength, and determine whether the current magnetic field strength exceeds a first preset threshold. If the current magnetic field strength exceeds a first preset threshold, it is determined whether the target smart wearable is in the state of strong magnetic interference.

[0008] Optionally, if the target smart wearable is under strong magnetic interference, the step of mapping the original magnetic field data to a preset level range and matching to obtain the current interference level includes: The current magnetic field strength is compared with multiple preset thresholds, including a first preset threshold, a second preset threshold, and a third preset threshold arranged in ascending order of value. If the current magnetic field strength is between the first preset threshold and the second preset threshold, then the current interference level is determined to be mild interference. If the current magnetic field strength is between the second preset threshold and the third preset threshold, then the current interference level is determined to be moderate interference. If the current magnetic field strength is greater than the third preset threshold, then the current interference level is determined to be severe interference.

[0009] Optionally, after determining that the current interference level is mild interference, the step of determining a filtering mode matching the current interference level and applying the filtering mode to the original data set to obtain a filtered data set includes: Acquire target data within a preset sliding window, wherein the target data includes the current magnetic field strength and associated sensor data from the original data set; The target data within the sliding window is subjected to extremum removal processing, and data points that deviate from the arithmetic mean by more than a preset standard deviation are removed; The arithmetic mean of the remaining valid data within the sliding window is calculated and written as the output value into the filtered data group, and the filtered data group is updated at a preset first time interval.

[0010] Optionally, after determining that the current interference level is moderate interference, the step of determining a filtering processing mode matching the current interference level and applying the filtering processing mode to the original data group to obtain a filtered data group includes: The model is constructed by building a state prediction model and an observation correction model through preset parameters. The state prediction model is used to characterize the relationship between the current state vector and the state vector at the previous time step, and includes a state transition matrix and process noise covariance. The observation correction model is used to characterize the relationship between the observation vector and the state vector, and includes an observation matrix and observation noise covariance. Performing a prediction includes obtaining the state vector of the previous time step, performing a prediction on the state vector of the previous time step using the state prediction model, and obtaining the predicted current state vector and the prediction error covariance. The update process includes determining the observation vector based on the original data set, combining the prediction error covariance, the observation vector, and the observation noise covariance, calculating the Kalman gain through the observation correction model, and correcting the predicted current state vector and update the error covariance based on the Kalman gain. The iterative output includes outputting the corrected current state vector as the filtered data set and updating the filtered data set at a preset second time interval.

[0011] Optionally, after determining that the current interference level is severe interference, the step of determining a filtering processing mode matching the current interference level and applying the filtering processing mode to the original data group to obtain a filtered data group includes: Enhance the process noise covariance and the observation noise covariance, and execute the model building step to the execution update step to obtain the data set to be processed; Obtain motion state data from the original data set and establish a data fusion matrix; Based on the data fusion matrix, the data group to be processed and the motion state data are weighted and calculated to obtain the filtered data group; The filtered data set is output and updated at a preset third time interval.

[0012] Optionally, the step of adjusting the working mode of the target smart wearable by combining the filtered data set with the current interference level includes: The target operating frequency is determined based on the current magnetic field strength corresponding to the current interference level, and the target operating frequency is dynamically mapped between a preset reference frequency and a preset minimum frequency. The system load rate and actual operating frequency are obtained. If the system load rate is lower than the preset load threshold, the actual operating frequency is increased until the target operating frequency or the preset load limit is reached, so as to adjust the working mode.

[0013] Secondly, this application provides a smart wearable control device resistant to strong magnetic interference, comprising: The data acquisition module is used to acquire raw data sets collected by various sensors, including raw magnetic field data, and to determine whether the target smart wearable is in a state of strong magnetic interference based on the raw magnetic field data. The interference determination module is used to map the original magnetic field data to a preset level range and match the current interference level if the target smart wearable is in a strong magnetic interference state. A filtering module is used to determine the filtering mode matching the current interference level, and to perform filtering on the original data group using the filtering mode to obtain a filtered data group. The mode adjustment module is used to adjust the working mode of the target smart wearable by combining the filtered data group with the current interference level, and to integrate the adjusted working mode with the current interference level into feedback information and output it.

[0014] Thirdly, this application provides an electronic device, comprising: One or more processors; One or more memory units; And one or more computer programs, wherein the one or more computer programs are stored in the one or more memories, and the one or more computer programs include instructions that, when executed by the one or more processors, cause the electronic device to perform the methods described above.

[0015] Fourthly, this application provides a computer-readable storage medium storing a program or instructions that, when executed, implement the above-described method.

[0016] The beneficial effects of this application are as follows: First, by collecting raw data from multiple sensors, it is determined whether the smart wearable device is under strong magnetic interference. Then, the raw magnetic field data is mapped to a preset level range to determine the current interference level. Based on different levels, the most suitable filtering mode for the current environment is selected to filter the raw data, thereby obtaining more accurate filtered data. Furthermore, by adjusting the operating mode of the target smart wearable device, system resource allocation and data processing efficiency are optimized. Finally, the adjusted operating mode and the current interference level are integrated into the feedback information output. This method can dynamically adjust the filtering mode and the operating mode of the smart wearable device, thereby effectively coping with strong magnetic interference, ensuring data accuracy and system reliability, while optimizing resource allocation, improving user experience, and enhancing the device's long-term performance. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the intelligent wearable control method against strong magnetic interference provided in the embodiments of this application; Figure 2 This is a virtual structural diagram of the intelligent wearable control device for resisting strong magnetic interference provided in this application; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] The present application will be further described below with reference to the accompanying drawings and embodiments.

[0019] The following will clearly and completely describe the concept, specific structure, and resulting technical effects of this application in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, features, and effects of this application. Obviously, the described embodiments are only a part of the embodiments of this application, not all of them. Other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are all within the scope of protection of this application. Furthermore, all connections / linkages involved in the patent do not simply refer to direct contact between components, but rather to the ability to form a better connection structure by adding or reducing connecting accessories according to specific implementation conditions. The various technical features in this application can be combined interactively without contradicting each other.

[0020] Smart wearables (such as smartwatches and smart bracelets), as wearable devices integrating multiple sensors, have been widely used in scenarios such as sports monitoring and health data collection. The accuracy of their sensors and the stability of their systems directly affect the user experience. In actual use, smartwatches are susceptible to strong magnetic interference from electronic devices and special working environments, which can lead to sensor data distortion, abnormal signal transmission, and unstable system operation.

[0021] Currently, there are two main solutions to the problem of strong magnetic interference: one is to add a metal shielding layer at the hardware level to reduce the impact of external magnetic fields on internal components through physical isolation; the other is to use a fixed algorithm at the software level to correct sensor data in order to offset the errors caused by interference.

[0022] It can be seen that the anti-interference measures of the two commonly used solutions lack dynamic adaptability. Specifically, when the external magnetic field strength fluctuates, the fixed hardware shielding layer cannot adjust the protection level according to the interference level. In mild interference scenarios, excessive shielding may lead to energy waste and functional redundancy, while in severe interference scenarios, it is difficult to provide sufficient shielding capability, causing sensitive components to be continuously affected by the magnetic field. This static protection mode cannot achieve a balance between shielding effectiveness and device power consumption in complex electromagnetic environments, affecting both battery life and the long-term stability of sensor data.

[0023] Meanwhile, the lack of a dynamic adjustment mechanism for commonly used anti-interference algorithms can cause data processing algorithms to become disconnected from real-time interference conditions. Fixed filtering algorithm parameters cannot cope with dynamic changes in magnetic field strength, making it difficult to effectively remove noise when the interference level increases, leading to increased sensor data errors. Conversely, when the interference weakens, over-filtering may cause data lag, affecting the real-time performance and accuracy of functions such as motion monitoring and health data collection. This passive compensation method cannot establish a dynamic correlation between the degree of interference and the processing strategy, resulting in the data acquisition accuracy of smartwatches remaining unstable in strong magnetic environments.

[0024] Based on this, please refer to Figure 1 , Figure 1 This is a flowchart illustrating the intelligent wearable control method against strong magnetic interference provided in this application embodiment. The diagram includes several key steps provided by this solution, which dynamically adjust the filtering processing mode and the working mode of the intelligent wearable device. Each key step is described in detail below: In step S1, raw data sets collected by multiple sensors are acquired, including raw magnetic field data. Based on the raw magnetic field data, it is determined whether the target smart wearable is in a state of strong magnetic interference.

[0025] The raw data set refers to the unprocessed initial data set collected by smart wearable devices (such as smartwatches) through their built-in sensors, which includes various types of data such as magnetic fields, motion, and health. The raw magnetic field data is the original value of the ambient magnetic field strength, which is monitored and output in real time by the magnetometer built into the smart wearable device.

[0026] Specifically, the target smart wearable first collects raw data sets through its built-in magnetic field monitoring module (such as a magnetometer) and other sensors (such as an accelerometer and gyroscope), with the core data being the real-time monitored raw magnetic field data. After receiving the raw magnetic field data, the data processing module compares it with a preset magnetic field strength threshold. If the raw magnetic field data exceeds the threshold, the target smart wearable is directly determined to be in a state of strong magnetic interference; if it does not exceed the threshold, it is determined to be in a normal state. By real-time acquisition of raw magnetic field data and threshold comparison, rapid identification of strong magnetic interference is achieved, providing triggering conditions for subsequent processes such as initiating shielding measures and optimizing sensor data.

[0027] More specifically, in this embodiment of the application, the step of acquiring a set of raw data collected by multiple sensors, the raw data set including raw magnetic field data, and determining whether the target smart wearable is under strong magnetic interference based on the raw magnetic field data includes: The original magnetic field data for the current period is obtained based on a preset sampling frequency; The preset sampling frequency refers to the sensor data acquisition interval pre-set by the smart wearable device, that is, the number of times raw magnetic field data is collected per unit time, which is used to control the real-time performance of data acquisition and energy consumption balance; the current period is a single data acquisition period corresponding to the preset sampling frequency. For example, when the sampling frequency is 10 times per second, the current period is 0.1 seconds.

[0028] Further, an analog-to-digital conversion is performed on the original magnetic field data to obtain the current magnetic field strength, and it is determined whether the current magnetic field strength exceeds a first preset threshold. Analog-to-digital conversion is the process of converting analog signals (such as continuous physical quantities like magnetic field strength) collected by sensors into digital signals (discrete values) so that the data processing module can perform calculations and analyses. The current magnetic field strength is the specific value of the magnetic field strength obtained after analog-to-digital conversion, which is the direct basis for judging strong magnetic interference. The unit is usually microtesla (μT). The first preset threshold is a magnetic field strength benchmark value preset in the data processing module, which is used to determine whether to trigger the judgment of strong magnetic interference. For example, the initial threshold of 50μT for mild interference.

[0029] Furthermore, if the current magnetic field strength exceeds a first preset threshold, it is determined whether the target smart wearable is under strong magnetic interference.

[0030] Specifically, the target smart wearable device first collects raw magnetic field data using a magnetometer within the current cycle according to a preset sampling frequency (e.g., 10 times per second), ensuring the real-time and continuous nature of the data. Next, the analog-to-digital conversion module converts the collected analog magnetic field signal into a digital form representing the current magnetic field strength, for example, converting the analog signal into a specific value "60μT". Subsequently, the data processing module compares this current magnetic field strength with a first preset threshold (e.g., 50μT). If the current magnetic field strength exceeds this threshold, the target smart wearable device is directly determined to be in a state of strong magnetic interference. In this step, the preset sampling frequency enables timed acquisition of magnetic field data, balancing real-time monitoring with device power consumption; the analog-to-digital conversion ensures the accurate conversion of the raw signal into a computable digital signal, providing a reliable data foundation for subsequent judgments.

[0031] More specifically, taking a smartwatch as an example, its preset sampling frequency is set to 10 times per second, meaning each current cycle is 0.1 seconds. In a scenario where the user is near an induction cooker (with an surrounding magnetic field strength of approximately 70 μT), the magnetometer collects raw magnetic field data within the 0.1-second current cycle, which is then converted from analog to digital to obtain the current magnetic field strength "70 μT". The data processing module compares this value with a first preset threshold "50 μT" and finds that 70 μT exceeds the threshold, thus determining that the smartwatch is in a state of strong magnetic interference. If the user is in a normal indoor environment (with a magnetic field strength of approximately 30 μT), the current magnetic field strength does not exceed the threshold, and is therefore considered to be in a normal state.

[0032] In step S2, if the target smart wearable is in a state of strong magnetic interference, the original magnetic field data is mapped to a preset level range to match and obtain the current interference level.

[0033] The preset level range refers to the pre-defined range of magnetic field strength values. Each range corresponds to a specific level of interference (such as mild, moderate, and severe), which is used to quantify the original magnetic field data into an executable interference level standard. The current interference level is the specific interference level identifier obtained by matching the preset level range into which the original magnetic field data falls.

[0034] Specifically, once the target smart wearable is determined to be under strong magnetic interference, the data processing module will call preset level range parameters. In one embodiment, the preset level range parameters can be mild interference (50μT-100μT), moderate interference (100μT-200μT), and severe interference (>200μT). The system then compares and maps the currently collected raw magnetic field data with these ranges. For example, if the raw magnetic field data is 150μT, the system will match it to the range corresponding to "moderate interference," thereby determining the current interference level to be moderate.

[0035] More specifically, in this embodiment of the application, the step of mapping the original magnetic field data to a preset level range and matching the current interference level if the target smart wearable is in a state of strong magnetic interference includes: The current magnetic field strength is compared with multiple preset thresholds, including a first preset threshold, a second preset threshold, and a third preset threshold arranged in ascending order of value. If the current magnetic field strength is between the first preset threshold and the second preset threshold, then the current interference level is determined to be mild interference. If the current magnetic field strength is between the second preset threshold and the third preset threshold, then the current interference level is determined to be moderate interference. If the current magnetic field strength is greater than the third preset threshold, then the current interference level is determined to be severe interference.

[0036] Among them, multiple preset thresholds refer to a set of pre-set magnetic field strength critical values ​​arranged from smallest to largest, used to delineate the boundaries of different interference levels, including a first preset threshold, a second preset threshold, and a third preset threshold; specifically, the second preset threshold is a magnetic field strength standard between the first preset threshold and the third preset threshold, used to define the dividing point between mild interference and moderate interference, and the third preset threshold is the highest value among the preset thresholds, used to define the dividing point between moderate interference and severe interference.

[0037] Specifically, when the target smart wearable is under strong magnetic interference, the data processing module calls multiple preset thresholds (arranged in ascending order as the first, second, and third preset thresholds) and compares the current magnetic field strength with these thresholds sequentially. If the current magnetic field strength is between the first and second preset thresholds, the current interference level is determined to be mild; if it is between the second and third preset thresholds, it is determined to be moderate interference; and if it is greater than the third preset threshold, it is determined to be severe interference. The entire process achieves precise matching of interference levels through clearly defined numerical ranges.

[0038] More specifically, by dividing the data into multiple threshold intervals, continuous magnetic field strength values ​​are transformed into discrete interference levels (mild, moderate, and severe), avoiding the limitations of single threshold judgment and enabling more precise quantification of interference levels. Different levels correspond to different anti-interference strategies, including but not limited to reducing the sampling frequency for mild cases and activating multi-layer shielding for severe cases, thereby reducing resource waste while ensuring anti-interference effectiveness and improving the system's response targeting and efficiency.

[0039] More specifically, in one embodiment, a smartwatch has several preset thresholds set as follows: a first preset threshold of 50 μT, a second preset threshold of 100 μT, and a third preset threshold of 200 μT. When a user carries the watch through a strong magnetic field area, the current magnetic field strength is detected as 150 μT. After comparing this with the thresholds, the system finds that 150 μT falls between the second preset threshold (100 μT) and the third preset threshold (200 μT), thus determining the current interference level to be moderate interference.

[0040] In step S3, the filtering processing mode matching the current interference level is determined, and the original data group is filtered using the filtering processing mode to obtain the filtered data group.

[0041] Among them, the filtering processing mode refers to the set of signal processing strategies preset according to different interference levels, including filtering algorithm types (such as mean filtering, Kalman filtering) and their corresponding filtering parameters, which are used to specifically eliminate magnetic field interference noise. Among them, the filtered data set is the optimized data set obtained after the original data set has been processed by the filtering mode. The magnetic field data noise is suppressed and can be directly used for subsequent sensor function calculations (such as compass and attitude detection).

[0042] Specifically, after the target smart wearable determines the current interference level (mild, moderate, or severe), it matches the corresponding filtering mode from the system's preset strategy library. In this embodiment, mild interference is matched with "basic mean filtering," moderate interference with "adaptive Kalman filtering," and severe interference with "multi-sensor fusion filtering." Subsequently, the data processing module calls the filtering mode to perform filtering calculations on the magnetic field data and associated motion data in the original data set, removing abnormal fluctuations introduced by strong magnetic interference, and finally outputting the noise-suppressed filtered data set.

[0043] More specifically, by dynamically matching the interference level with the filtering mode, lightweight algorithms, such as mean filtering, are used for mild interference to reduce power consumption while ensuring real-time performance; complex fusion algorithms, such as multi-sensor filtering, are used for severe interference to correct magnetic field deviations with the help of motion data, thereby improving data reliability. This differentiated processing avoids data lag caused by over-filtering and prevents noise interference left by insufficient filtering, ensuring that subsequent sensor functions (such as navigation and attitude recognition) can still work stably in strong magnetic environments.

[0044] More specifically, in one embodiment, in the policy library of a smartwatch, mild interference (50μT-100μT) corresponds to "5-window mean filtering," and moderate interference (100μT-200μT) corresponds to "adaptive Kalman filtering." When the user is in a moderate interference environment (current magnetic field strength 150μT), the system matches the "adaptive Kalman filtering" mode, iteratively correcting noise bias by fusing accelerometer motion data and magnetic field data. For example, if the original magnetic field data has fluctuations of ±20μT, after filtering, the fluctuation amplitude is reduced to ±5μT. The resulting filtered data set can accurately support the compass function, avoiding directional drift caused by interference.

[0045] More specifically, in this embodiment of the application, after determining that the current interference level is mild interference, the step of determining a filtering processing mode matching the current interference level and applying the filtering processing mode to perform filtering processing on the original data group to obtain a filtered data group includes: Acquire target data within a preset sliding window, wherein the target data includes the current magnetic field strength and associated sensor data from the original data set; The preset sliding window refers to a pre-defined fixed-length data buffer used to store multiple sets of continuously acquired target data. The window size is usually set according to the sampling frequency and real-time requirements, for example, containing 5 sets of continuous data. The target data is the set of raw data to be processed within the sliding window, including the current magnetic field strength acquired in the current period, as well as other sensor data associated with the magnetic field data (such as motion parameters of accelerometers and gyroscopes). The associated sensor data is auxiliary sensor information acquired synchronously with the magnetic field data, such as the device motion acceleration recorded by the accelerometer and the rotational angular velocity recorded by the gyroscope, which is used to improve the accuracy of data filtering.

[0046] Furthermore, extreme value removal processing is performed on the target data within the sliding window to remove data points that deviate from the arithmetic mean by more than a preset standard deviation multiple; Among them, outlier removal is the process of detecting and removing outliers from the target data within the sliding window. It uses statistical methods to identify and remove data points that deviate from the normal range, reducing extreme noise interference. The preset standard deviation multiple is a statistical threshold used to determine whether data is an outlier. For example, it is set to "3 times the standard deviation", which means that when a data point deviates from the arithmetic mean by more than 3 times the standard deviation, it is determined to be an outlier.

[0047] Furthermore, the arithmetic mean of the remaining valid data within the sliding window is calculated and written as the output value into the filtered data group, and the filtered data group is updated at a preset first time interval.

[0048] Among them, the effective data is the target data that remains after the extreme value removal process and is within the normal fluctuation range. It is the basis for calculating the filtered output value. The first time interval refers to the update cycle of the filtered data group, that is, the newly calculated average value is written into the filtered data group at fixed intervals to control the frequency of data output, such as updating once every 0.1 seconds.

[0049] Specifically, after determining the current interference level to be mild, the system first extracts target data from a preset sliding window. This data includes the current magnetic field strength and synchronously acquired data from associated sensors. Next, it performs extremum removal processing on the target data within the sliding window: calculating the arithmetic mean and standard deviation of all data points, and removing outlier data points that deviate from the mean by more than a preset standard deviation multiple (e.g., 3 times), such as jump values ​​caused by instantaneous strong magnetic fluctuations. Subsequently, it calculates the arithmetic mean of the remaining valid data, writes this mean as the filtered output value into the filtered data set, and periodically updates the filtered data set according to a first time interval to ensure the real-time nature and continuity of the data.

[0050] More specifically, a preset sliding window is used to achieve local smoothing of the data, avoiding the impact of fluctuations in a single data point on the results; extremum removal effectively eliminates extreme noise (such as instantaneous strong magnetic interference), reducing the distortion of the average value by outliers; arithmetic mean calculation further smooths data fluctuations, while the first time interval update balances the real-time performance of data output with system power consumption while ensuring the filtering effect. The overall processing is designed for mild interference scenarios, with low algorithm complexity and high computational efficiency. It can suppress mild noise without introducing significant processing delays, ensuring that smart wearable devices can maintain the stability and availability of sensor data in mild strong magnetic environments.

[0051] More specifically, in one embodiment, a smartwatch operates under mild interference (current magnetic field strength 70 μT, first preset threshold 50 μT, second preset threshold 100 μT), with a preset sliding window size of 5 data sets, a preset standard deviation multiplier of 2, and a first time interval of 0.1 seconds. Assume the target data within the sliding window contains 5 magnetic field strength values: 70 μT, 72 μT, 95 μT (outlier), 68 μT, and 71 μT. First, the arithmetic mean is calculated to be 75.2 μT, with a standard deviation of approximately 9.8 μT. The preset standard deviation multiplier of 2 corresponds to a threshold of 19.6 μT. The outlier 95 μT deviates from the mean by 19.8 μT (exceeding 19.6 μT) and is therefore discarded. The remaining valid data are 70 μT, 72 μT, 68 μT, and 71 μT. The arithmetic mean is calculated to be 70.25 μT, which is written as the output value into the filtered data set, and the calculation result for the next window is updated after 0.1 seconds.

[0052] More specifically, in this embodiment of the application, after determining that the current interference level is moderate interference, the step of determining a filtering processing mode matching the current interference level, and applying the filtering processing mode to perform filtering processing on the original data group to obtain a filtered data group includes: The model is constructed by building a state prediction model and an observation correction model through preset parameters. The state prediction model is used to characterize the relationship between the current state vector and the state vector at the previous time step, and includes a state transition matrix and process noise covariance. The observation correction model is used to characterize the relationship between the observation vector and the state vector, and includes an observation matrix and observation noise covariance. The state prediction model is a mathematical model used to predict the current state based on the state vector of the previous time step. Its core parameters include the state transition matrix (describing the change of the state vector over time) and the process noise covariance (characterizing the statistical properties of random disturbances during the prediction process), and its expression is as follows: ; in, This is the current state vector (containing magnetic field strength and sensor data). This is the state transition matrix; For the control matrix, This is process noise; This is the external control vector.

[0053] The observation correction model is used to correct the predicted state model by combining actual observation data. It includes the observation matrix (describing the mapping relationship between the state vector and the observation vector) and the observation noise covariance (characterizing the statistical properties of noise in the observation data), and its expression is: ; in, The observed value at the current moment, For the observation matrix, To observe noise.

[0054] The state vector contains vectors of key state parameters of the target system, such as continuously changing physical quantities like magnetic field strength and equipment acceleration in magnetic field filtering; the observation vector is the vector of actual measured values ​​converted from the original data set, containing the magnetic field strength at the current moment and associated sensor data (such as acceleration and angular velocity).

[0055] Specifically, after determining the current interference level to be moderate, the system first constructs a filtering model: a state prediction model and an observation correction model are constructed using preset parameters. The state prediction model includes the state transition matrix and the process noise covariance; the observation correction model includes the observation matrix and the observation noise covariance.

[0056] Performing a prediction includes obtaining the state vector of the previous time step, performing a prediction on the state vector of the previous time step using the state prediction model, and obtaining the predicted current state vector and the prediction error covariance. The prediction error covariance is a covariance matrix representing the deviation between the predicted state vector and the true state vector, reflecting the uncertainty of the prediction result. The expressions involved in this step include: the state vector at the previous time step. Covariance of prediction error Where Q is the process noise covariance, This represents the error covariance at the previous time step.

[0057] Specifically, the state vector of the previous moment (such as the magnetic field strength and acceleration of the previous moment) is obtained, substituted into the state prediction model, and the predicted current state vector is calculated through the state transition matrix. The prediction error covariance is then updated based on the process noise covariance.

[0058] The update process includes determining the observation vector based on the original data set, combining the prediction error covariance, the observation vector, and the observation noise covariance, calculating the Kalman gain through the observation correction model, and correcting the predicted current state vector and update the error covariance based on the Kalman gain. The Kalman gain is a weighting coefficient used to balance the reliability of the predicted state and the observation vector. It is calculated using the prediction error covariance, the observation matrix, and the observation noise covariance, and determines the degree to which the observation data corrects the predicted state. Its expression is: ; Where R is the observation noise covariance.

[0059] The update equation used is: ; The update error covariance is the covariance matrix of the deviation between the current state vector and the true state vector after observation correction. It is used to initialize the prediction error at the next time step, and its expression is: .

[0060] Specifically, observation vectors (such as the measured value of the current magnetic field strength) are extracted based on the original data set. These vectors are then combined with the prediction error covariance, the observation matrix, and the observation noise covariance to calculate the optimal weighting coefficients, i.e., the Kalman gain, using the Kalman gain formula. This gain is then used to correct the predicted current state vector, while simultaneously updating the error covariance, resulting in the corrected current state vector.

[0061] The iterative output includes outputting the corrected current state vector as the filtered data set and updating the filtered data set at a preset second time interval.

[0062] The second time interval is the update cycle of the filtered data group under moderate interference scenario, that is, the corrected state vector is written into the filtered data group at fixed intervals (such as once every 0.05 seconds) to balance real-time performance and computational resource consumption.

[0063] Specifically, the corrected current state vector is used as the output value of the filtered data set, and the filtered data set is updated periodically according to the second time interval (e.g., 0.05 seconds) to ensure the real-time performance and continuity of the data.

[0064] More specifically, in handling moderate interference, dynamic noise suppression is achieved through the Kalman filtering algorithm (a closed-loop process of state prediction + observation correction). Compared to mean filtering for mild interference, this method can more accurately track the time-varying characteristics of magnetic field data. The state prediction model predicts trends based on historical data, while the observation correction model dynamically adjusts based on real-time observations. The two are fused using Kalman gain weighting, effectively balancing prediction bias and observation noise, thus improving data stability under moderate interference. The second time interval (shorter than the first time interval for mild interference) ensures real-time performance while avoiding the waste of computational resources caused by high-frequency updates, achieving an optimized balance between filtering accuracy and system efficiency.

[0065] More specifically, in the embodiments of this application, after determining that the current interference level is severe interference, the step of determining a filtering processing mode matching the current interference level, and applying the filtering processing mode to perform filtering processing on the original data group to obtain a filtered data group includes: Enhance the process noise covariance and the observation noise covariance, and execute the model building step to the execution update step to obtain the data set to be processed; The data set to be processed refers to the intermediate data set obtained after processing the original data set through the Kalman filter model with enhanced noise covariance. It contains magnetic field data that has been preliminarily filtered, but may still contain some interference noise.

[0066] Further, the motion state data in the original data set is obtained, and a data fusion matrix is ​​established; Among them, motion state data are sensor data related to the device's motion in the raw data set, such as three-dimensional acceleration collected by accelerometer and rotational angular velocity collected by gyroscope, which are used to reflect the device's real-time motion posture, including but not limited to being stationary, walking, and rotating; The data fusion matrix is ​​used to characterize the weight relationship between the data group to be processed and the motion state data. The fusion ratio between the two is dynamically allocated through preset rules (such as motion intensity threshold). For example, when the motion state is "intense exercise", the weight of the motion state data is increased.

[0067] Furthermore, the data group to be processed and the motion state data are weighted and calculated based on the data fusion matrix to obtain the filtered data group; The filtered data set is output and updated at a preset third time interval.

[0068] The third time interval is the update cycle of the filtered data set under severe interference scenarios. It is usually shorter than the time interval for mild and moderate interference (e.g., updating once every 0.02 seconds) to adapt to the rapid fluctuations of magnetic field data under strong magnetic interference.

[0069] Specifically, when the current interference level is determined to be severe interference, the values ​​of process noise covariance (Q) and observation noise covariance (R) are first increased (e.g., from Q=0.01, R=0.1 to Q=0.05, R=0.5) to improve the model's tolerance to extreme noise. Then, the model building, prediction, and update steps are performed (consistent with moderate interference) to obtain the data set to be processed. Next, motion state data (such as x / y / z-axis acceleration from the accelerometer and angular velocity from the gyroscope) are extracted from the original data set, and a data fusion matrix is ​​established based on preset rules. Subsequently, a weighted calculation is performed based on the data fusion matrix: filtered data set = data set to be processed × weight 1 + motion state data × weight 2, where weight 1 and weight 2 are dynamically output by the fusion matrix; weight 1 can be the weight of the motion state data, and weight 2 can be the weight of the data set to be processed. Finally, the calculation result is output as the filtered data set and updated periodically at a third time interval (e.g., 0.02 seconds) to ensure data real-time performance.

[0070] More specifically, the anti-severe interference measures involved in this step, on the one hand, enhance the noise covariance (Q, R) to reduce the impact of extreme observations (strong magnetic interference) on the results of the Kalman filter model during the prediction and update stages; on the other hand, motion state data fusion utilizes the correlation between motion and magnetic field (e.g., magnetic field data should change synchronously when the equipment rotates) and eliminates outliers through multi-sensor cross-validation. For example, when strong magnetic interference causes a jump in magnetic field data but the motion state data shows the equipment is stationary, the fusion matrix will reduce the weight of the data group to be processed to avoid erroneous output.

[0071] More specifically, in one embodiment, a smartwatch, under a severe interference scenario (magnetic field strength 250 μT, exceeding the preset threshold of 200 μT), enhances the process noise covariance Q to 0.05 and the observation noise covariance R to 0.5. After performing Kalman filtering, the resulting data set is 240 μT (which may still be affected by residual interference). At this point, the motion state data in the original data set shows an acceleration of 0.3 m / s², and the data fusion matrix assigns weights accordingly: the data set to be processed has a weight of 0.7, and the motion state data (correlated with the magnetic field baseline value of 200 μT) has a weight of 0.3. Through weighted calculation: the filtered data set = 240 μT × 0.7 + 200 μT × 0.3 = 168 μT + 60 μT = 228 μT. This result is updated at a third time interval of 0.02 seconds, effectively suppressing the influence of the 250 μT strong magnetic interference and making the output data closer to the true magnetic field level.

[0072] In step S4, the working mode of the target smart wearable is adjusted by combining the filtered data set with the current interference level, and the adjusted working mode and the current interference level are integrated into feedback information and output.

[0073] The working mode refers to the set of operating configurations of the target smart wearable device in different scenarios, including sensor sampling frequency, system operating frequency, functional module activation status (such as compass, health monitoring), data update cycle, etc. The feedback information includes comprehensive status information on the current interference level and the adjusted working mode, which is used to display the device's operating status to users or internal system modules (such as the application layer).

[0074] Specifically, after obtaining the filtered data set and the current interference level, the system first inputs both as parameters to the working mode adjustment module. This module has a built-in preset rule library. For example, when the current interference level is mild, the sensor's default sampling frequency (e.g., 100Hz) is maintained, and all functional modules are enabled; when the interference level is moderate, the sampling frequency is increased to 200Hz, and unnecessary health monitoring functions are turned off to allocate more computing power to filtering; when the interference level is severe, the system switches to "anti-interference enhancement mode," increases the sampling frequency to 500Hz, and retains only the core compass function while pausing secondary functions such as motion trajectory recording.

[0075] Subsequently, the system adjusts the hardware configuration (such as sensor sampling rate) and software function switches of the target smart wearable according to the above rules, completing the dynamic switching of the working mode. Then, the adjusted working mode (such as "anti-interference enhancement mode") is integrated with the current interference level (such as "severe interference") into structured feedback information, which is output through the device display screen, APP notification or system log to ensure that the user or subsequent modules can keep track of the device status in real time.

[0076] In addition, the system can calculate the difference index between the filtered data set and the original data set, and compare it with a preset data difference threshold. For example, when the difference index is less than or equal to the data difference threshold, the filtering effect is considered good, and the interference has little impact on the core functions; when the difference index is greater than the data difference threshold, the original data is considered to be severely interfered with, and non-essential functions need to be further restricted to ensure the concentration of computing power.

[0077] More specifically, this step achieves dynamic optimization of equipment resources by linking interference levels with operating modes: maintaining functional integrity and low power consumption under mild interference; balancing accuracy and performance under moderate interference; and prioritizing the stability of core functions (such as the compass) under severe interference. This adaptive adjustment avoids resource waste or functional failure caused by a "one-size-fits-all" fixed configuration, improving the equipment's robustness in complex electromagnetic environments. Simultaneously, the output of feedback information enhances system transparency, allowing users to intuitively understand the interference status and equipment response strategies, reducing user confusion caused by functional abnormalities.

[0078] More specifically, in this embodiment of the application, the step of adjusting the working mode of the target smart wearable by combining the filtered data set with the current interference level includes: The target operating frequency is determined based on the current magnetic field strength corresponding to the current interference level, and the target operating frequency is dynamically mapped between a preset reference frequency and a preset minimum frequency. The target operating frequency is the operating frequency of the device's core processor, which is dynamically determined based on the current interference level. It is used to match the computing power requirements under different interference scenarios, and its value ranges between the preset base frequency and the preset minimum frequency. The preset base frequency is the default operating frequency of the device in an environment with no interference or slight interference, ensuring the smoothness of basic functions and the balance of power consumption; the preset minimum frequency refers to the minimum processor frequency (such as 100MHz) that the device needs to maintain basic operation, and is only enabled in extremely low load or low power scenarios. Dynamic mapping refers to the process of converting the current interference level into a target operating frequency using a preset algorithm. For example, for each increase in the interference level, the target frequency increases by 50% based on the reference frequency. In this embodiment, the algorithm used is as follows: Target frequency = preset minimum frequency + (preset reference frequency - preset minimum frequency) × [1 - (current magnetic field strength - reference magnetic field strength) / (maximum preset interference strength - reference magnetic field strength)]; Specifically, a linear mapping relationship is used to bind the current magnetic field strength to the target operating frequency, where the target operating frequency decreases linearly as the magnetic field strength increases. For example, the maximum preset interference strength of a smartwatch is 400μT. When the user enters a strong magnetic field environment (current magnetic field strength = 350μT), the formula is calculated as follows: Numerator = 350 - 200 = 150μT, Denominator = 400 - 200 = 200μT, Proportional term = 150 / 200 = 0.75, Dynamic adjustment term = 120 × (1 - 0.75) = 30MHz, Target frequency = 48 + 30 = 78MHz. At this time, the frequency drops from the reference 168MHz to 78MHz, reducing power consumption while ensuring the basic computing power for anti-interference filtering, thus adapting to the needs of strong magnetic interference scenarios.

[0079] Furthermore, the system load rate and actual operating frequency are obtained. If the system load rate is lower than a preset load threshold, the actual operating frequency is increased until the target operating frequency or the preset load limit is reached, so as to adjust the working mode.

[0080] Specifically, the built-in performance monitoring module collects the system load rate (current processor resource utilization ratio) and actual operating frequency (current processor operating frequency) in real time. Then, the system load rate is compared with a preset load threshold. If the actual load rate is lower than the threshold, for example, the current load rate is 50% and the preset threshold is 70%, it is determined that the device has computing power redundancy, and the frequency boosting mechanism is activated.

[0081] During the upgrade process, the system gradually increases the actual operating frequency in a step-by-step manner (e.g., 50MHz each time), and immediately re-detects the system load rate after each adjustment. If the load rate is still below the preset load threshold after the upgrade and the actual operating frequency has not reached the target operating frequency, the upgrade continues; if the system load rate has reached the preset load limit before reaching the target operating frequency, the frequency upgrade stops, and the current frequency is taken as the final adjustment result. Through this process, the actual operating frequency is dynamically adjusted to the critical value corresponding to the target value or the load limit, thereby optimizing the computing power configuration in the working mode.

[0082] By adopting this step, we can avoid the waste of computing power caused by fixed frequency, such as high-frequency operation consuming power under low load, and prevent insufficient computing power, such as forced overclocking under high load causing crashes. Ultimately, we can improve the stability and energy efficiency of the equipment in complex electromagnetic environments.

[0083] Reference Figure 2 , Figure 2 This is a virtual structural diagram of the intelligent wearable control device resistant to strong magnetic interference provided in this application. Secondly, this application provides an intelligent wearable control device resistant to strong magnetic interference, comprising: The data acquisition module 100 is used to acquire raw data sets collected by various sensors, including raw magnetic field data, and to determine whether the target smart wearable is in a strong magnetic interference state based on the raw magnetic field data. The interference determination module 200 is used to map the original magnetic field data to a preset level range and match the current interference level if the target smart wearable is in a strong magnetic interference state. The filtering module 300 is used to determine the filtering mode matching the current interference level, and to perform filtering on the original data group using the filtering mode to obtain the filtered data group. The mode adjustment module 400 is used to adjust the working mode of the target smart wearable by combining the filtered data group with the current interference level, and to integrate the adjusted working mode with the current interference level into feedback information and output it.

[0084] The intelligent wearable control device for resisting strong magnetic interference described in this application embodiment can execute the intelligent wearable control method for resisting strong magnetic interference provided in the above embodiment. The intelligent wearable control device for resisting strong magnetic interference has the corresponding functional steps and beneficial effects of the intelligent wearable control method for resisting strong magnetic interference described in the above embodiment. For details, please refer to the embodiments of the intelligent wearable control method for resisting strong magnetic interference described above. The embodiments of this application will not be repeated here.

[0085] This application also provides an electronic device, please refer to... Figure 3 , Figure 3This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include a processor and a memory, which can be connected via a bus or other means. The processor may be a Central Processing Unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips. The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the anti-strong magnetic interference smart wearable control method in the embodiments of this application. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory, thereby realizing the anti-strong magnetic interference smart wearable control method in the above method embodiments.

[0086] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. One or more modules are stored in the memory and, when executed by the processor, perform the intelligent wearable control method against strong magnetic interference as described in the above method embodiments. Specific details of the above electronic device can be understood by referring to the corresponding descriptions and effects in the above method embodiments, and will not be repeated here. Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it may include the processes of the embodiments of the above methods. The storage medium may be a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above types of memory.

[0087] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification. It should be noted that the foregoing embodiments are illustrative and not restrictive of this application, and that alternative embodiments may be devised by those skilled in the art without departing from the scope of the appended claims.

Claims

1. A smart wearable control method resistant to strong magnetic interference, characterized in that, Applied to a target smart wearable device equipped with multiple sensors, the method includes: Acquire raw data sets from multiple sensors, including raw magnetic field data, and determine whether the target smart wearable is under strong magnetic interference based on the raw magnetic field data; If the target smart wearable is under strong magnetic interference, the original magnetic field data is mapped to a preset level range to match and obtain the current interference level; Determine the filtering mode matching the current interference level, and apply the filtering mode to the original data group to obtain the filtered data group; wherein, mild interference is matched with basic mean filtering, moderate interference is matched with adaptive Kalman filtering, and severe interference is matched with multi-sensor fusion filtering. The working mode of the target smart wearable is adjusted by combining the filtered data set with the current interference level, and the adjusted working mode and the current interference level are integrated into feedback information and output. The step of adjusting the working mode of the target smart wearable by combining the filtered data set with the current interference level includes: determining the target operating frequency according to the current magnetic field strength corresponding to the current interference level, wherein the target operating frequency is dynamically mapped between a preset reference frequency and a preset minimum frequency; obtaining the system load rate and the actual operating frequency; if the system load rate is lower than a preset load threshold, increasing the actual operating frequency until it reaches the target operating frequency or a preset load upper limit, so as to adjust the working mode.

2. The intelligent wearable control method against strong magnetic interference according to claim 1, characterized in that, The step of acquiring raw data sets from multiple sensors, including raw magnetic field data, and determining whether the target smart wearable is under strong magnetic interference based on the raw magnetic field data includes: The original magnetic field data for the current period is obtained based on a preset sampling frequency; Perform analog-to-digital conversion on the original magnetic field data to obtain the current magnetic field strength, and determine whether the current magnetic field strength exceeds a first preset threshold. If the current magnetic field strength exceeds a first preset threshold, it is determined whether the target smart wearable is in the state of strong magnetic interference.

3. The intelligent wearable control method against strong magnetic interference according to claim 2, characterized in that, If the target smart wearable is under strong magnetic interference, the step of mapping the original magnetic field data to a preset level range and matching the current interference level includes: The current magnetic field strength is compared with multiple preset thresholds, including a first preset threshold, a second preset threshold, and a third preset threshold arranged in ascending order of value. If the current magnetic field strength is between the first preset threshold and the second preset threshold, then the current interference level is determined to be mild interference. If the current magnetic field strength is between the second preset threshold and the third preset threshold, then the current interference level is determined to be moderate interference. If the current magnetic field strength is greater than the third preset threshold, then the current interference level is determined to be severe interference.

4. The intelligent wearable control method against strong magnetic interference according to claim 3, characterized in that, After determining that the current interference level is mild interference, the step of determining the filtering mode matching the current interference level and applying the filtering mode to the original data group to obtain the filtered data group includes: Acquire target data within a preset sliding window, wherein the target data includes the current magnetic field strength and associated sensor data from the original data set; The target data within the sliding window is subjected to extremum removal processing, and data points that deviate from the arithmetic mean by more than a preset standard deviation are removed; The arithmetic mean of the remaining valid data within the sliding window is calculated and written as the output value into the filtered data group, and the filtered data group is updated at a preset first time interval.

5. The intelligent wearable control method against strong magnetic interference according to claim 3, characterized in that, After determining that the current interference level is moderate interference, the step of determining the filtering mode matching the current interference level and applying the filtering mode to the original data group to obtain the filtered data group includes: The model is constructed by building a state prediction model and an observation correction model through preset parameters. The state prediction model is used to characterize the relationship between the current state vector and the state vector at the previous time step, and includes a state transition matrix and process noise covariance. The observation correction model is used to characterize the relationship between the observation vector and the state vector, and includes an observation matrix and observation noise covariance. Performing a prediction includes obtaining the state vector of the previous time step, performing a prediction on the state vector of the previous time step using the state prediction model, and obtaining the predicted current state vector and the prediction error covariance. The update process includes determining the observation vector based on the original data set, combining the prediction error covariance, the observation vector, and the observation noise covariance, calculating the Kalman gain through the observation correction model, and correcting the predicted current state vector and update the error covariance based on the Kalman gain. The iterative output includes outputting the corrected current state vector as the filtered data set and updating the filtered data set at a preset second time interval.

6. The intelligent wearable control method against strong magnetic interference according to claim 5, characterized in that, After determining that the current interference level is severe interference, the step of determining the filtering mode matching the current interference level and applying the filtering mode to the original data group to obtain the filtered data group includes: Enhance the process noise covariance and the observation noise covariance, and execute the model building step to the execution update step to obtain the data set to be processed; Obtain motion state data from the original data set and establish a data fusion matrix; Based on the data fusion matrix, the data group to be processed and the motion state data are weighted and calculated to obtain the filtered data group; The filtered data set is output and updated at a preset third time interval.

7. A smart wearable control device resistant to strong magnetic interference, characterized in that, For use in target smart wearables, the target smart wearables are equipped with a variety of sensors, including: The data acquisition module is used to acquire raw data sets collected by various sensors, including raw magnetic field data, and to determine whether the target smart wearable is in a state of strong magnetic interference based on the raw magnetic field data. The interference determination module is used to map the original magnetic field data to a preset level range and match the current interference level if the target smart wearable is in a strong magnetic interference state. The filtering module is used to determine the filtering mode matching the current interference level, and to perform filtering on the original data group using the filtering mode to obtain the filtered data group; wherein, mild interference is matched with basic mean filtering, moderate interference is matched with adaptive Kalman filtering, and severe interference is matched with multi-sensor fusion filtering. The mode adjustment module is used to adjust the working mode of the target smart wearable by combining the filtered data set with the current interference level, and to integrate the adjusted working mode with the current interference level into feedback information and output it. The step of adjusting the working mode of the target smart wearable by combining the filtered data set with the current interference level includes: determining the target operating frequency according to the current magnetic field strength corresponding to the current interference level, wherein the target operating frequency is dynamically mapped between a preset reference frequency and a preset minimum frequency; obtaining the system load rate and the actual operating frequency; if the system load rate is lower than a preset load threshold, increasing the actual operating frequency until it reaches the target operating frequency or a preset load upper limit, so as to adjust the working mode.

8. An electronic device, characterized in that, include: One or more processors; One or more memory units; And one or more computer programs, wherein the one or more computer programs are stored in the one or more memories, the one or more computer programs including instructions that, when executed by the one or more processors, cause the electronic device to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a program or instructions that, when executed, implement the method as described in any one of claims 1 to 6.

Images