A wireless earphone voice data transmission abnormal mode recognition method and system

Abstract

This application provides a method and system for identifying abnormal patterns in voice data transmission of wireless headphones, including: acquiring instantaneous electromagnetic pulse information of the external electromagnetic environment; acquiring operating status information of the internal power supply unit; acquiring voice processing status information of the main control chip; in response to the acquisition of instantaneous electromagnetic pulse information, initiating an observation period, and within the observation period, establishing a correlation between the external instantaneous electromagnetic pulse and the internal processing abnormality based on the instantaneous electromagnetic pulse information, operating status information, and voice processing status information, so as to identify complex delay abnormal patterns. This application can establish a correlation between external instantaneous electromagnetic pulses and internal processing abnormalities, thereby identifying complex delay abnormal patterns, solving the problem that existing technologies cannot identify such complex fault modes.

Description

Technical Field

[0001] This application relates to the field of wireless communication technology, and more specifically, to a method and system for identifying abnormal patterns in voice data transmission from wireless headphones. Background Technology

[0002] When faced with transient high-intensity external electromagnetic pulse interference, existing wireless headphones can detect the momentary deterioration in signal quality and attempt to recover. However, this transient interference not only directly affects the communication link but also indirectly triggers temporary instability in the headphone's internal power management unit, which may have insufficient design margins. This, in turn, causes intermittent internal processing delays or data errors in the main control chip when processing voice data. This internal processing anomaly is not directly and continuously caused by the external interference but is a consequence of changes in the internal hardware state triggered by the transient external interference.

[0003] Existing identification methods struggle to establish a clear causal relationship between external transient interference events and subsequent internal processing anomalies of different natures, and are unable to identify complex, delayed failure modes caused by the combined effect of strong external transient interference and insufficient internal hardware margins. Summary of the Invention

[0004] This application discloses a method and system for identifying abnormal patterns in voice data transmission of wireless headphones, aiming to solve the problem in the prior art that it is difficult to establish a clear causal relationship between external instantaneous interference events and subsequent internal processing anomalies of different natures, thus making it impossible to identify complex and delayed fault modes.

[0005] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, this application discloses a method for recognizing abnormal patterns in voice data transmission from wireless earphones, including: Acquire instantaneous electromagnetic pulse information of the external electromagnetic environment, including the occurrence time, duration, and power distribution of the instantaneous electromagnetic pulse; Obtain the operating status information of the internal power supply unit, which includes the output voltage fluctuation and instantaneous current change of the power supply unit; Obtain the voice processing status information of the main control chip, which includes the internal clock frequency deviation of the main control chip, cache access latency, and voice decoder processing pause time; In response to the acquisition of transient electromagnetic pulse information, an observation cycle is initiated. Within this observation cycle, based on the transient electromagnetic pulse information, operating status information, and voice processing status information, a correlation is established between the external transient electromagnetic pulse and the internal processing anomaly in order to identify composite delayed anomaly patterns.

[0006] Secondly, this application also discloses a wireless earphone voice data transmission abnormality pattern recognition system, comprising: The instantaneous electromagnetic pulse information acquisition module is used to acquire instantaneous electromagnetic pulse information of the external electromagnetic environment. The instantaneous electromagnetic pulse information includes the occurrence time, duration and power distribution of the instantaneous electromagnetic pulse. The internal power supply unit operation status information acquisition module is used to acquire the operation status information of the internal power supply unit, which includes the output voltage fluctuation and instantaneous current change of the power supply unit. The main control chip voice processing status information acquisition module is used to acquire the voice processing status information of the main control chip. The voice processing status information includes the internal clock frequency deviation of the main control chip, the cache access delay, and the voice decoder processing pause time. The abnormal pattern recognition module is used to initiate an observation cycle in response to the acquisition of instantaneous electromagnetic pulse information. Within the observation cycle, based on the instantaneous electromagnetic pulse information, operating status information, and voice processing status information, it establishes the correlation between the external instantaneous electromagnetic pulse and the internal processing abnormality in order to identify composite delayed abnormal patterns. Beneficial effects

[0007] This application acquires instantaneous electromagnetic pulse information from the external electromagnetic environment, the operating status information of the internal power supply unit, and the voice processing status information of the main control chip. Upon acquisition of the instantaneous electromagnetic pulse information, an observation cycle is initiated. Within this observation cycle, a correlation is established between the external instantaneous electromagnetic pulse and internal processing anomalies based on this information, thereby identifying composite delayed anomaly patterns. This method effectively solves the technical problem in existing technologies that it is difficult to establish a clear causal relationship between external instantaneous interference events and subsequent internal processing anomalies of different natures. By comprehensively analyzing external electromagnetic interference and internal hardware status, this application can identify composite delayed fault modes triggered by strong external instantaneous interference and caused by insufficient internal hardware margins, overcoming the shortcomings of existing technologies that can only record external signal anomalies and internal processing timeouts in isolation. Therefore, this application can take targeted measures, such as performing additional stabilization operations on the internal power supply or processing unit after the external interference ends, or adjusting the data processing strategy according to the real-time operating status of the power management unit, thereby avoiding atypical anomalies in voice quality that persist after the external interference ends, significantly improving the user experience. Attached Figure Description

[0008] Figure 1 A flowchart illustrating a method for identifying abnormal patterns in voice data transmission from a wireless headset, as provided in this application; Figure 2 This application provides a schematic diagram of the structure of a wireless earphone voice data transmission abnormality pattern recognition system. Detailed Implementation

[0009] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0010] like Figure 1 As shown, this application proposes a method for identifying abnormal patterns in voice data transmission from wireless headphones, including: Acquire instantaneous electromagnetic pulse information of the external electromagnetic environment, including the occurrence time, duration, and power distribution of the instantaneous electromagnetic pulse; Obtain the operating status information of the internal power supply unit, including the output voltage fluctuation and instantaneous current change of the power supply unit; Obtain the voice processing status information of the main control chip, which includes the internal clock frequency deviation of the main control chip, cache access latency, and voice decoder processing pause time. In response to the acquisition of transient electromagnetic pulse information, an observation cycle is initiated. Within the observation cycle, based on the transient electromagnetic pulse information, operating status information, and voice processing status information, a correlation is established between the external transient electromagnetic pulse and the internal processing anomaly in order to identify composite delayed anomaly patterns.

[0011] This application, by comprehensively analyzing external instantaneous electromagnetic pulse information, internal power supply unit operating status information, and main control chip voice processing status information, can establish the correlation between external instantaneous electromagnetic pulses and internal processing anomalies, thereby identifying complex delayed anomaly modes and effectively solving the problem that existing technologies cannot identify such complex fault modes.

[0012] In order to make the technical solution of this application easier and clearer to understand, it is necessary to explain some key terms involved therein.

[0013] Transient electromagnetic pulse (EMI) information refers to a burst of electromagnetic energy with high intensity and a wide spectrum that occurs within an extremely short time. Its characteristics include the time of occurrence, duration, and power distribution. Such pulses may originate from electrostatic discharge, power supply transients, or external electromagnetic interference sources.

[0014] The operating status information of the internal power supply unit refers to the working status of the power management module inside the wireless headset, including fluctuations in its output voltage and changes in instantaneous current. These parameters directly reflect the stability of the power supply unit.

[0015] The voice processing status information of the main control chip refers to the internal working state of the wireless headset's main control chip when processing voice data, including the deviation of its internal clock frequency, the access latency of the cache, and the pause time of the voice decoder. These parameters are key indicators for measuring the processing efficiency and stability of the main control chip.

[0016] The composite delayed anomaly mode refers to a fault mode triggered by an external transient electromagnetic pulse, causing a series of chain reactions in the internal power supply unit and main control chip, ultimately manifesting as delayed or abnormal voice data transmission. The characteristic of this anomaly mode is its delayed nature; that is, the internal anomaly persists for a period of time after the external interference ends.

[0017] This application establishes a correlation between external instantaneous electromagnetic pulses and internal processing anomalies through multi-dimensional information fusion, thereby identifying complex delayed anomaly patterns.

[0018] Specifically, this method first requires acquiring instantaneous electromagnetic pulse information from the external electromagnetic environment. This can be achieved by integrating an electromagnetic field sensor inside the wireless headset. This sensor can monitor the surrounding electromagnetic environment in real time, recording the occurrence time, duration, and power distribution when a transient high-intensity electromagnetic pulse is detected. For example, a broadband radio frequency receiver can be used to extract the characteristics of the instantaneous electromagnetic pulse by performing spectrum analysis and time-domain analysis on the received signal. Another implementation method is to use a dedicated electromagnetic pulse detection circuit. When this circuit detects electromagnetic energy exceeding a preset threshold, it triggers an interrupt and records relevant parameters.

[0019] Simultaneously, it is also necessary to obtain the operating status information of the internal power supply unit. This can be achieved by setting voltage and current sensors at the output end and key nodes of the power supply unit. These sensors can monitor the output voltage fluctuations and instantaneous current changes of the power supply unit in real time. For example, a high-precision analog-to-digital converter (ADC) can be used to continuously sample the output voltage of the power supply unit and calculate its root mean square (RMS) value and peak-to-peak value to reflect voltage fluctuations. Instantaneous current changes can be measured using Hall effect sensors or shunt resistors.

[0020] In addition, it is necessary to obtain the voice processing status information of the main control chip. This can be achieved by integrating a performance monitoring module inside the main control chip. This module can monitor the internal clock frequency deviation, cache access latency, and voice decoder processing pause time of the main control chip in real time. For example, the internal clock frequency deviation can be obtained by comparing the phase difference between the internal clock of the main control chip and a high-precision reference clock. Cache access latency can be calculated by recording the start and end times of each cache access. Voice decoder processing pause time can be obtained by monitoring the decoder's operating status register or by injecting test data and measuring its processing time.

[0021] Upon acquiring transient electromagnetic pulse information, the system responsively initiates an observation cycle. The purpose of this observation cycle is to continuously monitor the status of the internal power supply unit and the main control chip for a period of time after the external transient interference occurs, in order to detect potential delay anomalies. During the observation cycle, the system establishes a correlation between the external transient electromagnetic pulse and internal processing anomalies based on the acquired transient electromagnetic pulse information, the operating status information of the internal power supply unit, and the voice processing status information of the main control chip. For example, a correlation can be determined by analyzing whether the voltage fluctuation of the internal power supply unit and the voice processing delay of the main control chip increase statistically significantly after the transient electromagnetic pulse occurs. If a significant correlation exists, it is identified as a composite delay anomaly pattern.

[0022] The wireless earphone voice data transmission abnormality pattern recognition method of this application can establish the correlation between external instantaneous electromagnetic pulses and internal processing abnormalities by comprehensively analyzing external instantaneous electromagnetic pulse information, internal power supply unit operation status information and main control chip voice processing status information, thereby identifying composite delay abnormality patterns.

[0023] Specifically, when instantaneous electromagnetic pulse information from the external electromagnetic environment is acquired, the system immediately initiates an observation cycle. During this observation cycle, the system continuously collects operational status information from the internal power supply unit and voice processing status information from the main control chip. By synchronously analyzing this multi-source heterogeneous data—for example, using time series analysis, correlation analysis, or machine learning models—the system can identify causal relationships or strong correlations between instantaneous electromagnetic pulse events and subsequent internal voltage fluctuations, current changes, clock frequency deviations, buffer access delays, and voice decoder pause times. For example, if, after an instantaneous electromagnetic pulse occurs, the output voltage of the internal power supply unit experiences a significant dip or fluctuation within a short period, and the subsequent pause time of the main control chip's voice decoder processing increases significantly, then a preliminary correlation can be identified. In this way, this application can effectively identify composite, delayed fault modes caused by the combined effect of external instantaneous interference and potential internal hardware margin deficiencies.

[0024] Compared to existing technologies, this application can establish a correlation between external transient electromagnetic pulses and internal processing anomalies, thereby identifying complex delayed anomaly patterns. Traditional anomaly identification methods typically can only detect external signal anomalies or internal processing timeouts in isolation, failing to link the two. For example, when a transient electromagnetic pulse causes a brief instability in the power management unit inside the headset, leading to a delay in the main control chip's voice processing, existing methods may only record two independent events: the external signal anomaly and the internal processing timeout, without identifying the causal chain between them.

[0025] This application acquires detailed information about transient electromagnetic pulses (occurrence time, duration, power distribution) and combines it with the real-time operating status of the internal power supply unit (output voltage fluctuations, instantaneous current changes) and the voice processing status of the main control chip (internal clock frequency deviation, cache access latency, voice decoder processing pause time). An observation cycle is initiated after the transient electromagnetic pulse occurs, enabling the capture of the delaying effects of external interference on the internal system. This multi-dimensional, cross-domain correlation analysis allows this application to identify composite, delayed fault modes caused by the combined effect of strong external transient interference and potential internal hardware margin deficiencies—something existing technologies cannot achieve. By identifying these composite delayed anomaly modes, the system can take more effective targeted measures, such as performing additional stabilization operations on the internal power supply or processing unit after the external interference ends, or adjusting the data processing strategy based on the real-time operating status of the power management unit. This significantly improves the stability of voice data transmission and user experience in complex electromagnetic environments for wireless headphones.

[0026] In some embodiments of this application, the steps of initiating an observation period in response to the acquisition of transient electromagnetic pulse information, and establishing a correlation between external transient electromagnetic pulses and internal processing anomalies within the observation period based on transient electromagnetic pulse information, operating status information, and voice processing status information, in order to identify composite delayed anomaly patterns, include: Upon receiving transient electromagnetic pulse information, the background electromagnetic noise intensity information of the current environment is obtained; Based on the background electromagnetic noise intensity information, the identification threshold for internal anomalies is adjusted, and the duration of the observation period is also adjusted. Within the adjusted observation period, based on instantaneous electromagnetic pulse information, the adjusted internal anomaly identification threshold, the operating status information of the internal power supply unit, and the voice processing status information of the main control chip, a correlation is established between external instantaneous electromagnetic pulses and internal processing anomalies to identify composite delayed anomaly patterns.

[0027] Specifically, upon receiving transient electromagnetic pulse information, the system first acquires the background electromagnetic noise intensity information of the current environment. This background electromagnetic noise intensity information can be understood as the general level of electromagnetic interference in the environment when no significant transient electromagnetic pulse event occurs, and its purpose is to provide a dynamic benchmark for subsequent anomaly identification. In practical applications, background electromagnetic noise intensity information can be monitored and acquired in real time through electromagnetic sensors or radio frequency receiving units integrated inside the headphones, for example, by averaging or statistically analyzing the signal strength of a specific frequency band.

[0028] The identification threshold and observation period are dynamically adjusted based on background electromagnetic noise intensity information. The identification threshold is the standard for judging whether the internal operating state or voice processing state is abnormal, aiming to distinguish between normal fluctuations and actual anomalies. The observation period refers to the time window during which the system continuously monitors the internal state to establish a correlation after an instantaneous electromagnetic pulse occurs, aiming to capture delayed anomalies. Specifically, when the background electromagnetic noise intensity is high, the identification threshold can be appropriately increased to avoid misjudging normal fluctuations caused by noise as anomalies; simultaneously, the observation period can be appropriately extended to ensure sufficient data collection even in complex noise environments, thereby more accurately identifying true delayed anomalies. Conversely, when the background electromagnetic noise intensity is low, the identification threshold can be appropriately decreased to improve sensitivity to weak anomalies; the observation period can be appropriately shortened to improve identification efficiency.

[0029] Therefore, within the adjusted observation period, the system will establish a correlation between external instantaneous electromagnetic pulses and internal processing anomalies based on instantaneous electromagnetic pulse information, adjusted internal anomaly recognition thresholds, operating status information of the internal power supply unit, and voice processing status information of the main control chip, in order to identify composite delayed anomaly patterns. This means that anomaly judgment and correlation establishment will no longer be based on fixed parameters, but will be adaptively adjusted according to real-time environmental conditions, thereby improving the accuracy and reliability of recognition.

[0030] This application addresses the problem of accurately identifying complex, delayed anomaly patterns in complex and variable electromagnetic environments by introducing the acquisition of background electromagnetic noise intensity information and dynamically adjusting the identification threshold and observation period based on this information. Specifically, when a transient electromagnetic pulse occurs, the system not only focuses on the pulse itself but also considers the overall electromagnetic background of the current environment. By evaluating the intensity of background electromagnetic noise, the system can intelligently adjust the sensitivity of anomaly judgment (identification threshold) and the time window for data collection (observation period). For example, in a high-noise environment, increasing the identification threshold can effectively filter out noise interference and prevent false alarms; extending the observation period helps to capture delayed true anomaly signals under noise masking. In a low-noise environment, decreasing the identification threshold can improve the detection capability of subtle anomalies, while shortening the observation period can accelerate the identification speed. This adaptive adjustment mechanism allows the anomaly pattern recognition process to better adapt to changes in the actual working environment, thereby more accurately distinguishing between anomalies directly caused by transient electromagnetic pulses and normal fluctuations caused by background noise or other factors, and establishing a more accurate correlation between external transient electromagnetic pulses and internal processing anomalies.

[0031] Through the above technical solution, this application can significantly improve the accuracy and robustness of abnormal pattern recognition in wireless headset voice data transmission. By dynamically acquiring background electromagnetic noise intensity information and adjusting the recognition threshold and observation period accordingly, the system can effectively avoid false alarms or missed alarms caused by improper fixed parameter settings in complex electromagnetic environments, making the anomaly recognition process more environmentally adaptable. This not only improves the recognition accuracy of complex delayed anomaly patterns but also optimizes the use of system resources. For example, in low-noise environments, the observation period can be shortened, thereby reducing computational load and energy consumption, providing users with a more stable and reliable voice data transmission experience.

[0032] In some embodiments of this application, the step of establishing a correlation between external transient electromagnetic pulses and internal processing anomalies to identify composite delayed anomaly patterns within the adjusted observation period, based on transient electromagnetic pulse information, adjusted internal anomaly identification thresholds, operating status information of the internal power supply unit, and voice processing status information of the main control chip, includes: Obtain the operating status information of the internal power supply unit and the voice processing status information of the main control chip; Spectrum analysis is performed on the operating status information of the internal power supply unit to extract the inherent fluctuation characteristics of the internal power supply unit; Timing pattern analysis is performed on the voice processing status information of the main control chip to extract the internal processing anomaly features of the main control chip; Obtain the spectral characteristics of dynamic background electromagnetic noise; The inherent fluctuation characteristics of the internal power supply unit, the abnormal internal processing characteristics of the main control chip, and the spectral characteristics of dynamic background electromagnetic noise are compared and contrasted. Based on the results of the differential comparison, the influence of external instantaneous electromagnetic pulses and dynamic background electromagnetic noise on the operating status information of the internal power supply unit and the voice processing status information of the main control chip is quantified. Based on the quantification results, a correlation is established between external instantaneous electromagnetic pulses and internal processing anomalies to identify complex delayed anomaly patterns.

[0033] Specifically, acquiring the operating status information of the internal power supply unit and the voice processing status information of the main control chip refers to the system continuously monitoring and collecting the output voltage fluctuations and instantaneous current changes of the power supply unit, as well as the internal clock frequency deviation, cache access latency, and voice decoder processing pause time of the main control chip. This information forms the basis for subsequent analysis. Spectral analysis of the operating status information of the internal power supply unit aims to reveal the energy distribution of the power supply unit at different frequencies, thereby identifying its inherent, periodic fluctuation patterns, such as power supply ripple or internal oscillations, which may exist even without external interference. Furthermore, temporal pattern analysis of the voice processing status information of the main control chip aims to identify specific time-series patterns that may occur during voice processing, such as periodic delays, sudden pauses, or abnormal clock drifts, which may indicate internal processing anomalies. In addition, acquiring the spectral characteristics of dynamic background electromagnetic noise can be understood as using sensors or receiving units inside the headset to monitor and analyze the intensity distribution of background electromagnetic noise in the current environment at different frequencies in real time, in order to understand the interference characteristics of the external environment.

[0034] In practical applications, the inherent fluctuation characteristics of the internal power supply unit, the internal processing anomaly characteristics of the main control chip, and the spectral characteristics of dynamic background electromagnetic noise are compared differentially. Methods such as correlation analysis, energy comparison, or pattern matching can be used to differentiate the impact of these signals from different sources on the internal operating state and voice processing state of the headset. Specifically, based on the differential comparison results, the influence ratio of external transient electromagnetic pulses and dynamic background electromagnetic noise on the operating state information of the internal power supply unit and the voice processing state information of the main control chip is quantified. This involves establishing a mathematical model or using machine learning algorithms to calculate the proportion or contribution of external transient electromagnetic pulses and dynamic background electromagnetic noise in causing internal anomalies based on the comparison results. Finally, based on the quantification results, a correlation is established between external transient electromagnetic pulses and internal processing anomalies to identify composite delay anomaly patterns. The purpose is to determine whether voice data transmission anomalies are mainly caused by external transient electromagnetic pulses based on the quantified influence ratios, thereby accurately identifying composite delay anomaly patterns caused by both external transient electromagnetic pulses and internal processing anomalies.

[0035] This application utilizes spectral analysis of the internal power supply unit's operating status information to identify and extract its inherent fluctuation patterns, thereby distinguishing them from fluctuations caused by external interference. Simultaneously, temporal pattern analysis of the main control chip's voice processing status information reveals specific temporal behavior patterns of internal processing anomalies. By acquiring the spectral characteristics of dynamic background electromagnetic noise and comparing them with inherent internal fluctuation characteristics and internal processing anomaly characteristics, this application can accurately separate and quantify the respective influence proportions of external instantaneous electromagnetic pulses and dynamic background electromagnetic noise on the headset's internal operating and voice processing states. This refined quantitative analysis enables the system to more accurately determine whether voice data transmission anomalies are caused by external instantaneous electromagnetic pulses, background electromagnetic noise, or inherent internal factors, thus establishing a more reliable correlation and effectively avoiding the inaccurate identification problems caused by confusing multiple interference sources in traditional methods.

[0036] Through the above technical solution, this application can achieve more accurate identification of abnormal modes in voice data transmission of wireless earphones. Specifically, by conducting in-depth spectral and temporal analysis of the internal operating state and voice processing state, and combining this with differential comparison and quantification of the spectral characteristics of dynamic background electromagnetic noise, this application can effectively distinguish the impact of external instantaneous electromagnetic pulses, background electromagnetic noise, and internal inherent factors on system performance. This significantly improves the accuracy and reliability of establishing the correlation between external instantaneous electromagnetic pulses and internal processing anomalies, thereby enabling more precise identification of complex delay anomaly modes and providing more refined and effective data support for fault diagnosis and performance optimization of wireless earphones.

[0037] In some embodiments of this application, the step of establishing the correlation between external transient electromagnetic pulses and internal processing anomalies based on quantization results to identify composite delayed anomaly modes includes: Obtain information on the complexity of the electromagnetic environment in which the headphones are currently located. This information includes the intensity level of background electromagnetic noise and the frequency of external instantaneous electromagnetic pulses. Obtain information on the aging level of the internal hardware of the headphones, including the cumulative working time of the internal power supply unit and the cumulative processing load of the main control chip; Based on information on the complexity of the electromagnetic environment and hardware aging, a correlation strength judgment threshold is calculated. The calculation process includes: lowering the correlation strength judgment threshold when the background electromagnetic noise intensity level is high; further lowering the correlation strength judgment threshold when the frequency of external instantaneous electromagnetic pulses is high; and lowering the correlation strength judgment threshold again when the cumulative working time of the internal power supply unit or the cumulative processing load of the main control chip exceeds a preset value. Compare the quantification results with the correlation strength judgment threshold; If the quantification result exceeds the correlation strength judgment threshold, a correlation is established between the external instantaneous electromagnetic pulse and the internal processing anomaly in order to identify the composite delayed anomaly pattern.

[0038] Specifically, before establishing the correlation between external transient electromagnetic pulses and internal processing anomalies, it is first necessary to obtain information on the complexity of the electromagnetic environment in which the wireless headset is currently located. This information mainly includes the intensity level of background electromagnetic noise and the frequency of occurrence of external transient electromagnetic pulses. The intensity level of background electromagnetic noise reflects the level of persistent interference in the environment, while the frequency of occurrence of external transient electromagnetic pulses indicates the density of transient interference events. This information can be obtained through real-time monitoring and analysis using electromagnetic sensors or radio frequency receiving units integrated within the headset.

[0039] Simultaneously, it is also necessary to obtain information on the aging level of the internal hardware of the headphones. This aging level information mainly includes the cumulative operating time of the internal power supply unit and the cumulative processing load of the main control chip. The cumulative operating time of the internal power supply unit can reflect the wear and tear of its components such as capacitors and batteries, while the cumulative processing load of the main control chip can reflect the fatigue level of its components such as transistors and interconnects. This information can be obtained and accumulated through firmware recording or sensor monitoring within the headphones.

[0040] Based on this, an association strength judgment threshold is calculated according to the acquired information on the complexity of the electromagnetic environment and the degree of hardware aging. This calculation process is dynamically adjusted: when the background electromagnetic noise intensity level is high, the environmental interference itself is strong, and even small internal anomalies may be associated with external pulses. Therefore, the association strength judgment threshold needs to be lowered to improve the sensitivity of identification. When the frequency of external instantaneous electromagnetic pulses is high, it indicates that instantaneous interference events are frequent, and the internal system is more susceptible to cumulative effects. In this case, the association strength judgment threshold should also be further lowered. In addition, when the cumulative working time of the internal power supply unit or the cumulative processing load of the main control chip exceeds the preset value, it indicates that the hardware is aging and its resistance to electromagnetic interference is reduced, making it more prone to anomalies. Therefore, the association strength judgment threshold needs to be lowered again to ensure that even weak associations can be identified.

[0041] Subsequently, the quantization results obtained in the above steps are compared with the calculated correlation strength judgment threshold. If the quantization result exceeds the correlation strength judgment threshold, it can be determined that there is a significant correlation between the external instantaneous electromagnetic pulse and the internal processing anomaly, thereby establishing a correlation to identify composite delayed anomaly patterns.

[0042] This application addresses the issue of insufficient recognition accuracy under complex and variable environments and hardware aging conditions by introducing information on the complexity of the electromagnetic environment and the degree of hardware aging, and dynamically adjusting the correlation strength judgment threshold accordingly. Specifically, when the headphones are in a complex electromagnetic environment with high noise or high pulse frequency, internal components are more susceptible to interference and exhibit abnormalities. In this case, lowering the correlation strength judgment threshold allows even relatively weak correlations between external pulses and internal anomalies to be effectively identified, avoiding missed detections due to environmental complexity. Simultaneously, considering that hardware aging reduces its anti-interference capability, the threshold is lowered again when the cumulative load of the internal power supply unit or main control chip reaches a certain level. This ensures that even with degraded hardware performance, the system can still sensitively capture the correlation between external interference and internal anomalies, preventing misjudgments or delayed judgments due to hardware aging. Therefore, through adaptive adjustment of the threshold, this application can more accurately reflect the actual impact of external instantaneous electromagnetic pulses on internal processing anomalies, thereby improving the robustness and accuracy of complex delayed anomaly pattern recognition.

[0043] Through the above technical solution, this application can significantly improve the accuracy and adaptability of abnormal pattern recognition in wireless headset voice data transmission. Specifically, by dynamically acquiring and utilizing information on the complexity of the electromagnetic environment and the degree of hardware aging, the system can adaptively adjust the correlation strength judgment threshold according to actual operating conditions. This allows the system to maintain sensitivity and accuracy in identifying the correlation between external instantaneous electromagnetic pulses and internal processing anomalies even when faced with adverse factors such as high-intensity background noise, high-frequency transient electromagnetic pulses, or internal hardware aging, effectively avoiding false negatives or missed positives that may occur with fixed threshold methods. Therefore, the solution of this application can more reliably identify complex delay anomaly patterns, providing a more solid technical guarantee for the stable operation and fault diagnosis of wireless headsets.

[0044] In some embodiments of this application, the step of obtaining information on the complexity of the electromagnetic environment in which the headphones are currently located, including the intensity level of background electromagnetic noise and the frequency of external instantaneous electromagnetic pulses, includes: Multiple radio frequency signal receiving units are set inside the earphone, and the radio frequency signal receiving units are distributed at different positions on the earphone shell at preset spatial intervals. Each radio frequency signal receiving unit independently samples the surrounding electromagnetic signals in real time and generates local electromagnetic field data with location and time stamps; Spatial consistency analysis and time series analysis are performed on local electromagnetic field data to identify the spatial locality and mobility characteristics of electromagnetic interference sources; Based on the results of spatial consistency analysis and time series analysis, the sampling frequency and sampling point distribution of electromagnetic signals are dynamically adjusted to obtain the intensity level of background electromagnetic noise and the occurrence frequency of external instantaneous electromagnetic pulses.

[0045] Specifically, placing multiple radio frequency (RF) signal receiving units inside the headphones refers to strategically deploying multiple sensors capable of sensing and receiving RF signals within the outer shell or internal structure of the wireless headphones. These RF signal receiving units are distributed at predetermined spatial intervals, for example, they can be configured in different locations such as the left and right earcups, headband, or earbuds of the headphones, to ensure comprehensive coverage of the electromagnetic environment in different directions and areas around the headphones. The purpose is to overcome the blind spots or insufficient local information that may exist in a single sensor in a complex electromagnetic environment through multi-point collaborative sensing, thereby obtaining more comprehensive electromagnetic field data. Each RF signal receiving unit independently samples the surrounding electromagnetic signals in real time and generates local electromagnetic field data with location and time stamps. Real-time sampling means that data acquisition is continuous and instantaneous, capable of capturing instantaneous changes in the electromagnetic environment. Local electromagnetic field data with location stamps means that the data collected by each receiving unit includes its precise physical location information inside the headphones, which is crucial for subsequent spatial analysis. Data with time stamps records the exact time of sampling, which is helpful for time series analysis and event synchronization. Its purpose is to provide high-precision, multi-dimensional raw data with spatiotemporal context for subsequent refined analysis.

[0046] In practical applications, spatial consistency analysis and time series analysis are performed on local electromagnetic field data to identify the spatial locality and mobility characteristics of electromagnetic interference sources. Spatial consistency analysis involves comparing local electromagnetic field data collected by different radio frequency signal receiving units at the same or similar times to assess the spatial distribution pattern and correlation of the electromagnetic field, thereby determining whether the electromagnetic interference source is local, point-source, or diffuse and wide-area. Time series analysis involves analyzing data collected by one or more radio frequency signal receiving units over a period of time to identify the dynamic changes of the electromagnetic interference source, such as its frequency of occurrence, duration, intensity fluctuations, and whether it has a movement trajectory. Its purpose is to extract the key characteristics of the electromagnetic interference source from the raw data, providing a basis for accurately assessing the complexity of the electromagnetic environment. Furthermore, based on the results of spatial consistency analysis and time series analysis, the sampling frequency and sampling point distribution of the electromagnetic signal are dynamically adjusted. This means that the system does not employ a fixed sampling strategy, but rather intelligently optimizes the data acquisition method based on the real-time identification results of the electromagnetic interference source characteristics (such as locality and mobility). For example, when a high-intensity, rapidly changing local interference source is identified, the sampling frequency of the receiving unit in the relevant area can be increased, and the sampling points may be adjusted (e.g., through virtual sampling points or focused sampling) to capture the interference source more precisely. Conversely, in areas with a relatively stable electromagnetic environment, the sampling frequency can be appropriately reduced to save power. The aim is to optimize resource utilization and more effectively obtain the intensity level of background electromagnetic noise and the frequency of external transient electromagnetic pulses while ensuring data accuracy.

[0047] This application's solution utilizes multiple radio frequency signal receiving units embedded within the earphone, distributed at preset spatial intervals, to acquire real-time local electromagnetic field data of the surrounding electromagnetic environment from multiple spatial dimensions. This data, marked with location and time stamps, lays the foundation for subsequent refined analysis. Spatial consistency analysis of this local electromagnetic field data effectively identifies the spatial locality characteristics of electromagnetic interference sources, such as determining whether the interference originates from a point source in a specific direction or is uniformly distributed background noise. Simultaneously, time-series analysis captures the dynamic changes and mobility characteristics of electromagnetic interference sources, such as the frequency and duration of instantaneous electromagnetic pulses. Based on these analysis results, the system can dynamically adjust the sampling frequency and sampling point distribution of the electromagnetic signals, allowing the data acquisition strategy to adapt to the complexity of the current electromagnetic environment. For example, when detecting high-frequency instantaneous electromagnetic pulses or rapidly moving interference sources, the system can increase the sampling frequency to more accurately capture these events; when the environment is relatively stable, the sampling frequency can be reduced to conserve energy. Therefore, this solution can overcome the limitations of traditional single sensor or fixed sampling strategies, providing more comprehensive, accurate, and spatiotemporally resolved information on the complexity of the electromagnetic environment. This provides a reliable input for subsequent calculation of the correlation strength judgment threshold, thereby improving the accuracy and robustness of composite delayed anomaly pattern recognition.

[0048] Through the above technical solution, this application can achieve refined and dynamic acquisition of information on the complexity of the electromagnetic environment in which the wireless headset is located. Compared with methods that rely solely on a single sensor or a fixed sampling strategy, this solution utilizes multiple distributed radio frequency signal receiving units, combined with spatial consistency analysis and time series analysis, to more accurately identify the spatial locality and mobility characteristics of electromagnetic interference sources. This enables the system to more accurately assess the intensity level of background electromagnetic noise and the frequency of external instantaneous electromagnetic pulses, thereby providing more reliable and detailed data support for subsequent correlation strength judgment threshold calculation. Therefore, this solution significantly improves the accuracy of establishing the correlation between external instantaneous electromagnetic pulses and internal processing anomalies, effectively avoiding misjudgments or omissions caused by insufficient or inaccurate environmental information, and thus improving the overall performance and robustness of complex delayed anomaly pattern recognition.

[0049] In some embodiments of this application, the steps of performing spatial consistency analysis and time series analysis on local electromagnetic field data to identify the spatial locality and mobility characteristics of electromagnetic interference sources include: Acquire head movement information of the headphone wearer; Obtain headphone attitude information; Based on head movement information and earphone posture information, the position and orientation of the local electromagnetic field data collected by each radio frequency signal receiving unit are calibrated to obtain calibrated local electromagnetic field data. Calculate the degree of difference between the calibrated local electromagnetic field data; Adjust the judgment threshold for spatial consistency analysis based on the degree of difference and preset calibration parameters; Based on the adjusted judgment threshold, spatial consistency analysis is performed on the calibrated local electromagnetic field data to identify the spatial locality characteristics of electromagnetic interference sources. Time series analysis of local electromagnetic field data is performed to identify the mobility characteristics of electromagnetic interference sources.

[0050] Specifically, acquiring head movement information of the headphone wearer can be understood as monitoring the dynamic changes in the wearer's head, such as rotation, tilt, and displacement, in real time through the inertial measurement unit built into the headphones or sensor data from smart devices connected to the headphones. The purpose is to provide necessary motion references for subsequent electromagnetic field data calibration.

[0051] Acquiring headphone attitude information refers to determining the precise orientation and tilt angle of the headphones in three-dimensional space using fused data from internal headphone attitude sensors (such as gyroscopes, accelerometers, and magnetometers). The purpose is to ensure that the calibration of the data collected by the RF signal receiving unit accurately reflects the actual spatial state of the headphones.

[0052] In practical applications, based on head movement and earphone posture information, the position and orientation of the local electromagnetic field data collected by each RF signal receiving unit are calibrated to obtain calibrated local electromagnetic field data. This involves using head movement and earphone posture data, and through coordinate transformation algorithms, to unify local electromagnetic field data collected at different times or locations into the same reference coordinate system, thereby eliminating measurement errors caused by earphone movement or posture changes. The purpose is to provide more accurate and consistent electromagnetic field data, laying the foundation for subsequent analysis.

[0053] Furthermore, the degree of difference between the calibrated local electromagnetic field data can be calculated using various statistical methods, such as calculating the root mean square error, correlation coefficient, or Euclidean distance between calibrated data collected by different radio frequency signal receiving units at the same or different times. The aim is to quantify the consistency level of the calibrated data and provide a basis for dynamically adjusting the analysis threshold.

[0054] Based on this, the judgment threshold for spatial consistency analysis is adjusted according to the degree of difference and preset calibration parameters. This means that when the difference in the calibrated data is large, it may indicate a complex electromagnetic environment or highly dynamic interference sources; in this case, the judgment threshold for spatial consistency analysis can be appropriately relaxed. Conversely, the threshold can be tightened. Preset calibration parameters can include parameters obtained from historical data statistics, expert experience, or machine learning model training. The aim is to enable spatial consistency analysis to better adapt to dynamically changing electromagnetic environments and headphone usage scenarios.

[0055] Therefore, based on the adjusted judgment threshold, spatial consistency analysis is performed on the calibrated local electromagnetic field data to identify the spatial locality characteristics of electromagnetic interference sources. This means that after considering the dynamic wearing factors of headphones and calibrating the data, the adjusted and more adaptive threshold is used to identify the spatial distribution pattern of electromagnetic field data, thereby more accurately determining whether the electromagnetic interference source is concentrated in a certain area, distributed in multiple points, or diffusely distributed.

[0056] Meanwhile, time series analysis of local electromagnetic field data to identify the mobility characteristics of electromagnetic interference sources refers to further analyzing the variation law of electromagnetic field data over time based on spatial consistency analysis, so as to determine whether the electromagnetic interference source is stationary, periodically moving, or randomly moving.

[0057] This application effectively solves the problem of inaccurate electromagnetic field data caused by headphone movement and posture changes in dynamic wearing environments by introducing the acquisition of headphone wearer's head movement information and headphone posture information, and calibrating the position and orientation of local electromagnetic field data collected by each RF signal receiving unit based on this information. Specifically, head movement information and headphone posture information provide the real-time dynamic state of the headphone in space, enabling the local electromagnetic field data to be accurately mapped to a unified reference coordinate system, thereby eliminating measurement errors introduced by changes in the relative position of sensors. In addition, by calculating the degree of difference between the calibrated local electromagnetic field data and dynamically adjusting the judgment threshold of spatial consistency analysis according to the degree of difference, the analysis process can adaptively cope with electromagnetic environments and data fluctuations of varying complexity, avoiding misjudgments or omissions that may be caused by fixed thresholds. It is precisely because of this dynamic calibration and adaptive threshold adjustment mechanism that subsequent spatial consistency analysis can more accurately identify the spatial locality characteristics of electromagnetic interference sources, while time series analysis can more reliably identify their mobility characteristics.

[0058] Through the above technical solution, this application can significantly improve the accuracy and robustness of identifying the spatial locality and mobility characteristics of electromagnetic interference sources. In some solutions, due to insufficient consideration of the dynamic wearing factors of headphones in actual use, the collected electromagnetic field data may have spatial and directional deviations, thus affecting the reliability of the analysis results. This application effectively eliminates these dynamic deviations by introducing head motion information and headphone posture information for data calibration, ensuring the spatial consistency of the data. At the same time, the analysis threshold is dynamically adjusted based on the degree of difference in the calibrated data, enabling the system to better adapt to complex and changing electromagnetic environments and avoiding recognition errors caused by fixed thresholds. Therefore, this application can more accurately characterize the real spatial distribution and movement patterns of electromagnetic interference sources, providing more reliable information on the complexity of the electromagnetic environment for subsequent abnormal pattern recognition, thereby improving the overall performance and user experience of abnormal pattern recognition for wireless headphone voice data transmission.

[0059] In some embodiments of this application, the step of performing spatial consistency analysis on the calibrated local electromagnetic field data based on the adjusted judgment threshold to identify the spatial locality characteristics of electromagnetic interference sources includes: Multi-scale spatial feature extraction is performed on the calibrated local electromagnetic field data to obtain local features at different spatial scales; Topological structure analysis is performed on the acquired multi-scale spatial features to construct a topological map of the locality features of the electromagnetic field; Based on the topological map, identify the local characteristics of electromagnetic interference sources, such as whether they present a single peak, a multi-peak distribution, or an irregular shape. The identified local features are compared with the adjusted judgment threshold to determine the spatial local features of the electromagnetic interference source.

[0060] Specifically, multi-scale spatial feature extraction of calibrated local electromagnetic field data refers to extracting the electromagnetic field intensity, gradient, or wave patterns exhibited at different spatial resolutions from the original local electromagnetic field data by employing different spatial filters or analysis window sizes. The aim is to comprehensively capture the influence range and intensity distribution of electromagnetic interference sources at different spatial scales, avoiding the omission of crucial information due to single-scale analysis. For example, wavelet transform, Gaussian filtering, or morphological operations can be used to process the electromagnetic field data at multiple scales to obtain local features ranging from fine to coarse.

[0061] Topological analysis of the acquired multi-scale spatial features can be understood as using mathematical methods, such as graph theory or network analysis, to model and analyze the connections, proximity, and overall structure between these local features at different scales. The aim is to construct a topological graph that can intuitively represent the spatial distribution and interactions of local electromagnetic field features, thereby revealing the intrinsic spatial structure of electromagnetic interference sources. In practical applications, this topological graph can be represented as a set of nodes (representing local features) and edges (representing relationships between features). By analyzing topological indices such as node degree, centrality, or clustering coefficients, a deeper understanding of the local structure of the electromagnetic field can be achieved.

[0062] Furthermore, based on the topological map, identifying the local characteristics of electromagnetic interference sources—whether they exhibit a single peak, multi-peak distribution, or irregular shape—involves analyzing the distribution density, connection patterns, and overall morphology of feature nodes in the topological map to determine the spatial concentration and distribution pattern of the electromagnetic interference sources. For example, a single peak may indicate a strong, concentrated interference source; a multi-peak distribution may indicate the presence of multiple discrete but related interference sources; and an irregular shape may indicate a diffuse and complex electromagnetic environment. The aim is to classify the physical characteristics and spatial behavior of electromagnetic interference sources, providing a more accurate basis for subsequent interference suppression or anomaly pattern recognition.

[0063] Finally, the identified local features are compared with the adjusted judgment threshold. This involves comparing the local features of the electromagnetic interference source obtained through topological analysis (e.g., quantitative indicators such as peak intensity, distribution range, and shape complexity) with the judgment threshold that has been pre-adjusted based on information such as background electromagnetic noise intensity and headphone orientation. The aim is to accurately determine the spatial local features of the electromagnetic interference source while considering the current environment and headphone status, thereby improving the accuracy and robustness of the identification.

[0064] This application introduces multi-scale spatial feature extraction to comprehensively capture the influence of electromagnetic interference sources at different spatial resolutions, avoiding information omissions that may occur with single-scale analysis. Based on this, a topological map of the local characteristics of the electromagnetic field is constructed through topological structure analysis, clearly revealing the intrinsic spatial structure of the electromagnetic interference source. This allows for more accurate identification of whether it exhibits a single peak, multi-peak distribution, or irregular shape. This refined identification process, combined with comparison with adjusted judgment thresholds, makes the determination of the spatial local characteristics of the electromagnetic interference source more precise and reliable. It is precisely this multi-level, structured analysis method that enables the effective differentiation and localization of different types of electromagnetic interference sources in complex electromagnetic environments, providing a solid foundation for subsequent anomaly pattern recognition.

[0065] Through the above technical solutions, this application can achieve more refined and comprehensive identification of the spatial locality features of electromagnetic interference sources. Compared with only general spatial consistency analysis, this solution avoids information omission and improves the completeness of feature capture through multi-scale feature extraction; through topological structure analysis, it reveals the intrinsic structure of the locality features of the electromagnetic field, making the judgment of the distribution pattern of interference sources more accurate; and by identifying single peaks, multi-peak distributions, or irregular shapes, it provides richer information on interference source classification. The introduction of these additional technical features significantly improves the accuracy and robustness of spatial locality feature identification of electromagnetic interference sources, thus providing a more reliable and detailed input for identifying abnormal patterns in wireless headset voice data transmission, and helping to more accurately diagnose and solve abnormal voice transmission problems.

[0066] In some embodiments of this application, the step of performing topological structure analysis on the acquired multi-scale spatial features to construct a topological map of electromagnetic field locality features includes: Continuously monitor the local electromagnetic field data collected by each radio frequency signal receiving unit, and update the multi-scale spatial characteristics in real time based on the local electromagnetic field data; When a significant change in multi-scale spatial features after real-time updates is detected, local or global reconstruction of the topology graph is triggered. During local reconstruction, the topology of only the local feature regions that have changed is updated, while the topology of the unchanged regions remains unchanged, resulting in a topology graph after local reconstruction. During global reconstruction, the topological structure of all multi-scale spatial features is reconstructed to obtain the globally reconstructed topological graph. Based on the topology graph after local or global reconstruction, output a topology graph that is updated in real time with the local characteristics of the electromagnetic field.

[0067] Specifically, continuous monitoring of the local electromagnetic field data collected by each radio frequency signal receiving unit refers to the system continuously acquiring electromagnetic field data from multiple radio frequency signal receiving units distributed on the earphone shell. This data is used to update multi-scale spatial characteristics in real time, ensuring an accurate reflection of the current electromagnetic environment. Real-time updating of multi-scale spatial characteristics can be understood as recalculating or adjusting the local electromagnetic field characteristics at different spatial scales based on the latest local electromagnetic field data, for example, by extracting features through methods such as sliding window averaging, wavelet transform, or Fourier analysis.

[0068] Specifically, when a significant change is detected in the multi-scale spatial features updated in real time, a local or global reconstruction of the topology graph is triggered. A significant change can refer to the Euclidean distance, cosine similarity, or specific statistics (such as mean or variance) of the feature vectors exceeding a preset threshold. The purpose of triggering local or global reconstruction is to optimize the use of computational resources while maintaining the accuracy of the topology graph.

[0069] In practical applications, during local reconstruction, only the topology of the changed local feature regions is updated, while the topology of the unchanged regions remains unchanged. This means the system can identify the specific regions where electromagnetic field changes occur and only make local adjustments to the topology graph for these regions, such as updating node connections or weights. The topology of the unchanged regions is preserved, avoiding unnecessary computation. Thus, the locally reconstructed topology graph can be obtained.

[0070] During global reconstruction, the topology of all multi-scale spatial features is reconstructed. Global reconstruction is typically triggered when there are large-scale, global changes in the electromagnetic environment, such as switching from one environment (e.g., indoors) to a completely different environment (e.g., a noisy outdoor area). In such cases, local reconstruction may not be sufficient to accurately reflect the overall change. By reconstructing all topologies, a globally reconstructed topology map can be obtained.

[0071] Finally, based on the locally reconstructed or globally reconstructed topology map, a real-time updated topology map of the local electromagnetic field characteristics is output. This output topology map reflects the local characteristics of the current electromagnetic environment, providing accurate and real-time basis for subsequent electromagnetic interference source identification and abnormal mode judgment.

[0072] This application effectively solves the problems of high computational resource consumption and poor real-time performance caused by frequent global topology reconstruction in dynamic electromagnetic environments by introducing a mechanism for continuous monitoring and real-time updating of multi-scale spatial features, and selectively triggering local or global topology reconstruction based on the significance of feature changes. Specifically, continuous monitoring ensures that the system can promptly perceive changes in the electromagnetic environment; real-time updating of multi-scale spatial features ensures accurate capture of current environmental characteristics. When a significant change is detected, a local or global reconstruction strategy is intelligently selected based on the scope and degree of the change. Local reconstruction, by updating only the topology of the affected region, greatly reduces the computational load and improves the response speed; while global reconstruction ensures that the topology map can comprehensively and accurately reflect the new electromagnetic field distribution when significant environmental changes occur. This adaptive reconstruction strategy enables the system to significantly reduce the computational load while maintaining the accuracy of the topology map, thereby improving the real-time performance and efficiency of abnormal pattern recognition in wireless headset voice data transmission.

[0073] Through the above technical solution, this application can significantly improve the efficiency and real-time performance of abnormal pattern recognition for voice data transmission in wireless headphones in complex dynamic electromagnetic environments. Compared to performing a comprehensive topology reconstruction for every environmental change, the adaptive local or global reconstruction mechanism proposed in this application can intelligently select the most suitable update strategy based on the actual changes in the local characteristics of the electromagnetic field. This not only significantly reduces unnecessary computational resource consumption and lowers the processing load on the main control chip, but also ensures the real-time accuracy of the electromagnetic field local characteristic topology map. As a result, the system can more quickly and accurately identify composite delay abnormal patterns caused by external instantaneous electromagnetic pulses and dynamic background electromagnetic noise, thereby improving the stability of voice data transmission and user experience of wireless headphones.

[0074] In some embodiments of this application, the step of triggering local or global reconstruction of the topology map when a significant change in the real-time updated multi-scale spatial features is detected includes: Continuously acquire the real-time updated multi-scale spatial features and calculate the difference between the real-time updated multi-scale spatial features and the multi-scale spatial features of the previous time step or a preset benchmark. Based on the difference amount, determine whether the difference amount falls into one of the preset multiple reconstruction strategy intervals. The multiple reconstruction strategy intervals include a local reconstruction interval and a global reconstruction interval. The local reconstruction interval corresponds to the local reconstruction strategy, and the global reconstruction interval corresponds to the global reconstruction strategy. When the difference falls into the local reconstruction interval, the local reconstruction of the topology graph is triggered. When the difference falls within the global reconstruction range, a global reconstruction of the topology graph is triggered. When the difference falls into the overlapping area set between the local reconstruction interval and the global reconstruction interval, it is maintained according to the previous reconstruction strategy, or a predictive selection is made according to the trend of change, so as to ensure the stability of the reconstruction decision.

[0075] Specifically, continuously acquiring real-time updated multi-scale spatial features refers to the system continuously extracting and updating local features at different spatial scales from the local electromagnetic field data collected by each radio frequency signal receiving unit. Based on this, the difference between the real-time updated multi-scale spatial features and the multi-scale spatial features at the previous moment or a preset benchmark is calculated. This difference can be understood as the degree of deviation between the current electromagnetic field topology and a previous state. For example, it can be obtained by calculating the Euclidean distance, cosine similarity, or weighted average rate of change of the feature vectors, with the aim of quantifying the magnitude of changes in the electromagnetic environment.

[0076] The system determines whether the difference falls within one of several preset reconstruction strategy intervals based on the magnitude of the difference. These intervals are defined according to empirical data or preset rules and are used to guide the system in selecting an appropriate reconstruction strategy. Specifically, the local reconstruction interval corresponds to a local reconstruction strategy. When the difference is small, indicating that the change in the electromagnetic environment is limited to a local area, the system triggers a local reconstruction of the topology map, updating only the topology of the changed local feature regions. The global reconstruction interval corresponds to a global reconstruction strategy. When the difference is large, indicating that the electromagnetic environment has undergone a global or drastic change, the system triggers a global reconstruction of the topology map, rebuilding the topology of all multi-scale spatial features.

[0077] In practical applications, to ensure the stability of reconstruction decisions, the system adopts a more cautious strategy when the difference falls within the overlapping area between the local and global reconstruction intervals. For example, it can maintain the previous reconstruction strategy: if the previous strategy was local, it continues with local reconstruction; if it was global, it continues with global reconstruction. Alternatively, the system can make predictive choices based on trends: if the difference shows a continuously increasing trend, it tends to choose global reconstruction; if the difference shows fluctuations but does not significantly exceed a certain threshold, it tends to maintain local reconstruction. The aim is to avoid frequently switching reconstruction strategies in critical states, thereby improving the system's robustness and stability.

[0078] This application effectively addresses the ambiguity and decision instability issues that may exist when judging changes in electromagnetic field characteristics by introducing a decision mechanism based on difference quantities, reconstruction strategy intervals, and overlapping regions. By continuously calculating the difference quantities of multi-scale spatial characteristics, the system can accurately quantify the degree of change in the electromagnetic environment. Based on preset reconstruction strategy intervals, the system can intelligently select local or global reconstruction according to the magnitude of the difference quantities, thereby avoiding unnecessary global reconstruction and improving resource utilization efficiency. Furthermore, by setting an overlapping region between the local and global reconstruction intervals and employing a mechanism within this region to maintain the previous strategy or make predictive choices, the scheme of this application can effectively smooth reconstruction decisions, prevent frequent strategy switching caused by critical value fluctuations, and thus ensure the stability and continuity of the topology reconstruction process, making the identification of the spatial locality and mobility characteristics of electromagnetic interference sources more accurate and reliable.

[0079] Through the above technical solution, this application enables refined management and stability improvement of electromagnetic field topology map reconstruction decisions. Specifically, by quantifying the differences in multi-scale spatial features and combining them with reconstruction strategy ranges, the system can intelligently select the most suitable reconstruction method based on the actual degree of change in the electromagnetic environment, avoiding excessive or insufficient reconstruction operations, thereby significantly improving the system's adaptability and processing efficiency to dynamic electromagnetic environments. Especially when the differences are at a critical state, by introducing a stable decision-making mechanism within the overlapping region, frequent oscillations in the reconstruction strategy are effectively avoided, enhancing the system's robustness, ensuring the accuracy and continuity of electromagnetic interference source identification, and thus improving the overall performance of abnormal pattern recognition in wireless headset voice data transmission.

[0080] Based on the same inventive concept, this application proposes a wireless earphone voice data transmission abnormality pattern recognition system, such as... Figure 2 As shown, the system includes: The instantaneous electromagnetic pulse information acquisition module 1 is used to acquire instantaneous electromagnetic pulse information of the external electromagnetic environment. The instantaneous electromagnetic pulse information includes the occurrence time, duration and power distribution of the instantaneous electromagnetic pulse. The internal power supply unit operation status information acquisition module 2 is used to acquire the operation status information of the internal power supply unit, including the output voltage fluctuation and instantaneous current change of the power supply unit. The main control chip voice processing status information acquisition module 3 is used to acquire the voice processing status information of the main control chip. The voice processing status information includes the internal clock frequency deviation of the main control chip, the cache access delay, and the voice decoder processing pause time. The abnormal pattern recognition module 4 is used to initiate an observation cycle in response to the acquisition of instantaneous electromagnetic pulse information. Within the observation cycle, based on the instantaneous electromagnetic pulse information, operating status information, and voice processing status information, it establishes the correlation between the external instantaneous electromagnetic pulse and the internal processing abnormality in order to identify composite delayed abnormal patterns.

[0081] Clearly, this system enables the effective identification of abnormal patterns in voice data transmission from wireless earphones, solving the problem of existing technologies being unable to identify complex delay abnormal patterns.

[0082] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for recognizing abnormal patterns in voice data transmission from wireless earphones, characterized in that, include: Acquire instantaneous electromagnetic pulse information of the external electromagnetic environment, including the occurrence time, duration, and power distribution of the instantaneous electromagnetic pulse; Obtain the operating status information of the internal power supply unit, including the output voltage fluctuation and instantaneous current change of the power supply unit; Obtain the voice processing status information of the main control chip, which includes the internal clock frequency deviation of the main control chip, cache access latency, and voice decoder processing pause time. In response to the acquisition of transient electromagnetic pulse information, an observation cycle is initiated. Within the observation cycle, based on the transient electromagnetic pulse information, operating status information, and voice processing status information, a correlation is established between the external transient electromagnetic pulse and the internal processing anomaly in order to identify composite delayed anomaly patterns.

2. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 1, characterized in that, The steps of responding to the acquisition of transient electromagnetic pulse information, initiating an observation period, and establishing a correlation between external transient electromagnetic pulses and internal processing anomalies based on transient electromagnetic pulse information, operating status information, and voice processing status information within the observation period to identify composite delayed anomaly patterns include: Upon receiving transient electromagnetic pulse information, the background electromagnetic noise intensity information of the current environment is obtained; Based on the background electromagnetic noise intensity information, the identification threshold for internal anomalies is adjusted, and the duration of the observation period is also adjusted. Within the adjusted observation period, based on instantaneous electromagnetic pulse information, the adjusted internal anomaly identification threshold, the operating status information of the internal power supply unit, and the voice processing status information of the main control chip, a correlation is established between external instantaneous electromagnetic pulses and internal processing anomalies to identify composite delayed anomaly patterns.

3. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 2, characterized in that, The step of establishing a correlation between external instantaneous electromagnetic pulses and internal processing anomalies within the adjusted observation period, based on instantaneous electromagnetic pulse information, the adjusted internal anomaly identification threshold, the operating status information of the internal power supply unit, and the voice processing status information of the main control chip, to identify composite delayed anomaly patterns, includes: Obtain the operating status information of the internal power supply unit and the voice processing status information of the main control chip; Spectrum analysis is performed on the operating status information of the internal power supply unit to extract the inherent fluctuation characteristics of the internal power supply unit; Timing pattern analysis is performed on the voice processing status information of the main control chip to extract the internal processing anomaly features of the main control chip; Obtain the spectral characteristics of dynamic background electromagnetic noise; The inherent fluctuation characteristics of the internal power supply unit, the abnormal internal processing characteristics of the main control chip, and the spectral characteristics of dynamic background electromagnetic noise are compared and contrasted. Based on the results of the differential comparison, the influence of external instantaneous electromagnetic pulses and dynamic background electromagnetic noise on the operating status information of the internal power supply unit and the voice processing status information of the main control chip is quantified. Based on the quantification results, a correlation is established between external instantaneous electromagnetic pulses and internal processing anomalies to identify complex delayed anomaly patterns.

4. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 3, characterized in that, The step of establishing the correlation between external instantaneous electromagnetic pulses and internal processing anomalies based on quantification results to identify composite delayed anomaly patterns includes: Obtain information on the complexity of the electromagnetic environment in which the headphones are currently located. This information includes the intensity level of background electromagnetic noise and the frequency of external instantaneous electromagnetic pulses. Obtain information on the aging level of the internal hardware of the headphones, including the cumulative working time of the internal power supply unit and the cumulative processing load of the main control chip; Based on information about the complexity of the electromagnetic environment and the aging of the hardware, a correlation strength judgment threshold is calculated. The calculation process includes: when the background electromagnetic noise intensity level is high, the correlation strength judgment threshold is reduced; when the frequency of external instantaneous electromagnetic pulses is high, the correlation strength judgment threshold is further reduced; when the cumulative working time of the internal power supply unit or the cumulative processing load of the main control chip exceeds the preset value, the correlation strength judgment threshold is reduced again. Compare the quantification results with the correlation strength judgment threshold; If the quantification result exceeds the correlation strength judgment threshold, a correlation is established between the external instantaneous electromagnetic pulse and the internal processing anomaly in order to identify the composite delayed anomaly pattern.

5. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 4, characterized in that, The step of obtaining information on the complexity of the electromagnetic environment in which the headphones are currently located, including the intensity level of background electromagnetic noise and the frequency of external instantaneous electromagnetic pulses, includes: Multiple radio frequency signal receiving units are set inside the earphone, and the radio frequency signal receiving units are distributed at different positions on the earphone shell at preset spatial intervals. Each radio frequency signal receiving unit independently samples the surrounding electromagnetic signals in real time and generates local electromagnetic field data with location and time stamps; Spatial consistency analysis and time series analysis are performed on local electromagnetic field data to identify the spatial locality and mobility characteristics of electromagnetic interference sources; Based on the results of spatial consistency analysis and time series analysis, the sampling frequency and sampling point distribution of electromagnetic signals are dynamically adjusted to obtain the intensity level of background electromagnetic noise and the occurrence frequency of external instantaneous electromagnetic pulses.

6. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 5, characterized in that, The steps of performing spatial consistency analysis and time series analysis on local electromagnetic field data to identify the spatial locality and mobility characteristics of electromagnetic interference sources include: Acquire head movement information of the headphone wearer; Obtain headphone attitude information; Based on head movement information and earphone posture information, the position and orientation of the local electromagnetic field data collected by each radio frequency signal receiving unit are calibrated to obtain calibrated local electromagnetic field data. Calculate the degree of difference between the calibrated local electromagnetic field data; Adjust the judgment threshold for spatial consistency analysis based on the degree of difference and preset calibration parameters; Based on the adjusted judgment threshold, spatial consistency analysis is performed on the calibrated local electromagnetic field data to identify the spatial locality characteristics of electromagnetic interference sources. Time series analysis of local electromagnetic field data is performed to identify the mobility characteristics of electromagnetic interference sources.

7. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 6, characterized in that, The step of performing spatial consistency analysis on the calibrated local electromagnetic field data based on the adjusted judgment threshold to identify the spatial locality characteristics of electromagnetic interference sources includes: Multi-scale spatial feature extraction is performed on the calibrated local electromagnetic field data to obtain local features at different spatial scales; Topological structure analysis is performed on the acquired multi-scale spatial features to construct a topological map of the locality features of the electromagnetic field; Based on the topological map, identify the local characteristics of electromagnetic interference sources, such as whether they present a single peak, a multi-peak distribution, or an irregular shape. The identified local features are compared with the adjusted judgment threshold to determine the spatial local features of the electromagnetic interference source.

8. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 7, characterized in that, The step of performing topological structure analysis on the acquired multi-scale spatial features to construct a topological map of electromagnetic field locality features includes: Continuously monitor the local electromagnetic field data collected by each radio frequency signal receiving unit, and update the multi-scale spatial characteristics in real time based on the local electromagnetic field data; When a significant change in multi-scale spatial features after real-time updates is detected, local or global reconstruction of the topology graph is triggered. During local reconstruction, the topology of only the local feature regions that have changed is updated, while the topology of the unchanged regions remains unchanged, resulting in a topology graph after local reconstruction. During global reconstruction, the topological structure of all multi-scale spatial features is reconstructed to obtain the globally reconstructed topological graph. Based on the topology graph after local or global reconstruction, output a topology graph that is updated in real time with the local characteristics of the electromagnetic field.

9. A method for identifying abnormal patterns in voice data transmission of a wireless headset according to claim 8, characterized in that, The step of triggering local or global reconstruction of the topology map when a significant change in the real-time updated multi-scale spatial features is detected includes: Continuously acquire the real-time updated multi-scale spatial features and calculate the difference between the real-time updated multi-scale spatial features and the multi-scale spatial features of the previous time step or a preset benchmark. Based on the difference amount, determine whether the difference amount falls into one of the preset multiple reconstruction strategy intervals. The multiple reconstruction strategy intervals include a local reconstruction interval and a global reconstruction interval. The local reconstruction interval corresponds to the local reconstruction strategy, and the global reconstruction interval corresponds to the global reconstruction strategy. When the difference falls into the local reconstruction interval, the local reconstruction of the topology graph is triggered. When the difference falls within the global reconstruction range, a global reconstruction of the topology graph is triggered. When the difference falls into the overlapping area set between the local reconstruction interval and the global reconstruction interval, it is maintained according to the previous reconstruction strategy, or a predictive selection is made according to the trend of change, so as to ensure the stability of the reconstruction decision.

10. A system for recognizing abnormal patterns in voice data transmission of wireless earphones, characterized in that, The system includes: The instantaneous electromagnetic pulse information acquisition module is used to acquire instantaneous electromagnetic pulse information of the external electromagnetic environment. The instantaneous electromagnetic pulse information includes the occurrence time, duration, and power distribution of the instantaneous electromagnetic pulse. The internal power supply unit operation status information acquisition module is used to acquire the operation status information of the internal power supply unit, including the output voltage fluctuation and instantaneous current change of the power supply unit. The main control chip voice processing status information acquisition module is used to acquire the voice processing status information of the main control chip. The voice processing status information includes the internal clock frequency deviation of the main control chip, the cache access delay, and the voice decoder processing pause time. The abnormal pattern recognition module is used to initiate an observation cycle in response to the acquisition of instantaneous electromagnetic pulse information. Within the observation cycle, based on the instantaneous electromagnetic pulse information, operating status information, and voice processing status information, it establishes the correlation between the external instantaneous electromagnetic pulse and the internal processing abnormality in order to identify composite delayed abnormal patterns.

Images