System and method of adaptive noise reduction earphone magnetic type charging box

Abstract

The invention relates to the field of audio equipment, and discloses a system and method of a self-adaptive noise reduction earphone magnetic attraction type charging box, and the method comprises the steps: when a noise reduction earphone is used, collecting motion data and environment noise in real time, judging the motion state of a user, calculating the energy entropy of the environment noise, and judging the scene noisy degree of the position where the user is located; the method comprises the following steps: selecting a noise reduction strategy for a noise reduction earphone in combination with a user motion state, analyzing historical use data of the noise reduction earphone, predicting an expected use time period of a user, predicting required residual electric quantity of the noise reduction earphone according to the expected use time period, comparing actual residual electric quantity, judging whether the electric quantity is sufficient or not, and generating an electric quantity risk signal if the electric quantity is insufficient; the method comprises the following steps: pausing real-time acquisition of motion data and environmental noise, establishing a model, selecting a noise reduction strategy of a noise reduction earphone in a prediction time period through the model, re-detecting the environmental noise and the motion state at a detection time point, updating the prediction model and a pre-selected noise reduction strategy, and dynamically adjusting the charging power when the noise reduction earphone is charged.

Description

Technical Field

[0001] The present invention relates to the technical field of audio equipment, and in particular to a system and method for a magnetic charging box for an adaptive noise-cancelling headset. Background Art

[0002] With the rapid development of science and technology and the continuous improvement of people's requirements for quality of life, noise-canceling headphones have become an indispensable electronic product in modern life. Especially in noisy environments, noise-canceling headphones can effectively isolate external noise and provide users with a relatively quiet space, which can greatly enhance the user experience whether it is commuting, studying or resting. However, in the actual use of noise-canceling headphones, users often encounter some problems. These problems not only affect the noise reduction effect, but also may cause inconvenience to users.

[0003] Traditional noise-cancelling headphones usually adopt a fixed noise-cancelling strategy and cannot be dynamically adjusted according to the actual environmental noise conditions of the user. For example, in a quiet library and a noisy street, users may need different noise-cancelling intensities to achieve the best noise-cancelling effect. However, fixed noise-cancelling strategies often cannot meet this demand, resulting in unsatisfactory noise-cancelling effects. At the same time, power management of noise-cancelling headphones is an important issue. Real-time detection of environmental noise and motion status requires continuous activation of sensors and processors, resulting in a surge in device power consumption. In the prior art, when the noise-cancelling headphones are in real-time detection mode, the battery life is shortened to less than 3 hours, which cannot meet the user's usage needs. Users may interrupt use due to low power during use, affecting the user experience. When charging, the charging power cannot be dynamically adjusted according to the actual usage of the headphones and user habits, resulting in low charging efficiency.

[0004] In view of the above problems, the present invention proposes a system and method for a magnetic charging box for adaptive noise reduction headphones. Summary of the invention

[0005] The object of the present invention is to provide a system and method for a magnetic charging box for adaptive noise-cancelling headphones to solve at least one of the above-mentioned problems of the prior art.

[0006] In a first aspect, the present invention provides a method for an adaptive noise reduction headset magnetic charging box, comprising the following steps: When the noise-cancelling headphones are used, the user's motion state is determined based on the collected motion data, the noise energy entropy of the environment is determined based on the collected ambient noise, and the noisiness of the scene at the user's location is judged based on the noise energy entropy; Comprehensively analyze the scene noise and motion status to select a noise reduction strategy; Determine whether the remaining power of the noise-cancelling headphones is sufficient within the predicted time; If it is insufficient, real-time collection is suspended during the forecast period, and a pre-selected noise reduction strategy is selected for the forecast period of the noise reduction headphones through the forecast model; Based on the motion data and environmental noise before the inspection time point, the prediction model is calibrated, and a calibrated pre-selected noise reduction strategy is obtained; The prediction time consists of several prediction periods, and the test time point is the end point of the prediction period.

[0007] In a second aspect, the present invention provides a system for a magnetic charging box of an adaptive noise reduction headset, comprising the following modules: Data collection module: when the noise reduction headphones are in use, the user's motion state is determined based on the collected motion data, and the noise energy entropy of the environment is determined based on the collected ambient noise; Strategy selection module: Based on the noise energy entropy, the noise level of the scene where the user is located is determined, and the noise reduction strategy is selected based on the user's motion status; Power analysis module: determines whether the remaining power of the noise-cancelling headphones is sufficient within the predicted time. If not, generates a power risk signal; Strategy optimization module: If a power risk signal is received, real-time collection is suspended during the prediction period, and a pre-selected noise reduction strategy is established for the prediction period of the noise reduction headphones through the prediction model. The prediction model is calibrated based on the motion data and environmental noise before the test time point, and the calibrated pre-selected noise reduction strategy is obtained; Charging management module: When the noise-canceling headphones are charging, the power level of the magnetic charging box and the noise-canceling headphones is analyzed, and the charging power is dynamically adjusted based on the expected usage period.

[0008] Beneficial effects of the present invention: 1. The present invention provides an accurate data basis for headphones by collecting motion data and environmental noise in real time. It can intelligently judge the noisiness of the scene according to the user's motion status and the energy entropy of the environmental noise, and select the most appropriate noise reduction strategy accordingly, thereby improving the noise reduction effect and ensuring that users can enjoy the best auditory experience in different scenarios, thereby enhancing the adaptability of the headphones and user satisfaction.

[0009] 2. The present invention can accurately predict the user's expected usage period and required remaining power by analyzing the historical usage data of the noise-cancelling headphones, effectively avoiding the situation of insufficient power. Once the power is insufficient, the system immediately generates a risk signal and suspends non-essential detection to save power. At the same time, a model is established to select the noise reduction strategy of the noise-cancelling headphones within the predicted period, and recalibrate at the inspection time point to ensure that the noise reduction effect is always optimal, thereby improving user experience and headphone usage efficiency.

[0010] 3. The present invention dynamically adjusts the charging power and flexibly selects the charging gear according to the power difference and demand, which not only meets the urgent charging needs, but also avoids battery aging, effectively balances the power reserves of the earphones and the charging box, and ensures the overall battery life of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 This is a flowchart of the steps of a method for an adaptive noise reduction headset magnetic charging box provided by an embodiment of the present invention; Figure 2 It is a flow chart of the steps for obtaining the expected usage period in a method for a magnetic charging box for an adaptive noise reduction headset provided in Embodiment 2 of the present invention; Figure 3 It is a flow chart of the steps of dynamically adjusting the charging power in a method for a magnetic charging box for an adaptive noise reduction headset provided in Embodiment 3 of the present invention; Figure 4 It is a flowchart of a system of a magnetic charging box for adaptive noise reduction headphones provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0013] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention. Embodiment 1

[0014] like Figure 1 As shown, a method for an adaptive noise reduction headset magnetic charging box provided by an embodiment of the present invention specifically includes the following steps: Step 1: When the noise-cancelling headphones are in use, the built-in sensors and microphone arrays are used to collect motion data and environmental noise in real time, determine the user's motion state, and calculate the environmental noise energy entropy; In some embodiments, when the noise reduction headphones are in use, data is collected at the current detection time point through motion sensors such as an accelerometer and a gyroscope built into the headphones, and the acceleration changes of the headphones on the x, y, and z axes and the rotation angular velocity of the headphones around the three coordinate axes are detected, and motion features are extracted from the collected acceleration and angular velocity data, and the extracted motion features are analyzed, and the user's current motion state is determined by a preset motion state classification algorithm, and the motion state includes stationary, walking, running, riding, etc.; It should be noted that the detection time points represent the time nodes used to collect motion data, and the intervals between adjacent detection time points are marked as detection mobile terminals, and the durations of adjacent detection time points are the same; In the current detection period, an array composed of several microphones is used to sample the ambient noise at a fixed sampling frequency fs, and the time point of sampling the ambient noise is marked as the sampling moment. The microphones include a feedforward microphone and a feedback microphone. The feedforward microphone is responsible for capturing the ambient noise transmitted into the earphone from the outside world, and the feedback microphone is responsible for monitoring the residual noise signal in the ear canal. The microphones are marked with a serial number i, i=1, 2, ..., n, where n represents the number of microphones; For example, a common sampling frequency fs=48000 Hz means that the sound signal is sampled 48000 times per second; The sound signal value collected by the microphone with serial number i at the kth sampling time in the current detection period is recorded as xi(k), where k=0, 1, 2, ...; The obtained noise signal is amplified and filtered to remove high-frequency noise and interference, and the formula Perform fast Fourier transform on the processed noise signal to obtain the frequency domain signal , where j represents the imaginary unit, , L represents the number of frequency points into which the signal is decomposed by the fast Fourier transform. =0, 1, 2, ..., L-1; According to the obtained frequency domain signal, through the formula Calculate the ratio of the energy of each frequency component to the total energy after Laplace smoothing to obtain the probability distribution of the frequency component ,in is the smoothing coefficient, ; It should be noted that the function of the smoothing coefficient is to avoid There are computational problems when Based on the obtained , through the formula The probability distribution of the frequency components of all microphones is jointly analyzed to calculate the energy entropy H of the ambient noise in the current detection period; It should be noted that the environmental noise energy entropy H represents the complexity of the environmental noise; Step 2: judging the scene noise level at the user's location based on the obtained environmental noise energy entropy, and selecting a noise reduction strategy for the noise reduction headphones based on the user's motion status; In some embodiments, the calculated environmental noise energy entropy H is compared with an environmental noise energy entropy low threshold Hd and an environmental noise energy entropy high threshold Hh; If the environmental noise energy entropy in the current detection period , judging that the scene noise level at the user's current location is quiet; If the environmental noise energy entropy in the current detection period , judging that the scene noise level at the user's current location is low noise; If the environmental noise energy entropy in the current detection period , judging that the scene noise level at the user's current location is highly noisy; Establishing a noise reduction strategy database including scene noisiness, user motion state, and corresponding noise reduction strategies, wherein the noise reduction strategy database stores the best noise reduction strategies corresponding to different combinations of environmental noise energy entropy ranges and user motion states; Based on the environmental noise energy entropy in the current detection period and the determined user motion state, query the noise reduction strategy database, and select a suitable noise reduction strategy for the noise reduction headphones according to the query result; For example, the preset low threshold value of environmental noise energy entropy Hd=3, if the environmental noise energy entropy in the current detection period , the scene noise level of the user's current location is judged to be quiet. If the user is stationary or riding at this time, the deep frequency domain filtering combined with the phase compensation algorithm can be used as the noise reduction strategy. If the user is walking or running at this time, the frequency band noise reduction combined with the IMU dynamic compensation algorithm can be used as the noise reduction strategy; For example, the preset low threshold value of environmental noise energy entropy Hd=3, the preset high threshold value of environmental noise energy entropy Hd=5, if the environmental noise energy entropy in the current detection period , determine that the scene noise level at the user's current location is low noise. If the user is stationary or walking at this time, the CNN neural network can be used to suppress medium and high frequency noise combined with traditional spectral subtraction as a noise reduction strategy. If the user is riding or running at this time, the enhanced human voice frequency band combined with the suppression of sudden noise can be used as a noise reduction strategy;‌‌ For example, the preset high threshold value of environmental noise energy entropy Hd=5, if the environmental noise energy entropy in the current detection period , it is judged that the scene noise level of the user's current location is high. Since the environmental noise energy is too high at this time, no matter what state the user is in, the full-function active noise reduction system combined with the neural network enhanced core algorithm can be used as the noise reduction strategy; The technical solution of an embodiment of the present invention is: when the noise-cancelling headphones are used, motion data and environmental noise are collected in real time through built-in sensors and microphone arrays, the user's motion state is judged and the environmental noise energy entropy is calculated, which provides a data basis for subsequently selecting an appropriate noise reduction strategy according to the user's motion state and environmental noise level, and the scene noisiness of the user's location is judged according to the obtained environmental noise energy entropy, and the noise reduction strategy is selected for the noise-cancelling headphones in combination with the user's motion state to meet the user's noise reduction needs in different scenarios.

[0015] Embodiment 2 like Figure 1 As shown, a method for an adaptive noise reduction headset magnetic charging box provided by an embodiment of the present invention specifically includes the following steps: Step 3: Analyze the historical usage data of the noise-cancelling headphones to predict the expected usage period of the user, predict the required remaining power of the noise-cancelling headphones based on the expected usage period, and compare the remaining power with the actual remaining power to determine whether the power is sufficient. If not, generate a power risk signal; In some embodiments, historical usage data of the noise cancelling headset is obtained, the historical usage data including the usage time period of the noise cancelling headset in each day within the historical usage time and the power consumption of the noise cancelling headset during the usage time period, the historical usage time represents a time period starting from the time node when the noise cancelling headset is first used by the user and ending at the current time node, and the usage time period of the noise cancelling headset represents a time period during the historical usage time when the noise cancelling headset is in use; It should be noted that there are several usage periods within each day of historical usage time; like Figure 2 As shown, the specific steps for obtaining the expected usage period are as follows; Based on any day in the historical usage time, obtain the interval duration between two adjacent usage periods in one day, where the interval duration represents the interval duration between the end point of the previous usage period and the start point of the next usage period, and compare the interval duration with the interval threshold; If the interval between two adjacent usage periods in a day is less than the interval threshold, the two adjacent usage periods are merged into one usage period, and the combined usage period takes the start point of the earlier usage period in the two usage periods as the starting point and the end point of the later usage period in the two usage periods as the end point; It should be noted that adjacent usage periods with a short interval within a day can be regarded as one usage period. The purpose of the above steps is to merge adjacent usage periods with a short interval to reduce subsequent statistical errors. Traverse each day in the historical usage time to obtain several usage time periods that are merged or do not need to be merged, mark the starting point of each usage time period as the start time, and count all the start times in the historical usage time; Take any two of the start times, perform difference processing and take the absolute value to obtain the start time difference between the two start times; For example, if the start time on a certain day of the historical usage time is 9:35, and the start time on another day is 9:37, then the start time difference between the two start times is 2 minutes; Compare the startup time difference between the two startup times with the interval threshold, and if the startup time difference between the two startup times is less than the interval threshold, the two startup times are classified into one startup time group; Traverse all the startup times in the historical usage time until all the startup times whose startup time differences are less than the interval threshold are classified into the startup time group. If there is a startup time that is not classified into any startup time group at this time, then this startup time itself is a startup time group; It should be noted that the purpose of the steps is to combine the startup times with similar time in the historical usage time into a startup time group, so as to facilitate the subsequent statistical analysis of a large number of startup times; Based on any startup time group, the number of startup times in the startup time group is obtained, and the ratio is processed with the total number of all startup times in the historical usage time to obtain the startup frequency value of the startup time group; Compare the obtained startup frequency usage value with the startup frequency usage threshold; If the startup frequency value of the startup time group is less than or equal to the startup frequency threshold, it means that the user uses the noise reduction headset less frequently during the time range included in the startup time group; If the startup frequency value of the startup time group is greater than the startup frequency threshold, it means that the user uses the noise reduction headset more times in the time range included in the startup time group, and the startup time group is marked as a frequent use time group; Based on the obtained frequent use time group, all the start-up times included in the frequent use time group are obtained, and the sum and average are taken to obtain the expected start-up time of the noise reduction headset, and the duration of the use period corresponding to all the start-up times included in the frequent use time group are obtained, and the sum and average are processed to obtain the expected use duration corresponding to the expected start-up time; Combine the expected startup time with the corresponding expected usage time to obtain the expected usage period of the noise-cancelling headphones within a day; It should be noted that the expected usage period of noise-canceling headphones is not necessarily only one, and users may use noise-canceling headphones multiple times a day; Obtain the power consumption of the noise cancelling headphones in the usage period corresponding to all the start-up times in the frequent use time group and sum them up to obtain the sum of power consumption, sum the duration of the usage period corresponding to all the start-up times in the frequent use time group to obtain the sum of usage duration, perform ratio processing on the obtained sum of power consumption and the sum of usage duration to obtain the expected power consumption rate corresponding to the expected usage period; The duration of all usage periods in the historical usage time is obtained and the average is calculated to obtain the average usage duration of the noise cancelling headphones during the historical usage time. The power consumption of the noise cancelling headphones in all usage periods is calculated and the average is calculated to obtain the average power consumption of the noise cancelling headphones during the historical usage time. The average power consumption is calculated and the average usage duration is calculated to obtain the average power consumption rate of the noise cancelling headphones. If the noise reduction headset is in use at the current time node, determine whether the current time node is within any expected usage period; If the current time node is within any expected usage period, obtain the interval between the current time node and the end of the expected usage period where the current time node is located, and multiply it by the expected power consumption rate of the expected usage period to obtain the required remaining power of the noise reduction headset; If the current time node is not within any expected usage period, obtain the duration of the current usage period of the noise cancelling headset at the current time node, perform a difference processing on the average usage duration of the noise cancelling headset and the duration of the current usage period, and perform a product processing on the obtained difference and the average power consumption rate of the noise cancelling headset to obtain the required remaining power of the noise cancelling headset; It should be noted that the current used period refers to the period between the time point when the noise reduction headset is started and used and the current time point; Obtain the actual remaining power of the noise cancelling headset at the current time node, perform subtraction processing on the remaining power required at this time, obtain the remaining power difference of the noise cancelling headset, and compare it with the remaining power difference threshold; If the remaining power difference of the noise cancelling headset is greater than or equal to the remaining power difference threshold, it indicates that the remaining power of the noise cancelling headset is sufficient as predicted based on the user's usage habits; If the remaining power difference of the noise cancelling headphones is less than the remaining power difference threshold, it is predicted based on the user's usage habits that the remaining power of the noise cancelling headphones is insufficient, and a power risk signal is generated; Step 4: If the generated power risk signal is received, the real-time collection of motion data and environmental noise is suspended, a prediction model is established to select a pre-selected noise reduction strategy for the prediction period, and the environmental noise and motion status are re-collected at the test time point, and the prediction model and the pre-selected noise reduction strategy are updated; In some embodiments, if a generated power risk signal is received, real-time detection of environmental noise and motion status is suspended within a predicted time; The prediction time consists of several prediction periods, and the end of the prediction period is marked as the test time point, which is also the starting point of the next prediction period. The duration of each prediction period is the same. Get the motion status of the noise-canceling headphones during each detection period during the current usage period and the ambient noise energy entropy , t=1, 2, 3, ..., m; where t represents the sequence number of the detection period in the current used period sorted by time, and m represents the number of detection periods in the current used period; The motion state is represented in discrete form. When the motion state is static, =0, when the motion state is walking, =1, when the movement state is running, =2, when the motion state is riding, =3; Calculate the probability of the noise-canceling headphones being in various motion states during the current usage period and compare them, and output the motion state with the highest probability as the motion state of the noise-canceling headphones during the prediction period; Construct a prediction model to predict the environmental noise energy entropy in the future c-th detection period , the specific formula is: ,in represents the rate of change, Represents the intercept, combined with the environmental noise energy entropy Fitting by least squares method: , ; According to the prediction model obtained by fitting calculation, the environmental noise energy entropy of the cth detection period within the prediction period is predicted and calculated to obtain the predicted environmental noise energy entropy range of each detection period within the prediction period, and the overlap with the judgment environmental noise energy entropy range of any scene noisiness is analyzed; Specifically, based on the determined environmental noise energy entropy range of any scene noisiness, the length of the overlapped range between the determined environmental noise energy entropy range and the predicted environmental noise energy entropy range is obtained, and the length of the predicted environmental noise energy entropy range is processed by ratio, so as to obtain the probability of the noise reduction headphones being noisy under the scene noisiness within the predicted period; Compare the noisy probabilities of the noise cancelling headphones under different scene noisiness levels, and output the scene noisiness level with the highest noisy probability as the predicted scene noisy level of the noise cancelling headphones within the predicted period; Substituting the obtained predicted motion state and predicted scene noisiness into the noise reduction strategy database for query, obtaining the pre-selected noise reduction strategy within the predicted period, and adjusting the noise reduction strategy of the noise reduction headset to the pre-selected noise reduction strategy; At the test time point of the prediction period, the environmental noise and motion state are collected by sensors, the environmental noise entropy is calculated, and the prediction model and motion state probability are updated according to the collected calculation results; Substitute the calculated environmental noise entropy and motion state into the noise reduction strategy database for query to obtain the actual noise reduction strategy. If the actual noise reduction strategy obtained is the same as the currently applied pre-selected noise reduction strategy, the noise reduction strategy is not changed. If the actual noise reduction strategy obtained is different from the currently applied pre-selected noise reduction strategy, the noise reduction strategy in the next prediction period is predicted according to the updated prediction model and motion state probability, and a new pre-selected noise reduction strategy is obtained and applied. It should be noted that the purpose of the steps is to reduce the energy consumption of the noise-cancelling headphones caused by real-time detection of environmental noise and motion status and frequent switching of noise reduction strategies, so that the noise-cancelling headphones can maximize the battery life to the required usage time predicted according to the user's usage habits; The technical solution of the embodiment of the present invention is: analyzing the historical usage data of the noise-cancelling headphones, predicting the user's expected usage period, and predicting the required remaining power of the noise-cancelling headphones based on the expected usage period, comparing the required remaining power with the actual remaining power, and judging whether the power is sufficient. If not, generating a power risk signal. If the generated power risk signal is received, suspending the real-time detection of environmental noise and motion status, establishing a model to predict the noise reduction strategy within the prediction period, re-detecting the environmental noise and motion status at the inspection time point, and updating the prediction model and the pre-selected noise reduction strategy.

[0016] Embodiment 3 like Figure 1 As shown, the method of a magnetic charging box for an adaptive noise reduction headset provided by an embodiment of the present invention further includes the following steps: Step 5. When the noise cancelling headphones are charging, analyze the power of the magnetic charging box and the noise cancelling headphones, and dynamically adjust the charging power based on the expected usage period; In some embodiments, if the noise cancelling earphone is in a charging state, the charging power of the magnetic charging box is dynamically adjusted; like Figure 3 As shown, the specific steps of dynamically adjusting the charging power are as follows; Specifically, the remaining power of the magnetic charging box and the used power of the noise cancelling headset are obtained by performing difference processing to obtain the charging power difference, and the difference is compared with the power difference threshold. If the charging power difference is greater than the power difference threshold, a sufficient power signal is generated, otherwise, an insufficient power signal is generated; Obtain the interval between the current time point and the starting point of the next expected usage period of the noise cancelling headset at the current time point, mark it as the required charging time, obtain the used power of the noise cancelling headset, perform ratio processing on the used power and the required charging time, and calculate the required charging rate of the noise cancelling headset; The required charging rate is compared with the charging rate threshold. If the required charging rate is greater than the charging rate threshold, it means that the charging demand is urgent and an urgent signal is generated. Otherwise, a non-urgent signal is generated. Selecting a suitable charging power for the magnetic charging box according to the generated signal; If the generated sufficient power signal and urgent signal are received, the noise reduction headset is quickly charged using a high-level charging power; If the generated sufficient power signal and non-urgent signal are received, the noise cancelling headset is normally charged using the middle charging power; If the generated low power signal is received, the noise cancelling headphones are slowly charged using a low-level charging power. At this time, the power reserve of the charging box itself must be prioritized, and only the basic power of the noise cancelling headphones is replenished; It should be noted that the purpose of the above steps is to avoid battery aging caused by long-term fast charging, give priority to emergency charging needs, and balance the power reserves of the earphones and the charging box to ensure the overall battery life of the system; The technical solution of the embodiment of the present invention is: when the noise-cancelling headphones are charging, the charging power is dynamically adjusted according to the power difference between the charging box and the headphones, the required charging time, and the required charging rate. When the power is sufficient and the charging demand is urgent, high-speed fast charging is adopted. When the power is sufficient but the charging demand is not urgent, normal charging is adopted at a medium gear. When the power is insufficient, low-speed slow charging is adopted to balance the power reserves of the headphones and the charging box, avoid battery aging, and ensure the overall battery life of the system.

[0017] Embodiment 4 like Figure 4 As shown, an embodiment of the present invention provides a system of a magnetic charging box for an adaptive noise reduction headset, which specifically includes the following modules: Data collection module: when the noise reduction headphones are in use, the user's motion state is determined based on the collected motion data, and the noise energy entropy of the environment is determined based on the collected ambient noise; In this embodiment, when the noise reduction headphones are in use, data is collected at the current detection time point through motion sensors such as the built-in accelerometer and gyroscope of the headphones, and the acceleration changes of the headphones on the x, y, and z axes and the rotation angular velocity of the headphones around the three coordinate axes are detected, and motion features are extracted from the collected acceleration and angular velocity data, and the extracted motion features are analyzed, and the user's current motion state is determined by a preset motion state classification algorithm; In the current detection period, an array consisting of several microphones is used to sample the environmental noise at a fixed sampling frequency fs; The obtained noise signal is amplified and filtered to remove high-frequency noise and interference, and the processed noise signal is subjected to fast Fourier transform to obtain a frequency domain signal. The energy ratio of each frequency component of the frequency domain signal after Laplace smoothing is calculated to obtain the probability distribution of the frequency component. The probability distribution of the frequency components of all microphones is jointly analyzed to calculate the energy entropy H of the ambient noise in the current detection period; Strategy selection module: Based on the noise energy entropy, the noise level of the scene where the user is located is determined, and the noise reduction strategy is selected based on the user's motion status; In this embodiment, the calculated environmental noise energy entropy H is compared with the environmental noise energy entropy low threshold Hd and the environmental noise energy entropy high threshold Hh; If the environmental noise energy entropy in the current detection period , judging that the scene noise level at the user's current location is quiet; If the environmental noise energy entropy in the current detection period , judging that the scene noise level at the user's current location is low noise; If the environmental noise energy entropy in the current detection period , judging that the scene noise level at the user's current location is highly noisy; Establishing a noise reduction strategy database including scene noisiness, user motion state, and corresponding noise reduction strategies, wherein the noise reduction strategy database stores the best noise reduction strategies corresponding to different combinations of environmental noise energy entropy ranges and user motion states; Based on the environmental noise energy entropy in the current detection period and the determined user motion state, query the noise reduction strategy database, and select a suitable noise reduction strategy for the noise reduction headphones according to the query result; Power analysis module: determines whether the remaining power of the noise-cancelling headphones is sufficient within the predicted time. If not, generates a power risk signal; In this embodiment, historical usage data of the noise cancelling headset is obtained, and the historical usage data includes the usage time period of the noise cancelling headset in each day within the historical usage time and the power consumption of the noise cancelling headset during the usage time period. The historical usage time represents a time period starting from the time node when the noise cancelling headset is first used by the user and ending at the current time node. The usage time period of the noise cancelling headset represents a time period during the historical usage time when the noise cancelling headset is in use; Based on any day in the historical usage time, obtain the interval duration between two adjacent usage periods in one day, where the interval duration represents the interval duration between the end point of the previous usage period and the start point of the next usage period, and compare the interval duration with the interval threshold; If the interval between two adjacent usage periods in a day is less than the interval threshold, the two adjacent usage periods are merged into one usage period, and the combined usage period takes the start point of the earlier usage period in the two usage periods as the starting point and the end point of the later usage period in the two usage periods as the end point; Traverse each day in the historical usage time to obtain several usage time periods that are merged or do not need to be merged, mark the starting point of each usage time period as the start time, and count all the start times in the historical usage time; Take any two of the start times, perform difference processing and take the absolute value to obtain the start time difference between the two start times; Compare the startup time difference between the two startup times with the interval threshold, and if the startup time difference between the two startup times is less than the interval threshold, the two startup times are classified into one startup time group; Traverse all the startup times in the historical usage time until all the startup times whose startup time differences are less than the interval threshold are classified into the startup time group. If there is a startup time that is not classified into any startup time group at this time, then this startup time itself is a startup time group; Based on any startup time group, the number of startup times in the startup time group is obtained, and the ratio is processed with the total number of all startup times in the historical usage time to obtain the startup frequency value of the startup time group; Compare the obtained startup frequency usage value with the startup frequency usage threshold; If the startup frequency value of the startup time group is less than or equal to the startup frequency threshold, it means that the user uses the noise reduction headset less frequently during the time range included in the startup time group; If the startup frequency value of the startup time group is greater than the startup frequency threshold, it means that the user uses the noise reduction headset more times in the time range included in the startup time group, and the startup time group is marked as a frequent use time group; Based on the obtained frequent use time group, all the start-up times included in the frequent use time group are obtained, and the sum and average are taken to obtain the expected start-up time of the noise reduction headset, and the duration of the use period corresponding to all the start-up times included in the frequent use time group are obtained, and the sum and average are processed to obtain the expected use duration corresponding to the expected start-up time; Combine the expected startup time with the corresponding expected usage time to obtain the expected usage period of the noise-cancelling headphones within a day; Obtain the power consumption of the noise cancelling headphones in the usage period corresponding to all the start-up times in the frequent use time group and sum them up to obtain the sum of power consumption, sum the duration of the usage period corresponding to all the start-up times in the frequent use time group to obtain the sum of usage duration, perform ratio processing on the obtained sum of power consumption and the sum of usage duration to obtain the expected power consumption rate corresponding to the expected usage period; The duration of all usage periods in the historical usage time is obtained and the average is calculated to obtain the average usage duration of the noise cancelling headphones during the historical usage time. The power consumption of the noise cancelling headphones in all usage periods is calculated and the average is calculated to obtain the average power consumption of the noise cancelling headphones during the historical usage time. The average power consumption is calculated and the average usage duration is calculated to obtain the average power consumption rate of the noise cancelling headphones. If the noise reduction headset is in use at the current time node, determine whether the current time node is within any expected usage period; If the current time node is within any expected usage period, obtain the interval between the current time node and the end of the expected usage period where the current time node is located, and multiply it by the expected power consumption rate of the expected usage period to obtain the required remaining power of the noise reduction headset; If the current time node is not within any expected usage period, obtain the duration of the current usage period of the noise cancelling headset at the current time node, perform a difference processing on the average usage duration of the noise cancelling headset and the duration of the current usage period, and perform a product processing on the obtained difference and the average power consumption rate of the noise cancelling headset to obtain the required remaining power of the noise cancelling headset; Obtain the actual remaining power of the noise cancelling headset at the current time node, perform subtraction processing on the remaining power required at this time, obtain the remaining power difference of the noise cancelling headset, and compare it with the remaining power difference threshold; If the remaining power difference of the noise cancelling headset is greater than or equal to the remaining power difference threshold, it indicates that the remaining power of the noise cancelling headset is sufficient as predicted based on the user's usage habits; If the remaining power difference of the noise cancelling headphones is less than the remaining power difference threshold, it is predicted based on the user's usage habits that the remaining power of the noise cancelling headphones is insufficient, and a power risk signal is generated; Strategy optimization module: If a power risk signal is received, real-time collection is suspended during the prediction period, and a pre-selected noise reduction strategy is established for the prediction period of the noise reduction headphones through the prediction model. The prediction model is calibrated based on the motion data and environmental noise before the test time point, and the calibrated pre-selected noise reduction strategy is obtained; In this embodiment, if the generated power risk signal is received, the real-time detection of environmental noise and motion status is suspended; Get the motion status of the noise-canceling headphones during each detection period during the current usage period and the ambient noise energy entropy , t=1, 2, 3, ..., m; where t represents the sequence number of the detection period in the current used period sorted by time, and m represents the number of detection periods in the current used period; The motion state is represented in discrete form. When the motion state is static, =0, when the motion state is walking, =1, when the movement state is running, =2, when the motion state is riding, =3; Calculate the probability of the noise-canceling headphones being in various motion states during the current usage period and compare them, and output the motion state with the highest probability as the motion state of the noise-canceling headphones during the prediction period; The prediction period represents a period for predicting the noise reduction strategy based on the environmental noise entropy and the motion state in the current used period at the current time point. The end of the prediction period is marked as the test time point, and the test time point is also the starting point of the next prediction period. The duration of each prediction period is the same; Construct a prediction model to predict the environmental noise energy entropy in the future c-th detection period , the specific formula is: ,in represents the rate of change, represents the intercept, through the environmental noise energy entropy The least squares fitting calculation was performed; According to the prediction model obtained by fitting calculation, the environmental noise energy entropy of the cth detection period within the prediction period is predicted and calculated to obtain the predicted environmental noise energy entropy range of each detection period within the prediction period, and the overlap with the judgment environmental noise energy entropy range of any scene noisiness is analyzed; Specifically, based on the determined environmental noise energy entropy range of any scene noisiness, the length of the overlapped range between the determined environmental noise energy entropy range and the predicted environmental noise energy entropy range is obtained, and the length of the predicted environmental noise energy entropy range is processed by ratio, so as to obtain the probability of the noise reduction headphones being noisy under the scene noisiness within the predicted period; Compare the noisy probabilities of the noise cancelling headphones under different scene noisiness levels, and output the scene noisiness level with the highest noisy probability as the predicted scene noisy level of the noise cancelling headphones within the predicted period; Substituting the obtained predicted motion state and predicted scene noisiness into the noise reduction strategy database for query, obtaining the pre-selected noise reduction strategy within the predicted period, and adjusting the noise reduction strategy of the noise reduction headset to the pre-selected noise reduction strategy; At the test time point of the prediction period, the environmental noise and motion state are collected by sensors, the environmental noise entropy is calculated, and the prediction model and motion state probability are updated according to the collected calculation results; Substitute the calculated environmental noise entropy and motion state into the noise reduction strategy database for query to obtain the actual noise reduction strategy. If the actual noise reduction strategy obtained is the same as the currently applied pre-selected noise reduction strategy, the noise reduction strategy is not changed. If the actual noise reduction strategy obtained is different from the currently applied pre-selected noise reduction strategy, the noise reduction strategy in the next prediction period is predicted according to the updated prediction model and motion state probability, and a new pre-selected noise reduction strategy is obtained and applied. Charging management module: When the noise-canceling headphones are charging, the power of the magnetic charging box and the noise-canceling headphones is analyzed, and the charging power is dynamically adjusted according to the expected usage period; In this embodiment, if the noise reduction headset is in a charging state, the charging power of the magnetic charging box is dynamically adjusted; The specific steps of dynamically adjusting the charging power are as follows: Specifically, the remaining power of the magnetic charging box and the used power of the noise cancelling headset are obtained by performing difference processing to obtain the charging power difference, and the difference is compared with the power difference threshold. If the charging power difference is greater than the power difference threshold, a sufficient power signal is generated, otherwise, an insufficient power signal is generated; Obtain the interval between the current time point and the starting point of the next expected usage period of the noise cancelling headset at the current time point, mark it as the required charging time, obtain the used power of the noise cancelling headset, perform ratio processing on the used power and the required charging time, and calculate the required charging rate of the noise cancelling headset; The required charging rate is compared with the charging rate threshold. If the required charging rate is greater than the charging rate threshold, it means that the charging demand is urgent and an urgent signal is generated. Otherwise, a non-urgent signal is generated. Selecting a suitable charging power for the magnetic charging box according to the generated signal; If the generated sufficient power signal and urgent signal are received, the noise reduction headset is quickly charged using a high-level charging power; If the generated sufficient power signal and non-urgent signal are received, the noise cancelling headset is normally charged using the middle charging power; If a low-battery signal is received, the noise cancelling headphones will be slowly charged using a low-level charging power. Priority will be given to protecting the power reserve of the charging box itself, and only the basic power will be replenished for the noise cancelling headphones.

[0018] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for an adaptive noise reduction headset magnetic charging box, characterized in that: The following steps are involved: When the noise-cancelling headphones are used, the user's motion state is determined based on the collected motion data, the noise energy entropy of the environment is determined based on the collected ambient noise, and the noisiness of the scene at the user's location is judged based on the noise energy entropy; Comprehensively analyze the scene noise and motion status to select a noise reduction strategy; Determine whether the remaining power of the noise-cancelling headphones is sufficient within the predicted time; If it is insufficient, real-time collection is suspended during the forecast period, and a pre-selected noise reduction strategy is selected for the forecast period of the noise reduction headphones through the forecast model; Based on the motion data and environmental noise before the inspection time point, the prediction model is calibrated, and a calibrated pre-selected noise reduction strategy is obtained; The prediction time consists of several prediction periods, and the test time point is the end point of the prediction period.

2. The method of the magnetic charging box of the adaptive noise reduction headset according to claim 1 is characterized in that: The method for determining the scene noisiness at the user's location is: The environmental noise energy entropy H is compared with the environmental noise energy entropy low threshold Hd and the environmental noise energy entropy high threshold Hh respectively; like , judging that the scene noise level at the user's current location is quiet; like , judging that the scene noise level at the user's current location is low noise; like , and judge that the scene noise level of the user's current location is highly noisy.

3. The method of the magnetic charging box for adaptive noise reduction headphones according to claim 1, characterized in that: The analysis process of whether the remaining power demand is sufficient is as follows: The actual remaining power of the noise cancelling headphones is compared with the required remaining power. If the actual remaining power is less than the required remaining power, the required remaining power is insufficient, and a power risk signal is generated.

4. The method for a magnetic charging box for an adaptive noise reduction headset according to claim 1, characterized in that: The method for obtaining the required remaining power is: Get the expected power consumption rate for all usage periods in the frequent usage time group; get the average power consumption rate of the noise-cancelling headset during the historical usage time; If the current time node is within the expected usage period, data processing is performed in combination with the expected power consumption rate of the expected usage period. If it is not within the expected usage period, data processing is performed in combination with the average power consumption rate of the noise cancelling headphones to obtain the required remaining power of the noise cancelling headphones.

5. The method of the magnetic charging box of the adaptive noise reduction headset according to claim 4 is characterized in that: The method for obtaining the expected usage period is as follows: Data processing is performed on all the start-up times and the durations of the corresponding usage periods contained in the frequency usage time group to obtain the expected usage period of the noise reduction headset in a single day.

6. The method for a magnetic charging box for an adaptive noise-cancelling headset according to claim 5, characterized in that: The method for obtaining the frequency-use time group is as follows: Get the time difference between two startup times in the historical usage time. If the time difference is less than the interval threshold, the two startup times are grouped into one startup time group. The number of startup times in the startup time group is processed to obtain a startup frequency value. If the startup frequency value is greater than a startup frequency threshold, the startup time group is marked as a frequent time group.

7. The method of the magnetic charging box for an adaptive noise reduction headset according to claim 1, characterized in that: The method for obtaining the pre-selected noise reduction strategy is as follows: Calculate the probability of the noise cancelling headset being in each motion state during the current usage period, and select the motion state with the highest probability as the predicted motion state of the noise cancelling headset; Combined with the historical environmental noise energy entropy, a prediction model is constructed by fitting with the least square method. The environmental noise energy entropy range within the prediction period is predicted and data analyzed. The probability of noise reduction headphones being noisy under the scene noisiness is calculated, and the scene noisiness with the highest probability of noise is used as the predicted scene noisiness of the noise reduction headphones. The pre-selected noise reduction strategy is obtained based on the predicted motion state and the predicted scene noisiness query.

8. The method for a magnetic charging box for an adaptive noise reduction headset according to claim 1, characterized in that: The calibrated pre-selected noise reduction strategy is obtained as follows: At the inspection time point of the prediction period, the environmental noise and motion status are collected, the environmental noise entropy is calculated, and the prediction model and motion state probability are updated. The actual noise reduction strategy is obtained based on the calculation results. If the actual noise reduction strategy is different from the currently applied pre-selected noise reduction strategy, the pre-selected noise reduction strategy is updated and applied based on the updated prediction model and motion state probability.

9. The method of the magnetic charging box of the adaptive noise reduction headset according to claim 1, characterized in that: The following steps are also included: When the noise cancelling headphones are charging, the power level of the magnetic charging box and the noise cancelling headphones is analyzed, and the charging power is dynamically adjusted based on the expected usage period; The method of dynamically adjusting the charging power is as follows: Get the remaining power of the magnetic charging box and the power used by the noise cancelling headphones, and determine whether the charging power is sufficient; If sufficient, analyze and obtain the required charging rate between the current time and the starting point of the next expected usage period. If the required charging rate is greater than the charging rate threshold, use the high-speed charging power for charging. Otherwise, use the medium-speed charging power for charging. If it is insufficient, use low-level charging power to charge.

10. A system for a magnetic charging box of an adaptive noise reduction headset, the system being used to implement the method as claimed in any one of claims 1 to 9, characterized in that: include: Data collection module: when the noise reduction headphones are in use, the user's motion state is determined based on the collected motion data, and the noise energy entropy of the environment is determined based on the collected ambient noise; Strategy selection module: Based on the noise energy entropy, the noise level of the scene where the user is located is determined, and the noise reduction strategy is selected based on the user's motion status; Power analysis module: determines whether the remaining power of the noise-cancelling headphones is sufficient within the predicted time. If not, generates a power risk signal; Strategy optimization module: If a power risk signal is received, real-time collection is suspended during the prediction period, and a pre-selected noise reduction strategy is selected for the prediction period of the noise reduction headphones through the prediction model. The prediction model is calibrated based on the motion data and environmental noise before the test time point, and the calibrated pre-selected noise reduction strategy is obtained; Charging management module: When the noise-canceling headphones are charging, the power level of the magnetic charging box and the noise-canceling headphones is analyzed, and the charging power is dynamically adjusted based on the expected usage period.

Images