Digital audio real-time transmission system based on body area network technology

By using a digital audio real-time transmission system based on body area network technology, combined with synchronized voice and video guidance and multi-location audio acquisition, the problem of quantifying the rehabilitation exercise effects of paralyzed patients has been solved. This enables dynamic, accurate, and real-time rehabilitation monitoring, ensuring the safety of the rehabilitation process.

CN121509873APending Publication Date: 2026-02-10DONGGUAN MINDONG ELECTRONIC TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202511361284.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-23
Publication Date
2026-02-10

AI Technical Summary

Technical Problem

Existing technologies make it difficult to quantify and monitor the therapeutic effects of rehabilitation exercises on paralyzed patients, which hinders the improvement of rehabilitation exercises.

Method used

The digital audio real-time transmission system based on body area network technology combines a voice prompt module, an audio acquisition module, a storage module, a preprocessing module, an analysis module, and a diagnostic module to achieve synchronized voice and video guidance, multi-location audio acquisition, sound quality optimization processing, and difference rate analysis, thereby enabling real-time monitoring of the patient's recovery trend.

Benefits of technology

It enables dynamic, precise, and real-time monitoring of patients' rehabilitation exercises, provides quantitative assessments, and ensures the safety and effectiveness of the rehabilitation process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121509873A_ABST
    Figure CN121509873A_ABST
Patent Text Reader

Abstract

The invention discloses a digital audio real-time transmission system based on a body area network technology, and relates to the field of digital audio transmission, and the system comprises a voice prompt module which is used for building a guiding voice catalog and playing guiding voices in the guiding voice catalog in sequence; the audio acquisition module is used for acquiring audio information when the patient listens to the guide audio and tries to make corresponding actions according to the guide audio; according to the invention, through combination of voice guidance and synchronous video, differential action instructions are dynamically generated, influence of repeated instructions on monitoring accuracy is avoided, multi-position audio accurate acquisition is realized based on a body area network technology, audio information quality is improved through combination of tone quality optimization processing, and through difference rate analysis and intensity correction, the real-time monitoring accuracy is improved. Action-related audio changes of a patient are accurately captured, rehabilitation trend safety diagnosis is realized in combination with continuous comparison, and a monitoring result is transmitted in real time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of digital audio transmission technology, specifically to a real-time digital audio transmission system based on body area network technology. Background Technology

[0002] A digital audio system is a system that digitally processes, transmits, stores, and plays back audio signals. It converts analog audio into digital signals, processes them through encoding and compression, and then uses devices to record, edit, transmit, and play them back. It is widely used in communications, entertainment, and other fields, and can improve audio quality and transmission efficiency.

[0003] Patent application No. 202410783797.0 discloses a real-time digital audio transmission system based on body area network technology, including an audio acquisition system, an audio encoding system, a central processing unit, an audio transmission system, an audio decoding system, and an audio playback system. The audio acquisition system and the audio encoding system are communicatively connected, the audio encoding system and the central processing unit are communicatively connected, the central processing unit and the audio transmission system are communicatively connected, the audio transmission system and the audio decoding system are communicatively connected, and the audio decoding system and the audio playback system are communicatively connected. This application aims to solve the problem that "existing technical solutions may become difficult to meet new requirements without restructuring, and may even lead to code corruption and chaos, which will increase labor costs."

[0004] However, for paralyzed patients, this group often has the characteristics of being able to hear and see, but the therapeutic effect of their rehabilitation exercises during the process of being guided to do rehabilitation exercises is currently difficult to quantify and monitor, so as to help improve the rehabilitation exercises.

[0005] To address this, we propose a real-time digital audio transmission system based on body area network (BAN) technology. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a digital audio real-time transmission system based on body area network technology, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions;

[0008] This invention discloses a real-time digital audio transmission system based on body area network technology, comprising:

[0009] The system comprises the following modules: a voice prompt module for creating a directory of guiding voice messages and playing them sequentially; an audio acquisition module for acquiring audio information as the patient listens to and attempts to perform actions according to the guidance audio; a storage module that follows the audio acquisition module, receiving and storing audio information during the sequential playback of the guidance audio; a preprocessing module that accesses the storage module, performs sound quality enhancement on each stored audio message, and iterates the corresponding original audio message with the enhanced audio message; an analysis module that receives the enhanced audio message, analyzes the audio information difference rate, and corrects the analysis results based on the overall intensity of the audio information; a diagnosis module that receives two consecutive corrected results from the analysis module, compares the latest corrected result with the other corrected result to diagnose whether the patient's recovery trend is safe; and an interaction module that acquires the audio information difference rate analysis results and correction results from the analysis module and the diagnosis results from the diagnosis module in real time, and transmits the analysis results, correction results, and diagnosis results to a preset receiving terminal.

[0010] Furthermore, the voice prompt module is equipped with a guidance voice library, which stores several guidance voices. Each guidance voice is marked with the name of a corresponding human limb, and the content of each guidance voice is to instruct the patient to perform a specified action. The guidance voices marked with the same human limb name have different corresponding instruction contents.

[0011] During the operation of the voice prompt module, the system user can manually select or manually determine a human limb name marker in the guidance voice library, and then randomly select a preset number of guidance voices based on the marker to build a guidance voice catalog.

[0012] When selecting guidance audio from the guidance audio library through random selection, the following conditions must be met: when the same human limb name marker is selected twice consecutively, and the corresponding guidance audio is randomly selected under the same human limb name marker, the repetition rate of the two randomly selected guidance audios does not exceed 50%. The voice prompt module is wired to the display screen. When the voice prompt module plays guidance audio in sequence, a preset guidance video that matches the currently playing guidance audio is played on the display screen to synchronously instruct the patient to perform the specified action.

[0013] Furthermore, the audio acquisition module is integrated with several miniature audio acquisition units. Each miniature audio acquisition unit has a built-in short-range wireless communication unit and a distributed network topology is constructed based on body area network technology. The miniature audio acquisition units are adaptively deployed along the inner surface of the patient's outer clothing and trousers.

[0014] The miniature audio acquisition device is deployed to cover the key acoustic conduction areas of the human body's limbs, including the shoulders, waist, and front of the thighs.

[0015] Furthermore, the storage module distinguishes storage areas based on the start and end times of playback of the guidance voice directory, so that the internally stored audio information is stored separately according to each constructed guidance voice directory.

[0016] The preprocessing module runs continuously following the operation of the storage module to update the stored audio information, so that when the storage module runs to update the stored audio information, the audio information of the previous set of stored information is simultaneously processed to enhance the sound quality.

[0017] The audio quality enhancement processing operations for audio information in the preprocessing module include:

[0018] ;

[0019] In the formula: To enhance the processed audio signal; This is the original audio signal that was acquired. The significance factor of the target signal; It is a noise suppression factor; It is a very small positive number;

[0020] in, The default value is This is used to avoid zero denominators in the formula. n represents the sampling point index, and k represents the discrete frequency point index. The above formula is used to output the audio information after sound quality enhancement processing.

[0021] Furthermore, the values ​​of the target signal saliency factor and the noise suppression factor follow the following order:

[0022] ;

[0023] In the formula: It is a positive coefficient used to adjust the degree of influence of its product object in the calculation of the significance factor of the target signal; It is a positive coefficient used to adjust the strength of the effect of its product object in the calculation of the noise suppression factor; The discrete index corresponding to the reference frequency of the target signal; The signal energy at discrete points; The sampling point interval; The average energy of the sliding window over the discrete frequency k; The frequency index jump value between adjacent sampling points;

[0024] in, ∈ (0.01, 0.5) ∈ (0.1,5), The value is positively correlated with the frequency concentration and energy continuity of the target signal; ∈ (0.05, 2), ∈ (0.02, 1), It is positively correlated with the intensity of energy abrupt changes and the amplitude of frequency jumps in noise.

[0025] Furthermore, the audio information received during the operation of the analysis module consists of the earliest and latest sets of audio information stored in the storage module based on iterative storage.

[0026] The audio information difference rate analysis logic is expressed as follows:

[0027] ;

[0028] In the formula: The difference rate between audio information X and audio information Y; The initial value is set to 0.7, and the range is 0.7 ± 0.1. The feature layer difference; Semantic layer difference;

[0029] Specifically, based on the above formula, the difference rate of each corresponding audio information in the earliest and latest audio information groups is calculated, and the average of the calculation results is recorded as the audio information difference rate. .

[0030] Furthermore, the correction logic for the audio information difference rate analysis results is expressed as follows:

[0031] ;

[0032] In the formula: The results of the audio information difference rate analysis are as follows; It is the average of the minimum and maximum audio amplitude values ​​of each audio message in the latest set of audio information and the earliest set of audio information. It is the average of the maximum audio amplitude values ​​of each audio message in the latest set of audio information and the earliest set of audio information.

[0033] in, When the value is greater than or equal to 1, .

[0034] Furthermore, if the latest correction result in the diagnostic module is greater than another correction result, the patient's recovery trend is determined to be safe; otherwise, the patient's recovery trend is determined to be deteriorating.

[0035] When a patient's recovery trend is determined to be deteriorating in the diagnostic module, the system user can actively adjust the currently designed rehabilitation training after the interaction module transmits the analysis results, correction results, and diagnostic results to the preset receiving end.

[0036] Furthermore, the interactive module transmits the analysis results, correction results, and diagnostic results to a preset receiving end, where the system user reads each result.

[0037] The analysis and correction results are recorded on the preset receiving end in the form of a line graph. The diagnostic results are cached synchronously on the preset receiving end, and each diagnostic result is marked with an output timestamp.

[0038] Furthermore, the voice prompt module is internally connected to a guidance voice library via a wireless network. The voice prompt module is also internally connected to an audio acquisition module and a storage module via a wireless network. The storage module is internally connected to a preprocessing module via a wireless network. The preprocessing module is internally connected to an analysis module via a wireless network. The analysis module is internally connected to a diagnostic module via a wireless network. The diagnostic module is internally connected to the interaction module via a wireless network.

[0039] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0040] This invention provides a real-time digital audio transmission system based on body area network (BNB) technology. During operation, the system dynamically generates differentiated action commands by combining voice guidance with synchronized video, avoiding repetitive commands that could affect monitoring accuracy. It leverages BNB technology to achieve precise multi-location audio acquisition, enhances audio quality through sound quality optimization, and accurately captures audio changes related to patient actions through difference rate analysis and intensity correction. Continuous comparison enables safe diagnosis of rehabilitation trends, and real-time transmission of monitoring results provides system users with reference and timely adjustments to rehabilitation plans. This ensures the dynamism, accuracy, and real-time nature of rehabilitation monitoring, guarantees the safety of the patient's rehabilitation process, and provides a quantitative assessment service for the effectiveness of rehabilitation treatment. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0042] Figure 1 This is a schematic diagram of the structure of a real-time digital audio transmission system based on body area network technology. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0044] The present invention will be further described below with reference to embodiments.

[0045] Example:

[0046] This embodiment describes a real-time digital audio transmission system based on body area network technology, such as... Figure 1 As shown, it includes:

[0047] The voice prompt module is used to create a guide voice catalog and play the guide voices in the catalog in sequence.

[0048] The voice prompt module has a built-in guidance voice library, which stores several guidance voices. Each guidance voice is marked with the corresponding human limb name, and the content of each guidance voice is to instruct the patient to perform a specific action. The guidance voices marked with the same human limb name have different corresponding instruction content.

[0049] During the operation of the voice prompt module, the system user can manually select or manually determine a human limb name marker in the guidance voice library, and then randomly select a preset number of guidance voices based on the marker to build a guidance voice catalog.

[0050] When selecting guidance voices from the guidance voice library through random selection, the following conditions must be met: when the same human limb name marker is determined twice consecutively, and the corresponding guidance audio is randomly selected under the same human limb name marker, the repetition rate of the two randomly selected guidance audios does not exceed 50%. The voice prompt module is wired to the display screen. When the voice prompt module plays guidance audios in sequence, a preset guidance video that matches the currently playing guidance audio is played on the display screen to synchronously instruct the patient to perform the specified action.

[0051] The audio acquisition module is used to collect audio information of the patient while listening to the guidance audio and attempting to perform the corresponding actions according to the guidance audio;

[0052] The audio acquisition module is composed of several miniature audio acquisition units. Each miniature audio acquisition unit has a built-in short-range wireless communication unit and is built into a distributed network topology based on body area network technology. The miniature audio acquisition units are adaptively deployed along the inner surface of the patient's outer clothing and trousers.

[0053] Among them, the miniature audio acquisition device should be deployed in accordance with the following principle: it should be deployed to cover the key acoustic conduction areas of the human body, including the shoulders, waist, and front of the thighs.

[0054] The storage module is used to follow the audio acquisition module and receive and store audio information during the sequential playback of the guide audio.

[0055] The storage module distinguishes storage areas based on the start and end settings of the playback of the guide voice catalog, so that the internally stored audio information is stored separately according to each guide voice catalog built.

[0056] The preprocessing module runs continuously following the storage module's operation of updating stored audio information, so that each time the storage module runs to update stored audio information, the previously stored audio information is simultaneously processed for sound quality enhancement.

[0057] The preprocessing module is used to access the storage module, perform sound quality enhancement processing on each stored audio information in the storage module, and iterate the corresponding original audio information with the audio information output after sound quality enhancement.

[0058] The audio quality enhancement operations in the preprocessing module include:

[0059] ;

[0060] In the formula: To enhance the processed audio signal; This is the original audio signal that was acquired. The significance factor of the target signal; It is a noise suppression factor; It is a very small positive number;

[0061] in, The default value is This is used to avoid zero denominators in the formula. n represents the sampling point index, and k represents the discrete frequency point index. The above formula is used to output the audio information after sound quality enhancement processing.

[0062] The above formula uses the target signal significance factor Enhance effective audio signals At the same time, with the help of noise suppression factors By reducing noise interference and using minimal positive numbers to avoid zero denominators, targeted enhancement of the acquired audio is achieved to adapt to the complex acoustic environment of key acoustic areas of the human limbs in a body area network. Based on the characteristics of motion-related audio acquired from the body area network, the target signal and noise are dynamically balanced. Furthermore, a mechanism ensuring that the repetition rate of the guidance audio under two consecutive identical limb markers is ≤50% ensures that the enhancement processing can adapt to audio changes caused by diverse movements, effectively improving audio quality in different action scenarios. The formula also includes... and All are audio signals (with consistent dimensions, such as pressure-related units). and Ɛ is a dimensionless adjustment factor, and Ɛ is a dimensionless minimum positive number. The whole system satisfies the dimensionless uniformity.

[0063] The values ​​of the target signal significance factor and the noise suppression factor follow the following rules:

[0064] ;

[0065] In the formula: It is a positive coefficient used to adjust the degree of influence of its product object in the calculation of the significance factor of the target signal; It is a positive coefficient used to adjust the strength of the effect of its product object in the calculation of the noise suppression factor; The discrete index corresponding to the reference frequency of the target signal; The signal energy at discrete points; The sampling point interval; The average energy of the sliding window over the discrete frequency k; The frequency index jump value between adjacent sampling points;

[0066] in, ∈ (0.01, 0.5) ∈ (0.1,5), The value is positively correlated with the frequency concentration and energy continuity of the target signal; ∈ (0.05, 2), ∈ (0.02, 1), It is positively correlated with the intensity of energy jumps and the amplitude of frequency jumps in noise;

[0067] The above formula dynamically adjusts the factor values ​​based on the frequency concentration and energy continuity of the target signal and the energy abruptness and frequency jump amplitude of the noise to adapt to different signal characteristics. By setting the positive correlation between the coefficients and the inherent characteristics of the signal, the factors can adaptively distinguish between the target signal and noise, thereby improving the sound quality enhancement accuracy in complex scenarios. All of them are dimensionless coefficients, and the proportional calculation of parameters such as signal energy and sliding window average energy is also dimensionless. Therefore, the factors are dimensionless as a whole, which matches the requirement for dimensions in the sound quality enhancement formula and is a reasonable design.

[0068] The analysis module is used to receive audio information after sound enhancement processing, analyze the audio information difference rate, and correct the audio information difference rate analysis results based on the overall intensity of the audio information.

[0069] The audio information received during the analysis module's operation phase consists of the earliest and latest sets of audio information stored in the storage module based on iterative storage.

[0070] The audio information difference rate analysis logic is expressed as follows:

[0071] ;

[0072] In the formula: The difference rate between audio information X and audio information Y; The initial value is set to 0.7, and the range is 0.7 ± 0.1. The feature layer difference; Semantic layer difference;

[0073] Specifically, based on the above formula, the difference rate of each corresponding audio information in the earliest and latest audio information groups is calculated, and the average of the calculation results is recorded as the audio information difference rate. ;

[0074] The above formula integrates feature layer difference and semantic layer difference by weighting coefficients to comprehensively evaluate the overall difference between the earliest and latest audio information. It breaks through the limitations of comparing a single feature and combines feature and semantic differences to more comprehensively reflect audio changes. In terms of parameter dimensions, it is a dimensionless weight. Both feature and semantic difference have been normalized to be dimensionless, so they are dimensionless values, which meet the quantitative requirements of difference rate.

[0075] The expression is:

[0076] ;

[0077] In the formula: For entropy weight, , The information entropy represents the k-th feature; For feature differences;

[0078] Including differences in time and space rhythm Wavelet frequency domain differences Psychoacoustic complex differences Spatial sound field differences ;

[0079] The differences in spatiotemporal rhythm, wavelet frequency domain, and spatial sound field are obtained based on any calculation method in the existing technology.

[0080] The psychoacoustic composite difference fusion of loudness, sharpness, and roughness three-dimensional perceptual features is obtained by weighted Euclidean distance normalization.

[0081] The expression is:

[0082] ;

[0083] In the formula: The semantic vectors of audio information X and audio information Y; The information entropy of a semantic vector;

[0084] The correction logic for the audio information difference rate analysis results is expressed as follows:

[0085] ;

[0086] In the formula: The results of the audio information difference rate analysis are as follows; It is the average of the minimum and maximum audio amplitude values ​​of each audio message in the latest set of audio information and the earliest set of audio information. It is the average of the maximum audio amplitude values ​​of each audio message in the latest set of audio information and the earliest set of audio information.

[0087] in, When the value is greater than or equal to 1, ;

[0088] The diagnostic module receives two consecutive corrected results from the analysis module, compares the latest corrected result with the other corrected result, and diagnoses whether the patient's recovery trend is safe.

[0089] If the latest corrected result in the diagnostic module is greater than another corrected result, the patient's recovery trend is determined to be safe; otherwise, the patient's recovery trend is determined to be deteriorating.

[0090] When the patient's recovery trend is determined to be deteriorating in the diagnostic module, after the interaction module transmits the analysis results, correction results and diagnostic results to the preset receiving end, the system user can actively adjust the currently designed rehabilitation training.

[0091] It should be noted that system users can proactively adjust the currently designed rehabilitation training, that is, change the type of rehabilitation training, such as changing muscle tissue massage to mechanical assisted training. The purpose is to change the rehabilitation training method to adapt to or optimize the patient's rehabilitation training effect.

[0092] The interaction module is used to acquire the audio information difference rate analysis results and correction results from the analysis module and the diagnostic results from the diagnostic module in real time, and transmit the analysis results, correction results and diagnostic results to the preset receiving end.

[0093] The interactive module transmits the analysis results, correction results, and diagnostic results to the preset receiving end, and the system user reads each result on the preset receiving end;

[0094] The analysis results and correction results are recorded on the preset receiving end in the form of a line graph. The diagnostic results are cached synchronously on the preset receiving end, and each diagnostic result is marked with an output timestamp.

[0095] The voice prompt module has a guidance voice library connected via a wireless network. The voice prompt module also has an audio acquisition module and a storage module connected via a wireless network. The storage module has a preprocessing module connected via a wireless network. The preprocessing module has an analysis module connected via a wireless network. The analysis module has a diagnostic module connected via a wireless network. The diagnostic module is connected to the interaction module via a wireless network.

[0096] In this embodiment, the voice prompt module runs to build a guide voice catalog and plays the guide voices in the catalog sequentially. The audio acquisition module then runs to collect audio information of the patient listening to the guide audio and attempting to perform corresponding actions according to the guide audio. The storage module further follows the audio acquisition module, receiving and storing audio information during the sequential playback of the guide audio. The preprocessing module then accesses the storage module to perform sound quality enhancement processing on each stored audio information and iterates the corresponding original audio information with the audio information output after sound quality enhancement. The analysis module receives the audio information after sound quality enhancement, analyzes the audio information difference rate, and corrects the audio information difference rate analysis result based on the overall intensity of the audio information. The diagnosis module simultaneously receives the correction results output by the analysis module twice in a row, compares the latest correction result with the other correction result to diagnose whether the patient's recovery trend is safe. Finally, the interaction module obtains the audio information difference rate analysis result and correction result from the analysis module and the diagnosis result from the diagnosis module in real time, and transmits the analysis result, correction result, and diagnosis result to the preset receiving end.

[0097] Through the system operation described in the above embodiments, patients are guided to perform diverse movements synchronously via voice and video. Audio related to these movements is accurately collected from multiple locations, processed, and analyzed to precisely determine the rehabilitation trend. Results are transmitted in real time, facilitating timely adjustments to the treatment plan, improving the accuracy and timeliness of rehabilitation monitoring, aiding in the scientific assessment of rehabilitation progress, and ensuring rehabilitation safety.

[0098] The following is an application example of the system described in the above embodiments:

[0099] Taking upper limb rehabilitation training for stroke patients as an example, the application process of this system is as follows:

[0100] 1. Create a guide voice catalog

[0101] The therapist (system user) selects "upper limb" as the name for the four limbs in the system and randomly selects three guidance voices marked "upper limb" from the guidance voice library of the voice prompt module to create a guidance voice catalog. This catalog includes: "Raise the left arm to shoulder height," "Extend the right arm forward and then bend the elbow," and "Raise both arms to the sides and hold for 3 seconds." If the therapist selects the "upper limb" mark again the next day and randomly selects three guidance voices, the two selections must not repeat more than one voice (repetition rate ≤ 50%).

[0102] 2. Synchronized playback instructions and audio capture

[0103] The voice prompt module plays the guiding voice messages in the catalog sequentially, while the display screen simultaneously plays demonstration videos of the corresponding actions (e.g., when playing "raise your left arm to shoulder height," the video shows the standard posture for that action). After hearing the voice messages and seeing the videos, the patient attempts to perform the corresponding actions.

[0104] At this point, miniature audio acquisition units (based on a distributed network built on a body area network) deployed along the inner surface of the patient's outer garment begin to operate, collecting audio information (such as muscle contraction sounds, joint movement sounds, etc.) during the patient's movements. The acquisition units cover key acoustic conduction areas such as the shoulders and waist, ensuring that acoustic signals related to upper limb movements are completely captured.

[0105] 3. Audio storage and preprocessing

[0106] The storage module divides the storage area separately according to the playback time of the audio catalog of this guide, and stores the audio information corresponding to the three actions collected separately (separate from the training audio of the previous day).

[0107] The preprocessing module simultaneously enhances the audio quality of the audio stored the previous day: by calculating the combined value of the target signal saliency factor and the noise suppression factor (combined with the original audio signal, using a very small positive number 10 to avoid the denominator being zero), the audio clarity is enhanced, and the processed audio replaces the original audio.

[0108] 4. Analyze audio difference rate

[0109] On day 10 of training, the analysis module retrieves the audio information from day 1 (the earliest group) and day 10 (the latest group) from the storage module and calculates the difference rate:

[0110] First, calculate the feature layer difference: combine the difference sub-items of four types of features, namely spatiotemporal rhythm, wavelet frequency domain, psychoacoustic composite, and spatial sound field, and determine the weights by combining the information entropy of each feature to obtain the feature layer difference.

[0111] Next, calculate the semantic layer difference: based on the semantic vector of the audio and its information entropy, obtain the semantic layer difference.

[0112] The two items are combined with a weight of 0.7. The difference rate of each corresponding audio is calculated and then averaged to obtain the initial difference rate.

[0113] The difference rate is obtained by combining the maximum and minimum average amplitude values ​​of each audio element in the two sets of audio and then correcting the difference rate.

[0114] 5. Diagnosis, Rehabilitation Trends, and Feedback

[0115] The diagnostic module compares the corrected difference rate between day 9 and day 10: if the difference rate on day 10 is greater than that on day 9, the patient's upper limb rehabilitation trend is considered safe (the changes in movement-related audio are consistent with the rehabilitation pattern); if it is less, the trend is considered to have deteriorated.

[0116] The interactive module sends the difference rate between the two tests to the therapist's terminal in the form of a line graph, while simultaneously caching the diagnostic results (marked with timestamps). After reviewing the data, if the trend worsens, the therapist can adjust the difficulty or frequency of movements in the guided voice catalog to optimize the training program.

[0117] In summary, during operation, the system in the above embodiments dynamically generates differentiated action commands by combining voice guidance with synchronized video, avoiding the impact of repetitive commands on monitoring accuracy. It also leverages body area network technology to achieve precise multi-location audio acquisition, enhances audio information quality through sound quality optimization, accurately captures audio changes related to patient actions through difference rate analysis and intensity correction, and achieves safe diagnosis of rehabilitation trends through continuous comparison. Real-time transmission of monitoring results provides system-side users with reference and timely adjustments to rehabilitation plans, ensuring the dynamism, accuracy, and real-time nature of rehabilitation monitoring, guaranteeing the safety of the patient's rehabilitation process, and providing quantitative assessment services for the effectiveness of patient rehabilitation treatment.

[0118] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time digital audio transmission system based on body area network technology, characterized in that, include: The voice prompt module is used to create a guide voice catalog and play the guide voices in the catalog in sequence. The audio acquisition module is used to collect audio information of the patient while listening to the guidance audio and attempting to perform the corresponding actions according to the guidance audio; The storage module is used to follow the audio acquisition module and receive and store audio information during the sequential playback of the guide audio. The preprocessing module is used to access the storage module, perform sound quality enhancement processing on each stored audio information in the storage module, and iterate the corresponding original audio information with the audio information output after sound quality enhancement. The analysis module is used to receive audio information after sound enhancement processing, analyze the audio information difference rate, and correct the audio information difference rate analysis results based on the overall intensity of the audio information. The diagnostic module receives two consecutive corrected results from the analysis module, compares the latest corrected result with the other corrected result, and diagnoses whether the patient's recovery trend is safe. The interaction module is used to acquire the audio information difference rate analysis results and correction results from the analysis module and the diagnostic results from the diagnostic module in real time, and transmit the analysis results, correction results and diagnostic results to the preset receiving end.

2. The digital audio real-time transmission system based on body area network technology according to claim 1, characterized in that, The voice prompt module has a guidance voice library, which stores several guidance voices. Each guidance voice is marked with the name of a corresponding human limb, and the content of each guidance voice is to instruct the patient to perform a specified action. The guidance voices marked with the same human limb name have different corresponding instruction content. During the operation of the voice prompt module, the system user can manually select or manually determine a human limb name marker in the guidance voice library, and then randomly select a preset number of guidance voices based on the marker to build a guidance voice catalog. When selecting guidance audio from the guidance audio library through random selection, the following conditions must be met: when the same human limb name marker is selected twice consecutively, and the corresponding guidance audio is randomly selected under the same human limb name marker, the repetition rate of the two randomly selected guidance audios does not exceed 50%. The voice prompt module is wired to the display screen. When the voice prompt module plays guidance audio in sequence, a preset guidance video that matches the currently playing guidance audio is played on the display screen to synchronously instruct the patient to perform the specified action.

3. The digital audio real-time transmission system based on body area network technology according to claim 1, characterized in that, The audio acquisition module is composed of several miniature audio acquisition units. Each miniature audio acquisition unit has a built-in short-range wireless communication unit and is built into a distributed network topology based on body area network technology. The miniature audio acquisition units are adaptively deployed along the inner surface of the patient's outer clothing and trousers. The miniature audio acquisition device is deployed to cover the key acoustic conduction areas of the human body's limbs, including the shoulders, waist, and front of the thighs.

4. The digital audio real-time transmission system based on body area network technology according to claim 1, characterized in that, The storage module distinguishes storage areas based on the start and end times of playback of the guide voice directory, so that the internally stored audio information is stored separately according to each guide voice directory that is constructed. The preprocessing module runs continuously following the operation of the storage module to update the stored audio information, so that when the storage module runs to update the stored audio information, the audio information of the previous set of stored information is simultaneously processed to enhance the sound quality. The audio quality enhancement processing operations for audio information in the preprocessing module include: ; In the formula: To enhance the processed audio signal; This is the original audio signal that was acquired. The significance factor of the target signal; It is a noise suppression factor; It is a very small positive number; in, The default value is This is used to avoid zero denominators in the formula. n represents the sampling point index, and k represents the discrete frequency point index. The above formula is used to output the audio information after sound quality enhancement processing.

5. A real-time digital audio transmission system based on body area network technology according to claim 4, characterized in that, The values ​​of the target signal significance factor and the noise suppression factor follow the following rules: ; In the formula: It is a positive coefficient used to adjust the degree of influence of its product object in the calculation of the significance factor of the target signal; It is a positive coefficient used to adjust the strength of the effect of its product object in the calculation of the noise suppression factor; The discrete index corresponding to the reference frequency of the target signal; The signal energy at discrete points; The sampling point interval; The average energy of the sliding window over the discrete frequency k; The frequency index jump value between adjacent sampling points; in, ∈ (0.01, 0.5) ∈ (0.1,5), The value is positively correlated with the frequency concentration and energy continuity of the target signal; ∈ (0.05, 2), ∈ (0.02, 1), It is positively correlated with the intensity of the energy jump and the amplitude of the frequency jump in noise.

6. A real-time digital audio transmission system based on body area network technology according to claim 1, characterized in that, The audio information received during the operation of the analysis module consists of the earliest and latest sets of audio information stored in the storage module based on iterative storage. The audio information difference rate analysis logic is expressed as follows: ; In the formula: The difference rate between audio information X and audio information Y; The initial value is set to 0.7, and the range is 0.7 ± 0.

1. The feature layer difference; Semantic layer difference; Specifically, based on the above formula, the difference rate of each corresponding audio information in the earliest and latest audio information groups is calculated, and the average of the calculation results is recorded as the audio information difference rate. .

7. A real-time digital audio transmission system based on body area network technology according to claim 6, characterized in that, The correction logic for the audio information difference rate analysis results is expressed as follows: ; In the formula: The results of the audio information difference rate analysis are as follows; It is the average of the minimum and maximum audio amplitude values ​​of each audio message in the latest set of audio information and the earliest set of audio information. It is the average of the maximum audio amplitude values ​​of each audio message in the latest set of audio information and the earliest set of audio information. in, When the value is greater than or equal to 1, .

8. A real-time digital audio transmission system based on body area network technology according to claim 1, characterized in that, When the latest correction result in the diagnostic module is greater than another correction result, the patient's recovery trend is determined to be safe; otherwise, the patient's recovery trend is determined to be deteriorating. When a patient's recovery trend is determined to be deteriorating in the diagnostic module, the system user can actively adjust the currently designed rehabilitation training after the interaction module transmits the analysis results, correction results, and diagnostic results to the preset receiving end.

9. A digital audio real-time transmission system based on body area network technology according to claim 1, characterized in that, The interactive module transmits the analysis results, correction results, and diagnostic results to the preset receiving end, and the system user reads each result on the preset receiving end; The analysis and correction results are recorded on the preset receiving end in the form of a line graph. The diagnostic results are cached synchronously on the preset receiving end, and each diagnostic result is marked with an output timestamp.

10. A digital audio real-time transmission system based on body area network technology according to claim 1, characterized in that, The voice prompt module is internally connected to a guidance voice library via a wireless network. The voice prompt module is also internally connected to an audio acquisition module and a storage module via a wireless network. The storage module is internally connected to a preprocessing module via a wireless network. The preprocessing module is internally connected to an analysis module via a wireless network. The analysis module is internally connected to a diagnostic module via a wireless network. The diagnostic module is internally connected to the interaction module via a wireless network.

Citation Information

Patent Citations

  • Digital audio real-time transmission system based on body area network technology

    CN118692475A