A low-power-consumption bluetooth earphone audio data transmission method and system

Abstract

This invention relates to the field of wireless communication technology and discloses a method and system for low-power Bluetooth headset audio data transmission. The method includes: acquiring an audio data stream and performing frequency domain conversion, spectrum segmentation, and amplitude calibration; calculating the compression ratio to form a preliminary compression scheme; weighted segmentation and iterative calibration of the audio data stream to obtain optimized compression blocks; extracting interference intensity based on channel state information; calculating transmission stability and interference disruption; generating compensation factors and configuring channel adaptation parameters; fusing channel adaptation parameters to locate redundant vectors; simplifying the optimized compression blocks to obtain simplified data units; and obtaining a burden-optimized version of the data through spectrum decomposition and high-frequency lightweight coding; parsing the receiver feedback signal to obtain sound quality indicators and signal-to-noise ratio; constructing a matrix to complete sound quality evaluation; locating audio defects based on the evaluation results; generating compensation data packets and verifying continuity to obtain the final audio stream. This invention solves the problem of low-power transmission stability.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, specifically to a low-power Bluetooth headset audio data transmission method and system. Background Technology

[0002] With the widespread adoption of smart audio devices and the extensive application of speech synthesis technology, low-power Bluetooth headsets have become core terminal devices for music playback, voice calls, and speech synthesis broadcasting. Existing Bluetooth headset audio transmission solutions mostly employ fixed compression ratios, basic channel adaptation, and simple retransmission mechanisms, making it difficult to balance low-power constraints with audio fidelity requirements.

[0003] The conventional transmission process of existing technologies involves encoding and compressing the audio data stream at a preset fixed compression ratio at the audio sending end, performing simple channel parameter adaptation based on basic channel state information, and performing simple retransmission operations when packet loss occurs. Throughout the process, the encoding strategy is not adjusted according to dynamic changes in the channel, nor is a closed-loop audio quality compensation mechanism built in conjunction with feedback from the receiving end.

[0004] However, existing transmission methods mostly rely on fixed compression strategies, failing to dynamically adjust compression granularity and coding complexity based on channel conditions, and also neglecting closed-loop audio quality compensation based on receiver status. In scenarios with strong interference and multiple concurrent device connections, packet loss, bit errors, and increased latency are prone to occur, leading to discontinuous audio in the synthesized speech. Furthermore, excessive computational and retransmission overhead directly shortens headphone battery life, failing to meet the requirements for long-term stable use. Therefore, existing low-power Bluetooth audio transmission solutions suffer from the technical deficiency of low-power transmission stability. Summary of the Invention

[0005] This invention provides a low-power Bluetooth headset audio data transmission method and system to solve the problem of low stability in low-power transmission in existing technologies.

[0006] In a first aspect, the present invention provides a low-power Bluetooth headset audio data transmission method, comprising: The audio data stream is acquired and preprocessed to obtain a preliminary compression scheme; Based on the preliminary compression scheme, the audio data stream is segmented and calibrated to obtain an optimized compression block; The channel state information is acquired and the interference intensity is extracted. The transmission stability is calculated based on the interference intensity. The interference disruption degree is calculated based on the transmission stability and a compensation factor is generated. The channel is configured based on the compensation factor to obtain the channel adaptation parameters. The channel adaptation parameters and the optimized compression block are fused to obtain a fusion mapping relationship. Based on the fusion mapping relationship, a simplification verification is performed to obtain a simplified data unit. The simplified data unit is subjected to spectral decomposition to obtain high-frequency components and low-frequency data. The high-frequency components and low-frequency data are then encoded and concatenated to obtain a burden-optimized version of the data. Based on the optimized version data, the feedback signal from the receiving end is extracted and parsed to obtain the sound quality index and signal-to-noise ratio. After validity verification, the sound quality evaluation result is obtained. Based on the sound quality assessment results, defective segments are located, spectral missing regions and temporal distortion intervals are extracted from the defective segments, compensation data packets are generated based on the spectral missing regions and temporal distortion intervals, and the continuity of the compensation data packets is verified to obtain the final audio stream.

[0007] In a second aspect, the present invention provides a low-power Bluetooth headset audio data transmission system, comprising: The initial compression module is used to acquire audio data streams and perform preprocessing to obtain a preliminary compression scheme; The data optimization module is used to segment and calibrate the audio data stream according to the preliminary compression scheme to obtain an optimized compression block; The channel adaptation module is used to acquire channel state information and extract interference intensity, calculate transmission stability based on the interference intensity, calculate interference disruption degree based on the transmission stability and generate compensation factor, and configure the channel based on the compensation factor to obtain channel adaptation parameters. A redundancy simplification module is used to fuse the channel adaptation parameters and the optimized compression block to obtain a fusion mapping relationship, and to perform simplification verification based on the fusion mapping relationship to obtain a simplified data unit. The encoding burden reduction module is used to perform spectral decomposition on the simplified data unit to obtain high-frequency components and low-frequency data, and to encode and concatenate the high-frequency components and low-frequency data to obtain a burden-optimized version of the data. The sound quality evaluation module is used to extract the feedback signal from the receiving end based on the load-optimized version data and parse it to obtain the sound quality index and signal-to-noise ratio. After validity verification, the sound quality evaluation result is obtained. The audio compensation module is used to locate defective segments based on the sound quality evaluation results, extract spectral missing regions and temporal distortion intervals from the defective segments, generate compensation data packets based on the spectral missing regions and temporal distortion intervals, perform continuity verification on the compensation data packets, and obtain the final audio stream.

[0008] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention achieves refined initial compression through frequency domain conversion, spectrum segmentation and amplitude calibration, and then obtains optimized compression blocks through spectrum weighting, dynamic segmentation and iterative calibration. Under the premise of ensuring that key information of speech synthesis and conventional audio is not lost, the amount of transmitted data is significantly reduced, thereby reducing headphone power consumption from the source. (2) Based on the channel state, the present invention calculates the interference intensity, transmission stability and interference damage in real time, dynamically generates compensation factors and configures channel parameters to form a channel adaptation mechanism, improves the transmission stability under complex interference environment, and avoids speech synthesis and audio stream interruption. (3) This invention reduces the processing burden through redundancy simplification and high-frequency lightweight coding, and constructs a closed loop for sound quality evaluation and defect compensation based on receiver feedback to accurately repair missing and distorted parts. Under low power consumption constraints, it ensures the integrity and smoothness of audio and speech synthesis content, achieving optimal synergy between power consumption, stability, and sound quality. (See attached figures) Figure 1 This is a schematic flowchart of a low-power Bluetooth headset audio data transmission method provided in the first embodiment of the present invention; Figure 2 This is a schematic diagram of a low-power Bluetooth headset audio data transmission system provided in the second embodiment of the present invention. Detailed Implementation

[0009] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0010] Reference Figure 1 The first embodiment of the present invention provides a low-power Bluetooth headset audio data transmission method, including the following steps: S1: Acquire audio data stream and perform preprocessing to obtain a preliminary compression scheme.

[0011] S2, according to the preliminary compression scheme, the audio data stream is segmented and calibrated to obtain an optimized compression block.

[0012] S3, acquire channel state information and extract interference intensity, calculate transmission stability based on the interference intensity, calculate interference disruption degree based on the transmission stability and generate compensation factor, configure the channel based on the compensation factor to obtain channel adaptation parameters.

[0013] S4, the channel adaptation parameters and the optimized compression block are fused to obtain a fusion mapping relationship, and a simplification verification is performed according to the fusion mapping relationship to obtain a simplified data unit.

[0014] S5, perform spectral decomposition on the simplified data unit to obtain high-frequency components and low-frequency data, encode and concatenate the high-frequency components and low-frequency data to obtain a burden-optimized version of the data.

[0015] S6. Based on the load-optimized version data, extract the feedback signal from the receiving end and parse it to obtain the sound quality index and signal-to-noise ratio. After validity verification, obtain the sound quality evaluation result.

[0016] S7. Based on the sound quality evaluation results, locate the defective segment, extract the spectral missing region and temporal distortion interval from the defective segment, generate a compensation data packet based on the spectral missing region and temporal distortion interval, perform continuity verification on the compensation data packet, and obtain the final audio stream.

[0017] In step S1, the audio data stream is acquired and preprocessed to obtain a preliminary compression scheme, including: S11, acquire the audio data stream, convert the audio data stream from a time domain signal to a frequency domain signal, and analyze the spectral characteristics of different frequency bands.

[0018] S12, according to the preset spectrum segmentation rules, the low-frequency band of the spectrum features is divided into narrow-band units, and the high-frequency band is divided into wide-band units.

[0019] S13, perform quantization calibration on the narrowband unit and the wideband unit according to the preset amplitude adjustment accuracy to obtain the distribution value.

[0020] S14, calculate the compression ratio based on the distribution values. If the compression ratio is less than a preset energy consumption threshold, extract the configuration parameters for generating the compression ratio to obtain a preliminary compression scheme.

[0021] In step S11, the audio data stream is acquired, the audio data stream is converted from a time-domain signal to a frequency-domain signal, and the spectral characteristics of different frequency bands are obtained by analysis.

[0022] It should be noted that the audio data stream is collected in real time by the audio acquisition module inside the Bluetooth headset. The acquired signals cover regular music playback data, voice call interaction data, and voice synthesis and broadcast data from smart terminals. The frequency range of the acquired signals fully covers the entire audible frequency range of the human ear, from 20Hz to 20000Hz, ensuring that the acquired data is complete and without omissions. The conversion of the time-domain signal to the frequency-domain signal is achieved using digital signal processing with Fast Fourier Transform (FFT). This transformation method can completely decompose the continuously changing time-domain voltage waveform into energy distribution values ​​at different frequency points, thereby obtaining spectral characteristics that can intuitively and comprehensively reflect the energy proportion of each frequency band of the audio. The spectral characteristics can clearly distinguish the distribution of low-frequency vocal components, mid-frequency instrumental components, and high-frequency overtone components, providing a stable and accurate data foundation for subsequent fine-grained processing operations such as spectrum segmentation and amplitude calibration. This avoids problems such as sound quality damage or excessive power consumption in subsequent compression processing due to incomplete extraction of frequency characteristics.

[0023] For example, when Bluetooth headphones play complete audio content including vocals, instrumental accompaniment, and high-frequency overtones, the original time-domain signal appears as a continuous and uninterrupted voltage fluctuation waveform, making it impossible to directly distinguish audio components of different frequencies. After being converted into a frequency-domain signal by a fast Fourier transform, the system can clearly analyze the energy distribution of the low-frequency vocal band from 0Hz to 4000Hz and the energy distribution of the high-frequency overtone band from 4000Hz to 8000Hz, forming a complete spectral feature covering the entire frequency band.

[0024] In step S12, the low-frequency band of the spectrum features is divided into narrow-band units and the high-frequency band is divided into wide-band units according to the preset spectrum segmentation rules.

[0025] It should be noted that the preset spectrum segmentation rules are fully constructed based on the physiological acoustic characteristics of the human ear's auditory masking effect. The construction process involves collecting a large amount of human hearing test data, analyzing the differences in the human ear's perceptual sensitivity and resolution in different frequency bands, and finally forming standardized segmentation rules. The rules clearly stipulate that the low-frequency band from 0Hz to 4000Hz is divided into narrow-band units, with the width of a single narrow-band unit set at 100Hz, and the high-frequency band from 4000Hz to 8000Hz is divided into wide-band units, with the width of a single wide-band unit set at 500Hz. This segmentation rule can fully preserve the core auditory information of the low frequencies that the human ear is sensitive to, while minimizing the amount of data processed in the high-frequency bands that the human ear is less sensitive to, thereby reducing the chip's computing power consumption and data transmission power consumption in Bluetooth headsets, and establishing an optimal balance between sound quality preservation and power consumption control.

[0026] For example, the system performs segmentation processing on a complete audio spectrum feature, dividing the low-frequency part into a finer division according to the standard of one unit per 100Hz, resulting in a larger number of narrow-band units, and dividing the high-frequency part into a broader division according to the standard of one unit per 500Hz, resulting in a smaller number of wide-band units.

[0027] It is worth noting that for the ultra-high frequency band of 8000 Hz to 20000 Hz, since human hearing perception is extremely insensitive and the playback capability of Bluetooth headset speakers is limited, this embodiment adopts a processing method of directly discarding or uniformly merging them into a wideband unit; preferably, the 8000 Hz to 20000 Hz band is divided into one or two wideband units, each unit with a width of not less than 2000 Hz, in order to further reduce the computation and transmission burden. This processing method also follows the preset spectrum segmentation rules constructed by the human hearing masking effect.

[0028] In step S13, quantization calibration is performed on the narrow-band unit and the wide-band unit according to a preset amplitude adjustment accuracy to obtain distribution values.

[0029] It should be noted that the preset amplitude adjustment precision is fully set based on the battery capacity limitations of Bluetooth headsets, the core requirements of low-power transmission, and the comprehensive indicators of human hearing sensitivity. The setting standard is to quantize and calibrate the original audio amplitude data, which is represented by 16 bits, into values ​​represented by 8 bits, thereby significantly reducing the transmission data bit width and transmission power consumption. For frequency band units with energy values ​​below the minimum perceptible threshold of the human ear, the system directly calibrates the amplitude value of the unit to 0, completely eliminating imperceptible redundant noise components. All calibrated values ​​are uniformly referred to as distribution values. Distribution values ​​maximize data volume compression while ensuring that the basic listening experience is not affected, providing accurate data for subsequent compression ratio calculations.

[0030] For example, the system converts the audio amplitude data that originally occupied 16 bits of storage space into 8-bit distribution values ​​according to a preset amplitude adjustment precision. For frequency band units with extremely low energy that cannot be perceived by the human ear, the system directly calibrates the amplitude to 0, completely eliminating invalid noise components and reducing the amount of data transmission.

[0031] In step S14, the compression ratio is calculated based on the distribution value. If the compression ratio is less than a preset energy consumption threshold, the configuration parameters for generating the compression ratio are extracted to obtain a preliminary compression scheme.

[0032] It should be noted that the compression ratio is calculated by dividing the total size of the calibrated data by the total size of the original audio frames. This calculation method directly reflects the degree of data compression; a smaller ratio indicates a higher degree of compression. The preset energy consumption threshold is fully set based on the Bluetooth chip's rated maximum transmission load, the chip's safe operating power consumption range, and the minimum bandwidth requirements for audio transmission. It represents the maximum compression ratio upper limit to ensure the Bluetooth chip operates in a low-power state. When the calculated compression ratio is less than this preset energy consumption threshold, the system determines that the current compression level fully meets the core requirements of low-power transmission. The system then extracts the bandwidth and quantization bits as core configuration parameters to form a preliminary compression scheme that meets the energy consumption limit target.

[0033] For example, the total data size of a certain frame of original audio data is 1000 bytes, and the total data size after calibration is 250 bytes. The calculated compression ratio is 0.25. The maximum compression ratio corresponding to the preset power consumption threshold is 0.3. Since 0.25 is less than 0.3, the system extracts the bandwidth of 100Hz and 500Hz and the quantization bit depth of 8 bits as configuration parameters to generate a preliminary compression scheme.

[0034] In step S2, the audio data stream is segmented and calibrated according to the preliminary compression scheme to obtain an optimized compression block, including: S21, Obtain the preliminary compression scheme, compare the compression ratio of the preliminary compression scheme with a preset compression threshold, and obtain the spectral energy difference.

[0035] S22, assign feature weights according to the spectral energy difference, determine the segmentation granularity according to the feature weights, and perform data segmentation on the audio data stream to obtain data compression blocks.

[0036] S23, perform amplitude calibration on the data compression block to obtain a calibration data block, and perform encoding and encapsulation on the calibration data block to obtain an optimized compression block.

[0037] In step S21, the preliminary compression scheme is obtained, and the compression ratio of the preliminary compression scheme is compared with a preset compression threshold to obtain the spectral energy difference.

[0038] It should be noted that the preset compression threshold is based on the minimum fidelity requirements for audio quality, the acceptable range of distortion for human hearing, and the complete settings of the Bluetooth audio transmission standard. This threshold is used to accurately determine whether the initial compression scheme carries the risk of over-compression leading to sound quality damage, and is a key critical value for distinguishing between reasonable and over-compression. The spectral energy difference is the energy difference between high-energy and low-energy frequency bands in the audio data stream. This difference is calculated by statistically analyzing the peak and valley values ​​of energy in each frequency band unit, and is used to comprehensively reflect the unevenness of the energy distribution of the audio signal. This provides a core basis for subsequent feature weight allocation, ensuring that the weight allocation closely matches the actual energy distribution of the audio.

[0039] For example, the preset compression threshold is set to 0.2, and the compression ratio of the initial compression scheme is 0.15, which is lower than the preset compression threshold. The system determines that the current audio data stream is at risk of over-compression. The system further calculates the energy difference between the high-energy frequency band of human voice and the low-energy frequency band of background noise to obtain the accurate spectral energy difference.

[0040] It should be noted that regardless of the comparison result between the compression ratio and the preset compression threshold, the spectral energy difference must be calculated. If the compression ratio is lower than the preset compression threshold, it indicates that there is a risk of over-compression. The difference between the peak and valley values ​​of energy in each frequency band unit is calculated. If the compression ratio is not lower than the preset compression threshold, it indicates that the current compression scheme is within a reasonable range. The peak and valley values ​​of energy are calculated in the same way to obtain the spectral energy difference, which is used for the subsequent allocation of feature weights.

[0041] In step S22, feature weights are assigned according to the spectral energy differences, the segmentation granularity is determined according to the feature weights, and data segmentation is performed on the audio data stream to obtain data compression blocks.

[0042] It should be noted that the feature weight allocation rule is entirely based on the magnitude of the spectral energy difference. For frequency bands with high energy values ​​and containing core audio information, the system assigns higher weight values; for frequency bands with low energy values ​​and containing only background noise, the system assigns lower weight values. This allocation method highlights the processing priority of core audio information. The data segmentation granularity is directly related to the feature weights. High-weight, high-energy frequency bands use finer segmentation durations, while low-weight, low-energy frequency bands use coarser segmentation durations. The segmentation duration is set according to the importance of the audio information. After segmentation, all audio segments are combined in chronological order to form a data compression block. The data compression block retains core information while eliminating redundant processing steps.

[0043] For example, for high-energy audio segments containing transient high notes of human voices, the system assigns a high weight of 0.8 and uses fine-grained segmentation of 5ms. For low-energy segments with stable background noise, the system assigns a low weight of 0.2 and uses coarse-grained segmentation of 20ms. All segmented audio segments are combined to form a data compression block.

[0044] In step S23, amplitude calibration is performed on the data compression block to obtain a calibration data block, and encoding and encapsulation are performed on the calibration data block to obtain an optimized compression block.

[0045] It should be noted that the number of iterations for amplitude calibration is determined based on the degree of dispersion of the amplitude data within the data compression block. The degree of dispersion is obtained by calculating the variance and standard deviation of the amplitude data. The higher the degree of dispersion, the more calibrations are performed, and the lower the degree of dispersion, the fewer calibrations are performed. The core objective of calibration is to ensure that the amplitude values ​​are evenly distributed within the quantization range, eliminating sound quality distortion caused by abrupt amplitude changes. After calibration, Bluetooth audio standard format encoding and encapsulation are performed. The encoding and encapsulation process removes padding bits, redundant bits, and invalid parity bits from the data, further compressing the data volume, ultimately resulting in an optimized compression block adapted for low-power transmission scenarios.

[0046] For example, for non-uniform segments with large amplitude data variance, the system performs 5 iterations of calibration; for segments with concentrated amplitude distribution, the system performs 1 iteration of calibration. After calibration, the data is encoded and encapsulated, redundant bits are removed, and an optimized compressed block is obtained.

[0047] In step S3, channel state information is acquired and interference intensity is extracted. Transmission stability is calculated based on the interference intensity. Interference disruption is calculated based on the transmission stability, and a compensation factor is generated. Channel adaptation parameters are configured based on the compensation factor, including: S31, acquire channel state information including received signal strength and signal-to-noise ratio, calculate the signal-to-noise ratio change, and extract the quantized interference intensity based on the signal-to-noise ratio change.

[0048] S32, calculate the unit bit error correction energy consumption as an evaluation dimension based on the channel state information and the coding complexity of the optimized compression block.

[0049] S33, calculate the transmission stability of the optimized compressed block based on the interference intensity and the evaluation dimension, and calculate the interference disruption degree based on the transmission stability and the interference intensity.

[0050] S34. Based on the interference damage degree and the preset compensation mapping relationship, perform numerical mapping on the interference damage degree to generate a gain compensation factor, and use the gain compensation factor to configure the channel parameters to obtain the channel adaptation parameters.

[0051] In step S31, channel state information including received signal strength and signal-to-noise ratio is obtained, the signal-to-noise ratio change is calculated, and the quantized interference intensity is extracted based on the signal-to-noise ratio change.

[0052] It should be noted that channel state information is obtained in real time through the Bluetooth headset's physical layer interface. The core information includes two key transmission indicators: received signal strength (RSS) and signal-to-noise ratio (SNR). Both indicators are core parameters reflecting the quality of wireless channel transmission, and none are missing or omitted. The SNR change is the difference between the currently monitored SNR and the standard SNR under stable transmission conditions. This difference directly reflects the degree of abrupt changes in channel interference. Interference intensity is obtained through a standardized mapping of the SNR change, with a strictly limited value range of 0 to 1. A higher value indicates a higher level of interference in the current wireless channel and poorer transmission stability. This quantification method can standardize the interference assessment criteria under different channel conditions.

[0053] For example, under stable transmission conditions, the channel signal-to-noise ratio is 30dB. Real-time monitoring shows that the signal-to-noise ratio drops to 15dB. The change in signal-to-noise ratio is 15dB. The system normalizes this change and maps it to an interference intensity of 0.6, representing that the current channel is at a medium to high level of interference.

[0054] It should be further explained that, under stable transmission conditions, the standard signal-to-noise ratio is the arithmetic mean of the signal-to-noise ratio values ​​collected continuously over a preset period of time, such as 1 second, after the Bluetooth headset and the audio source device establish a connection in an interference-free or low-interference environment. This standard signal-to-noise ratio can be dynamically updated during device pairing or idle periods, or a typical value provided by the chip manufacturer can be used as a fixed benchmark. In this embodiment, the average value of the first 100 signal-to-noise ratio samples at the initial connection is preferably used as the standard signal-to-noise ratio, and it is recalibrated every 10 seconds.

[0055] In step S32, the energy consumption per unit bit for error correction is calculated as an evaluation dimension based on the channel state information and the coding complexity of the optimized compression block.

[0056] It should be noted that the energy consumption per unit bit for error correction is calculated by dividing the total power consumption consumed during channel error correction by the total number of bits processed. This energy consumption value accurately reflects the energy cost required to transmit each bit of data for error correction. This energy consumption value increases in tandem with the increase in channel interference intensity and the increase in the coding complexity of the optimized compression block. The higher the coding complexity and the greater the interference intensity, the higher the energy consumption per unit bit for error correction. The system uses this energy consumption value as a core dimension for evaluating transmission stability, comprehensively reflecting the additional energy consumption of data transmission under interference conditions, and providing an objective basis for transmission stability calculation.

[0057] It should be noted that the coding complexity of the optimized compressed block is determined by comprehensively calculating the number of non-zero coefficients in the compressed block, the average code length of the entropy coding, and the sparsity of the transform domain coefficients. Specifically, the higher the proportion of non-zero coefficients, the longer the average code length, and the denser the coefficients, the higher the coding complexity. The coding complexity value is obtained by weighted summing of these three indicators and normalizing them to the range of 0 to 1.

[0058] It should be noted that the total power consumption during the channel error correction process is read from the power consumption monitoring register built into the Bluetooth chip. This register can accumulate the current consumption of the error correction processing circuit by multiplying it by the voltage. The error correction processing circuit includes FEC decoding, CRC verification, and retransmission control logic. If there is no hardware support, it can be estimated by multiplying the computational load of the error correction algorithm by the empirical power consumption value per unit of computation, such as the number of operations of the error correction code's check matrix.

[0059] For example, when the interference intensity is 0.6 and the optimized compression block coding complexity is at a high level, the system calculates that the energy consumption per unit bit for error correction is 1.5 times that of the standard transmission energy consumption. The system uses this value as the core evaluation dimension of transmission stability. The standard transmission energy consumption is the average energy consumption required to transmit a unit bit of data when there is no interference and the coding complexity is at its minimum, that is, when all compression blocks have zero coefficients. This value is measured by the Bluetooth headset during factory calibration or initial connection.

[0060] In step S33, the transmission stability of the optimized compressed block is calculated based on the interference intensity and the evaluation dimension, and the interference disruption degree is calculated based on the transmission stability and the interference intensity.

[0061] It should be noted that transmission stability is used to quantify and optimize the reliable transmission probability of compressed blocks under the current channel. It is calculated by subtracting the product of interference intensity and unit bit error correction energy consumption from a value of 1. When the product is greater than or equal to 1, the transmission stability value is 0. The transmission stability value ranges from 0 to 1. The closer the value is to 1, the better the transmission stability; the lower the value, the more prone the transmission is to problems such as stuttering, packet loss, and distortion. Before calculation, the unit bit error correction energy consumption needs to be normalized, mapping its value range to between 0 and 1. Specifically, it is calculated using actual units... Bit error correction energy consumption divided by the maximum allowable error correction energy consumption, which is determined by the maximum allowable power consumption of the Bluetooth chip's error correction processing circuit, is normalized and recorded as normalized error correction energy consumption. Interference destructiveness is used to quantify the potential destructive power of environmental interference on the optimized compressed block data structure. It is calculated by dividing the interference intensity by the sum of the transmission stability and a very small positive number, for example, 0.001, to avoid division by zero errors. This value can intuitively reflect the destructive effect of interference on data transmission. The larger the value, the more significant the destructive effect, providing an accurate basis for the subsequent generation of compensation factors.

[0062] In step S34, the interference damage degree is numerically mapped according to the interference damage degree and the preset compensation mapping relationship to generate a gain compensation factor. The channel parameters are then configured using the gain compensation factor to obtain the channel adaptation parameters.

[0063] It should be noted that the preset compensation mapping relationship is fully constructed based on a large amount of historical channel interference monitoring data and the correspondence between compensation parameters. The construction process involves statistically analyzing the optimal compensation parameters under different interference levels, establishing a one-to-one correspondence between interference levels and compensation parameters, and forming a standardized numerical mapping relationship. The gain compensation factor is obtained by accurately matching the interference level in the mapping relationship and is used to adjust the RF front-end transmit power and baseband modulation and demodulation parameters. During the channel parameter configuration process, the transmission mode is dynamically adjusted according to the gain compensation factor. The set of parameters obtained after adjustment is called the channel adaptation parameter. This parameter can adapt to the current channel state and balance transmission stability and device power consumption.

[0064] Specifically, the preset compensation mapping relationship is achieved by simulating interference signals of different intensities in a laboratory environment using a Bluetooth tester, with the interference severity ranging from 0.1 to 2.0 in a step of 0.1. For each interference severity value, the gain compensation factor is tested from 0.5 to 2.5 in a step of 0.1, and the bit error rate and packet loss rate at the receiver are recorded for each combination. The smallest gain compensation factor that makes the bit error rate less than one-thousandth and the packet loss rate less than five percent is selected as the optimal compensation parameter for that interference severity. The correspondence between the interference severity and the optimal compensation parameter is made into a lookup table and stored in the transmitter's memory. When needed, the gain compensation factor can be obtained directly by looking up the table.

[0065] For example, for an interference severity of 0.75, the system generates a gain compensation factor of 1.2 through a preset compensation mapping relationship. The system increases the radio frequency transmission power by 2dB or switches the modulation mode from 2Mbps to 1Mbps to obtain the final channel adaptation parameters.

[0066] In step S4, the channel adaptation parameters and the optimized compression block are fused to obtain a fusion mapping relationship. A simplification verification is performed based on the fusion mapping relationship to obtain a simplified data unit, including: S41, establish the fusion mapping relationship between the channel adaptation parameters and the optimized compression block, and locate the redundant vector in the optimized compression block under the current channel according to the fusion mapping relationship.

[0067] S42, extract the redundancy removal mask from the redundancy vector, and use the redundancy removal mask to perform mask simplification on the optimized compressed block to obtain a simplified data packet.

[0068] S43, calculate the hash check value of the simplified data packet, compare the hash check value with the preset integrity benchmark code, and obtain a qualified simplified data packet.

[0069] S44, if the bit length of the qualified simplified data packet is less than or equal to the preset maximum payload length, then the qualified simplified data packet is a simplified data unit.

[0070] In step S41, a fusion mapping relationship between the channel adaptation parameters and the optimized compression block is established, and the redundancy vector in the optimized compression block under the current channel is located according to the fusion mapping relationship.

[0071] It should be noted that the fusion mapping relationship is fully established based on the correspondence rules between channel state and audio data redundancy. The rules clearly define the audio data components that can be discarded under different channel states. When the channel adaptation parameters indicate a harsh transmission state of high interference and low bandwidth, the feature data corresponding to the ultra-high frequency overtone components and stereo side channel detail components are non-core redundant information. The feature set corresponding to this part of the information is the redundancy vector. The process of locating the redundant vector accurately identifies the data components that can be discarded without affecting the basic listening experience, avoiding the accidental removal of core audio information and ensuring the rationality and security of the simplification process.

[0072] For example, when the channel adaptation parameter indicates a high-interference, low-bandwidth mode, the system will locate the coefficients corresponding to the ultra-high frequency overtones above 8000Hz and the stereo side channel details in the optimized compressed block as redundant vectors.

[0073] In step S42, a redundancy removal mask is extracted from the redundancy vector, and the optimized compressed block is masked and simplified using the redundancy removal mask to obtain a simplified data packet.

[0074] It should be noted that the redundancy removal mask is a bit-level precise filtering identifier. This mask is generated using the bit position information of the redundant vector, accurately marking each bit position corresponding to the redundant vector without deviation or omission. The mask simplification process is implemented through digital logic operations. The process directly strips away the redundant bits marked by the mask, completely preserving the core audio data bits. This processing method minimizes the data volume and reduces the transmission load to the maximum extent without destroying the main audio frame structure. The data packet obtained after simplification is called a simplified data packet.

[0075] For example, the redundant vector corresponds to the padding bits and minor quantization precision bits at the end of the data block. The system generates a mask to mark these bits as invalid, and removes the redundant bits through logical operations to obtain a smaller, simplified data packet.

[0076] In step S43, the hash check value of the simplified data packet is calculated, and the hash check value is compared with the preset integrity benchmark code to obtain a qualified simplified data packet.

[0077] It should be noted that the preset integrity benchmark code is fully set based on the core frame header structure of the audio data, the check bit rules, and the correspondence of standard data. It is the sole criterion for determining whether the simplification process has damaged the core data structure. Specifically, before the audio data transmission begins, the original optimized compressed block without any simplification process is used to calculate its hash check value using the same hash algorithm. This value is then pre-stored in the sender's memory as the preset integrity benchmark code. In this embodiment, the hash check value calculated from the current simplified data packet is compared with the pre-stored benchmark code. If they are completely identical, it is determined that the simplification process has not damaged the core data structure. The hash check value is calculated from the core data of the simplified data packet and is used to reflect the overall data characteristics of the data packet. If the hash check value is completely identical to the preset integrity benchmark code, it indicates that the simplification process has not damaged the core data syntax structure, and the data packet can be decoded and played normally. This data packet is a qualified simplified data packet; otherwise, it is an invalid data packet.

[0078] For example, the system calculates the hash check value of the simplified data packet, which is completely consistent with the preset integrity benchmark code. If the data packet is found to be undamaged, it is marked as a qualified simplified data packet.

[0079] In step S44, if the bit length of the qualified simplified data packet is less than or equal to the preset maximum payload length, then the qualified simplified data packet is a simplified data unit.

[0080] It should be noted that the preset maximum payload length is based on the maximum capacity per time slot specified by the Bluetooth transmission protocol and the transmission capacity limited by the channel adaptation parameters. It is the maximum data packet length that the current channel can smoothly carry. When the data packet length is less than or equal to this value, non-blocking and packet loss-free transmission is possible. After the bit length of a qualified simplified data packet is compared with the preset maximum payload length and passes the comparison, the data packet is a simplified data unit that meets all transmission conditions and can directly enter the subsequent spectrum decomposition and encoding processing stage.

[0081] For example, the maximum payload length is preset to 27 bytes, and the length of a qualified simplified data packet is 25 bytes. Since the length is less than the preset value, the system determines the data packet as a simplified data unit.

[0082] In step S5, spectral decomposition is performed on the simplified data unit to obtain high-frequency components and low-frequency data. The high-frequency components and low-frequency data are then encoded and concatenated to obtain the burden-optimized version data, including: S51, compare the bit occupancy of the simplified data unit with the preset transmission limit. If it exceeds the preset transmission limit, perform spectral decomposition on the simplified data unit to obtain high-frequency components and low-frequency data.

[0083] S52, obtain the running status information fed back by the receiving end for the simplified data unit, determine the buffer status and latency tolerance based on the running status information, and set the lightweight coding rules according to the latency tolerance.

[0084] S53, the high-frequency components are compressed and encoded using the lightweight encoding rules to generate a high-frequency encoded stream and calculate the encoding computation overhead.

[0085] S54, compare the encoding calculation overhead with a preset processing burden threshold. If the encoding calculation overhead is less than the preset processing burden threshold, concatenate the high-frequency encoded stream with the low-frequency data to obtain a burden-optimized version of the data.

[0086] In step S51, the bit occupancy of the simplified data unit is compared with the preset transmission limit. If it exceeds the preset transmission limit, the simplified data unit is subjected to spectral decomposition to obtain high-frequency components and low-frequency data.

[0087] It should be noted that the preset transmission limit is based on the Bluetooth single-slot payload standard, and is fully set according to the current channel transmission capacity and low-power transmission requirements. It is the maximum number of data bits that a single time slot can carry. Exceeding this limit will result in transmission congestion and increased latency. Spectrum decomposition is achieved using a digital filter bank signal processing method. The low-frequency data obtained after decomposition carries the main audio energy and fundamental tone structure, which is the core auditory information and is preserved intact without compression. The high-frequency components carry the audio timbre details and overtone information, and are the main optimization targets for subsequent lightweight coding.

[0088] For example, the preset transmission limit is 40 bytes, and the simplified data unit bit occupancy is 45 bytes. If the limit is exceeded, the system obtains low-frequency reference data and high-frequency component characteristics by decomposing through a digital filter bank.

[0089] In step S52, the operating status information fed back by the receiving end for the simplified data unit is obtained, the buffer status and latency tolerance are determined according to the operating status information, and the lightweight coding rules are set according to the latency tolerance.

[0090] It should be noted that the running status information is the jitter buffer margin data fed back by the receiving end in real time. The buffer margin data directly reflects the receiving end's data buffering capacity and latency tolerance. The smaller the margin, the lower the receiving end's tolerance for transmission latency and the stricter the requirements for encoding processing latency. The lightweight encoding rules are set entirely based on the magnitude of the latency tolerance. Under low latency tolerance, a low-complexity, low-latency, and low-computing-power encoding strategy must be adopted to avoid audio stuttering and dropouts caused by time-consuming encoding processing. The rule settings are closely aligned with the actual running status of the receiving end.

[0091] For example, the receiver buffer has only 10ms remaining, with extremely low latency tolerance. The system sets adaptive differential quantization coding as a lightweight coding rule to ensure extremely low processing latency.

[0092] In step S53, the high-frequency components are compressed and encoded using the lightweight encoding rule to generate a high-frequency encoded stream and calculate the encoding computation overhead.

[0093] It should be noted that the encoding computation overhead is the total number of clock cycles consumed by the Bluetooth chip processor during the encoding process. This value can accurately quantify the degree of chip computing power consumption and instantaneous power consumption level of the encoding operation, and is a core indicator for judging whether the encoding process meets the low power requirements. Lightweight encoding processing significantly reduces the data volume of high-frequency components. The resulting compressed data stream is called a high-frequency encoded stream. The high-frequency encoded stream reduces the data volume by more than 50% while preserving the basic tonal details.

[0094] For example, the system uses lightweight coding rules to encode high-frequency components and generate a high-frequency coded stream. The coding computation overhead is 300 operation clock cycles.

[0095] In step S54, the encoding computation overhead is compared with a preset processing burden threshold. If the encoding computation overhead is less than the preset processing burden threshold, the high-frequency encoded stream and the low-frequency data are concatenated to obtain a burden-optimized version of the data.

[0096] It should be noted that the preset processing load threshold is fully set based on the Bluetooth chip's maximum safe computing capacity, the chip's low-power operation standards, and device heat dissipation limitations. It represents the maximum number of clock cycles the chip can safely operate on. When the encoding computation overhead is less than this threshold, the encoding operation will not cause a surge in chip power consumption, overload operation, or abnormal heat generation. The splicing operation combines the high-frequency encoded stream and low-frequency data in chronological order. The resulting data stream is called the load-optimized version, which simultaneously meets the requirements of low transmission load and low computational burden.

[0097] For example, the preset processing burden threshold is 500 operation clock cycles, and the encoding computation overhead is 300. Since it is less than the preset threshold, the system concatenates the high-frequency encoded stream with the low-frequency data, reducing the total length to 38 bytes, and obtains the burden-optimized version of the data.

[0098] In step S6, based on the load-optimized version data, the feedback signal from the receiving end is extracted and parsed to obtain the sound quality indicators and signal-to-noise ratio. After validity verification, the sound quality evaluation result is obtained, including: S61, the load-optimized version data is sent to the receiving end, and the feedback signal is extracted. The audio quality indicators of audio spectrum integrity and auditory clarity, as well as the channel signal-to-noise ratio value, are extracted from the feedback signal.

[0099] S62, determine the logical correspondence between the sound quality index and the signal-to-noise ratio according to the preset audio transmission quality verification rules, generate a valid identifier according to the logical correspondence, and remove invalid data with logical abnormalities according to the valid identifier.

[0100] S63, extract the indicator data corresponding to the valid identifier from the feedback signal, construct a difference measurement matrix with the indicator data and the preset sound quality threshold, and calculate the sound quality evaluation result by weighting the difference measurement matrix.

[0101] In step S61, the burden-optimized version data is sent to the receiving end, and the feedback signal is extracted. The audio quality indicators of audio spectrum integrity and auditory clarity, as well as the channel signal-to-noise ratio, are extracted from the feedback signal.

[0102] It should be noted that the feedback signal is a standardized evaluation data packet transmitted back in real time by the receiving end after completing audio decoding and playback. This data packet contains two core parameters: audio reproduction quality and channel transmission quality, with no omissions or gaps. Audio spectrum integrity reflects the completeness of frequency band coverage after audio decoding; a higher value indicates fewer missing frequency bands. Auditory clarity reflects the degree of audio reproduction distortion; a higher value indicates less distortion. These two indicators are core parameters for evaluating audio quality. The channel signal-to-noise ratio objectively reflects the physical transmission quality of the current wireless link and is a key indicator for judging transmission stability.

[0103] For example, the system analyzes the feedback signal and extracts an audio spectrum integrity level of 2, an auditory clarity level of 2, and a channel signal-to-noise ratio of 25dB.

[0104] In step S62, the logical correspondence between the sound quality index and the signal-to-noise ratio is determined according to the preset audio transmission quality verification rules, a valid identifier is generated according to the logical correspondence, and invalid data with logical abnormalities is removed according to the valid identifier.

[0105] It should be noted that the preset audio transmission quality verification rules are based on the physical laws of audio transmission, statistical results of a large amount of measured transmission data, and a complete demodulation logic. The rules clearly stipulate that when the signal-to-noise ratio is lower than the basic demodulation threshold, the sound quality index should be at a lower level. If a logical paradox occurs where there is a low signal-to-noise ratio but high sound quality, it is judged as invalid dirty data with logical abnormalities. Valid identifiers are used to mark logically compliant valid data. Removing invalid data can avoid the interference of abnormal data with subsequent sound quality evaluation results, ensuring the accuracy and objectivity of the evaluation results.

[0106] Specifically, the preset audio transmission quality verification rule is based on the collection of 10,000 sets of measured data under different channel conditions. Each set of data includes the signal-to-noise ratio (SNR) value at the receiving end, as well as the corresponding audio spectrum integrity and perceived clarity. Statistical analysis revealed that when the SNR is below 8 dB, the spectrum integrity never exceeds level 3 (out of 5), and the perceived clarity never exceeds level 2. Based on this, a logical verification rule is established: if the SNR in the feedback signal is below 8 dB but any of the sound quality indicators exceeds the above upper limit, it is judged as a logical anomaly. Similarly, when the SNR is above 20 dB, the sound quality indicator should not be lower than level 3; otherwise, it is also judged as an anomaly. This rule is pre-stored at the transmitting end for real-time filtering of invalid feedback data.

[0107] In step S63, the indicator data corresponding to the valid identifier is extracted from the feedback signal, the indicator data and the preset sound quality threshold are used to construct a difference measurement matrix, and the sound quality evaluation result is obtained by weighted calculation of the difference measurement matrix.

[0108] It should be noted that the preset sound quality threshold is based on high-fidelity audio standards, the requirements for a superior auditory experience, and the complete setting of Bluetooth audio transmission sound quality levels. It serves as the critical value for determining whether audio sound quality meets the standards. The difference measurement matrix is ​​used to quantify the difference between the actual indicators and the preset threshold in multiple dimensions. The matrix weighting calculation process assigns a higher weight coefficient to sound quality indicators and a lower weight coefficient to signal-to-noise ratio indicators. The value obtained after weighted calculation is the sound quality evaluation result, which can accurately locate the position and extent of audio sound quality defects.

[0109] For example, the preset sound quality threshold is a signal-to-noise ratio of 20dB and a sound quality classification of 4 levels. The actual indicators are a signal-to-noise ratio of 25dB and a sound quality classification of 2 levels. The sound quality evaluation result is obtained after constructing a difference metric matrix and weighting the calculation.

[0110] In step S7, defective segments are located based on the sound quality evaluation results. Spectral missing regions and temporal distortion intervals are extracted from the defective segments. Compensation data packets are generated based on the spectral missing regions and temporal distortion intervals. Continuity checks are performed on the compensation data packets to obtain the final audio stream, including: S71, based on the sound quality evaluation results, locate the defective segments with missing high-frequency information and waveform distortion, and extract the spectrum missing region and time-domain distortion interval from the defective segments.

[0111] S72, generate a frequency domain gain value based on the missing spectrum region, generate a waveform filling sequence based on the time domain distortion interval, and combine the frequency domain gain value and the waveform filling sequence to form a compensation data packet.

[0112] S73, the compensation data packet is embedded into the defect location of the audio data stream to obtain the compensated audio stream.

[0113] S74, calculate the continuity feature value of the compensated audio stream. If the difference between the continuity feature value and the feature value of the preset feature value template is less than the preset integrity threshold, the final audio stream is obtained.

[0114] In step S71, based on the sound quality evaluation results, defective segments with missing high-frequency information and waveform distortion are located, and the spectral missing region and temporal distortion interval are extracted from the defective segments.

[0115] It should be noted that the defective segments are audio passages that fail to meet the preset audio quality threshold in the audio quality assessment results. These defects are mainly characterized by excessive loss of high-frequency information and distortion of the time-domain waveform, representing the core defect locations affecting the audio listening experience. The missing frequency spectrum region is the high-frequency band between 8000Hz and 12000Hz, which is the core frequency band for audio overtone details. The time-domain distortion range is the time interval of waveform distortion caused by packet loss decoding errors. These two characteristics provide accurate positioning basis for audio compensation, without deviation or omission.

[0116] For example, based on the sound quality assessment results, the system accurately locates the audio from frame 120 to frame 150 as defective segments, extracts the 8000Hz to 12000Hz spectrum missing region, and the 5ms duration as the time domain distortion interval.

[0117] In step S72, a frequency domain gain value is generated based on the missing spectral region, a waveform filling sequence is generated based on the time domain distortion interval, and the frequency domain gain value and the waveform filling sequence are combined to form a compensation data packet.

[0118] It should be noted that the frequency domain gain value is used to increase the energy amplitude of the missing spectral region and reshape the energy distribution of the missing frequency band. The value is determined by the degree of spectral loss; the more severe the loss, the higher the gain value. The waveform filling sequence extracts waveform features from adjacent normal frames before and after the defective segment and uses linear interpolation logic to generate a smooth time-domain waveform, ensuring a natural and abrupt waveform transition. After synthesizing the frequency domain gain value and the time-domain waveform filling sequence according to the data structure specifications, a dedicated compensation data package is obtained, which can accurately repair audio defects.

[0119] For example, for the spectrum missing region from 8000Hz to 12000Hz, the system generates a frequency domain gain value of 2.5dB, and for the time domain distortion interval of 5ms, it generates a smooth filling sequence by interpolating the waveforms of adjacent normal frames to synthesize a compensation data packet.

[0120] In step S73, the compensation data packet is embedded into the defect location of the audio data stream to obtain the compensated audio stream.

[0121] It should be noted that the embedding operation corresponds exactly to the timeline of the defective segment. The embedding process uses a combination of data overwriting and data replacement to precisely embed the compensation data packet into the defective location, replacing the original distorted data. The embedding operation does not change the total duration or overall structure of the audio stream; it only repairs the defective portion. The audio stream obtained after embedding is called the compensated audio stream. This audio stream has completed defect repair, and the sound quality is significantly improved.

[0122] For example, the system precisely embeds the compensation data packet into the defect location between frames 120 and 150, replacing the original distorted audio data to obtain the compensated audio stream.

[0123] In step S74, the continuity feature value of the compensated audio stream is calculated. If the difference between the continuity feature value and the feature value of the preset feature value template is less than the preset integrity threshold, the final audio stream is obtained.

[0124] It should be noted that the continuity feature value reflects the smoothness of energy transition and phase change between frames of the compensated audio stream. A higher value indicates a more natural audio transition, without stutters or abrupt changes. The preset integrity threshold is set based on the human auditory smoothness standard, the requirements for smooth audio playback, and the Bluetooth audio transmission standard. It is the critical value for judging whether the compensated audio meets playback requirements. When the difference between the continuity feature value of the compensated audio stream and the preset feature value template is less than this threshold, it indicates that the compensation operation meets auditory requirements, and this audio stream is the final audio stream. The preset feature value template is set based on the Bluetooth audio transmission standard, the human auditory threshold for stutter-free perception, the requirements for smooth playback in low-power transmission scenarios, and the statistical results of the continuity features of a large number of normal, distortion-free audio samples. The system collects thousands of qualified audio data streams, calculates their inter-frame energy changes and phase smoothness, and takes the average of the continuity feature values ​​of all qualified samples as the preset feature value template, which serves as a standard reference value for whether the audio continuity meets the standard.

[0125] For example, the preset feature value template is 90, the actual continuity feature value of the compensated audio stream is 87, the difference between the two is 3, the preset integrity threshold is 5, 3 is less than 5, the system determines that the compensated audio meets the auditory requirements, and the final audio stream is obtained.

[0126] Reference Figure 2 The second embodiment of the present invention provides a low-power Bluetooth headset audio data transmission system, comprising: The initial compression module is used to acquire audio data streams and perform preprocessing to obtain a preliminary compression scheme; The data optimization module is used to segment and calibrate the audio data stream according to the preliminary compression scheme to obtain an optimized compression block; The channel adaptation module is used to acquire channel state information and extract interference intensity, calculate transmission stability based on the interference intensity, calculate interference disruption degree based on the transmission stability and generate compensation factor, and configure the channel based on the compensation factor to obtain channel adaptation parameters. A redundancy simplification module is used to fuse the channel adaptation parameters and the optimized compression block to obtain a fusion mapping relationship, and to perform simplification verification based on the fusion mapping relationship to obtain a simplified data unit. The encoding burden reduction module is used to perform spectral decomposition on the simplified data unit to obtain high-frequency components and low-frequency data, and to encode and concatenate the high-frequency components and low-frequency data to obtain a burden-optimized version of the data. The sound quality evaluation module is used to extract the feedback signal from the receiving end based on the load-optimized version data and parse it to obtain the sound quality index and signal-to-noise ratio. After validity verification, the sound quality evaluation result is obtained. The audio compensation module is used to locate defective segments based on the sound quality evaluation results, extract spectral missing regions and temporal distortion intervals from the defective segments, generate compensation data packets based on the spectral missing regions and temporal distortion intervals, perform continuity verification on the compensation data packets, and obtain the final audio stream.

[0127] It should be noted that the low-power Bluetooth headset audio data transmission system provided in this embodiment of the invention is used to execute all the process steps of the low-power Bluetooth headset audio data transmission method of the above embodiment. Each unit of the system corresponds one-to-one with each step of the method. The working principle, execution logic and beneficial effects are completely matched without omission or deviation. Therefore, it will not be described again.

[0128] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0129] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for transmitting audio data in a low-power Bluetooth headset, characterized in that, include: The audio data stream is acquired and preprocessed to obtain a preliminary compression scheme; Based on the preliminary compression scheme, the audio data stream is segmented and calibrated to obtain an optimized compression block; The channel state information is acquired and the interference intensity is extracted. The transmission stability is calculated based on the interference intensity. The interference disruption degree is calculated based on the transmission stability and a compensation factor is generated. The channel is configured based on the compensation factor to obtain the channel adaptation parameters. The channel adaptation parameters and the optimized compression block are fused to obtain a fusion mapping relationship. Based on the fusion mapping relationship, a simplification verification is performed to obtain a simplified data unit. The simplified data unit is subjected to spectral decomposition to obtain high-frequency components and low-frequency data. The high-frequency components and low-frequency data are then encoded and concatenated to obtain a burden-optimized version of the data. Based on the optimized version data, the feedback signal from the receiving end is extracted and parsed to obtain the sound quality index and signal-to-noise ratio. After validity verification, the sound quality evaluation result is obtained. Based on the sound quality assessment results, defective segments are located, spectral missing regions and temporal distortion intervals are extracted from the defective segments, compensation data packets are generated based on the spectral missing regions and temporal distortion intervals, and the continuity of the compensation data packets is verified to obtain the final audio stream.

2. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The process of acquiring audio data streams and preprocessing them to obtain a preliminary compression scheme includes: The audio data stream is acquired, converted from a time-domain signal to a frequency-domain signal, and the spectral characteristics of different frequency bands are obtained by analysis. According to the preset spectrum segmentation rules, the low-frequency band of the spectrum features is divided into narrow-band units, and the high-frequency band is divided into wide-band units; Quantization calibration is performed on the narrowband unit and the wideband unit according to a preset amplitude adjustment precision to obtain distribution values; The compression ratio is calculated based on the distribution values. If the compression ratio is less than a preset energy consumption threshold, the configuration parameters for generating the compression ratio are extracted to obtain a preliminary compression scheme.

3. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The step of segmenting and calibrating the audio data stream according to the preliminary compression scheme to obtain optimized compression blocks includes: The preliminary compression scheme is obtained, and the compression ratio of the preliminary compression scheme is compared with a preset compression threshold to obtain the spectral energy difference; Feature weights are assigned based on the spectral energy differences, the segmentation granularity is determined based on the feature weights, and data segmentation is performed on the audio data stream to obtain data compression blocks; Amplitude calibration is performed on the data compression block to obtain a calibration data block, and encoding and encapsulation are performed on the calibration data block to obtain an optimized compression block.

4. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The steps of acquiring channel state information and extracting interference intensity, calculating transmission stability based on the interference intensity, calculating interference disruption degree based on the transmission stability and generating a compensation factor, and configuring the channel based on the compensation factor to obtain channel adaptation parameters include: Obtain channel state information including received signal strength and signal-to-noise ratio, calculate the signal-to-noise ratio change, and extract the quantized interference intensity based on the signal-to-noise ratio change; The energy consumption per bit for error correction is calculated as an evaluation dimension based on the channel state information and the coding complexity of the optimized compression block. The transmission stability of the optimized compressed block is calculated based on the interference intensity and the evaluation dimension, and the interference disruption degree is calculated based on the transmission stability and the interference intensity. According to the preset compensation mapping relationship, the interference damage degree is numerically mapped to generate a gain compensation factor. The channel parameters are then configured using the gain compensation factor to obtain the channel adaptation parameters.

5. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The channel adaptation parameters and the optimized compression block are fused to obtain a fusion mapping relationship. A simplification verification is performed based on the fusion mapping relationship to obtain a simplified data unit, including: Establish a fusion mapping relationship between the channel adaptation parameters and the optimized compression block, and locate the redundant vector in the optimized compression block under the current channel based on the fusion mapping relationship; The redundancy removal mask is extracted from the redundancy vector, and the redundancy removal mask is used to perform mask simplification on the optimized compression block to obtain a simplified data packet; Calculate the hash check value of the simplified data packet, compare the hash check value with the preset integrity benchmark code, and obtain a qualified simplified data packet; If the bit length of the qualified simplified data packet is less than or equal to the preset maximum payload length, then the qualified simplified data packet is a simplified data unit.

6. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The process involves performing spectral decomposition on the simplified data unit to obtain high-frequency components and low-frequency data, and then encoding and concatenating the high-frequency components and low-frequency data to obtain a burden-optimized version of the data, including: The bit occupancy of the simplified data unit is compared with the preset transmission limit. If it exceeds the preset transmission limit, the simplified data unit is subjected to spectral decomposition to obtain high-frequency components and low-frequency data. Obtain the running status information fed back by the receiving end for the simplified data unit, determine the buffer status and latency tolerance based on the running status information, and set the lightweight coding rules according to the latency tolerance; The high-frequency components are compressed and encoded using the lightweight encoding rules to generate a high-frequency encoded stream and calculate the encoding computation overhead. The encoding computation cost is compared with a preset processing burden threshold. If the encoding computation cost is less than the preset processing burden threshold, the high-frequency encoded stream and the low-frequency data are concatenated to obtain a burden-optimized version of the data.

7. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The process involves extracting the feedback signal from the receiving end based on the load-optimized version data, parsing it to obtain sound quality indicators and signal-to-noise ratio, and then verifying its validity to obtain a sound quality evaluation result, including: The optimized version data is sent to the receiving end, and the feedback signal is extracted. Audio quality indicators such as audio spectrum integrity and auditory clarity, as well as channel signal-to-noise ratio values, are extracted from the feedback signal. The logical correspondence between the sound quality index and the signal-to-noise ratio is determined according to the preset audio transmission quality verification rules. A valid identifier is generated based on the logical correspondence. Invalid data with logical abnormalities is removed based on the valid identifier. The indicator data corresponding to the valid identifier is extracted from the feedback signal, and the indicator data and the preset sound quality threshold are used to construct a difference measurement matrix. The sound quality evaluation result is obtained by weighting the difference measurement matrix.

8. The low-power Bluetooth headset audio data transmission method according to claim 1, characterized in that, The process of locating defective segments based on the sound quality assessment results, extracting spectral missing regions and temporal distortion intervals from the defective segments, generating compensation data packets based on the spectral missing regions and temporal distortion intervals, and performing continuity checks on the compensation data packets to obtain the final audio stream includes: Based on the sound quality assessment results, locate defective segments with missing high-frequency information and waveform distortion, and extract the spectral missing region and temporal distortion interval from the defective segments; A frequency domain gain value is generated based on the missing spectrum region, and a waveform filling sequence is generated based on the time domain distortion interval. The frequency domain gain value and the waveform filling sequence are then combined to form a compensation data packet. The compensation data packet is embedded into the defect location of the audio data stream to obtain the compensated audio stream; Calculate the continuity feature value of the compensated audio stream. If the difference between the continuity feature value and the feature value of the preset feature value template is less than the preset integrity threshold, the final audio stream is obtained.

9. A low-power Bluetooth headset audio data transmission system, characterized in that, include: The initial compression module is used to acquire audio data streams and perform preprocessing to obtain a preliminary compression scheme; The data optimization module is used to segment and calibrate the audio data stream according to the preliminary compression scheme to obtain an optimized compression block; The channel adaptation module is used to acquire channel state information and extract interference intensity, calculate transmission stability based on the interference intensity, calculate interference disruption degree based on the transmission stability and generate compensation factor, and configure the channel based on the compensation factor to obtain channel adaptation parameters. A redundancy simplification module is used to fuse the channel adaptation parameters and the optimized compression block to obtain a fusion mapping relationship, and to perform simplification verification based on the fusion mapping relationship to obtain a simplified data unit. The encoding burden reduction module is used to perform spectral decomposition on the simplified data unit to obtain high-frequency components and low-frequency data, and to encode and concatenate the high-frequency components and low-frequency data to obtain a burden-optimized version of the data. The sound quality evaluation module is used to extract the feedback signal from the receiving end based on the load-optimized version data and parse it to obtain the sound quality index and signal-to-noise ratio. After validity verification, the sound quality evaluation result is obtained. The audio compensation module is used to locate defective segments based on the sound quality evaluation results, extract spectral missing regions and temporal distortion intervals from the defective segments, generate compensation data packets based on the spectral missing regions and temporal distortion intervals, perform continuity verification on the compensation data packets, and obtain the final audio stream.

Images