18 results about "Burst noise" patented technology

This invention discloses an underwater acoustic anti-interference transmission method and system based on sparse time-frequency feature mapping. The method includes: at the transmitting end, adaptively generating a fractional-order linear frequency-modulated waveform with a specific time-frequency shear slope based on the Doppler state of the underwater acoustic channel, achieving physical-layer focusing of transmitted energy; at the receiving end, constructing a hybrid observation model containing wide-block sparse channel components and narrow-block sparse burst noise components, and using a dual-channel variational Bayesian algorithm to jointly iteratively infer the posterior probability distribution of environmental burst noise in the fractional-order domain; finally, recovering the original signal through soft-threshold interference cancellation and fractional-order channel equalization. This invention effectively solves the communication failure problem caused by high-dynamic Doppler diffusion and marine biological impulse noise interference in underwater acoustics by actively mapping the waveform and using heterogeneous sparse joint inference at the receiving end, significantly improving transmission reliability in harsh underwater acoustic environments.

This invention discloses a method, apparatus, terminal device, and computer-readable storage medium for suppressing high-frequency noise. The method acquires a target audio signal, inputs it into a target noise model for classification, and obtains the noise type of the target audio signal. When the noise type is burst noise, multi-segment suppression processing is performed on the target audio signal based on a multi-segment noise suppression curve to obtain a reduced-noise frequency signal. The multi-segment noise suppression curve is obtained by segmenting a preset default noise suppression curve. This invention provides a high-frequency noise suppression scheme that segments and suppresses high-frequency burst noise, thereby ensuring voice call quality while effectively suppressing high-frequency burst noise.

The present disclosure relates to a noise reduction device, a noise reduction method, and a program that enable the appropriate reduction of noise. According to the present invention, when a sudden noise event is detected by a sensor signal from a sensor in an FF-NC device, a control signal supplied to an actuator in an FB-NC device is reduced. The present invention can be applied to an NC device.

The application provides a skin deformation-based biosignal unwinding method and system, which is applied to the field of data processing; the original time sequence signals of skin micro displacement, strain and vibration change are collected by a deformation detection device based on a smart bracelet, combined with the motion information obtained by an acceleration and angular velocity sensor, motion interference can be recognized and inhibited in real time in a dynamic wearing environment, low-frequency motion components are separated from the skin deformation signal through a time-frequency masking strategy, and further matching, reconstruction and post-processing are performed on the sub-signal components, the continuity and stability of the target physiological signal can be effectively restored, and outliers caused by phase jumps and burst noises are eliminated.

The invention belongs to the technical field of active noise reduction, and discloses a multi-modal sensing-based adaptive environmental noise active noise reduction system, which comprises the following modules: a multi-modal sensing module, a signal preprocessing module, an adaptive control module, a noise cancellation module and a feedback correction module. The multi-mode sensing module is arranged to comprehensively capture multi-source noise such as air sound structure vibration sound and burst noise, the counteracting rate of the low-frequency vibration noise in the frequency band of 20-500 Hz can be increased to 85% or above, the burst noise response time is shortened to be within 5 milliseconds, the problem that in the prior art, noise coverage is incomplete is effectively solved, and the noise coverage rate is improved. The algorithm convergence time can be controlled within 30 milliseconds during scene switching by matching with the improved variable step size LMS algorithm and scene feature weighted fusion logic, the noise reduction amount, namely the fluctuation amplitude of SNR improvement, is less than 3 decibels, the method can be adapted to various mainstream scenes in an office traffic room, and the problems of weak adaptive capacity and poor scene compatibility are solved.

The application discloses a call voice real-time noise reduction method and system based on dynamic noise perception, acquires a noisy original voice signal, carries out frame division and windowing pretreatment on the digital voice signal with noise, extracts the spectral characteristics of the noise, divides the noise into steady-state noise, non-steady-state noise and burst noise according to the spectral characteristics of the noise, constructs a lightweight neural network structure through model pruning and quantization, adjusts the parameters and strategies of the types of the steady-state noise, the non-steady-state noise and the burst noise to realize optimized reduction, carries out post-processing such as window removal and overlap addition on the voice signal after noise reduction, and outputs the voice signal after noise reduction; the application effectively removes the background noise in a complex noise environment through dynamic noise perception and an adaptive noise reduction strategy, significantly improves the voice quality, meets the real-time requirement, realizes real-time noise reduction processing under low delay conditions through lightweight model design, and meets the real-time requirement of voice communication and voice control.

The invention discloses an underwater acoustic anti-interference transmission method and system based on sparse time-frequency characteristic mapping, and the method comprises the steps: adaptively generating a fractional order linear frequency modulation waveform with a specific time-frequency shearing slope at a transmitting end according to the Doppler state of an underwater acoustic channel, and achieving the physical layer focusing of transmitted energy; at a receiving end, constructing a hybrid observation model comprising a wide block sparse channel component and a narrow block sparse burst noise component, and inferring posterior probability distribution of environmental burst noise in a fractional order domain through joint iteration by using a dual-channel variational Bayesian algorithm; and finally, recovering an original signal through soft threshold interference cancellation and fractional order channel equalization. Through active feature mapping of the waveform and heterogeneous sparse joint inference of the receiving end, the problem of communication failure caused by underwater acoustic high-dynamic Doppler diffusion and marine organism impulse noise interference is effectively solved, and the transmission reliability in a severe underwater acoustic environment is remarkably improved.

The present application relates to the field of wireless communication and private network coverage technology, in particular to a 800M wireless relay method suitable for subway station environment, which is applied to a communication system comprising terminals, base stations and relay devices. The method is as follows: in a low-power listening state, spatial radio frequency sampling sequence is obtained and the floor noise jump rate is calculated; based on the sampling sequence, the probability density function value under the assumption of pure noise and signal noise superposition is obtained, and the global cumulative statistics is recursively obtained; the optimal upper and lower bounds are calculated by combining the battery remaining capacity coefficient and the floor noise jump rate; the statistics is compared with the upper and lower bounds, if greater than or equal to the upper bound, the cluster forwarding mode is entered to forward signals, if less than or equal to the lower bound, the low-power listening is maintained, and if in the middle, the next sampling point is obtained and the iteration is returned. The present application realizes the adaptive cooperation of anti-electromagnetic impact and physical power limit under strong burst noise interference.

The embodiment of the invention provides an open earphone active noise reduction method and device, electronic equipment and a storage medium, and belongs to the technical field of earphone noise reduction. The method comprises the steps that a microphone collects an environment sound signal; the input compressor monitors the environment sound signal; in response to the change of the environment sound signal, the input compressor performs noise dynamic analysis on the environment sound signal to obtain a primary gain parameter; the first gain module performs first gain processing on the environment sound signal based on the primary gain parameter to obtain a primary noise reduction audio; the noise reduction filter carries out filtering processing on the primary noise reduction audio to obtain advanced noise reduction audio; a second gain module performs second gain processing on the advanced noise reduction audio to obtain a target output audio; and the sound output module outputs the target output audio. According to the embodiment of the invention, active noise reduction can be carried out on burst noise on the earphone.

The invention discloses an AI dialogue optimization method, system and device based on MAC CRC statistical characteristics and a medium, and relates to the technical field of intelligent voice interaction, and the method comprises the steps that an MAC layer carries out CRC on a received data packet, and generates an MAC CRC result; the MAC layer calculates at least one statistical characteristic parameter reflecting audio burst noise characteristics based on the MAC CRC verification result; the MAC layer synchronizes the MAC CRC check result and the statistical characteristic parameters to the application layer; the application layer judges whether the data packet is a suspicious data packet based on the MAC CRC check result; if the data packet is a suspicious data packet, the application layer decides whether to start application layer CRC verification on the data packet in combination with the statistical characteristic parameters and the real-time audio frame parameters of the data packet; and the application layer executes corresponding processing operation on the data packet according to the decision result of the CRC check. According to the invention, cross-layer collaborative optimization is realized, and the real-time performance, the accuracy and the system energy efficiency of voice interaction are remarkably improved without audio buffering.

The invention discloses a narrow pulse acquisition algorithm based on a JESD204B interface, and relates to the technical field of signal acquisition. The method comprises the steps of firstly calculating initial noise, a preset effective multiple and a preset effective number of times; then, the JESD204B interface data is subjected to preprocessing; then calculating the maximum amplitude in single-time received data through a three-stage pipeline; and finally, judging whether the acquired data is valid or invalid according to the set noise and the specific value, adaptively updating the noise value, and filtering burrs (including fixed noise and burst noise) with a certain width. According to the algorithm, the characteristic that data received by a JESD204B interface of a high-speed ADC is continuous acquisition points is utilized, the maximum amplitude in a single data packet is calculated, the maximum amplitude is compared with a preset noise value through a ratio method, and the validity / invalidity judgment of a narrow pulse signal is completed. The method has the advantages of low complexity, high anti-interference performance and flexible algorithm. The problems that in the prior art, the collected data volume is large, the data processing algorithm is complex, and burrs cannot be processed are solved.

This invention relates to the field of speech signal processing technology, specifically to a method and system for speech denoising and speech synthesis enhancement based on deep learning. It includes the following steps: complex spectrum data preprocessing; speech denoising processing, where each frame of preprocessed complex spectrum data is simultaneously input into a stationary background noise branch and a non-stationary burst noise branch for parallel denoising processing; speech synthesis enhancement processing; and complex spectrum data output. This invention guides branch allocation through a noise stationarity coefficient, diverting noisy speech frame by frame to a fixed receptive field depth separable convolutional branch and a dynamic receptive field deformable convolutional branch, achieving differentiated denoising between stationary background noise and non-stationary burst noise; and uses the same coefficient to drive a gating network to complete adaptive fusion of the dual-branch outputs, effectively overcoming the limitation of a single network structure in handling two significantly different types of noise, thus improving the accuracy and fidelity of speech denoising in complex scenarios.

The application relates to the field of data processing, in particular to a data acquisition method and system for an electric energy storage box, the method comprising: binning energy storage box data of each dimension to obtain a data range, and constructing an item set according to the data range; calculating a frequent coefficient according to the item set; calculating a weight coefficient of the energy storage data according to the frequent coefficient, and obtaining a filtered value after filtering the energy storage data according to the weight coefficient. The application improves the limitation of the weighted filtering on the processing of burst noise, and has the effect of improving the accuracy of the energy storage box data filtering.

The invention discloses a convenient floor impact sound detection method, and relates to the technical field of architectural acoustic detection, and the method comprises the steps: 1, synchronously collecting a vibration acceleration signal and an environmental noise sound pressure signal of a to-be-detected floor; step 2, performing spectral analysis on the vibration acceleration signal and the environmental noise sound pressure signal, and calculating a vibration noise coefficient by identifying an overlapped interference frequency band to realize frequency-band correction of the vibration signal so as to obtain a pure vibration characteristic signal; step 3, comparing the sensitive frequency band energy proportion and the vibration attenuation time of the pure vibration characteristic signal with a preset proportion threshold value and a time threshold value to judge the sound insulation grade; meanwhile, the time window duration is adjusted according to noise fluctuation, and the anti-jamming capability to burst noise is improved; through targeted signal processing and dynamic interference suppression, the accuracy and adaptability of floor impact sound insulation performance detection in different noise environments are improved.

This invention discloses a robust dynamic networking method for power line carrier communication that incorporates deep learning. Sites extract channel features using lightweight convolutional neural networks, while the central coordinator employs an attention-enhanced long short-term memory network to predict channel states. Optimal routes and dual-mode switching strategies are dynamically generated based on deep reinforcement learning. The central coordinator performs millisecond-level topology reconstruction, achieving network-wide synchronization through special synchronization beacon frames. The system automatically and incrementally updates model parameters daily, continuously optimizing network performance. Experimental results show that this method significantly improves the robustness of power line communication networks in complex noise environments, drastically reduces burst noise packet loss rate, and significantly shortens topology response time, providing highly reliable communication assurance for smart grids.

The invention discloses a power line carrier robustness dynamic networking method combined with deep learning. A site extracts channel characteristics through a lightweight convolutional neural network, a central coordinator predicts a channel state by adopting an attention mechanism enhanced long and short term memory network, and an optimal routing and dual-mode switching strategy is dynamically generated based on deep reinforcement learning. And the central coordinator executes millisecond-level topology reconstruction and realizes whole-network synchronization through a special synchronization word beacon frame. And the system automatically incrementally updates model parameters everyday, and continuously optimizes network performance. Actual measurement shows that the method remarkably improves the robustness of the power line communication network in a complex noise environment, the burst noise packet loss rate is greatly reduced, the topology response time is greatly shortened, and high-reliability communication guarantee is provided for an intelligent power grid.

This invention discloses a method for music signal analysis and dynamic lighting mapping based on multi-feature fusion, relating to the field of intelligent lighting control technology. The method reads the music signal, segments it by time granularity, and performs amplitude normalization, silent segment removal, and burst noise suppression to obtain an audio analysis segment sequence. Each audio analysis segment is then windowed and subjected to Fast Fourier Transform, dividing it into low-frequency, mid-frequency, and high-frequency bands. Amplitude information for each band is statistically analyzed, and a spectral state set is generated. A reference segment is determined based on the spectral state set, and brightness and hue adjustment indicators are identified, generating a lighting fluctuation primitive sequence. The lighting fluctuation primitives are then checked, merged, filled with gaps, and time-corrected in chronological order, expanding into a dynamic lighting recording sequence. Commands are encapsulated and sequentially issued using a lighting control protocol, and subsequent control commands are synchronously corrected based on the current playback time and dominant frequency band.

This invention discloses an active noise cancellation method for headphones and the headphones themselves. The method includes: acquiring motion state data of physical targets in the surrounding environment using a radar sensor mounted on the headphones; predicting an impending acoustic abrupt event and determining the corresponding estimated noise type when the motion state data meets preset approach warning conditions; extracting matching target filter coefficients from a noise cancellation strategy library based on the estimated noise type before the ambient noise reaches the headphone's feedforward microphone, and pre-loading these coefficients into the active noise cancellation processor to complete system pre-tuning; and directly processing the ambient noise using the pre-loaded target filter coefficients when the feedforward microphone detects the ambient noise, generating and outputting an inverse noise cancellation signal. This invention achieves advanced physical space perception and parameter pre-tuning through cross-sensor fusion, improving the response speed and noise cancellation stability to transient sudden noise.