118 results about "Signal compression" patented technology

Methods and systems implement a pleno-generation face video compression framework with bandwidth intelligence for generative models and compression. Heterogeneous-granularity facial description regularizes long-term dependencies between video frames and compensates for motion estimation errors caused by compact representations of motion information. A generative decoder reconstructs heterogeneous-granularity visual representations, providing auxiliary visual signals for attention-based recalibration of a GFVC-reconstructed face signal. A coarse-to-fine generation strategy avoids error accumulation. High efficiency for heterogeneous-granularity signal compression is achieved by two different entropy-based signal compression methods: heterogeneous-granularities feature representation from the key-reference frame as hyperpriors to optimize the entropy model for compressing heterogeneous-granularity feature from subsequent inter frames, and a feature difference operation for heterogeneous-granularities feature representation between key-reference and subsequent inter frames, such that the entropy model only compresses heterogeneous-granularities feature residual for redundancy reduction. Mixed-model dataset generation and training and model-specific dataset generation and training are also provided.

The invention relates to the technical field of voice processing, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a voice signal compression method, device, equipment and medium, which comprises the following steps: executing Fourier transform on an initial voice signal to extract an amplitude spectrum and a phase spectrum, respectively processing the amplitude spectrum and the phase spectrum to generate features, and splicing the features to obtain a compressed voice signal; the method comprises the following steps: carrying out quantization by adopting a residual vectorization mode to generate a compressed feature vector, executing entropy coding to obtain a compressed code stream, recovering the compressed feature vector at a decoding side, reconstructing a spliced feature through residual reverse quantization, enhancing feature expression through up-sampling operation, and restoring a voice signal through inverse Fourier transform. According to the method, a dual-path processing structure of amplitude features and phase features is constructed, feature compression is realized in combination with residual vectorization and entropy coding, inverse quantization and up-sampling enhancement are executed in a decoding stage, a spectrum energy structure and phase continuity are effectively reserved, and reconstruction precision and fidelity of voice signals are improved under the condition of low bit rate.

The invention discloses a signal feature compression method based on codebook discrete quantization and multi-task learning, and belongs to the technical field of crossing of signal processing and artificial intelligence. In order to solve the problem that in the prior art, compression efficiency, semantic retention and calculation complexity are difficult to consider in high-dimensional IQ signal compression, high-dimensional continuous features of an original IQ signal are extracted through an encoder; carrying out vector quantization by utilizing the learnable codebook to generate a discrete index and calculating quantization loss; inputting the discrete features into a decoder branch reconstruction signal to calculate reconstruction loss, and inputting a classifier branch prediction category to calculate classification loss; processing traditional features and coding features by adopting a multi-layer perceptron, and calculating comparison loss through cosine similarity; combining optimization quantization loss, reconstruction loss, classification loss and comparison loss to train a model; and finally outputting a discrete index as a compression feature. The method realizes efficient compression and classification semantic reservation, and is suitable for wireless communication, Internet of Things and other scenes.

The invention relates to the technical field of electric digital data processing, in particular to a charging pile remote verification method, system and equipment based on digital twinning and a medium, which are used for uniformly acquiring and processing various operating parameters and state information in the operating process of a charging pile to form standardized input; digital twin modeling is carried out, a virtual model corresponding to a real charging pile operation mechanism is constructed, and the state of the hardware logic and software protocol level is synchronously modeled; through real-time synchronization, alignment and mapping of time and variables are carried out on collected data and a virtual model, and timeliness and accuracy of virtual-real comparison are guaranteed; a state verification mechanism is adopted, operation deviation is calculated, potential faults are identified in combination with multi-period trends, and graded alarm information is generated; remote control is utilized, a control instruction is issued after abnormity is detected, reliable transmission is ensured by combining a secure channel, signaling compression and a receipt confirmation mechanism, feedback updating is completed through state verification and log archiving, and whole-process remote verification of the charging pile is achieved.

A compressor for compressing a control signal includes residual compression circuitry configured to sequentially receive a first control signal and a second control signal and perform compression based on a difference between the first control signal and the second control signal, coding mode determination circuitry configured to determine a coding mode based on a type of control signal, analog signal compression circuitry configured to perform compression based on indexing, and control signal compression circuitry configured to perform compression based on run-length encoding (RLE).

Aspects of the present disclosure relate to techniques for signal compression. Certain aspects described in this disclosure can be implemented in a method for wireless communication by a distributed unit (DU). The method generally includes compressing at least one reference signal (RS) in one or more first set of resource elements (REs) of a symbol and data in one or more second set of REs of the symbol, wherein the at least one RS is compressed using a first compression type and the data is compressed using a second compression type, and wherein the first compression type is different than the second compression type; and sending, to a radio unit (RU), the symbol having the at least one RS in the one or more first set of REs and the data in the one or more second set of REs compressed using the first and second compression types, respectively.

This invention discloses a sleep stage identification method, comprising: a multimodal data acquisition step; a preprocessing step, which normalizes the acquired physiological signals and segments them according to a set duration to obtain several original signal segments; a Williams RGB compression step, which maps the original signals to an RGB image to obtain a Williams RGB energy map; a feature extraction step, which inputs the Williams RGB energy map into a neural network model for feature extraction; and a comparison and classification step, which compares the feature values ​​with RGB standard thresholds to determine the corresponding sleep stage. This sleep stage identification method avoids directly processing the complex frequency domain features of physiological signals, thus reducing computational complexity. Three physiological signals within a certain time period only need to be represented by a single pixel in the image, achieving physiological signal compression and dimensionality reduction, further reducing computational load. This invention also effectively utilizes medical knowledge graphs to provide interpretable evidence for the sleep stage identification decision-making process.

The invention discloses a broadband random harmonic compression analysis method based on LSTM and multi-feature fusion, and belongs to the technical field of electric digital data processing. Comprising the following steps: collecting a broadband random harmonic signal for time domain analysis, and extracting a period; judging whether each period signal can be compressed or retained or not by using LSTM (Long Short Term Memory); carrying out down-sampling and coding compression on the compressible periodic section signal; decoding and reconstructing the signal through an LSTM decoder; the decoded and reconstructed signals are preprocessed, multi-dimensional features of the signals are extracted, and evaluation indexes are normalized; calculating the compression ratio of the dynamic fusion index and the signal; and dynamically displaying the evaluation index and the dynamic fusion index. According to the method, self-adaptive decision of compression or not is realized, the compression ratio is improved, and meanwhile, the spectrum fidelity and phase consistency are also kept; and parts which do not need to be compressed are reserved as original samples, so that real-time performance and error control are both considered.

A target detection device includes: a signal distribution unit to acquire a reception signal indicating a reflected wave of a pulse wave every time the pulse wave is emitted into space and divide acquired reception signals; signal compression units to compress respective reception signals after dividing by the signal distribution unit; and signal recovery units to recover a delay time of a reception signal after compression by any one signal compression unit among the signal compression units and a complex amplitude of the reception signal after compression and output a recovery signal indicating the delay time and the complex amplitude. Further, the target detection device includes a correlation processing unit to extract signal components correlated with each other from recovery signals output from the signal recovery units, and a target detection unit to detect a target on the basis of the signal components extracted by the correlation processing unit.

This invention discloses a method for compressing and transmitting pulse Doppler radar image signals, comprising the following steps: S1, acquiring pulse Doppler radar RD data and performing CFAR calculations on the RD data; S2, compressing the pulse Doppler radar RD data using the CFAR calculation results; S3, transmitting the compressed pulse Doppler radar RD data; S4, forming an image from a reconstructed RD map and map subframes using the pulse Doppler radar RD data. This invention can transmit the main information contained in the RD map with minimal loss of pulse Doppler radar RD map information using a smaller bandwidth.

The invention relates to the field of signal frequency detection in microwave photonics, and discloses a broadband instantaneous frequency measurement receiving system based on Brillouin spectrum compression, which utilizes a dual-tone pump optical carrier signal carried on an intensity modulator and an optical filter to generate a single Brillouin pump source regulated and controlled by unknown radio frequency. A stimulated Brillouin structure narrowed by an optical fiber ring cavity and a single-pumping reference source subjected to phase shift beat frequency, a signal in a 45GHz range can be compressed to a bandwidth of 2.5 MHz, and meanwhile, an original signal can be subjected to optical domain down-conversion processing by a reference source signal, so that a low-frequency signal easy to process electronically is obtained; besides, a double-fiber annular cavity design is adopted, the line width of the Brillouin resonant laser is further compressed through mode matching of interference of light waves in the cavity, test precision reduction caused by frequency spectrum compression is made up, the resolution ratio can reach + / -2.2 MHz, and finally the Brillouin broadband instantaneous frequency measurement receiving technology is achieved.

The invention discloses an electrophysiological signal acquisition and wireless transmission system and a signal acquisition method based on a microprocessor, and the electrophysiological signal acquisition method comprises the following steps: converting an acquired electrophysiological analog signal into an electrophysiological digital signal; performing data compression on the electrophysiological digital signal through a zero-copy DMA transmission architecture by adopting a dynamic reference value compression algorithm to obtain a compressed electrophysiological digital signal; the compressed electrophysiological digital signal is transparently packaged through a protocol to generate a layered data packet; the data packet is transmitted to an upper computer through a dual-mode data transmission module, and the data packet containing the electrophysiological signal is provided for the upper computer. The invention discloses a microprocessor-based electrophysiological signal acquisition and wireless transmission system and a signal acquisition method, which can avoid crosstalk caused by phase out-of-step in multichannel signal analysis, eliminate data handling overhead of an MCU (Microprogrammed Control Unit) kernel due to coherence distortion and realize multichannel strict synchronous sampling.

The application discloses a kind of detection communication integrated collision avoidance sonar signal design method and device, its method includes: modulating continuous frequency phase shift keying signal CFPSK, generates initial frequency and Chirp signal;Modulate continuous frequency phase shift keying signal, BPSK signal is modulated to Chirp by phase modulation two-way, then two-way signal is added, obtains final transmitting signal;After CFPSK signal is pulse width modulated, it is emitted by the transducer of collision avoidance sonar;Collecting collision avoidance sonar original echo signal, obtains original echo data;Original data collected is preprocessed, and two-way signal of virtual and real is obtained;Signal is compressed into short pulse bandwidth signal, generates the impulse response parameter of two groups of matched filters, each group is matched with virtual and real two-way signal to obtain four groups of matching results, and the final matching output is obtained from real number field to complex number field;The starting point of CFPSK signal is determined by detecting the peak position of matched filter output, and echo signal is accurately intercepted from received signal for communication decoding, and coherent detection is carried out on received signal and reference carrier.The application realizes the cooperative work of detection and communication function.

The application discloses a kind of array signal compression transmission and recovery method under multi-source scene, it is related to the technical field of wireless communication.The method comprises the following steps: obtaining the observation signal received by array matrix under multi-source scene, and the observation signal is compressed and transmitted to estimate, the received signal of array matrix is obtained;The received signal is transmitted to the preset signal recovery model as communication signal, and the signal recovery model includes signal iterative recovery module, denoiser module and iterative denoising module;The signal iterative recovery module carries out signal iterative recovery to the communication signal, the denoiser module carries out array parameter estimation to the communication signal in iterative recovery process, the iterative denoising module carries out parameter update according to estimated array parameter, and outputs the communication signal after parameter update.By the method of the application, the difficulty that the required calculation requirement is large caused by too many signal sources is overcome, and the signal can also be accurately recovered.

The invention discloses a method and a system for detecting weak signals and harmonic waves based on stochastic resonance of compression enhancement. The system comprises a vacuum system, an imprisoned field, a laser system, a fluorescence acquisition system and an external signal, the vacuum system comprises a vacuum cavity, an ion pump and a getter pump, and a calcium atom furnace and a surface electrode ion trap for trapping calcium ions are placed in the vacuum cavity; the trapping field comprises a radio frequency trapping field and a direct current trapping field, and the radio frequency trapping field is composed of a radio frequency signal source, a power amplifier and a spiral resonant cavity; the direct-current trapping field comprises a direct-current high-voltage power supply, a 6733 board card and an amplification chip. The fluorescence acquisition system comprises an imaging mirror, a CCD (Charge Coupled Device) camera and a photomultiplier; the external signal comprises a driving signal, a compression signal and a to-be-tested signal. According to the method, the compressed signal is used for replacing external random noise, ions are triggered to generate periodic phase change along with the to-be-detected signal, the to-be-detected signal is optimally matched with the bistable system by changing the strength of the compressed signal, and the signal-to-noise ratio and harmonic detection are improved to the maximum extent.

This invention discloses a method for extracting transformer vibration features based on multi-point collaborative sampling and time-frequency matrix reconstruction, comprising the following steps: S1. Establishing a finite element simulation model based on the transformer structure and acquiring original vibration signals from multiple locations on the surface of the tank; S2. Performing normalization preprocessing on the original vibration signals and dividing them into multiple data samples through sliding window sampling; S3. Multiplying each data sample by a window function, calculating the Fourier transform of each window, and constructing the short-time Fourier transform time-frequency matrix of the vibration signal by spectral splicing; S4. Based on the short-time Fourier transform time-frequency matrix, calculating the instantaneous frequency after signal compression, and mapping pseudo-frequency components within a set width to the instantaneous frequency according to the mapping relationship, thereby obtaining a high-resolution time-frequency matrix reconstructed by synchronous Fourier compression transform. This invention has the advantages of high feature extraction accuracy and high reliability.

The application belongs to the field of low-altitude unmanned aerial vehicle radar detection, and relates to a low-altitude unmanned aerial vehicle detection method, device, computer equipment and storage medium.The method comprises the following steps: initializing a radar system and receiving echo signals; performing full desquaring processing on the received echo signals to realize signal compression; performing improved Keystone transformation on the echo signals after full desquaring processing, wherein the improved Keystone transformation dynamically adjusts transformation parameters to adaptively compensate for the distance walking phenomenon caused by non-uniform target motion; performing distance-Doppler two-dimensional Fourier transformation on the echo signals after improved Keystone transformation; identifying real targets from the generated two-dimensional distance-Doppler spectrum and estimating the position and speed parameters of the real targets; and displaying the estimated position and speed parameters.The signal processing precision is improved, the target detection capability is enhanced, parameter estimation is realized, and the display intuitiveness is improved.

The invention discloses a lightweight semantic preserving coding and decoding and voice signal reconstruction method, which comprises the following steps of: acquiring an analog audio signal from an environment, inputting an original audio signal and an audio signal subjected to high-frequency filtering to two ends of a comparator, outputting an ultralow-bit-rate binary bit stream and transmitting the ultralow-bit-rate binary bit stream to a receiving end, and after the receiving end receives the signal, outputting the ultralow-bit-rate binary bit stream to the voice signal reconstruction end. The method comprises the following steps: inputting a voice signal to an end-to-end U-Net convolutional neural network of a bottleneck layer integrated LSTM module, outputting a reconstructed voice signal, training the network by using voice coding sample reconstruction loss and multi-scale short-time Fourier transform, and performing voice signal reconstruction on an ultra-low bit rate binary bit stream of a to-be-reconstructed audio after training. The technical contradiction that in resource-limited remote audio collection and distributed sensing tasks, node hardware resources are extremely limited, and ultra-low bit rate signal compression distortion causes that node microminiaturization deployment and high voice reconstruction quality cannot be considered at the same time is solved.

A method for compressing and transmitting an ECG signal by an ambulatory electrocardiogram (ECG) device, comprising: obtaining an ECG signal from a subject; determining whether the obtained ECG signal comprises ECG beat locations; pre-processing the obtained ECG signal with a de-trending analysis to generate a pre-processed ECG signal if the obtained ECG signal is determined to comprise ECG beat locations, or pre-processing the obtained ECG signal with a bandpass filter to generate a pre-processed ECG signal if the obtained ECG signal is determined to not comprise ECG beat locations; approximating the pre-processed ECG signal to generate an approximated ECG signal; quantizing the approximated ECG signal to generate a quantized ECG signal; generating a compressed ECG signal from the quantized ECG signal; transmitting the compressed ECG signal.

A power spectrum estimation method for compressed sampling of a generalized stationary HDR signal is disclosed. This method is used for processing aero-engine blade tip vibration monitoring signals. The method involves determining the threshold of a self-recovering analog-to-digital converter and the delay scheme for multi-channel compressed sampling based on prior information about the signal amplitude and frequency; compressing the signal using a multi-channel dual-threshold cooperative sampling method; recovering the correct amplitude information from the acquired signal using the dual-threshold information; constructing an indicator sequence and a data sequence based on the sampling delay scheme; and performing Fourier transform, element-square operation, inverse Fourier transform, and element-product calculation sequentially on the indicator sequence and data sequence to obtain the power spectrum estimate of the signal.

The invention relates to a mobile terminal emotion recognition acceleration method and system based on signal compression, and the method comprises the following steps: collecting a voice signal through a mobile terminal device, carrying out the feature extraction, generating a Mel spectrogram through an Fbank encoder, and carrying out the compression to obtain a compression feature; the mobile terminal device sends the compression feature to a server; the server restores the compression features to obtain a restored Mel spectrogram as voice modal input, and performs text recognition on the restored Mel spectrogram to obtain text information as text modal input; the server inputs the voice modal input and the text modal input into a multi-modal deep learning network at the same time for emotion recognition, and the multi-modal deep learning network achieves fusion of the voice modal input and the text modal input through a collaborative attention module. Compared with the prior art, the method has the advantages of high calculation and storage efficiency, low delay, accurate recognition result and the like.

Post training compression of 3DGS data is agnostic to training in a traditional signal compression perspective. Gaussian parameters are treated as signals. Pre-processing and transform coding techniques are used to compress the signals effectively. Firstly, lossless / lossy compression is performed on 3DGS geometry (positions, scales, rotations) using a point cloud coding-based (e.g., G-PCC, GeS) framework. Positions are compressed using occupancy tree coding. Scales and rotations are encoded as attributes using transform coding. The widely used block-based graph Fourier transform (GFT) is used to compress the attributes (base colors, spherical harmonic coefficients and opacities). In addition, a graph construction strategy is used for 3DGS data that computes the edge weights based on similarity (or dissimilarity) between the 3D Gaussian distributions using KL-divergence. Alternatively, positions can be encoded using occupancy tree (e.g., G-PCC, GeS) or AI-based PCC methods, and any subset of Gaussian parameters or the transformed coefficients of Gaussian parameters can be mapped into 2D frames and encoded by video coders.

This invention discloses a microprocessor-based electrophysiological signal acquisition and wireless transmission system and method, including the following steps: converting the acquired electrophysiological analog signal into an electrophysiological digital signal; compressing the electrophysiological digital signal using a dynamic reference value compression algorithm via a zero-copy DMA transmission architecture to obtain a compressed electrophysiological digital signal; encapsulating the compressed electrophysiological digital signal through protocol transparency to generate a layered data packet; and transmitting the data packet to a host computer via a dual-mode data transmission module, providing the host computer with a data packet containing the electrophysiological signal. This invention avoids crosstalk and coherence distortion caused by phase misalignment in multi-channel signal analysis, eliminates MCU core data transfer overhead, and achieves strictly synchronous sampling across multiple channels.

The invention discloses an array signal compression transmission and recovery method in a multi-source scene, and relates to the technical field of wireless communication. The method comprises the following steps: obtaining an observation signal received by an array matrix in a multi-source scene, compressing the observation signal, and carrying out transmission estimation to obtain a receiving signal of the array matrix; transmitting the received signal as a communication signal to a preset signal recovery model, wherein the signal recovery model comprises a signal iteration recovery module, a de-noising device module and an iteration noise reduction module; the signal iteration recovery module carries out signal iteration recovery on the communication signals, the de-noising device module carries out array parameter estimation on the communication signals in the iteration recovery process, and the iteration noise reduction module carries out parameter updating according to the estimated array parameters and outputs the communication signals after parameter updating. Through the method provided by the invention, the difficulty that the required calculation requirement is high due to excessive information sources is overcome, and meanwhile, the signal can be accurately recovered.

Embodiments of the present application relate to the field of digital signal processing, and disclose a radar digital signal compression and decompression method and circuit, an electrical device and a terminal. In the present application, a plurality of pulse sequences from a plurality of channels are obtained, a periodic binary modulation sequence corresponding to each channel is adopted, and the pulse sequences of the corresponding channels are digitally modulated and accumulated to obtain compressed radar digital signals. The period of the binary modulation sequence corresponding to each channel is a divisor of the total number of pulses in a frame. There are at least two binary modulation sequences satisfying the following constraint conditions: the periods of the binary modulation sequences corresponding to any two channels are different and are in a factor relationship, and are orthogonal within the least common multiple of each other's period, so that the storage overhead of the radar system is reduced without losing signal accuracy, reducing signal quality, or increasing additional cost.

The application discloses a method and system for detecting weak signals and harmonics based on compression-enhanced stochastic resonance, comprising a vacuum system, a trapped field, a laser system, a fluorescence collection system and an external signal; the vacuum system comprises a vacuum cavity, an ion pump and a suction pump, a calcium atom furnace and a surface electrode ion trap for trapping calcium ions are arranged in the vacuum cavity; the trapped field comprises a radio frequency trapped field and a direct current trapped field, the radio frequency trapped field is composed of a radio frequency signal source, a power amplifier and a spiral resonant cavity; the direct current trapped field comprises a direct current high-voltage power supply, a 6733 board card and an amplification chip; the fluorescence collection system comprises an imaging mirror, a CCD camera and a photomultiplier tube; the external signal comprises a driving signal, a compression signal and a signal to be detected. The application uses the compression signal to replace the external random noise, triggers the ion to have periodic phase change with the signal to be detected, changes the intensity of the compression signal, makes the signal to be detected and the bistable system reach the best matching, and maximizes the signal-to-noise ratio and the detection of harmonics.

The invention provides a method, a device and equipment for generating a nonlinear system model. The method comprises the following steps: establishing a piecewise linear nonlinear system model; obtaining an input signal compression matrix according to a corresponding relation between the input signal sequence of the nonlinear system and a preset LUT; and determining a model coefficient of the nonlinear system model according to the input signal compression matrix and the output signal matrix of the nonlinear system. According to the embodiment of the invention, the parameter calculation efficiency of the nonlinear system model is effectively improved, and the nonlinear system model is efficiently and accurately obtained.

Signal compression and noise shaping in a wireless communications system (WCS) is provided. Herein, a block compression circuit is integrated with a noise shaping circuit to concurrently perform downlink / uplink signal compression and noise shaping in the WCS. The block compression circuit performs block scaling compression on the downlink / uplink signal, which can cause a compression noise being distributed across an entire sampling bandwidth of the downlink / uplink signal. As such, the noise shaping circuit is configured to redistribute the compression noise from the entire sampling bandwidth to a selected portion of the sampling bandwidth. Accordingly, the redistributed compression noise can be effectively suppressed and / or filtered out when the downlink / uplink signal is received and decompressed. By concurrently performing block compression and noise shaping on the downlink / uplink signal, it is possible to achieve a good trade-off between compression ratio and latency, without compromising quality metrics of the downlink / uplink signal.

The amount of data sent in a unit time is reduced in a case of transmitting sensor data by using a transmission system involving band limitation.A detection circuit includes a plurality of detection elements, a plurality of scanning lines connected to a scan circuit, a plurality of signal lines connected to a detection circuit, and a control circuit. Each of the plurality of detection elements is connected to the corresponding scanning line and the corresponding signal line. The detection circuit is configured to sequentially read out detection values of signals of the plurality of detection elements through the plurality of signal lines. When a detection value of a signal of a detection element of interest is externally output, the control circuit outputs, as transmission difference data, difference data between a value based on a detection value of a signal of a detection element read out before the detection timing of the detection value of the signal of the detection element of interest and the detection value of the signal of the detection element of interest.

The invention provides a track irregularity signal compression and reconstruction method and system based on a compressed sensing theory, and mainly relates to the technical field of railway infrastructure health monitoring and big data processing. According to the method, the sampling accuracy in the engineering field is mainly improved, redundant information is reduced, the system bottleneck of storage and transmission is improved, a measurement matrix irrelevant to a sparse transformation base is introduced at a signal acquisition end, and a compression observation value far lower than the Nyquist rate is directly obtained. And then, an original track irregularity signal is reconstructed from a small number of observation values with high precision by solving a norm minimization problem at a data processing end. According to the method, the front-end data sampling rate, the hardware load and the data transmission and storage requirements are greatly reduced, meanwhile, it is guaranteed that the reconstructed signal meets the engineering analysis precision, and a core technical scheme is provided for a new-generation efficient and low-cost track detection system, real-time train-structure system dynamic analysis and the like.