73 results about "Quantization (signal processing)" patented technology

A video signal decoding device comprises a processor which: determines a particular quantizer for reconstructing a first quantized transform coefficient, the particular quantizer being one of a first quantizer and a second quantizer which are different from each other, the particular quantizer being determined on the basis of the state of the first quantized transform coefficient; reconstructs the first quantized transform coefficient on the basis of the particular quantizer to obtain a reconstructed transform coefficient; and updates the state of a second quantized transform coefficient that is reconstructed after the first quantized transform coefficient, wherein the first quantized transform coefficient and the second quantized transform coefficient are transform coefficients in the current block.

The invention relates to the technical field of signal processing, and particularly provides a full-quantization spectrum characteristic statistical method, system and device and a storage medium, and the method comprises the steps: discretizing a current sampling signal frequency band into a continuous frequency point unit sequence according to a preset rule, and mapping a signal intensity value into an intensity level according to the preset rule; if the signal intensity exceeds the current noise base value, determining the frequency point range of the current sampling signal by extracting the center frequency point and the effective bandwidth of the current sampling signal, and recording the intensity level mapped for all frequency point units in the range once; if the signal intensity does not exceed the current noise base value, recording the intensity grade of mapping for one time for a frequency point unit in the frequency point unit sequence of the current sampling signal; and counting all involved frequency point units in the frequency spectrum recording process and the occurrence frequency of each intensity level recorded on each frequency point unit. According to the invention, the problem of virtual high background noise caused by a traditional peak value holding mode is solved.

The invention discloses an AI-driven vehicle-mounted sound field real-time modeling and voice separation method, and relates to the technical field of voice signal processing. A main control unit comprising a time sequence synchronizer, a resource scheduler and a health monitor is constructed. 3D sound field modeling is carried out by adopting a lightweight STCN + bidirectional LSTM network, adaptive updating of the model is realized through EWC incremental learning, and a CNN-LSTM noise classification network and targeted suppression algorithms such as ANF / spectral subtraction are developed. The voice separation module adopts an improved Conv-TasNet architecture, 3D spatial constraint and a multi-task loss function are fused, and low delay is realized under INT8 quantization and pipeline processing. The system dynamically optimizes parameters through a real-time regulation and control unit, supports scene self-adaption, finally achieves a separation effect in a mixed noise scene, reduces the delay of the whole system, and effectively improves the definition and stability of vehicle-mounted voice interaction.

The invention discloses a controllable sparse migration attack method based on generator generation, and belongs to the crossing field of communication signal processing and artificial intelligence. The method aims at solving the problems that disturbance of existing countermeasure attacks in the field of electromagnetic signals is too concentrated, concealment is insufficient, and disturbance density is difficult to control. The technical scheme comprises the following steps: simultaneously outputting disturbance amplitude and disturbance probability by using a deep generation network; classifying key regions by calculating group frequency spectrum saliency positioning pairs and generating a group mask matrix; generating a binary position matrix through a mask quantization network in combination with the group mask and the disturbance probability; training the generative network by adopting a total loss function comprising adversarial loss and smooth loss; and finally, generating a sparse controllable adversarial sample through the trained network. According to the method, the adversarial sample which is high in concealment, high in mobility and controllable in disturbance proportion can be generated, and an effective means is provided for security assessment of the deep neural network.

The application provides a pulse density modulation interface implementation method and device of an analog-to-digital converter, electronic equipment and a storage medium. The method comprises the following steps: designing a circuit architecture of an analog-to-digital converter, and setting working variable parameters for components in the circuit architecture; determining a transfer function of an output signal and a transfer function of noise based on the working variable parameters; in the case that an oversampling rate (OSR) is a first set value, a filter order is a second set value, and a quantization bit number is a third set value, calculating an optimal solution of the transfer function of the noise; determining a value corresponding to the working variable parameters of the components according to the optimal solution; and selecting corresponding components for the circuit architecture based on the value, and generating a pulse density modulation interface of an analog-to-digital converter. The application can accurately design a PDM interface, and improves the signal processing performance of the interface.

The application belongs to the technical field of radar signal processing and imaging, and provides a differential coupling single-frequency time-varying low-quantization imaging method, which comprises receiving a SAR echo signal, constructing a single-frequency time-varying threshold signal, single-bit quantization processing of a reference channel, single-bit quantization processing of a phase shift channel, double-channel data fusion and harmonic cancellation, and imaging processing based on a synthesized signal. The application effectively reduces the quantization noise base and improves the detection capability of a SAR system on weak targets under the premise of not significantly increasing the hardware cost.

The application discloses a jitter auxiliary measurement precision enhancement method for space-based calibration, belongs to the technical field of electronic communication and signal processing, and comprises the following steps: receiving a continuous-time analog input signal from a space-based calibration system; injecting an artificial jitter signal with zero mean and subject to uniform distribution into the continuous-time analog input signal to generate a composite analog signal; performing multiple independent quantization observations on the composite analog signal by using a low-bit analog-to-digital converter, obtaining multiple quantization output values corresponding to discrete levels, calculating the arithmetic mean of the multiple quantization output values, and taking the arithmetic mean as a high-precision estimation value of the amplitude of the continuous-time analog input signal; and the jitter auxiliary measurement precision enhancement method for space-based calibration can greatly improve the measurement precision of the space-based calibration system without increasing the hardware power consumption and volume.

This invention discloses a method for covert processing of spaceborne signals based on Tent mapping, relating to the field of signal processing. In this invention, the transmitter generates a chaotic Tent sequence, which is then transformed by Box-Muller to obtain a Gaussian-distributed first interference signal. The chaotic sequence and the first interference signal are equally divided, and an identification signal containing a synchronization flag and a checksum is constructed based on the quantization value of the first data point of each chaotic sequence segment. An amplitude adjustment coefficient is determined for each segment of the target signal, and the adjusted first interference signal of each segment is superimposed with the corresponding segment of the target signal to obtain a mixed subsequence. An identification signal is added before each mixed subsequence to form a relay signal. At the receiver, a sliding window is used to match the synchronization flag in the received signal to locate the identification signal, the quantization value at the beginning of each segment is parsed, and a second interference subsequence is independently reconstructed. A coarse estimate is obtained by performing a two-dimensional correlation between time delay and frequency offset, and the received target signal is obtained through interference subspace projection cancellation. The method proposed in this invention has low hardware overhead, strong anti-interception capability, and high cancellation accuracy.

The invention discloses a lightweight lossless audio encoding and decoding method and system, and belongs to the technical field of audio signal processing. The method comprises the following steps: converting an original audio waveform into a compact coding expression through an encoder formed by connecting a TConv unit, a convolution unit, a down-sampling unit and a Local-Transform unit in series; a single finite scalar quantizer is adopted to discretize the coded representation, and a mixed quantization strategy is introduced to generate a single and consistent discrete token stream; and finally, a high-fidelity audio waveform is reconstructed through a decoder formed by connecting a Local-Transform unit, a TEConv unit, an up-sampling unit and an audio recovery unit in series. Through unit structure optimization and joint loss function training, indexes such as SDR, PESQ, STOI, WER and the like under multiple different code rates are all superior to those of a baseline model, model parameters are only 10.31-11.29 M, collaboration of lightweight deployment and high-fidelity reconstruction is achieved, resource-constrained equipment is adapted to downstream scenes such as voice communication, ASR and the like, and good engineering practicability is achieved.

The invention relates to the field of MIMO signal processing, and discloses an MIMO-oriented bit width quantization method, device, equipment and medium, and the method comprises the steps: obtaining a plurality of variables to be quantized, carrying out the quantization of the integer bit width and sign bit width of the variables, obtaining the initial population of the variables, and carrying out the iterative optimization of the initial population, in any iteration process, carrying out the quantization of the integer bit width and sign bit width of the variables, and carrying out the quantization of the integer bit width and sign bit width of the variables. The method comprises the steps of determining a reference individual in a current population, performing decimal bit width updating on other individuals based on the reference individual, performing decimal bit width updating on any individual after the decimal bit width updating based on the target function value difference between the individual and the other individual, and determining a final population after iterative optimization is completed. And determining a target individual in the final population, and quantizing the decimal bit widths of the variables based on the decimal bit width combination of the target individual. According to the technical scheme provided by the invention, the automatic and efficient distribution of the decimal bit width in the MIMO detector is realized, and the calculation complexity and the hardware overhead are reduced.

The application discloses an ADC based on a neural network enhanced heterogeneous channel architecture and relates to the technical field of ADC design.The ADC comprises two-way heterogeneous ADC channels, a delay module, a neural network circuit and a streaming training controller.The two-way heterogeneous ADC channels are a high-speed low-precision channel and a low-speed high-precision channel, and the two channels are connected to the same signal source.The delay module performs delay and alignment processing on the output data of the two-way channels, the multi-time dimension data formed after the high-speed channel is delayed is aligned with the single time dimension data of the low-speed channel, the high-speed channel data is taken as training data, the low-speed channel data is taken as label data, the neural network circuit is used for training to generate high-speed high-precision quantization output, and the streaming training controller globally manages data acquisition and batch processing operations and generates training data through downsampling, time offset and batch processing.The application can realize the combination of high speed and high precision and meet the complex signal processing requirement.

The application discloses a controllable sparse transfer attack method based on a generator, and belongs to the cross field of communication signal processing and artificial intelligence. The method aims to solve the problems of existing adversarial attacks, such as too concentrated disturbance in the electromagnetic signal field, insufficient concealment and difficulty in controlling the disturbance density. The technical scheme comprises the following steps: a deep generation network is used to simultaneously output a disturbance amplitude and a disturbance probability; a group mask matrix is generated by locating a region critical to classification through calculation of group spectral saliency; a binary position matrix is generated through a mask quantization network in combination with the group mask and the disturbance probability; a total loss function containing an adversarial loss and a smoothing loss is used to train the generation network; and finally, a sparse controllable adversarial sample is generated through the trained network. The application can generate an adversarial sample with strong concealment, high transferability and controllable disturbance proportion, and provides an effective means for security evaluation of a deep neural network.

The invention relates to the technical field of signal processing, and discloses a small signal amplification and measurement device based on a band-pass filter. The multi-stage band-pass filter comprises a multi-stage band-pass filter module, a pre-amplification module, an analog-digital conversion module and the like. The center frequency of the multi-stage band-pass filter is distributed according to logarithm intervals, the bandwidth is dynamically adjusted, the pre-amplification module realizes low-noise amplification, and the analog-to-digital conversion module adopts an oversampling technology and a Sigma-Delta modulator to improve the quantization resolution. The dynamic gain control module adjusts gain in real time, and the noise suppression module eliminates periodic interference. In addition, the device also has the functions of multi-sensor fusion, dynamic range expansion, signal integrity detection, automatic calibration, temperature compensation, parallel processing and the like. The device can accurately amplify and measure small signals and effectively suppress noise, is suitable for the fields of biomedicine, communication, industrial monitoring and the like, and overcomes the defects of a traditional device in the aspect of small signal processing.

The invention provides a quantizer, a modulator and a signal processing method, and is suitable for the technical field of integrated circuits. The quantizer comprises a signal processing circuit, a comparison circuit and a logic control circuit, the logic control circuit is electrically connected with the signal processing circuit and the comparison circuit, and the logic control circuit is configured to control the signal processing circuit to increase the difference value of the second differential signal and the first differential signal by a preset voltage. A first group of target input signals are formed and output to the comparison circuits, and based on preset logic control information and digital output codes output by each comparison circuit, the signal processing circuit is controlled to carry out successive approximation type quantization operation on one group of differential input signals, and the rest of multiple groups of target input signals are output in sequence; therefore, the quantizer can complete the middle-stepping type quantization curve. According to the embodiment of the invention, the logic mode of successive approximation type quantization of the quantizer is changed through the logic control circuit, the application scene of the quantizer is widened, and the quantizer has a wide market application prospect.

The application discloses a satellite communication-oriented differential time hopping estimation and multi-hop coherent combination method and device, and belongs to the field of digital signal processing. A sending end repeatedly sends information symbols in discontinuous time slots, and the time interval of each sending is randomly changed within a predetermined range according to an agreement between the two communication parties. After the time hopping signal is transmitted through an AWGN channel, a receiving end jointly estimates the time interval of adjacent two hops and the multi-hop carrier phase by using a cross-entropy iteration method. A mapping relationship between the time interval of the hop and the carrier phase and binary quantization bits is established, a binary quantization bit generation probability is iteratively updated, the estimated values of the time interval of the hop and the multi-hop carrier phase are regenerated, multi-hop coherent combination is performed, and the parameter estimated value with the minimum combination SNR loss is selected as a global optimal vector group. According to the global optimal vector group, multi-hop coherent combination is realized to obtain SNR gain. The application can reduce the complexity compared with a traditional traversal method.

The present application relates to the technical field of signal processing, and specifically provides a full-quantization spectrum feature statistical method, system, device and storage medium, comprising: discretizing a current sampling signal frequency band into a continuous frequency point unit sequence according to a preset rule, and mapping a signal intensity value into an intensity level according to a preset rule; if the signal intensity exceeds a current noise base value, determining a frequency point range of the current sampling signal by extracting a center frequency point and an effective bandwidth thereof, and recording the mapped intensity level of all frequency point units in the range once; if the signal intensity does not exceed the current noise base value, recording the mapped intensity level of a frequency point unit in the frequency point unit sequence of the current sampling signal once; and counting all involved frequency point units in the present spectrum recording process, and the occurrence number of each intensity level recorded on each frequency point unit. The present application solves the problem of false high background noise caused by the traditional peak value maintaining mode.

The invention relates to the technical field of wireless communication signal processing, and discloses a low-complexity predistortion method and device for a QV communication power amplifier, and the method comprises the steps: obtaining an amplitude quantization interval index through coordinate conversion, and carrying out the synchronous addressing to look up a difference coefficient table and a shift control table; in forward processing, a conventional hardware multiplier is replaced by a dynamic shifter and adder cascade architecture, and nonlinear correction is realized by performing exponential shift scaling on a differential coefficient. A feedback path adopts an error symbol extraction and symbol flipping rate statistical technology to construct a double-loop adaptive updating mechanism: a fast loop utilizes a symbol error to finely adjust a difference coefficient in real time, and a slow loop dynamically adjusts a shift parameter according to a flipping rate statistical result. On the premise of ensuring the linearization effect, the dependence on the hard core of the DSP multiplier is eliminated, and the hardware resource consumption and the system power consumption are reduced.

The invention discloses a displacement measurement system and method for generating virtual moire fringes based on a digital grating, and belongs to the technical field of optical measurement, and the system comprises a laser light source, a real object scale grating, a light field collection module, a signal processing device, a digital indication grating, a virtual interference generation module and a displacement calculation module. In a physical space, a gray scale image generated by a light field reflected by the real object scale grating is captured by the light field acquisition module and is converted into a digital signal, and filtering and high-precision light field information extraction are realized by using the signal preprocessing module and the phase resolving module. In a digital space, a complex amplitude distribution function is used as a simulation carrier of optical behaviors of the digital indication grating, and generation parameters of the digital indication grating are configured to be consistent with a real scale grating. And calling a virtual interference generation module to superpose the preprocessed physical light field and the virtual light field, comparing an ideal moire fringe period with a reference moire fringe period acquired in real time, and carrying out real-time compensation on the parameters of the digital indication grating based on a digital closed-loop measurement method of error quantization-parameter adjustment-active compensation. According to the method, by dynamically updating grating parameters of a digital light field model, cooperative compensation of multi-source errors and real-time closed-loop compensation of whole-process measurement errors are achieved, and the stable requirement of ultra-precision manufacturing for nanoscale measurement is met.

The invention provides a dual-frequency laser interference signal phase resolving and compensating method based on SDFT interpolation, and belongs to the technical field of digital signal processing. The method comprises the following steps: carrying out SDFT operation on an input signal, and updating a frequency spectrum in real time by utilizing a recursion formula; carrying out peak value detection in a set frequency range, and determining an integer index corresponding to a main peak; a Jacobsen algorithm is used to calculate a frequency offset estimator caused by a fence effect, a gain compensation correction factor is introduced to correct the frequency offset estimator, and system errors are eliminated, so that the accurate frequency of the signal is obtained; and finally, resolving phase information according to the accurate frequency difference between the measurement signal and the reference signal. According to the method, point-by-point updating of the frequency spectrum is achieved through SDFT, the improved interpolation compensation algorithm is combined, the real-time performance is guaranteed, meanwhile, quantization errors are effectively restrained, and high-precision displacement measurement of the double-frequency laser interference signals is achieved.

The application relates to a signal gain value adjusting method and device and a sound signal processing method, wherein the signal gain value adjusting method comprises the following steps: obtaining a target signal; identifying whether a target event exists in the target signal; if yes, adjusting the gain value of the target signal during the duration of the target event. The application can realize a linkage mechanism triggered based on the target event, can improve the quantization precision of the target event by virtue of the gain control, can reduce the operation times of the signal amplitude in the overall use time, can further reduce the invalid power consumption and the average power consumption in the overall use time, and can meet the requirements of high precision and low power consumption.

The invention relates to the field of passive microwave remote sensing signal processing, in particular to a radio frequency interference detection method and system based on third-order quantization tail probability. The method comprises the following steps: acquiring a passive microwave remote sensing signal and acquiring analog signal power; performing third-order quantization of which the threshold is Th on the analog signal, and obtaining a second-order moment of a third-order quantization signal; according to the Th, constructing a tail consistency ratio statistical magnitude; and obtaining a radio frequency interference judgment result Flag according to the TCR. The method has the advantages that the method can be embedded into a real-time radio frequency interference detection module of a satellite-borne passive microwave remote sensing system, has the advantages of low calculation complexity, small resource occupation and wide application range, and can effectively detect various types of radio frequency interference.

The present invention provides an encoder including a quantization circuit, a quantized data adjustment circuit and an encoding circuit. The quantization circuit is configured to perform a quantization operation on multiple blocks of current frame data in sequence, to generate multiple quantized data respectively corresponding to the multiple blocks. For each of the multiple blocks in the current frame data, the quantized data adjustment circuit adjusts multiple coefficients in the quantized data corresponding to the block according to an optimization level of the block, to generate adjusted quantized data. The encoding circuit is configured to encode the adjusted quantized data of each of the multiple blocks to generate encoded data.

The application provides a transient strong nonlinear characteristic signal phase space analysis method, a storage medium and computer equipment, belongs to the technical field of nonlinear dynamics, signal processing and impact engineering, solves the problems that the initial value sensitivity, bifurcation and chaos and other uncertainty evolution of a strong nonlinear non-stationary signal cannot be accurately described due to the fact that a traditional method is analyzed in a time domain / frequency domain range based on a linear theory framework, effective indexes for quantifying the difference of nonlinear dynamic behaviors of a system are lacked, and similarity analysis of a prototype and a scale comparison model cannot be realized at a nonlinear mechanism level. After preprocessing and blocking of a transient strong nonlinear signal, the signal is converted into a symbol sequence through phase space reconstruction, evolution paths and system similarity are revealed through complexity and symbol distance quantization analysis, and the application breaks through the limitation of traditional analysis and provides precise support for impact-resistant structure design.

The application relates to an active intelligent super-surface assisted MIMO communication system control method, a communication system comprising a base station, an active intelligent super-surface and a plurality of user terminals, a direct link between the base station and the user terminals being blocked by an obstacle, the method comprising: establishing a received signal model and a quantization noise model at the base station, and obtaining an uplink rate of each user terminal and a sum rate of all users; performing signal processing at the base station by using maximum ratio combining and an analog-to-digital converter with a precision lower than a preset threshold; establishing an active intelligent super-surface phase shift optimization problem according to the received signal model and the uplink rate, solving the optimization problem to obtain an optimal phase shift matrix, and controlling the active intelligent super-surface to operate according to the optimal phase shift matrix. Compared with the prior art, the application can overcome the multiplicative fading effect of a passive intelligent super-surface, and the problem of reduced communication performance caused by an undesirable phase shift, improve the communication system performance, and reduce system energy consumption and hardware cost.

This invention relates to the fields of underwater acoustic array signal processing and underwater acoustic measurement technology, and in particular to a long linear array calibration method based on dual-constraint quantization dynamic segmentation. The method includes: establishing a dual-constraint quantization segmentation model composed of the far-field formula and the passive sonar equations, and determining the maximum segment length; implementing dynamic non-uniform segmentation with the goal of minimizing calibration times, and pre-setting overlapping verification elements between adjacent segments; after independently calibrating each subarray, using the coordinate deviation of the overlapping points for segment-by-segment linear error compensation to align with the calibration benchmark; and judging spatial consistency by evaluating directivity indicators such as the 3dB width of the main lobe and the sidelobe suppression ratio of the entire array beamformation. If these are not satisfied, the compensation coefficient or the number of overlapping points is adaptively adjusted to form a closed-loop iterative optimization. This invention fundamentally solves the core physical contradiction of long linear array calibration, eliminates error accumulation, and improves calibration accuracy and overall array spatial consistency.

The invention discloses a voice recognition method and device based on artificial intelligence, and a medium, and relates to the technical field of voice signal processing, and the method comprises the steps: collecting voice data, carrying out the cutting, then carrying out the framing, constructing an encoder, carrying out the processing of the framed voice cutting, and outputting a latent variable; constructing an RVQ discretization quantizer to perform residual quantization on the latent variable to obtain a quantization vector, and defining commitment loss of the RVQ discretization quantizer; carrying out quantization vector output straight-through latent representation and index embedding construction for splicing, and constructing a content encoder to process splicing features and output voice content features; establishing a recognizer R by adopting a causal Conform to recognize the voice feature content and output the text content; and training the parameters in stages, and performing model deployment after training is completed. According to the method, low-delay identification without waveform reconstruction is realized, the deployment cost is reduced, and the method is suitable for long-term stable operation of a resource-constrained terminal.

This application discloses a neural signal encoding method, apparatus, device, and readable storage medium in the field of signal processing technology. After performing wavelet transform on the target matrix obtained from the quantization of the neural signal, this application selects features based on joint mutual information. In each feature selection using joint mutual information, it is possible to select approximately optimal features containing more important information. After feature selection based on joint mutual information, this application further simplifies the selected features. The simplified features retain as much information as possible, thereby achieving the goal of representing more information contained in the neural signal with as few features as possible, and thus encoding the neural signal more accurately. Therefore, this application can improve the effectiveness and accuracy of neural signal encoding results.

The invention provides a pseudo code resampling method for rapidly capturing a navigation signal of a GNSS (Global Navigation Satellite System) receiver, which comprises the following steps of: receiving an intermediate frequency digital signal output by a front-section radio frequency chip by an FPGA (Field Programmable Gate Array), mixing the intermediate frequency digital signal with a locally generated carrier wave, designing different sampling rates according to different integral accumulation time to sample the same pseudo code, calculating RMS values of I / Q paths of the signal, and calculating the RMS values of the I / Q paths of the signal; and the calculated RMS value is read by the DSP and written into the rapid capturing module, and different thresholds are set for the RMS value to re-quantify the signal which needs to be rapidly captured so as to obtain a more accurate capturing result. Based on the high, medium and low orbit adaptive loop signal processing technology, different sampling rates are set according to different integral accumulation times, the obtained RMS result is utilized to assist in quantization of signals of the rapid capture module, and more accurate Doppler and code phase capture results are obtained for different integral accumulation times.

The invention belongs to the technical field of digital signal processing in a radio frequency / microwave communication system, and particularly discloses a multi-frequency parallel compensation method for a nonlinear phase of a radio frequency device, which is characterized by comprising the following steps of: S1, acquiring a phase-frequency curve p (f) of an output signal of a radio frequency end of the radio frequency device; s2, converting the phase frequency curve p (f) into a plurality of amplitude-frequency responses amp (f), performing inverse Fourier transform on the amp (f) to generate a filter coefficient hp, and performing M-order truncation and N-bit fixed-point quantization on the hp to generate an all-pass filter coefficient hp (m); s3, performing orthogonal down-conversion on a digital intermediate frequency signal y (n) to be compensated to a baseband, inputting the digital intermediate frequency signal y (n) into the all-pass filter, and simultaneously completing phase compensation on all frequency components contained in the y (n) by utilizing a group delay characteristic; and the compensated baseband signal is recovered to the original intermediate frequency through quadrature up-conversion, and an output signal with a corrected nonlinear phase is obtained. Phase modulation can be carried out on multiple frequency signals at the same time, frequency measurement is not needed, and frequency and data alignment does not need to be considered.

The invention discloses a dynamic physical layer key generation method and system based on satellite channel reciprocity, and relates to the technical field of wireless communication physical layer security, and the method comprises the steps: calculating a relative radial speed based on the motion state parameter of a communication node, and estimating the maximum Doppler frequency shift; calculating the change rate of channel impact response through real-time channel measurement so as to evaluate the channel randomness; binary decision is carried out based on a quantitative reciprocity evaluation model, and a physical layer key generation mode or a standby security mode is started for dynamic decision; when a physical layer key generation mode is decided to be started, performing anti-frequency-offset joint detection and synchronization, extracting a multi-dimensional channel feature containing a multipath component, performing adaptive quantization according to a signal-to-noise ratio, performing information negotiation based on hybrid coding of an LDPC code and an LT code to correct inconsistency, and performing privacy enhancement processing to generate a final session key. Through dynamic decision and a complete signal processing flow, high-reliability and high-efficiency physical layer key generation of the satellite is realized.