37 results about "Distance spectrum" patented technology

The invention provides a high-precision angle estimation method for a distributed millimeter wave radar network, a medium and equipment, and belongs to the technical field of radar signal processing. According to the method, a large-aperture MIMO radar system is constructed based on a plurality of small millimeter wave radars, and an initial rough angle value of a target is obtained by using two-dimensional fast Fourier transform; accurately calculating the walking position of the distance spectrum peak value of each virtual antenna according to the distance resolution of the radar and the rough angle value, extracting all distance spectrum peak value data according to the walking position, and generating a one-dimensional angle vector; performing angle dimension fast Fourier transform on the one-dimensional angle vector to generate an angle spectrum; calculating a target angle estimation value according to the angle spectrum peak value position; the above process is iteratively circulated until the error between two adjacent angle estimation values is smaller than a preset threshold value. According to the method, different types of distributed millimeter wave radar network architectures can be flexibly adapted, the precision problem caused by target distance walking is effectively eliminated, and the angle estimation precision is greatly improved.

The invention provides a wind profile radar spectrum peak estimation method based on a three-parameter dynamic cost function, and the method comprises the steps: 1, carrying out the preprocessing of distance and Doppler spectrum data, and carrying out the three-point moving average smoothing; step 2, noise level estimation based on statistics is carried out; step 3, carrying out candidate spectrum peak selection based on space Doppler window smoothing; 4, performing dynamic weight calculation based on covariance matrix eigenvalue decomposition; step 5, constructing a three-parameter cost function of a distance spectrum peak value RSP power item, a continuity item and a spectrum width consistency item, and performing optimization; and step 6, outputting the optimal spectrum peak trajectory to form a complete wind speed profile trajectory. According to the method, the spectrum peak identification precision and stability in a complex environment can be improved, and the u and v wind correlation coefficients of the estimation method are verified to reach 0.880 and 0.879 through a data set disclosed by an atmospheric radiation measurement website.

The invention discloses a robust human body existence sensing method and system based on a millimeter wave radar in a complex interference environment, and aims to realize high-precision and high-robustness sensing of a personnel existence state in a complex indoor scene through a signal processing and multi-index fusion mechanism. The system takes a millimeter-wave radar as a core sensor, and through a signal processing flow comprising adaptive background modeling and coherent subtraction based on a complex number distance spectrum, a cleanliness index constructed by extracting multi-dimensional features from a micro-Doppler spectrum, and a dual-channel parallel detection mechanism for moving and static targets, the detection precision of the moving and static targets is improved. The core challenges that static detection is difficult, environment interference is sensitive, and system adaptability is insufficient are effectively overcome. According to the method, the environment confidence, the signal cleanliness, the moving target evidence and the static target evidence are fused for layered joint judgment, non-target movement and real human body movement can be accurately distinguished in the indoor environment with various typical interferences, and continuous and reliable detection of the full state of the person from moving to static is achieved.

A standing wave radar includes: a transmitter to set a frequency of a transmission signal to frequencies, and output the transmission signal taking the respective frequencies in a time-division manner; a receiver to receive a reflected signal taking the frequencies in a time-division manner, the reflected signal being the transmission signal reflected by an object; and a processor to obtain reflection coefficients of the frequencies by obtaining each reflection coefficient of the transmission signal and the reflected signal taking a same frequency, for the transmission signal and the reflected signal taking the frequencies, and to execute a first inverse Fourier transform process of calculating a first distance spectrum for the object by an inverse Fourier transform on the reflection coefficients, and a first distance measurement process of determining presence of the object and calculating a distance to the object, based on the first distance spectrum.

The application discloses a Parkinson gait evaluation method and system based on a millimeter wave radar signal. The method comprises the following steps: acquiring a millimeter wave radio frequency signal in an environment, performing data processing on the signal to obtain a distance spectrum of position change of a Parkinson patient's body and a Doppler spectrum reflecting the degree of change of the Parkinson patient's body; extracting a peak value at each moment from the distance spectrum through a filtering method, calculating a cross-correlation coefficient of the Doppler spectrum as an index of gait pattern matching, and dividing the Doppler spectrum according to each step; calculating gait features according to the filtered distance spectrum peak value and the divided Doppler spectrum; inputting the normalized gait features into a trained machine learning model for feature evaluation to obtain a UPDRS-III score of the patient. The application realizes non-invasive measurement and evaluation of Parkinson gait features in a low-cost manner, is convenient to operate, and is accurate in evaluation effect, thereby providing quantitative data and diagnosis reference for doctors.

The invention discloses a self-adaptive point cloud registration method based on mixed geometric constraint and statistical filtering, and belongs to the technical field of machine vision and three-dimensional point cloud processing. According to the method, an improved coarse-to-fine matching strategy is provided for solving the problems that a traditional distance signature algorithm is large in calculation redundancy, sensitive to noise and lack of scale adaptability. The method comprises the following specific steps: firstly, calculating normal vectors of a source point cloud and a target point cloud, carrying out rapid pre-screening by utilizing a normal included angle constraint, and removing candidate point pairs with inconsistent geometric features; secondly, a robust distance signature based on bilateral percentile truncation is constructed, and outlier noisy points in a distance spectrum are stripped through a statistical method; and finally, adopting a normalized dynamic matching threshold as a criterion to realize adaptive matching of point clouds with different densities, and solving a rigid transformation matrix through singular value decomposition (SVD). According to the method, the robustness and the calculation efficiency of point cloud registration in noise interference and partially overlapped scenes are remarkably improved.

The invention relates to vehicle radar systems having distributed antenna structures and to the evaluation of radar signals of a vehicle radar system of this type for environment detection, in particular speed and angle evaluation. Chirps in a frequency range are emitted in a time-offset manner by a radar system via different transmitting antennas and are detected simultaneously by receiving antennas. In addition, an additional chirp is continuously emitted in an additional frequency range, said chirp being detected and evaluated by receiving antennas which form a uniform linear antenna array. The range spectra of the receiving antennas of the uniform linear antenna array are coherently integrated and the speeds are determined therefrom by Doppler evaluations. The angles of the detected objects are also estimated. Both are allocated to the objects detected by the entire antenna structure over the determined distances. As a result, the further evaluation can be significantly accelerated, e.g. the number of angle hypotheses to be taken into account for each detected object can be limited.

A FMCW radar ranging method, system and medium based on power-weighted centroid, which comprises obtaining the beat signal of the FMCW radar and performing Fourier transform thereon to obtain an initial distance spectrum composed of multiple distance units; performing interpolation processing on at least two distance units corresponding to a preset near distance range in the initial distance spectrum to generate a refined distance spectrum with a higher frequency resolution than the initial distance spectrum; performing power-weighted centroid calculation on a continuous frequency spectrum interval corresponding to the preset near distance range based on the refined distance spectrum to obtain a first distance estimation value; performing peak value detection on the initial distance spectrum to obtain a second distance estimation value; performing time domain filtering processing on the first distance estimation values and the second distance estimation values obtained from continuous multiple frames of signals respectively; obtaining a target distance value based on the smoothed first distance estimation value and the smoothed second distance estimation value according to a preset fusion rule, and outputting.

The invention provides a millimeter wave radar automatic labeling method and system realized by using computer vision, and relates to the technical field of data processing, and the method comprises the steps: determining a corresponding motion video clip of a motion time period in video data and a corresponding radar frame sequence of the motion time period in millimeter wave radar data; determining a distance spectrum, a distance velocity spectrum and a distance angle spectrum of the target human body based on the radar frame sequence; determining a text label result of the target human body based on the action video clip; and determining radar labeling data corresponding to the target human body based on the distance spectrum, the distance velocity spectrum, the distance angle spectrum and the text label result of the target human body. According to the technical scheme, data labeling can be automatically completed through combination of millimeter wave radar data and video data, the amount of radar data needing to be collected is reduced, and manual labeling of a large amount of data is not needed. In addition, the generated annotation data is not limited to a single data format.

This invention discloses an indoor target localization method and system based on acoustic devices, belonging to the field of indoor measurement and positioning technology. The method includes: utilizing existing loudspeakers and microphone arrays to transmit ultrasonic linear frequency modulated continuous wave signals and receive target echoes; performing de-skew processing using a single reference signal in the digital domain, and then shortening the FFT operation length through a periodic extension reconstruction method; finally estimating the target distance and angle based on the distance spectrum and array signal, respectively. This invention achieves fully digital processing, eliminating the need for an additional analog front-end, significantly reducing hardware costs and computational complexity. It can run in real time on low-power microcontrollers and is suitable for embedded spatial perception in fields such as smart homes and robotics.

The invention relates to vehicle radar systems with distributed antenna structures and the evaluation of radar signals from such a vehicle radar system for environmental detection, in particular the evaluation of speed and angle. A radar system emits chirps in a frequency range with time delays via various transmitting antennas, which are simultaneously detected by receiving antennas. Additionally, another chirp is continuously emitted in a further frequency range, which is detected and evaluated by receiving antennas forming a uniform linear antenna array. The distance spectra of the receiving antennas of the uniform linear antenna array are coherently integrated, and the velocities are determined from this integration using Doppler analysis. Furthermore, the angles of the detected objects are estimated. Both of these values ​​are assigned to the objects detected by the entire antenna structure based on the determined distances.This can significantly accelerate further evaluation, e.g. by limiting the number of angle hypotheses to be considered for each recorded object.

The application discloses a synthetic aperture radar radio frequency interference suppression method and system based on S transform, a storage device and electronic equipment, and comprises the following steps: detecting interference-containing echo data by a method based on distance spectrum flatness and symmetry, and marking interference-containing pulses; calculating the amplitude spectrum of the interference-containing pulses, and performing smoothing and normalization processing; calculating the optimized window parameters after scaling the normalized amplitude spectrum; converting the interference-containing pulses to the time-frequency domain based on the optimized window parameters; positioning the interference position in the time-frequency domain through the OSTU algorithm, and eliminating the radio frequency interference through the wave-trap technology; inversely converting the processed interference-free pulses to the time domain based on the optimized window parameters; and combining all the processed interference-containing pulses and interference-free pulses to form new interference-free echo data. The application can effectively separate the radio frequency interference from the useful signal, thereby significantly reducing the influence of the radio frequency interference on the image quality of the synthetic aperture radar.

To improve the accuracy of distance estimation. [Solution] The distance estimation system 1 comprises a reference communication terminal 100 and a target communication terminal 10, each performing multi-carrier transmission, and an information processing device 500 that estimates the distance between the reference communication terminal and the target communication terminal according to the communication results between the reference communication terminal and the target communication terminal. The information processing device identifies a provisional distance estimate based on the received signal of the reference communication terminal transmitted and received between the reference communication terminal and the target communication terminal, removes the long-time delayed wave component from the received wave of the received signal based on the received signal from which the long-time delayed wave component has been removed, and estimates the distance based on the intensity of the distance spectrum within a predetermined search range of the distance including the provisional estimate.

A multifunctional sleep monitoring method based on a millimeter wave radar comprises the following steps: carrying out recombination, frame division and clutter removal on collected original ADC data, obtaining a target distance spectrum through one-dimensional FFT and inter-frame difference, and determining a target distance unit by adopting non-coherent accumulation. Extracting the phase information of the unit signal, unwrapping and differentiating to obtain a vital sign signal, separating breathing and heartbeat components by utilizing multi-mode decomposition and reconstruction, and obtaining the breathing frequency and the heart rate through Fourier transform; performing two-dimensional FFT and DBSCAN clustering on the data after inter-frame difference, generating three-dimensional point cloud data in combination with angle FFT, and inputting the three-dimensional point cloud data into a DGCNN and Transform cascade network to realize sleeping posture recognition; the system can analyze the vital signs and the sleeping posture state in real time, and when it is detected that apnea, abnormal heart rate or dangerous sleeping postures continuously exceed preset time, alarm information is automatically output. The method has the advantages of non-contact, high precision and real-time performance.

The present invention improves estimation accuracy of a distance.　A distance estimation system (1) comprises: a reference communication terminal (100) and a target communication terminal (10) that each perform multicarrier transmission; and an information processing device (500) that estimates a distance between the reference communication terminal and the target communication terminal according to a result of communication between the reference communication terminal and the target communication terminal. The information processing device: specifies a tentative estimation value of the distance on the basis of a received signal of the reference communication terminal transmitted and received between the reference communication terminal and the target communication terminal; removes a component of a long-delay wave from a received wave of the received signal on the basis of the tentative estimation value; calculates a distance spectrum of the received signal on the basis of the received signal from which the component of the long-delay wave has been removed; and estimates the distance on the basis of an intensity of the distance spectrum in a predetermined distance search range including the tentative estimation value.

This application provides a near-field target localization method, apparatus, electronic device, and storage medium, belonging to the field of communication technology. The method includes: acquiring an initial received signal of a preset symmetric coprime array; determining a first covariance matrix of the initial received signal based on the initial received signal; performing calculations based on the first covariance matrix to obtain an equivalent received signal of the initial received signal; determining a second covariance matrix of the equivalent received signal based on the equivalent received signal; performing spatial smoothing on the second covariance matrix to obtain a third covariance matrix of the equivalent received signal; performing a one-dimensional angle spectral peak search on the third covariance matrix to obtain a first angle between the near-field target and the central array element; and performing a one-dimensional distance spectral peak search on the first covariance matrix based on the first angle to obtain a first distance between the near-field target and the central array element. This application significantly reduces the computational overhead during near-field target localization.

The present invention relates to a method for processing radar data (100), wherein the radar data (100) represent measurement cycles (108) from several successive measurements (106), wherein a distance spectrum (112) is determined for each measurement (106), wherein a distance-velocity spectrum (124) is determined for a predefined distance range (114) of the distance spectra (112) over all measurements (106) of a current measurement cycle (108), wherein for at least one predefined distance range (114) of the distance spectra (112), a first distance-velocity spectrum (124) is determined over a first subset (120) of the measurements (106) of the measurement cycle (108), and at least one further distance-velocity spectrum (124) is determined over a further subset (126) of the measurements (106) that differs from the first subset (120). the measurement cycle (108) is determinedwherein after the determination (202) of the distance-velocity spectra (124) a target detection (204) is carried out, wherein the target detection (204) is carried out for at least the first subset (120) before an end of the current measurement cycle (108).

The invention relates to a detection method and system based on electromagnetic waves, an electromagnetic wave transmit-receive system is used for underground isomer target detection, a receiving end antenna and a transmitting end antenna of the electromagnetic wave transmit-receive system are dual-polarized antennas, distance spectrum analysis is carried out on a received signal, and a received signal distance spectrum is formed; according to the method, distance attenuation compensation is carried out on a received signal distance spectrum to form a received signal distance spectrum after distance attenuation compensation, and angle spectrum analysis can be carried out on the received signal distance spectrum to form a received signal distance angle spectrum under the condition that a dual-polarized antenna array is adopted; and based on the received signal distance spectrum or the distance angle spectrum after distance attenuation compensation, extracting polarization characteristics for isomer target identification, wherein the obtained polarization characteristics comprise classification information and other information of the isomer target. The method can be used for underground isomer target detection, and isomer target material and shape classification can be accurately identified.

The invention provides an anti-vibration detection method and device and a storage medium. The anti-vibration detection method comprises the following steps: acquiring sampling data of an echo signal by a radar sensor; performing fast time dimension calculation on the sampling data according to a time sequence to generate a corresponding distance spectrum, wherein the distance spectrum comprises a plurality of distance gates; selecting a range gate where an echo signal of a static object is located from the range spectrum as a reference range gate; acquiring amplitude and phase change information of the reference range gate in a slow time dimension to form a vibration waveform; at least compensating phases of other range gates based on the vibration waveform to obtain a compensated range spectrum; and carrying out slow time dimension calculation on the compensated distance spectrum, and carrying out target detection according to a calculation result. On the basis, zero-speed clutter energy caused by vibration of the base can be prevented from being leaked to a non-zero-speed area, the static clutter suppression effect of the radar sensor in a vibration environment is improved, and the accuracy of at least a detection result of a moving target is ensured.

The application discloses a Parkinson disease gait feature extraction method, a gait recognition method and a system based on a millimeter wave radar, and the feature extraction method comprises the following steps: collecting a millimeter wave radar signal, pre-processing the signal to obtain a Doppler spectrum and a distance spectrum of the signal; extracting a trunk speed curve from the Doppler spectrum of the signal by using an adjacent column constraint rule, extracting a trunk position change curve from the distance spectrum, and extracting basic gait information from the trunk speed curve and the trunk position change curve; extracting an envelope speed curve corresponding to sample motion from the Doppler spectrum, and screening out key gait features in combination with the basic gait information. The application does not depend on a multi-sensor wearing system, and fully reveals typical motion difference characteristics of Parkinson disease patients by targeted modeling of gait detail characteristics.

The application discloses a millimeter wave radar distance domain target detection method based on complex binary transformation, and realizes target distance detection by constructing a double-path symmetric complex binary network, and the implementation scheme comprises the following steps: 1. constructing samples containing different target distances and noise intensities as a training data set; 2. constructing a complex binary transformation network (CVBT) model; 3. constructing a loss function by adopting a GrandNorm strategy, solving the gradient disappearance problem in the back propagation of the binary neural network by using a straight-through estimator, and training the CVBT network parameters by using the training sample data set to obtain the CVBT network parameters; 4. performing complex binary transformation on radar echo by using the trained CVBT network, obtaining a distance spectrum, and performing constant false alarm rate detection on the distance spectrum to obtain the specific position of the target. The method realizes the purpose of replacing FFT for frequency domain analysis by adopting the binary neural network parameter training mode for the first time.

The application discloses a visual positioning and detection method for a battery assembling process, and particularly relates to the field of power battery production and detection, and is used for solving the problems that the main edge of the pole piece and the texture of the diaphragm are difficult to be stably aligned under light fluctuation and working condition switching, the deviation causes are difficult to be accurately determined, and the disposal action is easy to be mis-triggered; a unified feature caliber is constructed through homologous image acquisition and same-piece state binding, a unique deviation cause label is output by using the fusion determination of the sequence mapping inverse bit rate and the law distance spectrum, a disposal instruction is issued when the direction, position and trigger condition are consistent, the same caliber re-inspection after disposal, verification result alignment and template recursive update are combined, integrated closed-loop management and control of positioning detection, action execution, result review and version traceability are realized, and the assembling consistency and disposal certainty are improved.

The invention relates to the technical field of industrial process measurement and automatic instruments, in particular to a frequency-modulated continuous wave radar storage tank liquid level measurement method which comprises the following steps: acquiring an orthogonal baseband signal and generating a complex distance spectrum based on an uphill linear frequency-modulated signal, a downhill linear frequency-modulated signal and complementary phase coding; extracting reference echoes to calibrate the complex distance spectrum and determine a coding phase alignment angle, and subtracting the fixed structure echoes by using the fixed structure echo dictionary to obtain a fixed structure echo stripping complex distance spectrum and generate residual energy; complementary phase coding coherent combination is carried out based on the coding phase alignment angle, and candidate distance observation and candidate feature vectors are generated according to time delay consistency; and outputting liquid level estimation and confidence coefficient by adopting dynamic programming, performing change point detection on residual energy to generate a dictionary freezing mark and a dictionary rollback mark, generating an active disambiguation control instruction by combining the confidence coefficient, and updating emission scheduling parameters. The false peak and jump are effectively reduced, and the measurement stability and the adaptive anti-interference capability are improved.

The invention belongs to the field of signal processing and communication positioning, and discloses a star-flash anti-multipath phase ranging method based on fast Fourier transform and double-peak ratio re-check. The method comprises the following steps: acquiring phase response under a multi-frequency hopping channel of a star flash communication protocol, constructing a complex number of frequency domain signals, obtaining a time delay spectrum through fast Fourier transform, and mapping the time delay spectrum into a distance spectrum; in the distance spectrum, a threshold value is set according to amplitude distribution, and a first candidate peak exceeding the threshold value is searched for as a preliminary result; and then searching a second candidate peak, when the distance between the two candidate peaks is less than a set threshold and the amplitude ratio is greater than a preset threshold, determining that the first candidate peak is a pseudo peak, and outputting the distance corresponding to the second candidate peak as a final result. According to the method, a threshold screening and double-peak ratio re-checking mechanism is added under an original FFT distance measurement framework, false peak misjudgment caused by multipath interference is effectively suppressed, the distance measurement precision and robustness are remarkably improved, and the method is suitable for high-precision short-distance positioning application in a complex indoor environment.

The application discloses a kind of millimeter wave radar human perception detection method and system based on adaptive low operation complexity, the method first utilizes to obtain echo data, carries out discrete fourier transform, generates corresponding distance spectrum;Multiple distance spectra are accumulated in amplitude, to obtain the distance spectrum after accumulation.Calculate the cross-correlation coefficient between current frame and previous frame distance spectrum, pearson correlation coefficient and structural similarity index, and through adaptive linear combination method, generate comprehensive decision coefficient, according to the comprehensive decision coefficient to human existence is detected and judged.This method can dynamically adapt to human detection needs in different scenarios by combining various correlation analysis techniques, has lower computational complexity and higher robustness, especially suitable for human perception detection tasks in low-power application scenarios, so that the human perception detection capability of millimeter wave radar in low-power applications has been significantly improved, while maintaining high detection accuracy and reliability.

The application provides a method for detecting interference from a SAR single-view complex image, comprising the following steps: performing distance Fourier transform on a SAR image to be processed; performing equalization processing on a distance spectrum of the SAR image; equally dividing the distance spectrum into a plurality of subbands; performing inverse distance Fourier transform on each subband spectrum to obtain a plurality of subband images; calculating the contrast of each pixel in the subband domain; and comparing the calculated contrast with a threshold value, and detecting a pixel as interference if the contrast exceeds the threshold value. The application can effectively detect intermittent sampling repeater interference and common unintentional interference from a SAR image.

Method for determining at least one target information of a target object of a sensor system (2) which has several transmitting antennas (3) and several receiving antennas (4), wherein: - in a transmission process, several frequency-shifted electrical outgoing signals (40 to 43) are simultaneously transmitted from the several transmitting antennas (3) into an environment (18); - electrical receiving signals (54 to 57), which are based on the transmitted electrical outgoing signals (40 to 43), are received by the several receiving antennas; - on the basis of the received electrical receiving signals (54 to 57), the antenna positions of the several transmitting antennas (3) and the antenna positions of the several receiving antennas (4), a virtual antenna array (39) which has several virtual receiving antennas (61) is generated; - an environment reconstruction of the environment (18) is performed based on the virtual antenna array (39) is carried outand- which at least one target information is determined on the basis of the environmental reconstruction, wherein- a respective received electrical receive signal (54 to 57), which is assigned to a respective virtual receiving antenna (61) of the virtual antenna array (39), is mixed with a carrier signal (59) on which the several frequency-shifted electrical transmit signals (40 to 43) are based, and- for each virtual receiving antenna (61) a distance spectrum (62) is determined on the basis of the received signal (54 to 57) belonging to the respective receiving antenna (61) and mixed with the carrier signal, characterized in thatthat a respective distance spectrum (62) of a respective virtual receiving antenna (61) is decomposed into partial spectra (64 to 67) depending on a corresponding electrical outgoing signal (40 to 43) to the received signal (54 to 57) and by the transmitting antenna (3) emitting this electrical outgoing signal (40 to 43).

The invention discloses an image drift correction method and device, a storage medium and computer equipment. The method comprises the following steps: in an image acquisition process, performing multiple times of imaging on the same area of a sample at different rotation angles to obtain a plurality of initial images at different acquisition angles; for each initial image, extracting the position of an atomic centroid point, and calculating the distance and the included angle between all adjacent atomic centroid points to construct a distance spectrogram and carry out fitting to obtain at least four characteristic peak values, and calculating the distance and the nearest included angle between two nearest neighbor atomic centroid points in the initial image; solving the drift distance of the sample by using a global least square method through the drift distortion model; and performing pixel position compensation on the initial image based on the drift distance, and rotating the image to the same scanning coordinate system for image fusion to obtain a target image. According to the method, the atomic scale resolution level correction can be carried out on the transmission electron microscope image under the condition of unknown lattice constants.

The invention discloses a high squint SAR error estimation method and device, electronic equipment and a storage medium, and belongs to the technical field of radar signal processing, and the method comprises the steps: carrying out the distance spectrum division of a frequency domain echo signal of a to-be-measured target, and obtaining the steering vectors of a plurality of sub-spectrum echo signals; carrying out characteristic decomposition on the covariance matrix between the channels to obtain a noise subspace of each sub-spectrum segment; based on a local error estimation model, according to the noise subspace of the sub-spectrum segment and the steering vector of the echo signal of the sub-spectrum segment, solving an inter-channel error of each sub-spectrum segment; and fusing the inter-channel errors of the sub-spectrum segments to obtain a globally optimal channel error estimation value. According to the method, the frequency spectrum is divided in the distance spectrum section, so that the frequency spectrum broadening degree in each sub-spectrum section is effectively constrained, and the problem of estimation misalignment caused by frequency spectrum broadening in a satellite-borne large squint scene can be solved.

The application provides a wind profile radar spectrum peak estimation method based on a three-parameter dynamic cost function, comprising the following steps: step 1, distance and Doppler spectrum data preprocessing, three-point sliding average smoothing processing; step 2, noise level estimation based on statistics; step 3, candidate spectrum peak selection based on spatial Doppler window smoothing; step 4, dynamic weight calculation based on covariance matrix eigenvalue decomposition; step 5, construction of a three-parameter cost function of a distance spectrum peak value RSP power item, a continuity item and a spectrum width consistency item, and optimization; step 6, output of an optimal spectrum peak trajectory, and formation of a complete wind speed profile trajectory. The application can improve the spectrum peak recognition precision and stability in a complex environment, and the u and v wind correlation coefficients of the estimation method of the application reach 0.880 and 0.879 through the data set disclosed by an atmospheric radiation measurement website.