Method for calibrating radar echo signal on mobile platform

Abstract

The invention relates to the technical field of radar signal processing, and discloses a method for calibrating radar echo signals on a mobile platform, which comprises the following steps of: acquiring original echo signals of a radar on the mobile platform under different working conditions and preprocessing the original echo signals, acquiring multi-source IMU sensor and differential GPS information and performing time sequence fusion; carrying out platform motion parameter estimation on the preprocessed echo signal; and based on a preset reference signal model, performing feature extraction on the echo signal, and obtaining feature parameters for calibration through time-frequency processing and spatial domain adaptive filtering. Multi-source IMU (inertial measurement unit) data and differential GPS (global positioning system) data are fused to obtain six-degree-of-freedom continuous motion parameters of a mobile platform, and time synchronization and preprocessing of echo signals are combined to realize dynamic calibration of original radar echo signals. And the problem of amplitude, phase and Doppler frequency deviation of the echo signal caused by the attitude and the motion state of the mobile platform is solved.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, specifically a calibration method for radar echo signals on a mobile platform. Background Technology

[0002] The calibration method for radar echo signals on mobile platforms belongs to the field of radar signal processing. It mainly involves deploying a radar system on a mobile platform to receive electromagnetic wave signals reflected from targets. This method collects radar echo signals and combines them with the platform's attitude, motion state, and environmental information to perform time synchronization, preprocessing, and analysis on the signals to extract characteristic parameters such as amplitude, phase, and Doppler frequency. This provides basic data support for subsequent target detection, tracking, imaging, or navigation and positioning. Typically, multi-source information such as multi-channel radar arrays, inertial measurement units, and differential GPS is used. The echo signals are processed through time-frequency analysis and spatial filtering to acquire signal characteristics and perform preliminary correction, thereby meeting the radar system's requirements for signal reliability and continuity in complex environments.

[0003] In existing technologies, the attitude and motion state of mobile platforms change continuously during operation, but radar echo signals are usually only simply collected and roughly processed, resulting in systematic deviations in amplitude, phase and Doppler frequency of the echo signals, leading to insufficient accuracy in target tracking and signal analysis. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a calibration method for radar echo signals on mobile platforms, which solves the problem of systematic deviations in amplitude, phase, and Doppler frequency of echo signals, leading to insufficient accuracy in target tracking and signal analysis.

[0005] To achieve the above objectives, the present invention provides a method for calibrating radar echo signals on a mobile platform, comprising the following steps:

[0006] The system collects and preprocesses raw echo signals from radar on a mobile platform under different operating conditions, acquires multi-source IMU sensor and differential GPS information and performs time-series fusion, and estimates platform motion parameters from the preprocessed echo signals.

[0007] Based on a preset reference signal model, feature extraction is performed on the echo signal, and feature parameters for calibration are obtained through time-frequency processing and spatial adaptive filtering.

[0008] By combining the attitude information, motion state data, and environmental feature data of the mobile platform, error modeling, deviation analysis, and target tracking trajectory optimization are performed on the feature parameters.

[0009] Based on the error modeling and deviation analysis results, a calibration factor is generated, and the original echo signal is dynamically calibrated.

[0010] The calibrated echo signal is compared and verified with the reference signal model, and the calibration factor is iteratively updated based on the verification results.

[0011] By adopting the above technical solution, the six-degree-of-freedom continuous motion parameters of the mobile platform under different attitudes and motion states can be acquired in real time and correlated with the original radar echo signal. Through feature extraction, error modeling and calibration factor generation, dynamic calibration of the echo signal amplitude, phase and Doppler frequency can be achieved, thereby eliminating the systematic deviation caused by the platform motion and making the echo signal more accurate and reliable.

[0012] Preferably, the acquisition and preprocessing of the raw echo signal includes the following steps:

[0013] A radar receiving channel or array unit is deployed on a mobile platform to continuously collect the returned electromagnetic wave signals according to a preset sampling rate.

[0014] The acquired radar echo signals are normalized in amplitude and phase to reduce signal differences between channels.

[0015] A high-precision synchronous clock is used to synchronize the echo signal to ensure that the echo signal is time-aligned with the pulse data transmitted by the radar.

[0016] The echo signal is denoised, and inter-pulse alignment and amplitude-phase filtering are performed to generate a pre-processed echo signal.

[0017] Preferably, the acquisition of multi-source IMU sensor and differential GPS information and the subsequent time-series fusion includes the following steps:

[0018] Multiple inertial measurement units are deployed on the mobile platform, each unit being used to measure the platform's linear acceleration and angular velocity;

[0019] Data acquisition and synchronization of the output signals of each inertial measurement unit are performed to ensure time alignment of each channel;

[0020] Acquire differential GPS positioning signals and synchronously collect them with inertial measurement unit data;

[0021] The inertial measurement unit data and differential GPS data are fused to obtain the platform's six degrees of freedom motion parameters, including pitch, roll, yaw, heave, sway, and swell.

[0022] The fused motion parameters are filtered and interpolated to form a composite signal with a continuous time series.

[0023] Preferably, the estimation of platform motion parameters includes the following steps:

[0024] The preprocessed echo signal is correlated with the fused composite signal to form the time and platform attitude information corresponding to each echo data point;

[0025] Platform motion parameters are estimated from the correlated data to generate six-degree-of-freedom motion parameters for a continuous time series.

[0026] Preferably, the feature extraction includes the following steps:

[0027] Short-time Fourier transform is performed on the preprocessed echo signal to obtain echo spectrum data at multiple times and frequencies;

[0028] The improved STFT-CLEAN algorithm is used to process the echo spectrum data and remove the extended clutter spectrum caused by platform motion.

[0029] The processed echo spectrum data is compared with a preset reference signal model to extract amplitude, phase and Doppler frequency characteristic parameters.

[0030] By combining multi-channel radar beam data, a spatial adaptive filter bank is constructed, and the filter weights are adjusted to obtain characteristic parameters in the target direction.

[0031] Preferably, obtaining the characteristic parameters for calibration includes the following steps:

[0032] The echo signal data, after time-frequency processing and spatial adaptive filtering, is statistically analyzed to generate amplitude, phase, and Doppler frequency characteristic matrices.

[0033] The feature matrix is ​​organized according to time series and beam sequence to form a structured dataset for calibration processing;

[0034] The feature dataset is normalized and standardized to ensure that the feature parameters of different channels and different time points are consistent. The processed feature parameters are then labeled and stored along with the reference signal model.

[0035] Preferably, the error modeling, deviation analysis, and target tracking trajectory optimization of the feature parameters include the following steps:

[0036] The obtained characteristic parameters for calibration are compared with the reference signal model to establish an error model, and the amplitude, phase and Doppler frequency deviation are recorded.

[0037] Statistical analysis of the error model is performed to generate a sequence of deviation information for each target echo.

[0038] The deviation information is correlated with the attitude information and motion state data of the mobile platform to form a joint deviation dataset;

[0039] Construct a target trajectory prediction model and input joint bias data into the prediction model to update the trajectory points;

[0040] Correlation analysis and optimization of target tracks are performed to form a continuous time series of target tracking tracks.

[0041] Preferably, generating the calibration factor includes the following steps:

[0042] Based on the error model and deviation information, the correction parameters corresponding to each echo signal are calculated;

[0043] The calculated correction parameters are normalized and structured to form a set of calibration factors that can be directly applied to the echo signal.

[0044] The calibration factor is associated with the corresponding echo signal channel and time point to form a complete calibration factor data mapping;

[0045] The calibration factor data is filtered and interpolated to generate a continuous time series of calibration factors.

[0046] Preferably, the dynamic calibration process includes the following steps:

[0047] The generated calibration factor is applied to the original echo signal to perform amplitude and phase correction on the echo signal between each pulse;

[0048] Phase compensation processing is performed on the echo signal to ensure that the phase relationship between consecutive pulses remains consistent;

[0049] The corrected echo signal is subjected to false alarm suppression processing, and clutter false detection is reduced through track association and target screening methods;

[0050] The processed echo signal is synchronously correlated with characteristic parameters and platform motion data to form a continuous echo data sequence that can be used for target tracking.

[0051] Preferably, the iterative update includes the following steps:

[0052] The calibrated echo signal is matched with the reference signal model to generate an error dataset;

[0053] Based on the error dataset, the calibration factors are adjusted to generate new calibration factors;

[0054] The new calibration factor is applied to subsequent echo signal processing to perform closed-loop updates;

[0055] The calibration factor and echo signal data during the iterative update process are recorded and marked to form continuous time series data;

[0056] The updated calibration factor data is correlated with the characteristic parameters and platform motion parameters to provide input for the next cycle of dynamic calibration.

[0057] This invention provides a method for calibrating radar echo signals on a mobile platform. It offers the following advantages:

[0058] 1. In this invention, by fusing multi-source IMU inertial measurement unit data with differential GPS data, the six-degree-of-freedom continuous motion parameters of the mobile platform are obtained. Combined with time synchronization and preprocessing of the echo signal, dynamic calibration of the original radar echo signal is achieved, thus solving the problem of amplitude, phase and Doppler frequency deviation of the echo signal caused by the attitude and motion state of the mobile platform.

[0059] 2. In this invention, the extended clutter spectrum is removed by using short-time Fourier transform combined with an improved STFT-CLEAN algorithm on the preprocessed radar echo signal. A spatial adaptive filter bank is constructed by combining multi-channel radar beam data. The amplitude, phase and Doppler frequency characteristic parameters are extracted by adjusting the filter weights, thereby realizing a structured calibration feature dataset, which provides sufficient data support for error modeling and calibration factor generation.

[0060] 3. In this invention, a calibration factor generation method based on error modeling and deviation analysis is used, and closed-loop iterative updates are achieved based on the comparison between the calibrated echo signal and the reference signal model. Through amplitude and phase correction, phase compensation and false alarm suppression, continuous time series calibration echo data are formed. At the same time, the updated calibration factor is associated with characteristic parameters and platform motion parameters to provide input for the next cycle of dynamic calibration, thereby achieving continuous closed-loop adaptive adjustment. Attached Figure Description

[0061] Figure 1 This is a flowchart of the calibration method for radar echo signals on the mobile platform of the present invention;

[0062] Figure 2 This is a flowchart of the original echo signal acquisition process for the calibration method of radar echo signals on the mobile platform of the present invention.

[0063] Figure 3 This is a flowchart illustrating the feature extraction process of the calibration method for radar echo signals on a mobile platform according to the present invention.

[0064] Figure 4 This is a flowchart illustrating the error modeling and trajectory optimization process for the radar echo signal calibration method on the mobile platform of this invention.

[0065] Figure 5 This is a flowchart illustrating the calibration factor generation and dynamic calibration process of the calibration method for radar echo signals on the mobile platform of this invention. Detailed Implementation

[0066] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] Please see the appendix Figure 1 - Appendix Figure 5 This invention provides a method for calibrating radar echo signals on a mobile platform, comprising the following steps:

[0068] The system collects and preprocesses raw echo signals from radar on a mobile platform under different operating conditions, acquires multi-source IMU sensor and differential GPS information and performs time-series fusion, and estimates platform motion parameters from the preprocessed echo signals.

[0069] Furthermore, acquiring and preprocessing the raw echo signal includes the following steps:

[0070] A radar receiving channel or array unit is deployed on a mobile platform to continuously collect the returned electromagnetic wave signals according to a preset sampling rate.

[0071] The acquired radar echo signals are normalized in amplitude and phase to reduce signal differences between channels.

[0072] A high-precision synchronous clock is used to synchronize the echo signal to ensure that the echo signal is time-aligned with the pulse data transmitted by the radar.

[0073] The echo signal is denoised, and inter-pulse alignment and amplitude-phase filtering are performed to generate a pre-processed echo signal.

[0074] Furthermore, acquiring multi-source IMU sensor and differential GPS information and performing time-series fusion includes the following steps:

[0075] Multiple inertial measurement units are deployed on the mobile platform, each unit being used to measure the platform's linear acceleration and angular velocity;

[0076] Data acquisition and synchronization of the output signals of each inertial measurement unit are performed to ensure time alignment of each channel;

[0077] Acquire differential GPS positioning signals and synchronously collect them with inertial measurement unit data;

[0078] The inertial measurement unit data and differential GPS data are fused to obtain the platform's six degrees of freedom motion parameters, including pitch, roll, yaw, heave, sway, and swell.

[0079] The fused motion parameters are filtered and interpolated to form a composite signal with a continuous time series.

[0080] Furthermore, the estimation of platform motion parameters includes the following steps:

[0081] The preprocessed echo signal is correlated with the fused composite signal to form the time and platform attitude information corresponding to each echo data point;

[0082] Platform motion parameters are estimated from the correlated data to generate six-degree-of-freedom motion parameters for a continuous time series.

[0083] Specifically, the calibration method for radar echo signals on a mobile platform includes steps such as raw echo signal acquisition, acquisition of multi-source IMU sensor and differential GPS information, time synchronization and preprocessing, and estimation of platform motion parameters.

[0084] In the specific implementation process, radar receiving channels or array units are first deployed on the mobile platform to continuously acquire the returned electromagnetic wave signals. During the acquisition process, a preset sampling rate is used to ensure the temporal continuity and frequency domain integrity of the echo data. The acquired echo signals are normalized in amplitude and phase to eliminate amplitude and phase differences between different channels or different pulses. A high-precision synchronization clock, such as a pulse synchronization signal or a precision time protocol clock, is used to achieve time alignment of data from each channel and multiple pulses. The echo signals are then denoised, including digital filtering and inter-pulse alignment, to ensure the consistency of the continuous echo sequence. Subsequently, the pre-processed echo signals are labeled with the platform motion parameters to provide a complete data mapping for subsequent calibration and feature extraction.

[0085] Multiple inertial measurement units (IMUs) are deployed on the mobile platform, each measuring linear acceleration and angular velocity, forming a multi-channel data acquisition network. IMU output signals are synchronously acquired through the data acquisition module to ensure time alignment across channels. Differential GPS positioning signals are also acquired and synchronously integrated with the IMU data. The IMU data and differential GPS data are fused using an extended Kalman filter or unscented Kalman filter algorithm to generate the platform's six-degree-of-freedom motion parameters, including pitch, roll, yaw, heave, sway, and surge. The fused motion parameters are then filtered and interpolated to form a continuous time series, providing platform motion information for subsequent echo signal processing. Specifically, the following formula can be used:

[0086] ;

[0087] in, Indicates time The platform's six-degree-of-freedom state vector includes position, attitude angle, and their derivatives; For IMU measurement input vector, For differential GPS observation vectors, and These are process noise and observation noise, respectively; function and These are the state transition function and the observation function, respectively; the fusion algorithm achieves accurate mapping between echo data and platform motion state.

[0088] The acquired radar echo signals undergo time synchronization, preliminary preprocessing, and platform motion parameter estimation. First, the echo signals are timestamped using a high-precision synchronization clock to ensure time consistency across channels and multiple pulses. Then, amplitude and phase normalization are performed to eliminate systematic differences between channels and pulses. Noise denoising is then applied, including digital filtering, inter-pulse alignment, and amplitude and phase filtering, generating a continuous and stable echo sequence. Finally, the preprocessed echo signals are correlated with the fused IMU and differential GPS motion parameters by precisely timestamping each echo pulse sampling point and correlating it with the corresponding pitch, roll, and yaw parameters. The motion parameters of roll, pitch, heave, sway, and pitch are matched and written into the echo data structure to form a composite signal with motion attributes. This allows for accurate compensation of amplitude, phase, and Doppler frequency deviations caused by platform attitude and motion state during subsequent feature extraction, error modeling, and dynamic calibration. Each echo data point corresponds to time and platform attitude information, realizing the mapping relationship between the echo signal and the platform motion state. Motion parameters are estimated from the correlated data to generate six-degree-of-freedom motion parameters for a continuous time series, providing complete basic data for subsequent feature extraction, error modeling, and dynamic calibration. In this process, the synchronous mapping between the echo signal and motion parameters can be expressed as:

[0089] ;

[0090] in, This represents the original echo signal. Let the platform's motion state vector be... For time synchronization and preprocessed echo signals, This represents synchronization and preprocessing functions, including operations such as amplitude and phase normalization, denoising, pulse alignment, and data marking.

[0091] This embodiment achieves complete acquisition of radar echo signals, sensor information acquisition and fusion, time synchronization, preprocessing, and platform motion parameter estimation on the mobile platform through the above process. It provides a complete data foundation and processing conditions for subsequent feature extraction, error modeling, calibration factor generation, and dynamic calibration processing, while ensuring data continuity, time domain alignment, and inter-channel consistency.

[0092] Based on a preset reference signal model, features are extracted from the echo signal, and feature parameters for calibration are obtained through time-frequency processing and spatial adaptive filtering.

[0093] Furthermore, feature extraction includes the following steps:

[0094] Short-time Fourier transform is performed on the preprocessed radar echo signal to obtain echo spectrum data at multiple times and frequencies;

[0095] The improved STFT-CLEAN algorithm is used to process the echo spectrum data and remove the extended clutter spectrum caused by platform motion.

[0096] The processed echo spectrum data is compared with a preset reference signal model to extract amplitude, phase and Doppler frequency characteristic parameters.

[0097] By combining multi-channel radar beam data, a spatial adaptive filter bank is constructed, and the filter weights are adjusted to obtain characteristic parameters in the target direction.

[0098] Furthermore, obtaining the characteristic parameters used for calibration includes the following steps:

[0099] The echo signal data, after time-frequency processing and spatial adaptive filtering, is statistically analyzed to generate amplitude, phase, and Doppler frequency characteristic matrices.

[0100] The feature matrix is ​​organized according to time series and beam sequence to form a structured dataset for calibration processing;

[0101] The feature dataset is normalized and standardized to ensure that the feature parameters of different channels and different time points are consistent. The processed feature parameters are then labeled and stored along with the reference signal model.

[0102] Specifically, based on a preset reference signal model, features are extracted from the echo signal, and feature parameters for calibration are obtained through time-frequency processing and spatial adaptive filtering.

[0103] In the specific implementation process, the preprocessed radar echo signal is first subjected to a short-time Fourier transform to obtain echo spectrum data at multiple times and frequencies. The short-time Fourier transform can be expressed as:

[0104]

[0105] in Indicates the echo signal at time point The sampled values, This represents the sliding window function. Indicates the center position of the window. Indicates frequency index, The window length is indicated; the sliding window function can be either a Hamming window or a Gaussian window, and the window width and overlap rate can be adjusted according to the characteristics of the echo signal to balance time resolution and frequency resolution;

[0106] Subsequently, the improved STFT-CLEAN algorithm was used to process the echo spectrum data. This algorithm iteratively updates the residual spectrum and removes the extended clutter spectrum caused by platform motion, while simultaneously correcting the amplitude and phase of the remaining spectrum signal to accurately characterize the Doppler frequency and amplitude-phase information of the target signal. In the algorithm implementation, an initial residual spectrum can be selected. Then through iterative formulas Update the residual spectrum, in which Indicates the first The clutter spectral components removed in the next iteration These are the convergence control coefficients; during the iteration process, the termination condition can be determined based on the spectral amplitude threshold and frequency stability.

[0107] The processed echo spectrum data is compared with a preset reference signal model, and feature parameters are extracted through amplitude, phase, and Doppler frequency, among which amplitude features are... It can be expressed as the magnitude of the spectral amplitude and phase characteristics. Represented as complex spectrum phase value, Doppler frequency characteristics The spectral phase can be calculated using the rate of change of spectral phase over time. By combining multi-channel radar beam data, a spatial adaptive filter bank is constructed. The weights of each filter are adjusted using covariance matrix estimation and the minimum variance criterion to obtain characteristic parameters in the target direction. ,in This is the signal vector after beamforming. Here is the filter weight vector. This indicates the conjugate transpose operation; weight calculation can be solved by an optimization problem that minimizes the output power while maintaining the desired signal response constraint.

[0108] The echo signal data, after time-frequency processing and spatial adaptive filtering, is statistically analyzed to generate amplitude, phase, and Doppler frequency feature matrices. These matrices are then organized according to time series and beam sequences to form a structured dataset for calibration processing. The feature dataset is normalized and standardized to ensure consistency of feature parameters across different channels and time points, for example, through... For the characteristic matrix Standardization is carried out, among which This represents the mean. The standard deviation is represented; the processed feature parameters and reference signal model are labeled and stored, and associated with the platform motion parameters by channel and time index to form a complete data mapping relationship, realizing a continuous mapping from the original echo signal to the feature parameters and the platform motion state, providing basic data for subsequent calibration processing;

[0109] The entire feature extraction and feature parameter generation process transforms the echo signal from its original time-domain information into time-frequency and spatial domain representations through the above-mentioned steps, and associates it with motion parameters and reference models, thereby realizing data structuring and parameterization. This provides directly applicable feature data for subsequent error modeling, deviation analysis, and calibration factor generation.

[0110] By combining the attitude information, motion state data, and environmental feature data of the mobile platform, error modeling, deviation analysis, and target tracking trajectory optimization are performed on the feature parameters.

[0111] Furthermore, error modeling, deviation analysis, and target tracking trajectory optimization include the following steps:

[0112] The obtained characteristic parameters for calibration are compared with the reference signal model to establish an error model, and the amplitude, phase and Doppler frequency deviation are recorded.

[0113] Statistical analysis of the error model is performed to generate a sequence of deviation information for each target echo.

[0114] The deviation information is correlated with the attitude information and motion state data of the mobile platform to form a joint deviation dataset;

[0115] Construct a target trajectory prediction model and input joint bias data into the prediction model to update the trajectory points;

[0116] Correlation analysis and optimization of target tracks are performed to form a continuous time series of target tracking tracks.

[0117] Specifically, by combining the attitude information, motion state data, and environmental feature data of the mobile platform, error modeling, deviation analysis, and target tracking trajectory optimization are performed on the extracted feature parameters.

[0118] In the specific implementation process, an error model is first established and the difference information is recorded by comparing characteristic parameters such as amplitude, phase, and Doppler frequency with a reference signal model. In this process, the error parameter can be defined as: amplitude error. Phase error Doppler frequency error ,in These represent the observed amplitude, phase, and Doppler frequency parameters, respectively. These are the corresponding parameters under the reference signal model; in this way, error data records can be established for each target echo.

[0119] After error modeling is completed, statistical analysis is performed on the error model to generate a sequence of deviation information corresponding to the target echo. This statistical analysis includes calculating the mean and variance of the error of the same target in a continuous time series. For example, the phase error can be calculated as follows:

[0120] , ;

[0121] in Indicates the number of sampling points. For the first Phase deviation values ​​for each sample; This is the mean of the phase deviation, used to reflect the overall average deviation level; The variance of the phase deviation is used to reflect the degree of dispersion of the phase deviation. Through a similar method, the statistical characteristics of the amplitude error and the Doppler frequency error can be obtained respectively. These statistical characteristics form a sequence of deviation information for each target in the time and frequency domain, providing a quantitative description for subsequent optimization processing.

[0122] Furthermore, the deviation information is correlated with the attitude information and motion state data of the mobile platform to form a joint deviation dataset; the platform attitude information includes pitch angle, roll angle, and yaw angle, and the motion state data includes velocity vector and acceleration vector; assuming the platform is at time... The attitude state vector is:

[0123] ;

[0124] in For the platform at all times The pitch angle, that is, the rotation angle around the horizontal axis; For the platform at all times The roll angle, which is the angle of rotation about the longitudinal axis; For the platform at all times The heading angle, i.e., the rotation angle around the vertical axis; the motion state vector is:

[0125] ;

[0126] in For the platform at all times along Velocity in the axial direction; For the platform at all times along Velocity in the axial direction; For the platform at all times along Velocity in the axial direction;

[0127] By establishing mapping relationships The original error data can be used Transformed into joint deviation data associated with the platform's dynamic state. This allows us to distinguish between deviations caused by platform motion and deviations related to environmental factors;

[0128] After obtaining the joint deviation dataset, a target trajectory prediction model is constructed, and the joint deviation data is input into the model for waypoint updates. The target trajectory prediction model can employ the Kalman filter method, and its state update equation can be expressed as follows: The observation equation is expressed as ,in Indicates the first The target state vector at any given time, including position and velocity information. Indicates control input, Here is the state transition matrix. For the input control matrix, For process noise, Represents the observation vector. For the observation matrix, To address observation noise, during the update process, the joint bias dataset is used as a correction input to the characteristics of the observation noise, thereby improving the consistency between the trajectory prediction results and the actual target state.

[0129] The target trajectory is analyzed and optimized to form a continuous time series target tracking trajectory. First, the observation points of the same target at different times are matched using a data association algorithm, such as the nearest neighbor method or the joint probability data association method. After the matching is completed, the discrete trajectory points are interpolated and optimized using a smoothing filter method to ensure the continuity and stability of the trajectory. The resulting target tracking trajectory contains both the continuity information of the time series and the platform attitude, motion state and environmental characteristic parameters, so it can be directly applied in subsequent calibration and dynamic compensation, providing a reliable target motion representation for the overall system.

[0130] Based on the error modeling and deviation analysis results, a calibration factor is generated, and the original echo signal is dynamically calibrated.

[0131] Furthermore, generating the calibration factor includes the following steps:

[0132] Based on the error model and deviation information, the correction parameters corresponding to each echo signal are calculated;

[0133] The calculated correction parameters are normalized and structured to form a set of calibration factors that can be directly applied to the echo signal.

[0134] The calibration factor is associated with the corresponding echo signal channel and time point to form a complete calibration factor data mapping;

[0135] The calibration factor data is filtered and interpolated to generate a continuous time series of calibration factors.

[0136] Furthermore, the dynamic calibration process includes the following steps:

[0137] The generated calibration factor is applied to the original echo signal to perform amplitude and phase correction on the echo signal between each pulse;

[0138] Phase compensation processing is performed on the echo signal to ensure that the phase relationship between consecutive pulses remains consistent;

[0139] The corrected echo signal is subjected to false alarm suppression processing, and clutter false detection is reduced through track association and target screening methods;

[0140] The processed echo signal is synchronously correlated with characteristic parameters and platform motion data to form a continuous echo data sequence that can be used for target tracking.

[0141] Specifically, based on the results of the aforementioned error modeling and deviation analysis, a calibration factor is generated, and the generated calibration factor is used to perform dynamic calibration processing on the original echo signal to achieve amplitude and phase consistency correction and false alarm suppression of the echo signal.

[0142] In the specific implementation process, firstly, based on the error model and deviation information, the corresponding correction parameters are calculated for each echo signal sample; assuming that in time... The echo signal obtained at time is Its amplitude and phase are respectively expressed as and Then the amplitude deviation and phase deviation corresponding to the signal can be expressed as: and Based on the deviation modeling results, amplitude correction parameters are defined. and phase correction parameters They are respectively:

[0143] ;

[0144] in, Indicates the reference signal model at time... The range, Indicates the reference signal model at time... The phase; the parameters obtained through the above calculations. and The correction parameter pair that constitutes the echo signal;

[0145] Next, the calculated set of calibration parameters is normalized and structured; specifically, for calibration parameters of different channels and different time points, a matrix is ​​used to organize them and construct a set of calibration factors:

[0146] ;

[0147] in, Indicates the number of channels. and The first Each channel at time The corresponding amplitude correction factor and phase correction factor; after normalization, this matrix serves as a set of factors directly applied to signal calibration;

[0148] Based on this, discrete calibration factor data are smoothed using time series interpolation and low-pass filtering methods to generate continuous time series calibration factors. This process ensures the smoothness of the factors between adjacent pulses and avoids calibration jumps caused by instantaneous errors.

[0149] During the dynamic calibration process, the generated calibration factors are first mapped and associated with the corresponding channels and time points of the original echo signals; for each pulse of the echo signal... Amplitude and phase correction are applied:

[0150] ;

[0151] in, This represents the calibrated complex signal; the calibrated signal maintains consistency with the reference model in both amplitude and phase.

[0152] Furthermore, to ensure phase consistency between consecutive pulses, phase error is cumulatively compensated; specifically, the phase difference between two adjacent pulses is defined as... After calibration, the phase compensation function is used. Make corrections, among which As a compensation term, its calculation is derived from the smoothing trend of the calibration factor sequence;

[0153] False alarm suppression processing is performed on the corrected echo signal; this step, based on track association and target screening methods, removes clutter detection results that fail to form a stable track within a continuous time window; for example, this can be achieved by setting the length of the sliding time window. Only retain target points that appear continuously within the time window and are consistent with the platform motion data;

[0154] The processed echo signal is synchronously associated with the previously obtained feature parameters and platform motion data; through a unified time index and channel index, the multi-source data is fused into a continuous echo data sequence; this sequence can be directly used for subsequent target tracking, state estimation and multi-sensor information fusion processing.

[0155] The calibrated echo signal is compared and verified with the reference signal model, and the calibration factor is iteratively updated based on the verification results.

[0156] Furthermore, the iterative update includes the following steps:

[0157] The calibrated echo signal is matched with the reference signal model to generate an error dataset;

[0158] Based on the error dataset, the calibration factors are adjusted to generate new calibration factors;

[0159] The new calibration factor is applied to subsequent echo signal processing to perform closed-loop updates;

[0160] The calibration factor and echo signal data during the iterative update process are recorded and marked to form continuous time series data;

[0161] The updated calibration factor data is correlated with the characteristic parameters and platform motion parameters to provide input for the next cycle of dynamic calibration.

[0162] Specifically, after completing the dynamic calibration process, the calibrated echo signal is compared and verified with the reference signal model, and the calibration factor is iteratively updated based on the comparison results to achieve closed-loop adjustment of the calibration parameters in the time dimension.

[0163] In the specific implementation process, the calibrated echo signal is first matched with the reference signal model; assuming at time... The obtained calibrated signal is represented as Its amplitude and phase are respectively and The amplitude and phase of the reference signal model at this moment are respectively and The residual error between the two can be expressed as:

[0164] ;

[0165] in, For amplitude residuals, The phase residual is used to generate an error dataset by calculating the residuals for all channels and all time points. Its structural form is as follows:

[0166] ;

[0167] in, Indicates the number of channels. and The first Each channel at time The amplitude and phase residuals;

[0168] Based on this error dataset, the calibration factors are iteratively adjusted; let the amplitude and phase calibration factors obtained in the previous iteration be respectively... and Then in the first In this iteration, the new calibration factor is calculated using the following recursive formula:

[0169]

[0170]

[0171] in, and The step size factor is used to iteratively update the step size and control the update magnitude. Indicates at time Reference amplitude; Indicates at time The actual amplitude deviation, Indicates at time The phase deviation; through iterative calculations, a new set of calibration factors is generated. And applied to subsequent echo signal processing;

[0172] During the iterative update process, the calibration factors and corresponding echo signal data are recorded and marked. Specifically, a time series index is established to bind the calibration factors generated at each moment with the corresponding signal data, forming continuous time series data. This data not only includes the evolution trajectory of the amplitude and phase calibration factors, but also records the correction results of the signal under different iteration cycles.

[0173] Furthermore, the iteratively updated calibration factor data is associated with the aforementioned feature parameters and platform motion parameters. In terms of implementation, data alignment can be achieved through a unified timestamp and channel number, so that the calibration factor corresponds to the platform attitude, motion state, and environmental characteristics. This process provides input conditions for the next cycle of dynamic calibration, realizing a closed-loop update mechanism based on echo signals, calibration factors, and platform motion.

[0174] Through the above processing, the calibration factor is no longer a static parameter, but a continuous function that is dynamically adjusted with time and environment. This mechanism ensures that the entire system can maintain the adaptability and consistency of the calibration factor during long-term operation and provides a stable signal basis for subsequent target detection and trajectory tracking.

[0175] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A calibration method for radar echo signals on a mobile platform, characterized in that, Includes the following steps: The system collects and preprocesses raw echo signals from radar on a mobile platform under different operating conditions, acquires multi-source IMU sensor and differential GPS information and performs time-series fusion, and estimates platform motion parameters from the preprocessed echo signals. Based on a preset reference signal model, feature extraction is performed on the echo signal, and feature parameters for calibration are obtained through time-frequency processing and spatial adaptive filtering. By combining the attitude information, motion state data, and environmental feature data of the mobile platform, error modeling, deviation analysis, and target tracking trajectory optimization are performed on the feature parameters. Based on the error modeling and deviation analysis results, a calibration factor is generated, and the original echo signal is dynamically calibrated. The calibrated echo signal is compared and verified with the reference signal model, and the calibration factor is iteratively updated based on the verification results.

2. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The process of acquiring and preprocessing the raw echo signal includes the following steps: A radar receiving channel or array unit is deployed on a mobile platform to continuously collect the returned electromagnetic wave signals according to a preset sampling rate. The acquired radar echo signals are normalized in amplitude and phase to reduce signal differences between channels. A high-precision synchronous clock is used to synchronize the echo signal to ensure that the echo signal is time-aligned with the pulse data transmitted by the radar. The echo signal is denoised, and inter-pulse alignment and amplitude-phase filtering are performed to generate a pre-processed echo signal.

3. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The acquisition of multi-source IMU sensor and differential GPS information and its time-series fusion includes the following steps: Multiple inertial measurement units are deployed on the mobile platform, each unit being used to measure the platform's linear acceleration and angular velocity; Data acquisition and synchronization of the output signals of each inertial measurement unit are performed to ensure time alignment of each channel; Acquire differential GPS positioning signals and synchronously collect them with inertial measurement unit data; The inertial measurement unit data and differential GPS data are fused to obtain the platform's six degrees of freedom motion parameters, including pitch, roll, yaw, heave, sway, and swell. The fused motion parameters are filtered and interpolated to form a composite signal with a continuous time series.

4. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The estimation of platform motion parameters includes the following steps: The preprocessed echo signal is correlated with the fused composite signal to form the time and platform attitude information corresponding to each echo data point; Platform motion parameters are estimated from the correlated data to generate six-degree-of-freedom motion parameters for a continuous time series.

5. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The feature extraction process includes the following steps: Short-time Fourier transform is performed on the preprocessed echo signal to obtain echo spectrum data at multiple times and frequencies; The improved STFT-CLEAN algorithm is used to process the echo spectrum data and remove the extended clutter spectrum caused by platform motion. The processed echo spectrum data is compared with a preset reference signal model to extract amplitude, phase and Doppler frequency characteristic parameters. By combining multi-channel radar beam data, a spatial adaptive filter bank is constructed, and the filter weights are adjusted to obtain characteristic parameters in the target direction.

6. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, Obtaining the characteristic parameters used for calibration involves the following steps: The echo signal data, after time-frequency processing and spatial adaptive filtering, is statistically analyzed to generate amplitude, phase, and Doppler frequency characteristic matrices. The feature matrix is ​​organized according to time series and beam sequence to form a structured dataset for calibration processing; The feature dataset is normalized and standardized to ensure that the feature parameters of different channels and different time points are consistent. The processed feature parameters are then labeled and stored along with the reference signal model.

7. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The error modeling, deviation analysis, and target tracking trajectory optimization for the aforementioned feature parameters include the following steps: The obtained characteristic parameters for calibration are compared with the reference signal model to establish an error model, and the amplitude, phase and Doppler frequency deviation are recorded. Statistical analysis of the error model is performed to generate a sequence of deviation information for each target echo. The deviation information is correlated with the attitude information and motion state data of the mobile platform to form a joint deviation dataset; Construct a target trajectory prediction model and input joint bias data into the prediction model to update the trajectory points; Correlation analysis and optimization of target tracks are performed to form a continuous time series of target tracking tracks.

8. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The generation of calibration factors includes the following steps: Based on the error model and deviation information, the correction parameters corresponding to each echo signal are calculated; The calculated correction parameters are normalized and structured to form a set of calibration factors that can be directly applied to the echo signal. The calibration factor is associated with the corresponding echo signal channel and time point to form a complete calibration factor data mapping; The calibration factor data is filtered and interpolated to generate a continuous time series of calibration factors.

9. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The dynamic calibration process includes the following steps: The generated calibration factor is applied to the original echo signal to perform amplitude and phase correction on the echo signal between each pulse; Phase compensation processing is performed on the echo signal to ensure that the phase relationship between consecutive pulses remains consistent; The corrected echo signal is subjected to false alarm suppression processing, and clutter false detection is reduced through track association and target screening methods; The processed echo signal is synchronously correlated with characteristic parameters and platform motion data to form a continuous echo data sequence that can be used for target tracking.

10. The calibration method for radar echo signals on a mobile platform according to claim 1, characterized in that, The iterative update includes the following steps: The calibrated echo signal is matched with the reference signal model to generate an error dataset; Based on the error dataset, the calibration factors are adjusted to generate new calibration factors; The new calibration factor is applied to subsequent echo signal processing to perform closed-loop updates; The calibration factor and echo signal data during the iterative update process are recorded and marked to form continuous time series data; The updated calibration factor data is correlated with the feature parameters and platform motion parameters to provide input for the next cycle of dynamic calibration.

Images