Method and device for motion compensation for hovering unmanned airborne through-the-wall radar

Abstract

The application provides a motion compensation method and device for a hovering unmanned aerial vehicle-borne through-wall radar. By excavating inherent phase frequency linear characteristics of wall echo under platform motion and each subcarrier, the platform motion estimation problem is converted into a slope estimation problem in the frequency domain, and a closed-form least square estimator is derived to realize robust and unbiased time delay estimation, and then full-band coherent frequency domain compensation is completed. Through compensation, the motion-modulated wall echo is converted from time-varying clutter to approximately time-invariant background signal, and the separability of phase modulation caused by weak breathing is effectively improved. The method does not require additional sensors, does not limit the platform motion form, and avoids iterative optimization.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a motion compensation method, target detection method and device for hovering UAV-borne through-wall radar. Background Technology

[0002] Unmanned aerial vehicle (UAV)-borne through-wall radar extends traditional short-range vital sign detection technology to wide-area, high-altitude, non-contact penetrating sensing, providing new technical means for applications such as disaster relief, counter-terrorism operations, and situational awareness in confined spaces. Unlike ground-based fixed radar, UAVs are susceptible to airflow disturbances during flight, and their actual trajectories often deviate from the preset state. This causes the originally stationary wall echo to transform into time-varying strong clutter, and deeply couples the platform motion with the target's vital signs, thereby masking weak physiological modulation characteristics and severely limiting actual detection performance. How to effectively suppress time-varying wall clutter and achieve reliable separation of vital sign signals from platform motion has become a core problem that urgently needs to be solved for this type of system to move towards engineering applications.

[0003] Hovering UAV-borne through-wall radar provides a feasible means for non-contact vital sign detection in obscured and confined spaces. However, under real-world conditions, platform disturbances cause wall echoes to exhibit strong time-varying characteristics and deep coupling with the phase modulation of weak vital signs, posing significant challenges to through-wall radar systems in terms of motion compensation and signal separation. Existing UAV through-wall radar systems can be broadly categorized into synthetic aperture radar (SAR) mode and hovering mode. Compared to SAR mode, hovering UAV through-wall radar can achieve coherent observations in fixed geometric configurations without requiring precise aperture synthesis, thus exhibiting better robustness to platform motion and higher detection sensitivity for weak vital sign signals.

[0004] Based on UAV-borne through-wall radar target detection technology, existing research has proposed a dual-radar scheme, which separates platform motion by subtracting the reference signal from the ground reflected echo. However, the requirement for strict radar isolation and beam separation limits the practical applicability of this method. In addition, radar self-motion cancellation methods based on reflection from stationary obstacles and motion compensation methods combining inter-frame cross-correlation and range cell alignment have also been studied. However, their performance degrades under conditions of strong clutter, low signal-to-noise ratio, or when there are continuous time delay variations and frequency-related phase errors. Recently, a motion compensation framework customized for rotating UAV radar systems has been reported, but its signal model and compensation strategy are not directly applicable to the fixed-view hovering UAV through-wall radar configuration. Summary of the Invention

[0005] This application proposes a motion compensation method, target detection method, and device for hovering UAV-borne through-wall radar, which can solve one of the problems existing in the background art.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, a motion compensation method for hovering UAV-borne through-wall radar is provided, including:

[0008] Obtain the stepped-frequency echo signal of the stepped-frequency radar transmitted signal;

[0009] The stepped frequency echo signal is pulse-compressed to obtain stepped frequency echo data;

[0010] Based on the energy analysis of the stepped frequency echo data along the distance dimension, the wall distance gate index corresponding to the wall location is determined;

[0011] The wall echo signal is determined by the wall distance to the door index;

[0012] The wall echo signal is subjected to frequency domain transformation to obtain complex frequency domain data of the wall echo signal at each frequency point;

[0013] Extract the complex phase in a linear representation from the complex frequency domain data;

[0014] Solving the phase-frequency domain optimization problem constructed for the complex phase yields a slope-closed-form solution;

[0015] The estimated value of the platform's relative displacement is determined by the closed-form solution of the slope.

[0016] Furthermore, the step frequency echo signal is compensated using the estimated relative displacement of the platform.

[0017] Based on the above technical solution, by exploiting the inherent phase-frequency linearity characteristics of the wall echo and each subcarrier under platform motion, the platform motion estimation problem is transformed into a slope estimation problem in the frequency domain. A closed-form least squares estimator is derived to achieve robust and unbiased time delay estimation, thereby completing coherent frequency domain compensation across the entire frequency band. Through compensation, the motion-modulated wall echo is transformed from time-varying clutter into an approximately time-invariant background signal, effectively improving the separability of phase modulation caused by weak breathing. The proposed method requires no additional sensors, does not restrict the form of platform motion, and avoids iterative optimization.

[0018] In one possible design approach of the first aspect, pulse compression of the stepped frequency echo signal specifically includes:

[0019] The stepped frequency echo signal is pulsed repeatedly over a slow time period. Perform sampling;

[0020] And, utilizing The step-frequency pulse train is compressed using the point-fast discrete inverse Fourier transform to obtain the step-frequency echo data for each distance gate.

[0021] In one possible design approach of the first aspect, based on energy analysis of the stepped frequency echo data along the distance dimension, the wall distance gate index corresponding to the wall location is determined, specifically including:

[0022] Obtain the echo energy index of the step frequency echo data at each time and at each distance gate;

[0023] The echo energy index is averaged over a slow time to obtain the average energy index.

[0024] The wall distance to the door index is determined by the average energy index.

[0025] In one possible design approach of the first aspect, the wall echo signal is determined by the wall distance to the door index, specifically as follows:

[0026] Based on the wall distance gate index, the corresponding step frequency echo data is extracted within the determined wall distance gate and its neighborhood.

[0027] Furthermore, the extracted step-frequency echo data is weighted and synthesized to obtain the wall echo signal.

[0028] In one possible design of the first aspect, the complex phase is expressed as:

[0029]

[0030]

[0031] in, Let f represent the complex phase, k be the frequency, m be the time, b(m) be the slope, and f be the frequency. Indicates the noise phase. and Irrelevant, R wall0 Let be the nominal distance from the phase center of the UAV's onboard antenna to the plane of the wall, and c be the speed of light. Let be the relative displacement value of the platform at the m-th time.

[0032] The phase-frequency domain optimization problem is expressed as:

[0033]

[0034] Solving the phase-frequency domain optimization problem specifically involves using a least squares estimation framework to solve the phase-frequency domain optimization problem.

[0035] From the slope closed solution Determine the estimated relative displacement of the platform Specifically:

[0036]

[0037] The step-frequency echo signal is compensated using the estimated relative displacement of the platform, specifically as follows:

[0038] The radial distance disturbance caused by the projection of UAV jitter onto the line of sight in the step frequency echo signal is eliminated by using the platform relative displacement estimate.

[0039] In one possible design approach of the first aspect, the step-frequency echo signal is modeled using wall echo terms, human body echo terms, and noise terms.

[0040] Secondly, a target detection method for hovering UAV-borne through-wall radar is provided. This method for detecting vital signs is based on the motion compensation method for hovering UAV-borne through-wall radar described above. The method for detecting vital signs includes:

[0041] The compensated step-frequency echo signal is pulse-compressed to obtain range image data;

[0042] In addition, target signal extraction is performed on the distance image data.

[0043] In one possible design approach of the second aspect, target signal extraction is performed on the range image data, specifically including:

[0044] Based on the energy distribution characteristics of the target echo in the range image data, a range gate corresponding to the target position is selected in the range dimension; the echo signal at the corresponding target position range gate that varies with slow time is extracted to form a target slow time signal sequence;

[0045] Furthermore, frequency domain analysis is performed on the target slow time series to obtain the spectral characteristics of the target signal, which are used to characterize the periodic micro-motion characteristics of the target.

[0046] In one possible design approach of the second aspect, the target slow time series is subjected to frequency domain analysis, specifically by performing frequency domain analysis through Fourier transform, time-frequency analysis, or equivalent spectrum estimation methods.

[0047] Thirdly, an electronic device is provided, comprising: a processor, and a memory coupled to the processor, the memory for storing a computer program; the processor for executing the computer program stored in the memory such that the electronic device performs a motion compensation method for hovering UAV-borne through-wall radar as described in any possible implementation of the first aspect, or performs a target detection method for hovering UAV-borne through-wall radar as described in any possible implementation of the second aspect. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is an overall technical flow diagram provided in the embodiments of this application;

[0050] Figure 2 This is a diagram showing the motion estimation results of the unmanned aerial vehicle platform provided in the embodiments of this application;

[0051] Figure 3 These are motion compensation and clutter suppression results provided in the embodiments of this application, wherein (a) is the original echo, (b) is the echo after motion compensation, (c) is the clutter suppression result of the original echo, and (d) is the clutter suppression result of the motion-compensated echo;

[0052] Figure 4 This is a comparison diagram of respiratory spectrum results after motion compensation and clutter suppression provided in the embodiments of this application. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0054] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0056] Hovering UAVs carrying through-wall radar provide a feasible means for non-contact vital sign detection in obscured and confined spaces. However, under real-world conditions, platform disturbances can cause wall echoes to exhibit strong time-varying characteristics and become deeply coupled with the phase modulation of weak vital signs, posing a significant challenge to through-wall radar systems in terms of motion compensation and signal separation.

[0057] This embodiment proposes a motion compensation method based on phase-frequency linear regression for hovering UAV-borne stepped-frequency continuous-wave through-wall radar. This method exploits the inherent phase-frequency linearity characteristics of the wall echoes between the platform and each subcarrier under platform motion, transforming the platform motion estimation problem into a slope estimation problem in the frequency domain. A closed-form least-squares estimator is derived to achieve robust and unbiased time delay estimation, thereby completing coherent frequency domain compensation across the entire frequency band. Through compensation, the motion-modulated wall echoes are transformed from time-varying clutter into an approximately time-invariant background signal, effectively improving the separability of phase modulation caused by weak breathing. The proposed method requires no additional sensors, does not restrict the platform's motion, and avoids iterative optimization.

[0058] like Figure 1 The diagram shown is a technical flowchart of this embodiment. The following is a detailed description of this technical solution.

[0059] Step 1: Original signal model and data acquisition

[0060] Due to its advantages such as high resolution, strong penetration, strong anti-interference ability, and simple hardware implementation, stepped-frequency continuous wave (TFWS) signals are widely used in through-wall radar detection systems. (Stepped-frequency radar transmission signal) It can be represented as:

[0061]

[0062] in, Represents a rectangular window function. Indicates the pulse repetition time. Indicates the duration of the sub-pulse. Indicates the first The frequency of a single sub-pulse is given by t, where t is time. Assume the duration of a single sub-pulse is... If the displacement of the inner platform is negligible, then at slow time... The receiver echo can be modeled as a superposition of wall echo, human body echo, and noise:

[0063]

[0064] Where c is the speed of light. Indicates the distance from the drone to the wall. This indicates the distance from the drone to the target behind the wall. Indicates fast time noise. These are the scattering coefficients of the wall and the human body, respectively, and they typically satisfy... This means that the amplitude of wall clutter components is usually much stronger than the echo of human vital signs, making it extremely easy to mask micro-motion signals. (Regarding the formula...) Perform mixing and demodulation to obtain the first... Echoes at each frequency point:

[0065]

[0066] In the formula, This represents the slow-time noise after demodulation.

[0067] Will , Substituting into the above equation, we get:

[0068]

[0069] In the formula, ΔR w This indicates the additional equivalent propagation distance caused by the wall. This indicates the nominal distance from the phase center of the UAV's onboard antenna to the plane of the wall. The radial distance disturbance is caused by the projection of drone jitter onto the line of sight. This represents the initial distance between the phase center of the UAV-borne antenna and the scattering center of the human chest cavity. This represents the radial distance disturbance caused by the projection of slight movements of the human chest cavity onto the line of sight. It can be seen that the wall echo term includes... Caused slow-time phase modulation The wall clutter exhibits significant slow-time non-stationarity. Furthermore, the human echo term includes not only micro-motion modulations caused by vital signs... It also multiplied by a phase term related to platform jitter. This indicates that the platform disturbance and vital signs are coupled in phase, and since the amplitude of the platform disturbance is usually greater than the amplitude of the chest cavity micromovement, the phase fluctuations it causes will dominate the slow time change, so that the vital signs modulation is masked by strong clutter and platform disturbance together.

[0070] Step 2, Pulse Compression

[0071] To improve the distance resolution and signal-to-noise ratio of the echo signal, pulse compression of the stepped-frequency pulse train is required. In this embodiment, the echo signal is... The pulse repetition cycle occurs over a slow period of time. Sampling was performed, and the following methods were used: The point-fast discrete inverse Fourier transform is used to compress the step-frequency pulse train, and at the 1st... At time 1, the pulse compression... Step frequency echo data of a distance gate This can be expressed as:

[0072]

[0073] Step 3: Wall Echo Extraction

[0074] Since walls are typically located between the radar and the target, their echoes are characterized by fixed range and position, high energy, and high temporal stability, while the echoes from targets such as humans have relatively weak energy that changes over time. To provide a stable reference for subsequent platform motion estimation and phase compensation, this embodiment extracts the wall echoes from the pulse-compressed range image.

[0075] In this embodiment, energy analysis is performed along the distance dimension on the pulse compressed echo data obtained in step 2 at each slow-time sampling moment. Based on the echo amplitude or energy distribution characteristics within the distance gate, the distance gate index corresponding to the wall location is determined. Preferably, the robustness of wall echo localization can be improved by statistical averaging or consistency judgment of multiple slow-time sampling points. The following explanation uses a formula as an example:

[0076] Definition of the first Time, distance gate The echo energy index is:

[0077]

[0078] To improve robustness, statistical fusion can be performed across multiple slow-time sampling points. For example, slow-time averaging of the energy yields an average energy index.

[0079]

[0080] In the formula, Indicates the slow-time sampling point. Wall distance to door index. It can be determined by the maximum energy criterion:

[0081]

[0082] In the formula, This forms a candidate range gate set. Subsequently, echo data is extracted from the determined wall range gates and their neighborhoods. To reduce the impact of range sidelobes and inter-gate drift, echo data can be extracted within the wall range gate neighborhoods. ( The neighborhood half-width is used for weighted synthesis to obtain the wall echo signal. :

[0083]

[0084] in, For the neighboring region The weighting coefficients of each distance gate.

[0085] The wall echo signal mainly reflects the change in the propagation path between the radar platform and the wall, and can be used for subsequent frequency domain phase characteristic analysis and platform motion parameter estimation.

[0086] Step 4: Obtaining frequency domain data of the wall

[0087] To obtain the phase change characteristics of the wall echo at different frequency points, this embodiment constructs a frequency domain characterization based on the extracted wall echo signal.

[0088] In this embodiment, the wall echo data obtained in step 3 is used. The system is processed according to its step frequency sequence, and the complex frequency domain data of the wall echo at each frequency point are obtained through discrete Fourier transform or equivalent frequency domain transform. .

[0089]

[0090] The frequency domain data includes amplitude and phase information of the wall echo, reflecting the frequency correlation characteristics of the propagation path between the radar platform and the wall. Through the above processing, the wall echo can be mapped from the time or range domain to the frequency domain, providing basic data support for subsequent motion parameter estimation and phase compensation based on the frequency-phase relationship.

[0091] Step 5: Frequency-phase linear regression motion estimation and compensation

[0092] Based on the wall echo data extracted in step 4, observational Extract the first Time, Number Complex phase of the wall echo at each frequency point:

[0093]

[0094] In the formula, Indicates the noise phase, which is related to Irrelevant Let be the relative displacement value of the platform at time m. It can be seen that the phase and frequency of the wall components have a strictly linear relationship, which can be written in the form of a standard linear model:

[0095]

[0096] in, .Mode This indicates that the platform motion does not introduce arbitrary phase fluctuations in the frequency domain, but rather induces a slowly time-varying slope in the phase-frequency domain. This finding is crucial because it establishes a one-to-one and well-state mapping between the propagation delay caused by the platform and the phase-frequency slope. Therefore, the platform motion estimation problem can be equivalently transformed into a slope estimation problem in the frequency domain, providing a structurally robust theoretical foundation for motion compensation.

[0097] Based on The platform motion estimation can be simplified from each of the first... Identifying the slope in noisy phase observations at all frequencies at time t. Since wall echoes typically exhibit a high signal-to-noise ratio and strong inter-frequency coherence, utilizing frequency domain redundancy is crucial for achieving robust and accurate motion estimation. Therefore, we employ a least-squares estimation framework, derived from the linear model itself, which combines statistical efficiency with computational simplicity.

[0098] For each fixed number At that moment, And solve the following problems:

[0099]

[0100] make right Since the partial derivative is 0, we can obtain:

[0101]

[0102]

[0103] Define the mean Substitute it into the formula ,get:

[0104]

[0105] in For the center frequency, The average phase of the wall.

[0106] Further substitution Finally, the estimated slope is obtained. Closed-form solution

[0107]

[0108] This expression shows that the estimated slope is obtained by projecting the measured phase vector onto the frequency axis.

[0109] once Once determined, the two-way distance from the wall to the phase center of the platform antenna is... This can be derived. Using time zero as a reference, define... Then we have:

[0110]

[0111] Therefore, it is not necessary to know the exact absolute distance of the wall, and thus:

[0112]

[0113] in, This represents the estimated relative displacement of the platform at time m.

[0114] Finally, based on the echo Phase terms are introduced to compensate for motion at each frequency:

[0115]

[0116] Using the estimated The phase term generated by the platform displacement ΔR(m) is used to eliminate the jitter. This can be compared with equation (4), which does not contain ΔR(m).

[0117] As a result, the dominant wall echo is transformed from a time-varying clutter component into an approximately time-invariant background signal. This compensation also decouples the platform motion from the micro-motion phase modulation caused by the target, laying the foundation for reliable vital sign extraction in subsequent processing stages.

[0118] Step 6, Pulse Compression

[0119] After completing motion estimation and phase compensation based on the frequency domain phase characteristics of the wall, in order to further improve the distance focusing performance and signal-to-noise ratio of the echo signal, this embodiment performs pulse compression processing on the compensated echo data again.

[0120] In this embodiment, the stepped frequency echo data after phase compensation in step 5 is processed according to the system stepped frequency sequence. By performing discrete inverse Fourier transform on the frequency domain data, the compensated frequency domain echo is remapped to the range domain, thereby obtaining the motion-compensated range image data.

[0121] The pulse compression process described above can effectively eliminate the effect of phase mismatch caused by platform motion on distance focusing, providing high-quality input data for subsequent clutter suppression and target signal extraction.

[0122] Step 7: Clutter Suppression

[0123] Because echo signals in through-wall detection scenarios typically contain clutter components such as wall clutter, static scattering from the environment, and multipath interference, these clutter components exhibit strong correlation and stability in the slow time dimension, easily interfering with target signal extraction and parameter estimation. To further highlight the target echo characteristics, this embodiment performs clutter suppression processing on the range image data after obtaining motion-compensated and pulse-compressed data.

[0124] In this embodiment, after motion compensation and pulse compression, the range image data is processed in the slow time dimension by taking advantage of the slow or near-static nature of clutter signals. Static and quasi-static clutter components are suppressed by background estimation and elimination, time high-pass filtering, or equivalent clutter suppression methods, while retaining the target echo components that change over time.

[0125] The above-mentioned clutter suppression processing can effectively reduce the impact of residual echoes from the wall and environmental clutter on subsequent processing, significantly improve the contrast and signal-to-noise ratio of the target signal, and provide cleaner input data for target signal extraction and spectral feature analysis.

[0126] Step 8: Target signal extraction and target spectrum estimation

[0127] After clutter suppression processing, the echo data mainly retains the target echo components that change over time. To obtain the target's motion characteristics and physiological micro-motion information, this embodiment performs target signal extraction and spectrum analysis on the clutter-suppressed range image data.

[0128] In this embodiment, based on the energy distribution characteristics of the target echo in the range profile, a range gate corresponding to the target position is selected in the range dimension. The echo signal at this range gate that varies with slow time is extracted to form a target slow-time signal sequence. The target slow-time signal reflects the minute motion changes of the target in the radial direction. The following explanation uses a formula as an example:

[0129] After completing the clutter suppression process, let the processed range image data be... Based on the energy distribution characteristics of the target echo in the range dimension, the range gate index of the target is selected. For example, using the maximum energy criterion:

[0130]

[0131] in, This is the target candidate range gate set. The echo signal at each target range gate, varying with slow time, is extracted to form the target slow time series.

[0132]

[0133] Subsequently, frequency domain analysis is performed on the target slow time series, and the spectral characteristics of the target signal are obtained through Fourier transform, time-frequency analysis, or equivalent spectral estimation methods.

[0134]

[0135] in, For the target frequency The spectral values ​​at the specified location. These spectral features can be used to characterize the periodic micro-motion properties of the target, including but not limited to frequency components related to vital signs such as breathing and heartbeat, or other micro-motion characteristics.

[0136] Through the above-mentioned target signal extraction and spectrum estimation processing, stable acquisition of the micro-motion characteristics of the target can be achieved in complex through-wall detection environments, providing a reliable basis for target detection, identification and state assessment.

[0137] To further demonstrate the effectiveness of this embodiment, the following supplementary explanation is based on simulation results.

[0138] The simulation uses an SFCW radar system with a signal bandwidth of 510 MHz, containing 256 sub-pulses, with a start frequency of 1 GHz and a frequency step of 2 MHz. The simulation duration is 10 s, and the PRT of the echo signal is 0.1 s. In a Cartesian coordinate system, the UAV platform is located at the origin, and its motion is simulated using a sinusoidal model. The vibration amplitude and frequency in all three directions are set to 0.1 m and 2 Hz, respectively. The wall is parallel to the azimuth direction, with a thickness of 0.3 m and a relative permittivity of 6. The front surface of the wall is located 1 m in the range direction. A human target is placed 4 m behind the wall in the scenario, with a breathing frequency and amplitude of 0.3 Hz and 0.01 m, respectively. The simulation SNR is set to 10 dB.

[0139] The results of the drone motion estimation are as follows As shown. The method proposed in this embodiment can stably track this periodic motion, and its estimated curve almost coincides with the true value, with a corresponding RMSE of only [value missing]. m indicates that the time delay regression based on the frequency-phase linear relationship can make full use of the consistency constraints of the entire frequency point of SFCW, thereby significantly suppressing the errors caused by noise and cross-distance gate migration.

[0140] The results of motion compensation and clutter suppression are as follows: As shown. (a) shows the original echo signal. Due to the attenuation of electromagnetic waves in the wall, the front wall reflected signal has the maximum amplitude, and the RCS of the micro-moving human target is low, so the target signal is completely submerged in the wall clutter. (c) shows that after clutter suppression, the SCNR is -73.5 dB, and wall clutter still dominates. (b) illustrates the method proposed in this embodiment, which estimates the platform motion delay using the echo frequency-phase linearity characteristic and compensates for each frequency point in the frequency domain. Through this method, the platform's motion error is more accurately modeled and compensated, and the time-varying characteristics of the wall echo are effectively eliminated. The compensated clutter suppression results are as follows: As shown in (d), the target signal located at 5m is highlighted.

[0141] After motion compensation and clutter suppression are completed using the method in this embodiment, the target phase is extracted from the echo and spectral analysis is performed to obtain the following results: The normalized amplitude spectrum is shown. The black dashed line in the figure marks the true respiratory frequency of 0.3 Hz. It can be seen that the PFLRMC forms a sharp, amplitude-dominant peak at 0.3 Hz, with highly concentrated energy, indicating that the method in this embodiment can still effectively recover weak vital sign components even under strong front wall reflection and platform vibration. In contrast, the spectral energy of the original signal is mainly distributed in the low-frequency band and on a broadband noise floor, lacking a significantly identifiable peak near the target frequency, indicating that residual platform motion and time-varying wall clutter still dominate the phase sequence, causing the target respiratory component to be submerged.

[0142] This application also provides an electronic device, including: a processor, and a memory coupled to the processor, the memory being used to store a computer program; the processor being used to execute the computer program stored in the memory, so that the electronic device performs the method as described in any of the above embodiments.

[0143] Electronic devices can be computing devices such as desktop computers, laptops, handheld computers, and cloud servers. These electronic devices may include, but are not limited to, processors and memory.

[0144] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting various parts of the device via various interfaces and lines.

[0145] The memory can be used to store the computer program, and the processor implements various functions of the electronic device by running or executing the computer program stored in the memory and calling the data stored in the memory.

[0146] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0147] This application also provides a storage medium, which is a computer-readable storage medium. The computer program is stored in the computer-readable storage medium, and when executed by a processor, the computer program can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0148] This application also provides a computer program product, including: a computer program or instructions that, when the computer program or instructions are run on a computer, cause the computer to perform any of the above possible implementation methods.

[0149] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications are also considered to be within the scope of protection of this application.

Claims

1. A motion compensation method for hovering UAV-borne through-wall radar, characterized in that, include: Obtain the stepped-frequency echo signal of the stepped-frequency radar transmitted signal; The stepped frequency echo signal is pulse-compressed to obtain stepped frequency echo data; Based on the energy analysis of the stepped frequency echo data along the distance dimension, the wall distance gate index corresponding to the wall location is determined; The wall echo signal is determined by the wall distance to the door index; The wall echo signal is subjected to frequency domain transformation to obtain complex frequency domain data of the wall echo signal at each frequency point; Extract the complex phase in a linear representation from the complex frequency domain data; Solving the phase-frequency domain optimization problem constructed for the complex phase yields a slope-closed-form solution; The estimated relative displacement of the platform is determined by the closed-form solution of the slope; and, The step-frequency echo signal is compensated using the estimated relative displacement of the platform. The complex phase is expressed as: in, Let f represent the complex phase, k be the frequency, m be the time, b(m) be the slope, and f be the frequency. Indicates the noise phase. and Irrelevant, R wall0 Let be the nominal distance from the phase center of the UAV's onboard antenna to the plane of the wall, and c be the speed of light. Let m be the relative displacement value of the platform at time m. The phase-frequency domain optimization problem is expressed as: Solving the phase-frequency domain optimization problem specifically involves: The phase-frequency domain optimization problem is solved using a least-squares estimation framework. From the slope closed solution Determine the estimated relative displacement of the platform Specifically: ， The step-frequency echo signal is compensated using the estimated relative displacement of the platform, specifically as follows: The radial distance disturbance caused by the projection of UAV jitter onto the line of sight in the step frequency echo signal is eliminated by using the platform relative displacement estimate.

2. The motion compensation method for hovering UAV-borne through-wall radar as described in claim 1, characterized in that, Pulse compression of the stepped frequency echo signal specifically includes: The stepped frequency echo signal is pulsed repeatedly over a slow time period. Perform sampling; and, use The step-frequency pulse train is compressed using the point-fast discrete inverse Fourier transform to obtain the step-frequency echo data for each distance gate.

3. The motion compensation method for hovering UAV-borne through-wall radar as described in claim 1, characterized in that, Based on the energy analysis of the stepped frequency echo data along the distance dimension, the wall distance gate index corresponding to the wall location is determined, specifically including: Obtain the echo energy index of the step frequency echo data at each time and at each distance gate; The echo energy index is averaged over a slow time to obtain the average energy index. The wall distance to the door index is determined by the average energy index.

4. The motion compensation method for hovering UAV-borne through-wall radar as described in claim 1, characterized in that, The wall echo signal is determined based on the wall distance to the door index, specifically as follows: Based on the wall distance gate index, the corresponding step frequency echo data is extracted within the determined wall distance gate and its neighborhood; and, The extracted step-frequency echo data are weighted and synthesized to obtain the wall echo signal.

5. The motion compensation method for hovering UAV-borne through-wall radar as described in claim 1, characterized in that, The stepped frequency echo signal is modeled using wall echo term, human body echo term, and noise term.

6. A target detection method for hovering UAV-borne through-wall radar, characterized in that, The vital sign detection method for hovering UAV-borne through-wall radar is based on the motion compensation method for hovering UAV-borne through-wall radar as described in any one of claims 1-5, and the vital sign detection method for hovering UAV-borne through-wall radar includes: The compensated step-frequency echo signal is pulse-compressed to obtain range image data; and, Target signal extraction is performed on the distance image data.

7. The target detection method for hovering UAV-borne through-wall radar as described in claim 6, characterized in that, Extracting the target signal from the range image data specifically includes: Based on the energy distribution characteristics of the target echo in the range image data, a range gate corresponding to the target position is selected in the range dimension; the echo signal at the corresponding target position range gate that varies with slow time is extracted to form a target slow-time signal sequence; and, Frequency domain analysis is performed on the target slow time series to obtain the spectral characteristics of the target signal, which are used to characterize the periodic micro-motion characteristics of the target.

8. The target detection method for hovering UAV-borne through-wall radar as described in claim 7, characterized in that, The target slow time series is subjected to frequency domain analysis, specifically by using Fourier transform, time-frequency analysis, or equivalent spectrum estimation methods.

9. An electronic device, characterized in that, The electronic device includes: a processor, and a memory coupled to the processor. The memory is used to store computer programs; The processor is configured to execute the computer program stored in the memory, such that the electronic device performs the motion compensation method for hovering UAV-borne through-wall radar as described in any one of claims 1-5, or the target detection method for hovering UAV-borne through-wall radar as described in any one of claims 6-8.

Images