WO2021134449A1 - 一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 - Google Patents
一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 Download PDFInfo
- Publication number
- WO2021134449A1 WO2021134449A1 PCT/CN2019/130437 CN2019130437W WO2021134449A1 WO 2021134449 A1 WO2021134449 A1 WO 2021134449A1 CN 2019130437 W CN2019130437 W CN 2019130437W WO 2021134449 A1 WO2021134449 A1 WO 2021134449A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- detection
- target
- targets
- distance
- speed
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/06—Systems determining position data of a target
- G01S13/08—Systems for measuring distance only
- G01S13/32—Systems for measuring distance only using transmission of continuous waves, whether amplitude-, frequency-, or phase-modulated, or unmodulated
- G01S13/34—Systems for measuring distance only using transmission of continuous waves, whether amplitude-, frequency-, or phase-modulated, or unmodulated using transmission of continuous, frequency-modulated waves while heterodyning the received signal, or a signal derived therefrom, with a locally-generated signal related to the contemporaneously transmitted signal
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/50—Systems of measurement based on relative movement of target
- G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
Definitions
- the invention belongs to the technical field of radar detection, and particularly relates to a method, a device, a computer equipment and a storage medium for detecting a weak signal of an FMCW array radar moving multiple targets under strong clutter.
- FMCW frequency modulated carrier wave
- FMCW radar For FMCW radar, detecting weak signals of multiple moving targets in a strong clutter background is another difficult point. Strong clutter environments, especially non-uniform clutter environments, bring new challenges to the constant false alarm detection of moving weak signals. Many FMCW radar applications face the above problems. Take the current very hot vehicle radar as an example. The environment it faces is often more complicated (urban streets, many motor vehicles, non-motor vehicles, and pedestrians), and there are various clutters. (Including non-uniform clutter), there are multiple targets, and the echo signals of many targets are often very weak (pedestrians, etc.). Many authors have proposed some methods to improve the detection performance of FMCW radars, but many are based on triangular waves.
- Triangular waves need to match the distance and speed of the target, which in itself restricts the detection ability of weak targets and easily introduces false targets. Therefore, the problem of weak signal detection in motion under the background of strong clutter has not been well resolved. In engineering practice, the low false alarm rate and low false alarm rate of weak signal detection cannot be taken into account, resulting in low detection and recognition efficiency.
- the purpose of the present invention is to provide a method and device for detecting weak signal of FMCW array radar moving multiple targets under strong clutter, which effectively improves the detection performance of weak signal targets, and takes into account the low false alarm rate and low false alarm rate of weak signal detection. , The detection and recognition efficiency is high.
- the FMCW array radar moving multi-target weak signal detection method under strong clutter in an embodiment of the first aspect of the present invention uses sawtooth signal modulation, and the method includes:
- the first detection adopts an ordered statistic constant false alarm detection
- the third detection uses array information for spatial spectrum estimation, and uses the difference between the real target and the false target spatial spectrum estimation peak value to eliminate the false target;
- the first detection specifically includes:
- a preset number of protection units and calculation units are set on both sides of the distance unit, and then the calculation units are sorted in descending order;
- a comparator is used to compare the size of the measured distance unit with the threshold value of the first detection, and when the measured distance unit is greater than the threshold value, it is determined as the first target.
- the second detection specifically includes:
- the initial threshold is increased by a preset search step value as the comparison threshold for the next detection search.
- the number of second targets is lower than the preset condition, Decrease the initial threshold by a preset search step value as the comparison threshold for the next detection search; repeat the above comparison and judgment until the number of second targets meets the preset condition.
- the third detection specifically includes:
- the true and false target is judged on the MUSIC spectrum, and when the MUSIC spectrum peak is greater than the preset judgment threshold, it is judged as the third target, otherwise it is judged as the non-target.
- the preset decision threshold is 5-10 times the average value of all spectral peaks of MUSIC.
- the following expressions are used to calculate the distance and speed of the second target, wherein the calculation of the distance adopts the expression Among them, c represents the speed of light, T c represents a single frequency modulation cycle, F s represents the sampling rate, M bin represents the FFT unit where the target distance is located, B represents the signal bandwidth, and N FFT represents the distance FFT points; the calculation speed adopts the expression
- V bin represents the FFT unit where the target speed is located
- f represents the frequency of the transmitted signal
- K represents the number of frequency modulation cycles of the velocity measurement cycle
- T f represents the velocity measurement cycle.
- the covariance matrix Eigen decomposition has Wherein, U S is a vector from the largest eigenvalue corresponding feature spanned signal subspace, and U N is a value corresponding features small feature vectors spanned noise subspace; characteristic value using characteristic decomposition of the source number Estimate.
- the angle estimation is achieved by the minimum optimization search using the expression Calculate MUSIC spectrum using expression
- the frequency modulation working bandwidth of the FMCW array radar satisfies the expression Among them, d res represents the range resolution, c represents the speed of light, and B represents the frequency modulation working bandwidth; the single frequency modulation period of the FMCW array radar satisfies the expression Among them, V max represents the maximum measurable speed, T c represents a single frequency modulation period, and ⁇ represents the wavelength of the transmitted signal; the speed measurement period of the FMCW array radar satisfies the expression Among them, V res represents the velocity resolution, T f represents the velocity measurement period, and ⁇ represents the wavelength of the transmitted signal; the sampling rate of the FMCW array radar satisfies the expression Among them, F s represents the sampling rate, d max represents the maximum detectable distance, T c represents a single frequency modulation cycle, c represents the speed of light, and B represents the frequency modulation working bandwidth.
- the FMCW array radar moving multi-target weak signal detection device under strong clutter of the embodiment of the second aspect of the present invention, the FMCW array radar adopts sawtooth signal modulation, and the device includes:
- the Fourier transform module is used to perform fast Fourier transform processing on the difference frequency signals of different array elements and different frequency modulation periods to obtain the range spectra of different targets;
- the cancellation processing module is used to cancel the distance spectrum to eliminate the clutter signal generated by the stationary target;
- the first detection module is configured to perform the first detection on the data after the cancellation process to obtain the first target; the first detection adopts the ordered statistics constant false alarm detection;
- the second detection module is configured to perform a second detection on the first target so that the number of second targets obtained after the second detection meets a preset condition; the number of the second targets is smaller than the first The number of goals;
- the third detection module is used to perform a third detection on the second target to obtain a third target.
- the third detection uses array information for spatial spectrum estimation, and eliminates the difference between the real target and the false target spatial spectrum estimation peak False target
- the output module is used to output the distance, speed and angle information of the third target.
- the first detection module is specifically configured to:
- a preset number of protection units and calculation units are set on both sides of the distance unit, and then the calculation units are sorted in descending order;
- a comparator is used to compare the size of the measured distance unit with the threshold value of the first detection, and when the measured distance unit is greater than the threshold value, it is determined as the first target.
- the second detection module is specifically configured to:
- the initial threshold is increased by a preset search step value as the comparison threshold for the next detection search.
- the number of second targets is lower than the preset condition, Decrease the initial threshold by a preset search step value as the comparison threshold for the next detection search; repeat the above comparison and judgment until the number of second targets meets the preset condition.
- the third detection module is specifically configured to:
- the true and false target is judged on the MUSIC spectrum, and when the MUSIC spectrum peak is greater than the preset judgment threshold, it is judged as the third target, otherwise it is judged as the non-target.
- the preset decision threshold is 5-10 times the average value of all spectral peaks of MUSIC.
- the following expressions are used to calculate the distance and speed of the second target, wherein the calculation of the distance adopts the expression Among them, c represents the speed of light, T c represents a single frequency modulation cycle, F s represents the sampling rate, M bin represents the FFT unit where the target distance is located, B represents the signal bandwidth, and N FFT represents the distance FFT points; the calculation speed adopts the expression
- V bin represents the FFT unit where the target speed is located
- f represents the frequency of the transmitted signal
- K represents the number of frequency modulation cycles of the velocity measurement cycle
- T f represents the velocity measurement cycle.
- the covariance matrix Eigen decomposition has Wherein, U S is a vector from the largest eigenvalue corresponding feature spanned signal subspace, and U N is a value corresponding features small feature vectors spanned noise subspace; characteristic value using characteristic decomposition of the source number Estimate.
- the angle estimation is achieved by the minimum optimization search using the expression Calculate MUSIC spectrum using expression
- the frequency modulation working bandwidth of the FMCW array radar satisfies the expression Among them, d res represents the range resolution, c represents the speed of light, and B represents the frequency modulation working bandwidth; the single frequency modulation period of the FMCW array radar satisfies the expression Among them, V max represents the maximum measurable speed, T c represents a single frequency modulation period, and ⁇ represents the wavelength of the transmitted signal; the speed measurement period of the FMCW array radar satisfies the expression Among them, V res represents the velocity resolution, T f represents the velocity measurement period, and ⁇ represents the wavelength of the transmitted signal; the sampling rate of the FMCW array radar satisfies the expression Among them, F s represents the sampling rate, d max represents the maximum detectable distance, T c represents a single frequency modulation cycle, c represents the speed of light, and B represents the frequency modulation working bandwidth.
- the computer device of the embodiment of the third aspect of the present invention includes: one or more processors, a memory, and one or more computer programs, wherein the processor and the memory are connected by a bus, and the one Or a plurality of computer programs are stored in the memory and configured to be executed by the one or more processors, and when the processor executes the computer program, the above-mentioned FMCW array radar under strong clutter is realized Moving multi-target weak signal detection method.
- the storage medium of the embodiment of the fourth aspect of the present invention is used to store a computer program.
- the computer program runs on a computer device, the computer device executes the above-mentioned FMCW array under strong clutter. Radar moving multi-target weak signal detection method.
- This method eliminates the clutter generated by the stationary target through cancellation processing, and then detects the data from which the clutter is removed.
- the detection adopts three detections.
- the first detection ensures that the weak signal can pass the threshold to ensure a low false alarm rate;
- the second detection further narrows the scope and eliminates false targets to reduce the amount of calculation for the third detection;
- third The second detection uses the array information to estimate the spatial spectrum, and uses the difference between the real target and the false target spatial spectrum estimation peak to distinguish, and the remaining false targets can be eliminated.
- the detection performance of weak signal targets can be significantly improved, and the low false alarm rate and low missed alarm rate of weak signal detection are both considered, and the detection and recognition efficiency is high.
- FIG. 1 is a schematic flowchart of a method for detecting weak signals of moving multiple targets of FMCW array radar under strong clutter according to an embodiment of the present invention
- Figure 2 is a block diagram of the structure of the first detection
- Figure 3 is a schematic diagram of the third detection process
- FIG. 4 is a schematic diagram of a module of an FMCW array radar moving multi-target weak signal detection device under strong clutter according to an embodiment of the present invention
- Fig. 5 is a structural block diagram of a computer device provided by an embodiment of the present invention.
- the embodiment of the present invention provides a method for detecting weak signals of FMCW array radar moving multiple targets under strong clutter.
- the FMCW array radar adopts sawtooth signal modulation. As shown in FIG. 1, the method includes:
- S101 Perform fast Fourier transform processing on difference frequency signals of different array elements and different frequency modulation periods to obtain range spectra of different targets.
- the input array difference frequency signal data includes N array elements and K frequency modulation periods, and each frequency modulation period has M sampling points. Perform Fast Fourier Transform (FFT) processing on K frequency modulation periods of N array elements to obtain range spectra of different targets.
- FFT Fast Fourier Transform
- the function of the cancellation processing is to eliminate the clutter generated by the stationary target, leaving only the moving target.
- the clutter cancellation processing can eliminate the clutter signal generated by the stationary target, so that the number of echoes involved in subsequent calculations is reduced, thereby simplifying the following Separate work for multiple moving targets.
- Ordered statistics constant false alarm detection is actually a kind of adaptive threshold detection, which uses reference window samples to estimate the parameters of environmental noise to achieve the characteristic of constant false alarm rate.
- Environmental noise includes clutter and interference.
- the ordered statistics constant false alarm detection has better anti-interference ability than other types of constant false alarm detection.
- the first test specifically includes:
- the distance unit D For each distance unit D, set a preset number of protection units (Pn, Pn+1) and calculation units (X 1 -X N ) on both sides of the distance unit D, and then set the calculation unit Sort from largest to smallest; the preset sequence number can be the number of frequency modulation cycles K that represents the speed measurement cycle.
- the calculation unit Z with the preset sequence number K is selected and multiplied by the product factor T as the threshold value S for the first detection.
- D is the target unit under test. Because the power of the target may leak into the adjacent unit, the several units adjacent to the target are not used as the background clutter estimation, and are used as the protection unit P.
- a comparator is used to compare the size of the measured distance unit D with the threshold value S detected for the first time. When the measured distance unit D is greater than the threshold value S, D is determined as the first target.
- 1024-point FFT is performed on the echo difference frequency signal of a group of FM signals
- 1024 output data are obtained, and data 1-512 are used for detection (because the output of FFT is symmetrical), each data is detected one by one.
- the data represents a frequency point (that is, a certain distance). If the number of protection units is 4 and the number of calculation units is 8, then for the 20th unit, the protection units are 18, 19, 21, 22; the calculation units are 14, 15, 16, 17, 23, 24, 25, and 26.
- the threshold In order to detect the target from the weak signal, the threshold needs to be lowered for the first detection, so as not to miss the real target.
- the first detection is to detect the suspected target from the non-uniform clutter.
- S104 Perform a second detection on the first target so that the number of second targets obtained after the second detection meets a preset condition; the number of the second targets is smaller than the number of the first targets.
- the second detection is to eliminate some false targets. Due to the relatively low requirements of the first ordered statistic constant false alarm detection, some false targets may be mistaken for targets. The second detection further narrows the scope to eliminate false targets.
- the second test specifically includes:
- the mean value of the first target obtained in the first detection is multiplied by the preset coefficient to obtain the initial threshold value of the second detection. Calculate the average value of the threshold target detected for the first time, and multiply the average value by a preset coefficient to obtain the initial threshold value of the second detection search. In order to reduce the false alarm probability, both strong signal and weak signal detection are taken into account.
- the initial fixed threshold can be set to a small point, that is, the preset system can be set to be smaller.
- a comparator is used to compare the size of the measured distance unit with the initial threshold value of the second detection, and when the measured distance unit is greater than the initial threshold value of the second detection, it is determined as the second target.
- the initial threshold is increased by a preset search step value as the comparison threshold for the next detection search.
- the number of second targets is lower than the preset condition, Decrease the initial threshold by a preset search step value as the comparison threshold for the next detection search; repeat the above comparison and judgment until the number of second targets meets the preset condition.
- a preset search step value such as 10%
- the second initial threshold will be lowered by one search step value to continue searching until the target is searched; If the number of targets found in the second search is higher than the preset condition, it means that there are too many false targets, and the second initial threshold is increased by one search step value to search again.
- the preset conditions for the number of targets in the second test can generally be set as needed (depending on the hardware and requirements).
- the vehicle angle radar and the forward radar are obviously different.
- the angle radar has a short range, and the number of targets can be set to a small number (such as 10-20).
- the forward radar has a long range, and the hardware performance has sufficient margin. You can set the target number to a larger point (for example, 30-40).
- the number of targets obtained in the m-1th detection search is higher than the preset condition.
- the judgment threshold needs to be increased.
- the judgment threshold is increased by a preset search step value and then the mth detection search is performed, the target obtained by the detection search is found The number is lower than the preset condition, and the decision threshold needs to be lowered at this time.
- the range of lowering the judgment threshold cannot remain unchanged according to the original adjustment range, otherwise the data will not converge, and the number of targets obtained in the second detection will never meet the preset conditions.
- the adjustment range can be halved. That is, the preset search step value is increased or decreased by half. This method helps the number of second targets obtained in the second detection to converge and quickly meet the preset conditions.
- the preset search step value is 10%.
- the initial threshold of the target is 100.
- the detection threshold is 100, the number of second targets obtained in the second detection is higher than the preset condition, so the detection threshold needs to be increased.
- the number of second targets obtained in the second detection is lower than the preset condition.
- the detection threshold needs to be lowered.
- the target quantity can never meet the preset conditions. Therefore, the preset search step value needs to be halved.
- the third detection uses array information for spatial spectrum estimation, and uses the difference between the real target and the false target spatial spectrum estimation peak value to eliminate the false target.
- the third detection specifically includes the following steps:
- c represents the speed of light
- T c represents a single frequency modulation cycle
- F s represents the sampling rate
- M bin represents the FFT unit where the target distance is located
- B represents the signal bandwidth
- N FFT represents the number of FFT points in the distance.
- V bin represents the FFT unit where the target velocity is located
- f represents the frequency of the transmitted signal
- K represents the number of frequency modulation cycles of the velocity measurement cycle (also the number of velocity FFT points)
- T f represents the velocity measurement cycle (equal to KT c ).
- the above formula relates to the definition of speed in which the target approaching radar speed is positive.
- the array covariance matrix can be composed of the original data S (three-dimensional data: A frame of data has N elements, each element has K frequency modulation periods, and each frequency modulation period has a certain number of sampling points data)
- S FFT N ⁇ K
- N FFT dimension the first time is to perform the sampling point data of a single frequency modulation cycle
- the covariance matrix of the corresponding target is:
- U S is the subspace formed by the feature vector corresponding to the large eigenvalue, that is, the signal subspace
- U N is the subspace formed by the feature vector corresponding to the small eigenvalue, that is, the noise subspace.
- the eigenvalues after eigen decomposition can be used to estimate the number of information sources, and the correct estimation of the number of information sources is the premise of spatial spectrum estimation.
- the angle estimation here is realized by the minimum optimization search, namely
- the preset decision threshold is determined, after a large amount of data simulation and combined with actual measured data, the preset decision threshold can be set to 5-10 times the average of all spectral peaks of MUSIC.
- the false target can be eliminated and the distance, speed and angle of the real target can be obtained.
- the specific steps of the first detection, the second detection, and the third detection are all optimal solutions.
- the specific steps of the first detection, the second detection, and the third detection can also be implemented in other forms to achieve the corresponding effect of each detection.
- sawtooth wave FMCW signal modulation Compared with triangular wave FMCW signal modulation, sawtooth wave FMCW signal modulation has obvious advantages. It does not require distance and speed pairing, and can effectively improve the detection performance of weak signal targets. This application selects the sawtooth wave FMCW signal system.
- the signal waveform and hardware parameters of the FMCW array radar modulated by sawtooth signal need to meet the following conditions:
- d res represents the range resolution
- c represents the speed of light
- B represents the FM working bandwidth of the FMCW array radar. If the distance resolution needs to be within 1 meter, the FM working bandwidth should be greater than 150MHz.
- V max represents the maximum measurable speed
- V res represents the speed resolution
- T c represents a single frequency modulation cycle
- T f represents a speed measurement cycle, which generally includes K frequency modulation cycles. If the wavelength of the transmitted signal is 0.004 meters, the maximum measurable speed (relative radial velocity to the radar) is not less than 50 meters/second, and the speed resolution is not more than 1 meter/second, then T c should not be greater than 20 microseconds, and T f Should not be less than 2 milliseconds.
- sampling rate of the signal should satisfy the following relationship:
- F s represents the sampling rate
- d max represents the maximum detectable distance
- T c represents a single frequency modulation cycle
- c represents the speed of light
- B represents the frequency modulation working bandwidth. If B is 200 MHz, d max is 200 meters, and T c is 20 microseconds, then F s should not be less than 40/3 MHz.
- ⁇ res represents the angular resolution
- N represents the number of array elements
- d represents the distance between the array elements
- the angular resolution is approximately 14.324°.
- the accuracy of single pulse angle measurement can reach 2% of the angle resolution under ideal circumstances.
- there are many factors restricting the angle resolution in actual engineering which can generally reach 10%.
- this angular resolution is not enough.
- High-resolution spatial spectrum estimation provides a solution for small-array high-resolution angle measurement, which can break through the Rayleigh limit and obtain angle measurement accuracy far superior to traditional angle measurement methods.
- This method is used to detect the weak signal of the FMCW array radar moving multi-target under the strong clutter background. It is verified by simulation that the number of array elements is 8, the target signal-to-noise ratio is 0dB, and the clutter-to-noise ratio is 15dB. Between array elements) 0.1+0.1i, element error (within plus or minus 10%), amplitude and phase inconsistency (with amplitude plus or minus 1dB, phase plus or minus 10°)), 100 Monte Carlo simulation detection probability To reach 100%, the mean deviation and variance are not more than 0.01° and 0.02 respectively. This method not only can effectively detect the target (when the signal-to-noise ratio is 0dB, the probability of false alarms and missed alarms is extremely low), it also has high angle measurement accuracy.
- the embodiment of the present invention also provides an FMCW array radar moving multi-target weak signal detection device under strong clutter.
- the FMCW array radar adopts sawtooth signal modulation.
- the device includes:
- the Fourier transform module is used to perform fast Fourier transform processing on the difference frequency signals of different array elements and different frequency modulation periods to obtain the range spectra of different targets.
- the input array difference frequency signal data includes N array elements and K frequency modulation periods, and each frequency modulation period has M sampling points. Perform Fast Fourier Transform (FFT) processing on K frequency modulation periods of N array elements to obtain range spectra of different targets.
- FFT Fast Fourier Transform
- the cancellation processing module is used to cancel the distance spectrum to eliminate the clutter signal generated by the stationary target.
- the function of the cancellation processing is to eliminate the clutter generated by the stationary target, leaving only the moving target.
- the clutter cancellation processing can eliminate the clutter signal generated by the stationary target, and reduce the number of echoes involved in subsequent calculations, thereby simplifying Separation of multiple moving targets in the back.
- the first detection module is configured to perform the first detection on the data after the cancellation processing to obtain the first target; the first detection adopts an ordered statistics constant false alarm detection.
- the first detection module is specifically used for:
- a preset number of protection units and calculation units are set on both sides of the distance unit, and then the calculation units are sorted in descending order;
- a comparator is used to compare the size of the measured distance unit with the threshold value of the first detection, and when the measured distance unit is greater than the threshold value, it is determined as the first target.
- the second detection module is configured to perform a second detection on the first target so that the number of second targets obtained after the second detection meets a preset condition; the number of the second targets is smaller than the first The number of targets.
- the second detection is to eliminate some false targets. Due to the relatively low requirements of the first ordered statistic constant false alarm detection, some false targets may be mistaken for targets. The second detection further narrows the scope to eliminate false targets.
- the second detection module is specifically used for:
- the initial threshold is increased by a preset search step value as the comparison threshold for the next detection search.
- the number of second targets is lower than the preset condition, Decrease the initial threshold by a preset search step value as the comparison threshold for the next detection search; repeat the above comparison and judgment until the number of second targets meets the preset condition.
- the preset search step value needs to be increased or decreased by half. This method is helpful for the number of second targets obtained in the second detection. Meet preset conditions quickly.
- the third detection module is used to perform a third detection on the second target to obtain a third target.
- the third detection uses array information for spatial spectrum estimation, and eliminates the difference between the real target and the false target spatial spectrum estimation peak False target.
- the third detection module is specifically used for:
- c represents the speed of light
- T c represents a single frequency modulation cycle
- F s represents the sampling rate
- M bin represents the FFT unit where the target distance is located
- B represents the signal bandwidth
- N FFT represents the number of FFT points in the distance.
- V bin represents the FFT unit where the target velocity is located
- f represents the frequency of the transmitted signal
- K represents the number of frequency modulation cycles of the velocity measurement cycle (also the number of velocity FFT points)
- T f represents the velocity measurement cycle (equal to KT c ).
- the above formula relates to the definition of speed in which the target approaching radar speed is positive.
- the array covariance matrix can be composed of the original data S (three-dimensional data: A frame of data has N elements, each element has K frequency modulation periods, and each frequency modulation period has a certain number of sampling points data)
- S FFT N ⁇ K
- N FFT dimension the first time is to perform the sampling point data of a single frequency modulation cycle
- the covariance matrix of the corresponding target is:
- Eigen decomposition is performed on the covariance matrix of different targets respectively.
- U S is the subspace formed by the feature vector corresponding to the large eigenvalue, that is, the signal subspace
- U N is the subspace formed by the feature vector corresponding to the small eigenvalue, that is, the noise subspace.
- the eigenvalues after eigen decomposition can be used to estimate the number of information sources, and the correct estimation of the number of information sources is the premise of spatial spectrum estimation.
- the minimum optimization search is used to estimate the angle and calculate the MUSIC spectrum.
- the angle estimation here is realized by the minimum optimization search, namely
- the true and false target is judged on the MUSIC spectrum, and when the MUSIC spectrum peak is greater than the preset judgment threshold, it is judged as the third target, otherwise it is judged as the non-target.
- the preset decision threshold is determined, after a large amount of data simulation and combined with actual measured data, the preset decision threshold can be set to 5-10 times the average of all spectral peaks of MUSIC.
- the output module is used to output the distance, speed and angle information of the third target.
- the false target can be eliminated and the distance, speed and angle of the real target can be obtained.
- the FM working bandwidth of the FMCW array radar satisfies the expression Among them, d res represents the range resolution, c represents the speed of light, and B represents the frequency modulation working bandwidth; the single frequency modulation period of the FMCW array radar satisfies the expression Among them, V max represents the maximum measurable speed, T c represents a single frequency modulation period, and ⁇ represents the wavelength of the transmitted signal; the speed measurement period of the FMCW array radar satisfies the expression Among them, V res represents the velocity resolution, T f represents the velocity measurement period, and ⁇ represents the wavelength of the transmitted signal; the sampling rate of the FMCW array radar satisfies the expression Among them, F s represents the sampling rate, d max represents the maximum detectable distance, T c represents a single frequency modulation cycle, c represents the speed of light, and B represents the frequency modulation working bandwidth.
- the computer device 100 includes: one or more processors 101, a memory 102, and one or more computer programs, wherein the processor 101 Connected to the memory 102 via a bus, the one or more computer programs are stored in the memory 102 and configured to be executed by the one or more processors 101, and the processor 101 executes the
- the computer program implements the steps of the method for detecting the weak signal of the FMCW array radar moving multi-target under strong clutter provided by an embodiment of the present application.
- Computer equipment includes servers and terminals.
- the computer device may be a personal computer, a mobile terminal, or a vehicle-mounted device.
- the mobile terminal includes at least one of a mobile phone, a tablet computer, a personal digital assistant, or a wearable device.
- the present invention also provides a storage medium for storing a computer program.
- the computer program runs on a computer device, the computer device is caused to execute the strong clutter as provided in an embodiment of the present application.
- the steps of the FMCW array radar moving multi-target weak signal detection method are provided in an embodiment of the present application.
- a "computer-readable medium” can be any device that can contain, store, communicate, propagate, or transmit a program for use by an instruction execution system, device, or device or in combination with these instruction execution systems, devices, or devices.
- computer readable media include the following: electrical connections (electronic devices) with one or more wiring, portable computer disk cases (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable and editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
- the computer-readable medium may even be paper or other suitable medium on which the program can be printed, because it can be used, for example, by optically scanning the paper or other medium, followed by editing, interpretation, or other suitable media if necessary. The program is processed in a manner to obtain the program electronically, and then stored in the computer memory.
- each part of the present invention can be implemented by hardware, software, firmware or a combination thereof.
- multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system.
- a suitable instruction execution system For example, if it is implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: Discrete logic circuits, application-specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.
- a person of ordinary skill in the art can understand that all or part of the steps carried in the method of the foregoing embodiments can be implemented by a program instructing relevant hardware to complete.
- the program can be stored in a computer-readable storage medium. When executed, it includes one of the steps of the method embodiment or a combination thereof.
- the functional units in the various embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units may be integrated into one module.
- the above-mentioned integrated modules can be implemented in the form of hardware or software functional modules. If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium.
- the aforementioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
一种强杂波下FMCW阵列雷达运动多目标弱信号检测方法及装置,该方法包括:对不同阵元不同调频周期的差频信号分别进行快速傅里叶变换处理得到不同目标的距离谱(S101);对距离谱进行对消处理以剔除静止目标产生的杂波信号(S102);对对消处理后的数据进行第一次检测得到第一目标,第一次检测采用有序统计量恒虚警检测(S103);对第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件,第二目标的个数小于第一目标的个数(S104);对第二目标进行第三次检测得到第三目标,第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标(S105);输出第三目标的距离、速度和角度信息(S106)。该方法可显著提升强杂波环境下的运动多目标弱信号检测能力。
Description
本发明属于雷达检测技术领域,特别涉及一种强杂波下FMCW阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质。
相对于脉冲多普勒雷达,FMCW(frequency modulated carrier wave,调频连续波)雷达有特殊的优势,其近距离盲区一般不到1米(甚至可以到几厘米),这对于很多需要近距离测距的应用场景来说很有意义。
对于FMCW雷达,在强杂波背景下检测多运动目标的弱信号又是一个难点,强杂波环境,尤其是非均匀杂波环境给运动多弱信号的恒虚警检测带来新的挑战。在很多FMCW雷达的应用领域,都面临上述问题,以当前很热的车载雷达为例,其面临的环境往往比较复杂(城市街道,机动车、非机动车、行人很多),有各种杂波(包括非均匀杂波),有多个目标,很多目标的回波信号往往很弱(行人等)。很多学者提出了一些改善FMCW雷达检测性能的方法,但很多都是基于三角波的方法,三角波需要对目标的距离和速度进行配对,这本身就会制约弱目标的检测能力,容易引入虚假目标。因此,在强杂波背景下的运动弱信号检测问题没有得到很好的解决。工程实际中无法兼顾弱信号检测的低虚警率与低漏警率,导致检测识别效率不高。
本发明的目的在于提出一种强杂波下FMCW阵列雷达运动多目标弱信号检测方法及装置,该方法有效提升弱信号目标的检测性能,兼顾弱信号检测的低虚警率与低漏警率,检测识别效率较高。
为了实现上述目的,本发明第一方面实施例的强杂波下FMCW阵列雷达运动多目标弱信号检测方法,所述FMCW阵列雷达采用锯齿波信号调制,该方法包括:
对不同阵元不同调频周期的差频信号分别进行快速傅里叶变换处理得到不同目标的距离谱;
对所述距离谱进行对消处理以剔除静止目标产生的杂波信号;
对对消处理后的数据进行第一次检测得到第一目标;所述第一次检测采用有序统计量恒虚警检测;
对所述第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件;所述第二目标的个数小于所述第一目标的个数;
对所述第二目标进行第三次检测得到第三目标,所述第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标;
输出所述第三目标的距离、速度和角度信息。
在本发明的一个实施例中,所述第一次检测具体包括:
确定一个乘积因子;
对于每一个距离单元,在所述距离单元的两侧设定预设数量的保护单元和计算单元,然后对所述计算单元按从大到小顺序进行排序;
选定预设序号的计算单元和所述乘积因子相乘作为第一次检测的门限值;
利用比较器比较被测距离单元与所述第一次检测的门限值的大小,当被测距离单元大于所述门限值时,将其确定为第一目标。
在本发明的一个实施例中,所述第二次检测具体包括:
将第一次检测得到的第一目标的均值乘以预设系数得到第二次检测的初始门限值;
利用比较器比较被测距离单元与所述第二次检测的初始门限值的大小,当被测距离单元大于所述第二次检测的初始门限值时,将其确定为第二目标;
当第二目标的数量高于预设条件时,将初始门限值提高一个预设搜索步进值作为下一次检测搜索的比较门限值,当第二目标的数量低于预设条件时,将初始门限值降低一个预设搜索步进值作为下一次检测搜索的比较门限值;重复上述比较判断直至第二目标的数量满足预设条件。
在本发明的一个实施例中,所述第三次检测具体包括:
计算所述第二目标的距离和速度;
计算相应距离和速度的协方差矩阵;
分别对不同目标的协方差矩阵进行特征分解;
以最小优化搜索实现角度估计并计算MUSIC谱;
对所述MUSIC谱进行真假目标判决,当MUSIC谱峰大于预设判决门限时判定为第三目标,否则判定为非目标。
在本发明的一个实施例中,所述预设判决门限值为MUSIC所有谱峰均值的5-10倍。
在本发明的一个实施例中,计算所述第二目标的距离和速度采用如下表达式,其中,计算距离采用表达式
其中,c表示光速,T
c表示单个调频周 期,F
s表示采样率,M
bin表示目标距离所在的FFT单元,B表示信号带宽,N
FFT表示距离FFT点数;计算速度采用表达式
其中,V
bin表示目标速度所在的FFT单元,f表示发射信号频率,K表示测速周期的调频周期数,T
f表示测速周期。
在本发明的一个实施例中,对协方差矩阵
进行特征分解有
其中,U
S是由大特征值对应的特征矢量张成的信号子空间,而U
N是由小特征值对应的特征矢量张成的噪声子空间;利用特征分解后的特征值进行信源数目的估计。
在本发明的一个实施例中,所述FMCW阵列雷达的调频工作带宽满足表达式
其中,d
res表示距离分辨率,c表示光速,B表示调频工作带宽;所述FMCW阵列雷达的单个调频周期满足表达式
其中,V
max表示最大可测速度,T
c表示单个调频周期,λ表示发射信号波长;所述FMCW阵列雷达的测速周期满足表达式
其中,V
res表示速度分辨率,T
f表示测速周期,λ表示发射信号波长;所述FMCW阵列雷达的采样率满足表达式
其中,F
s表示采样率,d
max表示最大可探测距离,T
c表示单个调频周期,c表示光速,B表示调频工作带宽。
为了实现上述目的,本发明第二方面实施例的强杂波下FMCW阵列雷达运动多目标弱信号检测装置,所述FMCW阵列雷达采用锯齿波信号调制,该装置包括:
傅里叶变换模块,用于对不同阵元不同调频周期的差频信号分别进行快速傅里叶变 换处理得到不同目标的距离谱;
对消处理模块,用于对所述距离谱进行对消处理以剔除静止目标产生的杂波信号;
第一检测模块,用于对对消处理后的数据进行第一次检测得到第一目标;所述第一次检测采用有序统计量恒虚警检测;
第二检测模块,用于对所述第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件;所述第二目标的个数小于所述第一目标的个数;
第三检测模块,用于对所述第二目标进行第三次检测得到第三目标,所述第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标;
输出模块,用于输出所述第三目标的距离、速度和角度信息。
在本发明的一个实施例中,所述第一检测模块具体用于:
确定一个乘积因子;
对于每一个距离单元,在所述距离单元的两侧设定预设数量的保护单元和计算单元,然后对所述计算单元按从大到小顺序进行排序;
选定预设序号的计算单元和所述乘积因子相乘作为第一次检测的门限值;
利用比较器比较被测距离单元与所述第一次检测的门限值的大小,当被测距离单元大于所述门限值时,将其确定为第一目标。
在本发明的一个实施例中,所述第二检测模块具体用于:
将第一次检测得到的第一目标的均值乘以预设系数得到第二次检测的初始门限值;
利用比较器比较被测距离单元与所述第二次检测的初始门限值的大小,当被测距离单元大于所述第二次检测的初始门限值时,将其确定为第二目标;
当第二目标的数量高于预设条件时,将初始门限值提高一个预设搜索步进值作为下一次检测搜索的比较门限值,当第二目标的数量低于预设条件时,将初始门限值降低一个预设搜索步进值作为下一次检测搜索的比较门限值;重复上述比较判断直至第二目标的数量满足预设条件。
在本发明的一个实施例中,所述第三检测模块具体用于:
计算所述第二目标的距离和速度;
计算相应距离和速度的协方差矩阵;
分别对不同目标的协方差矩阵进行特征分解;
以最小优化搜索实现角度估计并计算MUSIC谱;
对所述MUSIC谱进行真假目标判决,当MUSIC谱峰大于预设判决门限时判定为第三目标,否则判定为非目标。
在本发明的一个实施例中,所述预设判决门限值为MUSIC所有谱峰均值的5-10倍。
在本发明的一个实施例中,计算所述第二目标的距离和速度采用如下表达式,其中,计算距离采用表达式
其中,c表示光速,T
c表示单个调频周期,F
s表示采样率,M
bin表示目标距离所在的FFT单元,B表示信号带宽,N
FFT表示距离FFT点数;计算速度采用表达式
其中,V
bin表示目标速度所在的FFT单元,f表示发射信号频率,K表示测速周期的调频周期数,T
f表示测速周期。
在本发明的一个实施例中,对协方差矩阵
进行特征分解有
其中,U
S是由大特征值对应的特征矢量张成的信号子空间,而U
N是由小特征值对应的特征矢量张成的噪声子空间;利用特征分解后的特征值进行信源数目的估计。
在本发明的一个实施例中,所述FMCW阵列雷达的调频工作带宽满足表达式
其中,d
res表示距离分辨率,c表示光速,B表示调频工作带宽;所述FMCW阵列雷达的单个调频周期满足表达式
其中,V
max表示最大可测速度,T
c表示单个调频周期,λ表示发射信号波长;所述FMCW阵列雷达的测速周期满足表达式
其中,V
res表示速度分辨率,T
f表示测速周期,λ表示发射信号 波长;所述FMCW阵列雷达的采样率满足表达式
其中,F
s表示采样率,d
max表示最大可探测距离,T
c表示单个调频周期,c表示光速,B表示调频工作带宽。
为了实现上述目的,本发明第三方面实施例的计算机设备,包括:一个或多个处理器、存储器以及一个或多个计算机程序,其中所述处理器和所述存储器通过总线连接,所述一个或多个计算机程序被存储在所述存储器中,并且被配置成由所述一个或多个处理器执行,所述处理器执行所述计算机程序时实现以上所述的强杂波下FMCW阵列雷达运动多目标弱信号检测方法。
为了实现上述目的,本发明第四方面实施例的存储介质,用于存储计算机程序,当所述计算机程序在计算机设备上运行时,使得所述计算机设备执行以上所述的强杂波下FMCW阵列雷达运动多目标弱信号检测方法。
该方法通过对消处理消除静止目标产生的杂波,然后对去掉杂波的数据进行检测。此时的检测采取3次检测,第一次检测确保弱信号能过门限保证低漏警率;第二次检测进一步缩小范围将虚假目标进行剔除以减小第三次检测的运算量;第三次检测利用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异进行区别,可以剔除剩余的虚假目标。经过三次检测,可以显著弱信号目标的检测性能,兼顾弱信号检测的低虚警率与低漏警率,检测识别效率较高。
图1为本发明实施例提供的一种强杂波下FMCW阵列雷达运动多目标弱信号检测方法的流程示意图;
图2为第一次检测的结构框图;
图3为第三次检测的流程示意图;
图4为本发明实施例提供的一种强杂波下FMCW阵列雷达运动多目标弱信号检测方装置的模块示意图;
图5为本发明实施例提供的计算机设备的结构框图。
下面详细描述本发明的实施例,所述实施例的示例在附图中示出,其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的,仅用于解释本发明,而不能解释为对本发明的限制。
在本发明的描述中,需要理解的是,术语“第一”、“第二”等仅用于描述目的,而不能理解为指示或暗示相对重要性。在本发明的描述中,需要说明的是,除非另有明确的规定和限定,术语“相连”、“连接”应做广义理解,例如,可以是机械连接,也可以是电连接;可以是直接相连,也可以通过中间媒介间接相连。对于本领域的普通技术人员而言,可以具体情况理解上述术语在本发明中的具体含义。此外,在本发明的描述中,除非另有说明,“多个”的含义是两个或两个以上。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为,表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分,并且本发明的优选实施方式的范围包括另外的实现,其中可以不按所示出或讨论的顺序,包括根据所涉及的功能按基本同时的方式或按相反的顺序,来执行功能,这应被本发明的实施例所属技术领域的技术人员所理解参照下面的描述和附图,将清楚本发明的实施例的这些和其他方面。在这些描述和附图中,具体公开了本发明的实施例中的一些特定实施方式,来表示实施本发明的实施例的原理的一些方式,但是应当理解,本发明的实施例的范围不受此限制。相反,本发明的实施例包括落入所附加权利要求书的精神和内涵范围内的所有变化、修改和等同物。
本发明实施例提供了一种强杂波下FMCW阵列雷达运动多目标弱信号检测方法,所述FMCW阵列雷达采用锯齿波信号调制,如图1所示,该方法包括:
S101、对不同阵元不同调频周期的差频信号分别进行快速傅里叶变换处理得到不同目标的距离谱。
输入的阵列差频信号数据包括N个阵元K个调频周期,每个调频周期有M个采样点。分别对N个阵元的K个调频周期进行快速傅里叶变换(FFT)处理得到不同目标的距离谱。该距离谱为3维立体数据。
S102、对所述距离谱进行对消处理以剔除静止目标产生的杂波信号。
对消处理的作用是为了消除静止目标产生的杂波,只留下运动目标,杂波对消处理可以剔除静止目标产生的杂波信号,使参与后续运算的回波数量减少,从而简化后面的多运动目标分离工作。
S103、对对消处理后的数据进行第一次检测得到第一目标;所述第一次检测采用有序统计量恒虚警检测。
有序统计量恒虚警检测实际上是一种自适应门限检测,它利用参考窗样本本来估计环境噪声的参数以达到恒虚警率的特性。环境噪声包括杂波以及干扰等。当参考环境单元为非均匀分布的多目标环境时,有序统计量恒虚警检测相比其他类的恒虚警检测具有较好的抗干扰能力。
如图2所示,为第一次检测的结构框图。第一次检测具体包括:
确定一个乘积因子T;对于每一个距离单元D,在距离单元D的两侧设定预设数量的保护单元(Pn、Pn+1)和计算单元(X
1-X
N),然后对计算单元按从大到小顺序进行排序;预设序号可选用表示测速周期的调频周期数K。选定预设序号K的计算单元Z和乘积因子T相乘作为第一次检测的门限值S。图中D为被测目标单元,因为目标的功率可能泄露到相邻的单元中,所以和目标相邻的数个单元不作为背景杂波的估计,作为保护单元P。
利用比较器比较被测距离单元D与第一次检测的门限值S的大小,当被测距离单元D大于门限值S时,将D确定为第一目标。
例如:对一组调频信号的回波差频信号进行1024点FFT后,得到1024个输出数据,取数据1-512进行检测(由于FFT的输出是对称的),对每一个数据逐一检测,一个数据就表示一个频点(也就表示某一个距离)。如果保护单元数为4,计算单元数为8,则对于第20个单元,保护单元为18、19、21、22;计算单元为14、15、16、17、23、24、25、26。
为了从弱信号中检测出目标,第一次检测需要降低门槛,以免漏掉真实目标。第一次检测是从非均匀杂波中检测出疑似目标。
S104、对所述第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件;所述第二目标的个数小于所述第一目标的个数。
第二次检测是用于剔除部分虚假目标。由于第一有序统计量恒虚警检测的要求比较低,可能一些虚假目标被误认为是目标。第二次检测进一步缩小范围将虚假目标进行剔除。
第二次检测具体包括:
将第一次检测得到的第一目标的均值乘以预设系数得到第二次检测的初始门限值。计算第一次检测过门限目标的均值,由均值乘以一个预设系数得到第二次检测搜索的初始门值,为了降低虚警概率,兼顾强信号和弱信号检测,第二次检测搜索的初始固定门限值可设置小点即预设系统可以设置得小一些。
利用比较器比较被测距离单元与所述第二次检测的初始门限值的大小,当被测距离单元大于所述第二次检测的初始门限值时,将其确定为第二目标。
当第二目标的数量高于预设条件时,将初始门限值提高一个预设搜索步进值作为下一次检测搜索的比较门限值,当第二目标的数量低于预设条件时,将初始门限值降低一个预设搜索步进值作为下一次检测搜索的比较门限值;重复上述比较判断直至第二目标的数量满足预设条件。
设定一个预设搜索步进值(比如10%),如果第二次检测的低于预设条件时,则将第二次初始门限降低一个搜索步进值继续搜索,直到搜索到目标为止;如果第二次搜索到的目标数量高于预设条件,则说明虚假目标太多,则将第二次初始门限提高一个搜索步进值重新搜索。
第二次检测中目标数量的预设条件,一般可以根据需要设置(取决于硬件和需求)。比如车载角雷达和前向雷达就明显不一样,角雷达的作用距离短,目标数目可以设置的少点(比如10-20个),前向雷达作用距离远,硬件性能也有足够的余量,可以将目标数量设置大点(比如30-40个)。
第m-1次检测搜索得到的目标数量高于预设条件,此时需要提高判决门限,当判决门限提高一个预设搜索步进值后进行第m次检测搜索时,发现检测搜索得到的目标数量又低于预设条件,此时需要降低判决门限。降低判断门限的幅度不能按原来的调整幅度不变,否则数据不会收敛,第二次检测得到的目标数量永远无法满足预设条件。此时可以将调整幅度进行折半处理。即将预设搜索步进值折半增减,该方式有助于第二次检测得到的第二目标的数量收敛快速满足预设条件。
例如:预设搜索步进值为10%。目标初始门限为100。当检测门限为100时进行第二次检测得到的第二目标数量高于预设条件,因此需要将检测门限提高,当检测门限提高一个预设搜索步进值改为100+10=110后进行第二次检测得到的第二目标数量又低于预设条件,此时需要将检测门限降低,此时如果降低一个预设搜索步进值110-10=100会导致第二次检测得到的第二目标数量始终无法满足预设条件。因此需要将预设搜索步进值进行折半处理。将降幅折半处理,即将检测门限设置为110-5=105,依次类推。
S105、对所述第二目标进行第三次检测得到第三目标,所述第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标。
如图3所示,第三次检测具体包括如下步骤:
S1051、计算所述第二目标的距离和速度。
计算第二目标的距离和速度采用如下表达式,其中,计算距离采用表达式
其中,c表示光速,T
c表示单个调频周期,F
s表示采样率,M
bin表示目标距离所在的FFT单元,B表示信号带宽,N
FFT表示距离FFT点数。
其中,V
bin表示目标速度所在的FFT单元,f表示发射信号频率,K表示测速周期的调频周期数(也是速度FFT的点数),T
f表示测速周期(等于KT
c)。上式涉及到速度的定义中目标接近雷达速度为正。
S1052、计算相应距离和速度的协方差矩阵。
这里需要区分两种情况:一种情况是不存在相干信号(即两个或两个以上的目标距离一样,速度也一样),此时的阵列协方差矩阵可以由原始数据S(三维立体数据:一帧数据有N个阵元,每个阵元有K个调频周期,每个调频周期有若干数目的采样点数据)经过两次FFT变换(第一次是对单个调频周期的采样点数据进行FFT,可以得到目标的距离信息;第二次是在第一次FFT的基础上再对相应目标距离的K个调频周期数据进行FFT,可以得到目标的速度信息)的数据S
FFT(N×K×N
FFT维)得到,即找到每个阵元的相应目标距离单元和速度单元的数据按照阵元排列成一向量X=[X
1 X
2 … X
N]
T,其中
X
n=S
FFT(n,V
bin,M
bin),n=1,2,···N;V
bin=1,2,···K,M
bin=1,2,···N
FFT。此时,相应目标的协方差矩阵为:
上标‘H’表示共轭转置。
另外一种情况是存在相干信号,此时的协方差矩阵需要将有目标的距离单元对应的没有速度目标的单元数据也进行累加求和,此时X
n成为一向量(X
n=S
FFT(n,1:K,M
bin),n=1,2,…N),X成为N×K的矩阵,此时,相应目标的协方差矩阵为:
S1053、分别对不同目标的协方差矩阵进行特征分解。
其中,U
S是由大特征值对应的特征矢量张成的子空间也即信号子空间,而U
N是由小特征值对应的特征矢量张成的子空间也即噪声子空间。利用特征分解后的特征值可以进行信源数目的估计,信源数目的正确估计是空间谱估计的前提。理想条件下数据空间 中的信号子空间与噪声子空间是相互正交的,即信号子空间中的导向矢量也与噪声子空间正交:a
H(θ)U
N=0。
S1054、以最小优化搜索实现角度估计并计算MUSIC谱。
由此得到经典MUSIC算法的空间谱估计公式:
S1055、对所述MUSIC谱进行真假目标判决,当MUSIC谱峰大于预设判决门限时判定为第三目标,否则判定为非目标。
对计算得到的MUSIC谱进行真假目标的判决:如果MUSIC谱峰大于预设判决门限就判定为第三目标,否则判定为非目标。其中预设判决门限的确定,经过大量数据仿真并结合实测数据,可将预设判决门限设置为MUSIC所有谱峰均值的5-10倍。
S106、输出所述第三目标的距离、速度和角度信息。
经过三次检测,可以剔除虚假目标并得到真实目标的距离、速度和角度。
在本实施例中,第一次检测、第二次检测以及第三次检测的具体步骤均为最优方案。在其他实施例中,第一次检测、第二次检测以及第三次检测的具体步骤还可以采用其他形式实现以达到每个检测对应的作用即可。
当然,在实际工程实现时,还需要解决很多问题,比如阵列误差、阵元互耦和通道幅相不一致的校正,矩阵的特征分解,信源数目的正确估计,信号解相干的处理,等等,在此不再赘述。
相对于三角波FMCW信号调制,锯齿波FMCW信号调制有明显的优势,它不需要进行距离和速度的配对,可有效提升弱信号目标的检测性能,本申请选用锯齿波FMCW信号体制。锯齿波信号调制的FMCW阵列雷达的信号波形和硬件参数需要满足如下条件:
(1)根据测距精度要求选择合适的调频工作带宽,距离分辨率和工作带宽的关系为:
其中,d
res表示距离分辨率,c表示光速,B表示FMCW阵列雷达的调频工作带宽。如果距离分辨率需要在1米以内,则调频工作带宽应大于150MHz。
(2)根据测速要求(包括速度精度要求和最大目标速度要求)和信号波长选择合适的调频周期和测速周期。对于锯齿波FMCW信号调制,其最大可测速度和速度分辨率分别需要满足下列关系:
其中,V
max表示最大可测速度,V
res表示速度分辨率,T
c表示单个调频周期,T
f表示测速周期,一般包含K个调频周期。如果发射信号波长是0.004米,最大可测速度(与雷达的相对径向速度)不小于50米/秒,速度分辨率不大于1米/秒,则T
c不应大于20微秒,T
f应不小于2毫秒。
(3)根据最大测量距离和调频工作带宽选择合适的采样率。信号的采样率应满足如下关系:
其中,F
s表示采样率,d
max表示最大可探测距离,T
c表示单个调频周期,c表示光速,B表示调频工作带宽。如果B是200MHz,d
max是200米,T
c是20微秒,则F
s应不小于40/3MHz。
根据测角要求选择合适的阵元数目。对于采用常规的测角方法来说,角度分辨率满足下面的关系(瑞利限):
其中,θ
res表示角度分辨率,N表示阵元数目,d表示阵元间距,θ表示目标与阵列法线的夹角(如果目标位于阵列法线,则θ=0)。如果阵元间距是半波长,目标位于阵列法线,阵元数是8,则角度分辨率大约是14.324°。对于单目标测角来说,理想情况下单脉冲测角精度可以达到角度分辨率的2%,当然,实际工程中还有很多因素制约角度分辨率,一般可以达到10%,对于有的应用场景可以满足要求,但是对于车载应用场景,这个角度分辨率是不够的。高分辨空间谱估计为小阵列高分辨测角提供了解决方法,其可以突破瑞利限,获得远优于传统测角方法的测角精度。
通过该方法进行强杂波背景下FMCW阵列雷达运动多目标弱信号检测,经仿真验证,阵元数为8,目标信噪比0dB,杂噪比15dB,在综合误差(存在互耦(相邻阵元间)0.1+0.1i、阵元误差(正负10%内)、幅相不一致(幅度正负1dB内、相位正负 10°内))下,100次蒙特卡洛仿真的检测成功概率达到100%,均值偏差和方差分别不大于0.01°、0.02。该方法不仅可以有效检测到目标(信噪比0dB时,极低的虚警概率和漏警概率),还有很高的测角精度。
本发明实施例还提供了一种强杂波下FMCW阵列雷达运动多目标弱信号检测装置,所述FMCW阵列雷达采用锯齿波信号调制,如图4所示,该装置包括:
傅里叶变换模块,用于对不同阵元不同调频周期的差频信号分别进行快速傅里叶变换处理得到不同目标的距离谱。
输入的阵列差频信号数据包括N个阵元K个调频周期,每个调频周期有M个采样点。分别对N个阵元的K个调频周期进行快速傅里叶变换(FFT)处理得到不同目标的距离谱。该距离谱为3维立体数据。
对消处理模块,用于对所述距离谱进行对消处理以剔除静止目标产生的杂波信号。
对消处理的作用是为了消除静止目标产生的杂波,只留下运动目标,杂波杂波对消处理可以剔除静止目标产生的杂波信号,使参与后续运算的回波数量减少,从而简化后面的多运动目标分离工作。
第一检测模块,用于对对消处理后的数据进行第一次检测得到第一目标;所述第一次检测采用有序统计量恒虚警检测。
第一检测模块具体用于:
确定一个乘积因子;
对于每一个距离单元,在所述距离单元的两侧设定预设数量的保护单元和计算单元,然后对所述计算单元按从大到小顺序进行排序;
选定预设序号的计算单元和所述乘积因子相乘作为第一次检测的门限值;
利用比较器比较被测距离单元与所述第一次检测的门限值的大小,当被测距离单元大于所述门限值时,将其确定为第一目标。
第二检测模块,用于对所述第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件;所述第二目标的个数小于所述第一目标的个数。
第二次检测是用于剔除部分虚假目标。由于第一有序统计量恒虚警检测的要求比较低,可能一些虚假目标被误认为是目标。第二次检测进一步缩小范围将虚假目标进行剔除。
第二检测模块具体用于:
将第一次检测得到的第一目标的均值乘以预设系数得到第二次检测的初始门限值;
利用比较器比较被测距离单元与所述第二次检测的初始门限值的大小,当被测距离单元大于所述第二次检测的初始门限值时,将其确定为第二目标;
当第二目标的数量高于预设条件时,将初始门限值提高一个预设搜索步进值作为下一次检测搜索的比较门限值,当第二目标的数量低于预设条件时,将初始门限值降低一个预设搜索步进值作为下一次检测搜索的比较门限值;重复上述比较判断直至第二目标的数量满足预设条件。
当数据不收敛,即第二次检测得到的目标数量始终无法满足预设条件时,需要将预设搜索步进值折半增减,该方式有助于第二次检测得到的第二目标的数量快速满足预设条件。
第三检测模块,用于对所述第二目标进行第三次检测得到第三目标,所述第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标。
第三检测模块具体用于:
计算所述第二目标的距离和速度。
计算第二目标的距离和速度采用如下表达式,其中,计算距离采用表达式
其中,c表示光速,T
c表示单个调频周期,F
s表示采样率,M
bin表示目标距离所在的FFT单元,B表示信号带宽,N
FFT表示距离FFT点数。
其中,V
bin表示目标速度所在的FFT单元,f表示发射信号频率,K表示测速周期的调频周期数(也是速度FFT的点数),T
f表示测速周期(等于KT
c)。上式涉及到速度的定义中目标接近雷达速度为正。
计算相应距离和速度的协方差矩阵。
这里需要区分两种情况:一种情况是不存在相干信号(即两个或两个以上的目标距离一样,速度也一样),此时的阵列协方差矩阵可以由原始数据S(三维立体数据:一帧数据有N个阵元,每个阵元有K个调频周期,每个调频周期有若干数目的采样点数据)经过两次FFT变换(第一次是对单个调频周期的采样点数据进行FFT,可以得到目标的距离信息;第二次是在第一次FFT的基础上再对相应目标距离的K个调频周期数据进行FFT,可以得到目标的速度信息)的数据S
FFT(N×K×N
FFT维)得到,即找到每个阵元的相应目标距离单元和速度单元的数据按照阵元排列成一向量 X=[X
1 X
2 … X
N]
T,其中
X
n=S
FFT(n,V
bin,M
bin),n=1,2,···N;V
bin=1,2,···K,M
bin=1,2,···N
FFT。此时,相应目标的协方差矩阵为:
上标‘H’表示共轭转置。
另外一种情况是存在相干信号,此时的协方差矩阵需要将有目标的距离单元对应的没有速度目标的单元数据也进行累加求和,此时X
n成为一向量(X
n=S
FFT(n,1:K,M
bin),n=1,2,…N),X成为N×K的矩阵,此时,相应目标的协方差矩阵为:
分别对不同目标的协方差矩阵进行特征分解。
其中,U
S是由大特征值对应的特征矢量张成的子空间也即信号子空间,而U
N是由小特征值对应的特征矢量张成的子空间也即噪声子空间。利用特征分解后的特征值可以进行信源数目的估计,信源数目的正确估计是空间谱估计的前提。理想条件下数据空间中的信号子空间与噪声子空间是相互正交的,即信号子空间中的导向矢量也与噪声子空间正交:a
H(θ)U
N=0。
以最小优化搜索实现角度估计并计算MUSIC谱。
由此得到经典MUSIC算法的空间谱估计公式:
对所述MUSIC谱进行真假目标判决,当MUSIC谱峰大于预设判决门限时判定为第三目标,否则判定为非目标。
对计算得到的MUSIC谱进行真假目标的判决:如果MUSIC谱峰大于预设判决门限就判定为第三目标,否则判定为非目标。其中预设判决门限的确定,经过大量数据仿真并结合实测数据,可将预设判决门限设置为MUSIC所有谱峰均值的5-10倍。
输出模块,用于输出所述第三目标的距离、速度和角度信息。
经过三次检测,可以剔除虚假目标并得到真实目标的距离、速度和角度。
所述FMCW阵列雷达的调频工作带宽满足表达式
其中,d
res表示距离分辨率,c表示光速,B表示调频工作带宽;所述FMCW阵列雷达的单个调频周期满足表达式
其中,V
max表示最大可测速度,T
c表示单个调频周期,λ表示发射信号波长;所述FMCW阵列雷达的测速周期满足表达式
其中,V
res表示速度分辨率,T
f表示测速周期,λ表示发射信号波长;所述FMCW阵列雷达的采样率满足表达式
其中,F
s表示采样率,d
max表示最大可探测距离,T
c表示单个调频周期,c表示光速,B表示调频工作带宽。
图5所示,本申请一实施例提供的计算机设备的具体结构框图,该计算机设备100包括:一个或多个处理器101、存储器102、以及一个或多个计算机程序,其中所述处理器101和所述存储器102通过总线连接,所述一个或多个计算机程序被存储在所述存储器102中,并且被配置成由所述一个或多个处理器101执行,所述处理器101执行所述计算机程序时实现如本申请一实施例提供的强杂波下FMCW阵列雷达运动多目标弱信号检测方法的步骤。计算机设备包括服务器和终端等。该计算机设备可以是个人计算机、移动终端或车载设备,移动终端包括手机、平板电脑、个人数字助理或可穿戴设备等中的至少一种。
为了实现上述实施例,本发明还提出一种存储介质,用于存储计算机程序,当所述计算机程序在计算机设备上运行时,使得所述计算机设备执行如本申请一实施例提供的强杂波下FMCW阵列雷达运动多目标弱信号检测方法的步骤。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为,表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分,并且本发明的优选实施方式的范围包括另外的实现,其中可以不按所示出或讨论的顺序,包括根据所涉及的功能按基本同时的方式或按相反的顺序,来执行功能,这应被本发明的实施例所属技术领域的技术人员所理解。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤,例如,可以被认为是用于实现逻辑功能的可执行指令的定序列表,可以具体实现在任何计算机可读介质中,以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用,或结合这些指令执行系统、装置或设备而使用。就本说明书而言,"计算机可读介质"可以是任何可以包含、存 储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下:具有一个或多个布线的电连接部(电子装置),便携式计算机盘盒(磁装置),随机存取存储器(RAM),只读存储器(ROM),可擦除可编辑只读存储器(EPROM或闪速存储器),光纤装置,以及便携式光盘只读存储器(CDROM)。另外,计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质,因为可以例如通过对纸或其他介质进行光学扫描,接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序,然后将其存储在计算机存储器中。
应当理解,本发明的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中,多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。例如,如果用硬件来实现,和在另一实施方式中一样,可用本领域公知的下列技术中的任一项或他们的组合来实现:具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路,具有合适的组合逻辑门电路的专用集成电路,可编程门阵列(PGA),现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成,所述的程序可以存储于一种计算机可读存储介质中,该程序在执行时,包括方法实施例的步骤之一或其组合。
此外,在本发明各个实施例中的各功能单元可以集成在一个处理模块中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现,也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时,也可以存储在一个计算机可读取存储介质中。
上述提到的存储介质可以是只读存储器,磁盘或光盘等。
在本说明书的描述中,参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。在本说明书中,对上述术语的示意性表述不一定指的是相同的实施例或示例。而且,描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
尽管已经示出和描述了本发明的实施例,对于本领域的普通技术人员而言,可以理解在不脱离本发明的原理和精神的情况下可以对这些实施例进行多种变化、修改、替换和变型,本发明的范围由所附权利要求及其等同限定。
Claims (19)
- 一种强杂波下FMCW阵列雷达运动多目标弱信号检测方法,其特征在于,所述FMCW阵列雷达采用锯齿波信号调制,该方法包括:对不同阵元不同调频周期的差频信号分别进行快速傅里叶变换处理得到不同目标的距离谱;对所述距离谱进行对消处理以剔除静止目标产生的杂波信号;对对消处理后的数据进行第一次检测得到第一目标;所述第一次检测采用有序统计量恒虚警检测;对所述第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件;所述第二目标的个数小于所述第一目标的个数;对所述第二目标进行第三次检测得到第三目标,所述第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标;输出所述第三目标的距离、速度和角度信息。
- 根据权利要求1所述的方法,其特征在于,所述第一次检测具体包括:确定一个乘积因子;对于每一个距离单元,在所述距离单元的两侧设定预设数量的保护单元和计算单元,然后对所述计算单元按从大到小顺序进行排序;选定预设序号的计算单元和所述乘积因子相乘作为第一次检测的门限值;利用比较器比较被测距离单元与所述第一次检测的门限值的大小,当被测距离单元大于所述门限值时,将其确定为第一目标。
- 根据权利要求1所述的方法,其特征在于,所述第二次检测具体包括:将第一次检测得到的第一目标的均值乘以预设系数得到第二次检测的初始门限值;利用比较器比较被测距离单元与所述第二次检测的初始门限值的大小,当被测距离单元大于所述第二次检测的初始门限值时,将其确定为第二目标;当第二目标的数量高于预设条件时,将初始门限值提高一个预设搜索步进值作为下一次检测搜索的比较门限值,当第二目标的数量低于预设条件时,将初始门限值降低一个预设搜索步进值作为下一次检测搜索的比较门限值;重复上述比较判断直至第二目标的数量满足预设条件。
- 根据权利要求1所述的方法,其特征在于,所述第三次检测具体包括:计算所述第二目标的距离和速度;计算相应距离和速度的协方差矩阵;分别对不同目标的协方差矩阵进行特征分解;以最小优化搜索实现角度估计并计算MUSIC谱;对所述MUSIC谱进行真假目标判决,当MUSIC谱峰大于预设判决门限时判定为第三目标,否则判定为非目标。
- 根据权利要求4所述的方法,其特征在于,所述预设判决门限值为MUSIC所有谱峰均值的5-10倍。
- 一种强杂波下FMCW阵列雷达运动多目标弱信号检测装置,其特征在于,所述FMCW阵列雷达采用锯齿波信号调制,该装置包括:傅里叶变换模块,用于对不同阵元不同调频周期的差频信号分别进行快速傅里叶变换处理得到不同目标的距离谱;对消处理模块,用于对所述距离谱进行对消处理以剔除静止目标产生的杂波信号;第一检测模块,用于对对消处理后的数据进行第一次检测得到第一目标;所述第一次检测采用有序统计量恒虚警检测;第二检测模块,用于对所述第一目标进行第二次检测使第二次检测后得到的第二目标的个数满足预设条件;所述第二目标的个数小于所述第一目标的个数;第三检测模块,用于对所述第二目标进行第三次检测得到第三目标,所述第三次检测采用阵列信息进行空间谱估计,利用真实目标和虚假目标空间谱估计峰值的差异剔除虚假目标;输出模块,用于输出所述第三目标的距离、速度和角度信息。
- 根据权利要求10所述的装置,其特征在于,所述第一检测模块具体用于:确定一个乘积因子;对于每一个距离单元,在所述距离单元的两侧设定预设数量的保护单元和计算单元,然后对所述计算单元按从大到小顺序进行排序;选定预设序号的计算单元和所述乘积因子相乘作为第一次检测的门限值;利用比较器比较被测距离单元与所述第一次检测的门限值的大小,当被测距离单元大于所述门限值时,将其确定为第一目标。
- 根据权利要求10所述的装置,其特征在于,所述第二检测模块具体用于:将第一次检测得到的第一目标的均值乘以预设系数得到第二次检测的初始门限值;利用比较器比较被测距离单元与所述第二次检测的初始门限值的大小,当被测距离单元大于所述第二次检测的初始门限值时,将其确定为第二目标;当第二目标的数量高于预设条件时,将初始门限值提高一个预设搜索步进值作为下一次检测搜索的比较门限值,当第二目标的数量低于预设条件时,将初始门限值降低一 个预设搜索步进值作为下一次检测搜索的比较门限值;重复上述比较判断直至第二目标的数量满足预设条件。
- 根据权利要求10所述的装置,其特征在于,所述第三检测模块具体用于:计算所述第二目标的距离和速度;计算相应距离和速度的协方差矩阵;分别对不同目标的协方差矩阵进行特征分解;以最小优化搜索实现角度估计并计算MUSIC谱;对所述MUSIC谱进行真假目标判决,当MUSIC谱峰大于预设判决门限时判定为第三目标,否则判定为非目标。
- 根据权利要求10所述的装置,其特征在于,所述预设判决门限值为MUSIC所有谱峰均值的5-10倍。
- 一种计算机设备,包括:一个或多个处理器、存储器以及一个或多个计算机程序,其中所述处理器和所述存储器通过总线连接,所述一个或多个计算机程序被存储在所述存储器中,并且被配置成由所述一个或多个处理器执行,其特征在于,所述处理器执行所述计算机程序时实现如权利要求1至9中任一项所述的强杂波下FMCW阵列雷达运动多目标弱信号检测方法。
- 一种存储介质,用于存储计算机程序,其特征在于,当所述计算机程序在计算机设备上运行时,使得所述计算机设备执行如权利要求1至9中任一项所述的强杂波下FMCW阵列雷达运动多目标弱信号检测方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201980099412.6A CN114270216A (zh) | 2019-12-31 | 2019-12-31 | 一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 |
PCT/CN2019/130437 WO2021134449A1 (zh) | 2019-12-31 | 2019-12-31 | 一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2019/130437 WO2021134449A1 (zh) | 2019-12-31 | 2019-12-31 | 一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021134449A1 true WO2021134449A1 (zh) | 2021-07-08 |
Family
ID=76685974
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/130437 WO2021134449A1 (zh) | 2019-12-31 | 2019-12-31 | 一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN114270216A (zh) |
WO (1) | WO2021134449A1 (zh) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113552533A (zh) * | 2021-07-19 | 2021-10-26 | 北京华力创通科技股份有限公司 | 一种空间谱估计方法、装置、电子设备及存储介质 |
CN113702965A (zh) * | 2021-08-31 | 2021-11-26 | 中国人民解放军海军航空大学 | 基于峰值聚优的改进积累方法及强弱目标同时检测方法 |
CN114002678A (zh) * | 2021-09-15 | 2022-02-01 | 丹阳市威鼎汽配有限公司 | 一种基于毫米波雷达动态保护单元的cfar检测方法 |
CN114002656A (zh) * | 2021-10-29 | 2022-02-01 | 中国人民解放军陆军军医大学 | 超宽带连续波生物雷达数据杂波抑制方法及系统 |
CN114019469A (zh) * | 2021-10-29 | 2022-02-08 | 中国人民解放军海军航空大学航空作战勤务学院 | 基于凹包判决的海面目标检测方法及设备、介质和产品 |
CN114065810A (zh) * | 2021-11-07 | 2022-02-18 | 西安电子工程研究所 | 基于谱峰个数的无人机与武装单兵分类识别算法 |
CN114265058A (zh) * | 2021-12-13 | 2022-04-01 | 中汽创智科技有限公司 | Mimo雷达目标测角方法、装置、电子设备及存储介质 |
CN114296047A (zh) * | 2021-12-29 | 2022-04-08 | 中仪知联(无锡)工业自动化技术有限公司 | 基于空域滤波的静止地物杂波对消方法 |
CN114488141A (zh) * | 2022-01-30 | 2022-05-13 | 北京理工大学 | 一种超宽带fmcw穿墙雷达的多目标成像方法 |
CN114690143A (zh) * | 2022-04-19 | 2022-07-01 | 珠海微度芯创科技有限责任公司 | 一种抑制雷达杂波的方法、装置、雷达和介质 |
CN115128590A (zh) * | 2022-06-24 | 2022-09-30 | 中国人民解放军国防科技大学 | 基于矩阵信息几何的自适应雷达目标检测方法和装置 |
CN115656998A (zh) * | 2022-11-22 | 2023-01-31 | 中国人民解放军空军预警学院 | 一种低样本数下阵列信号自适应检测方法与系统 |
CN115766361A (zh) * | 2022-10-17 | 2023-03-07 | 湖南鼎英信息科技有限公司 | 用于雷达通信一体化设备的前导序列处理方法及相关装置 |
CN116008948A (zh) * | 2023-03-27 | 2023-04-25 | 北京东远润兴科技有限公司 | 雷达定位方法、装置、设备及存储介质 |
CN117237833A (zh) * | 2023-11-16 | 2023-12-15 | 浙江凡双科技股份有限公司 | 基于自动门限提取的快速无人机图传信号识别方法及装置 |
CN117761631A (zh) * | 2024-02-22 | 2024-03-26 | 中国人民解放军空军预警学院 | 一种多通道融合检测的方法和装置 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116047462B (zh) * | 2023-03-31 | 2023-06-30 | 中国人民解放军空军预警学院 | 端射阵机载雷达最优阵元数目和阵元间距选取方法与装置 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050156780A1 (en) * | 2004-01-16 | 2005-07-21 | Ghz Tr Corporation | Methods and apparatus for automotive radar sensors |
CN102707285A (zh) * | 2012-05-28 | 2012-10-03 | 河海大学 | 车载毫米波防撞雷达系统的频域恒虚警检测方法 |
CN103257346A (zh) * | 2013-05-15 | 2013-08-21 | 桂林电子科技大学 | 一种汽车防撞雷达多目标探测方法与系统 |
US20150198697A1 (en) * | 2014-01-15 | 2015-07-16 | Panasonic Corporation | Radar apparatus |
CN105549001A (zh) * | 2015-12-02 | 2016-05-04 | 大连楼兰科技股份有限公司 | 车载毫米波雷达系统多目标检测方法 |
CN108169724A (zh) * | 2016-12-08 | 2018-06-15 | 南京理工大学 | 基于实信号数字波束形成的汽车防撞雷达系统及方法 |
CN108693523A (zh) * | 2017-12-06 | 2018-10-23 | 深圳大学 | 基于锯齿波线性调频雷达多目标测距测速的方法及系统 |
-
2019
- 2019-12-31 CN CN201980099412.6A patent/CN114270216A/zh active Pending
- 2019-12-31 WO PCT/CN2019/130437 patent/WO2021134449A1/zh active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050156780A1 (en) * | 2004-01-16 | 2005-07-21 | Ghz Tr Corporation | Methods and apparatus for automotive radar sensors |
CN102707285A (zh) * | 2012-05-28 | 2012-10-03 | 河海大学 | 车载毫米波防撞雷达系统的频域恒虚警检测方法 |
CN103257346A (zh) * | 2013-05-15 | 2013-08-21 | 桂林电子科技大学 | 一种汽车防撞雷达多目标探测方法与系统 |
US20150198697A1 (en) * | 2014-01-15 | 2015-07-16 | Panasonic Corporation | Radar apparatus |
CN105549001A (zh) * | 2015-12-02 | 2016-05-04 | 大连楼兰科技股份有限公司 | 车载毫米波雷达系统多目标检测方法 |
CN108169724A (zh) * | 2016-12-08 | 2018-06-15 | 南京理工大学 | 基于实信号数字波束形成的汽车防撞雷达系统及方法 |
CN108693523A (zh) * | 2017-12-06 | 2018-10-23 | 深圳大学 | 基于锯齿波线性调频雷达多目标测距测速的方法及系统 |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113552533A (zh) * | 2021-07-19 | 2021-10-26 | 北京华力创通科技股份有限公司 | 一种空间谱估计方法、装置、电子设备及存储介质 |
CN113702965A (zh) * | 2021-08-31 | 2021-11-26 | 中国人民解放军海军航空大学 | 基于峰值聚优的改进积累方法及强弱目标同时检测方法 |
CN114002678A (zh) * | 2021-09-15 | 2022-02-01 | 丹阳市威鼎汽配有限公司 | 一种基于毫米波雷达动态保护单元的cfar检测方法 |
CN114002656A (zh) * | 2021-10-29 | 2022-02-01 | 中国人民解放军陆军军医大学 | 超宽带连续波生物雷达数据杂波抑制方法及系统 |
CN114019469A (zh) * | 2021-10-29 | 2022-02-08 | 中国人民解放军海军航空大学航空作战勤务学院 | 基于凹包判决的海面目标检测方法及设备、介质和产品 |
CN114065810A (zh) * | 2021-11-07 | 2022-02-18 | 西安电子工程研究所 | 基于谱峰个数的无人机与武装单兵分类识别算法 |
CN114065810B (zh) * | 2021-11-07 | 2024-05-03 | 西安电子工程研究所 | 基于谱峰个数的无人机与武装单兵分类识别方法 |
CN114265058A (zh) * | 2021-12-13 | 2022-04-01 | 中汽创智科技有限公司 | Mimo雷达目标测角方法、装置、电子设备及存储介质 |
CN114296047A (zh) * | 2021-12-29 | 2022-04-08 | 中仪知联(无锡)工业自动化技术有限公司 | 基于空域滤波的静止地物杂波对消方法 |
CN114488141A (zh) * | 2022-01-30 | 2022-05-13 | 北京理工大学 | 一种超宽带fmcw穿墙雷达的多目标成像方法 |
CN114488141B (zh) * | 2022-01-30 | 2024-06-04 | 北京理工大学 | 一种超宽带fmcw穿墙雷达的多目标成像方法 |
CN114690143A (zh) * | 2022-04-19 | 2022-07-01 | 珠海微度芯创科技有限责任公司 | 一种抑制雷达杂波的方法、装置、雷达和介质 |
CN114690143B (zh) * | 2022-04-19 | 2023-01-06 | 珠海微度芯创科技有限责任公司 | 一种抑制雷达杂波的方法、装置、雷达和介质 |
CN115128590A (zh) * | 2022-06-24 | 2022-09-30 | 中国人民解放军国防科技大学 | 基于矩阵信息几何的自适应雷达目标检测方法和装置 |
CN115766361A (zh) * | 2022-10-17 | 2023-03-07 | 湖南鼎英信息科技有限公司 | 用于雷达通信一体化设备的前导序列处理方法及相关装置 |
CN115656998B (zh) * | 2022-11-22 | 2023-06-13 | 中国人民解放军空军预警学院 | 一种低样本数下阵列信号自适应检测方法与系统 |
CN115656998A (zh) * | 2022-11-22 | 2023-01-31 | 中国人民解放军空军预警学院 | 一种低样本数下阵列信号自适应检测方法与系统 |
CN116008948A (zh) * | 2023-03-27 | 2023-04-25 | 北京东远润兴科技有限公司 | 雷达定位方法、装置、设备及存储介质 |
CN116008948B (zh) * | 2023-03-27 | 2023-09-22 | 北京东远润兴科技有限公司 | 雷达定位方法、装置、设备及存储介质 |
CN117237833A (zh) * | 2023-11-16 | 2023-12-15 | 浙江凡双科技股份有限公司 | 基于自动门限提取的快速无人机图传信号识别方法及装置 |
CN117237833B (zh) * | 2023-11-16 | 2024-01-30 | 浙江凡双科技股份有限公司 | 基于自动门限提取的快速无人机图传信号识别方法及装置 |
CN117761631A (zh) * | 2024-02-22 | 2024-03-26 | 中国人民解放军空军预警学院 | 一种多通道融合检测的方法和装置 |
CN117761631B (zh) * | 2024-02-22 | 2024-05-07 | 中国人民解放军空军预警学院 | 一种多通道融合检测的方法和装置 |
Also Published As
Publication number | Publication date |
---|---|
CN114270216A (zh) | 2022-04-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021134449A1 (zh) | 一种强杂波下fmcw阵列雷达运动多目标弱信号检测方法、装置、计算机设备及存储介质 | |
CN111352102B (zh) | 一种基于调频连续波雷达的多目标个数检测方法及装置 | |
CN109375202B (zh) | 一种基于车载毫米波雷达的车辆测距测速方法 | |
EP3721258A1 (en) | Radar processing chain for fmcw radar systems | |
Rohling | Some radar topics: waveform design, range CFAR and target recognition | |
CN104614723A (zh) | 利用可变波形判别虚假目标的车辆用雷达及其判别方法 | |
WO2023030556A2 (zh) | 噪底估计值的确定方法、目标检测方法及装置、电子设备 | |
CN109932695A (zh) | 一种提高目标物识别速度的方法及装置 | |
CN113963441A (zh) | 一种基于跨域增强的毫米波雷达手势识别方法及系统 | |
CN111929685A (zh) | 基于线性调频连续波的雷达检测方法及装置、雷达装置和机动车 | |
CN113009441A (zh) | 一种雷达运动反射面多径目标的识别方法及装置 | |
CN105204017B (zh) | 一种基于正则优化的高分辨雷达角跟踪方法 | |
Özkan et al. | Ground multiple target tracking with a network of acoustic sensor arrays using PHD and CPHD filters | |
CN103901427A (zh) | 一种测速雷达多目标跟踪的方法和装置 | |
Pérez et al. | Range detection on time-domain FMCW radar signals with a deep neural network | |
CN114114197A (zh) | 多普勒模糊度的计算方法、多普勒速度的计算方法及装置 | |
CN105652256B (zh) | 一种基于极化信息的高频地波雷达tbd方法 | |
CN109917371B (zh) | 一种基于改进微波波形的微波雷达测量方法 | |
CN115267721B (zh) | 一种基于双频sar的地面动目标径向速度估计方法 | |
CN116593978A (zh) | 一种基于点云的手势检测方法、系统、存储介质 | |
CN110940973A (zh) | 一种用于雷达目标检测的角度测量方法及装置 | |
CN110907930B (zh) | 一种基于角度估计的车载雷达目标检测估计方法及装置 | |
Lei et al. | A TD-CF preprocessing method of FMCW radar for Dynamic Hand Gesture Recognition | |
CN116027288A (zh) | 生成数据的方法、装置、电子设备及存储介质 | |
Chang et al. | Modified position correlation method for high resolution radar target detection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19958635 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19958635 Country of ref document: EP Kind code of ref document: A1 |