CN106707258A - Multi-parameter estimation method for micro-motion target under non-Gaussian background - Google Patents
Multi-parameter estimation method for micro-motion target under non-Gaussian background Download PDFInfo
- Publication number
- CN106707258A CN106707258A CN201710123663.6A CN201710123663A CN106707258A CN 106707258 A CN106707258 A CN 106707258A CN 201710123663 A CN201710123663 A CN 201710123663A CN 106707258 A CN106707258 A CN 106707258A
- Authority
- CN
- China
- Prior art keywords
- target
- fine motion
- auto
- correlation entropy
- correlation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a multi-parameter estimation method for a micro-motion target under a non-Gaussian background. The method comprises the steps of: based on a single input multiple output (SIMO) radar system, calculating an autocorrelation entropy matrix of each receiving channel for each distance unit by using a Gaussian kernel function; then, performing mutual correlation processing on the autocorrelation entropy matrixes of every two channels to obtain multiple groups of mutual correlation entropy matrixes; fusing the obtained mutual correlation entropy matrixes in an arithmetic fusion mode; and finally, performing phase comparison monopulse positioning on multi-channel echo signals, thus positioning the micro-motion target. The method can realize multi-micro-parameter estimation and target positioning for the micro-motion target under the non-Gaussian background, and can obtain better estimation performance while estimating target parameters.
Description
Technical field
The present invention relates to radar micro-doppler technical field of imaging, the more particularly to fine motion target under non-gaussian background is more
Method for parameter estimation.
Background technology
In many detection systems, the detection and estimation of fine motion target suffer from important application.For example, in earthquake rescue
When can realize life signal detect；In air defense and monitoring field, it is possible to achieve unmanned plane wing is recognized；In Mechanical Property Analysis
Aspect, it is possible to achieve to the vibration detection of machinery.
Domestic and international many research institutions have carried out the research of fine motion target component estimation.Fourier analysis is to realize fine motion mesh
Mark the effective means that micro- perturbation parameter is estimated.The National University of Defense technology propose it is a kind of based on circulation coefficient correlation the fine motion cycle estimate
Meter method (W.P.Zhang, K.L.Li, et al, " Parameter estimation of radar targets with
macro-motion and micro-motion based on circular correlation coefficients[J]”,
IEEE Signal Processing Letters, 2013.), it does not change peak value number and size using cross-correlation is circulated
Characteristic, in the case where fine motion is with macroscopic motion, it is to avoid macroscopic motion influences on fine motion, reaches to the estimation in fine motion cycle
Purpose.University of Electronic Science and Technology proposes a kind of fine motion object detection method (Y.Jia, L.J.Kong, et al, " A in image area
novel method for detection of micro-motion target in image domain”,IEEE Radar
Conference, 2011.), it uses ULTRA-WIDEBAND RADAR, and the method being imaged using BP is realized determines target in image area
Position.
The above method shows good performance to the parameter Estimation of fine motion target under Gaussian Background, however, in not high
The performance of Fourier's analysis method is decreased obviously under this background.Therefore, the fine motion target component under research non-gaussian background is estimated
Method has important value in radar micro-doppler imaging field.
The content of the invention
The present invention provides the micro- moving target parameter estimation method under a kind of background suitable for non-gaussian, employs receipts more than single-shot
(SIMO) a kind of radar system, it is proposed that algorithm for estimating based on cross-correlation entropy.First, using gaussian kernel function, to each
Range cell, calculates the auto-correlation entropy of each receiving channel；Then, the auto-correlation entropy to each two interchannel is done at cross-correlation
Reason, obtains cross-correlation entropy；By way of arithmetic is merged, the cross-correlation entropy to obtaining does fusion treatment；Finally, by leading to more
Road echo-signal compares phase processor, it is possible to achieve to the positioning of target.
Technical solution of the present invention is as follows：The method for parameter estimation of compound fine motion target under a kind of non-gaussian background, including with
Lower step：
Step 1：Echo matrix is pre-processed
First to the N × L dimension echo matrixes A obtained by M passage of radar receiver1、A2、…、AMPressed on the fast time respectively
Row carries out windowing process, then is N points Fourier transformation (FFT), and wherein N is the total echo-signal sum of each passage, and L is each
The sampling number of echo-signal；Each echo-signal is compressed into the time domain impulse with target range information after Fourier transformation
Signal；All time domain impulsive signals signals are combined, the range-pulse domain matrix Z of M paths fine motion targets is obtained1、
Z2、…ZM；To Z1、Z2、…ZMRespectively on the slow time by row make Moveing target indication (MTI) filtering or remove average value processing, gone
Except the range-pulse domain matrix D of static background1、D2、…DM；
Step 2：Calculate auto-correlation entropy matrix：By the range-pulse domain matrix D corresponding to each passage1、D2、…DMEvery
Auto-correlation entropy treatment is carried out by row in individual range cell, the corresponding auto-correlation entropy matrix V of each passage is calculated1、V2、…VM；
Step 3：Cross correlation process and fusion
Assuming that k-th auto-correlation entropy matrix of passage is Vk, k=1,2...M, then the cross-correlation entropy of k passages and l passages can
To be expressed as：
Vkl[m]=xcorr (Vk(n),Vl(n))(k≠l)
To auto-correlation entropy matrix V1、V2、….VMOptional two are done cross correlation process as one group, are obtainedGroup cross-correlation entropy
Vil(i, l=1,2 ... M, and i ≠ l)；Arithmetic fusion is carried out to the last multigroup cross-correlation entropy for obtaining and obtains last cross-correlation
Entropy matrix V, wherein：V=∑s Vil, xcorr represents computing cross-correlation；
Step 4：The calculating of each parameter of fine motion target
Cross-correlation entropy matrix V is by L on slow time dimension1Point Fourier transformation obtains cross-correlation entropy-spectrum PV(ω), L1Represent slow
The points of time dimension Fourier transformation, frequency resolution that can be according to actual needs determines；Its corresponding imaging plane can be clear
The clear distance for clearly showing fine motion target, compound fine motion frequency, to PV(ω) carries out high-pass filtering, is gone using prior information
Except the interference of other band noises, Doppler-apart from plane is obtained, complete the identification to target.
Further, the step 2 is concretely comprised the following steps：The range-pulse domain matrix D for first respectively tieing up N × L1、
D2、…DMEach row be considered as discrete stationary random process x (n) 1 × N of ∈, and auto-correlation is carried out one by one to each sampled point
Entropy is calculated, and obtains the auto-correlation entropy vector vector of N-1 pointsM=1,2 ... N-1,
WhereinRepresent the corresponding auto-correlation entropy vector of each column data, the common L row of auto-correlation entropy vector；By all L row certainly
Joint entropy vector combination, obtains corresponding auto-correlation entropy matrix V1、V2、….VM；Wherein m pointsAuto-correlation entropy calculate
Formula is：
Wherein
Gaussian kernel function is represented, σ is the scale parameter value of core, x*(n-m) transposition of the x (n-m) in random process is represented.
Further, the step 4 is concretely comprised the following steps：Assuming that during fine motion target imaging, imaging point is in distance axis correspondence
Index value be R_index, be f_index1, f_index2, the then distance of target in the corresponding index value of Doppler frequency axle
And the computing formula of fine motion frequency is as follows：
Wherein, R represents fine motion target range, f1Represent target fine motion frequency 1, f2Target fine motion frequency 2 is represented,It is range resolution ratio, c is propagation velocity of electromagnetic wave, and B is transmitted signal bandwidth；Δ f is frequency resolution,T is transmission signal time width, L1For the Fourier transformation of slow time dimension is counted；
With reference to multichannel phase comparison monopulse method, the phase difference that target is produced in different interchannels is obtained, calculate target
Azimuth：
Wherein, d is antenna spacing, and λ is transmission signal wavelength, with reference to the fine motion target range R for measuring before, is realized to micro-
The positioning of moving-target.
The beneficial effects of the invention are as follows：
The present invention is proposed under the radar system that (SIMO) is received a kind of single-shot more, it is adaptable to the fine motion mesh under non-gaussian background
Mark Multiple Parameter Estimation Methods, the method can effectively detect and estimate multiple fine motion parameters of fine motion target, improve radar
While image quality, the loss of fine motion parameter is not resulted in.Compared to the conventional Fourier analysis side mentioned in technical background
Method and auto-correlation entropy processing method, in non-gaussian background, the present invention can show output signal-to-noise ratio higher, meanwhile, into
Many fine motion parameters as clearly indicating target in image.With reference to single channel than phase (PCM) technology, the present invention realizes fine motion
The space orientation of target.By the acquisition of above-mentioned parameter, the present invention ensure that the practical function of fine motion target detection system, be
Operating personnel make correct decision-making there is provided sound assurance.
Brief description of the drawings
Fig. 1 is cross-correlation entropy algorithm process flow chart.
Fig. 2 is LFMCW relationship between frequency and time schematic diagrames.
Fig. 3 is simulating scenes schematic diagram.
Fig. 4 is that conventional Fourier analyzes simulation result figure.
Fig. 5 is that auto-correlation entropy analyzes simulation result figure (before fusion).
Fig. 6 is that auto-correlation entropy analyzes simulation result figure (after fusion).
Fig. 7 is that cross-correlation entropy analyzes simulation result figure (before fusion).
Fig. 8 is that cross-correlation entropy analyzes simulation result figure (after fusion).
Fig. 9 is fine motion target positioning result.
Specific embodiment
Specific embodiment of the invention is given below according to a MATLAB examples of simulation.
To the parameter Estimation of fine motion target, its simulating scenes is as shown in figure 3, zero point of reference frame is located at position of transmitting antenna, 4
Individual reception antenna is placed at equal intervals along x-axis, and the horizontal interval between two neighboring reception antenna is half-wavelength d=λ/2, radar
Frequency 1GHz, the linear FM signal of bandwidth 250MHz centered on transmission signal, a width of 3.31ms during signal, transmission signal cycle
Number is 512.Fine motion target is located at (0,6) m, and the vibration frequency of target is respectively 4Hz and 8Hz, conventional Fourier analysis emulation
Result figure is as shown in Figure 4.The result imaging background strong jamming composition of conventional Fourier analysis spreads all over whole imaging plane, drops significantly
Low parameter Estimation performance, and there is the situation of fine motion parameter loss.
Treatment in accordance with the present invention step：
Step 1：Echo matrix is pre-processed
To the N × L dimensions obtained by the reception antenna of M=4 roads, (N is total number of echoes of each passage, and N=512, L are every first
The sampling number of individual echo-signal, L=256) echo matrix A1、A2、…、A4Carried out at adding window (by row) on the fast time respectively
FFT (Fourier transformation) is managed and be, the echo-signal in each cycle is compressed into sinc time domain impulsive signals, sinc has carried mesh
Target range information；N number of sinc signals that the aforesaid operations carried out on N number of frequency sweep cycle are obtained are combined again, obtains 4
The range-pulse domain matrix Z of paths fine motion target1、Z2、…Z4；To Z1、Z2、…Z4Make MTI (by row) on the slow time respectively
Filtering, obtains eliminating the range-pulse domain matrix D of the zero-frequency clutter such as static background1、D2、…D4；
Step 2：Auto-correlation entropy is calculated
According to analysis before to auto-correlation entropy ,-pulse domain matrix D of adjusting the distance1、D2、…D4(pressed on the slow time respectively
Row) auto-correlation entropy is calculated, obtain auto-correlation entropy matrix V1、V2、….V4；Wherein V1Corresponding auto-correlation entropy-spectrum imaging results figure is such as
Shown in Fig. 5,4 passage auto-correlation entropy matrix Vs1、V2、….V4By auto-correlation entropy matrix correspondence after the fusion obtained after arithmetic is merged
Target imaging result it is as shown in Figure 6.
Step 3：Cross correlation process and fusion
According to the analysis to cross-correlation entropy, to auto-correlation entropy matrix V1、V2、….V4Optional two are done cross-correlation as one group
Treatment, obtainsGroup cross-correlation entropy Vil(i, l=1,2 ... 4, and i ≠ l), its corresponding radar imagery plane such as Fig. 7 institute
Show.In order to obtain more preferable imaging effect and output signal-to-noise ratio higher, the present invention is to the last multigroup cross-correlation entropy for obtaining
Carry out arithmetic fusion and obtain last cross-correlation entropy matrix V, wherein：
V=∑s Vil
Step 4：The calculating of each parameter of fine motion target
Cross-correlation entropy matrix V is by L on slow time dimension1Point Fourier transformation treatment obtains cross-correlation entropy-spectrum PV(ω)；Its is right
The parameters such as the distance for showing fine motion target of the imaging plane energy clear and definite answered, compound fine motion frequency, to PV(ω) is carried out
High-pass filtering, removes other band noises and disturbs using prior information, obtains Doppler-apart from plane, can further improve
Performance is estimated in the image quality of target, lifting, and final imaging effect is as shown in Figure 8.
It is micro- in Fig. 8 (in order to more intuitively represent target component, the parameter that Fig. 8 shows is after imager coordinate is changed)
Pre-filter method point is R_index=10 in the corresponding index value of distance axis, is f_ in the corresponding index value of Doppler frequency axle
Index1=53, f_index2=109, range resolution ratio isFrequency resolution isTransmission signal time width T=3.31ms, slow time dimension Fourier transformation points L1=4096.Then target
Distance and fine motion frequency measurement be
With reference to multichannel phase comparison monopulse method, fine motion target positioning result is as shown in Figure 9.By than phase processor, we
Phase difference=0.0291 that target is produced in different interchannels, antenna spacing d=6.3mm, transmission signal wavelength X can be obtained
=1.25cm, the azimuth of target is
With reference to the fine motion target for measuring before apart from R=6.024m, the coordinate of fine motion target is：
From simulation result, the present invention provide suitable for the fine motion target Multiple Parameter Estimation Methods under non-gaussian background
Multiple fine motion parameters of fine motion target can not only be effectively detected and estimated, radar imagery quality and output signal-to-noise ratio is being improved
While, the loss of fine motion parameter is not resulted in, demonstrate correctness of the invention and validity.
Claims (3)
1. the method for parameter estimation of fine motion target is combined under a kind of non-gaussian background, is comprised the following steps：
Step 1：Echo matrix is pre-processed
First to the N × L dimension echo matrixes A obtained by M passage of radar receiver1、A2、…、AMRespectively by traveling on the fast time
Row windowing process, then N points Fourier transformation (FFT) is, wherein N is the total echo-signal sum of each passage, and L is each echo
The sampling number of signal；Each echo-signal is compressed into the time domain impulse letter with target range information after Fourier transformation
Number；All time domain impulsive signals signals are combined, the range-pulse domain matrix Z of M paths fine motion targets is obtained1、
Z2、…ZM；To Z1、Z2、…ZMRespectively on the slow time by row make Moveing target indication (MTI) filtering or remove average value processing, gone
Except the range-pulse domain matrix D of static background1、D2、…DM；
Step 2：Calculate auto-correlation entropy matrix：By the range-pulse domain matrix D corresponding to each passage1、D2、…DMEach away from
Auto-correlation entropy treatment is carried out by row on unit, the corresponding auto-correlation entropy matrix V of each passage is calculated1、V2、…VM；
Step 3：Cross correlation process and fusion
Assuming that k-th auto-correlation entropy matrix of passage is Vk, k=1,2...M, then the cross-correlation entropy of k passages and l passages can be with table
It is shown as：
Vkl[m]=xcorr (Vk(n),Vl(n))(k≠l)
To auto-correlation entropy matrix V1、V2、….VMOptional two are done cross correlation process as one group, are obtainedGroup cross-correlation entropy Vil
(i, l=1,2 ... M, and i ≠ l)；Arithmetic fusion is carried out to the last multigroup cross-correlation entropy for obtaining and obtains last cross-correlation entropy
Matrix V, wherein：V=∑s Vil, xcorr represents computing cross-correlation；
Step 4：The calculating of each parameter of fine motion target
Cross-correlation entropy matrix V is by L on slow time dimension1Point Fourier transformation obtains cross-correlation entropy-spectrum PV(ω), L1Represent the slow time
The points of Fourier transformation are tieed up, frequency resolution that can be according to actual needs determines；Its corresponding imaging plane can be clear bright
The true distance for showing fine motion target, compound fine motion frequency, to PV(ω) carries out high-pass filtering, and it is removed using prior information
Its band noise is disturbed, and obtains Doppler-apart from plane, completes the identification to target.
2. the method for parameter estimation of fine motion target is combined under a kind of non-gaussian background as claimed in claim 1, it is characterised in that
The step 2 is concretely comprised the following steps：The range-pulse domain matrix D for first respectively tieing up N × L1、D2、…DMEach row be considered as
One discrete stationary random process x (n) 1 × N of ∈, and carry out auto-correlation entropy calculating one by one to each sampled point, obtain N-1 points
Auto-correlation entropy vector vectorM=1,2 ... N-1, whereinRepresent every
The corresponding auto-correlation entropy vector of one column data, the common L row of auto-correlation entropy vector；All L row auto-correlation entropys vector is combined, is obtained
Corresponding auto-correlation entropy matrix V1、V2、….VM；Wherein m pointsAuto-correlation entropy computing formula be：
Wherein
Gaussian kernel function is represented, σ is the scale parameter value of core, x*(n-m) transposition of the x (n-m) in random process is represented.
3. the method for parameter estimation of fine motion target is combined under a kind of non-gaussian background as claimed in claim 1, it is characterised in that
The step 4 is concretely comprised the following steps：Assuming that during fine motion target imaging, imaging point is R_index in the corresponding index value of distance axis,
It is f_index1, f_index2 in the corresponding index value of Doppler frequency axle, then the calculating of the distance of target and fine motion frequency is public
Formula is as follows：
Wherein, R represents fine motion target range, f1Represent target fine motion frequency 1, f2Target fine motion frequency 2 is represented,Be away from
High Resolution, c is propagation velocity of electromagnetic wave, and B is transmitted signal bandwidth；Δ f is frequency resolution,T believes for transmitting
Number time width, L1For the Fourier transformation of slow time dimension is counted；
With reference to multichannel phase comparison monopulse method, the phase difference that target is produced in different interchannels is obtained, calculate the side of target
Parallactic angle：
Wherein, d is antenna spacing, and λ is transmission signal wavelength, with reference to the fine motion target range R for measuring before, is realized to fine motion mesh
Target is positioned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710123663.6A CN106707258A (en) | 2017-03-03 | 2017-03-03 | Multi-parameter estimation method for micro-motion target under non-Gaussian background |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710123663.6A CN106707258A (en) | 2017-03-03 | 2017-03-03 | Multi-parameter estimation method for micro-motion target under non-Gaussian background |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106707258A true CN106707258A (en) | 2017-05-24 |
Family
ID=58912116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710123663.6A Pending CN106707258A (en) | 2017-03-03 | 2017-03-03 | Multi-parameter estimation method for micro-motion target under non-Gaussian background |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106707258A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107167784A (en) * | 2017-07-05 | 2017-09-15 | 电子科技大学 | A kind of many human body target positioning and tracing methods based on multichannel phase comparison positioning |
| CN107450065A (en) * | 2017-08-07 | 2017-12-08 | 中国民航大学 | A kind of inexpensive SUAV surveillance radar |
| CN107728115A (en) * | 2017-09-11 | 2018-02-23 | 电子科技大学 | Ambient interferences suppressing method based on SVM after a kind of radar target imaging |
| CN108490427A (en) * | 2018-02-07 | 2018-09-04 | 浙江大学 | A kind of moving target indoor positioning and real-time tracing method |
| CN110161491A (en) * | 2019-06-28 | 2019-08-23 | 电子科技大学 | A kind of ranging and respiratory rate estimation method for faint life entity |
| CN111366919A (en) * | 2020-03-24 | 2020-07-03 | 南京矽典微系统有限公司 | Target detection method and device based on millimeter wave radar, electronic equipment and storage medium |
| CN111416595A (en) * | 2020-04-08 | 2020-07-14 | 北京航空航天大学 | Big data filtering method based on multi-core fusion |
| CN111609771A (en) * | 2020-04-19 | 2020-09-01 | 北京理工大学 | Laser fuse spacing method in aerosol environment |
| CN111639595A (en) * | 2020-05-29 | 2020-09-08 | 桂林电子科技大学 | Unmanned aerial vehicle micro-motion characteristic signal detection method based on weight-agnostic neural network |
| CN111708017A (en) * | 2020-05-27 | 2020-09-25 | 中国电子科技集团公司信息科学研究院 | Multi-radar joint detection method and device based on Gaussian kernel |
| CN113066539A (en) * | 2021-03-22 | 2021-07-02 | 上海商汤智能科技有限公司 | Prediction method and related device and equipment |
-
2017
- 2017-03-03 CN CN201710123663.6A patent/CN106707258A/en active Pending
Non-Patent Citations (2)
| Title |
|---|
| PAVLO MOLCHANOV ET AL.: ""Short-Range FMCW Monopulse Radar for Hand-Gesture Sensing"", 《IEEE INTERNATIONAL RADAR CONFERENCE》 * |
| 蓝方宇: ""基于微摄动与步态特征的人体探测算法研究"", 《中国优秀硕士学位论文全文数据库（电子期刊） 信息科技辑》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107167784A (en) * | 2017-07-05 | 2017-09-15 | 电子科技大学 | A kind of many human body target positioning and tracing methods based on multichannel phase comparison positioning |
| CN107450065A (en) * | 2017-08-07 | 2017-12-08 | 中国民航大学 | A kind of inexpensive SUAV surveillance radar |
| CN107728115A (en) * | 2017-09-11 | 2018-02-23 | 电子科技大学 | Ambient interferences suppressing method based on SVM after a kind of radar target imaging |
| CN107728115B (en) * | 2017-09-11 | 2020-08-11 | 电子科技大学 | SVM-based background interference suppression method after radar target imaging |
| CN108490427A (en) * | 2018-02-07 | 2018-09-04 | 浙江大学 | A kind of moving target indoor positioning and real-time tracing method |
| CN110161491B (en) * | 2019-06-28 | 2021-01-12 | 电子科技大学 | Ranging and respiratory frequency estimation method for weak life body |
| CN110161491A (en) * | 2019-06-28 | 2019-08-23 | 电子科技大学 | A kind of ranging and respiratory rate estimation method for faint life entity |
| CN111366919A (en) * | 2020-03-24 | 2020-07-03 | 南京矽典微系统有限公司 | Target detection method and device based on millimeter wave radar, electronic equipment and storage medium |
| CN111366919B (en) * | 2020-03-24 | 2022-05-13 | 南京矽典微系统有限公司 | Target detection method and device based on millimeter wave radar, electronic equipment and storage medium |
| CN111416595A (en) * | 2020-04-08 | 2020-07-14 | 北京航空航天大学 | Big data filtering method based on multi-core fusion |
| CN111416595B (en) * | 2020-04-08 | 2022-04-08 | 北京航空航天大学 | Big data filtering method based on multi-core fusion |
| CN111609771A (en) * | 2020-04-19 | 2020-09-01 | 北京理工大学 | Laser fuse spacing method in aerosol environment |
| CN111609771B (en) * | 2020-04-19 | 2021-05-18 | 北京理工大学 | Laser fuse spacing method in aerosol environment |
| CN111708017A (en) * | 2020-05-27 | 2020-09-25 | 中国电子科技集团公司信息科学研究院 | Multi-radar joint detection method and device based on Gaussian kernel |
| CN111708017B (en) * | 2020-05-27 | 2023-07-07 | 中国电子科技集团公司信息科学研究院 | Multi-radar joint detection method and device based on Gaussian kernel |
| CN111639595B (en) * | 2020-05-29 | 2022-03-18 | 桂林电子科技大学 | Unmanned aerial vehicle micro-motion characteristic signal detection method based on weight-agnostic neural network |
| CN111639595A (en) * | 2020-05-29 | 2020-09-08 | 桂林电子科技大学 | Unmanned aerial vehicle micro-motion characteristic signal detection method based on weight-agnostic neural network |
| CN113066539A (en) * | 2021-03-22 | 2021-07-02 | 上海商汤智能科技有限公司 | Prediction method and related device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106707258A (en) | Multi-parameter estimation method for micro-motion target under non-Gaussian background | |
| CN107966688B (en) | Broadband radar target speed ambiguity resolving method based on phase interference technology | |
| CN107167784B (en) | Multi-human-body target positioning and tracking method based on multi-channel phase comparison positioning | |
| CN101509972B (en) | Wideband radar detecting method for correcting correlation matrix based on high resolution target distance image | |
| CN105807267B (en) | A kind of MIMO radar extends mesh object detection method | |
| CN106226751B (en) | Maneu-vering target detection and tracking based on DP-TBD | |
| CN106443671A (en) | SAR radar moving target detecting and imaging method based on FM continuous wave | |
| CN105974376B (en) | A kind of SAR radio frequency interferences suppressing method | |
| CN108008386B (en) | A kind of distance based on single snap MUSIC algorithm is to processing method | |
| CN110568434B (en) | Multi-channel uniform acceleration SAR moving target two-dimensional speed estimation method | |
| CN107450055B (en) | High-speed maneuvering target detection method based on discrete linear frequency modulation Fourier transform | |
| CN105652246A (en) | Main lobe interference resisting method of radar based on blind source separation | |
| CN106249208B (en) | Signal detecting method under amplitude modulated jamming based on Fourier Transform of Fractional Order | |
| CN105572649A (en) | Radar target detection method based on sparse Fourier transform | |
| CN112782685B (en) | Multi-sound-source positioning and sound reconstruction method and system based on MIMO radar | |
| CN102914773A (en) | Multi-pass circumference SAR three-dimensional imaging method | |
| CN109031299B (en) | ISAR (inverse synthetic aperture radar) translation compensation method based on phase difference under low signal-to-noise ratio condition | |
| CN114594428A (en) | Intermittent sampling interference suppression method based on linear frequency modulation in pulse-to-pulse frequency coding | |
| CN112835009A (en) | Data set expansion method based on FMCW radar RAI | |
| CN103064084A (en) | Ambiguity solving method based on distance frequency domain | |
| Yu et al. | Ground moving target motion parameter estimation using Radon modified Lv's distribution | |
| CN111948618A (en) | Forward scattering target detection method and system based on satellite external radiation source | |
| CN108414992A (en) | A kind of object detection method based on phase information clutter map | |
| CN102830394B (en) | Weak target detection method based on multispectral accumulation | |
| Samczyński | Extended Generalized Chirp Transform for signal parameter estimation in bistatic passive pulse radars |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170524 |
|
| RJ01 | Rejection of invention patent application after publication |