Unlock instant, AI-driven research and patent intelligence for your innovation.

STAP training sample selection method based on system identification

A training sample and system identification technology, applied in the radar field, can solve problems such as waveform dissimilarity, missing similarity, and low available samples

Active Publication Date: 2017-09-19
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF3 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When the covariance matrices of the two signals are the same, the waveforms of the two signals may be completely dissimilar, so there may be samples with completely dissimilar waveforms in the selected training samples, and the traditional sample selection method based on waveform similarity tends to miss a large number of Available samples with low similarity

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • STAP training sample selection method based on system identification
  • STAP training sample selection method based on system identification
  • STAP training sample selection method based on system identification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0079] With sky-wave radar operating frequency f 0 =18.3MHz, pulse repetition period T=12ms, pulse accumulation number M=512, coherent integration time CIT=6.144s. In the echo data, it is known that the 435th range unit to be detected has a target with a Doppler frequency of -5.859, and its spectrum is as follows figure 2 shown. image 3 is the similarity between each distance unit data and the unit to be detected, Figure 4 Indicates the normalized output variance of each distance unit after filtering by the trained neural network.

[0080] If the training samples are selected based on similarity, theoretically, samples with a correlation coefficient close to 1 should be selected as much as possible. However, subject to the limitation of the number of optional samples, according to the reference "Zhang X, Yang Q, Deng W. Weak target detection within the nonhomogeneous ionospheric clutter background of HFSWR based on STAP [J]. International Journal of Antennas and Propagat...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an STAP training sample selection method based on system identification and belongs to the technical field of radar. The method comprises steps of firstly giving out a sample selection model based on system identification; based on this, further using the neural network to identity a CUT clutter model; using the identified model to carrying out filtering on other distance units; and finally, selecting a sample similar to a CUT clutter covariance matrix. According to the invention, a problem of lack of training samples in the STAP is solved; the estimated CUT clutter covariance matrix is quite accurate; and the clutter restraining performance is improved.

Description

technical field [0001] The invention belongs to the technical field of radar, and in particular relates to a method for selecting STAP training samples based on system identification. Background technique [0002] Space-time adaptive processing (STAP) is a key technology in radar and communication signal processing, and is widely used in radar and communication signal processing. STAP requires the design of the optimal weight vector , the output signal-to-noise ratio (SCNR) is maximized. Among them, s is the space-time steering vector of the target signal, R CUT is the clutter covariance matrix of the range unit to be detected. However in practice R CUT is unknown and needs to be estimated through the selected training samples. Let the clutter covariance matrix of the training samples be R TS , then the requirements for the selected training samples are: [0003] (1) Should satisfy R TS = R CUT . [0004] (2) There must be enough training samples, because the numb...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/08G06K9/00G01S7/41
CPCG06N3/08G01S7/414G01S7/417G06F2218/04
Inventor 胡进峰鲍伟伟曹健姚冯陈卓蔡雷雷
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA