Radar signal anti-interference low-loss recovery method of coding and decoding convolutional neural network

Abstract

The invention discloses a radar signal anti-interference low-loss recovery method of a coding and decoding convolutional neural network, and mainly solves the problem of target information loss after radar signal interference suppression in the prior art. The method comprises the following steps: 1) constructing a time-frequency signal low-loss recovery network; 2) generating a training data set composed of time frequency and mask data; 3) inputting data in the training data set into the time-frequency signal low-loss recovery network, and performing iterative training on parameters of the time-frequency signal low-loss recovery network by using a momentum method to reduce a loss function value; and 4) carrying out time-frequency domain anti-interference on the interfered radar echo, inputting the echo after anti-interference into the trained time-frequency signal low-loss recovery network to obtain a recovered time-frequency domain echo, and converting the recovered time-frequency domain echo into a time domain. The method has the capability of performing low-loss repair on defect signals after interference suppression in a time-frequency domain, has universality and robustness for time-frequency aggregation interference, and can be used for interference resistance of the synthetic aperture radar.

Description

technical field [0001] The invention belongs to the technical field of radar signal processing, and further relates to a radar signal anti-jamming and low-loss recovery method, which can be used for anti-jamming synthetic aperture radar. Background technique [0002] Radar systems can achieve high-precision detection of stationary or moving targets in complex environments. However, in the complex electromagnetic interference, the radar system is easily affected by various complex electromagnetic interference such as active interference signals, wireless communication signals and other radar signals in the working frequency band, which seriously restricts the high-resolution imaging effect of the radar. Therefore, the anti-jamming technology is radar. One of the core problems of signal processing technology. [0003] Since the coupling between the active mainlobe interference and the target in the time-frequency domain is significantly reduced, the active mainlobe interferen...

AI Technical Summary

Problems solved by technology

[0004] However, in this method, since the active interference such as slice forwarding suppression is related to the echo, the masking of the interference will lead to the loss of information of the echo in the same direction as the interference.

Even when the target and interference features are highly overlapped in the time-frequency two-dimensional space, the time-frequency filter anti-jamming processing may completely suppress the target echo, resulting in radar echo defects and loss of target information, which cannot meet the needs of radar imaging and subsequent target detection. identified needs

Images