Time-frequency domain interference suppression method based on automatic encoder

Abstract

The invention discloses a time-frequency domain interference suppression method based on an automatic encoder, and mainly solves the problem of weak universality of active main lobe interference in the prior art. The implementation scheme is as follows: 1) constructing a training data set; 2) building an automatic encoder network model; 3) training the automatic encoder network by using the training data set; 4) inputting the training data set into the trained automatic encoder network to calculate a segmentation threshold; 5) preprocessing the interference data into a form capable of being input into a network; (6) calculating an input and output difference value of interference data by using the trained automatic encoder network, and segmenting the difference value by using a segmentation threshold to obtain an interference positioning result; and 7) performing interference suppression on the interference data according to an interference positioning result. The method has the advantages of being high in adaptability to complex active interference of different interference forms, accurate in interference positioning result and low in target information loss, and can be used for synthetic aperture radar imaging.

Description

technical field [0001] The invention belongs to the technical field of radar signal anti-jamming, and further relates to a time-frequency domain interference suppression method, which can be applied to synthetic aperture radar anti-jamming. Background technique [0002] Since active main lobe jamming can greatly suppress radar imaging performance, anti-jamming technology is one of the core issues of radar signal processing technology. Traditional anti-jamming methods, such as adaptive beamforming technology, while suppressing active main lobe interference, will cause problems such as loss of target information in the same orientation as the interference, main lobe distortion and offset, etc., which cannot meet the imaging needs. On the other hand, during the synthetic aperture time, the pattern of strong pulsed active main lobe interference such as slice forwarding suppression and multiple false targets is flexible and changeable, and it exhibits strong coupling with the tar...

AI Technical Summary

Problems solved by technology

[0004] However, this kind of method is based on the assumption that the interference part in the time-frequency domain is the highest energy part, and suppresses the interference according to the energy, so it can only locate the interference part with the highest energy in the time-frequency domain. Other interference parts with relatively lower energy cannot be suppressed

At the same time, because this method cannot distinguish whether the echo is interfered, the same suppression operation will be taken for the echo that has not received interference, resulting in the loss of target information

Images