Radar radiation source signal depth pulse internal feature automatic extraction method

Abstract

The invention provides a radar radiation source signal depth pulse internal feature automatic extraction method. The method comprises the following steps: firstly, applying a specific sparse constraint to an auto-encoder to obtain a sparse auto-encoder; then, the sparse auto-encoder is optimized, a training scheme of the sparse auto-encoder is determined, the coding layer parameters are used for automatically extracting the radar signal depth intra-pulse characteristics, and the extracted characteristics can better achieve the classification and recognition of the radar radiation source signals within a large signal-to-noise ratio range.

Description

technical field [0001] The invention relates to the field of radar information processing, in particular to a method for automatically extracting deep intrapulse features of radar radiation source signals. Background technique [0002] The key to effectively sorting and identifying radar signals is to extract features that can reflect the essence of the signal, and the autoencoder (autoencoder, AE) under the deep learning theory aims at reconstructing the original input at the output layer, because it does not It needs additional supervision information to be able to extract the distributed features of the data, and can avoid the implicit subjectivity when designing features, and has become a hot direction that has attracted people's attention in recent years. In 2006, Hinton improved the prototype autoencoder structure and obtained a deep autoencoder (deep autoencoder, DAE). Bengio deepened the deep autoencoder and proposed a sparse autoencoder (sparse autoencoder, SAE), wh...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims at the problem of insufficient objectivity due to dependence on prior knowledge when extracting intrapulse features of radar signals, and pro

Images