Deep network model for radar signal sorting

Abstract

The invention relates to a deep network model for radar signal sorting. The deep network model comprises a radar signal sorting network obtained by training a pre-training model by using a training set, and the radar signal sorting network comprises a local feature learning sub-network, a cross-domain feature learning sub-network, a feature fusion module and a classifier; a to-be-classified radar signal is used as an input signal to be subjected to feature vector construction and then is respectively input to the local feature learning sub-network and the cross-domain feature learning sub-network, and the local feature learning sub-network is responsible for learning radar signal features at the current moment. The cross-domain feature learning sub-network is responsible for learning a sequential relationship between a radar signal at a current moment and radar signals at other moments, the feature fusion module completes fusion of local features and sequential features, and final features obtained after fusion are classified by a classifier. According to the invention, accurate classification of various radar signals can be completed, the sorting work of the radar signals is efficiently realized, and meanwhile, the deep network model has relatively high robustness and high efficiency.

Description

technical field [0001] The invention relates to the technical field of radar signal sorting, in particular to a deep network model for radar signal sorting. Background technique [0002] With the development of modern science and technology and the wide application of high technology in military warfare, the dependence on electronic equipment has increased sharply. Electronic countermeasures have become a key factor affecting the outcome of a war. Radar plays a vital role in modern warfare. , is the key equipment of modern weapons and equipment such as land-based, ship-borne, airborne and missiles, and also makes radar countermeasure technology a top priority in the development of national defense. The radar signal sorting method is the brain and core technology of radar reconnaissance equipment, and is the premise and basis for identifying, threatening, and jamming enemy radars. Radar signal sorting is an NP-complete problem, especially in complex electromagnetic environmen...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the shortcomings of traditional radar signal sorting technologies such as poor precision and low efficiency, and provide a deep network model for radar signal sorting. By constructing information-rich feature vectors, the radar signal sorting network can be trained The model learns local features and cross-domain feature information, so that the radar signal sorting network can accurately classify radar signals

Images