Radar anti-interference intelligent decision-making method based on backtracking DQN

The invention relates to a radar anti-interference intelligent decision-making method based on backtracking DQN, and the method comprises the steps: determining a corresponding interference state according to the interference type of a received echo signal; according to the interference state, the t...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: LIAO GUISHENG, HUANG LEI, LAN LAN, XU JINGWEI, ZHANG XIANG, JEON YOUNG-HWI, ZHU SHENGQI, LI XIMIN
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention relates to a radar anti-interference intelligent decision-making method based on backtracking DQN, and the method comprises the steps: determining a corresponding interference state according to the interference type of a received echo signal; according to the interference state, the trained DQN algorithm neural network is utilized to traverse the anti-interference methods in the anti-interference method library, and radar anti-interference intelligent decision is realized; wherein the DQN algorithm neural network is trained by using a method for updating a Q value based on backtracking, and the trained DQN algorithm neural network is obtained. According to the method, a radar anti-interference technology and a DQN algorithm in reinforcement learning are combined, a backtracking DQN algorithm is provided, a backtracking update Q value is added in an update formula of a target network, and the introduction of the backtracking update Q value can reduce the influence of optimal data change, improve