Anti-jamming decision method for anti-ship radar seeker based on embedding deep Q-network

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

With the increasing complexity of electronic countermeasures, radar seekers face challenges from complex electronic interference. This paper examines the scenario where an anti-ship radar seeker counters shipborne jamming aircraft, considering both common suppressive and deceptive jamming. Considering the Markov properties inherent in the radar anti-jamming decision-making procedure, this paper constructs a theoretical framework based on Markov Decision Process (MDP) and designs corresponding state space, action space, and reward function. This paper also proposes an improved Deep Q-Network algorithm-Embedding Deep Q-Network, which incorporates embedding layers to handle discrete input features, improving training performance. Simulation experiments confirm the effectiveness of the proposed method.

Original languageEnglish
Title of host publicationFifth International Conference on Signal Image Processing and Communication, ICSIPC 2025
EditorsShou Feng, Zhihao Zhang
PublisherSPIE
ISBN (Electronic)9781510694842
DOIs
Publication statusPublished - 8 Sept 2025
Externally publishedYes
Event5th International Conference on Signal Image Processing and Communication, ICSIPC 2025 - Zhengzhou, China
Duration: 16 May 202518 May 2025

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume13800
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference5th International Conference on Signal Image Processing and Communication, ICSIPC 2025
Country/TerritoryChina
CityZhengzhou
Period16/05/2518/05/25

Keywords

  • anti-jamming
  • anti-ship radar seeker
  • Deep Q-Network
  • Embedding layer

Fingerprint

Dive into the research topics of 'Anti-jamming decision method for anti-ship radar seeker based on embedding deep Q-network'. Together they form a unique fingerprint.

Cite this