Systems Engineering and Electronics ›› 2025, Vol. 47 ›› Issue (1): 12-21.doi: 10.12305/j.issn.1001-506X.2025.01.02

• Electronic Technology • Previous Articles     Next Articles

Neural network-based self-interference cancellation approach for full-duplex jammer

Hongyu ZHU1,2, Weidong HU1,2, Chao WANG1,*, Qingzhan SHI1, Ximeng ZHANG1,2, Naichang YUAN1,2   

  1. 1. College of Electronic Science and Technology, National University of Defense Technology, Changsha 410003, China
    2. National Key Laboratory of Automatic Target Recognition, National University of Defense Technology, Changsha 410003, China
  • Received:2023-10-17 Online:2025-01-21 Published:2025-01-25
  • Contact: Chao WANG

Abstract:

In view of the problem of cancellation the complex and strong nonlinear self-interference signals, the paper analyses the composition and non-linear characteristics of the self-interference signals generated by the leakage in full-duplex systems. And a digital domain neural network approach is proposed based on the idea of multi-level isolation and self-interference cancellation. The neural network quickly learns and perceives the parameters of the self-interference channel model, effectively canceling the leaked self-interference signals. Simulation results show that compared to traditional linear cancellation and digital cancellation methods, the proposed approach can rapidly adapt to changes in radar signals, exhibiting stronger self-interference suppression capability and providing a technological pathway for enhancing the performance of full-duplex jammer.

Key words: full-duplex jammer, high-power signal, non-linear characteristics, changes in radar signals, self-interference suppression capability

CLC Number: 

[an error occurred while processing this directive]