Systems Engineering and Electronics ›› 2024, Vol. 46 ›› Issue (11): 3893-3900.doi: 10.12305/j.issn.1001-506X.2024.11.31

• Communications and Networks • Previous Articles     Next Articles

Low-complexity iterative synchronization algorithm incorporating DNN for joint demodulation decoding

Yongsheng CUI1,2, Yafeng ZHAN1,*, Taiyi CHEN1,2, Xin FANG3   

  1. 1. Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing 100084, China
    2. Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
    3. Electric Power Research Institute, State Grid Jiangsu Electric Power Co., Ltd., Nanjing 210024, China
  • Received:2023-10-27 Online:2024-10-28 Published:2024-11-30
  • Contact: Yafeng ZHAN

Abstract:

In many wireless communication scenarios, including satellite communication, deep space communication, and covert communication, the received signal is markedly weak due to inherent limitations in factors such as transmission power and transmission distance. The existing joint demodulation decoding iterative synchronization algorithm applies channel coding gain to the entire signal reception process, which can effectively reduce the synchronization threshold of the receiver, but the computational complexity is high. This paper makes use of the morphological consistency characteristics of the iterative received target function and puts forward a synchronization optimization strategy based on deep neural network (DNN). In comparison to traditional iterative synchronization methods, this strategy can reduce computational complexity by 24% at a bit error rate of 1e-5. This research provides a new direction for the engineering application of iterative reception technology in higher data rate scenarios, while demonstrating the potential of deep learning in addressing complex communication environment issues.

Key words: joint demodulation decoding, iterative synchronization, deep neural network (DNN), maximum likelihood estimation

CLC Number: 

[an error occurred while processing this directive]