Blind Identification of Joint Source-channel Coding Based on Deep Learning

Blind identification technology has received significant attention in recent years. In this paper, we address the blind identification of low-density parity check (LDPC) code types in joint source-channel coding (JSCC). This paper proposes a blind identification algorithm for LDPC codes based on deep learning. According to the algorithm, the received encoded sequence is treated as a text sentence, and neural networks are used to identify the encoding method applied to the sequence. We consider two types of LDPC codes in JSCC: Double Protograph LDPC (DP-LDPC) and Double Spatially-Coupled LDPC (DSC-LDPC). Simulation results demonstrate that the proposed algorithm can accurately identify between these two types of codes.