LDPC Encoder Identification via Belief Propagation for Integrated Sensing and Communication Systems

Channel coding identification has attracted significant attention since it finds crucial applications in some key techniques for integrated sensing and communication (ISAC), e.g., interference cancellation and adaptive modulation and coding. In this paper, we propose a channel coding identification...

Full description

Saved in:
Bibliographic Details
Published in:IEEE International Conference on Communications workshops pp. 1094 - 1099
Main Authors: Wang, Hongyu, Wang, Fanggang, Liu, Yu
Format: Conference Proceeding
Language:English
Published: IEEE 28.05.2023
Subjects:
belief propagation
Channel coding identification
Conferences
integrated sensing and communication
low-density parity-check codes
maximum likelihood
Measurement
Modulation
Parity check codes
Performance gain
Probability
Symbols
ISSN:2694-2941
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Channel coding identification has attracted significant attention since it finds crucial applications in some key techniques for integrated sensing and communication (ISAC), e.g., interference cancellation and adaptive modulation and coding. In this paper, we propose a channel coding identification algorithm for low-density parity-check (LDPC) codes that exploits the belief propagation (BP) algorithm to improve the identification performance in the ISAC system. A maximum likelihood (ML) identifier is developed to determine the unknown LDPC encoder. In contrast to the existing algorithms, a novel log-likelihood function is proposed as the decision metric of the ML identifier, which is characterized by the conditional probabilities of the transmitted symbol. Particularly, these conditional probabilities are calculated accurately via the BP algorithm, enhancing the identification performance. Furthermore, we extend the proposed algorithm to the asynchronous scenario based on the designed decision metric. Numerical results show that the proposed algorithm achieves significant identification performance improvements compared with the existing algorithms.
ISSN:2694-2941
DOI:10.1109/ICCWorkshops57953.2023.10283632