Novel Blind Identification of LDPC Codes Using Average LLR of Syndrome a Posteriori Probability

Blind signal processing methods have been very popular recently since they can play crucial roles in the prevalent cognitive radio research. Blind encoder identification has drawn research interest lately. In this paper, we would like to tackle the blind identification of binary low-density parity-check (LDPC) codes for binary phase-shift keying (BPSK) signals. We propose a novel blind identification system which consists of three components, namely expectation-maximization (EM) estimator for signal amplitude and noise variance, log-likelihood ratio (LLR) estimator for syndrome a posteriori probabilities, and maximum average LLR detector. Monte Carlo simulation results demonstrate that our proposed blind LDPC encoder identification scheme is very promising even for low signal-to-noise ratio conditions.