Detailed evaluation of error floors of LDPC codes using the probabilistic algorithm

The performance of LDPC codes decoded by iterative algorithm depends on the structural properties of their underlying Tanner graphs. For discrete memoryless channels, error patterns dominating the frame error rate (FER) in the error floor region are termed trapping set. We have shown approximate examinations of small trapping sets of LDPC codes using the probabilistic algorithm. In this paper, we propose an efficient method for the detailed evaluation of the FER of LDPC codes in the error floor region. The accuracy of the FER lies on the failure probability of the number of trapping sets determined by our probabilistic algorithm and the FER estimated by the importance sampling.