Approximate examination of trapping sets of LDPC codes using the probabilistic algorithm

The performance of LDPC codes decoded by iterative algorithms depends on the structural properties of their underlying Tanner graphs. For general memoryless channels, error patterns dominating the bit and frame error probabilities at the error floor region are termed trapping sets. In this paper, we propose an effective method for finding small-size trapping sets of LDPC codes. In the proposed method, a probabilistic algorithm to find low-weight codewords is applied to finding small trapping sets of LDPC codes. Furthermore, we show numerical results of examining small trapping sets of (504, 252) and (1008, 504) LDPC codes.