Rate-compatible punctured low-density parity-check codes with short block lengths

• Published in: IEEE transactions on information theory Vol. 52; no. 2; pp. 728 - 738
• Main Authors: Jeongseok Ha, Jaehong Kim, Klinc, D., McLaughlin, S.W.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.02.2006; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Additive white noise; Additives; Applied sciences; Automatic repeat request; Blocking; Channels; Codes; Coding, codes; Communication channels; Comparative analysis; Complexity; Decoding; Distribution; Error correction codes; Exact sciences and technology; Gaussian; Information theory; Information, signal and communications theory; Low-density parity-check (LDPC) codes; Noise; Parity check codes; Performance loss; Piercing; Protocols; puncturing; rate-adaptive; Redundancy; Reliability theory; short block lengths; Signal and communications theory; Systems, networks and services of telecommunications; Telecommunications; Telecommunications and information theory; Time-varying channels; Transmission and modulation (techniques and equipments); AWGN channels; puncturing; rate-adaptive; short block lengths; Data transmission; Parity check codes; Error correcting code; Performance analysis; Adaptive method; Low-density parity-check (LDPC) codes
• DOI: 10.1109/TIT.2005.862118
• ISSN: 0018-9448

We consider the problem of rate-compatible puncturing of low-density parity-check (LDPC) codes over additive white Gaussian noise (AWGN) channels. In previous work, it was shown that "good" puncturing distributions exist for LDPC codes but the code length was large. In this correspondence, we give a procedure for determining the puncturing distributions for LDPC codes with short block lengths (a few thousand bits) and show that careful puncturing can produce good performance. We compare the performance of the rate-compatible punctured LDPC codes with dedicated LDPC codes across a range of rates and see that the rate-compatible codes have a favorable complexity/performance tradeoff.