Conference Paper

Protograph-Based LDPC Convolutional Codes for Correlated Erasure Channels

Univ. of California San Diego, La Jolla, CA, USA
DOI: 10.1109/ICC.2010.5502364 Conference: Communications (ICC), 2010 IEEE International Conference on
Source: IEEE Xplore

ABSTRACT We consider terminated LDPC convolutional codes (LDPC-CC) constructed from photographs and explore the performance of these codes on correlated erasure channels including a single-burst channel (SBC) and Gilbert-Elliott channel (GEC). We consider code performance with a latency-constrained message passing decoder and the belief propagation decoder. We give theoretical bounds on the code efficiency over the SBC and describe a construction that achieves this bound.We show that the designed codes with belief propagation (BP) decoding perform as well as the regular LDPC-CCs presented in the literature on the binary erasure channel (BEC) and the GEC, while achieving significant gains on the SBC. In the case of windowed decoding, our codes perform much better than the best known regular LDPC-CCs over the BEC and the GEC, with very low decoding latencies.

0 Followers
 · 
191 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Low implementation complexity, low-delay, and close-to-optimal performance of spatially-coupled LDPC codes over a wide variety of channels make them a very good candidate for upcoming wireless communication standards. However, due to the nature of the sliding window decoding architecture that is used to decode these codes, the associated error performance is considerably degraded over channels with memory, such as the burst erasure channel. In this case, using a block interleaver to break up the effects of the channel memory is not a viable option, since a block interleaver introduces a large amount of delay to the communication system and therefore takes back many of the advantages of using a sliding window decoder. In this paper, a reduced-delay communication system employing a convolutional interleaver is proposed. This scheme benefits from the inherent convolutional nature of the spatially-coupled codes and matches their structure with a low-delay convolutional interleaver. Thus, the resulting communication system exhibits very low overall delay. The performance of the proposed communication system is analyzed using the density evolution technique. Specifically, single-memory (minimum decoding delay), asymptotically (3,6)-regular spatially-coupled LDPC code ensembles are considered in the presence of burst erasures and the performance improvement of using a convolutional interleaver is demonstrated as a function of added interleaving delay in terms of iterative decoding thresholds.
    2012 International Symposium on Signals, Systems and Electronics (ISSSE); 10/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Desirable characteristics of spatially-coupled low-density parity-check (LDPC) codes including low implementation complexity, low-delay, and close-to-optimal performance over a wide variety of channels make them a noticeable candidate for the next-generation wireless communication standards. However, error performance of the sliding window decoding scheme used to decode these codes is considerably degraded over channels with memory, such as the burst erasure channel. Traditionally, communication over channels with memory is achieved through the use of block interleavers but as it introduces a large amount of delay, it is not always a viable option. In this paper, employing a convolutional interleaver which benefits from the inherent convolutional nature of the spatially-coupled codes with very low overall delay is considered. The performance of the proposed combination is analyzed using the density evolution technique. The performance improvement of using a convolutional inter-leaver is demonstrated as a function of the added interleaving delay in terms of iterative decoding thresholds.
    Signal Processing and Communications Applications Conference (SIU), 2013 21st; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Spatially-coupled LDPC codes have recently gained considerable amount of research interest because of their implementation advantages. In addition to their superior error performance over several communication channels, they are capable of low-delay decoding, achieved via a sliding window decoding architecture. On the other hand, window decoding is highly vulnerable against data transmission over correlated channels that introduce correlation among the received symbols. In this paper, we evaluate the error performance of two novel approaches that have been proposed in the relevant literature for robust spatially-coupled-coded data transmission over the correlated channels. In this scope, we consider the correlated Nakagami-m fading channel and show via simulations that convolutional-interleaver-based method, while maintaining the low-decoding delay, gives a better error performance.
    Signal Processing and Communications Applications Conference (SIU); 04/2014

Full-text (4 Sources)

Download
60 Downloads
Available from
May 31, 2014