Where academic tradition
meets the exciting future

You are here: TUCS > PUBLICATIONS > Publication Search > Decoding of DVB-T2 LDPC Codes ...

Decoding of DVB-T2 LDPC Codes on a Tile Processor: Optimisations and Performance Comparisons

Sudeep Kanur, Stefan Grönroos, Kristian Nybom, Jerker Björkqvist, Johan Lilius, Decoding of DVB-T2 LDPC Codes on a Tile Processor: Optimisations and Performance Comparisons . In: John Glossner, Stephanie Hamill, Lee Pucker (Eds.), Proceedings of SDR-WinnComm 2014: Wireless Innovation Conference on Wireless Communications Technologies and Software Defined Radio, Wireless Innovation Forum Conference on Wireless Communications Technologies and Software Defined Radio, 75–85, The Wireless Innovation Forum, 2014.

Abstract:

Multiple code rates and codeword lengths of 16200 bits and 64800 bits are used in broadcasting media by second generation Digital Video Broadcasting standard (DVB). Software defined radios can benefit from real-time decod-
ing of these codes on general purpose computing hardware. However, the iterative decoding algorithm of Low Density Parity Check (LDPC) codes is an NP-complete problem. To this end, multicore processors can be used to decode LDPC codes in parallel and the network bottleneck caused by irregular memory access can be handled
by balancing computational and communication load between the processors. Presented in this paper are the algorithms and the data structures used to implement log-domain decoding of the long LDPC codes specified by the DVB standards on a TILEPro64 multicore processor.
Two implementation methods are discussed and a maximum throughput of 1.09 Mbps and 970 kbps for rate 1/2 and 4/5 LDPC codes were obtained respectively. Despite the non-real time throughput, valuable insights were ob-
tained from the approach. The paper details the approach used and analyses the virtues and shortcomings of the implementation methods.

BibTeX entry:

@INPROCEEDINGS{inpKaGrNyBjLi14a,
  title = {Decoding of DVB-T2 LDPC Codes on a Tile Processor: Optimisations and Performance Comparisons },
  booktitle = {Proceedings of SDR-WinnComm 2014:  Wireless Innovation Conference on Wireless Communications Technologies  and Software Defined Radio},
  author = {Kanur, Sudeep and Grönroos, Stefan and Nybom, Kristian and Björkqvist, Jerker and Lilius, Johan},
  series = {Wireless Innovation Forum Conference on Wireless Communications Technologies and Software Defined Radio},
  editor = {Glossner, John and Hamill, Stephanie and Pucker, Lee},
  publisher = {The Wireless Innovation Forum},
  pages = {75–85},
  year = {2014},
  keywords = {LDPC, Tilera},
}

Belongs to TUCS Research Unit(s): Embedded Systems Laboratory (ESLAB)

Publication Forum rating of this publication: level 1

Edit publication