DocumentCode
539866
Title
DAVINCI Non-Binary LDPC codes: Performance and complexity assessment
Author
Gutierrez, Ismael ; Bacci, Giacomo ; Bas, Joan ; Bourdoux, André ; Gierszal, Henryk ; Mourad, Alain ; Pleftschinger, Stephan
Author_Institution
Samsung Electron. Res. Inst., Staines, UK
fYear
2010
fDate
16-18 June 2010
Firstpage
1
Lastpage
8
Abstract
Non-Binary Low Density Parity Check (NB-LDPC) codes are expected to reduce the performance gap between binary LDPC and Turbo Codes for both short and medium length codewords, retaining in both cases the best of each. This paper presents the performance evaluation of NB-LDPC when compared to state-of-art channel coding techniques, namely Convolutional Turbo Codes (CTC) and Quasi-Cyclic LDPC. The performance evaluation is twofold: first at the link level and then at system level. In both analyses, it has been shown that the NB-LDPC codes outperform the benchmark codes giving a Signal to Noise Ratio gain up to 0.5 dB and improving the average throughput 5%. Besides, a complexity analysis is also provided for NB-LDPC showing that, using a low complexity decoder, the complexity is approximately one order of magnitude above that required in case of CTC or QC-LDPC.
Keywords
channel coding; computational complexity; parity check codes; performance evaluation; turbo codes; DAVINCI nonbinary LDPC codes; channel coding techniques; complexity analysis; complexity assessment; complexity decoder; convolutional turbo codes; low density parity check codes; medium length codewords; performance assessment; performance evaluation; quasi-cyclic LDPC; AWGN; Complexity theory; Decoding; Encoding; Gain; Parity check codes; Throughput; DBTC; FEC; IMT-Advanced; LDPC; LTE; NB-LDPC; WiMAX;
fLanguage
English
Publisher
ieee
Conference_Titel
Future Network and Mobile Summit, 2010
Conference_Location
Florence
Print_ISBN
978-1-905824-16-8
Type
conf
Filename
5722335
Link To Document