Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.301590
Title: Low complexity bit-level soft-decision decoding for Reed-Solomon codes
Author: Oh, Min-seok
ISNI:       0000 0001 3455 5520
Awarding Body: University of Surrey
Current Institution: University of Surrey
Date of Award: 1999
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Reed-Solomon codes (RS codes) are an important method for achieving error-correction in communication and storage systems. However, it has proved difficult to find a soft-decision decoding method which has low complexity. Moreover, in some previous soft-decision decoding approaches, bit-level soft-decision information could not be employed fully. Even though RS codes have powerful error correction capability, this is a critical shortcoming. This thesis presents bit-level soft-decision decoding schemes for RS codes. The aim is to design a low complexity sequential decoding method based on bit-level soft- decision information approaching maximum likelihood performance. Firstly a trellis decoding scheme which gives easy implementation is introduced, since the soft-decision information can be used directly. In order to allow bit-level soft-decision, a binary equivalent code is introduced and Wolf's method is used to construct the binary-trellis from a systematic parity check matrix. Secondly, the Fano sequential decoding method is chosen, which is sub-optimal and adaptable to channel conditions. This method does not need a large amount of storage to perform an efficient trellis search. The Fano algorithm is then modified to improve the error correcting performance. Finally, further methods of complexity reduction are presented without loss of decoding performance, based on reliability-first search decoding using permutation groups for RS codes. Compared with the decoder without permutation, those schemes give a large complexity reduction and performance improvement approaching near maximum likelihood performance. In this thesis, three types of permutation, cyclic, squaring and hybrid permutation, are presented and the decoding methods using them are implemented.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.301590  DOI: Not available
Keywords: Error-correction; Communication; Storage systems
Share: