Novel near maximum likelihood soft decision decoding algorithm for linear block codes

The authors present a novel near maximum likelihood (ML) soft decision decoding (SDD) algorithm for linear block codes. The proposed algorithm can be subdivided into three distinct algorithms, each achieving a specific objective. The first algorithm achieves near ML decoding performance by utilising channel measurement information. The second and third algorithms maintain the improved decoding performance achieved, while at the same time reducing both the number of decodings and complexity required. The resultant algorithm is both general in nature and able to offer significant performance improvements. The theoretical results are verified by computer simulation.