Reduced complexity hard‐ and soft‐input BCH decoding with applications in concatenated codes

Error correction coding for optical communication and storage requires high rate codes that enable data throughput low residual errors. Recently, different concatenated schemes were proposed are based on binary BCH with error correcting capabilities. In this work, low-complexity hard- soft-input decoding methods such investigated. We propose three concepts to reduce the complexity of decoder. For algebraic we demonstrate Peterson's algorithm can be more efficient than Berlekamp–Massey single, double, triple codes. an inversion-less version a corresponding architecture. Furthermore, approach combines hard-input bit-flipping decoding. An acceptance criterion is utilised determine reliability estimated codewords. many received codewords stopping indicates hard-decoding result sufficiently reliable, costly omitted. To memory size soft-values, decoder stores only positions soft values small number code symbols. This method significantly reduces requirements has little adverse effect performance.