A parallel sliding window map decoder for a PR4 magnetic recording

Summary form only given. With regard to the design of turbo decoders, there are many issues. These include memory size, decoding delay, complexity, etc. A turbo decoder usually requires a huge memory size and the recursive computation in the MAP-based decoding of turbo codes usually causes a significant delay in decoding. The sliding window (SW) MAP decoder can reduce memory size. Parallel or pipelined decoding schemes can reduce decoding delay and improve throughput. We propose a parallel SW MAP decoder. In this scheme, we apply the SW MAP decoder to the sub-blocks in a parallel decoding scheme to reduce both the necessary memory size and the decoding delay. The scheme can be applied when the block length is long.