An efficient decoder for turbo product codes built using two-error-correcting eBCH codes

In this paper, we propose a method to reduce the number of test patterns (TPs) decoded in the Chase II algorithm when turbo-decoding product codes are constructed with two-error-correcting extended Bose-Chaudhuri-Hocqenghem (eBCH) codes. We classify TPs by their values and the relationship of syndromes, and then reject those TPs that have the same codeword. The method needs not to check whether errors are located at the least reliable bits (LRBs) positions to ensure that there are no same codewords decoded. For two-error-correcting eBCH (64,51,2), the simulation results show that over 50% TPs need not be decoded for moderate signal to noise ratios (SNRs) without any decoding performance loss.