Closed-Chains Error Correction Technique for Turbo Product Codes

In this work, we propose a new decoding algorithm to correct closed-chains error patterns in hard-input hard-output (HIHO) turbo product codes (TPCs). The proposed technique is based on correlating the horizontal and vertical component codes to estimate the location of the erroneous bits in the closed-chain of errors, then erasure decoding is used to correct the identified bit errors. Simulation results demonstrated that, for particular codes, a noticeable coding gain improvement of about 1.5 dB can be achieved when compared to the standard sequential HIHO decoding and about 0.8 dB when compared to the non-sequential HIHO decoding. The computational complexity of the proposed decoder can be substantially reduced at moderate and high signal-to-noise ratios by stopping the iterative process when it is not more beneficial to perform further iterations.