Design of short irregular LDPC codes based on a constrained Downhill-Simplex method [articol]

Abstract

In this paper we present an optimization procedure for the design of irregular Low-Density Parity-Check (LDPC) codes with short blocklength. For the optimization of the Symbol-Node Degree-Distribution (SNDD) of an irregular LDPC code we adapted the complete DHS-algorithm to the constrained problem. This is in contrast to [1], where the authors only applied a simplified version of the Downhill-Simplex (DHS) method. Furthermore our optimization procedure comprises several rounds including differently initialized simplezes in order to prevent from converging to a local minimum. Compared to simulation-results based on the simplified DHS-method provided in [1] the performance of our designed LDPC code shows a gain up to 0.3dB for the Bit-Error-Ratio (BER) and 0.2dB for the Word-Error-Ratio (WER).