Skip to Main Content
In this paper we propose a design-method for irregular LDPC codes in the context of channels, that can not sufficiently be described with the known parametric channel-models. We introduce a very flexible histogram-based channel-model. Furthermore a design-method based on the downhill simplex optimization is explained for the design of irregular LDPC codes in the context of histogram-based channel-models. We also propose an extension for soft-decision based LDPC decoders and explain the simulation of LDPC codes considering histogram-based channels-models. Following results based on a 2-state Markov-model prove the effectiveness of our design-method, the description of the channel and the decoding based on histograms.