Research On Terminated Protograph-based LDPC Convolutional Codes Algorithm And Hardware Implementation

Terminated Protograph-based LDPC Convolutional Codes(TP-LDPC-CCs),combining with protograph-based LDPC(P-LDPC)codes and LDPC convolutional codes,possess variable encoding constructed schemes and excellent decoding performance,thereby having a wide field of application.This dissertation investigates the algorithm and hardware implementation of TP-LDPC-CCs.The main works are summarized as follows:1.The key step of constructing TP-LDPC-CCs,denoted as edge spreading,is studied.The effect of the existing edge spreading methods on the performance of TP-LDPC-CCs is investigated.Then a novel edge spreading algorithm based on protograph-based extrinsic information transfer(P-EXIT)algorithm and differential evolution(DE)algorithm is proposed for TP-LDPC-CCs over AWGN(Additive White Gaussian Noise)channel.The P-EXIT analysis and simulation results indicate that the proposed edge spreading algorithm can achieve the best edge spreading mode and possess the best performance.2.Considering that the windowed decoding algorithm has low latency,hardware implementation of the windowed decoding scheme for TP-LDPC-CCs is studied.The designed windowed decoder has a partially parallel architecture and can achieve low latency.Moreover,the iterative process of windowed decoder is realized by the normalized Min-Sum algorithm based on the TDMP(Turbo decoding message passing),thereby reducing the consumption of hardware resources.And the designed windowed decoder is a general hardware architecture,which can be applied for different TP_LDPC-CCs patterns.3.The hardware testing platform of the designed windowed decoder for TP-LDPC-CCs is established.The BER performance simulation of this hardware platform for TP-LDPC-CCs over AWGN channel indicates that the result of the hardware platform is almost the same as that of the software platform.Thus,the correctness of the designed windowed decoder was verified.