Performance Analysis of Low Density Parity Check Code for Next Generation Communication System

Abstract

Channel coding provides the means of patterning signals so as to reduce their en-ergy or bandwidth consumption for a given error performance. Low density ParityCheck (LDPC) codes have been shown to have good error correcting performancewhich enables ecient and reliable communication. LDPC codes have linear decod-ing complexity but performance approaching close to shannon capacity with iterativeprobabilistic decoding algorithm. It has drawn the signicant attention of researchersdue to these reasons.In this dissertation, the performance of dierent error correcting code such asconvolution, Reed Solomon (RS), hamming, block code are evaluated based on dif-ferent parameters like code rate, bit error rate (BER), Eb/No, complexity, codinggain and compare with LDPC code. In general, message passing algorithm and thesum-product algorithm are used to decode the message. We showed that logarith-mic sum-product algorithm with long block length code reduces multiplication toaddition by introducing logarithmic likelihood ratio so that it achieves the highestBER performance among all the LDPC decoding algorithms. By applying min sumalgorithm we achieved further reduction in complexity and improvement of decodingperformance. The main contribution of this thesis is to modify min sum algorithm insuch a way that decoding performance can be improved with a shorter block lengthcode and reasonable complexity.The astonishing performance combined with proposed modied min sum decodingalgorithm make these codes very attractive for the next generations digital broad-casting system (ABS - S). This reduces the complexity of decoder with a signicantperformance improvement as compared to the DVB - S2 standard.