Low Complexity Error Correction in Low Density Parity Check (LDPC) Code Decoder and Encoder for Decode and Forward Cooperative Wireless Communication

• Main Author: JAM'AAH, SUUD
• Format: Thesis
• Language: English
• Published: Universiti Malaysia Sarawak (UNIMAS) 2021
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://ir.unimas.my/id/eprint/35926/1/Jam%E2%80%99aah.pdf; http://ir.unimas.my/id/eprint/35926/
• Institution: Universiti Malaysia Sarawak

Decoding high complexity is a major issue to design a decode and forward (DF) relay protocol. Thus, the establishment of a low complexity decoding system would be beneficial to assist DF relay protocol. To overcome this problem a DF protocol relay system model using LDPC code is proposed in this thesis. The results show that employing LDPC code for DF protocol relay system can achieve better error rate performance than that of using non-cooperative and other existing relay protocol systems which are AF, DF, and DF protocol relay system using Turbo code. Besides that by using LDPC code the decoding processing time can be reduced. The limitation of the available works on min-sum based LDPC code decoding algorithm specifically for DF protocol relay system motivated this research. The initial investigation on the existing LDPC code decoding algorithms assists to develop a low complexity LDPC code decoding algorithm. By using the optimization min-sum belief propagation approach, a low complexity min-sum (MS) based decoding algorithm called Variable Global Optimization Min-Sum (VGOMS) has been developed. A key aspect of this algorithm is balancing the trade-off between the problem of complexity reduction and error correction performance of the relay node LDPC decoder component. This algorithm only applies the optimization scaling factor at the bit node processing of the variable node operation. The Particle Swarm Optimization (PSO) search method is adopted to search the optimized scaling factor to obtain optimal error rate performance. The source to relay channel is modelled as a cooperative fading that is used extensively in cooperative communication. From the result simulation, the VGOMS algorithm outperforms than well-known existing LDPC code decoding algorithm min-sum, is comparable with Normalized min-sum (NMS), and Offset min-sum (OMS), and closed near to Sum-Product (SP) algorithm at higher SNR in terms of error rate performance. VGOMS also outperforms NMS, OMS, and SP in terms of check node operation complexity that consumes the most complex decoding operation. VGOMS has shown a better compromise between better error rate performance and low computation operational complexity. One major concerning the LDPC code for the DF protocol relay system is high encoding complexity. Thus, the LDPC code encoder model is developed to identify a low complexity encoding algorithm of LDPC code for the DF protocol relay system. In this model, eight different LU and QR encoding method variants are evaluated in terms of execution time, the number of nonzero, and the pattern of nonzero. The proposed LDPC code encoder model shows that the LUPQ encoding method achieved the lowest execution time and number of nonzero among LU and QR encoding methods. The performance results showed that LUPQ is the most suitable encoding algorithm for the DF protocol relay system using LDPC code as it performs low processing time of the LDPC encoder component.