LLR Based Analysis of Quasi-cyclic and Random Ldpc Codes Combined With Multi-user Mimo Systems

Abstract

Multiple input multiple output (MIMO) system is one of the most promising wireless technologies that can achieve wide coverage, high throughput and improved reliability. It employs multiple antennas at both transmitter and receiver sides in order to get high spatial diversity gain which in turn helps to mitigate fading and provides improved system reliability. MIMO systems also support high data transmission rate through spatial multiplexing thus results in higher system capacity. To further improve reliability and performance of MIMO systems, coding gain benefit can be added by combining MIMO systems with error correcting codes such as LDPC codes. Low density parity check (LDPC) codes can achieve capacity near Shannon‟s limit with good error correcting performance. In this thesis, we combined LDPC codes with orthogonal space time block codes (OSTBC) in MIMO systems. The performance of LDPC codes depends on the structure of the parity check matrix and decoding algorithms. In this thesis, bit-error rate (BER) performance of different LDPC codes like Quasi Cyclic (QC) and random which differs in their parity check matrix structures is analyzed. For low complexity decoding and better BER performance, log domain sum product algorithm (log-SPA) is used which has messages in the form of log-likelihood ratios (LLR). Performance of different LDPC codes with multi-user (MU-MIMO) system is also analyzed. Further, the approximate BER expressions are derived for above mentioned system with different antenna configurations.