Research On Key Techniques Of Low-Desity Parity-Check Codes In Wireless Communication Systems

Since the rediscovery of low-density parity-check (LDPC) codes in the middle of 1990's, LDPC codes are one of the hottest topics in the domain of forward error correction codes because of their excellent performance and linear decoding complexity. At the same time, the next generation wireless mobile communication system should reliably provide very high data transmission rate while supporting multiuser to communicate simultaneously. Because of this, powerful error correction codes must be employed to overcome noise and fading. Therefore, using error-correction codes for wireless communication to improve system performance has drawn great attention for researchers. Nowadays, many techniques have been extensively studied as key techniques for the next generation wireless mobile communication system, for example, orthogonal frequency division multiplexing (OFDM), multi-input multi-output (MIMO), adaptive modulation and coding (AMC), multiuser detection. This dissertation concentrates on the joint study of LDPC codes and these key techniques to further improve the performance of wireless communication systems.The first objective of this research is to design a new bit-interleaved coded modulation (BICM) scheme based on rate-compatible punctured LDPC codes. Two puncturing methods are employed in the proposed system: intentional puncturing designed by linear programming, and random puncturing. Then the performance of the proposed system is studied by simulation and the results are analyzed.In MIMO channels, spatial correlation and time-selective fading exist because of real physical environments and Doppler shifts. In order to overcome performance loss in concatenated LDPC and space-time block coding (STBC) system, a new decoder which joints space-time decoding and LDPC decoding is proposed. In the new decoder, we first suppress ISI in the output symbols from pre-processing. Then, we regard the modulation symbols from hard decisions of LDPC decoding as the estimated value of the transmitted symbols to further eliminate the influence of ISI (Inter-Symbol Interference). Moreover, the new decoder exploits diversity provided by LDPC code to compensate performance loss from spatial correlation between transmitter and receiver antennas. Therefore, the decoder is robust to performance loss caused by spatial and time-selective fading.MIMO OFDMA is a promising transmission scheme for wideband wireless mobile communication. In this dissertation, a LDPC-coded MIMO OFDMA system is proposed to cope with frequency-selective multipath fading for multiuser/multirate communication. In order to further improve performance by exploiting the multiuser diversity resulting from adaptive subcarrier allocation among users, we propose a new allocation criterion, which aims for the maximization of sum-rate of all users under the constraints of transmission signal power and bit error ratio. Furthermore, in order to guarantee quality of service (QoS) of each user, we propose a new low-complexity subcarrier allocation algorithm. In the proposed algorithm, an initial subchannel allocation is first performed according to the proposed subcarrier allocation criterion. A reallocation algorithm then guarantees each user's QoS requirement while maintaining fairness among the users and efficiency of system resource utilization. The performance penalty from QoS guarantees is also minimized.The last section of this dissertation presents a two-stage bit-flipping decoding algorithm for regular LDPC codes. Since the ultimate flipping bits are got through two selection processes, the probability of doing wrong flipping is greatly reduced. Moreover, a new bit-selection criterion which incorporates the syndrome and the absolute value of the log-likelihood ratio of the bit vector is proposed during the first stage of decoding. Simulation results show that our proposed decoding algorithm has excellent performance, and therefore it provides another kind of tradeoff between error-correcting performance and decoding complexity.