Enhanced PAPR reduction and channel estimation techniques in multi-carrier wireless communication system

摘要

The large and rapid development in the field of communications today represents one of the biggest interests among researchers. In fact, it has become a great revolution in the wireless communication and mobile Internet as an accelerating need to transfer data, such as voice, image, message, and video, in a faster, less expensive, and less complex manner. Hence, Orthogonal frequency division multiplexing (OFDM) and Multi Carrier-Code Division Multiple Access (MC-CDMA) systems are examples of multi-carrier wireless communication. In general, several benefits are gained when multi-carrier systems emerge as an attractive standard for various digital data over radio systems. However, OFDM and MC-CDMA still suffer from peak to average power ratio (PAPR), which is a major drawback in most multi-carrier communications systems. In addition, channel estimation plays a very important role in overcoming the effect of channel fading, which causes BER degradation and jamming pilot symbols. Thus, this thesis introduces two important modifications in OFDM transceiver. First, a novel rotating phase shift (RPS) based on signal scrambling to reduce PAPR in OFDM system had been proposed. The search algorithm was used to solve the convex problem in selecting PAPR-RPS best phase shift factor based on the cost of computational complexity. Second, the Channel Estimation (CE) expressions were derived as the objective function to study the effect of perfect BER in MC-CDMA/OFDM for a frequency selective Rayleigh fading environment. Besides, the CE scheme based on a Local Search Algorithm was suggested. The simulation results of RPS were compared with various PAPR reduction schemes, such as selective mapping (SLM) technique and Partial Transmission Sequence (PTS), and furthermore, different phase shifts with slight computational complexity had been analyzed. As a result from the simulations carried out, the RSP displayed significant approximate by 85% and 72% for PAPR reduction at around 1.5 dB compared to PTS and SLM techniques respectively. Furthermore, the local search channel estimation using number particles for each single swarm that had been used to identify the best fitness solution was computed from the average BER value. Besides, the proposed channel estimator was tested under fast fading channel in a multi-carrier communication system with and without interpolation methods. The simulation results demonstrated that the proposed channel estimation of the MIMO-OFDM system had significantly provide better BER performance compared to other techniques at different modulation types, Signal to Noise Ratio (SNR) values, and channel length. On top of that, the results showed that BER exhibited the best ratio at around 83% and 78% compared to MMSE and LSE algorithms, respectively. Finally, the proposed multi-carrier system, the MC-CDMA, and the OFDM were implemented in FPGA after it had been written by using the VHDL Language implemented in Xilinx ISE 14.1 and also employed the Virtex-4 FPGA board, which had been identical or close to the output of Matlab software environments.