Research On Technology And Performance Of Wireless Cooperative Relay

The future wireless mobile communication systems are expected to provide high-rate, high-quality and efficient data transmission. Diversity is an efficient way to improve communication reliability, which can be generally classified as spatial diversity, frequency diversity and time diversity. Multiple antennas technique can not be used in mobile terminals to exploit spatial diversity because of terminals'size constraint. But a new type of spatial diversity named cooperative diversity has recently received considerable attention due to their possible benefits of spatial diversity without extra terminal size. The broadcast nature of the wireless medium is the key property that allows for cooperative diversity among the transmitting terminals:transmitted signals can, in principle, be received and processed by any of a number of cooperative terminals which serve as virtual antennas to mitigate the effect of the fading and increase the system throughput. Cooperative relay technique becomes research hotspot now and is known as the most promising physical technologies for future broadband wireless mobile communication systems. Based on the different processing about the received signals at relays, cooperative relay can be categorized as amplify-and-forward (AF) relay, decode-and-forward (DF) relay and coded cooperation (CC) relay. In this thesis, the performance of dual-hop AF relay and DF relay systems are analyzed. Specially, the multi-user diversity and channel estimation in AF relay system is studied. The main research work and creative contents are as follows. Firstly, a novel performance analysis method to study the performance measures of amplify-and-forward (AF) relaying over non-identical Nakagami-m fading channels is proposed. The exact probability density functions (PDFs) of end-to-end signal-to-noise ratio (SNR) for variable gain relaying and fixed gain relaying are deduced, and the theoretical formulas of moment generating function (MGF) and moments for fixed gain relaying are presented. Since the exact PDF of end-to-end SNR for variable gain relaying is not tractable, its approximation is used to derive the formulas of MGF and moments. Using these statistic characteristics, the average symbol error probability (ASEP) and amount of fading (AoF) for fixed gain relaying are derived, and the approximate ASEP for variable gain relaying are given. The Monte Carlo simulations are used to verify the accuracy of deduced mathematics results and these results provide important theoretical reference to evaluate the actual performance of AF relay system.The second part of this thesis analyzes the performance measures of fixed and selective decode-and-forward (DF) relaying over non-identical Nakagami-m fading channel. The exact closed-form theoretical expressions for cumulative distribution function (CDF) of instantaneous end-to-end signal to noise ratio (SNR), outage probability (OP), average symbol error probability, moment of end-to-end SNR and amount of fading are presented. Additionally, we provide closed form outage probability in multiuser scenario with maximum SNR schedule. These derived theoretical results are verified by computer simulations and provide general perspective on DF relaying network, which are composed of sums of specially functions evaluated easily. These results are valuable in theory to evaluate the performance of DF relay system. In mobile cellur communication systems, it is practical to be equipped with multiple antennas for basestation. The outage performance of AF relay system in one source (basestation), multiple relays, multiple users scenario with maximum SNR schedule is provided, and the closed-form outage probabilities for variable gain relaying and fixed gain relaying network are derived. Based on these results, diversity order is deduced for variable gain relaying network. Simulation results validate the derived theoretical results.Finally, for the case of AF relay channel, the performance of three channel estimations including least square (LS) estimation, linear minimum mean square error (LMMSE) estimation and low rank minimum mean square error (Lr-MMSE) estimation with OFDM modulation technique is studied. The simulation results show that LMMSE estimation outperformes the other two estimations, but it is too complicated. The performance of LrMMSE estimation approaches that of the LMMSE estimation and lessens the number of the multiplications.