Research On Detection Based On The K^th Best-relay Selection Over Frequency Selection Fading Channels

Relay selection in cooperative communication systems, the problem maybe occurs that both the best relay and the second best relay are unavailable. Therefore, it is of vital performance to study the K^th best relay selection algorithm. When the channel is frequency selective fading, general relay selection algorithm may not be able to be applied directly. To solve these problems, the diversity order of K^th best relay selection over frequency selection fading channels is derived in this thesis and the equalization and detection algorithm are also investigated in cooperative communication. The main contents are as follows:1. The diversity order of K^th best relay selection based on the criterion of Max-Min over frequency selection fading channels is studied. The PDF and CDF of the received SNR and the end-to-end instantaneous SNR are combined to obtain the theory upper boundary of the BER in the K^th best relay. Then the diversity order is derived. Theoretical analysis and simulation results show that the diversity order is proportional to the number of the available relay and the paths of channel, and inversely proportional to the order number of the selected relay.2. The channel equalization technology in asynchronous cooperative communication is studied. Under the condition of two paths channel, aiming at the selection of two relay in single carrier transmission system and the performance of time domain equalization used in receiver is studied and simulated. Simulation results show that the interference between relays in asynchronous cooperative communication is eliminated greatly through the equalization and the slope of BER performance curves is equal to the diversity order.3. PSO algorithm is introduced to asynchronous cooperative systems. To solve the issue that the traditional PSO algorithm is easy to converge to the local optimal solution, the modified PSO algorithm is proposed in which the Hamming distance and bit flipping are combined to update the velocity. Simulation results show that the performance of ZF equalizer is improved obviously by using this proposed algorithm.4. SD algorithm is introduced to asynchronous cooperative systems. To reduce the complexity of the traditional SD algorithm, a modified iterative SD algorithm is proposed. The set to be searched is acquired through combining the initial solution and bit flipping, and two norm criteria is used to search the target solution from the set. In order to reduce the complexity in the iterative search process, a restrict-criterion is proposed. Simulation results show that SD algorithm can converge after 4 iterations and when the value of BER is 10-2, the BER performance is improved of about 9.9 dB compared with ZF algorithm.