Relay selection in AF and DF based cooperative systems

In wireless channels, the most disruptive effect is the fading which is caused by multipath propagation. One of the methods to avoid these fading effects is using multi input multi output (MIMO) systems. This method is also known as space or antenna diversity. Although using multiple antennas is feasible for base stations, it is not suitable for mobile units because of hardware complexity and cost. Thus cooperative communication techniques have become a popular alternative solution to MIMO systems. In cooperative communications, network nodes (users or relays) help the transmission of the message to the destination by using amplify and forward (AF) or decode and forward (DF) methods. This achieves spatial diversity gains in a distributed manner even if the users have single antennas. Recent research have shown that the relay/cooperative systems provide better error performance compared to non-cooperative systems at the same signal-to-noise ratio (SNR). The performance of cooperative systems can be further improved by increasing the number of relays. In the literature, the channels between the terminals are usually modelled as Rayleigh and Nakagami-m distribution.However when terminals are in motion, all these classical distributions are inadequate. For this case, experimental studies show that a new model which is named as multiplicative channel model or cascaded channel model is used instead of these classical distributions. In this thesis, the effects of relay selection on the error performance of the cooperative systems using AF and DF relaying is investigated over different channel models (Rayleigh, Nakagami-m, Cascaded channels). Error performance analysis for cooperative/relay systems is studied for several system and channel models. Specifically, by deriving end-to-end signal-to-noise ratio (SNR) and its statistics (e.g. probability density function (PDF), cumulative distribution function (CDF) and moment generating function (MGF)), outage probability and bit/symbol error rate (BER/SER) expressions are obtained. Also, application of error propagation reduction techniques and development of new methods are investigated. Error propagation occurs in DF based wireless cooperative communication systems due to the detection errors at the relay terminals affecting the performance at the destination in a negative way. In this thesis cooperative maximal ratio combining (C-MRC) and virtual noise (VN) methods are focused on.