Abstract
One of primary objectives of future wireless systems is to provide uniform high rate coverage. Essentially, a wireless system must take multipath fading, path loss and shadowing effect into consideration. In the pursuit of schemes that will provide a solution to these problems, various granular and distributed network architectures based on relaying techniques are emerging. In particular, three basic relaying schemes have been proposed including amplify-and-forward (AF), decode-and-forward (DF) and compress-and-forward (CF). This thesis has investigated the performance of various single-relaying schemes from the theoretical point of view and established a fundamental understanding of relay-assisted communication systems. An improved version of DF - Soft DF - has been evaluated through the derived upper bounds of Bit-Error-Rate (BER) and Block-Error-Rate (BLER). For the most complicated relaying scheme - CF, we proposed some novel techniques to efficiently improve its performance. Moreover, both theoretic and simulation results have shown that one single-relaying scheme cannot always outperform the others over all channel conditions. In this regard, the relaying schemes need to be deployed in a hybrid fashion to adapt to the current channel realization. We have provided two hybrid relaying schemes in this thesis and investigated the integration of the hybrid relaying schemes with hybrid retransmission protocols. The proposed integrated technique is named as H2-ARQ-Relaying schemes. Moreover, we have proposed two H2-ARQ-Relaying strategies that efficiently exploit a hybrid CF/DF forwarding scheme, where the relay, based on its decoding status, could dynamically switch between CF and DF and analyzed the performance analysis has been performed for both strategies mainly in terms of the obtained outage probabilities and throughputs. Considering the availability of Channel State Information (CSI) at the relay in practical implementations, we have proposed modified CF/DF-based H2-ARQ-Relaying strategies with limited CSI feedback accordingly.