AbstractsEngineering

The use of multiple input multiple output systems as well as the use of relays are without any doubt key elements of future wireless networks. Both techniques ensure high data rates and wide-range coverage. Due to the variety of possibilities for data and signal processing, especially in case of multiple users, the transmit strategies that achieve optimal rates are mostly unknown. In this thesis, we aim for finding transmit strategies that achieve at least near-optimal rates in multi-user channels with an amplify-and-forward relay. The channel models that we consider are the multiple-access relay channel as well as the broadcast relay channel. Moreover, we compare a time division multiple-access (TDMA)-based transmit scheme to a strategy where all stations use the channel at the same time (referred to as joint relaying). In a first step, we neglect the direct links between the transmitter(s) and the receiver(s) and derive the optimal sum-rates of TDMA and joint relaying for the multiple-access relay channel. Moreover, we show that TDMA achieves higher sum-rates than joint relaying. For the broadcast relay channel we derive a heuristic scheme. This scheme has a lower complexity than prevailing schemes and achieves approximately the same rates. Subsequently, we generalize the model by considering also the direct links between the transmitters and the receiver. For this generalization, we find upper and lower bounds on the sum-rate both for TDMA and joint relaying. It turns out that the superiority of TDMA persists for weak direct links but gets lost quickly if the direct links become stronger. Finally, we consider the use of finite alphabets in very noisy broadcast channels, where it can be shown that the optimal performance of Gaussian alphabets is also achieved by quadrature phase shift keying. As a consequence of this, the suboptimality of TDMA at low signal-to-noise ratio, which was already shown for Gaussian alphabets, persists if finite alphabets are used.