Abstract

Demand for high data rates is increasing rapidly, due to the rapid rise of mobile data traffic volume. In order to meet the demands, the future generation of wireless communication systems has to support higher data rates and quality of service. The inherent unreliable and unpredictable nature of wireless medium provides a challenge for increasing the data rate. Cooperative communications, is a prominent technique to combat the detrimental fading effect in wireless communications. Adding relay nodes to the network, and creating s virtual multiple-input multiple-output (MIMO) antenna array is proven to be an efficient method to mitigate the multipath fading and expand the network coverage. Therefore, cooperative relaying is considered as a fundamental element in the Long Term Evolution (LTE)-Advanced standard.
In this thesis, we address the problem of resource allocation in cooperative networks. We provide a detailed review on the resource allocation problem. We look at the joint subcarrier-relay assignment and power allocation. The objective of this optimization problem is to allocate the resources fairly, so even the cell-edge users with weakest communication links receive a fair share of resources. We propose a simple and practical algorithm to find the optimal solution. We assess the performance of the proposed algorithm by providing simulations. Furthermore, we investigate the optimality and complexity of the proposed algorithm.
Due to the layered architecture of the wireless networks, to achieve the optimal performance it is necessary that the design of the algorithms be based on the underlying physical and link layers. For a cooperative network with correlated channels, we propose a cross-layer algorithm for relay selection, based on both the physical and link-layer characteristics, in order to maximize the linklayer throughput. The performance of the proposed algorithm is studied in different network models. Furthermore, we investigate the optimum number of relays required for cooperation in order to achieve maximum throughput.
Buffering has proven to improve the performance of the cooperative network. In light of this, we study the performance of buffer-aided relay selection. In order to move one step closer to the practical applications, we consider a system with coded transmissions. We study three different coding schemes: convolutional code, Turbo code, and distributed Turbo code (DTC). For each scheme, the performance of the system is simulated and assessed analytically. We derive a closed form expression of the average throughput. Using the analysis results, we investigate the diversity gain of the system in asymptotic conditions. Further, we investigate the average transmission delay for different schemes.