This Is AuburnElectronic Theses and Dissertations

Relay Positioning for Energy Efficiency and Improved Performance of Cooperative Wireless Networks




Vellimalaip Sivakumar, Vignesh

Type of Degree



Electrical Engineering


Cooperative communications in wireless networks have gained much interest due to their ability to mitigate the effect of fading in wireless networks by achieving spatial diversity. One of the techniques used in cooperative communication systems, is by relaying data through relay nodes. The number of base stations (BS) in a network is small, so many users who are far from them require more power to transmit data. One issue is to find where to position a limited number of available relay nodes (RN) to reduce energy consumption of user nodes. Two ideas are proposed in this thesis as solutions to the above problem. First, we use a greedy algorithm to determine the position of the relay nodes. In this method, relays are placed at half way distance between the farthest user and the base station. Second, we use an approximation algorithm for positioning the relay nodes. In this method, relays are positioned at near optimal locations in an iterative manner. The algorithm tries to position the relay nodes at the periphery of the base station’s coverage area. We compare both the algorithms in terms of average energy consumption, average number of hops for a packet to reach the base station and average r-cover which is the coverage radius of RN. The approximation algorithm is shown to outperform the greedy algorithm in all the above mentioned metrics. We then investigate the application of two capacity enhancement techniques - Receiver Diversity (RD) and Transmitter zero forcing (TZF) - in cooperative wireless networks. The objective is to provide an analysis for the comparison of TZF and different RD strategies such as Maximal Ratio Combining (MRC), Equal Gain Combining (EGC) and Selective Combining (SC). We find that there is no case of dominance for the two techniques. TZF can be used when interference is weak while RD can be adopted when the interference is high. There always exists a trade-off between RD and TZF in network performance.