UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Network coding based cooperative communications Kandhway, Kundan

Abstract

Cooperative communications was proposed to enable spatial diversity in small and inexpensive devices. It allows the creation of virtual antenna array through the antennas of the participating users. The benefits offered by cooperation include increase in data rate, robustness against shadowing, decrease in overall transmit power of the system etc. However, when user cooperation is extended to include multiple users or multiple relays, the system suffers from loss in throughput due to increased number of channel use. To overcome this, cooperative communications schemes often make use of network coding which helps trade-off resource allocated for cooperation and system performance. In the first part of the thesis, we propose random network coding based user cooperation scheme in wireless networks. Our scheme is very effective in spreading the information of the pool of cooperating users so that the message can reach the destination via many alternative paths. Also, the proposed scheme is decentralized and the cooperating nodes act independent of the others. Results show that our scheme is resilient to inter-user channel noise and can achieve high diversity gain when number of cooperating users is large. We further enhance the performance of our scheme for bad user-destination channel by protecting the packets by convolutional coding. This version of our scheme performs better than traditional N user cooperation in terms of both outage and throughput for all user-destination channel conditions when inter-user channel is good. It also shows better robustness to inter-user channel than original scheme. In second part of this thesis, we consider analog network coding based bidirectional relaying system. We develop a scheme to optimally allocate power at the relay nodes such that overall data rate in transfer of messages between two user nodes is maximized under uncertain channel conditions. We have proposed an iterative solution for rate maximization problem and solve a geometric program at each step. Results show that bidirectional relaying can achieve significantly more data rate than conventional unidirectional relaying scheme at the cost of reduced diversity. Also, addition of more relays makes the system more robust to imperfections in channel.

Item Media

Item Citations and Data

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International