Feedback overhead-aware clustering for interference alignment in wireless network

다국어 초록 (Multilingual Abstract)

In wireless communications, interference alignment (IA) is an promising technique that can effectively control interference. Recently, clustered IA in practical network environment has been discussed. However, the intrinsic problem of how to form a cluster is still open. We focus on linking the clustering issue to the feedback overhead that cannot be overlooked in IA implementation. Before explaining the main idea of this thesis, we analyze the feedback rate from an outage perspective. This makes it possible to resolve the inconsistency between forward and reverse link resulting from assuming the feedback channel to be AWGN. Next, we formulate the optimal resource allocation problem for data transmission and feedback over the coherence block. The solution obtained from the optimization is directly applied to the algorithm as a parameter needed for cluster formation. Finally, feedback overhead-aware clustering algorithm that maximizes net spectrum efficiency is proposed. Through Monte-Carlo simulation, the proposed algorithm is shown to provide a better performance gain than conventional approaches.