Game theoretical resource allocations in multiple antenna interference networks

Abstract

During the past several decades, wireless communication has been one of the most exciting and challenging research areas in the communication field. The performance of most of nowday wireless networks has been ultimately limited by interference. For example, cell-edge mobiles in multi-cell cellular systems suffer from severe inter-cell co-channel interference (CCI) and have significant performance recessions. Similar issues also exist in ad hoc networks, wireless local area networks (WLAN), cognitive radio networks as well as many other networks. How to effectively allocate the radio resource to combat with interference in wireless interference networks, especially in multiple antenna wireless networks, has been an outstanding challenge in the design of wireless networks. In this thesis, we take a game theoretical approach to study the resource allocation problem in multiple antenna interference networks. Specially, we focus on two typical multiple antenna interference network scenarios, i.e., downlink power control in MIMO space-time coded cellular systems and transmit covariance design in K-pair MIMO interference channel. The first is the most representative example of the one-to-many MIMO interference channel and the results can be easily extended to any space-time coded one-to-many MIMO interference channel examples. To exploit the heterogeneous path loss and shadowing effect between cell-edge and cell-center mobiles and to mitigate the co-channel interference (CCI) at the cell-edge mobiles, we propose a novel space-time coded overlaying scheme to serve multiple cell-edge and cell-center mobiles simultaneously. The power allocation between cell-edge and cell-center mobiles is formulated as a non-cooperative game. We study the existence and uniqueness of the Nash Equilibrium (NE) in this game. We further propose a low-complexity distributive power allocation algorithm which only relies on the local channel statistics and has provable convergence to the NE in formulated game. The second is studied in the K-pair MIMO interference channel which can be fitted in with a lot of applications like ad hoc networks. To effectively control the interference generated at each transmitter, we consider the MIMO interference game with additional rank constraints imposed on each transmitter's covariance. We study the existence and uniqueness of the NE, the convergence of the best response dynamics as well as how to determine the rank constraints of the MIMO interference game with rank constraints in a holistic game theoretical framework.

Date of Award	2011
Original language	English
Awarding Institution	The Hong Kong University of Science and Technology