Mathematical optimization and game theoretic techniques for multicell beamforming

The main challenge in mobile wireless communications is the incompatibility between limited wireless resources and increasing demand on wireless services. The employment of frequency reuse technique has effectively increased the capacity of the network and improved the efficiency of frequency utilization. However, with the emergence of smart phones and even more data hungry applications such as interactive multimedia, higher data rate is demanded by mobile users. On the other hand, the interference induced by spectrum sharing arrangement has severely degraded the quality of service for users and restricted further reduction of cell size and enhancement of frequency reuse factor. Beamforming technique has great potential to improve the network performance. With the employment of multiple antennas, a base station is capable of directionally transmitting signals to desired users through narrow beams rather than omnidirectional waves. This will result users suffer less interference from the signals transmitted to other co-channel users. In addition, with the combination of beamforming technique and appropriate power control schemes, the resources of the wireless networks can be used more efficiently. In this thesis, mathematical optimization and game theoretic techniques have been exploited for beamforming designs within the context of multicell wireless networks. Both the coordinated beamforming and the coalitional game theoretic based beamforming techniques have been proposed. Initially, coordinated multicell beamforming algorithms for mixed design criteria have been developed, in which some users are allowed to achieve target signal-to-interference- plus-noise ratios (SINRs) while the SINRs of rest of the users in all cells will be balanced to a maximum achievable SINR. An SINR balancing based coordinated multicell beamforming algorithm has then been proposed which is capable of balancing users in different cells to different SINR levels. Finally, a coalitional game based multicell beamforming has been considered, in which the proposed coalition formation algorithm can reach to stable coalition structures. The performances of all the proposed algorithms have been demonstrated using MATLAB based simulations.