Comparison of Cooperative and Non-Cooperative Game Schemes for SINR-Constrained Power Allocation in Multiple Antenna CDMA Communication Systems

In this paper, we formulate the SINR-constrained power allocation problem in the wideband multiantenna CDMA cellular system as a game. Using the cooperative theory and Nash bargaining model, we compared it to the noncooperative game scheme proposed in (S. Koskie, Z. Gajic, IEEE trans. on Networking, 2005). We define a utility function which considers both power usage and QoS. We show that the resulting operating point using Nash bargaining model is fair and Pareto optimal, however, the Nash equilibrium operating point obtained in the noncooperative game scheme is not in general Pareto optimal. Also, we proposed a centralized power control algorithm based on conjugate gradient algorithm for multiantenna CDMA cellular systems, when the transmitted power and SINR which each user experience is constrained. In addition, we can extend the result to a fair rate optimization in a multiuser multiple antenna regime through a power control problem subject to SINR constraints.