In this paper we present a distributed, fast and fair resource allocation algorithm for a multiuser, wireless LTE OFDMA channel. Extending our previous work, we propose additional efficiency enhancements and discuss the protocol issues involved. We further explore the trade-offs between performance and costs and present an analysis of the overhead and time requirements. The proposed algorithm partitions the users of the wireless network into coalitions and, using the game theoretic concept of the Nash Bargaining Solution (NBS), offers a cooperative solution to the subcarrier allocation problem. The use of the NBS ensures that the fairness provided matches that offered by the widely accepted Proportional Fair (PF) scheduler. Simulation results show that the sum rate achieved can be tuned to be almost equivalent to the sum rate of the PF scheduler, while only requiring limited resources and exchange of information between nodes. At the same time, the extensive efficiency enhancements and its distributed nature render the algorithm fast and low-complexity enough to be suitable for implementation in a real-time wireless system.
展开▼
