Distributed Resource Allocation for Energy Efficiency in MIMO OFDMA Wireless Networks

This paper deals with the problem of distributed resource allocation in multiple-input multiple-output multi-carrier multiple-Access channel networks. The assignment between users and subcarriers is allocated together with the users' transmit powers for energy efficiency maximization, by means of a novel approach which merges the popular Dinkelbach's algorithm with the frameworks of distributed auction theory and stable matching. Two distributed algorithms are presented, which can be implemented in a fully decentralized way. The former is guaranteed to converge to the global optimum of the system energy efficiency, up to a threshold which can be set in advance, while the latter enjoys weaker optimality properties, but has an even lower computational complexity. Additionally, we develop a novel energy consumption model which explicitly accounts for the energy consumption due to feedback transmissions. Employing this new model, it is shown that the proposed distributed algorithms can even outperform centralized resource allocations which require a larger feedback energy consumption.