Frequency and power allocation schemes for heterogeneous networks including femto cells

In this paper we propose three novel schemes for spectrum management and compare their performance with that of the frequency reuse assignment technique. These schemes are based on reinforcement learning and power optimization algorithms. Simulation results show that our RL-based schemes yield up to 14% increase in capacity and provide better user satisfaction and capacity. Moreover, the RL-based schemes provide better performance in high cell loads and can accommodate femto cells with the same radio resources.