Distributed Q-Learning for Energy Harvesting Heterogeneous Networks

Distributed Q-Learning for Energy Harvesting Heterogeneous Networks is a solution proposed by SANSA, which is based on Q-learning algorithm and it is designed to increase the system throughput, offload the macro BSs and decrease the drop rate at the macro BS.

The simulation results are promising and show that the proposed approach is viable, as the algorithm meets most of our design goals and also improves the energy efficiency of the system. Future work in SANSA will consider the integration with the backhaul model and the enhancement of the decisions made by the distributed small cells.

More information on this solution introduced by SANSA will be available soon.