Reinforcement Learning Framework for Energy Efficient Wireless Sensor Networks

In recent years, there has been an increasing interest in Wireless Sensor Networks. Sensor nodes in Wireless Sensor Networks are energy constrained. Therefore, in WSNs one of the major design challenges is to minimize consumed energy at the sensor nodes. Hence, a number of routing schemes are designed that make efficient use of limited energy of the sensor nodes. Hierarchical routing protocols are best known in regard to energy efficiency. Our first aim is an implement of routing strategies using Q-routing algorithms and compared them for the energy efficient aspects. Our second step is to develop an energy efficient shortest path Q-routing algorithm using Reinforcement Learning property of ML techniques and compared it with an existing energy aware Q-routing algorithm. We have applied these algorithms on different topologies like heterogeneous, homogeneous and introduced a novel hybrid topology by combining heterogeneous and homogeneous topologies.