Realization of Autonomous Sensor Networks for Optimal Transmission and Efficient Energy Utilization using Artificial Intelligence
Wireless sensor networks (WSN) are most sought after for their ability to perform autonomously with least maintenance in unapproachable regions. An autonomous sensor network is one that is capable of decision making and implementing controls on its own without manual intervention. Hierarchical topology schemes are more common WSN topology that suits to compensate for the energy constrained nature of WSN. Also, to cope up with the resource constraints of WSN, intelligent data transmission is the current requirement. Hence, intelligent data gathering at node level is done by exploiting temporal correlation among sensed data. To increase the energy efficiency of the WSN, the node autonomously switches between and active modes and the duration in each mode is also computed dynamically using Q-Learning exploration - exploitation technique. This paper highlights the implementation and results of such autonomous WSN inreal-time.