Security enhanced optimal trajectory selection for mobile sink using reinforcement learning
What is it about?
The growing demand for energy efficient and maximized life time networks in highly critical applications like surveillance, military and medicine has opened up more prospects as well as challenges in the deployment of WVSNs. Multi-hop communication in WVSN results in overloading of intermediate sensor nodes due to its dual function in the network which results in hotspot effect. This can be mitigated with the help of mobile sinks and rendezvous points based route design. But mobile sinks has to visit every cluster head to gather data which results in longer traversal path and higher latency and power consumption related issues if not addressed properly will impact the performance of the network. Our objective is to analyze and determine the optimal trajectory for mobile sink node traversal with the help of a high quality transmission architecture integrated with reinforcement learning and isolation forest based anomaly detection to propose an energy efficient meta-heuristic approach to enhance the performance of network by reducing the latency and securing the network against possible attacks.
Why is it important?
The proposed model thus focuses on designing an efficient optimization technique to devise a trajectory for mobile sink traversal with the help of reinforcement learning framework.