Reinforcement Learning Based Clock Synchronization in WBAN

Abstract

Wireless Body Area Network (WBAN) is an application of wireless sensor network (WSN). WBAN therefore forms a comprehensive collection of devices that are not only capable of providing continuous information about the health status of a person but also offers helpful details about the activities and environment of the person. In this paper, we have evaluated TDMA based MAC protocol performance through several metrics and TDMA approach is used to avoid packet collision which leads to higher packet loss rate. Reinforcement Based Clock synchronization is the solution of problem like packet collision. After clocks of WBAN sensor nodes are synchronized, data can be transferred between sensor nodes and sink efficiently and rapidly. Reinforcement learning iteratively optimizes the clock synchronization technique. Experimental results indicate that the proposed algorithm is more efficient than existing techniques.