Add QLearning with Duty Cycle in Ns2

In WBAN, large number of sensors are deployed inside and outside of human body. People with the chronic disease require long-term hospitalization. To avoid from long-term hospitalization, WBAN is the ideal way. Main aim of IEEE 802.15.4 design is to provide low cost, low data rate, and low power protocol for communication in a sensor network. In IEEE 802.15.4, slotted ALOHA and CSMA/CA techniques are used to access the medium or channel. Slotted Aloha and CSMA/CA based techniques have benefits of low computations and low overheads, but suffers collisions due to blind transmission strategy in the beginning of time slots recently, Typically,WBAN nodes have limited energy capacity, constraints on the miniature size, low signal processing capabilities and low storage capacity. However, energy efficiency remains one of the most important issue in WBAN. Machine learning has been applied effectively in wireless sensor networks to improve their performance. In this proposal, I propose Adaptive wake up interval, which based on Q-Learning to reduce the unnecessary energy waste dynamic traffic in WBAN.