Coordinated Rate Control in Wireless Sensor Network

We consider the problem of achievable throughput maximization in wireless sensor network. Implementing a multi-agent system enables these nodes to perform sensing tasks in coordinated manner to achieve some desired system-wide objective. Our study results in a bandwidth allocation strategy that assigns a near-optimal transmission policy while adapting to the current traffic, channel load, and buffer occupancy. We present a reinforcement learning approach to rate control in wireless sensor network. A novel feature of our method lays in the incorporation of external coordination signals produced by a variant of belief propagation in graphical models