In-node cognitive power control in Wireless Sensor Networks

Reliability, interoperability and efficiency are fundamental in Wireless Sensor Network deployment. Herein we look at how transmission power control may be used to reduce interference, which is particularly problematic in high-density conditions. We adopt a distributed approach where every node has the ability to learn which transmission power is most appropriate, given the network conditions and quality of service targets. The status of the network is represented by the combination of three parameters: number of retransmissions, clear channel assessment attempts and the quantized average latency. The target is to maintain packet loss at the lowest possible level, whilst striving for minimum transmission power. The learning phase is managed by an ϵ-greedy strategy, which directs the physical layer of each node to choose between either a random action (exploration) or the best action (exploitation). We demonstrate as our learning sensors automatically discover the best trade off between power and quality.