Joint Voltage and Modulation Scaling for Energy Harvesting Sensor Networks

Energy harvesting is rapidly becoming a critical architectural component for CPS applications that use Wireless Sensor Networks (WSNs) technology. This paper presents an epoch-based approach for energy management in resourceconstrained WSNs that utilizing energy harvesting techniques. Each epoch represents a time period over which energy production can be reasonably predicted. We consider two energy harvesting hardware models, one that allows concurrent harvesting and execution, and one that does not. For both models we propose and analyze a resource management algorithm that combines energy harvesting with dynamic voltage scaling and dynamic modulation scaling. Our algorithm is optimal in the sense that it maximizes energy reserve levels at individual nodes. We have evaluated the performance of our approach with standard baseline algorithms. The results show that our algorithm outperforms the baseline algorithms under a variety of workloads and energy harvesting profiles.