Robust and Low Complexity Rate Control for Solar Powered Sensors

This paper is concerned with solar driven sensors deployed in an outdoor environment. We present feedback controllers which adapt parameters of the application such that a maximal utility is obtained while respecting the time-varying amount of available energy. We show that already simple applications lead to complex optimization problems, involving unacceptable running times and energy consumptions for resource constrained nodes. In addition, naive designs are highly susceptible to energy prediction errors. We address both issues by proposing a hierarchical control approach which both reduces complexity and increases robustness towards prediction uncertainty. As a key component of this hierarchical approach, we propose a new worst-case energy prediction algorithm which guarantees sustainable operation. All methods are evaluated using long-term measurements of solar energy in an outdoor setting. Furthermore, we measured the implementation overhead on a real sensor node.