• Energy Research

A central concern for large-scale sensor networks and the Internet of Things (IoT) has been battery capacity and how to recharge it. Recent advances have pointed to a technique capable of collecting energy from radio frequency (RF) waves called radio frequency-based energy harvesting (RF-EH) as a solution for low-power networks where cables or even changing the battery is unfeasible. The technical literature addresses energy harvesting techniques as an isolated block by dealing with energy harvesting apart from the other aspects inherent to the transmitter and receiver. Thus, the energy spent on data transmission cannot be used together to charge the battery and decode information. As an extension to them, we propose here a method that enables the information to be recovered from the battery charge by designing a sensor network operating with a semanticfunctional communication framework. Moreover, we propose an event-driven sensor network in which batteries are recharged by applying the technique RF-EH. In order to evaluate system performance, we investigated event signaling, event detection, empty battery, and signaling success rates, as well as the Age of Information (AoI). We discuss how the main parameters are related to the system behavior based on a representative case study, also discussing the battery charge behavior. Numerical results corroborate the effectiveness of the proposed system.

In this paper, a simple and flexible method for increasing the lifetime of fixed or mobile wireless sensor networks is proposed. Based on past residual energy information reported by the sensor nodes, the sink node or another central node dynamically optimizes the communication activity levels of the sensor nodes to save energy without sacrificing the data throughput. The activity levels are defined to represent portions of time or time-frequency slots in a frame, during which the sensor nodes are scheduled to communicate with the sink node to report sensory measurements. Besides node mobility, it is considered that sensors’ batteries may be recharged via a wireless power transmission or equivalent energy harvesting scheme, bringing to the optimization problem an even more dynamic character. We report large increased lifetimes over the non-optimized network and comparable or even larger lifetime improvements with respect to an idealized greedy algorithm that uses both the real-time channel state and the residual energy information.