Optimal microcontroller usage in reconfigurable intelligent surface: Batteryless IoT systems case study

To effectively enhance wireless communications in internet of things (IoT) systems using reconfigurable intelligent surfaces (RISs), it is crucial to have proper control over programmable passive elements for reflection angle manipulation and active elements that amplify and reflect signals in the desired direction. However, incorporating a high number of RIS elements requires an adequate number of microcontrollers, which results in increased system complexity and cost. This paper proposes a solution to improve the efficiency of RISs in enhancing wireless communications in IoT systems by investigating the trade-off between the number of RIS elements and microcontrollers. Our solution is a novel approach and differentiating factor from other works in the literature, named ”Module,” in which one microcontroller controls a single module. We define the modules as either active or passive, each containing an optimal number of active or passive elements. We determine the optimal number of elements to establish the size of a module using a non-linear energy harvesting model where a batteryless IoT (b-IoT) sensor harvests energy from a nearby base station (BS) radio frequency (RF) signals. Using the optimal size of a module, we find the optimal number of modules, i.e., the optimal number of microcontrollers for efficient RIS panel controlling. With the harvested energy from the BS RF signals, the b-IoT sensor transmits data from the BS to other IoT sensors, which enforces an information causality constraint in RIS module-assisted IoT systems. We formulate an optimization problem to minimize the energy consumption of the RIS modules while satisfying the non-linear energy harvesting model and information causality constraint. The formulated non-convex mixed-integer non-linear energy minimization problem is solved using iterative algorithms. Simulation results show that the RIS module-assisted energy harvesting enhances the performance of the IoT systems by around 100 % compared to benchmark models with no RIS panel.