Research Projects

The increasing demand for sustainable, maintenance-free, and energy-efficient Internet of Things (IoT) systems, especially in smart cities, smart buildings, and remote monitoring applications, has necessitated a shift from conventional battery-powered devices to self-sustaining technologies. Among various ambient energy sources, Radio Frequency (RF) energy harvesting has gained prominence due to the presence of RF signals from Wi-Fi, LTE, 5G, and other communication systems. However, conventional RF harvesters are typically narrowband, static in operation, and lack adaptability to dynamic wireless environments, limiting their efficiency and practical deployment. To address these challenges, the proposed project aims to develop an AI-assisted, reconfigurable RF energy harvesting system using Frequency Selective Surfaces (FSS) tailored for multi-band operation. The system will be capable of intelligently adapting to variations in the RF environment, harvesting energy from available ambient sources, and sustaining low-power IoT nodes. The integration of Artificial Intelligence (AI) will enable dynamic frequency tuning and decision-making, significantly improving energy harvesting efficiency in real-world scenarios. The primary scientific objectives of the project include: Designing and optimizing miniaturized, triband Frequency Selective Surfaces (FSS) to operate at 2.4 GHz (Wi-Fi), 3.5 GHz (LTE), and 5.8 GHz (5G). Integrating tunable components (e.g., varactors, PIN diodes) into the FSS unit cells for real-time reconfiguration. Developing a high-efficiency RF-to-DC rectifying circuit compatible with the FSS structure. Implementing AI algorithms to sense environmental RF spectrum and control surface reconfiguration. Demonstrating the operation of a batteryless IoT node powered entirely by harvested RF energy. The central hypothesis to be tested is: “An AI-integrated, reconfigurable Frequency Selective Surface can dynamically adapt to varying RF environments, thereby significantly improving ambient RF energy harvesting efficiency and enabling self-powered IoT devices.” To validate this, the project will involve the following key experiments: Electromagnetic simulation of FSS structures using full-wave solvers (e.g., CST, HFSS) to evaluate frequency response and tunability. Hardware prototyping of FSS panels embedded with tunable elements. Design and testing of rectifying circuits for multi-band RF-to-DC conversion. Development of AI control algorithms (e.g., reinforcement learning or rule-based models) to optimize the configuration of the FSS in real-time. Integration of the energy harvesting module with a low-power IoT sensor node and assessment under various indoor and semi-urban RF environments. If successful, the project is expected to make a significant contribution to the field of sustainable electronics and smart embedded systems. From a fundamental standpoint, it will advance the understanding of AI-assisted RF system reconfiguration and energy-aware surface engineering. From an application perspective, the outcome can enable batteryless operation of IoT nodes, reducing environmental impact and maintenance costs, while supporting the deployment of smart infrastructure in energy-constrained environments. The proposed system aligns with the goals of green IoT, smart cities, and digital sustainability, and holds potential for practical implementation in sectors such as building automation, wearable healthcare, industrial monitoring, and remote sensing.