Research Projects

The exponential surge in demand for ultra-fast, high-capacity, and low-latency wireless communication systems, particularly in the context of emerging 5G/6G and Internet of Things (IoT) ecosystems, necessitates transformative advancements in antenna technology. Substrate Integrated Waveguide (SIW) antennas have emerged as a promising platform for millimeter-wave (mmWave) applications due to their planar configuration, low profile, and high-quality factor. However, prevailing SIW architectures are plagued by inherent challenges such as restricted frequency agility, large footprint, mutual coupling in MIMO systems, and complex fabrication processes, which hinder their seamless deployment in compact and reconfigurable wireless devices. These limitations highlight the necessity of developing a novel antenna topology that tackles miniaturization, self-quadruplexing capability, isolation enhancement, and cost-effective manufacturing.The primary purpose of this project is to design a self-quadruplexing SIW antenna which is 3D-printed, compact, and highly isolated. This antenna will be precisely built for applications including mmWave 5G/6G connections and the Internet of Things. The results of this research work will lead the development of an integrative framework that combines additive manufacturing (also known as 3D printing) with design optimization driven by machine learning. This framework will make it possible to exercise precise control over antenna designs and electromagnetic performance indicators. In order to provide excellent isolation, compactness, and scalability for multi-band MIMO applications, the antenna that is being considered will have the inherent capability to operate over four discrete frequency channels that are self-quadruplexing.The hypothesis underpinning this research posits that a data-driven machine learning model, trained on an extensive parametric sweep of SIW geometries and material properties, can predict and optimize antenna performance parameters (S-parameters, isolation, gain, bandwidth) with high accuracy, thereby significantly expediting the design cycle. By embedding topology-specific machine learning algorithms into the design pipeline, it is anticipated that novel SIW cavity structures with enhanced field confinement and decoupling characteristics can be realized, surpassing the limitations of conventional empirical methods. The project methodology is defined by the following key experimental tasks: 1. Electromagnetic Modeling and Simulation , 2. Machine Learning Optimization Pipeline 3. Additive Manufacturing of SIW Prototypes 4. Experimental Validation and Measurement 5. System-Level Integration and Performance Evaluation. The anticipated research outcomes will offer significant insights into self-multiplexing strategies and advanced SIW miniaturization methods, thereby enriching the community’s understanding and propelling the translational potential of SIW antennas in 5G/6G and IoT infrastructures.