I am Dr. Amith, specializing in instrumentation and sensor systems, with a growing focus on integrating artificial intelligence to enhance decision-making. My work bridges the gap between traditional sensing technologies and modern data-driven methodologies, aiming to improve accuracy, efficiency, resilience, and system-level intelligence.

My core expertise lies in the design, analysis, and application of sensor-based systems. Over time, my research has evolved toward leveraging AI techniques to overcome inherent limitations of physical sensing, such as incomplete coverage, noise, and measurement uncertainty. Recently, I have focused on how AI can augment sensor systems by enabling predictive insights and more informed decision-making. For example, consider an apartment where temperature monitoring is required across multiple rooms. Even with a distributed network of sensors, some areas remain unmeasured, and collected data may be affected by noise or inconsistencies. Instead of relying solely on these physical measurements, I explore AI-based predictive models that learn from available data, such as readings from nearby sensors, temporal patterns, and environmental conditions. These models can estimate temperatures in unmonitored areas and refine overall system accuracy. This approach represents a shift from direct measurement to intelligent estimation, where sensors and AI work together to provide a more comprehensive and precise understanding of complex environments.

Beyond improving prediction accuracy, my research emphasizes privacy-preserving intelligent sensing, where the benefits of AI are achieved without compromising the confidentiality of sensitive sensor data. Modern sensor networks continuously collect large volumes of information, including environmental conditions, occupancy patterns, energy usage, and user behaviors. If such data are exposed or improperly shared, they can reveal sensitive information about individuals, organizations, or critical infrastructure, leading to privacy breaches, security vulnerabilities, and loss of trust. To address these challenges, I investigate AI methodologies that enable robust learning and inference while minimizing the exposure of raw sensor data. This includes designing frameworks that support secure data processing, distributed intelligence, and privacy-aware decision-making, allowing sensor systems to remain both intelligent and trustworthy.

Another key direction of my research is the development of sustainable and resource-efficient intelligent sensing architectures. As sensor deployments continue to expand across smart buildings, industrial facilities, cities, and critical infrastructure, it becomes increasingly important to reduce the dependence on dense sensor installations while maintaining high-quality monitoring and control. By combining physical sensing with AI-driven estimation and prediction, it is possible to infer measurements in locations where sensors are unavailable or impractical, thereby reducing hardware requirements, installation costs, maintenance efforts, communication overhead, and overall energy consumption. This approach not only improves the scalability and long-term sustainability of sensor networks but also extends their operational lifetime, enabling efficient monitoring of variables such as temperature and other environmental parameters without requiring exhaustive sensing coverage. The overarching goal is to design intelligent sensing systems that are accurate, energy-efficient, economically viable, and environmentally sustainable.

My research interests include sensor fusion, intelligent systems, data-driven modeling, AI-enhanced decision frameworks, privacy-preserving machine learning, and sustainable intelligent sensing architectures. I am particularly interested in developing scalable solutions that integrate sensing technologies with machine learning to address real-world challenges across various domains. The figure below illustrates that initial sensor deployments exhibit coverage gaps and noisy measurements, leading to inaccuracies and limited precision. By introducing an AI-based predictive model, these limitations are addressed through learned spatial and temporal patterns, enabling complete and high-resolution estimation across the entire environment while supporting secure, privacy-aware, and resource-efficient sensing.

Education

• PhD in Electrical and Computer Engineering, University of British Columbia, Canada.

• PhD in Biomedical Engineering Sensors and Instrumentation, Universiti Kebangsaan Malaysia, Malaysia, Best Thesis Nominee.

• M.Sc. in Computing (Network Concentration), Qatar University, Doha, Qatar, Passed with 4.0 CGPA

• B.Sc. in Electronics and Telecommunication Engineering, North South University, Bangladesh, Valedictorian.

Certification

• PMI-PMP - Project Management Professional 

• PMI-RMP - Risk Management Professional

• PMI-ACP - Agile Certified Practitioner

Committees

• Publication Chair at ISOPE, Click here

Get in touch at,

Email: amitk@qu.edu.qa

Google Scholar Link: Click Here