Internet of Things: IoT

The Internet of Things (IoT) represents a transformative technological paradigm where physical objects are embedded with sensors, software, and connectivity to collect, exchange, and act on data. By enabling seamless communication between devices, systems, and users, IoT is driving innovation across industries, enhancing efficiency, and improving quality of life.

At the Academy of Technology (AOT), research in IoT focuses on developing robust, scalable, and secure solutions that connect the digital and physical worlds. Projects involve designing sensor networks, edge-computing architectures, and intelligent gateways that process data closer to the source, reducing latency and improving system responsiveness. Faculty and students explore applications in smart cities, industrial automation, healthcare monitoring, precision agriculture, and environmental sensing. Security and data privacy are key research priorities. AOT’s work addresses vulnerabilities in IoT ecosystems through lightweight cryptography, secure communication protocols, and intrusion detection mechanisms tailored for resource-constrained devices. Energy-efficient design is another area of emphasis, with research aimed at extending the operational life of IoT devices through low-power hardware and adaptive communication strategies.
Applications of AOT’s IoT research are diverse and impactful. In healthcare, wearable sensors enable remote patient monitoring and early detection of health anomalies. In agriculture, IoT systems optimize irrigation, pest control, and crop management. In industrial settings, predictive maintenance systems use IoT data to reduce downtime and improve productivity.
Supported by dedicated IoT and embedded systems laboratories, AOT fosters interdisciplinary collaboration between electronics, computer science, and data analytics. Students gain hands-on experience with hardware prototyping, cloud integration, and data-driven application development. Future research directions at AOT aim to integrate IoT with artificial intelligence, blockchain, and 5G/6G communication technologies, creating intelligent, autonomous networks capable of real-time decision-making. Through its IoT research, AOT is contributing to a smarter, more connected, and sustainable world.

Researchers:

Dr. Subir Panja (CSE Department)
Mrs. Priyanka Bhattacharya (CSE Department)

Projects:

Fuzzy-logic-based IoMT framework for COVID19 patient monitoring

Enhancing Security Using Deep Learning for IoMT

Agronomy with IoT Devices: The Smart Solution for Detection of Diseases of Betel Leaves

Problem Description: COVID-19 patients in home isolation or care centers require remote monitoring to minimize infection risks and protect healthcare workers. Existing systems often lack real-time responsiveness and depend on costly hardware or extensive training. Therefore, an efficient and cost-effective framework is essential—capable of generating timely alerts during high-risk situations and enabling preventive action. The solution should include a user-friendly mobile application for patient monitoring, while also ensuring robust security measures to protect sensitive medical data and maintain patient privacy.

Focus of Research:

A real-time COVID-19 health monitoring system
Real Time analysis of the health conditions and alarms
Identification of risk status of patient using Fuzzy logic

Publications:

Panja, Subir, Arup Kumar Chattopadhyay, Amitava Nag, and Jyoti Prakash Singh, Fuzzy-logic-based IoMT framework for COVID19 patient monitoring, Computers & Industrial Engineering, Vol. 176, 108941, February 2023. https://doi.org/10.1016/j.cie.2022.108941

Problem Description: IoT systems often operate with limited computational resources, minimal bandwidth, and in remote locations, making them difficult to update and secure. These constraints lead to challenges in implementing real-time anomaly detection and controlling device accessibility. Additionally, the vulnerability of IoT devices to malware insertion poses serious security risks, including unauthorized data access and system manipulation. Without effective malware categorization and built-in protection mechanisms, these systems remain exposed to evolving cyber threats, highlighting the urgent need for a lightweight, secure, and intelligent solution.

Focus of Research:

To investigate various threats and attacks as well as the several potential research challenges for secure IoT ecosystems.
To design authentication and access control mechanisms for the IoT frame- work in order to prevent attacks before they happen.
To design efficient, lightweight techniques for anomaly detection using various ML/DL algorithms for resource-constrained IoT devices.

The aim is to develop effective methodologies for identifying malicious software through the utilization of diverse ML and DL algorithms that are appropriate for safeguarding the IoT environment.

Publications:

Hota, Ashlesha, Subir Panja, and Amitava Nag, Lightweight CNN-based malware image classification for resource-constrained applications, Innovations in Systems and Software Engineering, Vol. 21, no. 1 pp.1-14, 2025. https://doi.org/10.1007/s11334-022-00461-7
Panja, Subir, Subhash Mondal, Amitava Nag, Jyoti Prakash Singh, Manob Jyoti Saikia, and Anup Kumar Barman, An efficient malware detection approach based on machine learning feature influence techniques for resource-constrained devices, IEEE Access (2025). https://10.1109/ACCESS.2025.3526878
Panja, Subir, Nituraj Patowary, Sanchita Saha, and Amitava Nag, Anomaly detection in iot using extended isolation forest, International Symposium on Artificial Intelligence, pp. 3-14. Cham: Springer Nature Switzerland, 2022. https://doi.org/10.1007/978-3-031-22485-0_1
Panja, Subir, Kajal Yadav, and Amitava Nag, Anomaly detection at the iot edge in iot-based smart home environment using deep learning, Proceedings of International Conference on Advanced Computing Applications: ICACA 2021, pp. 119-125. Singapore: Springer Singapore, 2021. https://doi.org/10.1007/978-981-16-5207-3_11

Panja, Subir, Suman Das, and Arindrajit Pal, Assessing the Effectiveness of Anomaly Detection in IoT Data Streams with Machine Learning, 2024 4th International Conference on Computer, Communication, Control & Information Technology (C3IT), pp. 1-5. IEEE, 2024. https://10.1109/C3IT60531.2024.10829458

Student Projects:

Title: Assessing the Effectiveness of Anomaly Detection in IoT Data Streams with Machine Learning
Student: Suman Das
Supervisor: Dr. Subir Panja (CSE Department)
Year: 2024

Problem Description: Agricultural productivity is declining in India due to traditional and unscientific farming practices, especially affecting economically significant crops like betel leaves. Diseases such as leaf rot severely damage betel crops, and manual detection is inefficient and time-consuming. With increasing population and food demand, there is a pressing need for intelligent, automated systems to monitor crop health. This study addresses the problem by integrating Internet of Things (IoT) and image processing with machine learning, specifically a Support Vector Machine (SVM) classifier, to automatically detect and classify diseased betel leaves, providing a smart and cost-effective solution for farmers.

Focus of Research:

Integration of IoT in Smart Agronomy: The paper emphasizes how Internet of Things (IoT) technologies—sensors, drones, and cloud platforms—transform traditional farming into smart agronomy, enabling automated monitoring and disease detection in crops like betel leaves.
Automated Disease Detection Using Machine Learning: It presents a methodology using image processing and a Support Vector Machine (SVM) classifier to automatically identify and classify diseased betel leaves from healthy ones, reducing the need for manual inspection.
Development of a Scalable Decision Support System: The proposed system aims to aid farmers through a mobile-based application for real-time plant health analysis, with future plans to incorporate aerial drone imagery and expand to other crops.

Publications:

Sarkar, I., Karforma, S., Bhattacharya, P., Bose, R., & Roy, S., Agronomy with IoT Devices: The Smart Solution for Detection of Diseases of Betel Leaves, Universal Journal of Agricultural Research, 11(1), 98-109, 2023. https://10.13189/ujar.2023.110109