Below is a selection of some of the most impactful undergraduate and postgraduate projects I have guided. In total, I have mentored over 30 innovative student research projects across multiple universities. Due to space constraints, only a few highlights are shown here.
Cloud and Containerization
- Cloud Container Technology Evaluation: This project enabled students to evaluate and benchmark Docker and Kubernetes environments for real-time microservice deployment. They built CI/CD pipelines using Jenkins and GitHub Actions, integrated monitoring using Prometheus and Grafana, and tested fault tolerance and horizontal scaling strategies. The project aligned closely with industry use cases in DevOps and cloud-native architectures.
- Secure IoT Data Sharing via Cloud: Students developed a secure data transmission model using DACES encryption integrated into a fog-cloud computing environment. The solution was designed for smart healthcare use cases, allowing sensor data to be securely offloaded to the cloud for further analysis while preserving patient privacy.
IoT and Edge/Fog Computing
- Machine Learning for Fog IoT Services: Students used real-world sensor data to develop a predictive machine learning model (Random Forest and XGBoost) to optimize energy consumption and task scheduling in fog-based architectures. This was implemented using Raspberry Pi devices as edge nodes and AWS Greengrass as the fog layer.
- Fully Homomorphic Encryption for IoT: A postgraduate group built an IoT security architecture that combined lightweight FHE with a Discrete Bee Colony Optimization algorithm. The solution was simulated using MATLAB and tested on Arduino and ESP8266 boards, aiming to enable privacy-preserving database updates over public networks.
Wireless Networks and Protocols
- Power-Aware Routing in WSNs: Using NS2 and MATLAB simulations, students proposed and tested custom routing protocols designed to balance energy consumption and extend network lifetime. Their work has potential applications in remote environmental monitoring and agriculture sensor networks.
- Energy-Efficient Clustering in MANET: Students implemented genetic algorithm-based clustering in OMNeT++ to enable dynamic network topology adaptation. This work supports emergency rescue communications in infrastructure-less environments like disaster zones.
Cyber Security and Cryptography
- Elliptic Curve Cryptography Application: This project involved building a secure messaging system for mobile phones using ECC for key exchange and AES for message encryption. Students performed comparative benchmarking with RSA and demonstrated significantly better performance on low-power devices.
- Blockchain-Based Access Control: A master's thesis group implemented a role-based access control model on the Ethereum Rinkeby test network using Solidity smart contracts. Their system was validated in a simulated healthcare data-sharing scenario, ensuring tamper-proof audit trails and decentralized authentication.
Artificial Intelligence and Data Analytics
- AI-Enabled Intrusion Detection: Students developed a hybrid anomaly-based intrusion detection system using Random Forest for feature selection and SVM for classification. Using the CICIDS2017 dataset, the solution achieved over 92% accuracy and was tested in a Mininet-based virtual network lab.