Shiva Raj Pokhrel

This paper proposes and evaluates a federated deep learning framework using autoencoders for lightweight, privacy-preserving, and scalable real-time anomaly detection in resource-constrained IoT networks.

The paper proposes Q-AGNN, a Quantum-Enhanced Attentive Graph Neural Network, to improve intrusion detection by modeling network flows as graphs and leveraging quantum circuits to capture complex relational dependencies.

This paper compares Google's BBR-v3 Congestion Control Algorithm to eight others over SpaceX's Starlink network, demonstrating its fairness and throughput maximization in high-latency, variable satellite environments.