Research Projects

The advent of 6G technology brings unprecedented advancements to Unmanned Aerial Vehicle (UAV) networks, such as ultra-low latency, improved energy efficiency, and optimized resource management. Edge computing, meanwhile, brings computational power closer to the UAV's operational environment, enhancing responsiveness. Combining 6G with edge computing offers the potential to address several challenges faced by UAV networks, including the latency-energy tradeoff, scalability, data security, seamless connectivity, and reliability of UAV applications such as crop monitoring, disaster relief, delivery services, wildlife tracking, security and surveillance. However, developing a framework that allows UAVs to effectively share experiences within an edge network while leveraging 6G technology remains a complex challenge. Therefore, this proposal seeks to integrate these capabilities into UAV networks through a decentralized federated learning approach with a strong emphasis on UAV-to-UAV communication, security, and sustainability. Our focus is on three primary objectives: (i) equilibrium in latency, energy consumption, and resource allocation in UAV network models using 6G technology, (ii) developing a scalable and secure edge federation framework using a weighted federated average approach, and (iii) implementing a cluster-oriented pseudo-hierarchical decentralized learning architecture to ensure data integrity. To achieve these objectives, we will develop latency and energy-aware federated learning algorithms that dynamically adjust based on 6G network parameters. We will also design secure 6G-powered edge nodes to facilitate efficient UAV-to-UAV communication and ensure data privacy through a decentralized edge federation model, where the weighted federated average approach will be employed to enhance the security and robustness of the learning process. Finally, we will implement a cluster-based learning model that optimizes resource allocation and balances data processing across the network, thereby reducing the risk of a single point of failure in the UAV network. The evaluation of the proposed framework will be done in two stages. In the initial stage, we will validate the theoretical feasibility of the proposed model along with its accuracy using existing time series tiand datasets and consider various 6G network and system-based parameters, such as transmission bandwidth, energy consumption, latency, resource utilization, and throughput. In the final stage, an experimental testbed will be developed using 6G-supported UAVs and edge nodes to analyze data transmission delays, model convergence time, and robustness during operations.