ITU Scientists Won the Best Paper Award in International Conference

ITU scientists received the best paper award in EAI 7th International Conference on Industrial Networks and Intelligent Systems 2021 (EAI INISCOM 2021) with their work titled "A Robust Control Link for Aerial Mesh Networks in Contested Environments".

The work presented in an international conference by ITU Faculty of Computer and Informatics, Department of Artificial Intelligence and Data Engineering faculty members Assoc. Prof. Dr. Berk Canberk and Assist. Prof. Dr. Gökhan Seçinti together with our master’s students Fırat Rozkan Kılıç and Mehmet Özgen Özdoğan, whom they supervise within the framework of ITU-ASELSAN Academy Computer Engineering Master’s Program, won the best paper award.

In their paper titled "CLAN: A Robust Control Link for Aerial Mesh Networks in Contested Environments", our scientists propose a robust control link for aerial mesh networks jointly disseminating mesh and UAV control traffic, to ensure reliable control communication for aerial mesh networks, via multi-path, multihop control links in contested environment.

CLAN: A Robust Control Link for Aerial Mesh Networks in Contested Environments

Huge increase in the availability of commercial off-the-shelf unmanned aerial vehicles drastically shifts the way these devices operate and interact, enabling easy and affordable deployment of multiple drones to form a mesh network to perform aerial multi-robot missions such as 3D mapping, surveillance. This trend led to the development of unique applications and services relying on aerial mesh networks in contested environment, but also enables adversaries to improve their ability to counteract using simple techniques such as jamming.

In this paper, we propose a robust Control Link for Aerial mesh Networks, namely CLAN, jointly disseminating mesh and UAV control traffic, to ensure reliable control communication for aerial mesh networks, via multi-path, multihop control links in contested environment.

CLAN forms a dynamic tree topology, where control traffic is forwarded through multiple hops and utilizes multiple access technologies enabling multi-path end-to-end links in order to mitigate the effects of decreased signal power in longer ranges and the possibility of jamming attacks. Computer simulations of the contested environment show that the proposed CLAN algorithm that uses modified B.A.T.M.A.N. algorithm to selectively rebroadcast traffic control messages significantly reduces the traffic control message number by 90% while improving the connectivity of the nodes compared to single hop MAVLink communication up to 72%.

We congratulate our students and faculty members and wish them continued success.