PhD - Gent | More than two weeks ago
Single unmanned aerial vehicles (UAVs) are well established in aerial monitoring and imagery. Recently, new application domains are being investigated that require the collaboration of multiple UAVs in a drone swarm, e.g. flying ad-hoc networks that can be used in situations where no fixed infrastructure is available, or environmental monitoring in a.o. archaeology or agriculture. The advantages of a drone swarm compared to a single UAV are that the cost of a swarm of multiple small UAVs may be smaller than the cost of a single large UAV needed for the same task, and that a swarm can provide a higher fault tolerance as a single UAV can be added to or removed from the swarm with limited impact on the formation. Therefore, swarms can quickly adapt to changing conditions and maintenance or replacement of single UAVs in the swarm.
A prerequisite for drone swarms is the ability to communicate between the drones, so that the swarm can act as a distributed entity. Because of the relative movement of the different UAVs in the swarm, the channel between UAVs and between a UAV and ground control needs to be modelled as time-varying. Further, currently most applications use a line-of-sight (LOS) link for communication. However, the assumption that LOS is available limits the applicability of swarms, as LOS may be lost and only multipath propagation is available. In such a situation, the channel becomes doubly-selective.
The purpose of this PhD is to design the physical link for drone swarms for applications in doubly-selective channels. This encompasses the design of the signal, as well as the design of algorithms for channel estimation, equalization and synchronization that have good performance and low complexity.
The successful PhD student will join a team that has over 25 years of experience in research on communication and signal processing for communication.
Required background: Electrical Engineering or Computer Science Engineering with interest in signal processing for communication
Type of work: 80% modelling/simulation, 10% experimental, 10% literature
Supervisor: Heidi Steendam
Co-supervisor: Andre Bourdoux
Daily advisor: Andre Bourdoux, Heidi Steendam
The reference code for this position is 2021-122. Mention this reference code on your application form.