Conference Paper – Radio Lab

An Enhanced End-to-End Framework for Drone RF Signal Classification

eguffens — Mon, 01 Sep 2025 14:14:00 +0000

Smart RF jamming relies on long-term spectrum prediction, which requires accurate, high-resolution RF detection and classification over extended observation periods. Detecting and classifying drone RF signals is particularly challenging due to short dwell times, high hopping rates, and narrow instantaneous bandwidths. This paper presents an enhanced end-to-end framework designed to meet these requirements for smart RF jamming, delivering high-resolution and precise detection and classification. We demonstrate that our Residual Neural Network (ResNet)-based You Only Look Once (YOLO) model effectively detects and extracts RF features from previously unseen drone signals with high accuracy, even when trained solely on a synthetic RF dataset. Furthermore, our ResNet classifier outperforms existing models, achieving 99.29% accuracy at 0 dB signal-to-noise ratio (SNR) for drone RF signals.

Link to the paper

Researcher

Sanjoy Basak

Project

AIRFORC

UAVs as Relay Nodes: Positioning Method for Heterogeneous Wireless Networks

eguffens — Wed, 05 Jun 2024 15:53:00 +0000

In scenarios where Relay Nodes (RNs) need to be deployed to restore the connectivity of a partitioned wireless network, a positioning method limiting the number of RNs required is necessary. Optimally, the method must take into account the differences of transmission ranges between the different nodes – or the heterogeneity in range of the network. In particular, Unmanned Aerial Vehicles (UAVs) are considered to be great candidates as RNs with their high mobility and ability to be quickly deployed. However, their radio parameters will be different from the Initial Nodes (INs) of the network and there will be a diversity in propagation channel. Plus, even the network might be heterogeneous. With a mobile initial network, the placement method also have to be stable in terms of RN positions and number with respect to IN displacements. This paper introduces an approach to modify an effective RN placement algorithm for homogeneous networks to create a new algorithm capable of managing heterogeneity in the networks. The new algorithm is the Barycenter-focused Relay Positioning for Heterogeneous Wireless Networks (BRHEN) algorithm. In addition to lower the number of RNs compared to the initial algorithm, it also enhances its stability with respect to small IN displacements.

Link to the paper

Researcher

Eliott Guffens

Project

DAP21-06

Distributed Cooperative Jamming with Multi-Reference Known-Interference Cancellation

eguffens — Wed, 05 Jun 2024 15:09:00 +0000

Secure and reliable tactical communications within allied defense forces across a battlefield are often fundamental for achieving the forces’ operational targets against an adversary. Likewise, limiting the adversary’s capability to communicate securely and reliably promotes the host forces’ chances for operational success. As such, armed conflicts typically involve an underlying battle in the electromagnetic (EM) spectrum to facilitate one’s own communications and limit the opposition’s. In this work, we propose the use of distributed cooperative jamming for augmenting tactical communications and gaining a technological advantage in that underlying battle. Specifically, we propose a multi-reference known-interference cancellation (KIC) method that allows the host force tactical communication nodes to cancel known interference (KI) from multiple cooperative jammers simultaneously. Relying on simulations, we then study how cooperative jamming affects the opposing forces’ capabilities to use the EM spectrum in a simplified battlefield. Results show that cooperative jamming gives an advantage to those controlling the jammers, as the opposition’s use of the EM spectrum is obstructed for both communications and signals intelligence.

Link to the paper

Researcher

Vincent Le Nir

Projects

DAP21-01, IST-175

Assessment of the Antenna Mounting Location on a Military Vehicle for Jamming Applications Based on Computational Electromagnetics

eguffens — Wed, 20 Sep 2023 13:31:00 +0000

The placement of an antenna on the roof of a military vehicle-mounted jamming system presents many challenges. For that, further approaches are needed to study the effectiveness of placing antennas on these vehicles. Besides, due to the cost, time, and technical difficulties in measuring the suitable antenna locations on large tactical vehicles, there is a demand for methods to compute the optimal position for antennas before installation and for jamming coverage of 360° around the vehicle. In this paper, we analyze a simulation model based on an UWB (Ultra-Wideband) discone antenna placed on the roof of a MPP (Multi-Purpose Protected) military vehicle. The proposed discone antenna shows good UWB characteristics, and operates in the frequency band from 0.81 GHz to 6 GHz for a VSWR below 2:1 over the whole bandwidth. It displays a good radiation pattern behavior with a good average radiation efficiency of around 71 % when mounted on a military vehicle. In addition, the radiation pattern was compared for the stand-alone antenna in free space taking as a reference and at four locations on the roof of the military vehicles, at five frequencies. The antenna at the back of the roof of the vehicle has the most optimal results, with adequate far-field behaviors for jamming applications. The results confirm the significant advantage of an UWB discone antenna that could be installed on a military vehicle-mounted jamming system platform as an alternative potential candidate to conventional antennas. ‘This paper was originally presented at the NATO Science and Technology Organization Symposium (ICMCIS) organized by the Information Systems Technology (IST) Panel, IST-200RSY – the ICMCIS, held in Skopje, North Macedonia, 16-17 May 2023’.

Link to the paper

Researchers

Radouane Karli

Mathias Beacquaert

Project

DAP22-07