MA: Detection and classification of jammers in a distributed system
(suitable for HW/FP/MA; scope and depth will be adapted to ECTS and prior knowledge)
Description
In modern wireless communication systems, deliberate interference, also known as jamming, poses a significant threat. Such attacks can be classified as a type of denial of service (DoS) attack and jeopardize the integrity and availability of communication systems. Reliable detection and classification of jammers is essential in order to take appropriate countermeasures. Both the identification of the interference power and the detection of the jammer’s signal shape play a central role in this.
Research questions
- How effective are local decision-making processes in individual sensors compared to central data fusion in the detection and classification of jamming signals in a distributed sensor network?
- Which algorithms or methods for data processing and analysis offer the highest accuracy in distinguishing between different interference signal forms and power levels in a distributed sensor system?
- To what extent does the network architecture (e.g., number and placement of sensors) influence the ability of a distributed system to detect and classify jamming signals, and how can this architecture be optimized?
Objectives of the work
The project aims to develop innovative approaches for detecting and classifying jammers in a distributed sensor network and to compare their effectiveness. The focus is on two central approaches:
- Local decision per sensor and central fusion:
- Sensors make individual local decisions based on the data they receive.
- The local decisions are fused in a central unit to enable a meaningful overall analysis.
- Central decision using all sensor data:
- All raw data from the sensors is transmitted to a central unit.
- Global decision-making is carried out using all available data.
Work packages
The following sub-tasks can be derived from the description and research questions:
- Investigation of algorithms suitable for the detection and classification of jammers in a distributed network.
- Processing of time series of received IQ data to develop powerful classifiers.
- Division of an existing data set into training and test data, combined with appropriate labeling of the data.
- Comparison of the performance and efficiency of the two approaches by means of benchmarking in simulated and/or real scenarios.
Optional:
- Extension of the investigations to different topologies and sensor densities.
You are also welcome to discuss and contribute your own interests.
Your knowledge
- Fundamentals of information and communication technology
- Fundamentals of deep learning and neural networks
- Experience with AI/machine learning and tools such as Keras or TensorFlow
- Confident use of Python
References
[1] A. Mehrabian and G. Kaddoum, “Cooperative Jamming Detection Using Low-Rank Structure of Received Signal Matrix,” in IEEE Transactions on Communications, vol. 73, no. 11, pp. 12899-12912, Nov. 2025, doi: 10.1109/TCOMM.2025.3592583.
[2] Cortés-Leal A, Del-Valle-Soto C, Cardenas C, Valdivia LJ, Del Puerto-Flores JA. Performance Metric Analysis for a Jamming Detection Mechanism under Collaborative and Cooperative Schemes in Industrial Wireless Sensor Networks. Sensors (Basel). 2021 Dec 28;22(1):178. doi: 10.3390/s22010178.
If you are interested in this exciting research project, we look forward to receiving your application with a short resume at
