Through Wall Radar Imaging using Machine Learning
This project will investigate approaches that have been proposed to detect activities inside a building using through-the-wall radar imaging . First, the project will explore the main types of radar operating in the range of 1GHz to 3GHz, which can be used for the detection of activities behind walls. Second, the project will investigate and develop new machine learning (ML) algorithms for this purpose. Third, the project will assess the efficacy of such ML-based methods using real-world data and software simulations. The project will be carried out in collaboration with the Swiss startup company YOTASYS .
Stepped frequency and pulse compression radar are the most commonly used types of radar in through-the-wall radar imaging (TWRI). Even though the literature describes a wide range of potentially suitable methods, many open questions persist. In particular, the detectability of humans, wall modeling, and target differentiation have been identified as the main open questions in the field of TWRI. These challenges could potentially be addressed using ML algorithms—this project focuses on exactly this aspect.
 P. K. Nkwari, “Through-the-Wall Radar Imaging: A Review,” IETE Technical Review, Nov. 2018
- Looking for 1-2 Semester/Bachelor/Master students
- Contact: Christoph Studer
- Basic understanding of radar and signal processing
- Basic understanding of wireless communication
- Basic understanding of machine learning
- 20% Literature research
- 30% Theory
- 20% System-level simulation
- 30% Practical setup and tests