Background

I joined Ingeniørhøjskolen i Aarhus, now Aarhus University School of Engineering in 2005. In 2013 I moved to the Department of Engineering, now Department of Electrical and Computer Engineering, where I am an Associate Professor specializing in signal processing, machine learning and instrument development focused on geophysical applications, in particular groundwater.

Current research projects 

Surface Nuclear Magnetic Resonance (NMR) - Surface NMR is an electromagnetic method for ground based non-invasive measurements of groundwater systems in the upper 100 m of the subsurface. Surface NMR is the only geophysical method that can directly determine the water content. Further, the NMR signal also provides information on the geological layers, from which estimates of hydraulic conductivity can be obtained. One drawback of surface NMR is the often very low signal-to-noise ratio, which makes measurements difficult or impossible in many locations. My research in surface NMR focus on new signal processing methods, new instruments, and new surface NMR field methodologies. My research in surface NMR, TEM and DCIP is done in a close collaboration with the Hydrogeophysics Group at Department of Geoscience.

Transient electromagnetics (TEM) - TEM is used for ground based or airborne resistivity measurements of the subsurface. Resistivity is correlated with the geological layers and is used for, e.g., groundwater mapping, mineral prospecting and geotechnical investigations. My research centers on instrumentation and signal processing methods. Examples include suppression of motion noise in TEM signals acquired from moving platforms, and applications of machine learning to speed up numerical calculations or automatically detect and cull noise events in data.

Direct current induced polarization (DCIP) - DCIP is a geophysical method for mapping the frequency dependent resistivity of the subsurface. It is used for many purposes, one important application being mapping of polluted areas. Like all electrical and electromagnetic geophysical methods, DCIP is also susceptible to noise. My research in DCIP is concerned with instrument development and development of machine learning based methods for noise reduction.

Other projects - Several previous projects have centered on applications of digital signal processing in audio technology, mostly in collaboration with Danish audio companies. For the time being, I have no funded projects within audio research. Occasionally, I supervise master thesis projects in this area.

Teaching activities

My current teaching activities are supervision of bachelor, master and Ph.D. students and teaching courses on:

• Differential equations
• Discrete-time signal processing
• Advanced signal processing

Previously, I have been teaching a variety of full courses, e.g., applied linear algebra and adaptive signal processing, and I have organized study courses in convex optimization and microphone array signal processing.