Reserach

Henrik's research is in the area of system identification, machine learning, compressive sensing, signal processing, change detection and energy applications. Henrik is involved in a number of research projects, some which are described below.

Nonlinear Compressive Sensing

While compressive sensing has been one of the most vibrant research fields in the past few years, most development only applies to linear models. This limits its application and excludes many areas where compressive sensing could make a difference. One such area is phase retrieval. In this research project we aim to extend compressive sensing to the nonlinear phase retrieval problem.

Disaggregation

In many applications it is costly or impossible to measure the sources one-by-one and all that is available is some aggregated measurements of their contributions. Disaggregation is the task of separating a signal into its sources. It is a fairly new area and with applications in fields such as chemistry, biology, etc.

Hybrid System Identification

Hybrid system identification is an area within system identification which deals with the identification of models of hybrid systems. Hybrid system identification is an extremely complex task and as today most research has been dealing with the most basic type of hybrid model -- the piecewise affine model.