Compressive sensing applied to ultra-wideband radio channel measurements

This paper presents results of applying several compressive sensing approaches to measured radio channel frequency response data. We evaluate the relative error of the reconstructed signal using compressive sensing with respect to reference data derived from the measured data. In particular, we consider an ultra-wideband measurement setup in an EMC chamber that basically corresponds to a dominant two ray (line of sight and ground reflection) plus minor multipath scenario. The measured data (acquired without compressive sensing) are decimated according to the compressive sensing schemes under consideration. Moreover, some examples of recovered impulse responses are displayed to illustrate the main effects graphically.