Digital compensation of lowpass filters imperfection in the Modulated Wideband Converter compressed sensing scheme for radio frequency monitoring.

Highlights • Non-ideal lowpass filters are implemented to the MWC scheme for wideband monitoring. • All imperfections of the non-ideal filters are detailed: “amplitude” and “phase”. • Digital post-treatment method has been proposed to compensate non-ideal lowpass filter imperfections. • Simulations with several parameters of the MWC scheme result in their probability of good reconstruction and false alarm. • The obtained outputs of practical lowpass filters are as good as ideal filter output. Abstract This paper focuses on non-ideal filters in a Modulated Wideband Converter (MWC) scheme. The MWC is a system that can sample a sparse wideband signal at sub-Nyquist rate. Generally, the output of the ideal MWC components will ensure a perfect reconstruction. In practice, the reconstruction should be based on the output of non-ideal components, especially filters. The impact of non-ideal filters will trigger to a bad reconstruction. In this paper, a detailed study on non-ideal lowpass filters imperfection used in compressed sensing MWC scheme is synthesized. A digital post-treatment scheme with amplitude and phase compensation is proposed after real lowpass filtering step in order to have the filtered output as close as the ideal lowpass filter output. At last, reconstruction spectra obtained from different simulated lowpass filters are compared with different parameters of MWC. [ABSTRACT FROM AUTHOR]