Reduction of the order of anti-aliasing filters in embedded avionics data acquisition

Abstract

Research on the applications of compressed sensing has been mainly directed towards sub-Nyquist acquisition of a sparse signal followed by reconstruction, using different recovery algorithms, to get the original signal. In this paper, we present a scheme of compressed acquisition of a set of sparse signals and subsequent reconstruction of the signals at an effective sampling rate which is a multiple of the rated sampling rate of the analog to digital converters employed for data acquisition. We demonstrate that with a higher effective sampling rate the frequency specifications of the front-end anti-aliasing (AA) filter can be relaxed and it is possible to use a lower order AA filter, thereby simplifying the analog design on the embedded board. To verify the proof of concept we test the proposed acquisition architecture for sampling voltages proportional to the deflection of control surfaces of an in-flight aerospace vehicle.