Characterization of Analog-to-Information Converters

The dissertation deals with the characterization of sub-sampling devices, the Analog-to-Information Converters (AICs), recently proposed in research as an alternative to the traditional Analog-to-Digital Converters. By exploiting the Compressive Sampling paradigm, AICs stably recover the input signal, previously acquired with less samples than needed by the Nyquist-Shannon theorem, reducing simultaneously sampling frequency and acquisition memory. The dissertation research was carried out in order to design and develop an AIC prototype to be employed in a measurement instrument: a vector signal analyzer, that acquires and reconstructs sparse wideband signals. The aim of such an instrument is to guarantee proper monitoring, circumventing the trade-off between resolution and sampling frequency, typical of conventional data acquisition systems. With a view to this use, the characterization of AICs is of crucial importance, since their performance conditions the measurement instrument based on them, and, although several efforts have been made in research to propose several architectures, only in a few cases their performance is assessed.