Analog-to-information conversion by encoded multi-cosets.

This paper presents an Analog-to-Information Conversion (AIC) model to acquire and simultaneously compress, with Time-Interleaved Analog-to-Digital Converters (TI-ADC), wide-band analog signals that can be sparse in a variety of domains. Compared with current AIC acquisition methods also based on TI-ADC, the proposed scheme represents a useful generalisation because it embodies a recently theoretically proved Restricted Isometry Property (RIP)-compliant sampling operator structure in the Compressed Sensing (CS) framework and, hence, favours the universal acquisition requirement for stable information recovery. The proposed model achieves its enhanced measurement characteristics by means of a Digital Signal Processing (DSP)-based encoding multi-coset core that operates completely in the low-rate section. [ABSTRACT FROM AUTHOR]