The paper presents a novel system, the modulated wideband converter (MWC), for sub-Nyquist sampling of sparse wideband analog signals. The MWC uses periodic waveforms to alias the signal spectrum into baseband, followed by low-rate sampling. This approach allows efficient hardware implementation and low computational load, enabling spectrum-blind reconstruction of multiband signals. The system is designed to handle signals with unknown frequency support, which occupy only a small portion of a wide spectrum. The MWC's key advantage is its ability to sample at a rate much lower than the Nyquist rate, while still allowing for low-rate processing of individual bands without interpolation to the high Nyquist rate. The system is compared to previous methods, such as periodic nonuniform sampling and the random demodulator, which have limitations in bandwidth and computational complexity. The MWC's design ensures perfect recovery of the signal under certain conditions, and it is shown to be robust to noise, mismodeling, and quantization effects. The paper also discusses the practical implementation of the MWC, including the use of analog mixers, filters, and ADCs, and highlights the system's potential to overcome the bandwidth limitations of state-of-the-art analog conversion technologies. The MWC's ability to process signals with time-varying support and its efficient digital processing make it a promising solution for wideband signal sampling and reconstruction.The paper presents a novel system, the modulated wideband converter (MWC), for sub-Nyquist sampling of sparse wideband analog signals. The MWC uses periodic waveforms to alias the signal spectrum into baseband, followed by low-rate sampling. This approach allows efficient hardware implementation and low computational load, enabling spectrum-blind reconstruction of multiband signals. The system is designed to handle signals with unknown frequency support, which occupy only a small portion of a wide spectrum. The MWC's key advantage is its ability to sample at a rate much lower than the Nyquist rate, while still allowing for low-rate processing of individual bands without interpolation to the high Nyquist rate. The system is compared to previous methods, such as periodic nonuniform sampling and the random demodulator, which have limitations in bandwidth and computational complexity. The MWC's design ensures perfect recovery of the signal under certain conditions, and it is shown to be robust to noise, mismodeling, and quantization effects. The paper also discusses the practical implementation of the MWC, including the use of analog mixers, filters, and ADCs, and highlights the system's potential to overcome the bandwidth limitations of state-of-the-art analog conversion technologies. The MWC's ability to process signals with time-varying support and its efficient digital processing make it a promising solution for wideband signal sampling and reconstruction.