Photonics-enabled sub-Nyquist radio frequency sensing based on temporal channelization and compressive sensing

Abstract

A novel approach to sensing broadband radio frequency (RF) spectrum beyond the Nyquist limit based on photonic temporal channelization and compressive sensing is proposed. A spectrally-sparse RF signal with unknown frequencies is modulated onto a highly chirped optical pulse. An optical channelizer slices the modulated pulse spectrum, which is equivalent to temporally sampling the RF waveform thanks to the dispersion-induced wavelength-to-time mapping. This serial-to-parallel conversion avoids the use of a high-speed detector and digitizer. Furthermore, compressive sensing with optical random demodulation is achieved using a spatial light modulator, enabling the system to capture the wideband multi-tone RF signal with a sampling rate far lower than the Nyquist rate. It is demonstrated that the temporal channelization system with a channel spacing of 20 GHz achieves RF spectrum sensing with a high resolution of 196 MHz. With an equivalent sampling rate of only 25 GHz, a 50-GHz broadband two-tone RF signal can be captured and reconstructed by the system thanks to compressive sensing with a compression ratio of 4.