Photonic compressive sensing of sparse radio frequency signals with a single dual-electrode Mach-Zehnder modulator.

Abstract

A novel approach to realizing compressive sensing (CS) of sparse radio frequency (RF) signals based on photonic random demodulation (RD) is proposed. The key function of mixing the RF signal under test and the bipolar pseudo-random binary sequence (PRBS) in photonic RD is implemented with a single dual-electrode Mach-Zehnder modulator (DEMZM). By properly setting the DC bias of the DEMZM at Vπ and the voltages of the PRBS at ±Vπ/2, a pure desired multiplication term between the signal and the bipolar PRBS is obtained after an AC-coupled photodetector (PD), which not only simplifies the modeling of the CS link but also improves the recovery performance. A proof-of-concept experiment is demonstrated where a sparse signal with spectral components of 500 MHz and 950 MHz is successfully identified with a compression ratio of 20. Simulation results are also given to show the advantage of the given photonic CS scheme with bipolar random mixing.