Noise Mitigation in Spectrum Sliced Microwave Photonic Signal Processors

Abstract

A new noise mitigation design strategy for spectrum sliced microwave photonic filters is presented. It is based on making the spectral slice bandwidths narrower than the modulation frequency and making the spacing between slices larger than the photodetector bandwidth. Advantages of this approach are that it is particularly simple, it does not require additional components, and it does not distort the signal. A detailed theoretical model of the relative intensity noise in spectrum sliced photonic signal processors is presented, including the effects of modulation, dispersion, and multislice stochastic sources. The phenomenon of modulation-induced noise translation in spectrum-sliced photonic signal processors is also reported for the first time. Results show that the new design strategy for noise mitigation can realize a significant increase in the SNR for a filter operating at 6.3 GHz