Normalization detection scheme for high-speed optical frequency-domain imaging and reflectometry

Abstract

We introduce a new signal detection method that can effectively suppress the effect of relative intensity noise (RIN) in optical frequency-domain reflectometry or imaging (OFDR/OFDI) schemes to enhance the sensitivity and dynamic range. In this method, spectral interferogram signal is normalized digitally by a spectral reference signal that contains the real-time spectrum and the RIN information of the frequency-swept source. Unlike the conventional balanced detection method that suppresses only additive intensity noises, we found that our proposed scheme removes both the additive and convolutional contributions of the RINs in the final interferogram signals. Experimental demonstrations were performed using a stretched-pulse optical coherence tomography (SP-OCT) system where the high RIN of a supercontinuum source had been a serious drawback. We have experimentally verified the superiority of our proposed scheme in terms of its improved dynamic range in comparison to the balanced detection method. In addition, we have shown that the noise suppression performance is immune to the spectral imbalance characteristics of the optical components used in the system, whereas the common-mode noise rejection of the conventional balanced detection method is influenced by them.