Signal Processing Approach for Spectral Interferometry Immune to [Formula Omitted]/2 Errors

The paper develops a least-squares fitting approach for demodulating optical path difference (OPD) in spectral interferometric sensors. The proposed signal processing approach enables to substantially decrease the probability of abrupt [Formula Omitted]/2 demodulation errors. This is achieved by excluding from analysis a part of spectrum samples, providing less information, responsible for distinguishing the spectra of interferometrs, which OPDs differ at [Formula Omitted]/2. The amount of excluded samples is controlled by a numeric parameter. The dependency of optimal value of signal processing parameter on signal to noise ratio (SNR) of optical spectrum is derived. The proposed signal processing approach is tested experimentally, it’s efficiency in suppressing the [Formula Omitted]/2 demodulation errors and optimal parameter dependency on SNR were confirmed. It was demonstrated to provide a 4 dB effort in terms of minimal SNR providing error-free OPD demodulation.