Optimized Cavity Length Demodulation via Error Analysis in Graphene Fiber-Optic Fabry-Perot Sensor

Fiber-optic Fabry-Perot interference (FPI) sensors offer remarkable advantages in sensing applications, particularly in harsh environments, while errors from measurement and demodulation can deteriorate the sensing precision. To better understand the error origin, a systematic error model based on First Principles is established, including random and bias errors. An optimized cavity length demodulation (CLD) method, based on the neglected imperfect incidence spectrum, is proposed for cavity length extraction and spectrum decomposition. Demodulation experiments with a fixed cavity length were conducted on the graphene fiber-optic FPI sensor. The validity of the error model and optimized CLD (opt-CLD) method was verified. The random error mainly originates from the random wander of interference peak around the resolution of the spectrum analyzer, while the bias error mainly originates from the imperfect incidence spectrum. The graphene FPI sensor and opt-CLD method were applied to pressure sensing. The bias error was effectively eliminated, and the achieved random error was 20 times lower than that of the peak-peak demodulation (P-P) method. These findings in this article may provide valuable insights and solutions for the FPI sensor applications.