Research on fiber Bragg grating Fabry-Perots bending sensor based on principal component analysis

Optical fiber sensors have advantages such as small size and good biocompatibility, leading to widespread interest in the field of hand motion tracking. In this work, we designed a multi-core fiber Bragg grating Fabry-Perots (FBG-FPs) sensor based on FP interference theory. We inscribed short FBGs with different wavelengths in different fiber cores to form the FP, and then employed thermal diffusion technology for single-channel acquisition of the multi-core fiber FBG-FPs sensor. We also proposed a signal demodulation method for the FBG-FPs sensor based on the principal component analysis (PCA) algorithm and conducted experiments on strain, temperature, and bending sensing. The results indicate that the bending sensitivities of the three external cores exhibit strong dependence on bending, with a maximum sensitivity of 52.8 pm/m^-1. The direction and curvature of bending can be reconstructed using the wavelength shifts obtained from the PCA algorithm. This sensor features a compact structure, flexible fabrication, and low cost, improving the adaptability to individual differences in optical fiber smart gloves and enhancing multiplexing sensing capabilities, thus providing broad prospects for the development of human-machine interfaces and medical health monitoring.