Polarized low-coherence interferometry (PLCI) is a promising technology to absolute displacement measurement and sensor demodulation. To avoid jump errors induced by dispersion and achieve a high response, a dispersion compensation method based on fringe peak position estimation and polynomial fitting is proposed to get rid of the dependence on the priori model and birefringence parameters, and make this method very robust. The nearly linear resolution between the envelope and m -order PLCI fringe (PLCIF) centers are found and verified with both Gauss and non-Gauss profile broadband light sources. Then, the zero-order PLCIF is estimated and located to demodulate the measurement displacement. The demodulated results are significantly influenced by dispersion of the birefringence optical wedge and then polynomial fittings of sensing curve are conducted to raise the sensing accuracy. In comparison to the priori model and parameters based method, the proposed method can effectively avoid jump errors and has a higher accuracy than the directly envelope center method. Experimental results indicate that the measurement accuracy is higher than 19.5 nm which is enhanced by 6.88 times, the resolution is better than 2 nm, and its processing data rate can reach 35 kHz on an embedded system.

中文翻译：

---

基于条纹峰值估计和多项式拟合的PLCI色散补偿


偏振低相干干涉测量（PLCI）是一种很有前景的绝对位移测量和传感器解调技术。为了避免色散引起的跳跃误差并获得高响应，提出了一种基于条纹峰值位置估计和多项式拟合的色散补偿方法，摆脱了对先验模型和双折射参数的依赖，使该方法具有很强的鲁棒性。使用高斯和非高斯剖面宽带光源发现并验证了包络线和 m 阶 PLCI 条纹 (PLCIF) 中心之间的近线性分辨率。然后，对零阶PLCIF进行估计和定位，以解调测量位移。双折射光楔色散对解调结果影响较大，对传感曲线进行多项式拟合，以提高传感精度。与先验模型和基于参数的方法相比，该方法能够有效避免跳跃误差，并且比直接包络中心法具有更高的精度。实验结果表明，测量精度高于19.5 nm，提高了6.88倍，分辨率优于2 nm，在嵌入式系统上处理数据速率可达35 kHz。