Design and Analysis of High-Precision Weighted Linear Least-Squares Fitting Algorithm for Fluorescent Optical Fiber Temperature Sensor

Jian Yang,Xiangliang Jin,Yan Peng,Jun Luo
DOI: https://doi.org/10.1142/s0218126622500311
2021-01-01
Journal of Circuits Systems and Computers
Abstract:Microwave hyperthermia is a new method of treating cancer, where the therapeutic effect is determined by the heating temperature. Traditional active temperature sensors are interfered by high frequency so that the accuracy of temperature measurement cannot be guaranteed. It is of great significance to study the high-precision fluorescent optical fiber temperature sensor with complete insulation. This paper has realized a compact and practical fluorescent optical fiber temperature sensor after studying the optical path, circuit, data processing algorithm. In order to improve the accuracy of the system, the weighted linear least-squares fitting algorithm is improved in this paper. Through experimental tests, compared with the standard linear least-squares fitting algorithm and the unimproved weighted linear least-squares fitting algorithm, the accuracy of the algorithm is improved by about 98% and 65.5%, respectively. In addition, the response time is reduced by about 36.5%, compared with the unimproved weighted linear least-squares fitting algorithm. This algorithm fully meets the precision requirements of microwave hyperthermia.
What problem does this paper attempt to address?