SPRi传感器的数据处理方法

为了消除传感器输出数据中的高频噪声和低频漂移，以SPRi传感器的NaCl折射率样品和BSA生物样品检测的实验数据为例，研究零相移数字滤波、小波分析和希尔伯特-黄变换（HHT）3种数据处理方法及实际应用.结果表明，与零相移数字滤波相比，小波分析和HHT变换这两种时频分析方法在有用细节信息的提取方面更具优势.经这两种方法处理后，SPRi传感器的折射率分辨率达到1.4×10^-6 RIU，最小可检测0.025 g/L NaCl溶液和将原液稀释51 200倍的BSA抗体溶液.HHT变换具有自适应的特点，针对不同的实验数据，自动产生相应的基函数，并且操作简便，能够获得较理想的滤波效果，从而提高传感器的性能.

Data processing methods of SPRi sensor

Three kinds of data processing methods including zero-phase-shift digital filter, wavelet analysis, Hilbert-Huang transform (HHT) and their practical applications were analyzed in order to eliminate the high frequency noise and low frequency drift in the output data of sensors. Surface plasmon resonance imaging (SPRi) sensor experimental data of NaCl refractive index sample and the BSA biological sample detection were taken as an example. Results show that wavelet analysis and HHT, two time-frequency analysis methods, are more effective in extracting useful detail information compared with zero-phase-shift digital filter. After these two methods are processed, the refractive index resolution of the SPRi sensor goes to 1.4×10^-6 RIU, and the detections of limit are 0.025 g/L NaCl solution and the BSA antibody solution with 51 200 times dilution. HHT has the characteristic of self-adaptability, which can automatically generate the corresponding basis functions for different experimental data. It is easy to operate, and can obtain the ideal filtering result, thus improve the performance of the sensor.