光寻址电位传感器的噪声分析与信号处理方法研究

Research on noise analysis and signal processing method of light addressable potentiometric sensor

鉴于光寻址电位传感器(LAPS)的响应信号较弱,传统的抽取频域基波分量的方法易受信号漂移和随机噪声的影响,提出了一种基于频域分量均方根和卡尔曼滤波的两步信号处理方法。基于光寻址电位传感器件的理论模型构建了其等效电路模型,推导得到输出信号的表达式,分析了漂移与噪声产生的原因及其抑制方法。通过实验检测了不同pH值的溶液,并采集了系统输出的光电流信号。求取信号傅里叶变换后频域中的基波分量、二次谐波分量、三次谐波分量的谱线幅值的均方根,然后对归一化电流-偏压(I-V)特性曲线进行卡尔曼滤波。实验结果表明,相对于单纯抽取基波分量的方法,基于频域分量均方根和卡尔曼滤波的两步信号处理方法使检测结果的均方差(MSE)降低了97%,显著减少了信号漂移和随机噪声对检测结果的影响。

Considering that the response signal of light addressable potentiometric sensor（LAPS）is quite week and the traditional method for extraction of fundamental component of frequency domain is easily impacted by signal drift and stochastic noise,a kind of two-step signal processing method based on the mean square root of component of frequency domain and Kalman filtering was proposed.Based on the theoretical model of devices of light addressable potentiometric sensor,the equivalent circuit model was established;the expression of output signal was deduced,and the causes of drifting and noise as well as inhibition methods were analyzed.Through tests,the solutes with different pH values were detected,and the photocurrent signal outputted by systems was collected.The mean square roots of breadth value of spectral line of fundamental component,second harmonic component and third harmonic component in the frequency domain after Fourier transformation were calculated.Then,the Kalman filtering was implemented for characteristic curve of normalized current-bias（I-V）.The experimental results indicate that the two-step signal processing method based on the mean square root of component of frequency domain and Kalman filtering makes the mean square error（MSE）of testing result reduced by 97%,and the influences of signal drift and stochastic noise on testing result are obviously decreased compared to the simple method to extraction of fundamental component.