微弱光信号的极值检测及滤波算法研究

为有效提高微弱光电信号极值检测的精度，通过分析被测信号的变化规律，提出了高精度目标信号极值检测滤波算法，探讨了在理想条件和强干扰条件下的信号检测解决方法。该算法以改进的动态阈值法获得极值初始数据集，通过基于中值数绝对偏差的滑动滤波和加权平均滤波法相结合的方式对采集值进行滤波处理，最终获得信号的极值。实验结果表明，该算法测量结果和实际值相吻合，最小平均相对误差控制在0.1%左右，满足精度要求,测量精度及稳定性高于传统方法。该研究对不同波形微弱信号的高精度极值检测提供了参考依据。

Research on extremum detection and filtering algorithm of weak light signal

In order to effectively improve the accuracy of extreme value detection of weak photoelectric signal, through the analysis of the change rules of signals measured, a filtering algorithm for the extremum detection of the target signals with high precision was proposed, and the solutions to the signal detection under ideal conditions and strong interference conditions were also discussed. Firstly, in this algorithm, an improved dynamic threshold method was used to obtain the initial extremum data. Then the acquisition value was filtered by the combination of two kinds of filtering method, which were the median absolute deviation based moving filtering and the weighted average filtering. Finally, the extreme value of the signal was obtained. The test results show that the measurement results of the algorithm are consistent with the actual value, the minimum average relative error is controlled at about 0.1%, which meets the accuracy requirement. The accuracy and stability of this algorithm are higher than those of the traditional method. The research provides a certain reference for the high accurate measurement of peak and trough values of weak signals with different waveform.