基于实时阈值的CMOS有源像素传感器辐射致随机电报噪声自动检测

Auto-detection method for random telegraph signal noise detection for CMOS active pixel sensors based on real-time thresholds

研究了辐射环境下CMOS有源像素传感器产生的暗信号随机电报噪声(RTS)的特性,针对这一噪声提出一种实时自动检测算法,用以检测、重建暗信号RTS,并提取噪声关键参数。采用^(60)Cog射线进行的辐照实验测试结果表明,对于辐照后被观测像素发生的282次多级RTS转换,基于本文提出的实时阈值检测方法成功检测出了257次,检测效率达到91%,比采用暗信号幅值标准差为阈值的检测方法高出了16%。

Based on the study of radiation-induced random telegraph signal（RTS） noise in CMOS active pixel sensors, we propose a real-time auto-detection method for RTS detection, reconstruction, and parameters extraction. 60Co g-ray radiation experimental results reveal that, using the proposed method and employing the standard deviation of real-time Gaussian noise as the detection threshold, 257 out of 282 multilevel RTS transitions were successfully detected, thus, indicating a detection efficiency of 91%. This detection efficiency is 16% higher than that obtained in the detection method that uses signal standard deviation as the detection threshold.