探测器系统中基于 MHz ADC 的整形脉冲定时与表征:比较研究与深度学习方法
数据分析、统计与概率
2019-04-02 v3
摘要
基于模数转换器 (ADC) 的定时系统广泛用于先前高能物理探测器的设计中。本文中,我们提出一种基于深度学习的新方法,从有限的 ADC 样本集中提取时间信息。首先,针对三类变化(长期漂移、短期变化与随机噪声),结合仿真图示给出了传统曲线拟合方法的定量分析。接着,对曲线拟合与神经网络进行比较研究,以展示深度学习在该问题中的潜力。仿真表明,专用的网络架构能大幅抑制噪声 RMS 并改善非理想条件下的定时分辨率。最后,利用 ALICE PHOS FEE 卡开展实验。实验条件下,我们方法的性能优于曲线拟合 20% 以上。
引用
@article{arxiv.1901.07836,
title = {Timing and characterization of shaped pulses with MHz ADCs in a detector system: a comparative study and deep learning approach},
author = {Pengcheng Ai and Dong Wang and Guangming Huang and Ni Fang and Deli Xu and Fan Zhang},
journal= {arXiv preprint arXiv:1901.07836},
year = {2019}
}
备注
24 pages, 9 figures, 4 tables. This is the Accepted Manuscript version of an article accepted for publication in Journal of Instrumentation. Neither SISSA Medialab Srl nor IOP Publishing Ltd is responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at https://doi.org/10.1088/1748-0221/14/03/P03002