期刊
JOURNAL OF MAGNETIC RESONANCE
卷 299, 期 -, 页码 74-92出版社
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.jmr.2018.12.007
关键词
NMR spectrometer; System-on-Chip (SoC); Auto-tuning; Autonomous operation
资金
- National Science Foundation (NSF) [CCF-1563688]
This paper describes the development of a portable and self-optimizing NMR spectrometer based on a miniaturized custom analog front-end and a System-on-Chip (SoC)-based digital back-end. The SoC integrates a field-programmable gate array (FPGA) fabric with a hard processor running a Linux operating system, thus enabling fully-autonomous operation without the need for an external computer. In the proposed approach, data captured by the FPGA fabric during regular operation is transported to the hard processor using an integrated on-chip bus for further processing. The processed results are then used to automatically estimate parameter values that optimize a suitable cost function, such as signal-to-noise ratio (SNR) per unit time. Finally, the optimized values of both electrical and NMR-related tuning parameters (e.g., preamplifier gain and frequency response, pulse length and amplitude, operating frequency, etc.) are programmed back into the front-end and back-end hardware. Experimental NMR results from various samples in a similar to 0.1 T permanent magnet are presented to verify the operation of the proposed spectrometer. These demonstrate on-board Laplace inversion and automated frequency tuning to compensate for temperature changes. Preliminary N-14 NQR results are also presented. (C) 2018 Elsevier Inc. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据