LAUSR

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 ∼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 14N NQR results are also presented.