Contributions and Relevant Publications Clause Samples

Contributions and Relevant Publications. ‌ The main contribution of this research is the development of a portable landmine detection system using NQR and improvement of its accuracy based on ML. The specific outcomes are listed below: 1. Development of the portable NQR system Systems based on commercially available NQR spectrometers are generally too bulky to use on the field. Therefore, a portable device that can detectresearch department composition X (cyclotrimethylene trinitramine) (RDX)explosives was developed in this study. The device uses afield-programmable gate array (FPGA), a power-efficient dual-supply class-D power amplifier powered by batteries, and a low-impedance transmission/reception circuitry to ensure sufficient magnetic field excitation for mine detection. The device is capable of detecting RDX explosives at a lower cost and in a smaller form factor, while having the shorter measurement time and sameSNRas previous landmine detectors using commercially available components. (Chapter 3) 2. Improvement of detection accuracy based on ML The ML model was implemented to improve the SNR of NQR measurements for improvement of the detection rate and for use without electrical shielding. ML was applied to NQR signals from RDX acquired with the developed system. Results show that the ML method can indeed improve the detection accuracy of the NQR device. We also installed the trained classifier on the controller of the system and confirmed that it can be performed with little time penalty compared to the conventional classification method. (Chapter 5) 3. Lab experiments and field trials The developed device was evaluated in laboratory and outdoor experiments. The system was designed to be robust and stable for outdoor use. It can be used in an environment outside the laboratory withoutelectromagnetic (EM)shielding and is capable of acquiring the signal of RDX. (Chapter 4) The publications related to the key contributions of this work are stated below: 1. ▇▇▇ ▇▇▇▇▇▇▇, ▇▇▇▇ ▇▇▇▇▇▇, ▇▇▇▇▇ ▇▇▇▇▇▇, ▇▇▇▇▇ ▇▇▇▇-▇▇▇▇▇, and ▇▇▇▇▇▇▇▇▇▇ ▇▇▇▇▇▇, “Development of a low-cost, portable NQR spectrometer for RDX explosives detection,” IEEE Sens. J., vol. 21, no. 5, pp. 6922-6929, 2021. (Chapter4,5) 2. ▇. ▇▇▇▇▇▇▇, ▇. ▇▇▇▇▇▇, and ▇. ▇▇▇▇▇▇, “Improving detection of a portable NQR system for humanitarian demining using machine learning,” IEEE Trans. Geosci. Remote Sens., doi: 10.1109/TGRS.2021.3101226, early access stage. (Chapter6)