Search the digital library catalog

Abstract: In this Master’s thesis, we study in detail NFC technology, NFC standardization, Artificial intelligence and GUI topics.Our goal is to optimize the measurement process of the FeliCa listener phase window in order to define the best analog parameters for the NFC IC. Namely, by combining machine learning and test automation, we want to achieve a faster validation time of the NFC IC under test. Firstly, we introduce the main topics related to NFC and present the thesis alongside specific goals. Next, we present the NFC technology on a fewuse cases. We also cover international standards, emphasizing sections that directly affect our validation and, therefore, the understanding of our solution. The NFC coverage is described alongside data representation used to interpret data used in proprietary software. Artificial intelligence is integrated into this software for multidimensional data interpolation task. We also cover the basics of deep neural networks and the most common architectures used in the DL field. The neural network training process is also covered to understand how we can optimize neural network global minimum convergence. In the last part of the thesis, we also explain how proprietary software tool actually works and how it can outperform previous processes. In the end, specific solutions related to the tool are described in more detail to understand our software development decisions better. Finally, we present measurement results and evaluate our goals.
Keywords: NFC wireless communication, artificial intelligence, test coverage, FeliCa, graphical user interface, analog parameter extraction
Published in DKUM: 08.06.2022; Views: 1172; Downloads: 0