A ROM-less DDFS with Curvature Non-equal Division ParabolicPolynomial Interpolation Method and Frequency-Shift ReadoutCircuit for Marijuana Rapid Quantitative Test Microsystem

• Main Authors: Yun-Shen Liu, 劉運昇
• Other Authors: Chua-Chin Wang
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/3z63e8

碩士 === 國立中山大學 === 電機工程學系研究所 === 106 === This thesis was motivated by an MOST-sponsored project, ”Highly Sensitive Microsystem for Marijuana Rapid Quantitative Test”, where a frequency-shift readout circuit is required to carry out the expected functionality and a ROM-less DDFS (direct digital frequency synthesizer) based on curvature non-equal division parabolic polynomial interpolation method is also proposed to serve as a frequency generator therein to enhance the performance. The first topic investigates the marijuana rapid quantitative test system. The flexural plate wave (FPW) sensor and a control circuit are integrated, where digital to analog converter (DAC), DDFS, amplitude voltage converter (AVC) and peak detector are all included in the control circuit. By taking the advantage of FPW sensor characteristic, which is the resonant frequency will be shifted before and after titrating the target protein on FPW, a sinusoid signal with different frequency scanning over a pre-defined range is needed to seize the resonant frequencies, before and after titrating the protein. By computing the resonant frequency shift of the same FPW device, we can estimate the concentration by a look-up table searching. Since the sine wave generator is the key component of the readout circuit, a novel DDFS is proposed to improve the spurious free dynamic range (SFDR). The proposed non-equal division parabolic polynomial interpolation method depending on curvature variation analysis will generate a complete sine wave based on the synthesis of a quarter of a sine wave digital signals due to the symmetry characteristic. The proposed design not only reduces the absolute error between ideal sine wave and generated sine wave, but also improves the SFDR.