||As early as in the 1960s, drugs abuse has long been a serious problem in the world.The most popular one is marijuana, where 160 million people (79.4%) were reported to have addiction issues. The side effect is posed upon both mental and physical aspects.For instance, high heart beat rate, low blood pressure, short-term and working memory breakdown, coordination and concentration loss. It will lead to dangers during work, driving, and daily life. Thus, marijuana testing has become as urgent as that for alcohol.|
The first topic of this thesis proposes a highly sensitive portable microsystem for rapid marijuana detection, which is integrated with flexural plate wave microsonic sensor (FPW sensor). The system uses a single battery cell as the system power, and utilizes a direct digital frequency synthesizer (DDFS) to generate a frequency scanning signal, amplitude voltage converters, signal amplification circuits, and peak detectors, etc. The FPW sensor senses the resonance frequency shift caused by different concentrations of cannabinoid protein. It acquires the frequency, which is read by the control circuit and displayed on
the MCU. The linearity by physical experiments is up to 0.9992, and the maximum error is 12.378 KHz. The system is featured with no need for extra power supply, small size, fast detection, simple operation, which meets the demand of modern portable medical devices.
The second topic is to improve the accuracy of the peak detector circuit in the first topic. The peak detector uses an operational amplifier and a feedback circuit to compare the captured peak voltage automatically. Due to the mismatch of amplifier’s input stage, an input offset voltage will be generated to deteriorate the accuracy of the peak voltage acquisition. Therefore, the dynamic offset voltage cancellation technology is adopted,where the offset voltage is modulated to a higher frequency. Thus, the offset voltage is eliminated by using a low pass filter to increase the accuracy of the captured peak voltage. The input range is 0 V ~ 2.802 V and the precision can be as low as 0.558 mV as well as the offset voltage can be reduced to 0.3 mV by the simulations. The input range of previous work is 0 V ~ 1.985 V and the precision can be as low as 36.8 mV by measurement results.