A Generic FPGA Module for QCM Sensor Array Processing using Neural Network

QCM (Quartz Crystal Microbalance) multi-sensors are widely used in biosensors and chemical sensors. The sensing film materials coated on their electrodes will determine the characteristic of the sensors. In gas or odor classification, the combination characteristic of each sensor will form a pattern that will be used to infer the result. The related research has carried out sensor characterization, sensor surface modification, development of an electronic signal processing system, and using machine learning to process the sensor signal. However, the integration and optimization of the processing still need to be done to support the success of the application stage and the search for new techniques for faster, portable, and configurable multisensor signal processing. This paper will describe a conceptual idea of generic hardware integrating frequency counter and signal processing using a neural network for multisensor QCM Processing implemented using FPGA SoC.