Neural Network for Electronic Nose using Field Programmable Analog Arrays

Abstract

Electronic nose is a device detecting odors which is designed to resemblethe ability of the human nose, usually applied to the robot. The process ofidentification of the electronic nose will run into a problem when the gaswhich is detected has the same chemical element. Misidentification due tothe similarity of chemical properties of gases is possible; it can be solvedusing neural network algorithms. The attendance of Field ProgrammableAnalog Array (FPAA) enables the design and implementation of ananalog neural network, while the advantage of analog neural networkwhich is an input signal from the sensor can be processed directly by theFPAA without having to be converted into a digital signal. Direct analogsignal process can reduce errors due to conversion and speed up thecomputing process. The small size and low power usage of FPAA are verysuitable when it is used for the implementation of the electronic nose thatwill be applied to the robot. From this study, it was shown that theimplementation of analog neural network in FPAA can support theperformance of electronic nose in terms of flexibility (resource componentrequired), speed, and power consumption. To build an analog neuralnetwork with three input nodes and two output nodes only need twopieces of Configurable Analog Block (CAB), of the four provided by theFPAA. Analog neural network construction has a speed of the process0.375 μs, and requires only 59 ± 18mW resources.DOI:http://dx.doi.org/10.11591/ijece.v2i6.1501