Processing the Double Channel MGG Signal Based on Independent Component Analysis

Giant Magnetoimpedance (GMI) sensor is the hot research topics in sensor engineering. In this paper we setup a double channel acquisition system for MGG signal. Then FastICA algorithm is applied in the raw data and We get the spectrum after ICA processing. The results show that ICA is powerful tool for MGG signal processing to separate the noise signal. Introduction Recently giant Magnetoimpedance(GMI) sensor is the new tools for the researchers to acquire the magnetism signal. As before it is very difficult to acquire such signal in many fields, such as biosignal and magnetism signal, and such new material is allied in the sensor engineering. Giant magnetoimpedance in amorphous wires [1] will be widely exploited in the design of new magnetic field sensors [2]. It was reinvented quite recently that large variation of AC impedance with applied magnetic field can be observed also in the conventional soft magnetic metals such as Mo–permalloy wires [3] or mumetal strips [4]. In this paper GMI sensors are applied in the biosignal processing fields to acquire the MGG (Magnetogastrogram) for the first time, and ICA method is utilized in the processing tool. Data Acquisition One healthy male subject participated in the experiment. Magnetogastrogram data was collected from two locations on the skin beside the stomach simultaneously. We put up plastic tape on the skin first, and then attach the two sensors with another plastic tape to avoid the sensors touching skin directly. The sensors are connected to a USB acquisition card produced by National Instruments Corporation with a sampling frequency of 1000Hz for 8 minutes. FIG.1 DOUBLE CHANNEL ACQUISITION LOCATION Applied Mechanics and Materials Vol. 344 (2013) pp 129-131 Online available since 2013/Jul/31 at www.scientific.net © (2013) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMM.344.129