Sirma Orguc1, Harneet Singh Khurana1, Konstantina Stankovic2, Hae-Seung Lee, Anantha Chandrakasan1
13:30 - 15:00 | Tue 30 Oct | Ambassador A | B4L-B.1
An electromyogram (EMG) signal acquisition system capable of real time classification of several facial gestures is presented. The training data is collected from 10 individuals (5 female/5 male). A custom-designed sensor interface integrated circuit (IC) consisting of an amplifier and an ADC, implemented in 65nm CMOS technology, is used for single channel EMG signal acquisition [1]. Discrete wavelet transform (DWT) based algorithm is used for the feature extraction. Selecting specific wavelet decomposition levels reduces the dimensionality of the feature vector without compromising from the accuracy. Each classification loop takes a 300ms signal segment as the input for real time application concerns. A support vector machine (SVM) is used for the classification. Overall, the system is capable of identifying several facial gestures in real-time, which makes the system suitable for several applications such as monitoring sleep, stress, headaches and other orthodontic conditions.