Improving P300 Speller for Brain-Computer Interface by Using the Best Possible Classifier
سال انتشار: 1398
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 491
فایل این مقاله در 7 صفحه با فرمت PDF قابل دریافت می باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
این مقاله در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
ETECH04_064
تاریخ نمایه سازی: 27 بهمن 1398
چکیده مقاله:
In this paper, we discuss Brain-computer computer interface (BCI) and its problem of data transfer. The main object is to examine the classification of EEG signals whichare divided into two target and non-target classes and to improve the speed of BCI communication. Data is collected by the p300 speller, then a preprocessing treatment is performed on the recorded data. Since the EEG signals are low amplitude recording and averaging steps should be acted several times. Therefore, the BCI communication speed is slow. A new channel selection of electroencephalography is presented due to speed of communication. Due to the extraction of the P300 and distinction between the signal with P300 and signal without p300, we use the feature extraction. Then we classify signals using different classifiers such as RBF, SVM and ANN. Comparing analysisshows that SVM gives the best possible result among examining classifier for the BCI communication.
نویسندگان
Majid Ramedani
Faculty of Electrical Engineering Tarbiat Modares University Tehran, Iran
Behzad Farzanegan
Faculty of Electrical Engineering Amirkabir University of Technology Tehran, Iran
Mohammad Bagher Menhaj
Faculty of Electrical Engineering Amirkabir University of Technology Tehran, Iran