Preprocessing and Feature Extraction for Asynchronous Multi-Class Noninvasive Brain Computer Interface based on EEG signal

  • Hamzah AlZubi

Student thesis: Master's Thesis


A Brain Computer Interface (BCI) is a system which allows direct communication the brain and a computer. It can be used to allow paralyzed as well as healthy individuals to interact with and control the surrounding environment or to communicate simply by the conscious modulation of thought patterns. Although Noninvasive Electroencephalogram-based (EEG-based) BCI is showing a lot of promise, it is faced with a number of dif?cult challenges, especially from the perspective of signal processing, because the signals being observed by EEG are extremely weak and typically contain very high levels of noise. The asynchronous multiclass BCI problem is particular importance because it closely matches realistic operating conditions (as opposed to synchronous problems). However, it is a challenging problem and requires the development of appropriate machine learning and signal processing tools. This thesis develops an asynchronous multiclass noninvasive EEG-based BCI system which is based on a novel feature extraction method for asynchronous BCI. A set of prepossessing, feature extraction and classi?cation methods were implemented to help support and validate the proposed system and to benchmark it against the latest developed systems in BCI literature. The proposed system is tested on two separate datasets. The ?rst is a well-known and publicly available dataset, while the second dataset was collected locally using a retail EEG kit. The developed system showed robust and accurate classi?cation results for both datasets.
Date of Award2011
Original languageAmerican English
SupervisorWei Lee Woon (Supervisor)


  • Signal Processing
  • User-Computer Interface

Cite this