Abstract:

Brain computer Interfacing (BCI) is an emerging topic used in a wide range of fields from gaming

machines to health aids. BCI technology aims to establish a direct communication path between

the user's brain and any electronic device. A major challenge in building the MI-BCI system is to

produce robust, informative and discriminatory features that can be converted into device

commands. Automotive imaging is a BCI method in which the user's imagination of a leg

movement is acquired without actual physical movement. Among the various BCI strategies,

automotive images are the most popular BCI operating system due to its functionality and being

an independent BCI system. Typically, the electroencephalogram (EEG) is used to detect motor

image signals as it is an effective, cheap, fast and non-invasive method of analyzing brain signals.

The objective of the project is to develop a method for classification of movement / imagery of a

motor imagery signals and to build the classifier model to get better performance. For this purpose,

existing methods and proposed methods is going to be introduced and their phased performance

will be analyzed. The CSP will be embedded in class distortion which may be a hindrance to the

EEG signal coordinating.