Classification of Motor Imaginary Signals for Machine Commmunication - A Novel Approach for Brain Machine Interface Design

A great interest for brain machine interface (BMI) is in the arising nowadays, instigated by several promising scientific and technological outcomes. The methods of measuring, processing and classifying brain activities, in order to interpret neuronal signals into machine control is often regarded as a challenging possibility as the expected beneficiaries are patients affected by motor disorders and paralyses. Aiming at the usage of motor imaginary signals for machine control, the paper presents a BMI architecture with the combination of OPC (OLE process control), a standard communication protocol for the automation and machine control industry. We also analyzed the principle of real-time communication between simulated actuators and MATLAB which was used for signal processing and. Adaptive recursive bandpass filters and SFAM were used for the signals feature extraction and classifications respectively, whereas the communication with simulated actuators were realized with the usage of the Matricon OPC Server and MATLAB OPC Client. Test results also indicate the feasibility and effectiveness of implementing the proposed method in real time applications with reference of the control response time.