A Security System for Door Opened by Using Brain Signal

A Brain computer interface (BCI) has introduced a new zone for inventive applications for researchers and designers offering various channels of communication to those humans who are suffering of motor disabilities. Most of existing secure door opened systems are based on sensor and alarms. A possible solution is the use brain computer interface system (BCI). In addition, it is important to study the BCI framework. In this research, a security system for door opened by using brain signal framework developed based on time-frequency techniques is proposed. The time-frequency method for feature extraction namely empirical mode decomposition (EMD) was combined separately with different classification methods such as $K$ nearest neighbours (KNN), liner discriminate analysis (LDA) and Support vector machine (SVM). To validate this framework, experimentally recorded data from nine subjects were collected. The feature extraction-based method EMD with classification SVM obtained average accuracy of 84%. The proposed framework based on SVM is more precise resulting in the improvement of accuracy, compared to other classification methods.