Electroencephalography (EEG) signal based intent recognition has recentlyattracted much attention in both academia and industries, due to helping theelderly or motor-disabled people controlling smart devices to communicate withouter world. However, the utilization of EEG signals is challenged by lowaccuracy, arduous and time- consuming feature extraction. This paper proposes a7-layer deep learning model to classify raw EEG signals with the aim ofrecognizing subjects' intents, to avoid the time consumed in pre-processing andfeature extraction. The hyper-parameters are selected by an Orthogonal Arrayexperiment method for efficiency. Our model is applied to an open EEG datasetprovided by PhysioNet and achieves the accuracy of 0.9553 on the intentrecognition. The applicability of our proposed model is further demonstrated bytwo use cases of smart living (assisted living with robotics and homeautomation).
展开▼
