I Can See Your Brain: Investigating Home-Use Electroencephalography System Security

摘要: Health-related Internet of Things (IoT) devices are becoming more popular in recent years. On the one hand, users can access information their health conditions conveniently; on other they exposed to new security risks. In this paper, we presented, best our knowledge, first in-depth analysis home-use electroencephalography (EEG) IoT devices. Our key contributions twofold. First, reverse-engineered EEG system framework via which identified design and implementation flaws. By exploiting these flaws, developed two sets novel easy-to-exploit PoC attacks, consist four remote attacks proximate attack. a attack, an attacker steal user’s brain wave data through carefully crafted program while victim’s over-the-air without accessing device any sense when he is close victim. As result, all 156 brain–computer interface (BCI) apps NeuroSky App store vulnerable We also discovered that 31 free at least Second, proposed deep learning model joint recurrent convolutional neural network (RCNN) infer activities based reduced-featured stolen from devices, evaluation over real-world indicates inference accuracy RCNN reach 70.55%.