A Device-Free Intelligent Alarm System Based on the Channel State Information

Security has always been a hot issue of concern. Most of today's home security systems are based on cameras or sensors. These systems have challenges, such as involving in users' privacy and high initial cost, so they are not highly popular. In order to resolve these challenges, a single commodity Wi-Fi device is utilized to design a device-free intelligent alarm system. The proposed system relies on the cooperation of active and passive alarm modules so that the user is not required to wear any sensor. In the active alarm module, a novel algorithm is proposed to identify the user's tiny help-seeking action (e.g.special breath) using channel state information (CSI). Moreover, the passive alarm module employs the CSI-based prospect detection to identify the physical attack behavior. The proposed system uses the invisible Wi-Fi signal to identify the people's motion. Then, it uses the results of the activity recognition to identify the dangerous event, which has the characteristic of the concealment. A large number of experiments are carried out in the present study to investigate the accuracy and robustness of the proposed system. It is found that the average accuracy of the system identification is over 95%.