Abstract: An Internet-of-things (IoT)-based impact pattern analysis system for a smart security window includes: an ultra-small IoT device attached to a security window and detecting an impact; a station processing a signal obtained from the ultra-small IoT device and analyzing an impact pattern; and a user terminal outputting information related to the impact pattern through an interface. The station controller processes 3-axis acceleration data to extract a feature vector and determines an impact pattern by applying a learning model. According to the embodiment of the present disclosure, by automatically determining whether an intrusion occurs by analyzing impact patterns according to various conditions and circumstances, malfunction based on an impact other than an intrusion may be reduced and an unnecessary alarm may be prevented.

Abstract: An Internet-of-things (IoT)-based impact pattern analysis system for a smart security window includes: an ultra-small IoT device attached to a security window and detecting an impact; a station processing a signal obtained from the ultra-small IoT device and analyzing an impact pattern; and a user terminal outputting information related to the impact pattern through an interface. The station controller processes 3-axis acceleration data to extract a feature vector and determines an impact pattern by applying a learning model. According to the embodiment of the present disclosure, by automatically determining whether an intrusion occurs by analyzing impact patterns according to various conditions and circumstances, malfunction based on an impact other than an intrusion may be reduced and an unnecessary alarm may be prevented.