Effective Anomaly Detection in Smart Home by Integrating Event Time Intervals

Smart home IoT systems and devices are susceptible to attacks malfunctions. As a result, users’ concerns about their security safety issues arise along with the prevalence of smart deployments. In home, various anomalies (such as fire or flooding) could happen due cyber attacks, device malfunctions, human mistakes. These motivate researchers propose anomaly detection approaches. Existing works on focus checking sequence devices’ events but leave out temporal information events. This limitation prevents them from detecting that cause delay rather than missing/injecting To fill this gap, in paper, we novel method takes inter-event intervals into consideration. We an innovative metric quantify similarity between two event sequences. design mechanism for learning patterns sequences common daily activities. Delay-caused detected by comparing learned patterns. collect real-world testbed training testing. The experiment results show our proposed achieves accuracies 93%, 88%, 89% three