Anomaly-based cyberattacks detection for smart homes: A systematic literature review

Abstract

Smart homes, leveraging IoT technology to interconnect various devices and appliances to the internet, enable remote monitoring, automation, and control. However, collecting sensitive personal and business data assets renders smart homes a target for cyberattacks. Anomaly detection is a promising approach for identifying malicious behavior in smart homes. Yet, the current literature primarily discusses IoT-related cyberattacks and gives limited attention to detecting anomalies specific to the smart home context. Furthermore, there is a lack of datasets that accurately represent the complexity inherent in a smart home environment in terms of users with varying levels of expertise and diverse, evolving types of devices. Therefore, this paper presents a systematic literature review (SLR) that focuses on using anomaly detection to identify cyberattacks in smart home environments. The SLR includes an adapted taxonomy that classifies existing anomaly detection methods and a critical analysis of the current state of knowledge and future research challenges. Our findings show a growing interest in detecting cyberattacks with anomaly-based models in smart homes using centralized and network-based features. Ensemble and deep learning techniques are popular methods for detecting these anomalies. However, the limited diversity of cyberattacks in existing datasets and the absence of comprehensive datasets representing the complexity of smart home environments call for further research to improve the generalizability of detection models.