Preventing Cyberattacks on Smart Home Devices Using Machine Learning and Edge Computing

It seems that the Internet of Things (IoT) has revolutionized the way of life in today's world and, more so, in the context of smart homes.On the contrary, IoT technology comes with major cybersecurity challenges.The sheer volume of smart home devices raises the risk of being targeted by various cyber offenders through the possibility of unauthorized intrusion, information leaks, or false data injection.Such attacks are usually complex and evolve according to the environment.Thus, most security solutions, such as Intrusion Detection Systems (IDS), must be up to par.This paper looks into recent advances in Artificial Intelligence (AI) and Machine Learning (ML) on smart home security.In particular, it looks for advanced frameworks that will utilize deep learning and edge technologies to strengthen intrusion detection and response in real-time.The paper recommend use of AI4SAFE-IoT Architecture techniques, and more recently federated learning as potential candidates for improving security and privacy of the data.Further, in some cases, because the edge computation is present, communication latency is decreased, allowing for swift threat detection and responsive action to be taken.Hence such solution types meets the increasing requirements for novel and robust security architectures for smart homes as compared to other traditional methods.