Advanced Metering Infrastructures: Security Risks and Mitigation
This tackles security vulnerabilities in smart meter deployments for energy providers and consumers, but it appears incremental as it builds on existing intrusion detection approaches.
The paper addresses security and privacy risks in Advanced Metering Infrastructures (AMI) by proposing a novel Machine Learning Intrusion Prevention System (IPS) that uses graphical security models to handle zero-day attacks.
Energy providers are moving to the smart meter era, encouraging consumers to install, free of charge, these devices in their homes, automating consumption readings submission and making consumers life easier. However, the increased deployment of such smart devices brings a lot of security and privacy risks. In order to overcome such risks, Intrusion Detection Systems are presented as pertinent tools that can provide network-level protection for smart devices deployed in home environments. In this context, this paper is exploring the problems of Advanced Metering Infrastructures (AMI) and proposing a novel Machine Learning (ML) Intrusion Prevention System (IPS) to get optimal decisions based on a variety of factors and graphical security models able to tackle zero-day attacks.