An efficient intrusion detection scheme for cluster based AMI networks

Advanced Metering Infrastructure; it is an imperative component of the smart grid and a responsible ingredient to collecting, measuring and analyzing energy usage data which is transmitted to concentrator and control centre on utility side. The security of advanced metering infrastructure is one of the most concerning issue in smart grid implementation. Advanced Metering Infrastructure exhibits different types of security threats - like the privacy violations; altering pricing values; injecting false readings; denial of service and other kinds of malicious attacks. Such attacks always put down performance of smart grid system along with fiscal loss; causing failures; suspension of operations and create distrust between users and utilities. In this paper, state estimation based clustered framework is proposed for advanced metering infrastructure that is composed of smart meters, cluster head and vice cluster head. A Cluster head is responsible to collect data of smart meters from cluster and forward it to control centre, where trust value is evaluated by using state estimation method. Based on trust values, cluster of smart meter is declared by cluster head while malicious meters are kicked out. Meanwhile, vice cluster head concept is used to monitor the communication behavior of cluster head. Theoretical analysis and simulation results show that the proposed security scheme could prevent and detect malicious meter with high probability and success.