Efficient Privacy-Preserving Electricity Theft Detection with Dynamic Billing and Load Monitoring for AMI Networks

In advanced metering infrastructure (AMI), smart meters (SMs) are installed\nat the consumer side to send fine-grained power consumption readings\nperiodically to the system operator (SO) for load monitoring, energy\nmanagement, billing, etc. However, fraudulent consumers launch electricity\ntheft cyber-attacks by reporting false readings to reduce their bills\nillegally. These attacks do not only cause financial losses but may also\ndegrade the grid performance because the readings are used for grid management.\nTo identify these attackers, the existing schemes employ machine-learning\nmodels using the consumers' fine-grained readings, which violates the\nconsumers' privacy by revealing their lifestyle. In this paper, we propose an\nefficient scheme that enables the SO to detect electricity theft, compute\nbills, and monitor load while preserving the consumers' privacy. The idea is\nthat SMs encrypt their readings using functional encryption, and the SO uses\nthe ciphertexts to (i) compute the bills following dynamic pricing approach,\n(ii) monitor the grid load, and (iii) evaluate a machine-learning model to\ndetect fraudulent consumers, without being able to learn the individual\nreadings to preserve consumers' privacy. We adapted a functional encryption\nscheme so that the encrypted readings are aggregated for billing and load\nmonitoring and only the aggregated value is revealed to the SO. Also, we\nexploited the inner-product operations on encrypted readings to evaluate a\nmachine-learning model to detect fraudulent consumers. Real dataset is used to\nevaluate our scheme, and our evaluations indicate that our scheme is secure and\ncan detect fraudulent consumers accurately with low communication and\ncomputation overhead.\n