Valentin Tudor

Remote Control of Smart Meters: Friend or Foe?

The traditional electrical grid is transitioning into the smart grid. New equipment is being installed to simplify the process of monitoring and managing the grid, making the system more transparent to use but also introducing new security problems. Smart meters are replacing the traditional electrical utility meters, offering new functionalities such as remote reading, automatic error reporting, and the possibility for remote shutoff. This last feature is studied in this paper through two scenarios where the effects are outlined, both on a theoretical level and through a simulation. In the first scenario, the frequency property of the grid is the target to possibly cause a blackout. In the second scenario, the voltage is driven out of bounds by the adversary.

BES: Differentially Private and Distributed Event Aggregation in Advanced Metering Infrastructures

Significant challenges for online event aggregation in the context of Cyber-Physical Systems stem from the computational requirements of their distributed nature, as well as from their privacy concerns. In the context of the latter, differential privacy has gained popularity because of its strong privacy protection guarantees, holding against very powerful adversaries. Despite such strong guarantees, though, its adoption in real-world applications is limited by the privacy-preserving noise it introduces to the analysis, which might compromise its usefulness. We investigate the above problem from a system-perspective in the context of Advanced Metering Infrastructures, providing strong privacy guarantees together with useful results for event aggregation taking into account the distributed nature of such systems. We present a streaming-based framework, Bes, and propose methods to limit the noise introduced by differential privacy in real-world scenarios, thus reducing the resulting utility degradation, while still holding against the adversary model adhering with the original definition of differential privacy. We provide a thorough evaluation based on a fully implemented Bes prototype and conducted with real energy consumption data. We show how a large number of events can be aggregated in a private fashion with low processing latency by a single-board device, similar in performance to the devices deployed in Advanced Metering Infrastructures.

A study on data de-pseudonymization in the smart grid

In the transition to the smart grid, the electricity networks are becoming more data intensive with more data producing devices deployed, increasing both the opportunities and challenges in how the collected data are used. For example, in the Advanced Metering Infrastructure (AMI) the devices and their corresponding data give more information about the operational parameters of the environment but also details about the habits of the people living in the houses monitored by smart meters. Different anonymization techniques have been proposed to minimize privacy concerns, among them the use of pseudonyms. In this work we return to the question of the effectiveness of pseudonyms, by investigating how a previously reported methodology for de-pseudonymization performs given a more realistic and larger dataset than was previously used. We also propose and compare the results with our own simpler de-pseudonymization methodology. Our results indicate, not surprisingly, that large realistic datasets are very important to properly understand how an experimental method performs. Results based on small datasets run the risk of not being generalizable. In particular, we show that the number of re-identified households by breaking pseudonyms is dependent on the size of the dataset and the period where the pseudonyms are constant and not changed. In the setting of the smart grid, results will even vary based on the season when the dataset was captured. Knowing that relative simple changes in the data collection procedure may significantly increase the resistance to de-anonymization attacks will help future AMI deployments.

Harnessing the unknown in advanced metering infrastructure traffic

The Advanced Metering Infrastructure (AMI), a key component for smart grids, is expanding with more installed devices. Due to security and privacy concerns, the communication between these devices is encrypted, making it more secure against malicious third parties but also obscuring the ability of the network owner to detect any misbehaving user or equipment. We are investigating how to balance the need for confidentiality with the need to monitor the AMI. More specifically, we develop one important component for an AMI Intrusion Detection System (IDS), which can accurately determine the individual commands (but not their content) sent between AMI devices even when they are sent over an encrypted channel or in a protocol that the IDS cannot parse. We explain our methodology and propose features which summarize traffic characteristics. We conduct a feasibility study based on representative protocols in AMI and demonstrate the real utility of this IDS component. Our results are validated experimentally using two different datasets containing realistic traffic captured from two different AMI testbeds.