Tell Me How You Re-Charge, I Will Tell You Where You Drove To: Electric Vehicles Profiling Based on Charging-Current Demand

Brighente, Alessandro; Conti, Mauro; Sadaf, Izza

doi:10.1007/978-3-030-88418-5_31

Cited by 7 publications

(36 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We then investigate the battery degradation over time and its impact on EVScout 2.0 performances. Lastly, we show the superiority of EVScout 2.0, comparing its performance with those in [4], both on the old and new datasets.…”

Section: Introductionmentioning

confidence: 99%

“…However, the latter communication process (i.e., EV to EVSE) is not secure, as signals are exchanged without encryption or aggregation techniques. The signals exchanged during the charging process can hence be exploited as a sidechannel to extract information peculiar to each EV, allowing hence for their profiling and successive recognition [4]. This represents a threat towards users' privacy, as the connection of their EV to an EVSE monitored by a malicious user may lead to tracking their movement as well as information regarding their driving behavior.…”

Section: Introductionmentioning

confidence: 99%

“…In this paper, we propose EVScout2.0, a framework extending EVScout, the work proposed in [4], where we initially showed the feasibility of profiling EVs based on the current exchanged during the charging process. In particular, we extend the proposed framework by developing an enhanced feature extractor, which allows for higher classification scores compared to our previous work.…”

Section: Introductionmentioning

confidence: 99%

“…We perform a thorough evaluation of EVScout2.0, showing its profiling performance considering different training set sizes, as well as different unbalancing in the training-testing datasets. Compared to the work in [4], we provide a more comprehensive analysis of the performance of the attack. In particular, we first show the performance of the novel feature extractor and investigate the performance of EVScout2.0 for a varying number of features.…”

Section: Introductionmentioning

confidence: 99%

“…We then compare the performance of the different classifiers that can be exploited by EVScout2.0. Compared to [4] we also extend the number of classifiers and provide an in-depth description of the choice of their hyper-parameters. We then investigate the dependencies of EVScout2.0 on the number of training TSs needed for classifying EVs with sufficient confidence.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

EVScout2.0: Electric Vehicle Profiling Through Charging Profile

Brighente¹,

Conti²,

Denis³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

EVs (Electric Vehicles) represent a green alternative to traditional fuel-powered vehicles. To enforce their widespread use, both the technical development and the security of users shall be guaranteed. Privacy of users represents one of the possible threats impairing EVs adoption. In particular, recent works showed the feasibility of identifying EVs based on the current exchanged during the charging phase. In fact, while the resource negotiation phase runs over secure communication protocols, the signal exchanged during the actual charging contains features peculiar to each EV. A suitable feature extractor can hence associate such features to each EV, in what is commonly known as profiling.In this paper, we propose EVScout2.0, an extended and improved version of our previously proposed framework to profile EVs based on their charging behavior. By exploiting the current and pilot signals exchanged during the charging phase, our scheme is able to extract features peculiar for each EV, allowing hence for their profiling. We implemented and tested EVScout2.0 over a set of real-world measurements considering over 7500 charging sessions from a total of 137 EVs. In particular, numerical results show the superiority of EVScout2.0 with respect to the previous version. EVScout2.0 can profile EVs, attaining a maximum of 0.88 recall and 0.88 precision. To the best of the authors' knowledge, these results set a new benchmark for upcoming privacy research for large datasets of EVs.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

EVScout2.0: Electric Vehicle Profiling Through Charging Profile

Brighente¹,

Conti²,

Denis³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Decentralized Review and Attestation of Software Attribute Claims

2022

View full text Add to dashboard Cite

Software can be described, like human users and other objects, through attributes. For this work, we define software attributes as humanly verifiable, falsifiable, or judgeable statements regarding characteristics of said software. Much like attributes in general, software attributes require robust identities for their source but also for their target, meaning a software in general or a binary in particular. As software can be of critical importance, performing an independent review of attribute claims appears beneficial. We posit that decentralized platforms that were developed and refined over the past decade can bridge the gap between existing tools and methods for software review and their open, transparent, and accountable use for the benefit of users. In this work, we explore the feasibility and implications of decentralizing an independent review of software attribute claims. We envision the decentralization of a review process from initialization and execution to the persistent recording of results. We sketch the available design space by decomposing the overall process into a modular design and describe how each component covers overarching objectives. To illustrate practical implications and trade-offs, we present ETHDPR, a proof of concept implementation based on Ethereum and IPFS. Through a quantitative and qualitative evaluation, we show that a decentralized software review is practically feasible. We illustrate the flexibility of the proposed approach using a toy example of a software component in automotive systems. Lastly, we provide a discussion on fundamental limits and open issues of facilitating independent reviews via technological means.

show abstract