“Field surveillance of fuel dispensers using IoT-based metering and blockchains”

From Navigators

(Difference between revisions)
Jump to: navigation, search
(Created page with "{{Publication |type=inproceedings |title=Field surveillance of fuel dispensers using IoT-based metering and blockchains |author=Wilson S. Melo Jr, Luis V. G. Tarelho, Bruno A. Ro...")
 
Line 1: Line 1:
{{Publication
{{Publication
-
|type=inproceedings
+
|type=article
|title=Field surveillance of fuel dispensers using IoT-based metering and blockchains
|title=Field surveillance of fuel dispensers using IoT-based metering and blockchains
|author=Wilson S. Melo Jr, Luis V. G. Tarelho, Bruno A. Rodrigues Filho, Alysson Bessani, Luiz F. R. C. Carmo
|author=Wilson S. Melo Jr, Luis V. G. Tarelho, Bruno A. Rodrigues Filho, Alysson Bessani, Luiz F. R. C. Carmo

Latest revision as of 16:26, 29 September 2021

Wilson S. Melo Jr, Luis V. G. Tarelho, Bruno A. Rodrigues Filho, Alysson Bessani, Luiz F. R. C. Carmo

Journal of Network and Computer Applications, vol. 175, Feb. 2021.

Abstract: The field surveillance of fuel dispensers is an activity of Legal Metrology that checks these measuring instruments' correct behavior. However, it constitutes a complex challenge because malicious entities can tamper with fuel dispensers to get undue economic advantages. This paper proposes a distributed and decentralized solution. We use IoT-based vehicle simple meters to estimate the fuel amount in refilling events. Although these estimates can be inaccurate, we explore properties of the Law of the Large Numbers to evaluate the fuel dispenser's accuracy. We also use blockchains to avoid collusion attacks and provide a truly distributed and decentralized surveillance solution that implements statistical surveillance analysis as smart contracts. We develop a case study based on the vehicular fleet and fuel dispensers in São Paulo, Brazil. We perform our experiment using the Hyperledger Fabric platform with Byzantine fault-tolerant consensus. In a hypothetical scenario where vehicular meters present error rates below 5%, and each vehicle refuels more than ten times on average, we can identify tampered fuel dispensers with sensitivity and specificity over 95%. We also demonstrate that our blockchain deployment can support a workload of 600 concurrent clients with a throughput higher than 350 tps and latency lower than 1 s. These results attest to our framework suitability in terms of accuracy and performance. They provide promising perspectives on using our idea in the metrological surveillance of other measuring instruments.

Download paper

Download Field surveillance of fuel dispensers using IoT-based metering and blockchains

Export citation

BibTeX

Project(s):

Research line(s): Fault and Intrusion Tolerance in Open Distributed Systems (FIT)

Personal tools
Navigators toolbox