EXTENDED ANALYSIS USING NIST METHODOLOGY OF SENSOR DATA ENTROPY

IoT security is an unresolved issue at this time, which is not easy to achieve due to the multitude of types of sensors used. As almost any security solution involves cryptography elements, which by default requires random numbers generators, we aimed to analyze the possibility to extract entropy fr...

Full description

Saved in:
Bibliographic Details
Published inScientific Bulletin. Series C, Electrical Engineering and Computer Science no. 2; p. 121
Main Authors Răstoceanu, Florin, Ciubotaru, Bogdan-Iulian, Rădoi, Ionuț, Marian, Constantin Viorel
Format Journal Article
LanguageEnglish
Published Bucharest University Polytechnica of Bucharest 01.01.2021
Subjects
Cryptography
Entropy
Random numbers
Security
Sensors
Online AccessGet full text

Cover

More Information
Summary:IoT security is an unresolved issue at this time, which is not easy to achieve due to the multitude of types of sensors used. As almost any security solution involves cryptography elements, which by default requires random numbers generators, we aimed to analyze the possibility to extract entropy from different sensors data. The method proposed in this paper minimizes the effects of the diversity of different devices used and optimizes the entropy level according to consumed energy. The proposed solution was validated using the NIST methodology, which involves 10 different estimators and a large dataset.
ISSN:2286-3540