Smart System for Detecting Unauthorized Entry into a Smart Home

The article covers the matter of the risk assessment of unauthorized entry into a smart home. Thereby the authors model and compare two approaches to the issue considered. The first technique actualizes a fuzzy system for the level detecting intrusion danger and warns about an unauthorized entry. Bu...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2020 International Conference Quality Management, Transport and Information Security, Information Technologies (IT&QM&IS) s. 63 - 67
Hlavní autori: Sedova, Nelly A., Arkhipova, Zoya V., Lavrov, Evgeniy A., Ledovskikh, Irina A., Bazhenov, Ruslan I.
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 07.09.2020
Predmet:
adaptive neuro-fuzzy system
detecting unauthorized entry
Face recognition
Linguistics
Sensors
smart home
Smart homes
smart system
Temperature measurement
Temperature sensors
Vibrations
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:The article covers the matter of the risk assessment of unauthorized entry into a smart home. Thereby the authors model and compare two approaches to the issue considered. The first technique actualizes a fuzzy system for the level detecting intrusion danger and warns about an unauthorized entry. But the second technique deals with a neuro-fuzzy system construction for assessing the level of unauthorized entry to the smart home. The authors present a structure and methods of training fuzzy and neuro-fuzzy systems to evaluate the level of unauthorized entry to the smart home. They carry out a series of experiments on the models of the designed systems. According to the results of comparative analysis, the authors recognize the application feasibility of a neuro-fuzzy system to assess the level of unauthorized entry to a smart home with a fuzzy logic inference algorithm based on Takagi Sugeno fuzzy model. Besides, the authors run a neuro-fuzzy system computer modelling/simulation to identify its best values. In particular, they declare a list of the best membership functions for generating input linguistic variables. They also propose a hybrid learning technique and the method of linear coefficients for developing terms of the output variable/signal.
DOI:10.1109/ITQMIS51053.2020.9322947