A method to detect Internet of Things botnets

Submitted by aekwall on Mon, 05/20/2019 - 9:00am

Title	A method to detect Internet of Things botnets
Publication Type	Conference Paper
Year of Publication	2018
Authors	Prokofiev, A. O., Smirnova, Y. S., Surov, V. A.
Conference Name	2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)
Date Published	jan
Publisher	IEEE
ISBN Number	978-1-5386-4340-2
Keywords	authorisation, Botnet, Botnet detection, botnets, common characteristics, composability, computer network security, cybersecurity, Human Behavior, Internet of Things, invasive software, IoT, IoT devices, IoT Security 2018, logistic regression, Logistics, main security problems, Malware, Metrics, performance evaluation, Protocols, pubcrawl, Resiliency, security, security of data, unauthorized access, widespread botnets
Abstract	The main security problems, typical for the Internet of Things (IoT), as well as the purpose of gaining unauthorized access to the IoT, are considered in this paper. Common characteristics of the most widespread botnets are provided. A method to detect compromised IoT devices included into a botnet is proposed. The method is based on a model of logistic regression. The article describes a developed model of logistic regression which allows to estimate the probability that a device initiating a connection is running a bot. A list of network protocols, used to gain unauthorized access to a device and to receive instructions from common and control (C&C) server, is provided too.
URL	https://ieeexplore.ieee.org/document/8317041
DOI	10.1109/EIConRus.2018.8317041
Citation Key	prokofiev_method_2018

Groups:

Science of Security VO