Professional Documents
Culture Documents
1 s2.0 S2665917424000114 Main
1 s2.0 S2665917424000114 Main
1 s2.0 S2665917424000114 Main
Measurement: Sensors
journal homepage: www.sciencedirect.com/journal/measurement-sensors
A R T I C L E I N F O A B S T R A C T
Keywords: Using IoT and cloud-based resources that have high-speed storage and processors with maximum bandwidth is
Cybersecurity obligatory. Live interaction between heterogeneous resources is supported physically and virtually through a
Feed forward variety of modes, which allows heterogeneous resources to interact in real time. These connected devices are
Encasement-based parameters
often mentioned as pervasive connectivity that can be accessed public often. There are several interactions such
Internet of things
Pervasive connectivity
as sensors, billing operations and much more services that are highlighted for Internet of Things (IoT) based
communication. Existing IoT environments are application oriented data that are mostly sensitive which are
ubiquitous collective from various IoT devices. Data classifications are implemented in the surrounding area to
facilitate various decisions. The Internet is one of the primary needs in everyone’s workplace where multiple
advanced handheld devices or laptops are mostly accessed along with the internet for all aspects. Machine
learning does the data sharing based on the growth of information collected where some of the intrusion and
cyber attacks from IoT devices were not comparatively good in error detection or accurate classifications. Using
the proposed algorithm Xception based Feedforward Encasement (XBFE) based Parameters for Cybersecurity
along with IoT Devices where cyclic communication among hidden layers that can focus on unsupervised
monitoring such that the feature mapping and scaling can filter along with maximum or minimum usage. The
main research idea is to use the UNSW-NB15 dataset to analyze cyber attacks from 49 features to enhance the
result.
1. Introduction systems could not possibly identify the detection of attacks. Moreover
performance was not effective as well as cost was highly increasing for
Sectors which are not technically improved are deployed with IoT the number of parameters for false rate alarm. In the traditional methods
devices and most of them are included in areas of agriculture [1]. The the scalable and reliable results were tested for large internet protocols
cost of these devices are lesser and the capability of computing is also that are gathered from a huge network.
limited. These devices have several constraints which makes the man
ufacturers to perform the design process in such a way to handle them.
Features providing appropriate security is necessary for making these 1.1. Applications of IoT
devices to function effectively. While Cyber attacks and intrusion
detection were technically handled by a number of classification algo The devices are supported even when the resources are limited and
rithms using machine learning [2], single perception from existing they provide huge support to other technologies. Communication
among devices is done in an automatic way here and their applications
* Corresponding author.
E-mail addresses: rvinothrs@gmail.com (R. Sudharsanan), mra.it@rmkec.ac.in (M. Rekha), prithabe28@gmail.com (N. Pritha), barathganagandhi@gmail.com
(G. Ganapathy), anrashoni2@gmail.com (G. Arokia Nerling Rasoni), uthayakumargs@stjosephstechnology.ac.in (G.S. Uthayakumar).
https://doi.org/10.1016/j.measen.2024.101035
Received 2 March 2023; Received in revised form 3 July 2023; Accepted 15 January 2024
Available online 8 February 2024
2665-9174/© 2024 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
R. Sudharsanan et al. Measurement: Sensors 32 (2024) 101035
2
R. Sudharsanan et al. Measurement: Sensors 32 (2024) 101035
3
R. Sudharsanan et al. Measurement: Sensors 32 (2024) 101035
Table 2
Proposed model summary for training and testing.
Proposed Model Summary
4
R. Sudharsanan et al. Measurement: Sensors 32 (2024) 101035
are functional mainly due to usage of the Internet. Tremendous amount [2] L. Tseng, L. Wong, S. Otoum, M. Aloqaily, J.B. Othman, Blockchain for managing
heterogeneous Internet of Things: a perspective architecture, IEEE Netw 34 (1)
of data generated is shared by usage of machine learning methods which
(Jan. 2020) 16–23.
help in detection of errors as well as classification in an accurate [3] I. Al Ridhawi, S. Otoum, M. Aloqaily, Y. Jararweh, T. Baker, Providing secure and
manner. Xception based Feedforward Encasement (XBFE) algorithm is reliable communication for next generation networks in smart cities, Sustain. Cities
employed for monitoring of devices additional to feature mapping as Soc. 56 (May 2020).
[4] J. Canedo, A. Skjellum, Using machine learning to secure IoT systems, in: Proc.
well as filter scaling process. Dataset named UNSW-NB 15 is incorpo 14th Annu. Conf. Privacy Secur. Trust (PST), Dec. 2016, pp. 219–222.
rated for analyzing the cyber attacks based on the feature selected. The [5] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, M. Ayyash, Internet of
experimental results denote the proposed system efficiency based on the Things: a survey on enabling technologies, protocols and applications, IEEE
Commun. Surveys Tuts. 17 (4) (2015) 2347–2376, 4th Quart.
enhanced results obtained. [6] A. Ferdowsi, W. Saad, Generative Adversarial Networks for Distributed Intrusion
Detection in the Internet of Things, 2019.
Declaration of competing interest [7] G. Caminero, M. Lopez-Martin, B. Carro, Adversarial environment reinforcement
learning algorithm for intrusion detection, Comput. Network. 159 (Aug. 2019)
96–109.
The authors declare that they have no known competing financial [8] S. Chen, et al., Automated poisoning attacks and defenses in malware detection
interests or personal relationships that could have appeared to influence systems: an adversarial machine learning approach, Comput. Secur. 73 (Mar. 2018)
326–344.
the work reported in this paper. [9] M. Moh, R. Raju, Machine learning techniques for security of internet of Things
(IoT) and fog computing systems, in: 2018 International Conference on High
Data availability Performance Computing & Simulation (HPCS), 2018, pp. 709–715. Orleans,
France.
[10] M. Bagaa, T. Taleb, J.B. Bernabe, A. Skarmeta, A machine learning security
No data was used for the research described in the article. framework for iot systems, IEEE Access 8 (2020) 114066–114077.
[11] J. Cañedo, A. Skjellum, Using machine learning to secure IoT systems, in: 2016
References 14th Annual Conference on Privacy, Security and Trust (PST), 2016, pp. 219–222.
Auckland, New Zealand.
[12] F. Hussain, R. Hussain, S.A. Hassan, E. Hossain, Machine learning in IoT security:
[1] S. Zafar, S. Jangsher, O. Bouachir, M. Aloqaily, J.B. Othman, QoS enhancement current solutions and future challenges, IEEE Communications Surveys & Tutorials
with deep learning-based interference prediction in mobile IoT, Comput. Commun. 22 (3) (thirdquarter 2020) 1686–1721.
148 (Dec. 2019) 86–97.