Machine Learning Enhanced Entropy-Based Network Anomaly Detection

Timčenko, Valentina; Gajin, Slavko

doi:10.4316/aece.2021.04006

Cited by 13 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several significant research works [4], [5], [6], [7], [8] have been proposed for IoT anomaly detection, with and without SDN. For instance, Del-IoT [9] is an IoT anomaly detection approach that employs a deep ensemble model to address data imbalance issues in network traffic datasets.…”

Section: Related Workmentioning

confidence: 99%

In-network DDoS detection and mitigation using INT data for IoT ecosystem

Altangerel,

Tejfel

2023

Infocommunications journal

View full text Add to dashboard Cite

Due to the limited capabilities and diversity of Internet of Things (IoT) devices, it is challenging to implement robust and unified security standards for these devices. Additionally, the fact that vulnerable IoT devices are beyond the network’s control makes them susceptible to being compromised and used as bots or part of botnets, leading to a surge in attacks involving these devices in recent times. We proposed a real-time IoT anomaly detection and mitigation solution at the programmable data plane in a Software-Defined Networking (SDN) environment using Inband Network telemetry (INT) data to address this issue. As far as we know, it is the first experiment in which INT data is used to detect IoT attacks in the programmable data plane. Based on our performance evaluation, the detection delay of our proposed approach is much lower than the results of previous Distributed Denial-of-Service (DDoS) research, and the detection accuracy is similarly high.

show abstract

Section: Related Workmentioning

confidence: 99%

In-network DDoS detection and mitigation using INT data for IoT ecosystem

Altangerel,

Tejfel

2023

Infocommunications journal

View full text Add to dashboard Cite

show abstract

“…Different machine learning algorithms have been applied to anomaly detection in IoT systems and achieved positive outcomes [8], [9], [10], [11], [15], [16], [17], [19,20,21]. One of the most common approaches is the deep learning algorithm [8], [9].…”

Section: Related Workmentioning

confidence: 99%

An Intelligent Detection of Malicious Intrusions in IoT Based on Machine Learning and Deep Learning Techniques

Iftikhar¹,

Khan²,

Al-Madani³

et al. 2022

CSJM

View full text Add to dashboard Cite

The devices of the Internet of Things (IoT) are facing various types of attacks, and IoT applications present unique and new protection challenges. These security challenges in IoT must be addressed to avoid any potential attacks. Malicious intrusions in IoT devices are considered one of the most aspects required for IoT users in modern applications. Machine learning techniques are widely used for intelligent detection of malicious intrusions in IoT. This paper proposes an intelligent detection method of malicious intrusions in IoT systems that leverages effective classification of benign and malicious attacks. An ensemble approach combined with various machine learning algorithms and a deep learning technique, is used to detect anomalies and other malicious activities in IoT. For the consideration of the detection of malicious intrusions and anomalies in IoT devices, UNSW-NB15 dataset is used as one of the latest IoT datasets. In this research, malicious and normal intrusions in IoT devices are classified with the use of various models. %Moreover, improved results are provided and compared with CorrAuc [1] for training accuracies, cross-validation accuracies, execution time, precision, recall and F1 score.

show abstract

“…Fractional fourier entropy is used to detect hyperspectral anomalies by Ran Tao et al [13], and Joshua Garland et al [14] apply permutation entropy to identify abnormal data records of paleoclimate. Huffman-multi-scale entropy is utilized to detect satellite momentum wheel anomalies [15], and Valentina et al [16] take advantage of machine learning enhanced entropy and effectively detect network anomalies. Different entropy shows excellent results within a certain range of their application [17,18].…”

Section: Introductionmentioning

confidence: 99%

“…And if the signal does not meet the Shannon entropy uncertainty condition of fractional Fourier transform, the anomaly will not be detected by fractional Fourier entropy [13]. Machine learning enhanced entropy depends on easier detection of the hidden traffic patterns [16]. Fundamentally, almost each type of entropy has its own limitations or defects [19,20].…”

Section: Introductionmentioning

confidence: 99%