Attack and anomaly detection in IoT sensors in IoT sites using machine learning approaches

Hasan, Mahmudul; Islam, Md. Milon; Zarif, Md Ishrak Islam; Hashem, M. M. A.

doi:10.1016/j.iot.2019.100059

Cited by 619 publications

(298 citation statements)

References 10 publications

Supporting

Mentioning

295

Contrasting

Unclassified

Order By: Relevance

“…Figure 4 shows the performance of recall and f-measure over proposed method. [15] 99.00 99.00 LR [15] 98.00 98.00 ANN [15] 99.00 99.00 SVM [15] 98.00 98.00 RF [15] 99.00 99.00 sProposed ECC-MAC-MQTT 99.64 99.45 From these results, it can be seen that proposed ECC-MAC-MQTT protocol is secure against different security attacks.…”

Section: Precision Of Proposed Ecc-mac-mqtt Methodsmentioning

confidence: 89%

“…DT [15] 99.40 99.02 LR [15] 98.30 98.01 ANN [15] 99.40 99.04 SVM [15] 98.20 98.21 RF [15] 99.40 99.0 Ensemble [17] 99.54 Not available NB [17] 91. 2.…”

Section: Methodsologymentioning

confidence: 99%

“…To strengthen the application layer, MQTT messages like PUBREC and PUBLISH, but this method focused on the features of Connect and ConnectAck of CONNECT and CONNACK messages. M. Hasan, et al, [15] predicted the anomalies and attacks on the IoT networks by using various machine learning algorithms such as Decision Tree (DT), logistic regression (LR), Artificial Neural Network (ANN), Random Forest (RF) and Support Vector Machine (SVM). The experiments were carried out on collected micro-services that was communicated using MQTT protocol.…”

Section: Literature Reviewmentioning

confidence: 99%

“…In this section, the quantitative analysis of proposed method is carried out with existing techniques namely fuzzy based Secure-MQTT [14], several machine learning algorithms [15], ensemble intrusion detection techniques (i.e. NB, ensemble) [17] and ensemble learning algorithm [18] in terms of metrics such as accuracy, F-measure, recall, FPR and precision.…”

Section: Performance Evaluation Of Proposed Ecc-mac-mqttmentioning

confidence: 99%

See 3 more Smart Citations

Authentication and Access control Mechanism using Elliptical Curve Cryptography with Message Authentication Code

Fathima

2019

IJEAT

View full text Add to dashboard Cite

Nowadays, Internet of Things (IoT) provided smart and inter-connected devices, which are ranging from tiny sensors to complex nodes of Fog and Cloud, communication protocols and several networking technologies. IoT plays a significant role in various application areas such as e-commerce, healthcare, research, governance and so on, and these application easily accessible for end users. Although the IoT has many benefits, there are some security challenges associated with the vulnerabilities presents in this area. This research work introduced a new authentication and access control model using Message Authentication code (MAC) with Elliptical Curve Cryptography (ECC) to address the security issues in Message Queuing Telemetry Transport (MQTT). In cryptography, a message can be authenticated by MAC (i.e. to verify that whether the message came from authenticate sender and the content of the message is changed or not). If any changes occur to the message content, the verifier can able to detect those changes using MAC values, which preserves the authenticity as well as data integrity. The validated results stated that the proposed ECC-MAC-MQTT shows better performance when compared with existing techniques in terms of accuracy, precision, recall and f-measure.

show abstract

Section: Precision Of Proposed Ecc-mac-mqtt Methodsmentioning

confidence: 89%

“…DT [15] 99.40 99.02 LR [15] 98.30 98.01 ANN [15] 99.40 99.04 SVM [15] 98.20 98.21 RF [15] 99.40 99.0 Ensemble [17] 99.54 Not available NB [17] 91. 2.…”

Section: Methodsologymentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Performance Evaluation Of Proposed Ecc-mac-mqttmentioning

confidence: 99%

See 2 more Smart Citations

Authentication and Access control Mechanism using Elliptical Curve Cryptography with Message Authentication Code

Fathima

2019

IJEAT

View full text Add to dashboard Cite

show abstract

“…For example a 97.4% of identification success for real traffic data has been obtained by D'Angelo et al [18] using U-BRAIN [19]. In the specific field of IoT devices anomaly and attacks detection Hasan et al obtained up to 99.4% of identification success using Decision Tree, Random Forest, and Artificial Neural Networks [20].…”

Section: Introductionmentioning

confidence: 99%

Towards Near-Real-Time Intrusion Detection for IoT Devices using Supervised Learning and Apache Spark

Morfino

Rampone

2020

Electronics

View full text Add to dashboard Cite

In the fields of Internet of Things (IoT) infrastructures, attack and anomaly detection are rising concerns. With the increased use of IoT infrastructure in every domain, threats and attacks in these infrastructures are also growing proportionally. In this paper the performances of several machine learning algorithms in identifying cyber-attacks (namely SYN-DOS attacks) to IoT systems are compared both in terms of application performances, and in training/application times. We use supervised machine learning algorithms included in the MLlib library of Apache Spark, a fast and general engine for big data processing. We show the implementation details and the performance of those algorithms on public datasets using a training set of up to 2 million instances. We adopt a Cloud environment, emphasizing the importance of the scalability and of the elasticity of use. Results show that all the Spark algorithms used result in a very good identification accuracy (>99%). Overall, one of them, Random Forest, achieves an accuracy of 1. We also report a very short training time (23.22 sec for Decision Tree with 2 million rows). The experiments also show a very low application time (0.13 sec for over than 600,000 instances for Random Forest) using Apache Spark in the Cloud. Furthermore, the explicit model generated by Random Forest is very easy-to-implement using high- or low-level programming languages. In light of the results obtained, both in terms of computation times and identification performance, a hybrid approach for the detection of SYN-DOS cyber-attacks on IoT devices is proposed: the application of an explicit Random Forest model, implemented directly on the IoT device, along with a second level analysis (training) performed in the Cloud.

show abstract

Machine Learning Techniques for IoT Data Analytics

Afshan

Rout

2021

Big Data Analytics for Internet of Things

View full text Add to dashboard Cite

Attack and anomaly detection in IoT sensors in IoT sites using machine learning approaches

Cited by 619 publications

References 10 publications

Authentication and Access control Mechanism using Elliptical Curve Cryptography with Message Authentication Code

Authentication and Access control Mechanism using Elliptical Curve Cryptography with Message Authentication Code

Towards Near-Real-Time Intrusion Detection for IoT Devices using Supervised Learning and Apache Spark

Machine Learning Techniques for IoT Data Analytics

Contact Info

Product

Resources

About