Selection of effective machine learning algorithm and Bot-IoT attacks traffic identification for internet of things in smart city

Shafiq, Muhammad; Tian, Zhihong; Du, Xiaojiang; Guizani, Mohsen

doi:10.1016/j.future.2020.02.017

Cited by 305 publications

(149 citation statements)

References 23 publications

Supporting

Mentioning

148

Contrasting

Unclassified

Order By: Relevance

“…In order to discretize the input time series x without losing the relative distance information, Buckman [3] proposed thermometer encodings. For an index j ∈ 1, .…”

Section: Proposed Methodsmentioning

confidence: 99%

“…e IoT consumer products are no longer just the product only; it is the data, the product, the infrastructure, and the algorithms. ese IoT products have switched to connected technologies from analog one, therefore, introducing novel risks for consumers regarding potential safety, privacy, and security issues for discriminatory data [2][3][4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An IoT Time Series Data Security Model for Adversarial Attack Based on Thermometer Encoding

Yang

Abbasi

Algarni

et al. 2021

Security and Communication Networks

View full text Add to dashboard Cite

Nowadays, an Internet of Things (IoT) device consists of algorithms, datasets, and models. Due to good performance of deep learning methods, many devices integrated well-trained models in them. IoT empowers users to communicate and control physical devices to achieve vital information. However, these models are vulnerable to adversarial attacks, which largely bring potential risks to the normal application of deep learning methods. For instance, very little changes even one point in the IoT time-series data could lead to unreliable or wrong decisions. Moreover, these changes could be deliberately generated by following an adversarial attack strategy. We propose a robust IoT data classification model based on an encode-decode joint training model. Furthermore, thermometer encoding is taken as a nonlinear transformation to the original training examples that are used to reconstruct original time series examples through the encode-decode model. The trained ResNet model based on reconstruction examples is more robust to the adversarial attack. Experiments show that the trained model can successfully resist to fast gradient sign method attack to some extent and improve the security of the time series data classification model.

show abstract

“…In order to discretize the input time series x without losing the relative distance information, Buckman [3] proposed thermometer encodings. For an index j ∈ 1, .…”

Section: Proposed Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An IoT Time Series Data Security Model for Adversarial Attack Based on Thermometer Encoding

Yang

Abbasi

Algarni

et al. 2021

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…The dataset, NSL-KDD, which was used in the experiments, is formally recognized for the traditional network attack detection system. The study [13] proposed a framework to select the effective machine learning algorithm for identifying malicious activities in IoT environments. They mentioned that naïve Bayes is an effective algorithm for the anomaly detection system in their experiments.…”

Section: Related Workmentioning

confidence: 99%

Machine Learning-Based IoT-Botnet Attack Detection with Sequential Architecture

Soe

Feng

Santosa

et al. 2020

Sensors

137

View full text Add to dashboard Cite

With the rapid development and popularization of Internet of Things (IoT) devices, an increasing number of cyber-attacks are targeting such devices. It was said that most of the attacks in IoT environments are botnet-based attacks. Many security weaknesses still exist on the IoT devices because most of them have not enough memory and computational resource for robust security mechanisms. Moreover, many existing rule-based detection systems can be circumvented by attackers. In this study, we proposed a machine learning (ML)-based botnet attack detection framework with sequential detection architecture. An efficient feature selection approach is adopted to implement a lightweight detection system with a high performance. The overall detection performance achieves around 99% for the botnet attack detection using three different ML algorithms, including artificial neural network (ANN), J48 decision tree, and Naïve Bayes. The experiment result indicates that the proposed architecture can effectively detect botnet-based attacks, and also can be extended with corresponding sub-engines for new kinds of attacks.

show abstract

“…Packets. ML algorithms are being in use for the identification of intrusion disturbing the organizations or its system [2,6,[40][41][42]. In this paper, a rough set approach is used to identify the attack on data packets.…”

Section: Rough Set Approach For Identification Of Attack On Datamentioning

confidence: 99%

Identification of Attack on Data Packets Using Rough Set Approach to Secure End to End Communication

Nazir

Mukhtar

2020

Complexity

View full text Add to dashboard Cite

Security has become one of the important factors for any network communication and transmission of data packets. An organization with an optimal security system can lead to a successful business and can earn huge profit on the business they are doing. Different network devices are linked to route, compute, monitor, and communicate various real-time developments. The hackers are trying to attack the network and want to draw the organization’s significant information for its own profits. During the communication, if an intrusion or eavesdropping occurs, it will lead to a severe disfigurement of the whole communication network, and the data will be controlled by wrong malicious users. Identification of attack is a way to identify the security violations and analyze the measures in a computer network. An identification system, which is effective and accurate, can add security to the existing system for secure and smooth communication among end to end nodes and can work efficiently in the identification of attack on data packets. The role of information security is to design and protect the entire data of networks and maintain its confidentiality, integrity, and availability for their right users. Therefore, there is a need for end to end security management, which will ensure the security and privacy of the network and will save the data inside networks from malicious users. As the network devices are growing, so the level of threats is also increasing for these devices. The proposed research is an endeavor toward the detection of data packets attack by using the rough set theory for a secure end to end communication. The experimental work was performed by the RSES tool. The accuracy of the K-NN was 88% for the total objects of 8459. For cross validation purposes, the decision rules and decomposition tree algorithms were used. The DR algorithm showed accuracy of 59.1%, while the DT showed accuracy of 61.5%. The experimental results of the proposed study show that the research is capable of detecting data packets attack.

show abstract

Selection of effective machine learning algorithm and Bot-IoT attacks traffic identification for internet of things in smart city

Cited by 305 publications

References 23 publications

An IoT Time Series Data Security Model for Adversarial Attack Based on Thermometer Encoding

An IoT Time Series Data Security Model for Adversarial Attack Based on Thermometer Encoding

Machine Learning-Based IoT-Botnet Attack Detection with Sequential Architecture

Identification of Attack on Data Packets Using Rough Set Approach to Secure End to End Communication

Contact Info

Product

Resources

About