Semisupervised-Learning-Based Security to Detect and Mitigate Intrusions in IoT Network

Ravi, Nagarathna; Shalinie, S. Mercy

doi:10.1109/jiot.2020.2993410

Cited by 64 publications

(36 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [18], author proposed a novel SDRK-ML algorithm, that is, supervised DNN, and further extended it to an unsupervised clustering technique. e algorithm is placed between IoT and cloud layers to make it work better.…”

Section: Related Workmentioning

confidence: 99%

Feature Entropy Estimation (FEE) for Malicious IoT Traffic and Detection Using Machine Learning

Diwan

Choubey²,

Hota³

et al. 2021

Mobile Information Systems

View full text Add to dashboard Cite

Identification of anomaly and malicious traffic in the Internet of things (IoT) network is essential for IoT security. Tracking and blocking unwanted traffic flows in the IoT network is required to design a framework for the identification of attacks more accurately, quickly, and with less complexity. Many machine learning (ML) algorithms proved their efficiency to detect intrusion in IoT networks. But this ML algorithm suffers many misclassification problems due to inappropriate and irrelevant feature size. In this paper, an in-depth study is presented to address such issues. We have presented lightweight low-cost feature selection IoT intrusion detection techniques with low complexity and high accuracy due to their low computational time. A novel feature selection technique was proposed with the integration of rank-based chi-square, Pearson correlation, and score correlation to extract relevant features out of all available features from the dataset. Then, feature entropy estimation was applied to validate the relationship among all extracted features to identify malicious traffic in IoT networks. Finally, an extreme gradient ensemble boosting approach was used to classify the features in relevant attack types. The simulation is performed on three datasets, i.e., NSL-KDD, USNW-NB15, and CCIDS2017, and results are presented on different test sets. It was observed that on the NSL-KDD dataset, accuracy was approx. 97.48%. Similarly, the accuracy of USNW-NB15 and CCIDS2017 was approx. 99.96% and 99.93%, respectively. Along with that, state-of-the-art comparison is also presented with existing techniques.

show abstract

Section: Related Workmentioning

confidence: 99%

Feature Entropy Estimation (FEE) for Malicious IoT Traffic and Detection Using Machine Learning

Diwan

Choubey²,

Hota³

et al. 2021

Mobile Information Systems

View full text Add to dashboard Cite

show abstract

“…Ravi and Shalinie [64] employ cloud computing techniques in conjunction with SDN to mitigate DDoS attacks on IoT servers. They have introduced a new mechanism called Learning-Driven Detection Mitigation (LEDEM), which detects DDoS attacks with the aid of a semi-supervised machine learning algorithm.…”

Section: Flow Filteringmentioning

confidence: 99%

“…Bull et al [72] Xu et al [55] Salva-Garcia et al [40] & Molina Zarca et al [57] Rafique et al [61] Bawany and Shamsi [31] Yang et al [60] Rafique et al [74] Nair et al [65] Galeano-Brajones et al [68] Ravi and Shalinie [64] Rate limiting Sharma et al [69] Traceback Chen et al [45] Request prioritization Sarwar et al [71] Flow filtering MTD Sahay et al [73] Honeypots MTD Luo et al [75] Hybrid SDN-Fog…”

Section: Flow Filteringmentioning

confidence: 99%

“…Xu et al [55] Krishnan et al [63] Rafique et al [61] Sarwar et al [71] Rafique et al [74] Smart Homes Bhunia and Gurusamy [25] Sharma et al [69] IoT Data Centers Bawany and Shamsi [31] Industrial IoT Yan et al [42] Ship communication systems Sahay et al [73] Generic Bull et al [72] Özçelik et al [56] Yin et al [54] Krishnan et al [70] Salva-Garcia et al [40] & Molina Zarca et al [57] Nguyen et al [58] Rathore et al [59] Yang et al [60] Houda et al [66] Luo et al [75] Chen et al [45] Nair et al [65] Galeano-Brajones et al [68] Ravi and Shalinie [64] The techniques that rely on the analysis of network flow statistics to detect anomalous actions are efficient in containing volumetric attacks, which are based on high traffic rates [80]. Concerning this, there are solutions based on the following strategies: rate limiting, cosine similarity, traceback, and flow prioritization, as outlined in Table 5.…”

Section: Sdn Control Planementioning

confidence: 99%

See 1 more Smart Citation

A Taxonomy of DDoS Attack Mitigation Approaches Featured by SDN Technologies in IoT Scenarios

Silva

Neto

et al. 2020

Sensors

View full text Add to dashboard Cite

The Internet of Things (IoT) has attracted much attention from the Information and Communication Technology (ICT) community in recent years. One of the main reasons for this is the availability of techniques provided by this paradigm, such as environmental monitoring employing user data and everyday objects. The facilities provided by the IoT infrastructure allow the development of a wide range of new business models and applications (e.g., smart homes, smart cities, or e-health). However, there are still concerns over the security measures which need to be addressed to ensure a suitable deployment. Distributed Denial of Service (DDoS) attacks are among the most severe virtual threats at present and occur prominently in this scenario, which can be mainly owed to their ease of execution. In light of this, several research studies have been conducted to find new strategies as well as improve existing techniques and solutions. The use of emerging technologies such as those based on the Software-Defined Networking (SDN) paradigm has proved to be a promising alternative as a means of mitigating DDoS attacks. However, the high granularity that characterizes the IoT scenarios and the wide range of techniques explored during the DDoS attacks make the task of finding and implementing new solutions quite challenging. This problem is exacerbated by the lack of benchmarks that can assist developers when designing new solutions for mitigating DDoS attacks for increasingly complex IoT scenarios. To fill this knowledge gap, in this study we carry out an in-depth investigation of the state-of-the-art and create a taxonomy that describes and characterizes existing solutions and highlights their main limitations. Our taxonomy provides a comprehensive view of the reasons for the deployment of the solutions, and the scenario in which they operate. The results of this study demonstrate the main benefits and drawbacks of each solution set when applied to specific scenarios by examining current trends and future perspectives, for example, the adoption of emerging technologies based on Cloud and Edge (or Fog) Computing.

show abstract

“…It is estimated that in 2020, between 25 and 50 billion devices will be connected to the network (Jaślan, 2015). The integration of data, processes, people and devices (Figure 1) along with artificial intelligence opens up many new opportunities in the area of acquiring knowledge and using it (Ravi & Shalinie, 2020). Processes that combine existing data with objects and individuals ultimately form the socalled intelligent spaces (Brachman, 2017), collections of beings of various types, which interacting with each other generate information and knowledge resources.…”

Section: Internet Of Things and Connection With Artificial Realitymentioning

confidence: 99%

Hybrid Reality in the Internet of Things as an Environment for Transferring Knowledge

Woźniakowski

Orłowski

2020

IJREL

View full text Add to dashboard Cite

The paper presents features and applications of hybrid reality technology and the Internet of Things. The analysed thesis states that such a solution is a favourable environment for the dissemination of knowledge through the concept of interaction of objects of postulated environment with the theoretical model of knowledge objects. The article discusses an initial version of the potential information system method. The method, in the form of a mathematical formula, makes it possible to evaluate objects in the postulated space. The article presents the case study implementing a similar environment in the service of advanced industrial equipment. The results of the analysis indicate a high potential of the postulated solution, which requires further refinement and research. At the same time, the practice of the implementation case points to technological and organisational threats which should be neutralised to ensure a high probability of the project’s success.

show abstract

Semisupervised-Learning-Based Security to Detect and Mitigate Intrusions in IoT Network

Cited by 64 publications

References 24 publications

Feature Entropy Estimation (FEE) for Malicious IoT Traffic and Detection Using Machine Learning

Feature Entropy Estimation (FEE) for Malicious IoT Traffic and Detection Using Machine Learning

A Taxonomy of DDoS Attack Mitigation Approaches Featured by SDN Technologies in IoT Scenarios

Hybrid Reality in the Internet of Things as an Environment for Transferring Knowledge

Contact Info

Product

Resources

About