Crossfire Attack Detection Using Deep Learning in Software Defined ITS Networks

Narayanadoss, Akash Raj; Truong-Huu, Tram; Mohan, Ponnusamy; Gurusamy, Mohan

doi:10.1109/vtcspring.2019.8746594

Cited by 30 publications

(21 citation statements)

References 8 publications

(13 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition, the authors proposed an MC-CNN model to maximize feature information for better recognition. The authors in [ 37 ] proposed an automatic learning approach based on SDN capabilities. Advanced learning methods using ANN, LSTM, and CNN to build the learning model.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Deep Learning-Driven SDN Enabled Mechanism for Secure Communication in Internet of Things (IoT)

Javeed

Khan

Ahmad

2021

Sensors

View full text Add to dashboard Cite

The Internet of Things (IoT) has emerged as a new technological world connecting billions of devices. Despite providing several benefits, the heterogeneous nature and the extensive connectivity of the devices make it a target of different cyberattacks that result in data breach and financial loss. There is a severe need to secure the IoT environment from such attacks. In this paper, an SDN-enabled deep-learning-driven framework is proposed for threats detection in an IoT environment. The state-of-the-art Cuda-deep neural network, gated recurrent unit (Cu- DNNGRU), and Cuda-bidirectional long short-term memory (Cu-BLSTM) classifiers are adopted for effective threat detection. We have performed 10 folds cross-validation to show the unbiasedness of results. The up-to-date publicly available CICIDS2018 data set is introduced to train our hybrid model. The achieved accuracy of the proposed scheme is 99.87%, with a recall of 99.96%. Furthermore, we compare the proposed hybrid model with Cuda-Gated Recurrent Unit, Long short term memory (Cu-GRULSTM) and Cuda-Deep Neural Network, Long short term memory (Cu- DNNLSTM), as well as with existing benchmark classifiers. Our proposed mechanism achieves impressive results in terms of accuracy, F1-score, precision, speed efficiency, and other evaluation metrics.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hybrid Deep Learning-Driven SDN Enabled Mechanism for Secure Communication in Internet of Things (IoT)

Javeed

Khan

Ahmad

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In [27], the authors propose a bio-inspired SDNbased IDS for cross fire attacks. [28] presents a novel SDN originated IoT network security architecture, SeArch, leveraging deep learning models for intelligent threat detection in IoT network that scores 95.5% of average accuracy employing Stacked Autoencoder (SAE) and Self Organizing Map (SOM). The work done in [29], demonstrates an artificial neural network-based approach to train a network packet inspector for the identification of malicious packets from the IoT devices.…”

Section: Background and Related Workmentioning

confidence: 99%

“…To increase the effectiveness of IDS, there is a need to shift all values to a scaled version as it depurates the effect of gross influence. This process is referred as normalization [28]. MinMaxScaler function have been utilized to perform the normalization on the feature vectors of Bot-IoT dataset.…”

Section: ) Data Normalizationmentioning

confidence: 99%

SDN orchestration to combat evolving cyber threats in Internet of Medical Things (IoMT)

Liaqat

Akhunzada

Shaikh

et al. 2020

Computer Communications

View full text Add to dashboard Cite

Internet of Medical Things (IoMT) is now worth a billion dollar market. While offering enormous benefit, the prevalent and open environment of IoMT ecosystem can be a potential target of varied evolving cyber threats and attacks. Further, extensive connectivity of IoMT devices and their dynamic massive heterogeneous communication can create a new attack surface for sophisticated multivector malware attacks. There is a dire need to protect the forthcoming IoMT industrial revolution from varied evolving cyber threats and attacks. The authors propose a hybrid DL-driven SDN-enabled IoMT framework leveraging Convolutional Neural Network (CNN) and Cuda Deep Neural Network Long Short Term Memory (cuDNNLSTM) for a timely and efficient detection of sophisticated multivector malware botnets. For comprehensive evaluation, a state-of-the-art IoMT dataset and standard performance metrics have been employed. For verification purpose, we compare our proposed framework with our constructed hybrid DL-driven architectures and benchmark algorithms. Our proposed technique outperforms in terms of detection accuracy and testing efficiency. Finally, we also perform 10-fold cross validation to utterly show unbiased results.

show abstract

“…They compared the accuracy of neural network models on both standard data and infected data. In the same year, the authors in [17] employed a neural network model to address the network attacks issue that may arise in intelligent transport systems. They developed a machine learning model that shows the temporal relationship between the traffic data that flows in the smart transport system network, so that this neural network can differentiate between legitimate traffic data to that of coordinated network data.…”

Section: Related Workmentioning

confidence: 99%

An Empirical Comparison on Malicious Activity Detection Using Different Neural Network-Based Models

2020

View full text Add to dashboard Cite

The internet is growing at a rapid pace offering multiple web-based applications catering to the changing needs and demands of customers. Nevertheless, extensive use of internet services has potentially exposed the threats of data security and reliability. With technological advancements, cyber threats have also become more sophisticated with the blend of distinctive forms of attacks to cause potential damage. The increase in the number and variety of cyber attacks is inevitable; hence it is imperative to improve the efficiency of the cyber security systems. This research aims to compare different neural network models to distinguish malicious acts from non-malicious ones. The examined models are trained, validated, and tested using two datasets(cyber-physical subsystem dataset and KDD dataset). The performance of the studied models is measured using the confusion matrix. For the cyber-physical subsystem dataset, binary classification and multi-class classification are used for evaluating the models. In the KDD dataset, binary classification is the only classification approach because the dataset contains two classes, regular (normal actions) and harmful (malicious actions). In general, the results in binary classification are more encouraging than in multi-class classification. Among all the models, the PNN model achieves the best performance, while the GRNN model is the fastest one. Although PNN's runtime is slightly higher than the GRNN model, we can claim that the PNN is the best model for our data because a trade-off between the performance and run time can be obtained.

show abstract

Crossfire Attack Detection Using Deep Learning in Software Defined ITS Networks

Cited by 30 publications

References 8 publications

A Hybrid Deep Learning-Driven SDN Enabled Mechanism for Secure Communication in Internet of Things (IoT)

A Hybrid Deep Learning-Driven SDN Enabled Mechanism for Secure Communication in Internet of Things (IoT)

SDN orchestration to combat evolving cyber threats in Internet of Medical Things (IoMT)

An Empirical Comparison on Malicious Activity Detection Using Different Neural Network-Based Models

Contact Info

Product

Resources

About