Collaborative Intrusion Detection for VANETs: A Deep Learning-Based Distributed SDN Approach

Shu, Jiangang; Zhou, Lei; Zhang, Weizhe; Du, Xiaojiang; Guizani, Mohsen

doi:10.1109/tits.2020.3027390

Cited by 118 publications

(55 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Network Intrusion Detection System (NIDS) is often designed for specific use cases. In the literature, Federated Learning (FL) method has been proposed for intrusion detection in Wireless Edge Network (WEN) [32], [33], IoT [21]- [23], [34]- [39], Industrial IoT (IIoT) [24], [40]- [42], industrial Cyber-Physical System (CPS) [43], Medical CPS [44], Wireless Fidelity (Wi-Fi) network [45], large-scale distributed Local Area Network (LAN) [46], [47], satellite-terrestrial integrated networks [48], Cloud [49], edge computing [50], vehicular network [26], [51], [52]. We acknowledge that FL methods have been proposed for intrusion detection in IoT networks [21]- [23], [34]- [39].…”

Section: Review Of Related Workmentioning

confidence: 99%

“…Edge computing can be combined with DL to bring intelligence closer to where data are being generated, thereby addressing the issues of data privacy, high communication cost, large memory space requirement, short training time, and high latency [25]. Localized DL (LDL) and Distributed DL (DDL) methods achieve edge intelligence without data aggregation [26]. However, the classification p erformance of these methods is usually low in zero-day (previously unknown) botnet attack scenarios because a single IoT edge device has limited training samples.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Federated Deep Learning for Zero-Day Botnet Attack Detection in IoT-Edge Devices

Popoola

Ande²,

Adebisi

et al. 2022

IEEE Internet Things J.

191

View full text Add to dashboard Cite

Deep Learning (DL) has been widely proposed for botnet attack detection in Internet of Things (IoT) networks. However, the traditional Centralized DL (CDL) method cannot be used to detect previously unknown (zero-day) botnet attack without breaching the data privacy rights of the users. In this paper, we propose Federated Deep Learning (FDL) method for zero-day botnet attack detection to avoid data privacy leakage in IoT edge devices. In this method, an optimal Deep Neural Network (DNN) architecture is employed for network traffic classification. A model parameter server remotely coordinates the independent training of the DNN models in multiple IoT edge devices, while Federated Averaging (FedAvg) algorithm is used to aggregate local model updates. A global DNN model is produced after a number of communication rounds between the model parameter server and the IoT edge devices. Zero-day botnet attack scenarios in IoT edge devices is simulated with the Bot-IoT and N-BaIoT data sets. Experiment results show that FDL model: (a) detects zero-day botnet attacks with high classification performance; (b) guarantees data privacy and security; (c) has low communication overhead (d) requires low memory space for the storage of training data; and (e) has low network latency. Therefore, FDL method outperformed CDL, Localized DL, and Distributed DL methods in this application scenario. Index Terms-Cybersecurity, botnet detection, federated learning, deep learning, deep neural network, Internet of Things. I. INTRODUCTION B OTNET attack is a serious cyber security challenge facing the Internet of Things (IoT) [1]-[3]. In our context, a botnet is a network of compromised devices that is used to launch cyber attack against critical infrastructures [4]. This cyber attack may be in form of Denial of Service (DoS), Distributed DoS (DDoS), reconnaissance, or data theft [5].

show abstract

Section: Review Of Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

Federated Deep Learning for Zero-Day Botnet Attack Detection in IoT-Edge Devices

Popoola

Ande²,

Adebisi

et al. 2022

IEEE Internet Things J.

191

View full text Add to dashboard Cite

show abstract

“…Shu et al [208] demonstrate an IDS for Vehicular Ad Hoc Networks (VANETs) by installing a distributed SDN controller on each base station to distinguish regular network traffic and malicious network traffic. Using the full network flow information, they used GAN to jointly train numerous SDN controllers for the entire VANET without directly trading their sub-network flows.…”

Section: Unsupervised DL Based Ids In Sdnmentioning

confidence: 99%

Intrusion Detection System in Software-Defined Networks Using Machine Learning and Deep Learning Techniques –A Comprehensive Survey

Rayhan¹,

Islam²,

Shatabda³

et al. 2022

Preprint

View full text Add to dashboard Cite

<div>At present, the Internet is facing numerous attacks of different kinds that put its data at risk. The safety of information within the network is, therefore, a significant concern. In order to prevent the loss of incredibly valuable information, the Intrusion Detection System (IDS) was developed to recognize the outbreak of a stream of attacks and notify the network system administrator providing network security. IDS is an extrapolative model used to detect network traffic as routine or attack. Software-Defined Networks (SDN) is a revolutionary paradigm that isolates the control plane from the data plane, transforming the concept of a software-driven network. Through this data and control plane separation, SDN provides us the opportunity to create a manageable and programmable network, allowing applications in the top plane to access physical devices via the controller. The controller functioning inside the control plane executes network modules and establishes flow rules to forward packets in the switches residing in the data plane. Cyber attackers target the SDN controller to subdue the control plane, which is considered the brain of the SDN, providing a plethora of functionalities such as regulating flow control to switches or routers in the data plane below via southbound Application Programming Interfaces (APIs) and business and application logic in the application plane above via northbound APIs to implement sophisticated networks. However, the control plane becomes a tempting prospect for security attacks from adversaries because of its centralization feature. This paper includes an in-depth overview of the notable published articles from 2015 to 2021 that used Machine Learning (ML) and Deep Learning (DL) techniques to construct an IDS solution to provide security for SDN. We also present two detailed taxonomic studies regarding IDS, and ML-DL techniques based on their learning categories, exploring various IDS solutions to secure the SDN paradigm. We have also conducted brief research on a few benchmark datasets used to construct IDS in the SDN paradigm. To conclude the survey, we provide a discussion that sheds light on continuous challenges and IDS issues for SDN security.</div>

show abstract

“…Deep learning is doing wonders when it comes to intrusion Detection Systems both in WSN and VANETs. To suppress the inter-vehicle attacks and malicious nodes detection, deep learning based distributed approach is widely adapted for abnormal activity detections [36]. Another fault detection technique is Distributed data mining method based on DNN [2] and routing technique is IP Base node degree of WC [37].…”

Section: ) Deep Learningmentioning

confidence: 99%

Role of Machine Learning in WSN and VANETs

Gillani

Niaz

Tayyab

2021

IJECER

View full text Add to dashboard Cite

Dynamic nature of Vehicular Ad-hoc Networks (VANETs) and Wireless Sensor Networks (WSN) makes them hard to deal accordingly. For such dynamicity, Machine learning (ML) approaches are considered favourable. ML can be described as the process or method of self-learning without human intervention that can assist through various tools to deal with heterogeneous data to attain maximum beneﬁts from the network. In this paper, a quick summary of primary ML concepts are discussed along with several algorithms based on ML for WSN and VANETs. Afterwards, ML based WSN and VANETs application, open issues, challenges of rapidly changing networks and various algorithms in relation to ML models and techniques are discussed. We have listed some of the ML techniques to take additional consideration of this emergent ﬁeld. A summary is given for ML techniques application with their complexities to cover on open issues to kick start further research investigation. This paper provides excellent coverage of state-of-the-art ML applications that are being used in WSN and VANETs with their comparative analysis.

show abstract

Collaborative Intrusion Detection for VANETs: A Deep Learning-Based Distributed SDN Approach

Cited by 118 publications

References 23 publications

Federated Deep Learning for Zero-Day Botnet Attack Detection in IoT-Edge Devices

Federated Deep Learning for Zero-Day Botnet Attack Detection in IoT-Edge Devices

Intrusion Detection System in Software-Defined Networks Using Machine Learning and Deep Learning Techniques –A Comprehensive Survey

Role of Machine Learning in WSN and VANETs

Contact Info

Product

Resources

About