TY - JOUR
T1 - An SDN perspective IoT-Fog security
T2 - A survey
AU - Javanmardi, Saeed
AU - Shojafar, Mohammad
AU - Mohammadi, Reza
AU - Alazab, Mamoun
AU - Caruso, Antonio M.
N1 - Funding Information:
Dr. Mohammad secured £311k for the ESKMARALD project (funded by NCSC-UK) and £40k for the APTd5G project (funded by EPSRC/UKI-FNI) in 2022. He secured €750k (PI) for AUTOTRUST (funded by the European Space Agency) in 2021. Mohammad secured $40k (funded by the Ecudiarain government) for IoT and Industry 4.0 resource allocation in 2020. Mohammad secured €275k (PI) for PRISENODE project (Horizon 2020 Marie Curie Individual project) in 2019 and €60k (PI) for an SDN security and privacy project (funded by the University of Padua) in 2018. He contributed to some Italian projects in telecommunications like GAUChO - A Green Adaptive Fog Computing and Networking Architecture (€400k), S2C: Secure, Software-defined Cloud (€30k), and SAMMClouds- Secure and Adaptive Management of Multi-Clouds (€30k) collaborating among Italian universities. He received a Ph.D. from the Sapienza University of Rome, Rome, Italy, in 2016 with an “Excellent” degree. He received the honored B.Sc. in CS at Iran University of Science and Technology, Tehran, Iran, in 2006. He was a programmer/software analyzer at the National Iranian Oil Company (NIOC) and Tidewater ltd in Iran from 2008 to 2013. He published over 180 refereed top-tier articles in prestigious venues such as IEEE TII, IEEE TCC, IEEE TNSM, IEEE T-ITS, IEEE Network, Computer Networks, and FGCS. He is an Associate Editor in IEEE Transactions on Network and Service Management, IEEE Transactions on Intelligent Transportation Systems, IEEE Systems, IEEE Consumer Electronics Magazine, and Computer Networks Journals. He published three books on Cybersecurity Applications and Network Security, which appeared in Springer recently.
Publisher Copyright:
© 2023 Elsevier B.V.
PY - 2023/6
Y1 - 2023/6
N2 - The utilization of the Internet of Things (IoT) has burst in recent years. Fog computing is a notion that solves cloud computing's limitations by offering low latency to IoT network user applications. However, the significant number of networked IoT devices, the large scale of the IoT, security concerns, users’ critical data, and heterogeneity in this extensive network significantly complicate the implementation. The IoT-Fog architecture consists of fog devices (servers) at the fog layer, which decreases network utilization and response time due to their closeness to IoT devices. However, as the number of IoT and fog devices under the IoT-Fog architecture grows, new security concerns and requirements emerge. Because incorporating fog computing into IoT networks introduces some vulnerabilities to IoT-Fog networks, the nodes in the fog layer are the target of security threats. Software-Defined Networking (SDN) is a novel paradigm that decouples the data plane from control plane, resulting in better programmability and manageability. Attack defense mechanisms can be implemented in the IoT-Fog network without SDN. But SDN paradigm provides the IoT-Fog with some characteristics that facilitate counterattacks. This survey briefly explains some works that utilized the SDN features in the IoT-Fog network for security threats in the IoT-Oriented fog layer. To this end, we examine IoT-Fog, SDN, and SDN-based IoT-Fog networks. We describe security threats in IoT-Fog networks and briefly explain the vulnerabilities and attacks in the fog layer. Then, we describe the fog layer's most common IoT-Fog security defense mechanisms. Following that, we present the SDN features, explore how SDN can help defensive mechanisms in IoT-Fog networks, and categorize the works based on the SDN features they use. We explain their features and present a comparison between them. Finally, we discuss the disadvantages of SDN in IoT-Fog networks.
AB - The utilization of the Internet of Things (IoT) has burst in recent years. Fog computing is a notion that solves cloud computing's limitations by offering low latency to IoT network user applications. However, the significant number of networked IoT devices, the large scale of the IoT, security concerns, users’ critical data, and heterogeneity in this extensive network significantly complicate the implementation. The IoT-Fog architecture consists of fog devices (servers) at the fog layer, which decreases network utilization and response time due to their closeness to IoT devices. However, as the number of IoT and fog devices under the IoT-Fog architecture grows, new security concerns and requirements emerge. Because incorporating fog computing into IoT networks introduces some vulnerabilities to IoT-Fog networks, the nodes in the fog layer are the target of security threats. Software-Defined Networking (SDN) is a novel paradigm that decouples the data plane from control plane, resulting in better programmability and manageability. Attack defense mechanisms can be implemented in the IoT-Fog network without SDN. But SDN paradigm provides the IoT-Fog with some characteristics that facilitate counterattacks. This survey briefly explains some works that utilized the SDN features in the IoT-Fog network for security threats in the IoT-Oriented fog layer. To this end, we examine IoT-Fog, SDN, and SDN-based IoT-Fog networks. We describe security threats in IoT-Fog networks and briefly explain the vulnerabilities and attacks in the fog layer. Then, we describe the fog layer's most common IoT-Fog security defense mechanisms. Following that, we present the SDN features, explore how SDN can help defensive mechanisms in IoT-Fog networks, and categorize the works based on the SDN features they use. We explain their features and present a comparison between them. Finally, we discuss the disadvantages of SDN in IoT-Fog networks.
KW - IoT-Fog network
KW - IoT-Oriented fog structure
KW - SDN paradigm
KW - Security threat
UR - http://www.scopus.com/inward/record.url?scp=85152492754&partnerID=8YFLogxK
U2 - 10.1016/j.comnet.2023.109732
DO - 10.1016/j.comnet.2023.109732
M3 - Article
AN - SCOPUS:85152492754
SN - 1389-1286
VL - 229
SP - 1
EP - 19
JO - Computer Networks
JF - Computer Networks
M1 - 109732
ER -