Reducing attack surface of edge computing IoT networks via hybrid routing using dedicated nodes

Author Identifier

James Jin Kang

ORCID : 0000-0002-0242-4187

Leslie Sikos

ORCID : 0000-0003-3368-2215

Document Type

Book Chapter

Publication Title

Secure edge computing: Applications, techniques and challenges


CRC Press


School of Science / ECU Security Research Institute




Kang, J. J., Sikos, L. F., & Yang, W. (2021). Reducing attack surface of edge computing IoT networks via hybrid routing using dedicated nodes. In M. Ahmed & P. Haskell-Dowland (Eds.), Secure edge computing: Applications, techniques and challenges (pp. 101-115). CRC Press.


Network bandwidth capacity and processing performances have increased due to the increasing availability of affordable, expandable and energy-efficient networking technologies such as Low-Power Wide Area Networks (LPWANs) emerged with IoT networks in the public and private network spaces. Numerous applications with smart devices and health-related applications have grown in popularity with widespread uptake of Internet of Things (IoT) technologies. This has raised security concerns given the confidential nature of information that many IoT devices carry. A compromise of this information such as MAC/IP addresses or personal data can lead to privacy breach issues. IoT networks are vulnerable to man-in-the-middle (MITM) attacks which exhibit two observable features: 1) time delay in the session that is measurable and 2) data travel times which are unusual when compared with previous normal transactions. This chapter describes a novel scheme that aims to improve the ability to detect these attacks. This is achieved through a hybrid routing mechanism that specifies dedicated nodes for routing traffic between IoT devices and users while minimizing the burden of workload on the network. Having dedicated devices with increased battery and computational capacity would offer three main benefits: 1) to be able to define secure pathways within the network and avoiding routing through suspicious or untrusted nodes/networks, 2) help stabilize travel time (by reducing fluctuations) within a trusted time server (TTS) that would result in more accurate time estimations and 3) allow for packets to be inspected through security checks. The proposed model may improve IoT network security by bypassing high-risk network segments.

Access Rights

subscription content