Wormhole Attack Detection System for IoT Network: A Hybrid Approach

Snehal A Bhosale,S S Sonavane

doi:10.1007/s11277-021-09395-y

Abstract

Many errors in data communication cause security attacks in Internet of Things (IoT). Routing errors at network layer are prominent errors in IoT which degrade the quality of data communication. Many attacks like sinkhole attack, blackhole attack, selective forwarding attack and wormhole attack enter the network through the network layer of the IoT. This paper has an emphasis on the detection of a wormhole attack because it is one of the most uncompromising attacks at the network layer of IoT protocol stack. The wormhole attack is the most disruptive attack out of all the other attacks mentioned above. The wormhole attack inserts information on incorrect routes in the network; it also alters the network information by causing a failure of location-dependent protocols thus defeating the purpose of routing algorithms. This paper covers the design and implementation of an innovative intrusion detection system for the IoT that detects a wormhole attack and the attacker nodes. The presence of a wormhole attack is identified using location information of any node and its neighbor with the help of Received Signal Strength Indicator (RSSI) values and the hop-count. The proposed system is energy efficient hence it is beneficial for a resource-constrained environment of IoT. It also provides precise true-positive (TPR) and false-positive detection rate (FPR).

Highlights

Internet of Things (IoT) is an emerging technology nowadays with a wireless interconnection of sensory devices in the existing infrastructure
Sensory devices are uniquely identified by Internet Protocol (IP) addresses viz IPv4 and IPv6
Under performance-based metrics, most of the Intrusion Detection System (IDS) proposed in the literature fail to get effective results in the wormhole attack detection in IoT

Summary

Introduction

Internet of Things (IoT) is an emerging technology nowadays with a wireless interconnection of sensory devices in the existing infrastructure. Most of the researchers in this field claim that more than 30 billion devices are expected to connect to the internet by 2022. Sensory devices are uniquely identified by IP addresses viz IPv4 and IPv6. IPv4 has limitations of providing IP addresses to the network with a large number of devices whereas IPv6 protocol offers an infinite number of unique IP addresses. The performance of all these smart devices is being affected by battery power, memory, communication ranges, size, etc. For the optimal performance of the network, all the above constraints are considered by avoiding the use of bulky and battery consuming encryption or security algorithms [3]

Methods

Results

Conclusion