An Enhanced Intrusion Detection System for IoT DDoS Attack

Abstract

Smart farming, as an integral part of the Internet of Things (IoT), is gaining popularity as a means of meeting the world's growing demand for food. There are many approaches for smart farms to use technology and connected devices. For example, from receiving real-time information on crop status and soil moisture to operating drones to assist with tasks such as spraying pesticides. However, the use of various internet-connected devices introduces some vulnerabilities in the smart farming ecosystem. Intruders can exploit these vulnerabilities to manipulate and remotely disrupt the flow of data from field sensors. This will have negative consequences, especially in high-risk situations such as harvesting, where real-time monitoring is required. This study uses both qualitative and quantitative approaches to analyse the intended research objectives for Denial of Service (DOS) attack mitigation in smart farms. Furthermore, the study presents an enhanced Intrusion Detection System (IDS) utilising disparate metrics to validate the performance and efficacy of the IDS deployed. Additionally, it asserts that distinct IDS methodologies achieve an accuracy of 0.95, precision of 0.92, recall of 0.90, F1 Score of 0.91, and ROC curve of 96. The study concludes that the application of the IDS measurement method can be effectively employed to resolve instances of Denial of Service (DOS).