A Comprehensive Review of Recent Types of Flooding Attack and Defense Methods in IoT-Based Smart Environments

Abstract

In an attempt to completely transform people's lives, smart cities have implemented a number of modifications. Nevertheless, even though smart cities greatly improve people's quality of life and provide significant convenience, there are still more unaddressed cyber security risks, such as malicious cyberattacks and information leaks. The efficient design of the defense model is crucial for safeguarding smart city cyberspace, as present cyber security advancements are not keeping up with the rapid uptake of these technologies worldwide. The present study describes in detail the architecture of a smart city and the sophisticated types of attacks that could target it. Also, the study examines the current literature on IoT security in smart cities and provides an overview of the concepts of cyber security, learning-based defense methods, and smart cities. In particular, a number of learning methods models such as Instance Supervised Learning, Supervised Sequence Learning, Semi-Supervised Learning, Reinforcement Learning, And Hybrid Learning methods were quickly examined. Additionally, the review illustrates the testing datasets used to test and evaluate the performance of the proposed defense methods.