A Survey on the Latest Intrusions and Their Detection Systems in IoT-Based Network

Abstract

The Internet of Things is a boon to mankind, with a plethora of capabilities to provide, yet, like two sides of a coin, the technology’s lack of data security could be a major drawback. By 2030, it is expected that about 25.44 billion IoT devices will be connected around the world, but nearly 98% of IoT communications are reported to be unencrypted, revealing confidential and personal data on the network. Thus, the vulnerabilities of the system are exposed to a variety of attacks, impairments, and misuses caused by difficulties such as resource constraints, heterogeneity, lack of standardization, and architecture, among others. Bringing together our informative findings, hypotheses, and challenging outcomes, the paper aims to help IoT developers handle dangers and security problems for improved protection. The solutions that are discussed here mostly rely on machine learning (ML), but other techniques such as blockchain (BC) technology, fog computing (FC), edge computing (EC), and certain open research topics are also explored. In addition, our study also addresses scopes for future directions in securing the IoT network and finding potential solutions to various threats and attacks.