Artificial Intelligence Based Cyber Threat Detection in Smart City Networks

Abstract

Modern smart cities use lots of devices, known as Internet of Things (IoT) devices to support important systems such as transport, healthcare and power distribution over the internet. This connection that can makes things run smoothly but it also brings the big cybersecurity worries. When systems are connected with online it leads to create various problems. Even the small weaknesses in devices like sensors or smart meters can let attackers into the wider city network. This paper looks at the cybersecurity challenges in cities showing how the different smart city systems are all connected so if one fails it can cause problems in other areas. It also highlights the growing risk to data privacy as cameras, sensors and digital platforms are always collecting information about citizens. To deal with these threats’ security measures like encryption, authentication and access control need to be built in from the start not added. The study found that using machine learning to detect intrusions can be 95% accurate, in smart city IoT networks. Also having layers of security can reduce successful attacks by nearly 60% but using these in real-time is hard because of scalability and limited computing power.