Article Preview
TopIntroduction
The Internet of Things network can be formed by interconnecting of physical objects in network that used build sensors to obtain data from environment and embedded connectivity to shares and exchanges information through the internet. ‘Things’ can be any heterogeneous devices such as computing devices, digital machines, electronic gadgets, objects, cameras, mobile devices. The smart object is the major element of the IoT networks. The object has been integrated with intelligence and is able to be dynamically interconnected together and forms a network, gather information from the environment and intermingle with other objects, disseminate data and information. The IoT network consists of millions of interconnected devices and huge volume of data (Borgia, E. 2014).
The advanced Sensors integrated with IoT devices which are monitoring and detection by sensing and sharing information from the environment. There are several real-time AI applications with cloud and an edge computing solution using the IoT networks. Nowadays IoT emerged as an important useful innovation in our day today life which uses smart systems, framework and intelligence devices to sense and gather information around us and provides a comfortable environment. In real time we can deploy IoT technologies in smart city, smart healthcare, smart vehicles, smart agriculture, smart homes, smart energy efficient applications, smart industries (Kumar, S., et al., 2019).
The below figure 1 exhibits the application areas of IoT devices in detail.
Figure 1. Example Application areas of IoT
The main objective of the Internet of Things networks create systems with heterogeneous interoperable devices, Providing security for sharing of data between IoT devices and also providing protection mechanisms against intrusion attack for the network(Patel, K. K., et al., 2016). Security threats to IoT networks are obvious since due to the number of devices connected with various standards (Choudhary, S., et al., 2019). An intrusion detection system works in the network layer which can used to analyze packets of data and provides a results, also examine the data packets in any protocol layers, and uses converge of technologies and protect IoT network from various types of attacks (Gendreau, A. A., et al., 2016). To manage and discover security vulnerabilities it necessary to Monitor and analyzing user information, networks, and services of IoT network in a timely manner by using the tools such as passive traffic collection and analysis (Rubio-Loyola, J., et al., 2008).
IoT network is an interconnected with intelligent low powered devices which communicate themselves with less computing power, Hence the providing security, authentication, scalability integration and privacy is a major concern in IoT. The IoT systems are prone to intrusion attacks, specifically denial-of-service (DoS) attacks and also the distributed denial-of-service (DDoS) attacks. These attacks can cause substantial damage to the applications with Smart IoT network environment. It is challenging task to provide security IoT systems.
The IoT network poses an ability to interconnect a range of heterogeneous devices such as intelligent objects, embedded intelligent sensors, context-aware computations,laptops,mobiles,computing networks and smart objects that diverge in their design, systems, protocols, intelligence, applications, vendors, and sizes. Nowadays due to the increased growth of smart IoT networks there is significant increase in the number of connected devices in the IoT which lead to more vulnerable to DDoS attacks to grow in numbers and frequency. These DDoS attacks prevent the access to smart IoT infrastructure which leads to Network unavailability. The main objective of this paper is Deep-GAN (Generative Adversarial Networks) method to create modeling by employing DL techniques with auto encoder to detect malicious attacks in IoT networks.