Research Projects

In the era-of information and communication technology, people rely on modern networks like Internet-of-Things (IoT), Cloud, and Software-Defined-Networking (SDN) for their day-to-day activities. Due to the wide applications of IoT networks in healthcare, smart cities, agriculture, intelligent transportation, Industries, and government organizations, the number of cyber-attacks and threats are increasing day by day. Moreover, the number of devices connected to IoT networks are increasing significantly, and it is expected that more than 75 billion devices including smart watch, refrigerator, washing machine, CCTV camera, etc. will be connected to IoT networks by 2025. The devices connected to IoT networks are less secured with limited memory, and hence they are always vulnerable to attackers. Attackers compromise a large number of IoT devices to form a large IoT botnet that can be used to launch sophisticated attacks such as Distributed Denial of Attacks (DDoS). Nowadays, botnet-based network attacks are prevalent, and many organizations, including Amazon, eBay, Twitter, Shopify, and many others, incur a substantial financial loss for IoT botnet attacks. Therefore, detection and prevention of botnet-based IoT attacks are significant research problems of modern network security. Identification of IoT botnets and botnet-based attacks are very challenging due to the heterogeneous nature of the devices. So there is of prime importance to identify botnet-based attacks at the earliest after lunching and prevent the attacks before significantly impacting the network. Some recent techniques like Deep learning and SDN will effectively solve my proposal's objectives. Deep learning methods handle massive network traffic in identifying botnet-based attacks. So, deep learning-based methods will be very effective for identifying attack traffic by analyzing massive network data. Moreover, to protect the IoT network from significant damage, botnets are detected at the earliest and blocklist the IPs of the devices connected to the botnet. Although this objective is very challenging, Software Define Networking can play a significant role in controlling the IoT network traffic flows generated from malicious devices to prevent attacks.