DDoS Attack Detection with Deep Learning Algorithm for SNMP, NetBISO, and DNS

Abstract

Abstract: In this day and age of advanced technology, devices that are connected to the Internet and can think are a big part of both our everyday lives and the work we do in factories. The number of Internet of Things devices has been steadily increasing from one year to the next, and it is expected that by 2030, there will be 126 billion of them. On the other hand, the number of distributed denial of service, or DDoS, attacks on the internet's surface has gone up as the number of Internet of Things devices has grown. Because IoT devices are limited in what they can do, it's important to come up with some advanced security techniques to protect the DDoS surface. Because of this, people who want to take control of an Internet of Things device can attack it. This thesis uses the CICDoS2019 dataset to improve how bugs are handled and build a new taxonomy that can handle DDoS attacks better. In the end, this will make the defense against these kinds of attacks stronger. In this paper, the DNN and the LSTMs methods to find distributed denial of service threats (SNMP, NetBIOS, DNS). With our suggested method, accuracy rates of 99.99% have been reached.

.

Keywords SNMP, NetBIOS, DNS, LSTM, DDN, Deep Learning