Denial of service detection and prevention in set - top box using machine learning

Denial of Service (DoS) is one of the most occurring and the most dangerous
attacks in the Security world. DoS is an attack meant to shut down a machine
or network, making it inaccessible to its intended users. DoS attacks
accomplish this by
ooding the target with trac, or sending it information
that triggers a crash. DoS attacks do not typically result in the theft or loss
of signicant information or other assets, they can cost the victim a great
deal of time and money to handle.
Set Top box is an endpoint of entire broadcasting unit. It is deployed in
the user's side & it contains some information about the customer. Nowadays
Set Top Box (STB) came with internet connection facility & we know that
every device that connected to the internet has a chance to be vulnerable to
the security threats. So we need a system to provide security to the STB.
In this project I have developed a Tool for real time DoS detection using
supervised Machine Learning Algorithm. Here I used Decision tree classier
as my machine learning model for detecting DoS attack in Set Top Box. I
have considered a multi class detection for DoS attack including ICMP Flood,
UDP Flood, TCP Flood and Ping Death. So my proposed system is aims to
detect these four types of DoS attack in Set-Top Box live network trac.

MSC CS 2017-2019 INT Dr. Elizabeth Sherly