Security Solutions for Next-Generation Firewalls

Abstract

These days, internet is used for almost everything. There are several mediums (channels) available for people to search for information that is relevant either in terms of business or entertainment perspective. These mediums can be used by a hacker to steal information and perform an attack on any network infrastructure using various techniques. Hence, securing the network is essential for cyber-security. Accordingly, various types of attacks like denial of service (DoS) attacks, backdoors, buffer overflow, guess the password, Smurf DoS etc. are quite prevalent and are commonly used by an attacker. This paper presents the complete process to detect and re-mediate / mitigate these attacks using automatic detection techniques. The Paper includes ways to improve network security using the feature of Intrusion Prevention System in Firewall by detecting and analyzing the packet flows. The main objective of the project is to detect the attack, analyze it and if it is risky then drop the attacking packet before it harms the network. Also, the paper presents the error and accuracy rate of detecting attacks using the different machine learning algorithms which can be used in the IPS.