Sharif Digital Repository

One of the most important requirements in deploying a security system is to ensure the effectiveness and absence of bypass patterns. This is especially important for attack-based detection systems. One of the systems that has recently attracted the attention of network administrators is Web Application Firewall (WAF). The purpose of this thesis is to propose a deep learning approach to identify the pattern of SQL Injection (SQLi) attacks which could potentially bypass a WAF. We delve into the problem of detecting SQLi attacks among a very large dataset of existing SQL queries. To this end, we use one of the latest implementation of Recurrent Neural Network (RNN) called Long Short-Term Memory... 

Due to increasing web-based attacks against web applications and inefficiency of intrusion detection and prevention systems for detecting and preventing web attacks in the application layer, web application firewalls (WAF) developed to deal with this problem. There are most common attacks affecting today’s web applications like SQL Injection (SQLi), Cross-Site Scripting (XSS) and Logical attacks. The Logical attack focuses on the abuse or exploitation of a web application’s logic flow, and unlike SQLi and XSS, it depends on the web application functionality. Nowadays, there are a lot of methods for designing a WAF which are divided into two categories: Signature-based and Anomaly-based... 

Software defined networks are developed to provide programmability and a centralized view in networks by decoupling control plane from data plane. Software defined networks are now well received,and these networks are evolving every day. This is while more attention has been paid to widen the application of these networks and eliminating the shortcomings in their performance. On the other hand, in very large networks, the issue of efficiency and processing speed is of great importance. However, performance in these networks is not satisfactory, especially in single controller based SDN due to the complex processing of packets in a unique controller. Security needs are also of great... 

In recent years, due to the drawback of rule-based firewalls in detecting unknown attacks, using neural networks has got more attention to be used in firewalls. As the computation load of neural networks are so much there is a need to decrease the processing time and power consumption as they are under load 24/7. Although there have been huge achievements in the usage of graphics processing units (which contain numerous processing cores) in neural networks, their high power consumption has made the scientists think about an alternative to implement neural networks. Field Programmable Gate Array (FPGA) is one of the most serious candidates to be used for implementing neural networks. The goal...