Please use this identifier to cite or link to this item:
https://dspace.iiti.ac.in/handle/123456789/10391
Title: | SYN flood and saturation attack detection and mitigation in SDN networks |
Authors: | Patel, Kanishk Hubballi, Neminath [Guide] |
Keywords: | Computer Science and Engineering |
Issue Date: | 27-May-2022 |
Publisher: | Department of Computer Science and Engineering, IIT Indore |
Series/Report no.: | BTP584;CSE 2022 PAT |
Abstract: | Recent works have shown that the interaction between control and data plane in the Soft ware Defined Networks can be chocked by an adversary with saturation attack. This attack is generated by sending large number of new flows to a switch exploiting the switch-controller communication. A switch sends a packet-in message to the controller if a new flow is seen. A flux of new flows results in a large number of packet-in messages at the controller. In this paper, we present SaturationGuard which mitigates this attack by adopting an early attack de tection method. An anomaly detection method deployed at the controller observes the patterns of packet-in messages and identifies the attack. In particular, we capture normal interaction between switch and controller using the arrival rate of packet-in messages with a probability distribution. To mitigate the attack, we propose to throttle the bandwidth of the affected switch port in proportion to the arrival rate of new flows. We implement a proof of concept solution with Mininet and an external controller and show that SaturationGuard is effective in handling the saturation attacks with early stage detection. |
URI: | https://dspace.iiti.ac.in/handle/123456789/10391 |
Type of Material: | B.Tech Project |
Appears in Collections: | Department of Computer Science and Engineering_BTP |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
BTP_584_Kanishk_Patel_180001025.pdf | 1.38 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
Altmetric Badge: