Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/519232
Title: | Traceback of distributed denial of service attacks in software defined networking |
Researcher: | Fenil E |
Guide(s): | Mohan Kumar P |
Keywords: | Artificial Neural Network Deep Neural Network Software-Defined Networking |
University: | Anna University |
Completed Date: | 2022 |
Abstract: | Software-Defined Networking (SDN) allows various protocols for transmission in a radio system that can be configured, modified, and combined with a large number of bands. Research and development in wireless communications have advanced rapidly, and it is considered an important technological possibility. Accurate newlineDistributed Denial of Service (DDoS) defense classification and detection is an newlineemerging technology in software defined networks. Even though DDoS defense newlinedetection is a trendy topic, it is very ra re to realize the accuracy of the detection. To overcome the attacks this research focused on three novel methods to address the objectives by trust factor calculation to generate a dictionary of network traffic parameters. In the first phase, Artificial Neural Network (ANN) with a trust factor is constructed, wherein a decision engine is used to incorporate anomaly detection and newlineabuse detection into the suggested technique. In the second phase Deep Learning RADAR (DL-RADAR) collects multiple data with the help of centralized logic and powerful compatibility. The handled data from the controller has been used to train the Deep Neural Network (DNN) which consists of an Auto Encoder (AE) and Soft Regression (SR) layer. AE is used to extract the data from the collected data and SR is used as a classifier to classify the attacks and based on the classification, attacks have been detected. In the third phase, it is possible to incorporate Efficient Attention Modules (EAM) such as EAM-S, wherein S stands for spatial, and EAM-T, wherein T stands for temporal, by sharding Blockchain with 3D-Residual Network (ShChain 3D-ResNet). Such EAMs acquire attentive ratings for numerous spatial channels and multiple temporal frames while implementing symmetrical requirements. newline |
Pagination: | xvi,188 |
URI: | http://hdl.handle.net/10603/519232 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 89.22 kB | Adobe PDF | View/Open |
02_prelim_pages.pdf | 2.93 MB | Adobe PDF | View/Open | |
03_contents.pdf | 154.81 kB | Adobe PDF | View/Open | |
04_abstracts.pdf | 65.94 kB | Adobe PDF | View/Open | |
05_chapter1.pdf | 408.04 kB | Adobe PDF | View/Open | |
06_chapter2.pdf | 564.34 kB | Adobe PDF | View/Open | |
07_chapter3.pdf | 473.12 kB | Adobe PDF | View/Open | |
08_chapter4.pdf | 532.18 kB | Adobe PDF | View/Open | |
09_chapter5.pdf | 607.16 kB | Adobe PDF | View/Open | |
10_chapter6.pdf | 628.72 kB | Adobe PDF | View/Open | |
11_annexures.pdf | 193.55 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 105.58 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: