Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/466967
Title: | Efficient intrusion detection models using deep learning techniques for fog computing environment |
Researcher: | Kalaivani, K |
Guide(s): | Chinnadurai, M |
Keywords: | Engineering and Technology Computer Science Computer Science Interdisciplinary Applications Deep Learning Fog Computing CNN |
University: | Anna University |
Completed Date: | 2022 |
Abstract: | The Internet of Things (IOT) is an emerging technology that newlineintegrates the Internet and physical smart objects. The integration of IoT and newlinecloud computing provides infrastructure, servers, and storage required for newlinereal-time operations and processing. Although IoT applications benefit from newlinecloud computing, the cloud computing paradigm confronts some issues such newlineas high latency, bandwidth, network failure, and reliability. To solve these newlinedifficulties, Fog computing paradigm is introduced as an extension to cloud newlinecomputing by offering processing, storage, and networking connection at the newlineedge between data centres in cloud computing environments and end devices. newlineIt reduces delays in communication between end-users and the cloud via fog newlinedevices for time-critical applications. However, the Fog computing environment is vulnerable to a variety of malicious attacks including DoS attacks. Attackers may send malicious data packets to fog devices that can lead to unexpected loss. As a result, an newlineeffective Intrusion Detection System (IDS) is required to ensure the secured newlineoperation of fog without compromising efficiency. Most IDSs have a low newlineaccuracy and high false alarm rate when it comes to anomaly detection. Deep newlinelearning techniques are a subset of machine learning that has recently evolved newlineand is being used to construct an IDS capable of automatically detecting and newlineclassifying attacks at the network level. It can also efficiently detect modern newlinenetwork attacks.In the present research work, deep learning based intrusion newlinedetection models are proposed. A hybrid deep learning intrusion detection newlinemodel for fog computing environment ICNN-FCID (Integrated Convolutional newlineNeural Network for Fog Computing Environment) is proposed that integrates newlinedeep learning models of Convolutional Neural Network (CNN) and Long newlineShort Term Memory (LSTM) for detecting the intrusions from the network newlinetraffic. newline newline |
Pagination: | xviii,134p |
URI: | http://hdl.handle.net/10603/466967 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 20.27 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 1.4 MB | Adobe PDF | View/Open | |
03_content.pdf | 96.36 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 90.29 kB | Adobe PDF | View/Open | |
05_chapter 1.pdf | 917.24 kB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 318.48 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 866 kB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 416.11 kB | Adobe PDF | View/Open | |
09_annexures.pdf | 153.4 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 99.93 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: