Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/547908
Title: | Certain investigation on attack detection in iot networks using deep learning techniques |
Researcher: | Padmashree A |
Guide(s): | Krishnamoorthi M |
Keywords: | Convolutional Neural Network Deep Learning Internet of Things |
University: | Anna University |
Completed Date: | 2023 |
Abstract: | The Internet of Things (IoT) has grown significantly over the past ten newlineyears, and this has sparked ground-breaking advancements in the network newlinebusiness. Common IoT systems include smart cities and homes, wearable newlinetechnology, traffic control, health services, and resource efficiency. The newlineeffectiveness of such smart devices could affect the end - user, corrupt their newlineprivate data, and expand threats and risks online. Cyber-attacks make this newlineIoT-based smart city development dangerous. Consequently, it is needed to newlinedevelop an efficient system model that can protect IoT devices from attacks newlineand threats. To enhance product safety and security, the IoT-enabled newlineapplications should be monitored in real- time. newlineDeep Learning techniques are employed to identify attacks in IoT newlinenetworks. It plays an important role in the improving security of IoT networks newlineby effectively and accurately detecting attacks. It analyzes large volume of newlinedataset from the devices and the networks, identify patterns and anomalies newlinethat results in potential attack. This helps to reduce the impact of attacks and newlineprevent them from spreading to other devices or systems. Overall, the role of newlineDeep Learning (DL) in network attack identification in IoT is becoming newlineincreasingly important to address the security challenges posed by the newlineproliferation of IoT devices. By leveraging the power of these advanced newlinetechniques, it is possible to improve the accuracy and effectiveness of newlinenetwork security systems and better protect against potential cyber-attacks. newlineIn the first research work, the feature selection is done based on newlineCorrelation based Feature Selection (CFS) method on 5% of the records of newlineBot-IoT dataset. Data cleaning, standardization and label encoding is done to newlineperform data preprocessing. newline |
Pagination: | xv,136p. |
URI: | http://hdl.handle.net/10603/547908 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 99.35 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 585.8 kB | Adobe PDF | View/Open | |
03_contents.pdf | 495.05 kB | Adobe PDF | View/Open | |
04_abstracts.pdf | 126.64 kB | Adobe PDF | View/Open | |
05_chapter1.pdf | 228.36 kB | Adobe PDF | View/Open | |
06_chapter2.pdf | 225.83 kB | Adobe PDF | View/Open | |
07_chapter3.pdf | 268.43 kB | Adobe PDF | View/Open | |
08_chapter4.pdf | 572.69 kB | Adobe PDF | View/Open | |
09_chapter5.pdf | 525.92 kB | Adobe PDF | View/Open | |
10_chapter6.pdf | 343.06 kB | Adobe PDF | View/Open | |
11_annexures.pdf | 104.83 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 99.64 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: