Please use this identifier to cite or link to this item:
http://dspace.dtu.ac.in:8080/jspui/handle/repository/19143
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | KARAMVEER | - |
dc.date.accessioned | 2022-06-07T06:15:31Z | - |
dc.date.available | 2022-06-07T06:15:31Z | - |
dc.date.issued | 2022-05 | - |
dc.identifier.uri | http://dspace.dtu.ac.in:8080/jspui/handle/repository/19143 | - |
dc.description.abstract | The Internet consists of multiple interconnected systems/networks, one of which being the “Internet of Things”. In spite of their flexibility, numerous IoT devices/gadgets are technically weak in terms of security, which makes them an ideal target for a variety of security breaches, including botnet assaults. IoT applications in the smart city are currently being targeted by advanced persistent threats (APT). Botnets are a piece of malware that permits hackers to take control of several systems and carry out destructive operations. IoT-based botnet assaults have become increasingly common as a result of the development of IoT gadgets, which are more readily hacked than desktop PCs. To combat this new danger, advanced approaches for identifying attacks initiated from infected IoT devices and distinguishing between day and milliseconds duration assaults must be developed. This study aimed to find, assess, and present a comprehensive overview of experimental works on IoT botnet detection research. The identification methods used to identify IoT botnets, their stages, and the botnet stealth strategies were all investigated in this study. The writers examined the nominated study as well as the major approaches used in it. The authors analyzed the botnet stages when detection is done and categorized the detection methods depending on the strategies utilized. The authors examined current research gaps and proposed future research topics as a consequence of this investigation and proposed a network-based anomalous detector that leverages deep learning to identify aberrant network traffic flowing from exploited IoT nodes by extracting network behavioral snapshots. On the UNSW dataset with a slew of neural network architectures and hidden layers, the suggested model combining CNN and LSTM has been trained and assessed. To test our strategy, I employed a dataset of various commercial IoT nodes infiltrated with iv Mirai and BASHLITE, two popular IoT botnets. The results of our tests showed that our suggested strategy could correctly and quickly detect assaults as they were launched from hacked IoT nodes that were members of a botnet. | en_US |
dc.language.iso | en | en_US |
dc.relation.ispartofseries | TD-5731; | - |
dc.subject | IOT BOTNET DETECTION | en_US |
dc.subject | IoT-DEVICES | en_US |
dc.subject | CNN MODELS | en_US |
dc.subject | LSTM | en_US |
dc.title | IOT BOTNET DETECTION | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | M.E./M.Tech. Computer Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
KARAMVEER M.Tech..pdf | 1.95 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.