This repository contains the code and resources for the project "Defence Against ARP Spoofing using Deep Learning". The project aims to mitigate ARP spoofing attacks using a combination of XGBoost feature extraction and a proposed CNN-LSTM architecture. This README provides an overview of the project, its components, and its current status.
ARP (Address Resolution Protocol) spoofing is a type of cyber attack where an attacker impersonates another device on a local network by sending fake ARP messages. This can lead to various security vulnerabilities and breaches. Our project focuses on developing a robust defense mechanism against ARP spoofing attacks using deep learning techniques.
We utilized the XGBoost classifier to extract the most relevant features from a publicly available dataset. XGBoost is a powerful machine learning algorithm that excels at feature selection and classification tasks. By employing XGBoost, we were able to identify key features that contribute significantly to detecting ARP spoofing attacks.
To effectively address the problem of ARP spoofing, we introduced a novel CNN-LSTM architecture. This architecture leverages the strengths of Convolutional Neural Networks (CNNs) for feature extraction from spatial data and Long Short-Term Memory (LSTM) networks for capturing temporal dependencies within the data. The combined power of these two architectures enhances the accuracy and robustness of the detection mechanism.
We are thrilled to announce that our project titled "Deep Learning based framework for dynamic Detection and Mitigation of ARP Spoofing attacks" has been accepted for presentation at the 14th International Conference on Computing, Communication, and Networking Technologies (ICCCNT), hosted by IIT Delhi. The conference will provide us with the opportunity to share our findings, insights, and innovations with the research community and industry experts.
Currently, our project is in the publication phase. As part of our commitment to transparency and knowledge sharing, we plan to release the complete codebase of our project once the publication process is successfully completed.
For inquiries, issues, or suggestions, please feel free to reach out via [email protected] or [email protected].