Author : Bharanidharan V 1
Date of Publication :1st June 2023
Abstract: The continuous increase of cyber attacks using false data injection presents a huge threat to organizations in all industries by violating the security of their systems. These attacks can cause huge losses in terms of finance, reputation as well as legal consequences. This paper proposes a unique approach for network traffic detection at the packet level by utilizing Federated Deep Learning, a robust machine learning model that collects decentralized data along with weighted hybrid technique.The proposed framework dynamically detects and classifies false data injection attacks and retrieves the control signal using the acquired information.This approach can help to reveal characteristics of the attacks including the direction magnitude and ratio of the injected false data. Using this information the signal retrieval module can easily recover the original control signal and remove the injected false data. Furthermore, the resulting model can potentially detect zero day attacks that have not been achieved before.
Reference :