IMCCRT 2023 4220 Title- DDoS Attack Detection from Optimal Feature Set in eHealth Cloud Environment
Title- DDoS Attack Detection from Optimal Feature Set in eHealth Cloud Environment – A Module using LSTM
Author's name- ANINDYA BOSE
Abstract- Distributed denial of service (DDoS) is a common type of cyber attack in cloud computing scenario. Though the healthcare cloud was comparatively immune to attack, in recent times there had been instances of DDoS attack which caused lot of damage. It is essential to identify the attacks at the earliest. In this work the authors considred three types of DDoS attacks – ICMP Flood, TCP-SYN Flood, and LAND Flood attack in healthcloud environment. The authors proposef a Long Short Term Memory (LSTM) based model to identify the data traffic as either as Flood or as Normal. The optimal feature set of the data traffic was found using ensemble based technique. The acciracy for the three types of attacks were between 98% - 99.5%. The authors compared the result with the outcome obtained with classical machine learning algorithm and found that the proposed system performed better for all the three types of attacks.