A Stacked Ensemble Method with Adaptive Attribute Selection to Detect DDoS Attack in Cloud-Assisted WBAN

The present era of technology improves the health care services and Wireless Body Area Network (WBAN) is one of them. It is a technology which uses wireless sensors to gather the vital signs from human body for monitoring. The WBAN is often connected with cloud to overcome processing and storage limitations. However, using cloud with WBAN opens up the door for various attacks. DDoS is one of the major threat which directly affects the availability of patient data, and harness the adoption of cloud-assisted WBAN technology. Therefore, in this work we propose a approach to detect DDoS attack in cloud-assisted WBAN and ensures the availability of patients data. The proposed approach is based on Adaptive and Supreme Attribute Selection with Stacked Ensemble Classification (ASAS-SEC). All the requests intended to use the patient data stored on cloud must have to pass through ASAS-SEC mechanism. The request classified as benign are only allowed to access the patient data and DDoS requests are passed to Intensive Care Unit (ICU), where the source of the attack is identified and blocked. Publically available NSL-KDD dataset is utilized to evaluate the proposed ASAS-SEC approach and results shows that proposed approach outperforms other state-of-the-art approaches and achieves classification accuracy of 98.86%, F1-Score of 98.3%, and false alarm of 0.017.