A new secure proxy-based distributed virtual machines management in mobile cloud computing

• Main Authors: Annane, Boubakeur, Ghazali, Osman, Alti, Adel
• Format: Article
• Language: English
• Published: 2019
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://repo.uum.edu.my/26451/1/IJACR%209%2043%202019%20222%20231.pdf; http://repo.uum.edu.my/26451/; http://doi.org/10.19101/IJACR.PID10
• Institution: Universiti Utara Malaysia

The mobile cloud computing as an excellent paradigm offers on-demand services, whereas users can be confident once using them. Nevertheless, the existing cloud virtualization systems are not secure enough regarding the mediocre degree of data protection, which avoids individuals and organizations to engage with this technology. Therefore, the security of sensitive data may be affected when mobile users move it out to the cloud exactly during the processing in virtual machines (VMs). Many studies show that sensitive data of legitimate users’ VMs may be the target of malicious users, which lead to violating VMs’ confidentiality and privacy. The current approaches offer various solutions for this security issue. However, they are suffering from many inconveniences such as unauthorized distributed VM access behavior and robust strategies that ensure strong protection of communication of sensitive data among distributed VMs. The purpose of this paper is to present a new security proxy-based approach that contains three policies based on secured hashed DiffieHellman keys for user access control and VM deployment and communication control management in order to defend against three well-known attacks on the mobile cloud environment (co-resident attacks, hypervisor attacks and distributed attacks). The related attacks lead to unauthorized access to sensitive data between different distributed mobile applications while using the cloud as a third party for sharing resources. The proposed approach is illustrated using a healthcare case study. Including the experimental results that show interesting high-efficiency protection and accurate attacks identification.