A Secure Body Sensor Network Architecture with CP-ABE Based Fine-Grained Data Access Control

Abstract

Body Sensor Network (BSN) is a network of wirelessly connected physiological sensors. BSNs offer promising innovative medical and non-medical applications. In such applications and networks, security is essential, critical, and challenging. In this context, data privacy and data security are major requirements. The ultimate goal of data privacy is to implement an efficient fine-grained data access control. Ciphertext-Policy Attribute-Based Encryption (CP-ABE) schemes offer a vital option for achieving fine-grained data access control in BSNs. In this work, we propose a secure cluster-based BSN architecture optimized for an energy-efficient operation. The new architecture is designed to allow flexible operation by 1) implementing CP-ABE either on sensor nodes or on the BSN controller, and 2) targeting specific performance merits through proposed enhancement approaches. Three CP-ABE schemes are investigated, the original CP-ABE (BSW), an enhanced CP-ABE (ECP- ABE), and a CP-ABE scheme with constant ciphertext length (CCP-ABE). Thorough simulations are conducted on the MICAz sensor node platform. The performance evaluation, for typical BSN applications, shows that ECP-ABE outperforms BSW. Interestingly, CCP-ABE provides only a marginal advantage in cluster lifetime compared to ECP-ABE defeating the purpose of sacrificing flexible access structure for constant ciphertext length.