Provably secure certificateless aggregate signcryption with offline/online decoupling for resource constrained WBANs
摘要
Wireless Body Area Networks (WBANs) form the backbone of the Internet of Medical Things (IoMT), enabling real-time, continuous, and remote monitoring of critical physiological parameters. However, the transmission of highly sensitive health data over open wireless channels, coupled with the extreme energy and computational constraints of 8/16-bit biosensors, demands a cryptographic solution that simultaneously guarantees provable security, ultra-low latency, and minimal resource consumption. To address this challenge, we propose COAS: a provably secure, lightweight offline/online certificateless aggregate signcryption scheme tailored for WBANs. COAS eliminates the certificate management overhead of Public Key Infrastructure (PKI) while resolving the key escrow vulnerability of Identity-Based Cryptography (IBC) through a certificateless framework. Its core innovation is an offline/online decoupling mechanism that shifts over 90% of the cryptographic workload, specifically, all elliptic curve scalar multiplications, to idle preprocessing phases, reducing real-time signcryption latency to just 4.4 ms. At the network layer, COAS employs homomorphic aggregation to compress n individual signcryptexts into a constant-size credential, cutting wide-area network (WAN) bandwidth by 14% and enabling the Medical Server to verify an entire batch in a single step. We formally prove in the Random Oracle Model (ROM) that COAS achieves IND-CCA2 confidentiality under the Computational Diffie-Hellman (CDH) assumption and EUF-CMA unforgeability under the Elliptic Curve Discrete Logarithm Problem (ECDLP). Extensive experiments on a heterogeneous testbed (Raspberry Pi Zero W, Android gateway, and cloud server) demonstrate that COAS reduces computational latency by 42.5%, communication overhead by 14%, and total energy consumption by 38.6% compared to state-of-the-art protocols. By unifying provable security, GDPR-compliant privacy via (t, k)-threshold tracing, and hardware-aware efficiency, COAS establishes a new benchmark for scalable, secure, and deployable IoMT architectures.