B. Sengupta and A. Lakshminarayanan, "Fast Verification of Online/Offline Threshold Signatures for 5G IoT," 2024 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS), Guwahati, India, 2024, pp. 1-6, doi: 10.1109/ANTS63515.2024.10898365.
Abstract:
IoT devices are authenticated using cryptographic credentials (e.g., secret keys). These credentials must be protected. In the enterprise world, storing credentials on network-accessed hardware security modules (HSMs) is common practice. With 5G multi-access edge computing (MEC), it is now possible to host HSMs on 5G MEC. A single HSM however is a vulnerable point of failure. We can increase resilience by using threshold cryptography, wherein credentials are never stored in full in any one HSM, but split among several HSMs. 3GPP Release-17 has support for multi-SIM devices, enabling split credentials stored on multiple mobile network providers.Threshold cryptography incurs additional time-cost though (an order of magnitude slower), which will be unacceptable for latency-sensitive applications. Crutchfield et al. introduced an online/offline threshold signature scheme using chameleon hash function, where the signature-generation phase is split into online and offline phases with most of the signature-generation cost delegated to the offline phase. This speeds up the online phase, making signature generation much faster.However, this speedup comes at a cost of an additional chameleon hash evaluation during signature verification. In this work, we split the signature-verification phase also into online and offline phases. In the offline signature-verification phase, a verifier precomputes (and stores) certain results, such that evaluating the hash function in the online signature-verification phase requires significantly less computation. To be precise, this technique saves around two-third of the time required for hash evaluation during the signature-verification phase. Experimental results validate our analysis.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the National Research Foundation - Industry Alignment Fund - Pre-Positioning: 5G-AMSUS
Grant Reference no. : A20F8a0044