Answers are grounded in this paper and cite their sources.
Internet of Underwater Things (IoUT) deployments need secure communication over acoustic channels characterized by limited bandwidth, high propagation delay, packet loss and intermittent connectivity. This study presents a post-quantum-oriented blockchain authentication framework for hierarchical underwater networks, using a study-specific NTRU-743 configuration for confidentiality, SHA-256 for integrity, HMAC-SHA-256 for authentication tags and HKDF-SHA-256 for key derivation, so one hash primitive underlies every construction, and an Ethereum-compatible permissioned ledger with Proof-of-Authority validators at the buoy, satellite and base-station layers. Underwater nodes hold no ledger; they submit authenticated records through opportunistic synchronization with their parent gateways. BAN logic and Scyther verify authentication, freshness, nonce secrecy and session-key secrecy under stated assumptions, with the SPDL model and settings released so the results can be reproduced. All performance figures come from an executable simulator released with the paper, in which cryptographic costs are measured and the channel, mobility, availability and validator planes are simulated; energy is a stated model output. Per authentication the protocol carries 1.41 KB of control-field payload, expanding to 8.56 KB on the acoustic link through NTRU ciphertext expansion, plus 1.38 KB validator and 1.18 KB synchronization traffic, totaling 11.13 KB. Processing takes 7.33 ms (95% CI 7.30–7.37) and 4.03 mJ, while acoustic transmission of the same exchange costs 8.42 J: the dominant cost is airtime, not computation. Over 100–1,000 nodes the design scales linearly, cluster utilization holding at 9.13% and validator utilization rising only from 0.08% to 0.80%. Replay detection attains 98.7% recall at a 0.00% false-positive rate against 86.6% and 7.95% for a certificate baseline re-implemented under identical conditions. Comparisons with figures published by prior schemes are descriptive, since those systems were not re-executed.
No takes yet. Share an insight, caveat, or question.
Rupa et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: