• HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
Friday, November 20, 2026
BIOENGINEER.ORG
No Result
View All Result
  • Login
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
No Result
View All Result
Bioengineer.org
No Result
View All Result
Home NEWS Science News Technology

Hierarchical Blockchain and Zero-Knowledge Architecture Cuts IoT Authentication Latency and Energy

by
November 20, 2026
in Technology
Reading Time: 6 mins read
0
Hierarchical Blockchain and Zero-Knowledge Architecture Cuts IoT Authentication Latency and Energy

Hierarchical Blockchain and Zero-Knowledge Architecture Cuts IoT Authentication Latency and Energy

Share on FacebookShare on TwitterShare on LinkedinShare on RedditShare on Telegram

Connecting millions of microcontrollers to operational infrastructure requires verifying device identities without overwhelming embedded hardware or distributed networks. In a study published in Results in Engineering, Haozhe Zhou, Hang Lei, and Maolin Yang introduced HybridTrust, a three-tier hierarchical architecture that coordinates physical unclonable functions, fog-level batch Bulletproofs, and cloud-level succinct zero-knowledge proofs. Evaluated across a heterogeneous testbed of 16 physical machines and up to 500 emulated containers representing up to 10,000 devices, the system achieved a mean end-to-end authentication latency of 160 to 220 milliseconds for offered loads up to 8,000 authentications per second and 240 milliseconds at 10,000 authentications per second. Across these loads, the architecture demonstrated an authentication throughput of up to 8,500 operations per second while requiring 0.3 millijoules of device-side energy per authentication event, reducing energy use by 92 percent compared to a public-key blockchain baseline and 94 percent compared to a per-device zero-knowledge baseline.

Industrial installations supporting automated manufacturing, smart transportation, and medical telemetry routinely supervise deployments containing between 100,000 and 1,000,000 heterogeneous microcontrollers. These deployed endpoints operate under stringent physical limitations, typically possessing roughly 100 kilobytes of random-access memory, 10 kilobytes of persistent storage, and strict power budgets that preclude intensive cryptographic operations. Conventional centralized systems frequently rely on gateway proxies or certificate authorities, yet standard gateways establish vulnerable centralized points of failure lacking verifiable evidence of authorization. Conversely, migrating authentication directly onto decentralized ledgers confronts what the investigators formalized as a security, scalability, and resource-efficiency trilemma: shared ledgers impose linear verification expenses when evaluating every device transaction independently, overwhelming both on-chain throughput and constrained microcontroller processors.

The engineering team addressed this structural bottleneck by dividing authentication responsibilities across a computational gradient spanning devices, edge fog nodes, and cloud consensus infrastructure. At the hardware foundation, each microcontroller derives dynamic operational secrets from an onboard static random-access memory physical unclonable function paired with a fuzzy extractor using Bose–Chaudhuri–Hocquenghem error-correcting codes. By extracting 512-bit response vectors and generating 384-bit helper data records, the device reconstructs a stable 128-bit secret without retaining persistent private keys in non-volatile flash storage. Because helper data and domain-scoped authentication keys are encrypted and stored across local edge nodes rather than locally on endpoints, extracting physical memory dumps from captured field hardware exposes neither secret keys nor helper syndromes.

To execute an authentication event, an endpoint fetches its helper data, reconstructs its root secret, derives an ephemeral session key and commitment via hash-based key derivation functions, and delivers a keyed message authentication code over a lightweight User Datagram Protocol transport. The neighboring fog node decrypts its domain-scoped verification key, confirms the freshness of the request timestamp against a configurable window, inspects its seen-nonce cache to bar replay attempts, and natively validates the message authentication code. This initial edge verification requires only two key derivation expansions and two cryptographic hash evaluations, ensuring that the microcontroller never executes computationally demanding asymmetric public-key operations or zero-knowledge proof generation routines.

Once individual requests pass initial admission filtering, the fog node organizes up to 64 verified requests into unified batches. Instead of transmitting individual device transactions to the broader network, the fog processor generates a single Bulletproof zero-knowledge argument asserting that all admitted requests correspond to enrolled devices, confirmed absent from sorted Merkle revocation structures, and compliant with current tenant policy versions. Incorporating an algebraic hash function designed for arithmetic circuit efficiency allows the batch proof to demonstrate joint membership and non-revocation without embedding expensive SHA-256 or message authentication code circuits. The edge node then signs a cryptographic attestation over the resulting batch root, records a batch authentication record in its regional Raft-managed ledger for local auditing, and dispatches the bundle upstream.

At the cloud tier, an independent service operating within a hardware-secured infrastructure natively verifies the batch Bulletproofs before admitting them into a global verification pipeline. Rather than embedding non-native Bulletproof arithmetic directly inside a recursive circuit, the cloud layer natively confirms the multi-scalar multiplications of all received batch proofs in one pass. It then constructs a succinct Groth16 argument of knowledge over the admitted fog attestation signatures. This design allows the root blockchain validators, communicating through Tendermint Byzantine fault tolerant consensus, to confirm only one constant-size Groth16 proof per authentication window, thereby maintaining an on-chain verification workload that remains asymptotically constant regardless of total device population or batch counts.

Physical testbed validation combined eight ESP32 development boards, four Raspberry Pi 4B microcomputers, one RISC-V edge board, and three enterprise x86 cloud servers, extended by 480 device-emulation containers hosting up to 21 simulated instances each alongside 20 containerized fog nodes. Network emulation introduced 10 to 40 millisecond one-way wide-area latencies and up to one percent packet loss between edge and cloud layers to mimic realistic wide-area operational networks. Under moderate workloads of 2,000 authentications per second across 5,000 devices, HybridTrust recorded 160 milliseconds mean latency and a 95th-percentile latency of 210 milliseconds. Under heavy workloads of 8,000 authentications per second, mean latency reached 220 milliseconds with a 95th-percentile metric of 320 milliseconds, climbing to 240 milliseconds at 10,000 authentications per second.

When compared alongside architectural alternatives under identical testbed parameters, the hierarchical system established distinct trade-offs against conventional single-layer baselines. A public-key blockchain configuration relying on per-device smart contract validation produced average latencies of 1.2 seconds at moderate load and escalated to 2.6 seconds under heavy traffic, saturating near 1,000 authentications per second. A single-layer zero-knowledge design requiring endpoints to generate local proofs averaged 1.5 seconds at 2,000 authentications per second and saturated at 800 authentications per second. A ledger-less trusted gateway baseline proved 12 to 14 percent faster in raw latency, averaging 140 milliseconds at moderate load, but lacked ledger-backed audit trails and cryptographically verifiable authorization, while consuming 0.04 to 0.05 millijoules more energy per authentication than the proposed architecture.

The researchers also investigated operational robustness against intentional disruptions and node corruption. In fault-injection experiments where the proportion of Byzantine fog nodes reached 32 percent, legitimate endpoints achieved a 96 percent authentication success rate through client-side timeouts and automatic fail-over retries across neighboring domain nodes, while average latency increased by approximately 20 percent. When malicious fog nodes increased to 50 percent, legitimate success rates fell to 87 percent, reflecting availability degradation rather than cryptographic failure. Under simulated denial-of-service and Sybil floods where fraudulent requests matched legitimate traffic at a one-to-one ratio, the system retained an 84 percent authentication success rate with mean latency remaining below 235 milliseconds, as edge nodes eliminated bogus identifiers during initial local checks before proof construction.

Ablation experiments demonstrated the proportional performance contributions of each architectural module under a steady load of 4,000 authentications per second. Removing cloud-side succinct aggregation and forcing validators to process Bulletproofs directly on-chain introduced severe validator bottlenecks, whereas retaining state caching at the edge reduced repeated remote lookups. Selecting a batch capacity of 64 devices balanced mathematical amortization benefits against packet assembly delay, avoiding the excessive collection latencies observed under larger batch sizes.

To assess behavior beyond the 10,000-device experimental threshold, the authors formulated an analytical scaling model factoring endpoint processing, batch generation, cloud aggregation, and wide-area network propagation delays. Assuming linear scaling of independent edge domains and dedicated network provisioning, the calibrated model projected that average authentication latency could remain below one second across populations reaching 1,000,000 devices. However, the authors explicitly noted that this analytical projection has not been validated in a physical hardware deployment of that magnitude. Unmodeled real-world disruptions, including transient wide-area congestion, uneven device density across regions, edge node churn, and packet loss retries, could degrade end-to-end performance under practical operating conditions.

The authors detailed additional structural limitations governing the current implementation. Inherent in the symmetric edge verification protocol is a residual trust assumption: because a home edge node maintains the domain-scoped key of its enrolled endpoints, a compromised fog processor could theoretically generate unauthorized admissions for valid enrolled devices that did not initiate a request, although malicious admissions remain auditable on regional ledgers and cannot admit revoked or unregistered identities. Furthermore, the single-block fuzzy extractor parameters require near-full physical entropy from hardware arrays to preserve 128-bit security, and current error-correction tolerance of 5.9 percent covers nominal ambient temperatures near 25 degrees Celsius rather than extreme industrial fluctuations spanning minus 40 to 85 degrees Celsius. Addressing these operational boundaries, future investigations plan to explore device-side one-time authenticators to eliminate edge key trust, dynamic multi-block error correction across wider thermal ranges, and larger physical hardware validations.

Subject of Research: Hierarchical blockchain and zero-knowledge proof architecture for resource-constrained IoT device authentication

Article Title: HybridTrust: PUF-rooted hierarchical blockchain with layered zero-knowledge proofs for scalable IoT device authentication

Article References: Zhou, H., Lei, H., & Yang, M. (2026). HybridTrust: PUF-rooted hierarchical blockchain with layered zero-knowledge proofs for scalable IoT device authentication. Results in Engineering, 32, Article 113821. https://doi.org/10.1016/j.rineng.2026.113821

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113821

Keywords: Internet of Things, Device Authentication, Hierarchical Blockchain, Physical Unclonable Functions, Zero-Knowledge Proofs, Bulletproofs, Groth16 SNARK, Edge Computing, Fog Computing, Microcontrollers, Fuzzy Extractors, Cryptographic Verification, System Scalability, Energy Efficiency, Byzantine Fault Tolerance, Network Security

News Source: Denise Maddox. (November 20, 2026). Hierarchical Blockchain and Zero-Knowledge Architecture Cuts IoT Authentication Latency and Energy. Scienmag.

Tags: BulletproofsByzantine fault tolerancecryptographic verificationDevice AuthenticationEdge Computingenergy efficiencyfog computingFuzzy ExtractorsGroth16 SNARKHierarchical BlockchainInternet of Thingsmicrocontrollersnetwork securityPhysical Unclonable FunctionsSystem scalabilityzero-knowledge proofs
Share12Tweet7Share2ShareShareShare1

Related Posts

Batch-Invariant Spectral Network Identifies Edible Insect Species Across Variable Processing Conditions

Batch-Invariant Spectral Network Identifies Edible Insect Species Across Variable Processing Conditions

November 20, 2026
Self-Sensing Well Cement Monitors Downhole Integrity and Predicts Failures Using Machine Learning

Self-Sensing Well Cement Monitors Downhole Integrity and Predicts Failures Using Machine Learning

November 20, 2026

Plant Extract Cleans Fish Industry Effluent but Raises Organic Load and Toxicity

November 20, 2026

Minimum Wage Differences Prompt Chinese Companies to Relocate Establishments Across Provinces

November 20, 2026

POPULAR NEWS

  • Pressurized Isopropanol Extracts Antioxidants and Enzyme Inhibitors From Prickly Pear Seeds

    Pressurized Isopropanol Extracts Antioxidants and Enzyme Inhibitors From Prickly Pear Seeds

    29 shares
    Share 12 Tweet 7
  • Integrated Photonic Quantum Converter Enhances Digital Phase Estimation in Metrology

    29 shares
    Share 12 Tweet 7
  • Endoscopic Ultrasound Expands Diagnostic and Guided Interventional Roles in Pancreatic Disease

    29 shares
    Share 12 Tweet 7
  • Full-Length TDP-43 Protein Propagates From Motor Cortex to Muscle in Rats

    29 shares
    Share 12 Tweet 7

About

We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.

Follow us

Recent News

Infiltrating Monocytes Trigger Early Retinal Damage Before Restoring Homeostasis in Mice

Nerve stimulation aids walking nearly a decade after drop foot surgery

Intensified Combination Therapies Improve Outcomes in Advanced EGFR and TP53 Mutated Lung Cancer

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 89 other subscribers
  • Contact Us

Bioengineer.org © Copyright 2023 All Rights Reserved.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Homepages
    • Home Page 1
    • Home Page 2
  • News
  • National
  • Business
  • Health
  • Lifestyle
  • Science

Bioengineer.org © Copyright 2023 All Rights Reserved.