Introduction
The Metaverse is increasingly reliant on Digital Twins (DTs)—virtual models that mirror physical entities in real-time. For these DTs to be effective in critical sectors like manufacturing or healthcare, the data streaming from IoT sensors must be immutable and verifiable. While blockchain technology offers a solution for data integrity, its traditional "proactive" consensus mechanisms are often too resource-intensive for power-constrained IoT devices. This paper, originally presented at ICDCS 2023, introduces 2LDAG, a two-layer directed acyclic graph architecture, and Proof-of-Path (PoP), a reactive consensus protocol designed to bridge the gap between IoT constraints and blockchain security.
The Challenge: Heavy Chains for Light Devices
Standard blockchain implementations require nodes to participate in global consensus, necessitating significant storage (often hundreds of gigabytes) and high bandwidth. IoT devices, characterized by limited memory and battery life, cannot maintain a full copy of a ledger or verify every transaction across a network. Existing "lightweight" alternatives often compromise on decentralization or security. The core challenge lies in providing the security guarantees of a blockchain without the overhead of proactive, global verification.
Proposed Architecture: 2LDAG
The authors propose a decoupled architecture consisting of two distinct layers:
- Physical Layer: Individual IoT nodes maintain local storage. Instead of storing the entire network's history, a node only stores its own generated data blocks and a list of its immediate physical neighbors.
- Logical Layer: Data blocks are organized into a Directed Acyclic Graph (DAG). When a node creates a new block, it embeds the cryptographic "digests" (hashes) of the most recent blocks from its neighbors. This creates a distributed web of pointers, where trust is established through the interconnectedness of the graph rather than a single linear chain.
Reactive Consensus: Proof-of-Path (PoP)
The defining innovation of this research is the shift from proactive to reactive consensus. In traditional systems, every node verifies every block immediately. In the PoP protocol, verification is triggered only when a user or a Digital Twin needs to audit a specific data point.
To verify a block, a validator must identify a path of $\gamma + 1$ nodes (where $\gamma$ represents the maximum number of tolerable malicious nodes) that point back to the target block. This "path of trust" ensures that even if some nodes are compromised, the integrity of the data can be mathematically confirmed.
To optimize this process, the authors developed two key algorithms:
- Weighted Path Selection (WPS): This algorithm navigates the logical DAG to find the most efficient verification path, specifically avoiding "micro-loops" caused by high-frequency data generators.
- Trust Path Selection (TPS): A caching mechanism that allows nodes to store previously verified headers, significantly accelerating subsequent audits.
Performance and Results
The protocol was evaluated in a simulated environment of 50 wireless IoT nodes and compared against Practical Byzantine Fault Tolerance (PBFT) and IOTA.
- Storage Efficiency: 2LDAG demonstrated storage overhead two orders of magnitude lower than the benchmarks. Because nodes only store their own data, storage requirements grow locally rather than globally.
- Communication Overhead: The system achieved a reduction in communication costs by three orders of magnitude. In scenarios with low audit frequency, the overhead was nearly zero, as the reactive protocol avoids unnecessary background chatter.
- Security and Scalability: The system remains resilient even when up to 49% of nodes are malicious. Furthermore, because block generation is independent and not gated by global consensus, the system maintains high throughput suitable for high-velocity IoT streams.
Conclusion and Future Work
The 2LDAG and PoP protocol represent a significant step toward making the Metaverse's data infrastructure sustainable for IoT. By moving to an on-demand verification model, the system preserves the cryptographic rigor of blockchain while drastically reducing the resource burden on edge devices.
The authors acknowledge that future work is needed to address network "churn" (nodes frequently joining or leaving) and to test the protocol on physical low-power hardware, such as ESP32 or ARM Cortex-M devices, to further validate its real-world viability.
Summary updated