A Scalable IoT Protocol via an Efficient DAG-Based Distributed Ledger Consensus

Bumho Son, Jaewook Lee, Huisu JangSustainability 2020 · DOI

TL;DR

This paper introduces a novel, lightweight distributed ledger update algorithm for IoT systems, leveraging Bayesian inference to improve upon IOTA's Tangle. The proposed method offers faster computation times and maintains similar security to existing MCMC-based approaches, addressing scalability and centralization concerns in IoT blockchain adoption.

What Problem Does This Paper Address?

The Internet of Things (IoT) is rapidly expanding, but its inherent heterogeneity and massive data generation create significant security vulnerabilities. While blockchain technology offers a promising solution for enhancing IoT security and reliability, practical limitations, particularly in scalability, hinder its widespread adoption. Existing blockchain solutions struggle with the high transaction volume generated by IoT devices. IOTA's Tangle, a Directed Acyclic Graph (DAG)-based distributed ledger, attempts to address this by eliminating transaction fees and offering better scalability. However, IOTA's current implementation relies on a centralized "coordinator" node for security and its Markov Chain Monte Carlo (MCMC) tip selection algorithm faces challenges with efficiency and potential centralization. This paper aims to tackle these issues by proposing a more efficient and decentralized tip selection algorithm for DAG-based distributed ledgers in IoT environments.

Methods

The core of this paper is a new distributed ledger update algorithm that utilizes Bayesian inference to select transaction "tips" in a DAG structure. Unlike IOTA's MCMC algorithm, which considers a broader range of transactions and relies on a coordinator, this proposed algorithm focuses on a "subtangle" – the recent history of confirmed transactions relevant to the current node.

Here's a breakdown of the approach:

  • Bayesian Inference: The algorithm models the probability of a node being malicious using Bayes' rule. It updates the "posterior distribution" (the probability of a node being malicious given new evidence) based on a "prior distribution" (initial belief) and a "likelihood distribution" (how likely the new evidence is given a hypothesis).
  • Subtangle Focus: Instead of processing the entire Tangle, each node only considers its local subtangle. This significantly reduces computational overhead.
  • Tip Selection: The algorithm probabilistically selects tips (transactions to approve) based on the updated posterior distribution. It aims to favor nodes that are likely to be honest and avoid those exhibiting malicious behavior.
  • Likelihood Distribution Categories: The algorithm categorizes nodes into three sets (A, B, and C) based on their presence in the prior distribution and the current subtangle. Different likelihood distributions are defined for each set, considering factors like cumulative weight and changes in cumulative weight to detect potential malicious activity.
  • Decentralization: By removing the reliance on a central coordinator and focusing on local subtangles, the algorithm promotes a more decentralized network.

Datasets

Not discussed in the paper. The paper focuses on a theoretical proposal and simulation-based evaluation rather than training on specific datasets.

Key Insights and Ideas

  • Bayesian Inference for Malicious Node Detection: Applying Bayesian inference to estimate the probability of a node being malicious is a novel approach for securing DAG-based ledgers. It allows for a more nuanced assessment of node trustworthiness than simply relying on cumulative weight.
  • Subtangle Efficiency: Restricting the analysis to the subtangle is a key innovation for improving scalability. This drastically reduces the computational burden on individual IoT nodes, which often have limited resources.
  • Decentralization through Local Analysis: The proposed algorithm's reliance on local subtangles and Bayesian inference inherently reduces the need for a central coordinator, addressing a major drawback of current IOTA implementations.
  • Dynamic Maliciousness Assessment: The algorithm's ability to detect maliciousness not just by cumulative weight but also by changes in cumulative weight is crucial. This allows it to identify sudden shifts in behavior, which are characteristic of certain attack strategies.

Key Results

The paper presents simulation results that demonstrate the effectiveness of the proposed algorithm:

  • Faster Computation Time: Figure 3 shows that the proposed algorithm has a significantly shorter elapsed time per iteration compared to the MCMC algorithm. This is attributed to its focus on the subtangle rather than the entire tangle.
  • Similar Security: The authors claim the proposed algorithm provides the same resistance to possible attacks as the MCMC algorithm, due to its ability to minimize the probability of selecting malicious nodes.
  • Scalability: The focus on subtangles and reduced computation directly contributes to improved scalability, making it more suitable for resource-constrained IoT environments.
  • Cumulative Weight Behavior: Figures 1 and 2 show that cumulative weights increase linearly, and the number of tips remains relatively stable, similar to existing methods, indicating predictable behavior under normal conditions.

Limitations and Future Work

  • Centralization Concerns: While the proposed algorithm aims to reduce centralization, the paper acknowledges that the current IOTA system's stability relies on a coordinator node. The proposed algorithm's ability to fully eliminate this reliance in practice requires further investigation.
  • Economic Incentives: The paper mentions that the economic incentive for choosing the algorithm is insufficient. This aspect is not fully addressed in the proposed solution and remains an area for future research.
  • Further Research: The authors suggest that further work is needed to study the stability of the system and to develop a consensus algorithm suitable for this ledger expansion method. They also propose exploring its application in other scenarios, such as fast and efficient storage for image data streams (e.g., CCTV).

Summary updated

Abstract

The Internet of Things (IoT) suffers from various security vulnerabilities. The use of blockchain technology can help resolve these vulnerabilities, but some practical problems in terms of scalability continue to hinder the adaption of blockchain for application in the IoT. The directed acyclic graph (DAG)-based Tangle model proposed by the IOTA Foundation aims to avoid transaction fees by employing a different protocol from that used in the blockchain. This model uses the Markov chain Monte Carlo (MCMC) algorithm to update a distributed ledger. However, concerns about centralization by the coordinator nodes remain. Additionally, the economic incentive to choose the algorithm is insufficient. The present study proposes a light and efficient distributed ledger update algorithm that regards only the subtangle of each step by considering the Bayesian inference. Experimental results have confirmed that the performance of the proposed methodology is similar to that of the existing methodology, and the proposed methodology enables a faster computation time. It also provides the same resistance to possible attacks, and for the same reasons, as does the MCMC algorithm.
BibTeX
@inproceedings{son2020scalable,
  author = {Bumho Son and Jaewook Lee and Huisu Jang},
  title = {{A Scalable IoT Protocol via an Efficient DAG-Based Distributed Ledger Consensus}},
  booktitle = {Sustainability},
  year = {2020},
  doi = {10.3390/su12041529},
}
Figure