Source-linked AI summary
When Internet of Things Meets Blockchain: Challenges in Distributed Consensus
Bin Cao, Yixin Li, Lei Zhang, Long Zhang, Shahid Mumtaz, Zhenyu Zhou, Mugen Peng
TL;DR
IoT needs decentralized mechanisms that address trust, security, cost, and scalability concerns, but traditional blockchain consensus mechanisms have important limitations. This paper reviews PoW, PoS, Tangle, and Hashgraph, compares their suitability for IoT, and examines DAG consensus under varying transaction arrival rates. It concludes that DAG mechanisms are more suitable for large-scale IoT, while highlighting unresolved centralization and traffic-dependent confirmation-delay challenges.
Problem
IoT requires consensus mechanisms that address trust, security, maintenance-cost, and scalability concerns within resource-constrained, large-scale systems.
Method
The paper introduces blockchain consensus, reviews PoW, PoS, Tangle, and Hashgraph, compares their IoT characteristics, and simulates DAG confirmation under different transaction arrival rates.
Results
DAG-based consensus has lower transaction fees and resource consumption and higher transaction throughput than PoW and PoS for large-scale IoT.
Takeaways & Limitations
DAG consensus is presented as more suitable for large-scale IoT, but its remaining issues require further research before broader application.
Takeaways & Limitations
DAG confirmation delay becomes very large at low transaction arrival rates, while coordinator assistance reduces delay but introduces centralization and is limited to private or consortium blockchains.
Abstract
from arXiv · showhide
Blockchain has been regarded as a promising technology for Internet of Things (IoT), since it provides significant solutions for decentralized network which can address trust and security concerns, high maintenance cost problem, etc. The decentralization provided by blockchain can be largely attributed to the use of consensus mechanism, which enables peer-to-peer trading in a distributed manner without the involvement of any third party. This article starts from introducing the basic concept of blockchain and illustrating why consensus mechanism plays an indispensable role in a blockchain enabled IoT system. Then, we discuss the main ideas of two famous consensus mechanisms including Proof of Work (PoW) and Proof of Stake (PoS), and list their limitations in IoT. Next, two mainstream Direct Acyclic Graph (DAG) based consensus mechanisms, i.e., the Tangle and Hashgraph, are reviewed to show why DAG consensus is more suitable for IoT system than PoW and PoS. Potential issues and challenges of DAG based consensus mechanism to be addressed in the future are discussed in the last.
I. INTRODUCTION
IoT offers broad opportunities but faces trust, security, maintenance-cost, and scalability problems in centralized systems. Blockchain is presented as a decentralized approach whose consensus mechanism supports trustworthy coordination.
- IoT is described as a disruptive technology with applications across daily activities, business models, products, services, and research.
- Centralized IoT cloud systems can enable provider control over devices and hinder trust between different business agencies.
- Centralized IoT data centers are vulnerable to DDoS attacks, potentially affecting all services because of their centralized topology.
- Updating software across millions of IoT devices creates high maintenance costs in the centralized model.
- Poor scalability in centralized topologies can produce large delays when massive IoT connections generate surging service requests.
- Blockchain is a peer-to-peer distributed ledger that uses consensus to establish trust and agreement, while encryption supports security.
B. Advantages of Blockchain for IoT
Blockchain is presented as a way to reduce centralized IoT weaknesses and support secure, auditable cooperation and autonomous transactions. The paper connects these benefits to applications including supply chains, healthcare, and energy trading.
- Blockchain-based decentralization can reduce hot-spot burden and the probability of a single point of failure in IoT systems.
- Consensus mechanisms and encryption algorithms can strengthen IoT security, while smart contracts can support autonomous device trading and actions.
- A public distributed ledger can provide an auditable platform for cooperation among IoT businesses.
- In supply chains, blockchain records can log container deliveries so entities can track movement, delays, and missing assets.
- Consortium blockchain is used in a localized peer-to-peer electricity-trading framework to improve transaction security without relying on a third party.
- A blockchain-enabled IoT workflow broadcasts transactions, validates them, elects a block producer, and distributes identical ledger copies.
II. CONSENSUS MECHANISM IN BLOCKCHAIN
Consensus mechanisms address trust in blockchain by determining who may insert the next block and by distributing transaction agreement without a third party. The paper evaluates whether different designs meet IoT requirements.
- The section compares consensus designs against the needs and design criteria of IoT systems.
- Consensus mechanisms determine which participant has the right to insert the next block into the blockchain.
- Consensus enables information to be announced orderly to all users without third-party involvement.
- Proof of Work and Proof of Stake are identified as the most widely used consensus mechanisms.
A. Blockchain 1.0 : Proof of Work
Proof of Work selects block producers through computing-power competition, whereas Proof of Stake uses coin age. The paper argues that both designs fit IoT poorly because of resource demands and concentration risks.
- Proof of Work miners compete through hashing, and the first to find a hash below the announced target can generate the new block and receive bonuses.
- High Proof of Work difficulty prevents forking but causes deteriorated and meaningless energy consumption.
- Proof of Work is considered unsuitable for IoT because IoT devices have very limited available resources.
- Proof of Stake uses coin age, calculated as transaction-output value multiplied by the time since creation, to determine block-winning probability.
- Because winning probability follows coin age, Proof of Stake may favor wealthy miners, producing oligopolies or near-monopolies and powerful third parties.
C. Limitations of PoW and PoS for IoT
PoW and PoS use single-chain consensus that creates resource, fee, throughput, and confirmation-delay bottlenecks for IoT. DAG consensus is proposed to address these shortcomings by allowing transactions or blocks to be added while processing earlier transactions.
- Single-chain PoW and PoS slow new-block access to prevent forking, creating bottlenecks for IoT systems.
- Traditional consensus consumes scarce IoT resources, including PoW computing power and PoS coin age.
- Transaction fees burden IoT micropayments because traditional consensus uses fees to compensate miners.
- Bitcoin provides 7 TPS and Ethereum 20 to 30 TPS, limiting response to exponential IoT-device growth.
- DAG consensus permits users to insert blocks at any time after processing earlier transactions, generating simultaneous branches.
III. TYPICAL DAG BASED CONSENSUS
The Tangle and Hashgraph are presented as two typical DAG-based consensus mechanisms. Tangle records transactions in a DAG, where users approve earlier tips and cumulative weight determines transaction validity.
- A. The Tangle: Tangle is a DAG-based distributed ledger and the mathematical foundation of IOTA, a cryptocurrency for the IoT industry.
- A. The Tangle: Tangle branches can eventually merge, increasing overall throughput, while each new vertex typically approves two childless tips.
- A. The Tangle: Higher arrival rates of new transactions can accelerate confirmation because new vertices approve earlier transactions.
- A. The Tangle: A user creates and signs a transaction vertex, selects two non-conflicting tips using MCMC, solves a low-difficulty puzzle, and broadcasts it.
- A. The Tangle: Tangle selects branches with the largest overall cumulative weight through MCMC tip selection, analogous to choosing the longest chain.
B. Hashgraph
Hashgraph combines gossip-based information dissemination with virtual voting to establish transaction order and reach eventual consensus without PoW. Its voting process elects famous witnesses from graph-connected events.
- B. Hashgraph: Hashgraph targets Byzantine fault tolerance, asynchrony, decentralization, eventual consensus with probability one, and high consensus speed without PoW.
- B. Hashgraph: Gossip protocol has each user randomly announce known transactions to another user at fixed intervals, eventually spreading every event.
- B. Hashgraph: Gossip exchanges have low overhead, comprising positional information, a 64-byte signature, and transactions of about 100 bytes.
- B. Hashgraph: Virtual voting selects famous witnesses from first events in each round by voting and checking based on graph connectivity.
C. Comparisons
The article reports that DAG-based consensus is more suitable for large-scale IoT than PoW and PoS, but identifies centralization concerns and traffic-sensitive confirmation delays as unresolved issues.
- C. Comparisons: Comparisons in Table I evaluate DAG-based consensus against PoW and PoS to demonstrate advantages and limitations for IoT.
- C. Comparisons: DAG-based consensus has lower transaction fees and resource consumption, while achieving much higher transaction throughput than PoW and PoS.
- C. Comparisons: DAG-based consensus remains more suitable for large-scale IoT than PoW and PoS according to the reported comparisons.
- C. Comparisons: Remaining issues include centralization concerns in Tangle and confirmation delays that vary significantly with traffic load.
- C. Comparisons: These issues must be addressed before DAG-based consensus can be applied broadly in IoT systems.
A. Analysis Model
The paper proposes a Markov-chain model for analyzing DAG consensus using cumulative weight as the confirmation criterion, while identifying challenges in low-traffic regimes and large state spaces.
- Analysis approach: A Markov-chain model uses cumulative weight as the confirmation criterion to analyze transition probabilities, cumulative-weight growth, TPS, and confirmation delay.The model examines N-step transitions from the current state to a finality state.
- Model limitations: The Markov-chain model remains difficult to scale because its transition-probability matrix is hard to capture in large networks with many system states.Transition probabilities also differ according to the consensus design, including between Tangle and Hashgraph.
- Low-traffic behavior: Low transaction arrival rates increase confirmation delay because cumulative weight grows slowly.The simulation links low arrival rate to slow cumulative-weight growth and large confirmation delay.
- Low-traffic behavior: A coordinator issuing zero-value transactions accelerates cumulative-weight growth and can reduce confirmation delay under low traffic.The coordinator is a third-party-controlled entity introduced to process unconfirmed transactions.
- Low-traffic behavior: Coordinator assistance may introduce centralization, so the approach is limited to private or closed settings such as consortium blockchains.The paper identifies the coordinator’s third-party status as conflicting with blockchain’s decentralized design.
C. Mobile Blockchain
Wireless communication is a potential bottleneck for blockchain-enabled IoT because existing consensus studies often assume wired or perfect communication conditions.
- Mobile Blockchain: Wireless-channel fluctuations can constrain blockchain-enabled IoT consensus processes and make communication a system bottleneck.The paper frames this challenge across multiple communication layers.
- Mobile Blockchain: Physical-layer analysis should relate SINR and communication throughput to IoT deployment, protocol design, and confirmation delay.Relevant design factors include node distribution, block size, and transaction frequency.
2) Upper layer:
Upper-layer blockchain strategies must accommodate constrained IoT resources and distributed verification, while the paper identifies DAG consensus as a basis for further optimization.
- Upper layer:: DAG consensus lets each participant verify, store, and update the ledger in a distributed manner.This distributed role allocation differs from relying on a separate centralized verifier.
- Upper layer:: Because IoT devices have limited power and memory, energy-saving and caching strategies should balance each participant’s workload.The paper suggests allowing resource-limited devices to issue transactions while resourceful devices process them and generate the blockchain.
- Upper layer:: Incentive mechanisms are needed to motivate suitable IoT devices to participate in consensus despite selfish and rational behavior.The paper identifies game theory as a way to model distributed strategy selection and cites auction-based PoW offloading as an example.
- Upper layer:: The paper’s broader review compares PoW, PoS, and DAG consensus, simulates transaction-arrival effects, and summarizes DAG challenges for IoT deployment.The identified challenge areas include analytical modeling, major drawbacks, mobile blockchain, and optimization strategy.