Research
DEXO: A Secure and Fair Exchange Mechanism for Decentralized IoT Data Markets
The paper is the authors' accepted version of an article published in IEEE Internet of Things Journal, vol. 12, no. 11, pp. 16095–16111, 2025 (doi: 10.1109/JIOT.2025.3535671).

- arXiv
- 2511.18498
- Published
- 2025-11-23
- Authors
- Yue Li, Ifteher Alom, Wenhai Sun, Yang Xiao
AI summary
Overview
Research area: Blockchain-enabled data markets, decentralized oracle networks, trusted execution environments (TEE), and cryptographic fair exchange for Internet of Things (IoT) and mobile data.
Technical level: Advanced. The paper builds on ideal-functionality abstractions (smart contracts, attested execution, secret sharing, commitments), threshold security assumptions, and smart contract engineering, and it assumes familiarity with blockchain and applied cryptography.
Scope (one sentence): The paper designs, proves, and prototypes DEXO — a decentralized data exchange mechanism that combines TEE-based secret sharing with smart-contract fair exchange so that IoT/mobile data can be sold confidentially, verifiably, and fairly without a trusted broker.
The paper is the authors' accepted version of an article published in IEEE Internet of Things Journal, vol. 12, no. 11, pp. 16095–16111, 2025 (doi: 10.1109/JIOT.2025.3535671). Authors are Yue Li, Ifteher Alom, Wenhai Sun, and Yang Xiao (University of Kentucky and Purdue University).
What This Paper Is About
IoT and mobile device data are valuable, but selling them usually requires a trusted third party such as a data broker, which is a single point of failure and has repeatedly leaked or resold subscriber data without consent (the paper cites the T-Mobile location-data case). Existing blockchain approaches either place trust in a broker's off-chain server, or use decentralized oracle networks (DONs) that publish plaintext data on-chain and cannot support confidential, large-volume data or fair exchange.
DEXO's goal is a marketplace where a buyer can verify that data really came from an attested device and an agreed processing routine, where data stays confidential end to end (even from the exchange nodes), and where neither buyer nor seller can cheat the other — while keeping most data transfer off-chain for scalability.
Key Contributions
-
A new decentralized exchange mechanism (DEXO). DEXO extends the DON model to a data marketplace and, according to the authors, is the first to accomplish end-to-end data confidentiality, source verifiability, fault tolerance, and fair exchange of off-chain data together.
-
TEE-based confidential and verifiable data supply. Each contributing device runs a DEXO-ratified trusted application inside its TEE, which pre-processes and formats raw data, produces attestation-backed signatures, and splits the data into N secret shares — one per DEXO node — so no single node ever sees plaintext.
-
Smart-contract fair exchange for off-chain delivery. A DEXO contract enforces atomicity: encrypted shares are delivered off-chain while the decryption key is released on-chain via cryptographic commitments, with payments routed to the original data sources and a dispute path for both parties.
-
Proof-of-concept implementation and evaluation. A prototype uses Ethereum as the smart contract platform and ARM TrustZone as the device TEE, and the authors report better on-chain gas cost per unit of data consumed than a popular data exchange mechanism, with moderate off-chain overhead for individual providers.
Main Findings
-
Security holds under a minority-corruption assumption. The analysis assumes F out of N DEXO nodes are compromised, with F < ½N and F < t ≤ N − F. Under these conditions, the authors state that the original data remains confidential to individual DEXO nodes at all times and that the consumer is guaranteed to reconstruct the data.
-
Collusion resistance is bounded by the sharing threshold. A consumer colluding with fewer than t DEXO nodes cannot obtain any portion of the original data without full payment; symmetrically, a P-DApp cannot take payment without providing the full requested data, including when colluding with individual nodes that tamper with shares.
-
The threshold assumption is positioned relative to Chainlink. The paper notes that its F < ½N bound covers the case of existing oracle networks such as Chainlink, where at most 6 out of 21 oracle nodes could be compromised.
-
Comparative positioning against ten schemes. A comparison table scores DEXO against Ocean Protocol, DataBroker DAO, OpenSea, Rarible, FairSwap, OptiSwap, PrivacyGuard, Town Crier, DECO, and Chainlink on five criteria (data source verifiability, end-to-end data confidentiality, decentralization/fault tolerance, fair exchange of off-chain data, mobile/IoT data support). DEXO is the only entry marked as satisfying all five; the authors note that DataBroker DAO is the only compared data-exchange scheme credited with fair exchange, and that none of the compared data-exchange or NFT schemes is credited with source verifiability.
-
Off-chain delivery is the scalability lever. Unlike DON schemes where the entire requested data must be delivered on-chain, DEXO delivers data off-chain with on-chain settlement, which the authors identify as the key to handling large data volumes.
-
Evaluation is reported qualitatively in the available text. The paper states a moderate deployment cost and significantly improved blockchain operation efficiency (gas cost per unit of data consumed) versus a popular data exchange mechanism; specific gas figures, dataset sizes, and off-chain latency numbers are not reported in the excerpt provided here.
Methodology in Plain English
The design splits labor between an on-chain contract and an off-chain network of N nodes.
On the supply side, device owners join a data-providing DApp (a "P-DApp") that has a normal frontend server plus its own smart contract on the blockchain. Each participating device runs an approved program inside its TEE. That program cleans up and formats the raw sensor data, signs it together with proof that the program and the hardware are genuine, and then splits the data into N pieces using Shamir's Secret Sharing, sending one piece to each DEXO node. A single intercepted piece is useless; only t of N pieces reconstruct the data.
On the exchange side, when a buyer requests a data item, each node encrypts its share and publishes a Merkle root hash of the encrypted shares to the contract — a commitment that it cannot later change. The buyer pays into the contract. Nodes then deliver encrypted shares off-chain, and the contract releases the decryption keys only once payment is locked and delivery is verifiable. If something goes wrong, either side can abort or dispute, following the style of the OptiSwap fair-exchange construction — modified so that each node commits to an individual share rather than to the whole dataset.
The paper formalizes the pieces as ideal functionalities (a smart contract functionality, a TEE attested-execution functionality, a secret-sharing functionality, and a single-message commitment functionality), composes them into the full protocol in stages, and then argues security and complexity before building the prototype.
Why This Matters
Impact on research. DEXO argues that DON-style decentralization and confidential data markets can be combined rather than traded off, and it treats data provenance — proving that data came from a genuine device running an approved transformation — as a first-class marketplace property. It also shows how fair-exchange constructions built for single-seller digital goods can be adapted to a multi-node, secret-shared setting.
Real-world applications:
- Mobile network planning and city traffic management, where telecom and municipal buyers need trustworthy device-origin data without exposing subscribers.
- Healthcare analytics on data from personal and wearable devices, where confidentiality from intermediaries is a legal and ethical requirement.
- Training data pipelines for foundational AI models, where the buyer needs verifiable provenance and volume but not plaintext exposure.
- Any sensor-data sale where the seller wants payment guaranteed before disclosure and the buyer wants a refund path if the data does not match the advertised specification.
Industry relevance. The paper speaks directly to oracle providers (it explicitly frames its design as an extension of DON infrastructure), to telecoms and data brokers under scrutiny for unauthorized resale, and to blockchain platforms whose on-chain gas costs constrain data-heavy applications. The Ethereum plus ARM TrustZone prototype indicates the design targets currently deployable hardware rather than hypothetical TEEs.
Future Directions
- Pricing and market design. The paper fixes a price per data item in the P-DApp's contract and explicitly leaves richer mechanisms such as auctioning to future work.
- Data quality beyond format conformance. The authors state that DEXO does not provide broader data-derived quality control (for example, assessing the contextual utility of the data) and leave that judgment to consumers.
- Scaling the node set and trust assumptions. The guarantees depend on F < ½N and on a fixed, pre-determined set of N nodes; how membership, rotation, or misbehavior-driven ejection would interact with the secret-sharing threshold is not addressed in the available text.
- Broader TEE deployment and attestation trust. The design assumes TEEs are available on mobile and IoT devices and that the attestation service is trusted; how it behaves across a wider range of device TEE implementations, or under TEE compromise, is left open.
Target Audience
Researchers and graduate students in blockchain security, applied cryptography, and privacy-preserving data markets; engineers designing oracle networks, decentralized data marketplaces, or TEE-based IoT applications; and technical decision-makers at telecoms, data brokers, or blockchain platforms evaluating alternatives to brokered data sales. Readers need a working grasp of smart contracts, secret sharing, and TEE attestation to follow the protocol details, though the problem framing and comparison table are accessible to a broader technical audience.
Authors’ abstract
Opening up data produced by the Internet of Things (IoT) and mobile devices for public utilization can maximize their economic value. Challenges remain in the trustworthiness of the data sources and the security of the trading process, particularly when there is no trust between the data providers and consumers. In this paper, we propose DEXO, a decentralized data exchange mechanism that facilitates secure and fair data exchange between data consumers and distributed…
The full abstract on arXiv