Enforce trust in data collected by oracles and exploit it for traceability.
NGI-ONTOCHAIN:PROJECT-2
OntochainPrj2
As deployment of multiple sensors for a single measurement comes with an increasing cost, we propose to deploy multiple sensors over what we define as correlation area. The idea is that off-chain measured data are correlated according to a certain statistical correlation model. Therefore, once the correlation model has been inferred, a value claimed by a sensor can be trusted if the correlation computed from other sensors is above a certain level. To enforce fault tolerance and defence against byzantine nodes, we propose to query a randomly selected subset of the available nodes in the correlation area. As random selection is introduced, the only way an attacker can control data inserted in the smart contract is to control the majority of sensors. A multi-signature scheme is hence introduced, where oracles sign their contribution if a certain correlation level is obtained. We also propose to reward proper behaviour of sensor nodes with assets that they are allowed to spend.
The proposed solution enhances the automation level in the ONTOCHAIN ecosystem, as it allows to eliminate human interaction for those tasks for which a user is not motivated to contribute (see e.g., verify a temperature value.
https://ontochain.ngi.eu/content/dart
Not available yet
Not available yet
Country: Italy
NGI Project: Ontochain
Status: Early research demo
Category: Decentralized solutions (including blockchain and distributed ledger technologies)