The mechanism employed in a blockchain network for data trading is done through a block of code which is known as a Smart Contract. The use of a smart contract allows the transacting parties to set their terms of settlement rights before the execution of the trade, and as soon as the trading parties come to a consensus and input their digital signatures, the smart contract is executed automatically. A major advantage of employing the use of a Smart Contract for data trading between the buyer and seller is that it removes the need for the interference of a third party and can speed up the process of settlement by removing the after-trade infrastructure that usually fragments and allowing for the implementation of a flexible cycle of settlement. Furthermore, blockchain technology offers an opportunity for the trading parties to speed up the process of trading, necessary for time-limited transactions.
Simulation provided a means to examine how data trading will take place between parties and depending on different seller rights scenarios. The focus was on agent-based simulation, and results obtained from the simulation provide an instantiation of data trading using a smart contract with protected seller's rights. Data trading based on the seller rights was a new concept that enabled the seller to protect his data from being sold to any buyer but only those who meet the terms and conditions defined in the smart contract.
The figure below presents the interaction between the market simulation and the trading simulation. In this relatively simple experiment, the marker dictated the number of randomly generated buy and sell requests. The framework also includes buyer types to support different types of buyers and sellers. The propensity to trade was generated from either a connected simulation or machine-learning generated synthetic data.
The exchange contained buyers and sellers and their respective preferences, which were then routed to a matching algorithm with the help of a transaction agent. The algorithm then determined the matching characteristics, and the transaction was subsequently routed on an approved or denied direction. Trade confirmation was based on the algorithm’s decision depending on the terms and conditions. The model will execute the smart contract functionality based on random values of parameters and produce the output as a successful data trading transaction or deny the transaction depending on the match found.
Conceptual hybrid data trading model
A novel data trading approach was presented in this paper – one where trading was controlled by seller preferences. The approach followed the principles of seller’s rights protection and control over the data sharing available in a community of users.
A data trading platform was developed in order to understand the core elements of the trading process and subsequently calibrate the trading model. Hybrid simulation in AnyLogic was then used to explore and test trading behavior in different market environments, with four simple scenarios presented. The Process Modeling library was chosen to define the data trading transaction with inputs applied for testing the model functionality.
A hybrid approach was shown to combine market and technology simulations and enable system developers to test robust future scenarios. Each transaction had been simulated and tested individually and the behavior of every transaction was recorded. The preferences of each entity involved in the transaction are compared and tested for similarity between seller and buyer preferences. All transaction metrics were compared to the real system and ideal scenarios to test the algorithm's functionality. The output associated with these transactions gave a clear picture of what the real system might achieve during the data trading process and how the algorithm will respond to buyers and sellers having a range of preferences.