Blockchain-Based Communication for Digital Twins

• Blockchain technology can safely and reliably track the creation process of digital twins. To ensure the data security in the case of untrusted multiple parties sharing data and to improve users’ satisfaction with the use of digital assets in daily transactions, in this article, the digital twin technology is combined with the blockchain technology, and for multiple scenarios in blockchain networks, a credible and efficient edge computing resource allocation method is proposed based on deep reinforcement learning (DRL) theory. The log storage system of the blockchain is mainly used as the interface for writing and reading data operation logs, and a hybrid storage strategy is put forward for the log storage system. As for the trusted resource allocation under the decentralized model of the blockchain network, to prevent the system from directly offloading the computing tasks submitted by the user each time to the edge server, first, the task is submitted to the system, and then a more reasonable resource allocation strategy is implemented with the user satisfaction as the standard. The simulation experiment results show that the improved environment after the introduction of the log storage system reduces the storage overhead by about 75% compared with the blockchain benchmark environment. When the number of servers is greater than 5, compared with other resource allocation algorithms, user satisfaction of the resource allocation algorithm based on DRL theory has been significantly improved, and compared with Q-learning algorithm, its user satisfaction has increased by 15%. In conclusion, the digital twins use credible source data as input. In this process, the blockchain ensures the security of data management, and finally, the data analysis is performed to predict events and evaluate related factors. The resource allocation method based on DRL realizes credible resource utilization based on recorded data on the blockchain.

Ämnesord

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

Nyckelord

Blockchain

Data security

Digital twins

Edge computing

Resource allocation

Computation theory

Deep learning

Digital storage

Engineering education

Information management

Learning algorithms

Reinforcement learning

Allocation methods

Block-chain

Creation process

Log storages

Reinforcement Learning theories

Resource allocation algorithms

Resources allocation

Storage systems

Users' satisfactions

Security of data

Publikations- och innehållstyp

ref (ämneskategori)

kap (ämneskategori)