Digital Twin Empowered Service Provisioning in Mobile Edge Computing

Project: Research

View graph of relations


Digital twin (DT) has emerged as a breakthrough technology to revolutionize diverse fields, including manufacturing, Internet of Things, autonomous driving, health care, and education. The global DT industry was valued at $6.5 billion in 2021 and is expected toreach $125.7 billion in 2030 at an annual growth rate of 39.48% according to a market research in July, 2022. Empowered by the DT technology, the digital twin of a physical object can be used to keep the historical data, simulate the behaviours, and providepredictive decisions of the object. The study of DT however is in its early stage. With the DT development, there is growing interest in DT-assisted service provisioning in mobile edge computing (MEC) networks. Unfortunately, existing methods and techniques ofservice provisioning in MEC are not applicable to DT-assisted service provisioning in MEC due to lack of continuously monitoring the state information of objects and prediction mechanisms on the behaviours of objects in future. It thus is urgently neededto develop new foundations, theories, algorithms and techniques for DT-assisted service provisioning in MEC networks.This project will address the following challenges: (i) how to build DT modeling to achieve synchronizations between DTs and their objects? (ii) how to predict the behaviours of objects based on their DT data? (iii) how to develop performanceguaranteedalgorithms for DT placements and migrations? and (iv) how to devise efficient online algorithms for delay-sensitive services through predicting the number of DT replicas and the DT replica placements? To this end, we will propose a novel metric to measure the DT states of objects. We will formulate several DT-assisted service provisioning problems with different optimization objectives, including budgetconstrained DT state synchronization, DT placements and migrations, machine learning based prediction mechanisms to predict or simulate behaviours of objects, the number of DT replicas needed and their placements, and fidelity-aware query evaluations. We will develop efficient algorithms and prediction mechanisms for the problems.In summary, this project will develop a suite of novel solutions to DT-assisted service provisioning in MEC networks. It will conduct cutting-edge research on the development of DT technologies for edge computing. It will contribute substantially to deepen ourunderstanding on key technologies and critical challenges of DTs in real-world applications. It will also contribute to developing advanced algorithms and effective prediction mechanisms. This project will tackle challenges, lay theoretical foundations,invent efficient algorithms, and develop DT-enabling technology for service provisioning in MEC.


Project number9043510
Grant typeGRF
StatusNot started
Effective start/end date1/01/24 → …