The digital twin(DT)is envisaged as a catalyst for pioneering ecosystems of service provision within an immersive environment born from the convergence of virtual and physical realms.Specifically,DT could enhance the ...The digital twin(DT)is envisaged as a catalyst for pioneering ecosystems of service provision within an immersive environment born from the convergence of virtual and physical realms.Specifically,DT could enhance the performance of edge-intelligent connected vehicular networks by allocating network resources efficiently based on the key performance indicators(KPIs)of vehicular data traffic.Consequently,this work addresses the key challenge of computation and spectrum resource allocation for vehicular networks.To allocate the optimal resource allocation,we subdivided the problem into:traffic classification,collective learning,and resource allocation scheme.In order to do so,this paper concentrates on two crucial vehicular applications:brake application and lane-change application.We utilize a random forest model to collectively learn vehicular data traffic in the upcoming time slot.Thereafter,a time-ahead resource allocation algorithm is proposed to improve the quality of service(QoS)by intelligently offloading vehicular data traffic to a DT-based integrated fiber-wireless(Fi-Wi)connected vehicular network.We evaluate the performance of the resource allocation strategy in terms of resources required by the network alongside the packet loss rate.It was observed that there was a 44.74%increase in cost as the total computation resources increased from F=50 to 100 GHz,whereas the PLR of the network decreased by 71.43%.展开更多
Digital twins for wide-areas(DT-WA)can model and predict the physical world with high fidelity by incorporating an artificial intelligence(AI)model.However,the AI model requires an energy-consuming updating process to...Digital twins for wide-areas(DT-WA)can model and predict the physical world with high fidelity by incorporating an artificial intelligence(AI)model.However,the AI model requires an energy-consuming updating process to keep pace with the dynamic environment,where studies are still in infancy.To reduce the updating energy,this paper proposes a distributed edge cooperation and data collection scheme.The AI model is partitioned into multiple sub-models deployed on different edge servers(ESs)co-located with access points across wide-area,to update distributively using local sensor data.To reduce the updating energy,ESs can choose to become either updating helpers or recipients of their neighboring ESs,based on sensor quantities and basic updating convergencies.Helpers would share their updated sub-model parameters with neighboring recipients,so as to reduce the latter updating workload.To minimize system energy under updating convergency and latency constraints,we further propose an algorithm to let ESs distributively optimize their cooperation identities,collect sensor data,and allocate wireless and computing resources.It comprises several constraint-release approaches,where two child optimization problems are solved,and designs a largescale multi-agent deep reinforcement learning algorithm.Simulation shows that the proposed scheme can efficiently reduce updating energy compared with the baselines.展开更多
文摘The digital twin(DT)is envisaged as a catalyst for pioneering ecosystems of service provision within an immersive environment born from the convergence of virtual and physical realms.Specifically,DT could enhance the performance of edge-intelligent connected vehicular networks by allocating network resources efficiently based on the key performance indicators(KPIs)of vehicular data traffic.Consequently,this work addresses the key challenge of computation and spectrum resource allocation for vehicular networks.To allocate the optimal resource allocation,we subdivided the problem into:traffic classification,collective learning,and resource allocation scheme.In order to do so,this paper concentrates on two crucial vehicular applications:brake application and lane-change application.We utilize a random forest model to collectively learn vehicular data traffic in the upcoming time slot.Thereafter,a time-ahead resource allocation algorithm is proposed to improve the quality of service(QoS)by intelligently offloading vehicular data traffic to a DT-based integrated fiber-wireless(Fi-Wi)connected vehicular network.We evaluate the performance of the resource allocation strategy in terms of resources required by the network alongside the packet loss rate.It was observed that there was a 44.74%increase in cost as the total computation resources increased from F=50 to 100 GHz,whereas the PLR of the network decreased by 71.43%.
基金supported by National Key Research and Development Program of China(2020YFB1807900).
文摘Digital twins for wide-areas(DT-WA)can model and predict the physical world with high fidelity by incorporating an artificial intelligence(AI)model.However,the AI model requires an energy-consuming updating process to keep pace with the dynamic environment,where studies are still in infancy.To reduce the updating energy,this paper proposes a distributed edge cooperation and data collection scheme.The AI model is partitioned into multiple sub-models deployed on different edge servers(ESs)co-located with access points across wide-area,to update distributively using local sensor data.To reduce the updating energy,ESs can choose to become either updating helpers or recipients of their neighboring ESs,based on sensor quantities and basic updating convergencies.Helpers would share their updated sub-model parameters with neighboring recipients,so as to reduce the latter updating workload.To minimize system energy under updating convergency and latency constraints,we further propose an algorithm to let ESs distributively optimize their cooperation identities,collect sensor data,and allocate wireless and computing resources.It comprises several constraint-release approaches,where two child optimization problems are solved,and designs a largescale multi-agent deep reinforcement learning algorithm.Simulation shows that the proposed scheme can efficiently reduce updating energy compared with the baselines.
