To satisfy mobile terminals ’( MTs) offloading requirements and reduce MTs’ cost,a joint cloud and wireless resource allocation scheme based on the evolutionary game( JRA-EG) is proposed for overlapping heterogeneou...To satisfy mobile terminals ’( MTs) offloading requirements and reduce MTs’ cost,a joint cloud and wireless resource allocation scheme based on the evolutionary game( JRA-EG) is proposed for overlapping heterogeneous networks in mobile edge computing environments. MTs that have tasks offloading requirements in the same service area form a population. MTs in one population acquire different wireless and computation resources by selecting different service providers( SPs). An evolutionary game is formulated to model the SP selection and resource allocation of the MTs. The cost function of the game consists of energy consumption,time delay and monetary cost. The solutions of evolutionary equilibrium( EE) include the centralized algorithm based on replicator dynamics and the distributed algorithm based on Q-learning.Simulation results show that both algorithms can converge to the EE rapidly. The differences between them are the convergence speed and trajectory stability. Compared with the existing schemes,the JRA-EG scheme can save more energy and have a smaller time delay when the data size becomes larger. The proposed scheme can schedule the wireless and computation resources reasonably so that the offloading cost is reduced efficiently.展开更多
如何在IPv6 over MPLS网络中提供端到端的服务质量?针对这个问题介绍了一个6PE-DiffServ解决方案(6PE,IPv6 Provider Edge Routers)。该方案使用6PE技术进行组网,通过统一的DiffServ策略的部署和BB-6PE(Bandwidth Broker for 6PE netwo...如何在IPv6 over MPLS网络中提供端到端的服务质量?针对这个问题介绍了一个6PE-DiffServ解决方案(6PE,IPv6 Provider Edge Routers)。该方案使用6PE技术进行组网,通过统一的DiffServ策略的部署和BB-6PE(Bandwidth Broker for 6PE network)的集中管理,给由IPv4向IPv6过渡的网络提供DiffServ服务。6PE-DiffServ方案的可行性和有效性在CNGI-QoS实验网中得到了验证。展开更多
This paper presents a novel compact memory in the processing element （PE） for single-instruction multiple-data （SIMD） vision chips. The PE memory is constructed with 8×8 register cells, where one latch in the...This paper presents a novel compact memory in the processing element （PE） for single-instruction multiple-data （SIMD） vision chips. The PE memory is constructed with 8×8 register cells, where one latch in the slave stage is shared by eight latches in the master stage. The memory supports simultaneous read and write on the same address in one clock cycle. Its compact area of 14.33 μm^2/bit promises a higher integration level of the processor. A prototype chip with a 64×64 PE array is fabricated in a UMC 0.18 μm CMOS technology. Five types of the PE memory cell structure are designed and compared. The testing results demonstrate that the proposed PE memory architecture well satisfies the requirement of the vision chip in high-speed real-time vision applications, such as 1000 fps edge extraction.展开更多
基金The National Natural Science Foundation of China(No.61741102,61471164)
文摘To satisfy mobile terminals ’( MTs) offloading requirements and reduce MTs’ cost,a joint cloud and wireless resource allocation scheme based on the evolutionary game( JRA-EG) is proposed for overlapping heterogeneous networks in mobile edge computing environments. MTs that have tasks offloading requirements in the same service area form a population. MTs in one population acquire different wireless and computation resources by selecting different service providers( SPs). An evolutionary game is formulated to model the SP selection and resource allocation of the MTs. The cost function of the game consists of energy consumption,time delay and monetary cost. The solutions of evolutionary equilibrium( EE) include the centralized algorithm based on replicator dynamics and the distributed algorithm based on Q-learning.Simulation results show that both algorithms can converge to the EE rapidly. The differences between them are the convergence speed and trajectory stability. Compared with the existing schemes,the JRA-EG scheme can save more energy and have a smaller time delay when the data size becomes larger. The proposed scheme can schedule the wireless and computation resources reasonably so that the offloading cost is reduced efficiently.
文摘如何在IPv6 over MPLS网络中提供端到端的服务质量?针对这个问题介绍了一个6PE-DiffServ解决方案(6PE,IPv6 Provider Edge Routers)。该方案使用6PE技术进行组网,通过统一的DiffServ策略的部署和BB-6PE(Bandwidth Broker for 6PE network)的集中管理,给由IPv4向IPv6过渡的网络提供DiffServ服务。6PE-DiffServ方案的可行性和有效性在CNGI-QoS实验网中得到了验证。
基金Project supported by the National Natural Science Foundation of China(Nos.60976023,61234003)the Special Funds for Major State Basic Research Project of China(No.2011CB932902)
文摘This paper presents a novel compact memory in the processing element （PE） for single-instruction multiple-data （SIMD） vision chips. The PE memory is constructed with 8×8 register cells, where one latch in the slave stage is shared by eight latches in the master stage. The memory supports simultaneous read and write on the same address in one clock cycle. Its compact area of 14.33 μm^2/bit promises a higher integration level of the processor. A prototype chip with a 64×64 PE array is fabricated in a UMC 0.18 μm CMOS technology. Five types of the PE memory cell structure are designed and compared. The testing results demonstrate that the proposed PE memory architecture well satisfies the requirement of the vision chip in high-speed real-time vision applications, such as 1000 fps edge extraction.
