IEEE Journal on Selected Areas in Communications · 2019 · 44 citations · 37 references
ComputingComputer ScienceInformation TechnologyComputer NetworksDigital CommunicationVirtualizationCachingLatency OptimizationResource ProvisioningResource SchedulingVirtual Network Functions5G Mobile Networks5G NetworksBig Data ArchitectureNetwork Performance OptimizationQos-aware ProvisioningResource VirtualizationWireless Communication Channels
The multimedia transmission represents a typical big data application in the fifth-generation (5G) wireless networks. However, supporting multimedia big data transmission over 5G wireless networks imposes many new and open challenges because multimedia big data services are both time-sensitive and bandwidth-intensive over time-varying wireless channels with constrained wireless resources. To overcome these difficulties, in this paper we propose the information-centric virtualization architectures for software-defined statistical delay-bounded quality of service (QoS) provisioning over 5G multimedia big data wireless networks. In particular, our proposed schemes integrate the three 5G-promising candidate techniques to guarantee the statistical delay-bounded QoS for multimedia big data transmissions: 1) information-centric network (ICN), to derive the optimal in-network caching locations for multimedia big data; 2) network functions virtualization (NFV), to abstract the PHY-layer infrastructures into several virtualized networks to derive the optimal multimedia data contents delivery paths; and 3) software-defined networks (SDNs), to dynamically reconfigure wireless resources allocation architectures through the SDN-control plane. Under our proposed architectures, to jointly optimize the implementations of NFV and SDN techniques under ICN architectures, we develop the three virtual network selection and transmit-power allocation schemes to: 1) maximize single user's effective capacity; 2) jointly optimize the aggregate effective capacity and allocation fairness over all users; and 3) coordinate non-cooperative gaming among all users, respectively. By simulations and numerical analyses, we show that our proposed architectures and schemes significantly outperform the other existing schemes in supporting the statistical delay-bounded QoS provisioning over the 5G multimedia big data wireless networks.
37
Existence and Uniqueness of Equilibrium Points for Concave N-Person Games
Judah B Rosen · Econometrica · 1965 · 2.7K citations · Full text
Xindong Wu, Xingquan Zhu, Gongqing Wu et al. · IEEE Transactions on Knowledge and Data Engineering · 2013 · 2.4K citations
Big Data Deep Learning: Challenges and Perspectives
Xuewen Chen, Xiaotong Lin · IEEE Access · 2014 · 1.3K citations · Full text