Page 123 - ITU-T Focus Group IMT-2020 Deliverables
P. 123

ITU-T Focus Group IMT-2020 Deliverables                                3


            Virtualized WiFi: A WiFi network testbed deployed at the NICT Headquarters in Tokyo [Ref.6.3.3-4]. The
            virtualized WiFi can control the connectivity of a target service by provisioning dedicated base station (BS)
            resources to the target service or the target WiFi slice. All the decisions on BS selection and handover are
            separated from the BSs and terminals and put together into a centralized controller, while consistent layer-2
            paths for the terminals in the backhaul OpenFlow network are also cooperatively configured. The Virtualized
            WiFi has potential to connect to core of other virtualized networks.
            Abbreviations and Acronyms

            •       RISE (Research Infrastructure for large-Scale network Experiments)
            •       JOSE (Japan-wide Orchestrated Smart/Sensor Environment)
            •       HIMALIS (Heterogeneity Inclusion and Mobility Adaptation through Locator ID Separation)
            •       SDN (Software-Defined Networking)
            •       VM (Virtual Machine)

            •       BS (Base Station)
            •       WiFi (Wireless Fidelity)
            •       MA (Mobility Anchor)
            •       JGN-X (JGN eXtreme)

            References
            [6.3.3-1]    RISE: A Wide-Area Hybrid OpenFlow Network Testbed, IEICE Transactions on Communications,
                        Vol. E96-B, No. 1, Jan. 2013.
            [6.3.3-2]    JOSE: An Open Testbed for Field Trials of Large-scale IoT Services, NICT Journal, Vol. 62 No. 2,
                        Mar. 2016.
            [6.3.3-3]    An ID/locator split architecture for future networks, IEEE Communications Magazine, Vol. 48,
                        No. 2, Feb. 2010.

            [6.3.3-4]    WiFi Network Virtualization to Control the Connectivity of a Target Service, IEEE Transactions
                        on Network and Service Management, Vol. 12, Issue 2, June 2015.

            6.3.4   A network slicing prototype for 5G
            6.3.4.1    5G Network Slicing Architecture Model

            Network slicing is a key feature in 5G network which enables the operator to create networks customized to
            provide optimal solutions for different market scenarios, which have diverse requirements, with respect to
            the functionality, performance and resource isolation. We intend to standardize the slice information model,
            the interfaces for slice management and slice control as well as the intra-slice/inter-slice interface definitions
            through the introduction of this section. And this section is well aligned with the end-to-end slice described
            in Section 8.2, as well as the network slicing scenario in Section 9.4.

            The slicing architecture should be easily extensible for various requirements in 5G network. A basic network
            slice has direct relationships with network resources, network services, network functions, and likely other
            network slices. The slice virtualizes its resources and provides the service with the resources, and implements
            various network control/mgmt functions, as shown in
            Figure 6.3.4-1. Network resources are utilized to realize network services. The slice model is recursive and
            extensible both horizontally and vertically to support different scenarios. The construction of network slice
            is recursive, which means a network slice may be a virtual resource in another network slice. The construction
            of network slice is also scalable by hierarchical abstraction and the proprietary engineering methods. As an
            example, the connectivity slice can be constructed of access network slices and transport network slices,
            which only need to expose certain abstracted interfaces to the connectivity slice as the virtual resources
            without  revealing  all  the  details.  As  a  result,  there  will  be  interactions  among  slices,  either  in  a
            vertical/hierarchical level or horizontal/peer level.



                                                                                                         117
   118   119   120   121   122   123   124   125   126   127   128