End-to-end network slicing will be important for 5G

Telecoms.com periodically invites expert third parties to share their views on the industry’s most pressing issues. In this piece Mr. Sun Dong, Marketing Director, ZTE Asia & CIS Region, looks at the significance of network slicing in the 5G era.

As the key infrastructure of the digital society, 5G will not only serve individuals but also accelerate digital transformation in many aspects of business and life. 5G can become a ubiquitous platform to satisfy diverse requirements for man-to-man, man-to-machine, and machine-to-machine communications.

While 4G provides broadband service to individual users, 5G needs to serve more diverse application scenarios and raise challenges on network capability and reliability. In a smart factory, serious damage could happen if service interruption occurs. With automated driving, human life is at stake and the network needs to provide an ultralow latency with 99.999% reliability. For VR/AR applications, the network needs to provide more than 1 Gbps bandwidth. The IoT has low requirements on network bandwidth and latency, but the network needs to provide up to 1 million connections per square kilometre. These scenarios place crucial and diverse requirements on 5G networks.

It is not feasible economically to build a new network for each type of service. But a single network could not simultaneously fulfil all the requirements of ultra-high bandwidth, ultra-low latency, and ultra-high reliability at the same time. It also introduces potential risks without service isolation.

Network slicing becomes an inevitable choice to solve the dilemma. Network slicing enables flexible slicing of 5G network resources into multiple virtual networks to meet specific customers’ requirements. In addition, network resources can be fully shared and dynamically balanced among services and therefore the network becomes scalable. Network slicing offers operators the capability and flexibility to develop new business models, and thus becomes a basic 5G network requirement.

An end-to-end 5G network slice consists of RAN, core network, bearer sub-slices and the management system to support its lifecycle management. Slicing and flexible deployment of AAUs, CUs, and DUs help to adapt to different scenarios.

The cloud-based deployment of CUs facilitates centralized management of radio resources. The co-location of DUs and CUs and the deployment of service anchors close to users improve the transmission latency.

The NFV-based 5G core network further introduces service-based architecture, which decouples network functions from hardware. The architecture implements components-based functions and adopts a stateless design with lightweight and open interfaces. It becomes more agile, scalable, flexible, and open.

The network slices are created by physical network virtualization. SDN architecture with unified management and control can achieve IP and optical layer synergy. This allows an open and programmable physical network to support innovation in network architecture and future services. The degree of intra-slice isolation depends on the slicing technology used. For example, FlexE and FlexO technologies could build rigid pipes that ensure strict isolation between slices with rapid forwarding implemented at the underlying layer. The technologies give flexibility to address different service requirements on bearer networks.

At 5G Asia 2018 in Singapore, 18 – 20 September, ZTE will demonstrate their cutting-edge 5G solutions. Meet them at 5G Asia.