5G QoS can already distinguish service types so precisely. Why is network slicing still required?

This is because each of 5G’s three use cases (eMBB, URLLC, and mMTC) have different requirements on 5G networks. eMBB, URLLC, and mMTC are short for Enhanced Mobile Broadband, Ultra-Reliable and Low-Latency Communication, and Massive Machine-Type Communications, respectively.

• eMBB: relates to peak rates, capacity, spectral efficiency, mobility, and energy efficiency.
• URLLC: relates to ultra-low latency, high reliability, and mobility.
• mMTC: relates to the number of connections.

5G must guarantee Service Level Agreements (SLAs), enable independent operation, and provide security isolation for different services in all industries if these industries are to hit digitalization goals.

The figure below illustrates further details.

Traditional QoS-based networks do not provide independent operation and security isolation that are tailored to different service requirements. They are also unable to provide efficient SLA assurance for all services.

With this in consideration, network slicing has been proposed. With network slicing, an operator’s physical network can be sliced into multiple E2E (RAN, transport network, and core network) virtual logical networks that are isolated from one another and provide customizable network capabilities for industry customers. These customizable network capabilities include latency, bandwidth, number of connections, and reliability. This way, multiple logical networks can run on top of an operator’s network, tailoring to the specific requirements of different services.