Private or public 5G network - which is better for Industrial IoT?

Manufacturers have several options to deploy 5G network in factories, including all private network, private integrated with public network and network slicing from public network. Use cases and the extent of control would be the key decision-making factors. Enterprises to adopt 5G network are facing two options – private network or network slicing of the public network.

Self-managed 5G private network

A private network is defined as networks with radio, core, and transmission resources dedicated to the enterprise, and under the control of the enterprise. A private 5G network offers several benefits for enterprises. First, the network can be optimized and tailored specifically for industrial applications such as QoS and latency. Enterprises then take control of network deployment and management. This process ensures sensitive company data to stay local. Enterprises could either build a private network for self-operation, or deploy a private network service at the customer premise by 5G service providers.

In terms of deployment models, enterprises can build the network with or without integration with the public network. When enterprises need external wide-area mobile coverage, integration with the public network might bring more benefits than deploying all coverage as a private network. Many integration models are available for enterprises. They will have the option to look at how they want to architect and construct their networks in a more efficient way.

Network slicing of public network

Another way to manage sensitive company data under 5G network is through network slicing, a technology that allows multiple logical networks to be created on the top of a shared physical infrastructure. “Network slicing technology enables network owners to customize their network to the needs of diverse use cases within the factories,” said Jens Jakobsen, HMS Labs Development Manager, HMS Networks.

Through network slicing, Jakobsen provided an example, parts of the network can be customized to provide lowest possible latency, while other parts can be customized for highest throughput. Augmented Reality, for instance, requires high throughput, data speed and low latency, while smart electricity meters ask for high-reliability data. Enterprises can then create different silos of networks for various applications in the industrial setting.

In addition to optimizing the needs of the services or segment a cluster 5G network are serving, network slicing also brings great elasticity, robustness, secure and stable operations for enterprises, as well as flexibility and efficiency for service maintenance. Network slicing isn’t the latest technology. It has existed since the time of 3G and 4G. With the arrival of 5G, however, the demand for the technology has truly risen.

Private or public 5G network?

“It depends on the use cases that want to be implemented and the specific requirements of the industry, “suggested Erik Josefsson, Vice President and Head of Advanced Industries, Ericsson. “For many applications, a public network is a great alternative, for example using network slicing. In cases where local management of the network and/or high availability is required, a private network would be right solution,” said Josefsson. After all, the first commercial 5G networks are deployed for mobile broadband. Maybe they’re not supporting all the features needed for industrial communication networks.

Considering the current status of 5G network, maybe it’s too early for enterprises to commit building a 5G private network. But, building a private 5G network will undoubtedly become more popular among enterprises. “Private 5G networks are capable of being deployed in licenses, dedicated and shared spectrum. The current 5G IIoT applications are expected to rely on the public networks; however, after large scale deployment on standalone 5G networks, private 5G networks are expected to witness significant roll-outs,” said Akshay Deshpande, Senior Research Analyst at MarketsandMarkets Research.