Viewpoint: How open APIs give 5G the support it needs

We are surrounded by IoT solutions, from phones to smart meters to smart vehicles that are connected to the internet and constantly share data, writes Zoltan Varnai of Nokia.

But while today’s cellular IoT devices are primarily connected via 3G and LTE networks, the development of 5G will broaden the realm of possibilities when it comes to IoT. Take a robotic manufacturing line in which every millisecond counts, as the movements of the manufacturing line requires rapid measuring and instant feedback on results for the controller.

Software control

Controlling software is at the heart of this automated production line. Synced to the IoT platform, the control software submits queries and requests before receiving hordes of data from IoT devices via open APIs. Enabling these APIs to access the network services they need requires a network exposure platform. Today’s 3G and 4G IoT API exposure platform SCEF (Service Capability Exposure Function) can handle enormous amounts of IoT devices; however, use cases still remain within the boundaries of 3G and 4G latency.

With regard to operator IoT services, only 5G can achieve the super-high reliability and ultra-low latency needed for our robotic manufacturing use case. Moreover, a 5G Network Exposure Function (NEF) is required to provide secure and speedy access to IoT control and management of APIs in a single, robust dedicated API gateway. The response time of API requests are crucial for Ultra-Reliable Low Latency Communication (URLLC) IoT use cases, such as robotic assembly. The 5G NEF platform must therefore deliver outstanding performance.



In order to provide a seamless upgrade path for operators, SCEF is bundled into its 5G NEF functionality (in one of the biggest telecom vendor solution), allowing operators to easily evolve from 4G API exposure to 5G exposure.

The power of network slicing

End-to-end network slicing is a key use case for helping operators maximize and monetize 5G.

A robot production line has strict availability requirements and latency, which can readily be met by a network slice. The dedicated network slice attributes must be set up according to the needs of the manufacturing line massive IoT load and low latency requirements.

For example, massive User Plane Function (UPF) and Session Management Function (SMF) resources must be allocated in the edge of the operator’s telco cloud in the closest location to the end points.

From a use case point of view, several non-time critical 5G network functions may be in the central cloud like UDM (Unified Data Management) and AUSF (Authentication Server Function). Since the manufacturing plant location is fixed from the 5G mobile network point of view, the slice-related Access and Mobility Management function (AMF) can be instantiated in the central telco cloud as well.

The slice has a dedicated NEF, which only accepts API requests from the line’s IoT control software – any other applications queries are denied. The slice specific instantiation of NEF also has a preferred location in the edge to reduce latency, as well as its own API Gateway function to enable access to the slice specific APIs. The control software data collection and decisions to interact with the production line are done through APIs. The API requests include parameters and data that are coded to specific http/2 protocol.

Similarly, the API responses to control software are in http/2, as this protocol is widely used for RESTFul (Representational State Transfer) APIs.

The 5G core network may also receive lots of requests to its master-NEF from several different use case applications. While these may apply to the entire network, the slice-specific NEF will not be affected and will, therefore, consistently maintain high performance. Even if the master-NEF were to become overloaded, the slice-specific NEF would remain functional.

This boosts the service’s security since other applications are unable to access the slice related APIs. Similarly, since the gateway is dedicated to its own use case, API response time is enhanced.

Thanks to this, slice-specific network exposure will be a critical capability that operators can utilize to score new revenue and support many of the latest IoT use cases.

Catalyst

5G end-to-end network slicing promises to be a powerful catalyst for innovative use cases, and 5G network exposure fits this concept through slice-specific NEFs. 5G programmable networks will become an essential platform for high-productivity automated manufacturing that can achieve greater efficiency and higher-quality products, so manufacturers, system integrators, and third party cloud application developers can work to build an open ecosystem that is based on telco APIs.

Zoltan Varnai works in the Nokia Software Core marketing team. His professional passion is marketing the core network with a focus on 5G core programming capabilities.

Alun Williams

Web Editor of Electronics Weekly, he is the author of the Gadget Master, Eyes on Android and Electro-ramblings blogs and also covers space technology news. He has been working in tech journalism for worryingly close to thirty years. In a previous existence, he was a software programmer.

Leave a Reply

Your email address will not be published. Required fields are marked *

*