From Nokia and MWC23 in Barcelona: 5G satellites, edge computing, artificial intelligence, low latency and digital twins for wireless networks

5G Obs – From Nokia and MWC Barcelona, Spotlight On Telco Trends 2023+: Satellites, Edge, Digital Twins

The telco techno trends for 2023 and beyond from Nokia–one of the top and reliable vendors of hardware and software for telecommunications,—overlap with the big themes of the industry’s main events, and firstly upcoming MWC23.

Connectivity reaching new heights with 5G satellite access, edge computing allowing networks to run artificial intelligence (AI) and other technologies with the lowest latency, “Networks – and not only – as a Service“; digital twins for the wireless networks and, last but not least, a Federation of the cloud environments, or Network of networks, are the topics emerging across the globe.

Some trends from last year are still central: industrial private networks for settings, open-Ran   or multi-vendor networks  and, just beginning to take off, the industrial metaverse.

Connectivity from satellites

grew strongly, especially over the last couple years. A well-known example is how Ukraine maintained access to the Internet, whilst another is how private connectivity expanded in remote places thanks to constellations of Low orbit satellites, LEOs, which are smaller , lighter and much cheaper to make, launch and manage than those with Intermediate orbit (MEOs) or the geostationary ones (GEOs).

“As we look ahead,” Claudio Santoianni, Marketing & Corporate Affairs Director of Nokia Italia said,“ we expect a boom in satellite access for Non-Terrestrial Networks (NTNs) that use spaceborne/airborne vehicles for transmission as well as devices that access satellite connectivity directly.”

The institutional organization establishing the standards for 5G, that is 3GPP,  is currently working at defining NTNs for advanced 5G, a technology being seen as an integrated part of 6G to provide connectivity everywhere.

5G NTN satellite access creates many possibilities, including rural and remote 5G coverage, filling gaps in existing networks, global 5G connectivity in areas without terrestrial coverage, global mobile broadband and IoT coverage with low-cost connectivity, Fixed wireless access, IoT low data rate services for long battery life and connectivity for planes, ships and in disaster areas.

Networks edge orchestration

is bound to take center stage. Having computing capacity at the edge allows to run e.g. AI and Software-as-a-Service (SaaS) applications close to the physical end devices producing or consuming the data, such as controllers in the case of manufacturing. Having computing hosts close to the IIoT sensors or other measuring devices facilitates data collection, processing, storage and analytics.

The proximity that the edge cloud and edge computing allow  means furthermore low latency, availability and reliability  for the user applications and delivers requisite performance for high bandwidth and latency sensitive use cases like IIoT, augmented/virtual reality (AR/VR) and Industry 4.0.

” With 5G monetization as the prize, the edge cloud computing technology will proliferate rapidly as communications service providers (CSPs) deploy 5G networks with dozens of central and thousands of distributed cloud-edge sites,” Santoianni said.

The edge cloud computing environment will evolve, using Network as Code, multi cloud, open API exposure and AI/ML with closed loop intent driven orchestration. These building blocks will create a converged framework at the network edge to meet a multitude of user demands delivered with high agility and lower operational cost.

Edge orchestration is therefore critical and edge ecosystems will evolve to become an essential pillar of CSPs’ digital transformation journey, enabling them to gain market relevance beyond just connectivity. Or as  STL Partners in London tells us: “We believe that 5G  needs  edge computing more than the latter needs 5G networks”.

Networks-and-more-as-a-Service (N+aaS)

will build more capabilities on top of Core SaaS.

In a previous Nokia blog, Core Network Software-as-a-Service (Core SaaS) was described as offering hardware, software and services bundled into a pay-as you-grow subscription. Core SaaS offers simplicity and a more predictive, OPEX-led approach.

In 2023 and beyond, services will be distributed, deployed and run across multiple resources such as public clouds, edge clouds, networks, and devices all working together to provide a single service or set of services. CSPs will evolve from Core SaaS to N+aaS   providers with cloud, connectivity, context, and data assets offered to enterprises.

N+aaS builds on Core SaaS, going beyond basic connectivity to offer additional value in the form of positioning, presence and other network-driven insights that are abstracted for digital services to consume. The current way of delivering Core SaaS from public clouds will be extended to use local resources to meet the needs of future applications in augmented reality, gaming or automation that require local anchors to provide low latency, efficient data transfer and for enhanced security and privacy.

Digital Twins: borrowed from industrial technology

are a qualitative leap for the last generation of networks.

In the telco world, a digital twin is a virtual representation of a network (services and applications) based on real-time data from multiple sources like ML data lakes, edge clouds, IoT devices, subscriber data, sensors and more. The aim is to use simulation and machine learning to visualize and predict the effects of different scenarios without having to implement them in physical networks.

As CSPs adopt and accelerate digital transformation to address complex 5G consumer and industry vertical use cases, digital twins can monitor and augment such complex systems in real-time. This will help CSPs to better understand the network, processes, and customers – and how they impact one another.

Early use cases being explored at Nokia Core Networks include: Network monitoring with anomaly prediction and self-healing, network function software update and version management service, visual network planning and configuration in Digital Sandbox for impact analysis prior to network deployment, simulation of energy consumption and cost of running the services based on function of €/MW, simulation of least cost routing to simulate cost (for CSP) of routing calls / data via different interconnect carriers instead of performing actual routing, simulation of interface failure and traffic re-routing to represent impact of failure on network and services.

Towards a  Network of networks or Cloud federation

that will open up new possibilities for already existing technologies.

To facilitate N+aaS to deliver services comprising multiple assets from many diverse sources, CSPs will need to seamlessly support the sharing of arbitrary resources, from arbitrary application domains with arbitrary consumer groups across multiple administrative domains.

Network of networks or Cloud Federations will be key for achieving such a complex sharing of resources from multiple cloud environments such as public clouds, private clouds, and hybrid cloud, as well as on-premises data centers.

A vast range of dynamic resources and services sharing are possible, such as cloud bursting, telemetry and observability data of events and alarms, data sharing collaboration based on regulatory requirements, disaster recovery models etc.  Any type of organizational collaboration could be facilitated by a secure method to selectively share data with specific partners.

With the adoption of federated cloud ecosystems, users can take advantage of increased reliability, the flexibility to deploy assets on multiple cloud providers according to their business requirements and services that leverage multiple assets as distributed service chains.

However, cloud federation is an emerging topic, so much effort is still needed to seamlessly integrate multiple assets with the right security and entitlement for users.

It may be interesting to remark how broader and faster the boundaries between industries, and especially telecommunications and manufacturing, begin to fade as to technologies and  use cases.