Boost your 5G cloud-native transformation journey with our latest insights and learnings

Today, we are experiencing an evolution of mobile networks – not only on the market expectations placed on new use cases, but also on how they will be delivered. This will demand the introduction of a number of new technologies to today’s networks, such as: cloud-native service-based architecture (SBA), cloud infrastructure, continuous integration and continuous deployment (CI/CD), edge computing, network slicing, automation and orchestration, Voice over New Radio (VoNR) and network exposure. We see exciting prospects ahead of us, but these opportunities are not without their challenges.

The adoption of 5G standalone (SA) technology will be pivotal to deliver to the promise of 5G and the 5G Core network is the engine that will enhance user experience, simplify operations and enable agile new service creation. However, a core network for 5G SA goes way beyond the new 3GPP 5G Core architecture and you are not alone in wondering how to get started. Or, if you have already taken the first step toward deploying 5G Core, you may be considering how to expand, evolve and efficiently operate your network. The realization of this platform for innovation goes beyond the adoption of new technologies and will require a complete transformation of your network, employee skillset and current processes.

From deploying dozens of 5G Core networks for SA with leading service providers (soon to reach 30 commercial live networks empowered by Ericsson’s dual-mode 5G Core solution), we have learnt that there are six strategic domains that service providers need to tackle for a successful cloud-native core network transformation. These domains are cloud native infrastructure, 5G Core, 5G voice, automation and orchestration, operations and life cycle management, and security.

In these last two years, since the first release of our guide series for building cloud-native 5G Core networks, technologies and products have evolved, and we have accumulated lots of experiences and learnings from deploying 5G Core. With the 2.0 version of our guide series, we want to share with you some of this knowledge, as well as insights from some of our partner service providers, to help you get better prepared for your own core network transformation for 5G.

Let me give you a spoiler of what you will find in the series.

The first thing that is important to have in mind is that it will take time to get the 5G platform in place, and it will be a multi-step journey with new functionalities being constantly added to the network. Idealizing and deploying new use cases and applications in 5G will be a dynamic process and it is crucial to bring application developers into the ecosystem as early as possible to start seeing faster innovation.

We are not only building a new core network for 5G and previous generations, but we are also creating the foundation for 6G and beyond.

We are seeing three different waves of 5G SA adoption from a core network perspective, as depicted in the following figure.

Figure 1: The waves of 5G SA adoption from the core network perspective

 

Wave 1: Service providers are mainly focusing on deploying 5G SA services at a small scale, covering selected cities or regions within cities, and for that, a small-scale core network fits the need. There are also service providers that opt to start deploying 5G Core network functions (NFs) just for the packet core domain and leaving others, such as subscriber data management, policy, and signaling, for a second step.

In this initial wave, we are also seeing some service providers take their first steps into some “additional technologies”, such as network slicing, edge computing and exposure, though still quite on experimental level.

Wave 2: Service providers start to expand their network, not only to new geographical locations or by growing its capacity, but also by adding new NFs, such as subscription data management NFs to an already-live cloud-native packet core network.

This second wave of 5G SA also focuses on deploying tools for more efficient network management. As the network expands, the need for automation and orchestration capabilities increases, as does the need for a broader review of operational structures and ways of working. Modernization of some network domains, such as moving into a bare-metal cloud infrastructure or deploying VoNR services, may also be covered in this phase.

Wave 3: Usually running in parallel to the two previous waves, this third wave focuses on enriching the network with new capabilities that will enable more innovation and monetization of the

5G network capabilities that will enable, for example, higher penetration in enterprise and industry segments, and expand the complete ecosystem with application developers. Here, we are talking about the deployment in scale of capabilities such as enriched network slicing solutions, edge computing and network APIs’ exposure.

Maybe you already recognize your organization within one of these waves? Or maybe in more than one?

Although it is possible to navigate the different waves on a trial and experimentation of new technologies level, for your live network deployment we  recommend that  you do not attempt to embrace all these at once, but instead carefully plan when and how to embark on each phase of the transformation journey according to your specific business and network needs, and your technology maturity.

 

Building a cloud-native infrastructure

A solid foundation to the cloud-native 5G Core network starts with its infrastructure. There are several key considerations to ensure success when building a cloud-native infrastructure

With 5G, the idea is to build a single infrastructure for all types of IoT and enhanced mobile broadband services, from gaming to automated manufacturing. The solution is a shared infrastructure with orchestrated network slices to efficiently deploy multiple services, making working telco cloud infrastructure a prerequisite to running many upcoming 5G services.

For the cloud infrastructure, it will be important to have a horizontal platform capable of running multi-tenant virtual network functions (VNFs) and cloud-native functions (CNFs) and to have those orchestrated.

One of the key learnings we have discovered from our virtualization journey with the NFVI solution is that system-verified solutions work best and are the most efficient, while customer-unique and best-of-breed cloud infrastructure stacks come with additional R&D work, costs and a longer time-to-market. To provide support for cloud-native telecom applications in the fastest and most cost-efficient way, we recommend introducing a Kubernetes over bare-metal cloud infrastructure.

This solution has no virtual infrastructure manager (VIM); instead, the Kubernetes-based container as a service (CaaS) platform runs directly on the underlying hardware. For Ericsson and many of our partners, it’s clear that this is the most competitive solution for cloud-native applications, as it simplifies the architecture and uses the underlying hardware more efficiently, with significant total cost of ownership (TCO) savings.

Figure 2: Making a cloud infrastructure solution system-verified

 

In the guide series we explore further the topic of how to build the cloud native infrastructure over bare metal.

Get to the core of the network: the 5G Core

Moving from an Evolved Packet Core (EPC) to a cloud-native 5G Core is a stepped journey that will vary significantly depending on specific business drivers and technological realities from your organization.

Despite the growth rate of 5G services, new 5G networks will co-exist side by side with 4G networks for many years, while at the same time, efficiencies must be improved to capture new opportunities. Keeping separate core networks to manage these different access technologies will significantly increase your TCO and will eventually become unmanageable from an operational perspective. It will also limit the overall benefits you can gain from the cloud-native platform if this only manages the 5G SA traffic portion.

A multi-access, one-core solution is an effective way to ease the introduction of new 5G Core network architecture, as it allows an efficient co-existence with 4G by balancing cost optimization with performance-optimized network deployments. To address this need, we have developed the “Ericsson dual-mode 5G Core”, a solution that combines EPC and the new 5G Core NFs into a common cloud-native platform.

 

Another aspect to highlight is that the 5G Core involves different distinctive domains with its own needs and challenges. While the focus of many (wave 1 see figure 1 above) deployments has been in the packet core domain, the subscription and data management (SDM) and the signaling domains are of equal importance to achieve the expected efficiency and benefits expected from 5G Core and these domains are usually, but not necessarily, in the scope of the second wave of the 5G Core deployments.

The Ericsson’s dual-mode 5G Core, and the specifics of the cloud-native SDM and signaling domains are covered in the guide series.

Migrating to this target architecture is a journey on its own and different options exist . In the guide series we also outline the paths service providers choose most often to evolve from EPC to a dual-mode 5G Core, and closely examine some key building blocks for a cloud-native 5G Core network that you will need to address to become successful in the future 5G business landscape.

 

Enable voice services over 5G

5G subscriptions also need voice, needless to say, right?

The mobile voice service, as defined by the telecommunication standards body 3GPP, is an integral part of 5G smartphones and other voice-capable 5G devices. Voice must be supported from the start in a 5G network, as a 5G smartphone will connect to the new 5G system only if it detects that voice services are supported by the network. There are several ways to achieve this, even before the 5G RAN is fully deployed.

5G voice provides many enhancements such as improved speech and video calling or improved security. These will be fundamental for the realization of many 5G use cases, like gaming and mission critical communications. To achieve this, a fully operational VoLTE capable network using IP Multimedia System (IMS) is required. This means that service providers that have not yet fully embraced VoLTE in their networks have to include that in their network evolution plans to 5G.

We see the evolution to 5G voice following three steps:

Step 1: Dual Connectivity – the introduction of voice in 5G smartphones

Step 2: EPS fallback after the first introduction of voice in the 5G system

Step 3: Voice over 5G New Radio – the target solution for voice in the 5G system

Figure 4: Overview of network evolution steps to enable voice calls on 5G devices

 

Not forgetting that roaming, SMS, MMS, interconnected networks and emergency services also need to adapt to a 5G network context to maintain regulatory and legacy service support.

In the 5G voice network evolution guide we’ve explored further this evolution, its benefits and our recommendations for you to consider.

 

Automating and orchestrating the 5G Core network

5G is designed to be orchestrated and automated. To achieve the full potential of 5G, your network must enable you, your consumers, and partners to use it innovatively.

The network evolution and digital transformation are happening in parallel, and orchestration and automation are key enablers of the journey toward zero-touch networks and business operation.

5G will be distributed over multiple clouds, with NFs expected to be deployed in cloud-native environment to run in the most optimized way. However, manually onboarding, configurating, and life cycle managing those functions will be unmanageable and not sustainable due to their complexity. It is imperative to reduce the number of manual tasks associated with E2E processes. Automating manual tasks will not only decrease operational complexity, but also significantly reduce opex and time to service.

Different autonomous mechanisms are introduced by the industry to reduce the complexity of the 5G Core network, and its services’ management. These mechanisms will also make the 5G operating ecosystem more flexible and self-operative.

Figure 5: Zero-touch requires bringing together automation, openness and flexibility

 

Moving from a manual to an automated operating network is a journey with many challenges, and one that should be undertaken in steps.

Working with leading service providers around the world we have experienced some things that are common for a successful transformation project.

1. Execute each step as a minimal viable product and don’t increase scope and features. Capabilities will evolve based on experience and learnings from previous steps.
2. Each step should have its own business case and follow up. With many initiatives going on in parallel, it is sometimes a challenge to measure the exact causes and effects.
3. Establishing clear governance is vital. Use cross-functional teams to deliver benefits that foster motivation and engagement.

To learn more, check the Automating and orchestrating the 5G Core network guide.

 

Before you have a ‘formula 1’ core network in place, get ready to operate and life cycle manage it

Operating the cloud-native core platform will require a greater operational transformation and automation adoption than seen in previous network generations

As with any new technology adoption, the cloud-native 5G Core network will have to adapt and coexist with the legacy network for many years. In previous technology shifts, from 2G to 3G to 4G, the foundation was quite similar and most existing operational procedures could be reused or adapted with some basic levels of automation added. This time, however, that will not be the case.

5G operational procedures will need to support massive data traffic, maintain stringent service level agreements (SLAs), multi-access edge computing and exposure. Therefore, processes requiring manual activities will not be able to react to the number of decisions and actions required to maintain the expected customer service experience.

Another important aspect to be considered is how to automate the life cycle management of the network. To enable a faster time to launch new services, and to implement the latest improvements to the network, implementing CI/CD of software is paramount.

The ultimate goal of CI/CD is to allow for automated, repeatable and low-risk updates of all components and layers of the architecture stack. This includes the cloud infrastructure, software applications for the core network and IT domains like OSS and BSS, management of network slices for end-to-end services and the exposure of APIs.

To meet service provider requirements, new releases from suppliers need to be continuously delivered, integrated, verified, validated and deployed. This flow should be fast, and each step automatically triggered by input from the previous one. With updated software made available to service provider repositories, automation will pull new and accepted deliveries into the next environment, facilitating a flow between test and production, or central and national operations.

As you can see, this is not a simple task, but one that is necessary to reap the fill benefits of 5G and beyond and that should be an integral part of the overall operation transformation journey.

And back to that, based on our experience, and the feedback from our partner service providers, a good operational transformation strategy is imperative to address the opportunities and challenges that cloud-native adoption brings. This strategy needs to build on three main pillars:

1. Processes: Managing, automating, orchestrating and life cycle managing a hybrid core network in a cost-efficient way will require a suite of new E2E operational processes to move away from siloed stacks to a service-oriented approach. These new processes should leverage automation capabilities, achieving overall network visibility.
2. People: Models should be transformed from a typical domain (vertical) to cross-domain (horizontal) to enable the necessary E2E competence for leveraging the capabilities of a cloudified network. This requires new skills to be developed and organizational borders must be reviewed to match the new processes.
3. Platform: Revamped modern solutions are required to address both old and operational challenges while scoping an architecture with legacy, virtualized and cloudified components at the core. They should invest in technology that can move them closer to a zero-touch network, for quick creation and delivery of innovative services.

Figure 6: The three pillars of operational transformation

 

To learn more, read our guide: “Transforming operations on the way to 5G”

 

Keeping 5G networks secure

Now that all is in place and working as it should, let’s not forget to ensure it stays like that. Safeguard your network.

Security is critical for 5G businesses, industries and society.

Society’s critical functions are continuing to move to the digital domain, and are relying on telecom networks. New capabilities have become vital for both industry and society but cyberattacks are on the rise. As the value and volume of exposed personal information increases, laws and regulations aimed at controlling network access and protecting user data are becoming more stringent.

Regulators consider safeguarding these networks vital. However, the threat landscape for 5G is more complex than previous generations as it converges with traditional IT, enabling IT threat actors to attack telecom networks in a similar way. In addition, networks often have new functionalities, such as network slicing for service separation and isolation, as well as an increased use of AI/ML for automation. When this is paired with edge computing, which is placing cloud resources closer to the access, new challenges arise – even while enabling mission-critical, low-latency applications.

There are four layers that we see as key to building a secure, trustworthy network and services: standard, product development processes, network deployments and network operations. 

Security is driven by risk and is about continuous risk-management and risk-appetite decisions for context. Building trust in the critical infrastructure platform, deployment and operations is vital. Risk-driven and adaptive security management automation solutions that provide enhanced security visibility to the managed context will be increasingly needed.

Figure 7: Four areas for building trust in mobile networks

 

At Ericsson, we have a long history of systematically incorporating security and privacy considerations into all relevant aspects and phases of our products’ life cycles. Our efforts follow a well-established internal control framework known as the Security Reliability Model (SRM). The SRM enables a managed, risk-based approach to achieve security and privacy by design, where requirements are tailored to the target environment and demands.

To learn more on how to secure deployment and operations for 5G Core networks, the operational benefits of automating security and our unique solutions to secure the core network, its traffic and subscriber’s privacy, read our guide: “Mastering complete 5G network security” guide.

 

You next steps: 

Imagine if you could navigate the learning curve of the cloud-native transformation faster, while leveraging the experiences from 5G pioneers. With Ericsson, it is Possible. 

This post gives just a glimpse of what you will find in our extensive “cloud-native 5G Core network guide series 2.0”.  

Continue exploring further by reading the different guides that digs deeper into each of the six strategic domains we spoke about here. They do not cover just technology, but in them we have compiled insights and recommendations from our vast experience launching 5G SA networks with 5G service provider pioneers. 

Enjoy the reading!