Skip navigation
Male technician checking server

Network architecture

Explore what it is, how it works and what to expect next

The network’s architecture provides the framework for the various components that make up today’s versatile mobile networks, such as its functions, interfaces and information elements. This includes defining the design of its components, how they interact and how they can be deployed across the network

Network architecture should ensure that networks can perform and evolve as required. Often hierarchical in nature, a well-defined architecture ensures that new functionality can be introduced, optimized and scaled without needing to change existing components or other domains. This lays the foundation to realize several types of networks for society-, business-, and mission-critical networks.

The basics explained

Then and now: how architecture is evolving

Today’s mobile networks are undergoing a transformational shift. New capabilities are emerging that realize business-critical functionality such as differentiated connectivity, programmability, exposure and autonomous networks. This is changing how networks are designed, deployed, operated and monetized – driving a need for design choices that are more flexible, scalable and sustainable.

From vertical to horizontal

Networks are evolving beyond siloed, vertically integrated nodes towards cloud-native, horizontal deployments where capabilities can be shared and scaled on-demand across network layers.

From static to autonomous

Demands for increased service agility and assurance is driving the evolution of autonomous networks, based on network programmability, intent-driven operations and AI-driven adaptibility in real time.

From closed to open

Networks are becoming more open than ever, exposing capabilities via network APIs and enabling integration with enterprise- and other architectures.

Five design pillars of a performant network architecture

Network architecture should bring a long-term perspective to the evolution, supporting the introduction of new technologies through compatibility, modularity and open interfaces. At the same time, network architecture should reuse and build on what is already available.

A network architecture should be able to work in different usage scenarios, such as high to low capacity, wide area to local area, very dense coverage unsatisfactory coverage, indoor and outdoor, etc. It needs to be scalable and flexible, with the ability to scale upwards and downwards depending on demand. The architecture must also take into consideration the cost of acquiring, installing and operating a network.

The network must be available 24/7 for authorized users, delivering the expected level of service and ensuring the necessary safeguards against unauthorized access. Whatever happens, it should always ensure that there is a solution to either prevent or mitigate the problem. The architecture must also be able to allow authorities to lawfully intercept the network for crime prevention and detection.

The network’s infrastructure must be capable of delivering the expected user experience every time they access the network – balancing price, cost and performance. The term user should be interpreted in the widest context – anything from a person to an actuator, a sensor, etc.

The architecture’s success is measured by the value it creates, a culmination of its performance, security, resilience, efficiency and innovation value through exposure. The architecture should ensure that functionality and technical interfaces are aligned with business needs and act as an enabler for a shorter time to customer. Networks must also work towards a more sustainable environment; for example, via lower power consumption, comply to material regulations and alternative energy solutions, such as solar power.

How architectures are defined, built and operated

An effective mobile network is the product of three different kinds of architecture: a functional architecture, an implementation architecture, and a deployment architecture. Together, these architectures contribute to how the network is defined on an abstract level, how it’s built and how it’s eventually designed, configured, and operated.

Functional architecture

The functional architecture defines the abstract entities, the functions they perform, their services offered to other entities, and the procedures i.e. how the system works. This sets what the implementation needs to do, while allowing a range of implementation choices.

Implementation architecture

The implementation architecture guides the development process with the choice of development tools, internal structures, and software lifecycle management i.e. how the system is built. The implementation should be flexible to allow as many kinds of deployment as possible.

Deployment architecture

The deployment architecture comprises the design, configuration, and operation of the network within the capabilities of the available implementations i.e. how the network is created and operated. Central to a deployment is a network topology with different types of network sites.

Learn more about the three kinds of network architecture in the blog post: six talking points for designing the next wireless generation

The building blocks of modern architectures

To enable networks to meet the diverse needs of industries, public safety and defense sectors, today’s architectures need to be able to provide various types of functionalities across different physical locations.

At Ericsson, we achieve this through a layered approach that separates functional network domains running horizontally across locations. This makes it possible to deploy network functionality across multiple various network topologies that suit the designated purpose, enabling scale and flexibility of individual parts without needing to change the entire system. This approach allows for complex problems to be divided into manageable components that can be handled by the relevant area. For practical and traditional reasons, a network is often deployed in several sites forming a vertical instantiation of functionalities.

""

Ericsson’s global network architecture is divided across five functional (horizontal) domains and five topological (vertical) sites. Learn more below.

Functional domains (horizontal)

The global network architecture is divided into five functional horizontal domains (with an additional applications domain external to the network):

The Application domain contains applications external to the network that interact with the network and utilize its exposed capabilities for communication, execution and storage. Such applications could be developed for various use case areas including public sector, enterprise and mission critical.

Developer platforms typically implement functionalities, such as developer-friendly tools, while exposing standardized APIs that enable developers and enterprises to access and consume aggregated network services as building blocks for applications. These capabilities are often streamlined into domain-specific solutions for use cases including fleet management, smart industries, banking and fraud prevention.

The API aggregation implements functionality for aggregating network APIs across a large number of CSPs, exposing aggregated API services to external customers (typically developer platforms) through a standardized interface.

The management, orchestration, monetization and exposure domain contains functionality to manage and control the network, handle the business management interaction with customers to the network, as well as enabling exposure of network functionality to external applications to be consumed through Network APIs.

An integral part of service exposure is to make network services (QoS, location, control and payload services) available in a convenient, controlled and secured way. This includes OSS/BSS functions that support commercial and operational processes from service onboarding to charging and retirement.

It shall be noted that AI is leveraged in this area through AI-driven automation. AI will be present throughout the mobile networks but not always for direct external consumption to its users.

The access, mobility and network application domain contains network functions to secure several types of network access as well as network-integrated functions. This includes the functional entities of radio access networks (RAN), core networks, IP Multimedia Subsystem (IMS), user data management (UDM), etc., all standardized primarily by 3GPP, but also by other standardization bodies like O-RAN Alliance.

The cloud infrastructure and data handling domain contains functionality for secure processing and storage of network- and application functionality. Data handling supports all network domains in collecting, storing, distributing and processing of data providing support for e.g. AI.

Most functions in the network are realized as software that executes in an execution environment with an operating system and/or a cloud platform that runs on some hardware. Many network functions are executed in this domain.

The Transport domain contains functionality for transmission and transport primarily between domains but also within domains. Enabling advanced use cases requires continuous evolution of the transport network, supporting higher bandwidth, new spectrum and centralized RAN. The convergence of transport technologies, such as IP/MPLS networking protocols, can provide operational cost savings for communication service providers (CSPs).

Network sites (network topology)

Mobile networks are built up of many different types of sites – from local access sites to centrally deployed functionality on a national or even global level.

How this topology is designed and deployed is network dependent and relates to relevant use cases. For example, typically a national CSP network has a somewhat different network topology with different names compared to a mission critical network or enterprise-private local area networks.

This domain comprises devices and nodes that transmit to the network and are normally outside the control of the CSP. This could be a smartphone, a robotic arm within a factory or a device that provides a network set-up in an enterprise, such as a router connected to the mobile network for internet access.

Access sites comprise cell sites that comprises antenna systems, radio base stations, backhaul networks and even satellite-based access. In an enterprise, the concept of sites are also used and could be the placement of an access point e.g. an Ericsson Radio Dot.

Aggregation or distribution sites provide greater execution and transport efficiency, as well as the aggregation of functionality. This also includes local breakout for deployments such as software-defined wide area networks (SD-WAN).

National- or central sites are typically centralized within a CSP’s network or within a local dedicated network (typically private network) when connecting several sites in the same network.

A global site is a centralized site that is publicly accessible from anywhere, for example a large data center that may be located either nationally, regionally or globally.

Future network architecture

Dive deeper into the architectural capabilities network domains

Learn more

Network architecture deployment examples

3GPP based telecom networks are highly suitable as the base network architecture for many different scenarios, enabling multiple use cases for several customer types.

Explore the architectural blueprints for each use case on our future network architecture page.

General public networks

General public networks are provided by CSPs and cater to both consumer and business needs. They encompass a wide range of services, including internet and data access, communication services, emergency services, and emerging technologies like XR and cloud gaming.

Wide-area dedicated networks

Wide-area dedicated networks offer coverage over general wide areas or specific geographical regions. While they provide similar services to general public networks, they may be tailored to include enterprise solutions such a support for virtual private networks (VPNs), public safety services, and mission critical communications, like those required in rail transportation.

Local dedicated networks

Local dedicated networks are designed for specific, localized environments with unique demands, such as mining sites, factories, hospitals, airports, stadiums, and university or enterprise campuses. Local dedicated networks can be deployed through using the infrastructure of a CSP or as a dedicated infrastructure with exclusive resources.

6G and future network architecture

Future 6G networks are already in the research phase, with commercial deployments expected around 2030. 6G will bridge physical things, people and activities into a fully cyber-physical world and in doing so will form the bedrock of a more human-friendly, sustainable and efficient future society.

Learn more about future 6G architecture on our future network architecture page.

Stay up to date with the latest 6G developments on our 6G pages.

""

Overview of the target network for 6G in the 2030 timeframe across selected areas.

Trends shaping future network architecture

The evolution towards 6G networks is already taking place across four key areas:

Enablers of monetization and exposure

A top priority for today’s CSPs is to monetize 5G capabilities to improve top line revenue capture. This ongoing journey will continue in 6G to make it possible for 6G networks from day one to reuse and expand on the evolution of 5G exposure and monetization functionality. This includes network API exposure, as well as differentiated connectivity models to support several different service models.

Automation of network operations

AI and intent-based autonomy will continue to redefine service and network automation, including the automation of processes handling the full lifecycle of customers, partners, suppliers, products, orders etc., both from commercial and operational aspects. rApps will be key to redefining network automation, enabling CSPs to manage networks more efficiently and improve end user experience.

Cloud-native design and deployment

The introduction of the Service-Based Architecture (SBA) with 5G, inspired by cloud-native design principles, provides a key design foundation for future 6G architectures. While it is fundamental to leverage existing investments in the 5G Core as part of the future 6G architecture, potential enhancements could involve simplifying the SBA, reducing network function dependencies, and strategically bundling functionalities to manage the growth of network functions and service-based interfaces.

Network architecture evolution

The 6G architecture will be based on the horizontal separation of the network functions from the underlying platform and overlying end-to-end management and exposure layers. This will enable new 6G RAN and evolved 5G core network functions to benefit from the fast evolution of cloudification, IT frameworks, automation, open interfaces and artificial intelligence (AI)/machine learning (ML).

Related insights

6G network architecture – a proposal for early alignment

Learn why early alignment on a common set of 6G architectural principles is crucial to achieve the 6G vision, use cases and requirements within the 2030 timeframe.

The path forward: hybrid 5G Core network architecture

Learn more about the future hybrid core network, capable of supporting legacy 4G and 5G non-standalone devices alongside 5G standalone devices and 6G traffic.

Six talking points for 6G architecture

Our experts lay out the six key talking points for designing future 6G systems, including considerations for deployment options, network interfaces, intent-based autonomy, and exposure.

Explore related topics