Using AI and ML to Develop the Zero-touch Networks of the Future

The telecom industry is in the midst of a massive shift to new service offerings enabled by 5G and edge computing technologies. With this digital transformation, networks and network services are becoming increasingly complex: RAN, Core, and Transport are only a few of the network’s many layers and integrated components. Today’s telecom engineers are expected to handle, manage, optimize, monitor, and troubleshoot multi-technology and multi-vendor networks. The biggest challenge is balancing the innovation that pushes for new technologies, layers, and nodes with the need to provide robust, high-quality products and services 24/7, 365 days a year.

What are Zero-Touch Networks?

Networks are the backbone for the internet that enables us to communicate globally as we know it today. Service providers help to connect these networks to enable communication across the world. However, the deployment and maintenance of this technology require talented engineers, businesses, and a long history of innovation. Most of us have heard of artificial intelligence and how computers and servers can manage to run themselves without any manual intervention. It is time we bring this idea to the fundamentals of how the Internet runs as we know it today.

A Zero-touch (ZT) network does not require any human to interact with the everyday activities of a commercial, self-sustaining network. According to Ericsson, Zero-Touch Networks have end-to-end network programmability with APIs for service deployments and the full automation of life-cycle processes. Data-driven decision-making and AI are critical components of this design. Zero-touch networks will be built using cloud infrastructure, Software-Defined Networking (SDN), Network Function Virtualization (NFV), machine learning, and 5G developments. This ties into the 5G revolution in networking that is under development today that will bring increased bandwidth, capacity, and expansion for future advancements. One key feature of 5G includes Networking Slicing, which enables network resources to be distributed and deployed easily to meet a certain business purpose or customer. Zero-Touch Networks will be able to easily address different and new business developments, enterprise organizations, consumers, and industry segments with fast turn-around time and return on investment.

Automation is a key enabler of Zero-Touch Networks. This is because automation enables us to automatically configure different components and change the state of a currently deployed service without manually typing in commands or restarting a server. Distributed cloud infrastructure and optimized network applications do expose application programmable interfaces (API) to these network layers. Resources and services are abstracted to enable automation on the slice level. Machine learning and artificial intelligence can be applied to this automation to detect when changes in the deployed setup, network slice, or configuration need to be changed.

What is Network Function Virtualization (NFV)?

This is a relatively new technology in the realm of virtualized networks, although it has been around for a number of years now. Basically, NFV is the next disruptive wave in virtualized networks. NFV is a new architectural solution that focuses on the functions or methods, that networks typically perform, such as Network Address Translation (NAT), Intrusion Prevention Systems (IPS), Access Control Lists (ACLs), Quality of Service (QoS), and many other features. Traditionally, this type of functionality was implemented on top of standard x86 hardware resources. With virtualization, interfaces between these types of network functionalities become instead interfaces between software. This gives us the potential to achieve better automation, service deployment of commercial networks, upgrades, and maintenance.

Benefits of NFV

There are many benefits to utilizing network function virtualization in your data center. NFV can greatly reduce capital expenses which are used to purchase new equipment, hardware, and support. NFV also reduces operation expenses, such as rack space, power, cooling, and other requirements needed to roll out and manage network services. Time to market can be accelerated because the time needed to deploy new networking equipment is reduced and the business can seize new market opportunities and improve return on investment. Finally, NFV can deliver agility and efficiency to quickly scale up or down services to address changing business demands.

What is Software Defined Networks?

While discussing the topic of zero-touch in the network function virtualization, it is also crucial to define Software-defined Networks (SDN). This is a technology that has been around for some time. According to Cisco, it is simply virtualized network hardware components, such as routers, switches, and network interfaces that live on, for example, Virtual Machines instead of traditional hardware.

Traditionally, every time a new service is added, the devices have to be configured and deployed manually by skilled engineers who have done it before. All participating network nodes and devices have to be configured with very similar rules. This causes complex environments with many requirements to be difficult to maintain in a risk-free manner.

Zero-Touch in the 5G Revolution

The 5G revolution will bring much faster speeds, increased load capacity and a number of users, and new use cases.

NFV Will Enable 5G

NFV is a key enabler of this 5G infrastructure because it helps to virtualize all the different appliances and interfaces in the network. A key component of 5G includes Network Slicing, which divides a common bed of physical infrastructure and hardware into several logical, separate networks.

5G Network Slicing

Network slicing is a key feature that will be utilized in 5G networks. A network slice consists of a set of networking resources that are joined together to serve a specific business use case or a customer. The slice is created, configured, and removed by management-level functions. These new types of services are deployed utilizing templates and abstract components or building blocks.