5G networks are significantly more complicated than previous network generations.

In addition to the technology advancements, complexity comes from the fact that 5G networks will be hybrid networks. Network components of different generations will be operating side by side. This means they will be composed of traditional physical networks, virtualized network functions, and cloud infrastructure.

Standalone (SA) 5G introduces both a new cloud-native network core as well as a new radio access technology, known as the 5G New Radio (NR). However, in the initial 5G networks being deployed today, Non-Standalone (NSA) 5G is being implemented that uses the new 5G radio and the 4G core. In the past, mobile networks have been configured with elements from the same generation. However, in 5G, operators can mix elements from different mobile network iterations in various configurations.

5G will also generate vast quantities of data. This will be a result in part of more connected devices, the bandwidth and speed enabled in 5G and compounded by the demand for more services. All this data will need to be smartly sampled, filtered, and analyzed.

Operators will need to utilize automation to help drive efficiencies across the network, making it a critical enabler to improving customer service and streamlining processes. Automation also underpins a closed-loop approach to network monitoring. Having a network that can auto-heal will reduce the number of issues and complaints. If the network is running smoothly there will be less need for a customer to contact support. Predictive analytics and proactive troubleshooting will also be essential here and are driven by automation.

Automation permeates nearly every area of the 5G network and service assurance is no different. In the article, we will drill down into each area of our expertise and examine how automation will be the kingmaker for 5G.

Let us work through the layers of the network and see where automation makes its mark.


Network visibility

The primary function of a network visibility layer is to control the flow of packets of data and traffic across the network. In the cloud-native 5G environment, the network packet broker (as part of a more comprehensive network visibility solution) is deployed as a Container-based Network Function (CNF). These need to be dynamic so that they can adapt to changes on the fly, which is not possible without automation. Any network changes that happen in the assurance layer are automatically reflected in the visibility later, and specifically in the visibility system of virtual packet brokers and virtual filters. The network visibility layer can be used to smartly sample the user plane traffic, which reduces the data load being sent to the monitoring tools while still enabling the operator to ensure high-quality services and a fully optimized network. Therefore, a unified network visibility and service assurance solution are important. Operators are able to launch, synchronize and update elements or whole parts of the system together and deliver an end-to-end view of the network.

Controlling the flow of traffic also requires a smart and automated solution. Smart load-balancing, filtering, and sampling all have automated components that can react to anomalies or trends in the traffic and adjust the flow of traffic as required, without human intervention. This is crucial for large-scale operators looking to optimize their network to deliver a superior customer experience automatically.


Service Assurance

Automation will be critical to the 5G network, which will need an automated service assurance solution to help monitor the traffic. A significant part of 5G is the improved levels of customer service and experience. In order to deliver this, operators need to be deploying a service assurance solution that provides them closed-loop functionality. The loop refers to the feedback of information through the network, which monitors, identifies, adjusts, and optimizes automatically. This ability to self-optimize is delivered through automated capabilities. Without automation, the network would not be able to recognize the anomalies in the network data and automatically heal without human intervention.

It is precisely this fact which makes automation so crucial for 5G. There is just too much network data for humans to comprehend, monitor, and manage. There needs to be an automated solution that removes as many repetitive actions as possible and enables the network to function with greater efficiency.

Another aspect to consider is in the set-up and instantiation of any new service. An automated service assurance platform is critical in making this a smooth process. Service assurance tools need to integrate seamlessly with an operator’s orchestration system. In 5G, the network edge will support multiple service types and will need to be dynamically managed. Automation will also be an essential part of the orchestration process, as when new services are deployed, monitoring and visibility tools can be launched automatically alongside them.

What is important to remember is that service assurance requires monitoring of network traffic. With the growth of data in 5G, automated services will be needed to manage alerts when service levels drop below a certain threshold. Anomaly detection solutions automatically and continuously scan the network and break down the data according to different dimensions and network entities. This entity-level data is then compared to baseline data for the same entity, which is calculated using Machine Learning. When a significant anomaly occurs on a specific entity and alert is automatically generated. This method has advantages over network-wide KPI-based analytics which often miss isolated issues that only affect specific parts of the network. Highlighting such anomalies helps to cut human intervention to a minimum so that only the most severe issues need to be managed manually. The goal is to have a self-healing network, and for this automation is the key.


Network Insights

Network Insights are about delivering a superior customer and service experience to the end-user. This means proactive troubleshooting and anomaly detection. In order to gather these insights for the 5G core, the operator will need to collect data from multiple sources including network events and counters alongside service assurance probe data and network visibility data. This data will then need to be smartly correlated and analysed to highlight anomalies in the data and service degradations. Again, automation will be the critical enabler to making these happen. Automatic alarms will trigger when service degradations occur. Operators will also be able to detect trends in customer experience and reduce potential churn using AI and machine learning.

The growing trend towards encryption in 5G has left operators with a blind spot regarding their users’ Quality of Experience. Automation, and the use of AI and machine learning will again be critical for operators in gaining insights into encrypted traffic. 5G delivers new and complexed use cases including video streaming, tethering, and gaming, all of which require ultra-low reliable latency. For an operator to meet these service level agreements they need to know how the network is functioning and if degradations occur to be able to perform fast automated drill downs for real-time root cause analysis.

In order to crack this encrypted code, service assurance providers can gather statistics on specific metrics alongside a crawler, which calculates KQIs and, at the end of the session, will prompt the user to enter the real customer experience observed. Thousands of traffic samples are then analyzed by machine learning algorithms, the KPI’s are fine-tuned, and anomalies detected in the traffic patterns.

Automation is in nearly every aspect of an insights solution, from predictive troubleshooting to generating KQIs for encrypted traffic. All of this information is data that can then be fed back into the system and further fine-tune the algorithms which program the automated functions.


Network Data Analytics Function (NWDAF)

A key component for any 5G automation solution is a Network Data Analytics Function (NWDAF) as defined by 3GPP. This is a network analytics capability built into the general framework of the network architecture. Its purpose is to create a centralized data and analytics function, taking raw data from anywhere in the core network and use it to provide analytics information.

Service assurance providers will have to develop a “3-in-1” solution which leverages the same data collected from the network for three main purposes. The first is to develop NWDAF-based automation and closed-loop use cases according to 3GPP specifications. The second is to create a Northbound interface for NFV-O, OSS and BSS feeds for non-3GPP closed-loop solutions. Finally, it should create user interfaces for semi-automated analytics and troubleshooting workflows.

Together, these create a highly agile and customizable solution, fitting both operators’ needs as their 5G core functions are implemented and as the NWDAF function evolves.



It is clear that automation permeates network intelligence in its entirety. Any closed-loop, smart visibility solution will have many automated network elements and components. How the operator utilizes automation is their challenge. Ensuring they monitor and fix their network with minimal disruption while gaining highly accurate insights is a challenge which will need to be addressed. Perhaps automation is the answer to that too?


RADCOM Network Intelligence is a fully automated and containerized solution for 5G. The combined solution is a dynamic and multi-functional. To learn more about how RADCOM delivers an intelligent, containerized on-demand solution for network analysis with full network visibility from the RAN to the core click here.