Automation is a critical component of any smart, containerized, closed-loop assurance solution. For the operator, the challenge will be how to utilize automation within their networks and deliver the valuable and complex new use cases that 5G has to offer.
5G deployments are well and truly underway. As of March 2020, there were 106 known 5G deployments by Mobile Network Operators (MNOs) spread across seven different regions. This does not account for the private network implementations, which are also taking place in increasingly more significant numbers.
BT has confirmed that they are fully managing the UK’s first live 5G factory for Bosch. The factory utilizes Artificial Intelligence, machine learning, robotics, IoT, wearables, big data analytics, and augmented reality for a fully automated, intelligent, and dynamic manufacturing process. Lufthansa Technik has gained a spectrum license for operating a private LTE and 5G network to enable remote inspections of engine parts. This avoids having to completely dismantle the engine and send the parts to a lab for inspection, saving time and resources.
To ensure these deployments are agile and efficient, operators are utilizing Containerized Network Functions (CNFs), which among other things, enable smooth orchestration and interoperability between functions and scalability. Many operators are exploring the use of public or hybrid clouds to provide more business and commercial agility. Orchestration is performed by Kubernetes (K8) to deploy and configure these CNFs, which is what facilitates the network-wide automation.
However, if operators want to deliver the levels of service required to keep customers happy in 5G, they will need an automated assurance solution. In this blog, we will investigate the role automated assurance has to play in 5G and how operators can leverage assurance to deliver some of the most valuable use cases.
Automated assurance for 5G
5G means a significant increase in the quantities of data, devices, and traffic on the network. To effectively monitor the network, operators need an automated assurance platform that is fully cloud-native, meaning it can be deployed on public or private cloud infrastructure, containers or virtual machines, bare metal, or any combination of these.
In order to deliver the complex use cases, a new 5G core architecture has been created. As a result, operators need to collect data not only from packet feeds using virtual probes, but also from event-based feeds (Network events and Event Detailed Records). Therefore, operators require the ability to correlate data from multiple sources, meaning that different types of data are collected from different parts of the network in order to deliver a complete end-to-end solution.
As part of the new network architecture mentioned above, automated assurance must be ready for a Service-Based Architecture (SBA). This means supporting both Service-Based Infrastructure (SBI) and non-SBI interfaces as well as Control and User Plane Separation (CUPS). This offers the ability to decipher encrypted traffic by collecting and combining information from multiple sources and multiple event types, which is then processed and decrypted, if necessary, to deliver network insights leveraging Machine Learning and AI.
Another critical feature for automated assurance is using a microservices architecture, which allows efficient scaling, updating, and even the complete replacement of any of the assurance components. Additionally, Lego-like assurance architecture enables smooth orchestration and seamless integration with K8. As mentioned, K8 controls the containerized functions, which is essential to delivering automation and a closed-loop environment.
In order to deliver the closed-loop environment, operators must deploy an assurance solution that utilizes cutting edge technologies in AI and machine learning. 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. While still in its infancy, a closed-loop approach is an end goal for operators as it will be an essential requirement for managing their 5G networks. Without automated assurance, the network would not be able to recognize the anomalies in the network data and automatically heal without human intervention.
A fully cloud-native assurance solution will also enable automated service discovery, automatic payload discovery, automated tracing policy configuration, and automated enrichment data discovery. This is accomplished via integration with the Network Repository Function (NRF) in 5G and your assurance platform’s Web API. The NRF provides real-time information about 5G network functions so that the assurance solution can subscribe to, and store data in a central location, enabling the assurance solution to keep up to date with the network inventory at all times.
This functionality also works the other way. So that network functions can “discover” the assurance solution, which enables the network function to register via a Web API to the assurance components and ensure data collection and tracing policies are updated.
Valuable use cases for the operator
Having an automated assurance solution is the key to unlocking some of the valuable use cases which 5G has to offer. Aside from enabling the operator to deliver an enhanced customer experience, automated assurance potentially opens new and lucrative revenue streams for the operator. Indeed, automated root cause analysis and anomaly detection will help the operator to solve network issues in real-time, which will improve the customer and service experience proactively.
As mentioned, an automated assurance offering delivers end-to-end network visibility, meaning from the RAN to the network core. Approximately 70% of all network issues occur in the RAN, which, if not fixed, can lead to service outages having a critical impact on the customer experience. Operators looking to avoid this pitfall must deploy a cloud-native automated assurance solution that can collect data from the RAN as well as from the network core. Without this data, operators cannot draw a concrete conclusion to the root cause of issues. Therefore, RAN monitoring becomes a critical use case in 5G for delivering improved customer experience.
Perhaps the most talked about use cases is that of network slicing, which in turn enables its series of advanced use cases. To deliver network slicing, the operator must deploy a Network Data Analytics Function (NWDAF), which is explicitly designed for the centralized collection and analytics of information in 5G, acquiring data close to the source, and seamlessly integrating with an operator’s cloud environment. In Release 15 the initial use case for the NWDAF was defined as network slicing, serving the Policy Control Function (PCF) and Network Slice Selection Function (NSSF). Now as part of Release 16 the following consumers of NWDAF data have been added alongside the PCF and NSSF:
- Access and Mobility Management Function (AMF)
- Session Management Function (SMF)
- Network Exposure Function (NEF)
- Unified Data Management (UDM)
- Application Function (AF)
- Operation, Administration, and Maintenance (OAM)
Additional closed loop use cases can be implemented based on NWDAF data such as:
- Load level information
- Service experience
- NF load
- Network performance
- Abnormal behaviour
- UE mobility
- UE communication
- User data congestion
- QoS sustainability
In addition to 3GPP-defined use cases, proprietary automated closed loop use cases may be implemented based on the NWDAF which can be extended above and beyond 3GPP defined interfaces and APIs.
The ability to slice the network into virtual portions and deliver specific SLAs for different services is one of the most exciting aspects of 5G. With network slicing, an operator could feasibly provide Ultra-Reliable Low-Latency Communication (URLLC) for use cases such as driverless cars, while at the same time offering super-fast download speeds for gamers and avid video streamers. Recent global events have also highlighted the need to use a portion of the network for Mission-Critical Communications, which is a super assured area of the network ensuring governmental and emergency services can communicate effectively in times of national crisis. These complex use cases require sophisticated monitoring, which is only possible with an automated assurance solution. The end-to-end view of the network enables the operator to balance the network and optimize resources where they are required ensuring the SLA for each slice is met.
Conclusion
5G is complex and delivers with it many challenges. However, these challenges are easily overcome when you bring automation and automated assurance into the picture. Having an assurance platform that can correlate data from multiple sources that seamlessly integrate into an operators’ cloud environment will deliver a full-end-to-end containerized solution.
RADCOM is the leading provider of cloud-native, containerized, and automated service assurance solutions with AI-driven insights and complete network visibility. RADCOM ACE delivers Automated, Containerized, and End-to-End visibility of the network.
This blog post may contain forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. To read more about forward-looking statements, please click here.
