Between infrastructure and spectrum licenses, mobile operators worldwide will spend over $1 trillion on 5G by 2025, GSMA Intelligence predicts. What will they - and their customers - get for that money? Gigabit speeds, latency as low as 1 millisecond and the ability to support up to 1 million devices per square kilometer, to name just a few key advantages over 4G LTE.
But just as Rome wasn’t built in a day, neither will ‘true’ 5G: a fully standalone network with a cloud-native, service-based architecture. With the exception of greenfield operators, everyone will spend the next several years with a mix of 4G, 5G non-standalone (NSA) and 5G standalone infrastructure. This complexity makes it challenging for operators to provide end-to-end service assurance - critical to ensure a return on that $1 trillion investment.
For example, a big chunk of 5G revenue opportunities come from the enterprise market, where service level agreements (SLAs) will be the norm. To capitalize on those opportunities, operators must be able to ensure that their network meets SLAs for connected vehicle services, mission-critical Internet of Things (IoT) applications and other demanding but lucrative use cases.
Consumers have high expectations for 5G, too, especially as mobile operators position the next generation of mobile as an alternative to copper and fiber access. In this context, the most game-changing aspect of 5G standalone is its latency and speed (as low as 1 ms and up to 20 Gbps), which is why BearingPoint predicts the market for 5G fixed wireless access will generate $58 billion in extra revenues for mobile operators globally by 2023.
Respect the 4G past, move to the 5G future
End-to-end service assurance requires tools that can access, analyze and correlate key performance indicator (KPI) and key quality indicator (KQI) data from sources that are spread across a multi-technology, multi-domain and multi-vendor environment. 5G’s virtualized and cloud-native network functions create an even larger, diverse set of variables that affect service quality, including cloud infrastructure, also known as the ‘undercloud’. Operators must be able to detect issues in the undercloud and correlate them with network and service quality impacts to fully manage performance and troubleshoot issues.
In addition to providing operators with a single, integrated view of the entire 5G network and service performance, end-to-end service assurance tools help ensure a good user experience during handoffs between 4G and 5G. These tools also make it possible for operators to determine whether the new 5G standalone core is affecting 4G network customers. That’s important because even as they deploy 5G, operators will continue to investing heavily in 4G upgrades: about $275 billion through 2025, according to GSMA Intelligence.
These are just a few examples of the juggling act that operators must perform as they evolve to 5G standalone service assurance without ditching tools that still work - all without creating a "Frankenstein" solution in the process. One smart move: focus on creating a set of modular andadaptive service assurance functions. Some of those functions will come from their existing 4G (and even 3G/4G) service assurance solutions that can be used for 5G, too. The biggest challenge will be support for northbound application programming interface (API) requirements for orchestration, automation and real-time KPI delivery. Operators will need precise, split-second insights and in context with the network or service, which traditional static approaches can’t deliver.
Another smart move: take a “stepped” approach (Figure 1), where assurance components are added incrementally and selectively. This accelerates the move to an open, multi-vendor, orchestrated 5G network while also helping to optimize 4G and 3G networks.
Operators also need active, passive, fiber and RAN monitoring solutions that fit seamlessly into any cloud or physical network operations environment. These solutions augment existing service assurance capabilities and use standard, open streaming interfaces, like Kafka, to provide results. Open APIs and flexible workflow engines powered by artificial intelligence (AI) will be required to provide automated insight for orchestrated networks.
Figure 1: At each step, 5G service assurance is essential to reduce risk and accelerate reward.
By following this strategy, operators can quickly build a “fit-for-purpose” 5G service assurance platform while also laying a solid foundation for automated operations. For example, the combination of APIs and a Kafka data bus enables real-time insights that machine learning can use to detect, correlate and alert operators of service-impacting faults and anomalies. AI can then be used for problem detection and diagnostics to support closed-loop control, helping operators avoid being swamped by the enormous volumes of KPIs that a 5G standalone service assurance solution can generate.
AI-powered automation will be increasingly necessary for meeting SLA requirements and maximizing customer satisfaction. By analyzing data faster than humans can, AI has the potential to cut both mean-time-to-detect (MTTD) and mean-time-to-restore (MTTR) by half. That’s a future everyone can look forward to.
