Invariably, discussion of advanced radio access networks (RANs) is dominated by one thing: performance and how to maximize capacity in a service area using available spectrum. As the industry moves to 5G, the need to support reliability, low-latency, high-speed mobility, extreme density and user-level radio resource control adds several new dimensions to the discussion.

In areas of high demand, operators need ultra-dense networks based on a layered architecture containing macro and small cells radiating in multiple different frequency bands. The challenge is how to coordinate the use of these radios to minimize interference and to ensure the user is served by the "best" access cells. This is driving the development of new Centralized RAN (C-RAN) architectures -- also sometimes described as vRAN, Elastic RAN, Cloud RAN, and so on -- that can coordinate radio resource management across the service area.

An enormous amount of work is, rightly, being put into developing these new RAN operating models to ensure 5G networks can meet a wide variety of service types with diverse performance requirements. At the same time, the move to Centralized RAN also has a major impact on performance monitoring and service assurance. And with the emergence of so-called "open RAN" initiatives by groups such as the Open RAN Alliance at this year's MWC, the importance of being able to monitor and assure performance over standard interfaces is now even greater. One area that needs attention, and which has potential to improve service experience, is the fiber plant needed to connect centralized baseband processing to remote radio units. These "fronthaul" links can use passive or active optical transport. Some of the key aspects of this new C-RAN monitoring environment are:

There are several variations on the C-RAN model, each with different fronthaul transport requirements and performance trade-offs. 5G, for example, introduces the Centralized Units or CUs and Distributed Units or DUs with different functional splits. This makes deploying and operating C-RAN challenging in some respects. The extent to which operators, and their vendors, can make progress, the better their chances of making a successful transition to a new RAN operating model that has the cost profile and performance needed to support a new innovation s-curve.

— Gabriel Brown, Principal Analyst, Heavy Reading

This blog is sponsored by EXFO.