Building the road to 5G requires real backbone
By Paul Brooks, Viavi Solutions
The legend of 5G grows mightier by the day. Network service providers, businesses and even consumers are pinning their hopes on the vision of gigabit speeds to mobile subscribers — not to mention support for the Internet of Things (IoT) and easy-to-deploy wireless solutions for fixed broadband. The race to high-speed networks is evident worldwide, as we witness a dramatic acceleration in gigabit internet to end users. Seemingly everything drives the need for speed. But before we can reach the promised land of 5G, mobile and core networks will require some upgrades to shoulder the burden of the market’s ever-increasing demands for bandwidth.
Approximately every eight years, a new higher-speed premium rate for Ethernet is defined, lighting a path to market for early adopters and prompting development of new generations of optical transport equipment. Just a few short years ago, 100G was the great frontier for backbone links. The standardisation of 100G in 2010 revealed a clear blueprint for building cost-effective interfaces based on technology that wouldn’t mature for another five years. At that time, 25G-based signalling served as the basis for 100G technology.
Now, service providers and their ecosystems are accelerating the development, testing and deployment of 400G. The IEEE standard for 400G, 802.3bs, is expected to be ready around 2018. In the meantime, early adopters can go to market with 25G-based technology, but should have a migration plan to 50G PAM-4 electrical interfaces; all while maintaining the appropriate photonic interoperability. Even with this approach, the path to higher speeds brings considerable challenges:
- Photonic interfaces will use PAM-4 signalling, a significant departure from conventional NRZ (non-return to zero); and
- PAM-4 will require significant improvements in signal-to-noise ratio (SNR) and linearity as well as bandwidth.
Moreover, higher speeds create a highly non-linear test scenario. Newer, more insightful network testing will be required to validate margin and diagnose issues through the coding and PAM-4 modulation. No longer can testing be confined to just one of the layers; it must cover the link from the physical layer through to the Ethernet. Test results need to be able to reconcile where the issues lie and fully validate the margin implications of the forward error correction (FEC) channel. For all of these new testing requirements to work successfully, instrumentation should be able to test and validate all core elements in parallel.
Beyond beefing up mobile backhaul, another crucial consideration when building out 5G networks is planning connectivity for billions of soon-to-be ubiquitous IoT ‘things’ alongside more the traditional services. Smart homes, connected cars and all the many different devices on the edge of the 5G network will have differing propagation characteristics, latency requirements, connectivity and security needs. It’s a delicate balancing act.
A concept called ‘network slicing’ in 5G holds the key to this connectivity conundrum. To enable smooth delivery of voice, data and IoT services running in parallel on the cell site, advances in radio access network (RAN) technology are needed as well, including adoption of network functions virtualisation (NFV) and software-defined networking (SDN) technology. With the migration to NFV/SDN technology, mobile operators can fully leverage network slicing, where multiple cloud-based network functions are automated and programmed from the RAN. In the cloud 5G network layers, or slices, can be designated to meet the conflicting requirements of different IoT systems without additional hardware deployments.
To gain the most flexibility and control possible with network slicing, operators should consider migrating to fibre-based fronthaul with a centralised-RAN (C-RAN) architecture. By moving baseband processing units to a central location, a larger number of remote radio units can be served more efficiently, particularly over a larger geographic area. And, due to its high-power, low-loss and greater bandwidth characteristics, fibre is ideally suited to enabling these multiple network layers for various IoT services.
While it’s true that the road to 5G will present significant challenges due to this inherent network fragmentation, the potential for new and improved service offerings is limited only by our ingenuity and creativity. But we must all work together to build our road on a solid foundation.
• Paul Brooks is Viavi Solutions’ 400G technology lead and a member of the IEEE committee working to develop the 400G standard, which will be essential to support tomorrow’s networks.