The 5G fork in the road for network transport—and which path to take (Reader Forum)

0
18


The hype of 5G is ushering in the potential for drastic network change and is driving service providers to prepare for the future. Today’s networks provide ubiquitous all-purpose IP connectivity to support all services. 5G, by definition, supports three major service classes: enhanced mobile broadband (eMBB), ultra-reliable and low latency communications (uRLLC), and mass machine-type communications (mMTC) – and must have the ability to provide configurable levels of isolation between each service transported. 

The question facing mobile network operators (MNOs) is, what path do they take to help support these new services? While the answer depends on organizational beliefs and business timing, three distinct strategies emerge for 5G network transport.

Keeping it business as usual (at least to start with)

This strategy is underpinned by the belief that the only 5G service being deployed in the near future is enhanced mobile bandwidth. This class of MNO is extending their network’s IP-connectivity to support higher bandwidths, adding routing, switching and optical transport capacity and connectivity where required. These providers survived in the pre-5G era with this strategy and plan to stay the course. 

This approach enables MNOs to buy time to uncover the next killer 5G service before committing to significant network modernization investments. However, it may be difficult for them to catch up to competitors in the long term if the more advanced 5G services take-off in the short-term. 

So, what’s the next best option? 

A proactive approach to network modernization 

What separates MNOs with this strategy from their business as usual counterparts is the understanding that, in addition to increased bandwidth, early variants of new 5G services, which require URLLC, mMTC and service isolation, will start to be deployed in the near term. Hence the network rolled out today for enhanced broadband services must also be flexible enough to rapidly support these new services.

In this type of network, the IP layer handles all service functionality, including network slicing, while Dense Wave Division Multiplexing (DWDM) supports physical connectivity where needed, such as longer, higher capacity routes. 

This strategy takes a proactive approach to network modernization and brings together a fully functional IP layer with a simple DWDM transport layer, reducing capital and operational expenditures. In theory, mechanisms at the IP layer can meet all of the 5G service needs and the optical layer just provides big transport pipes for these services, resulting in a network which is less complex to operate. 

But the reality is much different. Even with new deterministic IP transport mechanisms like segment routing, it is difficult to fully guarantee service isolation and complete deterministic behavior for all operational scenarios. 

So what about an even better option? 

Make room for multilayer optimization 

Like the IP for everything camp, the final strategy of multilayer optimization focuses on future-proofing networks so that they can support services which require deterministic behavior, low latency, IoT and service isolation as well as enhanced bandwidth. 

The objective of this approach is to fully guarantee service isolation and completely deterministic behavior under all operational scenarios. Achieving this requires seamless integration between fully functional IP and optical transport layers. 

The IP layer remains the same as the previous approach. However, a fully functional optical layer is introduced that supports OTN, DWDM and ROADMS (reconfigurable optical add/drop multiplexers). The introduction of Flexible Ethernet (FLexE) provides a flexible lower latency deterministic mechanism for mapping IP transport onto the inherently deterministic optical transport layer. 

The benefit of this approach is threefold. It reduces costs by optimizing resources across the IP and optical transport layers. TDM mechanisms such as FLexE and OTN provide deterministic service behavior and ensure hard isolation between services. And, with the right tools and expertise, it maximizes revenues by offering a complex suite of services that feature different levels of bandwidth, latency, availability and duration.

Of the three strategies listed, multilayer optimization is likely to be the best approach, combining economic efficiency with multi-service transport effectiveness. Moreover, it offers the ability to deliver stratified communications services that emulate the successful cloud services model. This approach should be strongly considered for MNOs looking to unlock the flexibility needed to engineer a 5G multi-service network and reap the rewards of its associated revenue.



Source link