Mobile operators are currently engaged in a 5G phony war. Those in saturated markets are eagerly launching 5G services by adding a few New Radio (NR) base stations to their 4G networks, ramping up the data rates for highly selected locations and engaging in a great deal of marketing (some, like AT&T with 5GE, are even labelling 4G+ services ‘5G’).
In reality, these deployments do no add anything that couldn’t have been achieved with Gigabit LTE. The grid of sites remains the same in most cases, as do the applications (faster video, mainly) which the network supports. Most MNOs are sticking to the same pricing models too, as the UK demonstrates – Vodafone introduced unlimited data (with the somewhat interesting variation of speed-based charging), initiating a price war that Three UK, the usual challenger in the UK market, has unsurprisingly confirmed it wil follow (it will offer its subscribers 5G at “no extra cost” compared to 4G).
The lack of new services, which might have generated new revenue, or of a price premium, means two things for MNOs. They need to plan their moves to ‘true 5G’ sooner rather than later. At that point, they will deploy a cloud-native 5G core (current NR Non-Standalone networks use the 4G core), which will have the flexibility and scalability to support a wide variety of different use cases, and to harness network resources with optimal efficiency, matching them dynamically to users and applications. Eventually, a cloud-based RAN will enhance those effects of flexibility and adaptability, and even before that, the 5G RAN will become denser, supporting capabilities of the upcoming next 3GPP standards releases, such as very low latency.
These changes will be essential to the MNOs which are otherwise facing the prospect of deploying 5G to support better quality YouTube viewing, with marginal additional revenue. With the next generation of deployments, especially the cloud-native core, they will have the option to launch new revenue-generating services, especially for industrial users, and to run the network in far more cost-efficient way, while adding more midband and high band spectrum – these evolutions helping to reduce the cost of delivering all that YouTube video and so protecting margins.
Some will chase even greater cost efficiencies by outsourcing the hosting of their cloud-native cores and RANs to public cloud providers, which will truly introduce the scalability of cloud economics to the mobile network. That will be a bridge too far for operators in the first phase of 5G, when the virtualized RAN, in particular, will still need considerable optimization and specialist skills. But as these virtual network functions (VNFs) become more standardized and automated, there will be no reason not to migrate them to the public cloud. Some cloud core suppliers, like Affirmed and NEC, already offer implementations of their products on AWS, for instance.
It may take a decade, but the economics of ‘true 5G’ will be significantly different from those of 4G or the current ‘phony 5G’. But it will require a massive industry effort to adapt to the norms of the cloud world to achieve that, and the transition will often be a painful one.