The True 5G Value Proposition: It’s not about Speed and Latency!

By | May 4, 2018

5G FutureIt is common to pin the 5G value proposition on its performance, mainly capacity* and latency. These features will enable applications that operators can monetize. However, I will argue here that capacity and latency are secondary to realizing the full 5G value proposition. Capacity and latency make for big headlines and are easy to grasp by the common subscriber. But the true value of 5G resides elsewhere: it lies in the network architecture of 5G. It is not that capacity and latency are not important. However, if we think of 5G purely from capacity and latency perspective, we could misread how the market could evolve.

The facts about capacity and latency

5G promises up to 20 Gbps of peak throughput and 1 msec latency. By comparison, while LTE provides 1 Gbps and latency that theoretically could reach below 20 msec. But it is important to understand how 5G achieves this step function in performance. Doing so leaves us with a number of questions the answer to which could have profound implications on service provider strategies for network upgrades. 

Capacity is a function of the performance of the air interface. 5G achieves capacity through a combination of optimizing existing technologies and scaling them to new level. For example, LTE started with capability to aggregate up to 5 carriers of maximum 20 MHz. This was later increased to 32 carriers. On the other hand 5G will deal with channels up to 100 MHz in bandwidth across low and high spectrum allocations. The incremental spectral efficiency of 5G over LTE is actually not very high. But the ability to combine carriers, scale antenna systems, and densify the network is a function of the architecture of 5G.

In contrast to capacity, latency is a function of the end-to-end network. 5G brings optimizes the air interface to reduce latency. But more importantly, its flexible architecture ultimately allows 5G to meet its latency targets. For example, the architecture makes it possible to implement edge computing without breaking some of the fundamental paradigms of the technology as is the case in LTE.

The business case challenge

Many are pessimistic on the future of 5G. Even mobile network operators (MNOs) have publicly expressed doubts on the commercial viability of 5G. It is very difficult to estimate the revenue impact of the virtual and augmented reality applications. Similarly, the connected and autonomous vehicle market is still at its infancy. There are many alternative solutions for industrial Internet connectivity, and so forth. All these applications are predicated on the need for high capacity and/or low latency.

The fact is that nobody really knows what end users will be willing to pay and how much service providers can charge for 5G applications. Talking about these and other applications as drivers for 5G while ignoring organic growth of capacity is telling. The promise of 5G is “build it and they will come.” But “they” may never come as it’s been proven that consumers have limits on what they are willing to spend.¬†High uncertainty about revenue potential lead to market jitters. Thus, thinking about 5G purely from the perspective of capacity and latency may lead us to miss-read potential market development.

Source: Huawei

The 5G architecture framework

The value of 5G resides in the core network and not in the access network. In 1G to 4G networks, the access network grabbed headlines with the introduction of digital technology, data services, and IP architecture. Speed was all that mattered. The core network, while important to the practitioners, is not a headline grabber for the average consumer.

This changes with 5G: the value resides in the core network. It is the core network that will allow different services to run cost effectively. Virtualization of the core network is critical to enabling MNOs launch new services cost effectively. The functional distribution and flexibility of the architecture will be instrumental in enabling different deployment scenarios. All this will be difficult to replicate on massive scale in LTE. Edge computing, heterogenous networks are two examples where 5G will provide a step function improvement over LTE.

5G still provides a more flexible radio access network architecture than 4G. But the 5G architecture allows for much greater flexibility to scale networks cost effectively. It also provides a better framework to integrate multiple technologies under one umbrella.

5G value proposition: efficient scalability

The 5G architecture allows scalability much more efficiently than 4G. There benefits will be realized not only in greater set of services and applications, but also in operating costs. The catch is that to realize the cost savings, scale will be important as otherwise operators will continue to extend the lifecycle of the LTE network.

Scalability and cost of operation are two key drivers for 5G. They are perhaps more important than capacity and latency which are products of scalability and flexibility of the architecture. If we think of 5G through this prism, we could draw an alternative scenario to market evolution.

Market evolution

It takes a major player to make a wireless market. In 4G it was Verizon that decided to switch from CDMA to LTE. AT&T quickly followed as its 3G network struggled to meet the iPhone capacity demand. Japan and Korea with significant CDMA networks also aligned on LTE. Combined, they made the market for LTE as the rest of the world joined. By contrast, Clearwire / Sprint adopted WiMAX but that market failed as no market makers participated.

The 5G evolution has been fragmented to date. Verizon is leading with fixed wireless access use case. Japan and Korea are focusing on mobile broadband. China is focusing on IoT. But with the T-Mobile – Sprint merger, this could change. The combination of low-band and high-band (2.5 GHz) spectrum is ideal to kick start a 5G network with all its capacity and latency benefits. But more importantly, it will allow the new entity to scale more efficiently than it could with LTE. The threat to Verizon and AT&T comes not just from spectrum and service performance, but more critically in a competitor that could scale efficiently over the long term. This could play in favor of 5G, and be the catalyst for new series in investment across the industry.

Could the combined T-Mobile-Sprint be 5G market maker? We’ll have to wait and see.


* I use ‘capacity’ to imply speed or throughput.

2 thoughts on “The True 5G Value Proposition: It’s not about Speed and Latency!

  1. David Smith

    My hunch as well is that the 5G core is the key to 5G adoption, but I think there would be a lot of value in a dedicated post laying out key reasons. Network slicing, yes, but why? Core virtualization, yes, but the ePC can be virtualized. We need a better understanding of this.

    1. Frank Rayal Post author

      David, the 5G core allows distributing functions in a way that the 4G core / EPC cannot. The EPC centralizes some functions, but 5G allows them to be distributed. This enables edge compute, for example, in a way not possible in 4G/LTE.
      Good idea for a future post. Will table that!

Comments are closed.