Analyst Angle: network efficiency evolution from simple access to appropriation for enterprise use cases

Cellular network efficiency increases significantly with each generation, especially as the value chain discovers new technologies to improve spectrum use. The following chart illustrates how 3G, 4G, and 5G started and what their peak achievements have been. The top speed for 5G will likely surpass several Gbps when mmWave spectrum is introduced.

Contrary to 3G and 4G, 5G is expected to introduce support for a much broader spectrum range, including from <1 GHz to 60 GHz, and including licensed, unlicensed, and shared spectrum. It should also be noted that the most advanced implementations of 4G, especially considering pre-5G additions, may not be very different compared to early 5G systems. However, the introduction of new frequency bands, support for mmWave spectrum, and the capability to tie these together is what is going to set 5G apart. Current usage trends indicate that mobile broadband is either used for intensive applications (e.g., video) or bursty, low-traffic use cases (e.g., social media). The design of current LTE networks offers a democratic approach for allocating resources, which depends on the position of the user with respect to the cell site. For example, a user at the cell edge consuming video will likely be allocated a smaller channel compared to a user conducting a mobile chat at the center of the cell. This may have been a suitable approach in the early stages of the mobile broadband evolution, but today’s networks need to handle a heterogeneous set of services, each of which needs to be treated differently.

An analogy can be made with road traffic. In early generations (both roads and cellular networks), roads were few, narrow, and catered to few vehicles or packets. These roads widened and traffic increased with 3G. However, it was with 4G that all roads were replaced with highways and increasing traffic flowed without problems. The difference is that not all traffic needs highways to reach its destination; small roads are still necessary to cater to local traffic or low-speed traffic. The same can be said with 5G, which will create a network of connections that will range from highways to narrow channels, catering to high-speed traffic (e.g., video) and occasional low-speed use (e.g., IoT sensors). In both roads and cellular networks, both types of routes are necessary to cater to increasing traffic efficiently. Moreover, network—and road—control needs to be automated and tightly orchestrated to optimize flows.

This advanced network will also need new devices that can provide a better experience to end users, in a similar way that cars have advanced from simple functions to become data centers on wheels. 4G—and in time, 5G—now require advanced smartphones where even machine learning and AI are being used to improve user experience. If 4G unleashed the power of the smartphone, it could also be said that 5G will introduce immersive AR/VR devices and immersive experiences where the user interface becomes seamless using natural human senses, whether that be voice, touch, or gesture.

5G introduces structural evolution of the cellular network
5G introduces technical functionality that is structurally different from previous technologies and, in a way, disrupts the linear evolutionary methodology exercised from 2G to 4G. 5G is expected to be much more efficient, even when compared to 4G, for both capacity and latency:

• 5G will introduce flexible numerology and extensible transmission time interval (TTI) that will allow for more efficient use of the network resources. It will allow the network to accommodate different services with different performance requirement and QoS factors. For example, while mission-critical services like autonomous driving and tactile Internet will be privileged and served through high-performance network slices (high-speed, low-latency), lower-priority applications, such as massive IoT applications, will be carried over lower-performance network slices.
• 5G will allow handsets or devices to use cloud or edge processing by allowing low-latency connections. Several applications, including AR and VR, require cloud processing capabilities that are not available today due to latencies on the order of 100 ms to 200 ms. 5G will lower these numbers to 1 ms to 10 ms, thus allowing on-device applications to use cloud processing.

The faster nature of 4G and Gigabit LTE has also made the operation of the cellular network more efficient. The availability of a faster connection means that devices need to be “on-net” for less time, thus allowing more time for other devices, and at the same time, reducing battery consumption in smartphones. For example, a 1-hour YouTube video corresponds roughly to 800 MB of data. In an early LTE network where a typical speed of 10 Mbps is achieved, this video would require 10 minutes of an uninterrupted cellular connection that uses the best part of a basestation radio’s resources. In a typical Gigabit LTE scenario, where typical speeds of 300 Mbps are achievable, the same clip would take 21 seconds to download. Apart from the massive improvement in user experience, a higher speed drastically frees up resources to accommodate more users and richer content.

Significant provisions are also introduced for lower latency in 5G, including in radio access and the design of the core network. Latency-sensitive applications can be targeted with localized, edge computing deployments that reduce latency to a matter of milliseconds compared to more than 50 ms today. 5G New Radio (NR) introduces radio physical layer improvements that reduce RAN latency by a factor of 10 compared to LTE, bringing the theoretical limit down to 1 ms. However, the most important aspect is that the network is tightly orchestrated and automated and comes with a prioritization framework.

Network appropriation for enterprise use cases
The structural changes outlined above introduce the ability to appropriate network resources according to each use case, which is also a completely new way to design and utilize cellular networks. These now become enabling platforms, rather than dumb pipes, for use cases that we cannot yet fully understand or expect. To do so, telcos are now beginning their transition from managing individual network domains, or even elements, separately, to treating their network as a platform. The 5G NR interface extends this functionality to the access domain where flexible numerology, flexible TTI, and cloud technologies extend this granularity toward end-user devices.

The capability to manage the network as a platform that extends from the core to the access layer means that future use cases can be catered to, rather than being designed for a static set of use cases that are being discussed today. This is vital for the success of 5G and has also been the reason 2G and 4G networks were vast successes: the first brought voice to the mass market, while the second introduced mobile broadband to consumers. On the other hand, 3G was poised to introduce video calls and Push to X, both of which never materialized. The industry is now fully aware that future technologies need to be open to new types of innovation, most of which will come from the market, rather than the telco or vendor community.

In short, 5G will bring some unique network characteristics, in addition to high bandwidth and low latency, including traffic prioritization, automated operations, and real-time scheduling. These characteristics will enable 5G to offer abundant bandwidth to handle the increasing amount of traffic generated by people and machines. 5G will also allow this traffic to flow faster, more efficiently, and more reliably. Thanks to these characteristics, 5G will enable the creation of a very highly productive environment across a broad range of industries.

In conclusion, cellular technology innovation has been critical in creating new business opportunities and in improving global economic growth, and this is just the tip of the iceberg. 5G is expected to be one of the most innovative technologies out there, enabling substantial economic growth across various industries. It will allow networks to be more efficient, and this efficiency will have a significant impact on business productivity while optimizing overall network resources.

About the Analyst
Dimitris Mavrakis, Research Director, manages ABI Research’s telco network coverage, including telco cloud platforms, digital transformation, and mobile network infrastructure. Research topics include AI and machine learning technologies, telco software and applications, network operating systems, SDN, NFV, LTE diversity, and 5G.

References:

1. Waverman, L., Meschi, M., and Fuss, M. “The Impact of Telecoms on Economic Growth in Developing Countries.” In Africa: The Impact of Mobile Phones. The Vodafone Policy Paper Series, no. 2 (2005): 10–23. See also: http://corpo.videotron.com/static/site/static/documents/en/Mobile-GDP-benefits-07-06.pdf
2. https://www.gsma.com/publicpolicy/wp-content/uploads/2012/11/gsma-deloitte-impact-mobile-telephony-economic-growth.pdf
3. https://en.wikipedia.org/wiki/Arellano%E2%80%93Bond_estimator
4. https://www.gsma.com/mobileeconomy/
5. https://www.qualcomm.com/invention/5g/economy