It’s no secret that 5G is big tech news these days. As an analyst, I spend considerable time with the largest carriers and cellular networking infrastructure providers in the world, including their silicon providers. I also spend time with the CBRS Alliance as well as the Linux Foundation networking and edge working groups. In all of these interactions it is a challenge to navigate the alphabet soup that is carrier networking. It’s a confluence of hardware, software, topology considerations and spectrum coverage, littered with three to five letter acronyms. After the recent cancellation of Mobile World Congress in Barcelona due to coronavirus concerns, my principal Patrick Moorhead and I had quite a laugh attempting to decipher the slides presented in some of the follow-up briefings. Today I will attempt to provide a decoder ring to help make sense of it all, pulling from several articles that I’ve written in the past.
Core and radio access network elements defined
Let’s start with the basic core and radio access network elements. Core network components serve as the central part of a cellular network, knitting together mobile, fixed and converged connectivity to ensure a more consistent user experience. For 5G, they include a higher degree of hardware disaggregation from a compute and storage perspective as well as software programmability over 4G LTE networks. Companies that deliver industry-standard server platforms such as Dell EMC, Hewlett Packard Enterprise and Lenovo have made significant inroads into the telecommunications space, and have disrupted the sector from a cost and deployment agility perspective. Some of the more recent capabilities to come to core networking include machine learning (ML), artificial intelligence (AI), virtualization (NFV) and software-defined networking tools (SDN). There is a degree of whitewashing with some if not all of these platforms, but the benefits are real. They include faster time to deployment, self-healing for improved uptime and network slicing aimed at guaranteeing the quality of new service and new monetization opportunities for carriers and service providers.
Radio access network (RAN) components, on the other hand, play an important role in how your smartphone or mobile device communicates across a cellular network. RAN includes base stations and antenna arrays. Base stations are fixed points of communication within a cellular network that are designed to cover a specific geographic area. Based on the need for radio coverage, they can take three forms: macro-cells, which cover a wide area and are typically found on towers, micro-cells, which are used for densification of coverage in highly populated areas and can be found mounted to street light poles and traffic signals, and pico-cells, which boost coverage within buildings such as the Sprint Magic Box.
From an outbound perspective, these various cells communicate with smartphones, IoT devices and sensors and, soon, automobiles. From an inbound perspective, they connect to the core network. Antenna arrays amplify coverage and can be attached to base stations that serve as both main radio units (RUs) and remote radio units (RRUs). For 5G, there are also new requirements, such as Massive MIMO, needed to deliver significantly lower latency and improved throughput. MIMO stands for “multiple-input, multiple-output” and combines a large number of antennas to improve broadcast efficiency. Base stations and antennas combine to constitute what’s referred to as the radio access network, which connects devices to a core network.
Have you ever wondered about the various “flavors” of RAN? Virtualized RAN, or vRAN, refers to the ability to combine multiple 5G New Radio elements via software with a common core to more cost effectively support higher user capacity. C-RAN usually refers to either cloud RAN, a scenario where an operator doesn’t own the infrastructure, or centralized RAN, where an operator needs to minimize the overall footprint of a deployment based on the lack or cost of physical real estate. You may or may not be familiar with OpenRAN, O-RAN, and an effort by some U.S. government officials to mandate deployment. OpenRAN is an initiative launched through the Telecom Infra Project (TIP) association to build radio access network elements upon more open standard, general purpose hardware. OpenRAN utilizes SDN to improve operator agility and reduce capital expenditure. O-RAN, on the other hand, represents a collaboration between the O-RAN Alliance and the Linux Foundation to drive the adoption of open source software into the same radio access portion of the network. The good news is that late last month, TIP and the O-RAN Alliance agreed to form a pact to ensure 5G harmonization and share resources with the aim of lowering development costs. These efforts are compelling, but I don’t believe governments should mandate the use of technology platforms. Rather, the decision should be left to operators making the investment. However, I expect to see this RAN effort blended with more purpose-built infrastructure from the likes of Ericsson, Huawei, Nokia and Samsung as 5G networks deploy around the world.
Deployment topologies, spectrum, and software-defined tools
I’m often asked about “fake 5G,” since some carriers and service providers push the limits of the nomenclature when marketing new services. While some of the confusion stems from densification efforts of current 4G LTE networks, the best way to dissect this query is to examine the topology of a 5G network from a Standalone (SA) and non-standalone (NSA) perspective. SA refers to an end-to-end 5G network that will deliver all of the promises of speed and latency from core to RAN. Meanwhile, NSA allows 5G networks to fallback and rely on some elements of existing 4G LTE networks to bolster throughput as an interim step, typically with an LTE core. With initial 5G deployments globally, most will be NSA since carriers will also have to support two spectrum frequency ranges. Range 1 extends current 4G LTE and is referred to as New Radio (NR). It supports “sub 6” with a spread of 450 MHz to 6,000 MHz. Range 2 sits at a much higher level, with a spread of 24GHz to 52GHz, and is referred to as millimeter wave (mmWave). Each frequency range has certain propagation characteristics. For example, mmWave can cover densely populated areas but only over short distances. Sub 6, although highly occupied today, can be coupled with technologies such as massive MIMO antennas to deliver reliable, cost effective and highly scalable mobile broadband access.
The big bet by many of the tier one carriers in the U.S. is on mmWave, since it will deliver incredible performance and low latency. However, the nature of the higher spectrum band requires more base station deployments, making it a more costly endeavor. In contrast, much of the rest of the world is focused on deploying 5G services in the lower and mid-band spectrum ranges, given its lower cost and wider coverage area. SDN also plays a big role because of the inherent virtualized nature of 5G over 4G LTE. This will allow operators to monetize network slices that guarantee levels of performance for certain applications. These include low latency mobile gaming for consumers, as well as enterprise services such as factory automation, augmented reality (AR) assisted field service and other digital transformation capabilities.
CBRS Alliance and its democratization of licensed spectrum
The Citizens Broadband Radio Service (CBRS) and the related OnGo Wireless initiative promise to deliver a flexible model for the sharing of wireless spectrum in the 3.5 GHz band, which has been reallocated from military radar applications. Spectrum drives the overall capacity for wireless communication and is one of the scarcest and most expensive resources carriers invest in beyond core infrastructure. I liken spectrum to lanes on a highway. The more lanes, the higher the capacity. The game changer with CBRS is that carriers and service providers that want to offer wireless services won’t have to pay expensive spectrum licenses up front for roughly half of its availability under the General Authorized Access tier (GAA). This stewardship by the CBRS promises to level the playing field and should result in massive innovation and new use cases.
OnGo Wireless is positioned to deliver “uncompromised connectivity” for in-building, public space and Industrial IoT deployments. While I believe the recent introduction of Wi-Fi 6 is compelling and will continue to meet the needs of many with its unlicensed spectrum and corresponding economics, OnGo should serve a growing niche for private LTE and eventually 5G networks. Two big private network advantages in my mind are the ability to fine-tune overall network performance for specific application needs (such as lower latency for video) and to support remote and highly distributed worksites that might not have access to fiber or carrier-delivered fixed wireless access (FWA) services. Applications in education, hospitality, retail, manufacturing and transportation are especially compelling.
We love acronyms in the technology industry, and I hope this article helped to define some of the alphabet soup tied to carrier networking. I’m often reminded of the movie “The Imitation Game,” which chronicles the story of mathematician Alan Turing’s effort to decode the German Enigma codes during World War II. Its comparison to 5G may be a distant parallel, but one thing is certain—the next generation cellular networking standard will unlock transformative use cases for consumers and enterprises alike.