October 27, 2017
Earlier this week, I attended the Georgetown University McDonough School of Business and the Wireless Technology Association conference, “Massive Broadband Network Densification: Unleashing the Opportunities of 5G.” Panels focused on the technology, business economics, and policy aspects of fifth generation (5G) wireless deployment in the United States. I participated on the business economics panel, entitled, “Who Pays? Funding America’s 5G Infrastructure.”
As with past generations of wireless service, but even more so, wireline providers will be essential to the successful deployment of 5G networks. In particular, 5G will require increasingly dense fiber deployed closer and closer to end user locations in order to offload the extremely high volume of anticipated data demand from wireless cell sites. 5G will also require hyper-fast, secure fiber connections among the core network data centers that will host switching and other intelligent network management capabilities, as well as the edge network servers that will run 5G-enabled applications.
Deployment of 5G networks will be an extraordinarily complex and costly endeavor. It involves many moving parts requiring significant coordination among network, equipment, and application providers. 5G will be more heterogeneous than previous generations. The constraints of spectrum availability and different geographies, 5G will offer network providers a diverse set of tools, from both fixed and mobile networks, to meet growing demand efficiently where it arises. This means that policymakers have a complex task– they must monitor many moving parts while ensuring network, equipment, and application providers can roll out 5G as effectively and expeditiously as possible. The U.S. was a global leader in 4G deployment and it must remain a leader to reap the social and economic benefits of 5G.
What are the many moving parts that must come together to make 5G a reality? Here are just a few. While I am likely missing some things and vastly oversimplifying for the sake of brevity, this brief survey of the 5G landscape should provide a sense of the complexity of the effort.
-
- First, 5G involves technical standards for transmission and network management in existing and new frequency bands, especially new ultrahigh frequency bands known as millimeter wave spectrum. Providers intend to use these bands for data-intensive 5G applications, mostly in dense areas where demand is concentrated. However, while millimeter-wave spectrum represents a component of 5G, it is only one among many parts.
-
- Second, there is cell densification: the deployment of as many as several hundred thousand small cell sites in order to locate wireless transmission equipment. The millimeter spectrum bands, in particular, enable very large data throughput but have a relatively short transmission range. As a result, providers must deploy large numbers of cells close to end users.
-
- Third, providers will continue to use fourth generation, or 4G, networks for years to come. 5G standards will be backward compatible, and for many years, 4G and 5G will coexist. Today there are several hundred thousand wireless cell sites in the U.S. used to provide 4G and earlier generation wireless services. About 50,000 cell sites today are small cells since there has been some cell densification to support 4G; but most are “macro cells” transmitting over much wider geographies. For example, Cisco projects that in 2021 4G will represent 93 percent of U.S. mobile traffic, while 5G will represent 6 percent. Over time, we will come to rely more and more on standalone 5G service, and eventually providers will retire 4G equipment and repurpose 4G spectrum. However, the macro cell sites and spectrum currently used for 4G services will continue to be part of the nation’s wireless architecture during and after the migration to 5G.
-
- Fourth, 5G will involve an evaluation of the mobile network architecture: in order to gain cost efficiencies and service delivery improvements, providers will virtualize many of the switching and network management functions, moving them from macro cell sites to remote data centers in the network core. At the same time, many computing functions will move from cloud-based data centers in the network core toward the edge of the network, possibly located at today’s macro cells, in order to reduce transmission delays – or “latency” – for real-time 5G-enabled applications.
-
- Fifth, network connections are evolving. Traditional devices, like smartphones and tablets, will need to accommodate new 5G spectrum bands and dynamically choose among them, all the while maintaining reasonable power consumption and consumer prices. At the same time, a greater share of connected devices will consist of the so-called Internet of Things, or “machine-to-machine” connections. These connected “things” will comprise a whole range of devices, from connected TVs to autonomous vehicles to large numbers of low power, low-bandwidth sensors deployed over very wide geographic areas. Cisco projects that from 2016 to 2021, machine-to-machine connections will grow from 1.2 billion to 2.9 billion, and connected TVs will grow to 535 million to 646 million, together representing 80 percent of all connected devices.
-
- Sixth, there will be many new 5G-enabled applications, some of which are reasonably foreseeable, such as wireless video, autonomous vehicles, energy efficiency, telehealth, and virtual reality; and many of which we have not yet even imagined. Applications will have a great deal to say about how providers design and manage networks, with some applications demanding varying degrees of extremely low latency, high data throughput, and traffic prioritization.
-
- Finally, and this is critical, there is deep fiber. Fiber connections from the nation’s wireline broadband providers will be essential to the deployment and continued operational success of 5G networks. It has always been the case that mobile networks depend on wireline backhaul connections to cells cites, with nearly all wireless traffic traversing fixed, usually wireline networks. With 5G, the migration to fiber that began with 4G intensifies. Driving denser fiber deeper in the network will be essential to achieving the kind of throughput and latency required for 5G services and applications. It will also be essential to efficient network management.
What will deployment of 5G networks cost and who will pay for it? It is hard to put a precise number on it, but estimates are in the hundreds of billions of dollars. Data provided by CTIA, the Wireless Association indicates that wireless capital expenditures have increased with each successive generation of wireless. An Accenture study commissioned by CTIA suggests the network infrastructure cost would be $275 billion over seven years. At the same time, an increasing portion of wireline capital expenditures will support small cell deployments and core network connectivity. A Deloitte study suggests U.S. providers need to deploy at least $130 billion – $150 billion of fiber over five to seven years to upgrade and extend the broadband network. The proposed deployment would constitute a multi-use asset, covering consumer, enterprise, and wireless users; it would include at least $20 billion for small cell densification.
Given the high costs, along with the unprecedented complexity involved in deploying this next generation network, policy makers at all levels of government have a challenging task. They must be flexible and forward looking in order to facilitate expeditious rollout. They must recognize the significant benefits 5G and more fiber deployment offer to their constituents. Federal and local policy makers will have to coordinate with standards bodies, network providers, equipment manufacturers, and application developers. They will have to ensure efficient access to resources, such as spectrum and right of way. They will have to clear away existing obstacles, such as excessive siting requirements, legacy network obligations, and discriminatory Title II business and network management restrictions; and they must take care not to create any new obstacles. If we get this right, the U.S. will again be among the world leaders and will reap the economic benefits of global leadership in broadband and information technology.