Remarks of David J. Redl
Assistant Secretary of Commerce for Communications and Information
July 24, 2018
It’s Getting Crowded in Here! Understanding the Spectrum Implications of IoT and 5G
Thank you, Keith.
I’m very pleased to speak to you today, and I thank NTIA’s Institute for Telecommunication Sciences and NIST for the opportunity to join in a conversation about the important work we’re doing on spectrum research to help us grapple with the technical and policy challenges of our increasingly connected world.
First, I want to take a moment to express my gratitude for Keith Gremban’s leadership of ITS. Director of a Federal lab is tough job. You have to not only have a solid grounding in the sciences, but also be able to share that technical expertise with the lay person in a way they actually understand. Keith is one of those rare talents. He’s also a great leader, and a terrific evangelist for the important work taking place here that will play a role in shaping our digital future. Thank you for your public service.
Last month, NTIA hosted a spectrum symposium to discuss the Administration’s approach to spectrum policy. As I said at the time, we are taking a strategic approach toward our spectrum goals that will engage all stakeholders. We hope to soon put forth a framework for building a longer-term, more sustainable approach to spectrum policymaking, recognizing that we will have to look beyond the old model of redeploying spectrum from one use to another.
That’s not to say that we are abandoning our tried and true approach. A big part of our focus will still be on ways to find more spectrum for commercial services, including our push to ensure that America wins the race to 5G. But increasingly this will have to be done through spectrum-sharing scenarios, which fortunately are becoming more viable thanks in part to work being done by NTIA’s Office of Spectrum Management and ITS engineers.
The theme of this year’s symposium addresses the challenges we face as we work to make even more from our limited spectrum resources. Simply put, our airwaves are getting more crowded. Analysts at Gartner predict that there will be 20 billion IoT devices deployed by 2020, up from about 11 billion now, and widespread 5G deployment is just around the corner.
As more devices come online and the paradigm shifts to increased spectrum sharing, we’ll be packing more and more into our finite resources than ever before. That will lead to a number of potential problems. These are issues that OSM and ITS engineers are thinking about now, so that consumers, businesses and federal agencies will be able to continue to use the services they need in the future. I’d like to take a look at a few of these problem areas and talk about how we’re approaching them.
First, more devices means more noise. The essence of a successful spectrum system is the ability to discern a specific signal amongst a sea of other signals, and every device that starts using spectrum resources adds to an increasing level of background noise. If you’ve ever been at a crowded bar and had to yell just to be heard, you understand this phenomenon.
The last significant study of radiofrequency noise was done by ITS in the 1970s. Today, ITS is developing and deploying the technology to monitor spectrum activity 24 hours a day, 7 days a week. This spectrum monitoring capability will be used to determine baseline noise levels and track changes, detect and locate sources of interference, and enforce spectrum sharing agreements, among other capabilities. Later this year, ITS will deploy a neighborhood-scale prototype system to monitor spectrum usage in collaboration with the University of Colorado, Boulder.
A university campus is a great place to get a broad sample of different types of scenarios, ranging from quiet times between semesters, to the hectic days in the fall when students return to campus, and extreme crowd situations such as football Saturdays. A college campus also has the advantage of providing a fairly high density and broad assortment of RF noise sources such as lab equipment, maintenance equipment, vehicles, and even drones, in addition to the usual lighting, cell phones, Wi-Fi hot spots, wireless headsets, and public safety radios.
A longitudinal study of noise levels should prove to be interesting and informative. And, there are many other potential uses for spectrum monitoring technology, including for interference detection, occupancy studies, and enforcement. In the future, ITS will be increasing the size and scale of the monitoring effort and the range of frequencies covered.
When we are talking about noise, we are really just talking about one type of potential interference. Interference, in its many forms, is a potentially significant problem in a crowded environment, and one likely to become more challenging in a world where spectrum is more frequently shared.
Within the federal government, we have begun to focus on what we are calling “Interference Prevention, Detection and Resolution” or “IPDR.” Yes – it’s another acronym. But it reflects an effort to capture the full scope of the challenge we face in detecting the sources of interference and then moving to resolve that interference.
Prevention is a key aspect of this work. In order to head off potential interference before it happens, we must explore ways to make systems more resilient. This can take many different forms, whether it is developing technology to protect against interference or simply setting expectations among users, working in advance we can avoid the time-consuming task of resolving interference after it happens. In a sense, this kind of pre-engineered interference prevention can be considered a form of automated “enforcement” — it just builds the enforcement capability into the radio systems themselves.
As part of our IPDR efforts, NTIA led a workshop of federal users in February, bringing together some 60 spectrum managers from more than a dozen federal agencies. Participants identified areas that needed solutions as well as potential improvements. Ideas that were discussed included the potential of a common database among agencies for interference-incident reporting, as well as researching how automation could be applied effectively in the future, to identify interfering devices in a congested environment or for small-area, high-security events.
We have established a working group that will set priorities and make recommendations on how to proceed in this area. This is a federal-only effort at this point, so it’s just the tip of the iceberg in terms of potential examination of interference issues across the entire spectrum-based economy.
Dynamic Protection Areas
In additional to studying how to resolve interference issues, NTIA is developing tools to make spectrum management more agile. Among these new tools is the introduction of a technique known as dynamic protection areas (DPAs), which will be used to help avoid interference in the 3.5 GHz Citizens Broadband Radio Service (CBRS) band.
DPAs allow regulators to eschew rigid geographic exclusion zones, in favor of increasing the flexibility for multiple services and systems to operate while sharing spectrum in the same geographic areas. DPAs in the CBRS band will function using environmental sensing capability (ESC) and spectrum access system (SAS) technologies, which combine the ability to detect a specific system’s signal, then trigger automated responses to prevent other radio systems from interfering with that signal.
In the case of 3.5 GHz, the ESC equipment will be able to detect when a radar system is active in the DPA, and the SAS system will then trigger changes in the way that commercial CBRS systems operate in that area. These changes, or operational adjustments, might include reducing the commercial systems’ transmission power, moving to another channel or even temporarily shutting down some transmissions.
The dynamic aspect of a DPA means that mitigation measures are triggered only when a radar system might receive harmful interference. At all other times, the wireless systems can operate without mitigation restrictions. So the geographic area in which mitigation is triggered changes and “flows” over time to maximize the use of the band.
The bottom line is that the development of DPAs will help make CBRS commercially viable to deploy across the United States. Just a few years ago, the coastal exclusion zones to protect the radar systems would have undercut CBRS’s potential as a national service.
This is an exciting development from both a spectrum management and an economic standpoint. Technically, this is a major step forward in testing and fielding dynamic spectrum management technologies, which we have envisioned for many years. Economically, DPAs offer a pathway forward for commercial services to share spectrum and operate in potentially more bands while protecting important existing federal and other operations. These are exactly the kind of creative approaches to sharing that we believe will be critical to making more spectrum available.
I’d like to point out that the work to develop DPAs came out of a collaborative process involving NTIA, the Department of Defense, the FCC, and industry – all working together to develop a technological solution to a regulatory problem. It’s an encouraging development and could be a model for solving other complex issues in spectrum management.
Our work on interference and spectrum management will help us get ready for the predicted growth in IoT adoption and is especially critical for smart city planners around the country. After all, if traffic lights and traffic sensors are connected, their failure could result in a significant safety hazard. Here is where ITS’ expertise is helping prepare to meet this challenge. ITS engineers are developing a simulation-based tool to enable city planners to investigate the impact of IoT on urban infrastructure.
Essentially, the concept is to model the RF and information characteristics of a range of IoT devices, develop representative scenarios, then employ probability simulations to assess the effects on spectrum availability and capacity as the number and density of devices increases.
Smart cities will be the embodiments of ultra-dense deployments, and urban planners will be looking for the kind of tools that ITS is developing to help them effectively introduce new capabilities.
While we’re excited about the promise of IoT, we’re also taking a close look at the security issues that accompany the rapid expansion of this technology. Specifically, botnets and the cyberattacks they enable pose a significant threat on a major scale because of the sheer number of unsecured devices. Earlier this year, the Departments of Commerce and Homeland Security delivered a report to the President on how to improve the resilience of the Internet and reduce the threat of botnets. The Departments are now working on a draft roadmap of actions that the government and private stakeholders can take to make our networks more secure against botnet attacks, which we expect to release in the fall.
Last week, we took a first step forward by launching a multistakeholder process to develop greater transparency of software components. What we’ve recognized is that many companies are adding “smart” features or connectivity through software that’s made up of existing components, modules, and libraries from the open source and commercial software world. And consumers, particularly enterprise consumers, in many cases don’t know exactly which components are in the software in the devices they employ, which can create obstacles when protecting against security risks. We’ve brought together software vendors, purchasing organizations, and security solutions providers to encourage better coordination and increase awareness of solutions and new approaches.
NTIA has used these open processes to make progress on a number of issues, including cybersecurity vulnerability disclosure, secure updates of IoT devices, and providing more transparency about data collected by mobile apps. If anyone here feels like they have important contributions to make in the area of software component transparency, I invite you to participate – we welcome your input.
There’s a lot to look forward to at this conference, including a discussion about the economic and policy implications of improved propagation modeling later today, and a keynote address tomorrow from Deputy Administrator Heidi King of the National Highway Traffic Safety Administration. It’s always special to hear smart people talk about the technological and engineering developments that will power the next generation of communications and this symposium has many of these opportunities.
We see so often that technology can rise to the occasion to solve technical problems in the spectrum world. For example, Multiple-Input Multiple-Output (“MIMO”) technology does reduce some noise issues. Software defined networking and dynamic user bandwidth allocation will make systems more adaptable, resilient, and amenable to spectrum sharing.
Looking to the future, 5G will support services and applications spanning from Gigabyte throughput to very low latency secure communications. These solutions, however, will undoubtedly create new problems that we haven’t yet identified. This is why the radio science work that NTIA and ITS are doing is so important, and why we’re so lucky to have these organizations working on behalf of the American people.
More than ever, Americans depend on radio technologies in all aspects of their lives. They’re how we connect to each other, how we learn, and how we call for help. The work that this group is doing will improve the lives of Americans – and people around the world – for years to come. Thank you all for being here, I look forward to hearing from all of you over the next few days.