Realizing the Vision of Everywhere, Anytime Communication

Posted by Dan Reed
Corporate Vice President, Technology Strategy and Policy & Extreme Computing Group

Mobile and intelligent devices have become essential everyday tools in most of our lives as evidenced by the nearly 5 billion active cell phones in the world today and the rapidly emerging Internet of Things.

For many of us, it’s hard to remember a time when we couldn’t immediately respond to a time-sensitive e-mail,  access the latest video from a smartphone or connect our phones to our cars or other devices. However, the reality is that this seemingly seamless functionality comes at a cost. Every new download, upload or connection adds strain on a wireless communications network which is based on usable radio spectrum. That spectrum is intrinsically limited by the laws of physics and practical economics.

This week, in an effort to address the limitations of the current approach to managing wireless spectrum, the Federal Communications Commission will obtain comments in response to its Notice of Inquiry on dynamic radio spectrum access. We strongly support the FCC’s commitment to exploring more intensive and efficient use of the nation’s radio spectrum. It’s critical that we transform how spectrum is managed to accommodate the explosive growth in wireless data traffic that we’re experiencing today and that we anticipate will grow exponentially tomorrow (see chart).

Projected wireless data traffic growth through 2014 (source: FCC).

In support of the FCC’s efforts, Microsoft submitted comments highlighting our belief that the FCC must consider not only reallocation of spectrum for licensed use but also innovative technologies – such as dynamic spectrum access – that can make higher and better use of limited spectrum.

The Notice of Inquiry period builds on the FCC’s ruling in September, when the commission unanimously voted to finalize rules enabling the use of TV white spaces for wireless broadband connectivity. We applaud the FCC’s Inquiry, as it has the potential to allow more consumers to have broadband access and spur the kind of growth and innovation in wireless technologies that leads to true everywhere, anytime communication.

Microsoft has spent many years investigating new technologies that can increase productive use of spectrum, and we are committed to fostering the needed regulatory shifts that will make more underutilized spectrum available. Our ongoing “WhiteFi” trial project on our Redmond, Wash. campus was one of the first white spaces-based networks to use an experimental license from the FCC and covers the nearly 600-acre campus.

In addition to our comments, Microsoft Research sponsored a paper written by a leading academic researcher, Professor Dirk Grunwald of the University of Colorado, Boulder. Professor Grunwald’s paper expands upon the FCC’s developing belief that regulatory policies will need to promote greater spectrum efficiency through the use of new radio technologies.

He argues that the best targets for novel spectrum management techniques are underutilized spectrum bands (such as TV white spaces) that abut current unlicensed allocations. He also provides several examples of other spectrum bands that the FCC could open up for secondary uses—such as the 3550 to 3650 megahertz band because of its proximity to the lightly licensed 3650 to 3700 megahertz band.

As Professor Grunwald argues, “Revolutionizing the way that spectrum is managed is essential to meeting the future need for high-speed wireless broadband and enabling new wireless applications. This will entail adopting policies and technologies that allocate spectrum at different times, locations, and frequencies more efficiently.”

It’s important to note that the challenge of effective spectrum allocation is not specific to the U.S. market. It is an issue that affects our global economy and interactions. Last week, I had the privilege of speaking to the International Telecommunications Union in Geneva, Switzerland about the growing global strain on available spectrum, where there were thoughtful ideas and exchanges on the shared problems of spectrum management. In speaking with others in industry, it is clear we all see the challenges ahead, although our proposed solutions may differ.

It’s critical that policymakers around the world recognize this dilemma and pursue nimble and efficient approaches to the developing spectrum crisis. Real-time spectrum access is key, as is allowing access to more spectrum using a mix of spectrum allocations, enabling complementary shorter-range and wide-area networks and increasing reliance on automated solutions. Adopting these advances will enhance the user experience by enabling literally billions of new devices as they go live over the course of the next decade on the Internet of Things.

In many countries, including the United Kingdom, Finland and Japan, researchers are developing trial technologies that explore how to facilitate more dynamic spectrum sharing. These include cognitive radios, online databases that automate spectrum allocation and data networks that can dynamically modulate their transmission power.

Everywhere, anytime communication is a notable result of recent computing advances, but it’s dependent upon available bandwidth, and that bandwidth is finite. Spectrum is, in many ways,  like a natural resource that has to be managed judiciously, especially if we are to continue to advance the digital economy and leverage technology to drive innovation—to create new services, new business models, new ways of communicating and living for the betterment of society.

Comments (1)

  1. Robert Syputa says:

    In addition to the enabling wireless access and database spectrum provisioning technologies, any wireless market requires vibrant supply ecosystem and 'adoption mechanisms'.

    Success in adoption and QoS that aides it is only partly dependent on the frequency band.  Frequency sets the stage because it impacts the signal propagation characteristics that greatly impacts the user experience. However, a supply ecosystem has to develop that weeds out the compatibility over large coverage areas and propels critical mass that results in lower unit pricing.

    Mobile operators have argued, with some success, that in order to get the up-front commitments to capital needed for wide area deployments, spectrum must be licensed in ways that benefit them.

    From a different perspective, studies of spectrum use across wide range from 200MHz to nominally 10 GHz has shown consistently that Wi-Fi achieves the highest concentrated utilization rates: on a per Hz, per unit area (when measured with sufficient granularity), per time basis, 802.11 Wi-Fi achieves 3X-5X higher utilization of spectrum than 2G-3G or what is expected of early 4G networks (those before massive use of microcell architectures , eg. any current plans and deployments).  During MCW in Barcelona, members of CTIA including AT&T said that the industry  should make more use of WiFi than femtocells because it has shown to be cost effective, significantly due to the fact that it makes use of unlicensed spectrum. Like femtocell, Wi-Fi is typically user deployed and optimized.  users can figure out where to put APs to get good reception.. and that doesn't cost operators.  Users benefit, operators benefit.  Sounds symbiotic until you account for the fact users pay for access at both ends.  Why wouldn't operators love the allocation and use of more unlicensed spectrum?  Despite the conclusive results, we all know why.

    As you mentioned, the FCC is pursuing licensing of unfettered blocks of spectrum because the government desperately needs the money to be diverted to other (than domestic infrastructure) uses.  As much as many think it should be clear that wireless access is among the most rewarding uses of OUR spectrum, the ledger shows we are in debt and thus the money will tend to be plundered.  

    The FCC has considered freeing more license-exempt spectrum.  I don't have a current understanding of how discussions have progressed (or dropped off the table), but 30-90 MHz was discussed last fall.

    How does the gap between what operators want, what the government must have (more money to spend or pay back debt), and what is supposed to be the overriding purpose of the use of 'the public's' airwaves?

    I have suggested to various groups that one feasible direction is to advocate for paired use of license-exempt and licensed spectrum where it can be organized into nearby or complimentary blocks. If needed, such as would be the case for cognitive radio/adaptive spectrum utilization, standards groups and suppliers would be required to develop dual/multi-MAC approaches.  However, this is already the case with proposed systems and chips to be used for WiMAX+LTE, TDD-FDD LTE, and proposed TD-LTE systems and devices that would be compatible with FDD and conventional licensed access networks.

    While it is technically feasible to develop white spaces devices and systems that use the same SDR platforms and would allow roaming of devices across licensed networks, there is no requirement for commercial compliance.  

    The nature of the incumbent industry is that of 'The 3G BORG': it assimilates the technologies and business methods that it finds attractive by taking them into their sphere of control.  That is only 'human nature' or, extended, a monopolists nature.  Verizon, AT&T and other operators around the globe already have their sights set on using cognitive radio to their benefit.. most likely to similarly offload network choking capacity at little or low cost to drone subscribers.

    A  better way forward, imo, is to seek out paired use of licensed and unlicensed spectrum with the requirement that devices that work on licensed spectrum must also work on the unlicensed portion and be made available on open markets without service contract requirements.

    What drives the market is a combination of the right technology, the right ecosystem mass and momentum, the right amount and ease of use spectrum, and the right amount of deployed capital and ability to roam everywhere.  Having only two or three of the ingredients will fail.