Let’s spend a moment in the no-man’s-land of broadcasting: all those TV white spaces (TVWS). Together, they represent a substantial amount of unused spectrum available for unlicensed use in TV broadcast channels.
Pockets of vacant spectrum left between television channels could be incredibly powerful: They can be used to deliver widespread broadband internet via unlicensed Wi-Fi-like devices, for example. In contrast to licensed spectrum, TVWS enable these devices to reuse portions of otherwise empty spectrum, creating a cost-effective model for expanding broadband access.
Companies like Microsoft and Google have championed whites space technology, deploying programs and building tools to drive more TV white space utilization. But not all is rosy here. Although public databases from the FCC, Spectrum Bridge, and Google’s Spectrum Database [discontinued] have been designed to report empty broadcasting channels and manage white space usage, these databases are complex and have drawn criticism.
The truth is, nobody has quite figured out how to leverage TVWS well.
We’re sensing something
So, might there be a better option? Can an external source of sensing data validate where vacant channels exist and compare those databases to usage measurements?
Yes. Our sensors are able to detect and measure broadcast television and white space activity, providing an independent, sensing-only component on a massive scale. As an example, we compared white space channels in San Francisco and San Diego. Here’s what we learned:
Our data and algorithms illustrate broad trends, like these, but they also provide more granular insights. As we add more data and increase our sample sizes, we can begin to answer questions like:
Which technologies (3G/4G) and which spectrum bands are deployed by different operators — and in what markets?
Are there geographical differences between markets and operators in the way spectrum is deployed?
How has deployment changed over time?
Or, perhaps more powerfully:
How is an operator’s deployment likely to change in the future?
We know that network forecasting can be tricky, but with more data comes more clarity. We’ll pass on what we learn in future installments and keep you updated.