Spectrum in general is underutilized: Rohan Murty
Latest News »
Bengaluru: Back in 2008, 25-year-old Rohan Murty embarked on an unusual road trip. He travelled to various US cities around the Washington, DC metro area on his own, often walking for miles from one city to another. It was far from a thrill-seeking adventure—he was a man on a mission. Murty drove more than a thousand miles through forests and in the middle of nowhere.
Using a technology system he had built at Harvard University—where he was at the time pursuing his PhD in computer science—Murty was attempting to measure the actual usage of licensed and unlicensed spectrum—which is typically used by telecom companies, TV and radio stations, as well as for technology such as Wi-Fi. He stopped every few miles, in order to validate the correctness of his research and the software he had built.
What he found out during that study would lay the basis for his research into technology that is referred to as “White Space Network” or White-Fi. In layman’s terms, White-Fi refers to unutilized television broadcast frequencies or “white spaces” that are potentially capable of delivering low-cost Internet connectivity to rural and remote parts of a country. “What I found out during that trip was that spectrum in general is severely underutilized,” says Murty.
Murty jokingly recalls that at one point, he was even stopped by a state police officer, who wondered why some “scrawny, brown kid” was walking around suspiciously with an unusual technical device on a US highway.
A vocal proponent of White-Fi—which he believes has the potential to deliver affordable Internet connectivity to hundreds of millions of Indians across the remotest parts of the country—Murty is of the view that the Indian government needs to make a stronger push for White-Fi and does not agree with the current stand of top telcos such as Airtel and Vodafone, which have so far opposed the use of white spaces. Companies like Microsoft have also in recent months urged the government to fight for White-Fi and oppose the telecom lobby, citing the examples of countries like the US and UK where unlicensed spectrum usage has been permitted.
In an interview, Murty, son of Infosys Ltd founder N.R. Narayana Murthy and a junior fellow at the Society of Fellows at Harvard University, spoke primarily on his research on White-Fi, how his work focused on architecture for dynamic spectrum access in wireless networks and other work that he is doing on the sidelines. Aside from helping manage investments for the Murthy family office, Catamaran Ventures, he has also made some angel investments under the radar—including a small investment in Bengaluru-based start-up Floh, which is a singles’ network that helps urban and educated singles find life partners.
What do you make of current debates around spectrum usage in India? The telcos have so far fiercely opposed the use of white spaces and unlicensed spectrum...
Let me zoom back a little. Historically we’ve always thought of spectrum as a binary sort of thing—either somebody has exclusive access to it or nobody has exclusive access to it. So, it’s either licensed or unlicensed. Example of licensed, is TV, radio, cellphone companies, etc. Wi-Fi is an example of unlicensed spectrum, it’s the technology that Wi-Fi operates over—that’s one example. Wi-Fi is just a piece of technology though, not a piece of spectrum. Now, let’s take an example—let’s say you have a wire between your computer and the Internet. And you pump data through. Think of the wire as a pipe and data as water. The more volume of water you pump, the more data you are pumping through. But there is a limit before the pipe explodes, in some sense. So, what do you do when the pipe saturates? All you have to do is instal one more pipe—or in this case, one more wire. So, in some sense you can have infinite bandwidth—because you just keep installing more wires. And that works. But when you move to the wireless world, this is not the right analogy. The wireless world is like a highway—the width is fixed, you can’t increase the width at all. That is a limitation of nature. So, the capacity of a wireless world is bound—there is a maximum you can extract out of it. So, now you have these lanes on the highway—when you licence some part of the spectrum to someone, you basically say “hey, this lane is yours”. Nobody else can touch it. And then there is another lane which is for the public and they use Wi-Fi for it. Now, what happens is that more and more cars are getting into this lane and there is only one lane. And we start saturating this lane. What do we do? We look across at the licensed lane, which is registered to someone and we see that nobody is using it frequently and it’s just sitting there. But the law states that you can only use this one lane. And all this while more and more cars are coming and we feel there is scarcity. But the truth is that this is artificial scarcity. So, the work I ended up doing was to see if those unused lanes could be used without interfering with someone else’s licensed lane and on how to detect the presence of others in that lane. When I detect you’re coming, then I just get out and go back in the other lane.
India seems to be lagging countries such as the US and the UK in terms of embracing the opportunity here…
Absolutely. India is behind in terms of thinking of new forms of spectrum usage. I do believe that is the case. If you go talk to the FCC (Federal Communications Commission) in the US and Ofcom in the UK, just look at some of the progressive things these guys are doing.
In India, we’ve seen companies like Google and Facebook attempt moon shots with permission from the government and provide Internet access through balloon projects such as Loon to remote parts of the country. But we’ve surprisingly seen opposition on White-Fi.
What does the government and the telecom lobby need to do on this matter? Is there a general lack of understanding of how this technology can benefit the masses?
There is no one-size-fits-all solution to this problem. I don’t think the answer is that telcos will solve everything. If they claim that, I don’t agree with that. I don’t think balloons will solve everything either. All these different technologies will play a role and they’ll fit different needs. I think it will fulfil an important need. You cannot solve everything (with just White-Fi) as well. Let’s assume that spectrum operated by Wi-Fi was operated by a telco—almost certainly it wouldn’t have taken off. Every time you would have turned your Wi-Fi modem on, you would pay the telco money. Today you don’t pay anybody anything. So, I don’t agree with the view that telcos will solve every problem.
What is the potential of White-Fi in terms of connecting millions to affordable Internet?
If spectrum is a good owned by the people, then the right thing is you want to raise revenue from it and at the same time want to maximize the usage of spectrum. Let’s assume that you raise revenue from it and the spectrum is actually not being used. Then you have a losing proposition... I don’t know if (White-Fi) can solve all (connectivity) problems, but it can certainly have a positive impact and become one of the many technologies that helps. And fundamentally, in my mind, it alters the way we think about spectrum. These are important policy decisions which will alter people’s lives down the line. I do admire the US on this—they’ve been much more open-minded on this. I think there is a case to be made here as well. I think it is inevitable that there will be an increasing wave of opportunism in terms of spectrum use. It’s like what Andy Grove said once: if you have a great piece of technology, you can send lawyers and bureaucrats to delay it. Ultimately the technology goes around it.
Can you talk about some of your work in this area?
Aside from my thesis on this subject, I even have a follow-on paper that was published. It’s all there in my thesis—it’s called SATYA, which was an algorithm, a system that we built with one of my professors at Harvard, where we showed that even on top of this, you can actually raise revenues. We showed how it lowers the barrier for access to spectrum, while also increasing overall revenues. This will be an inevitable part of our future—and that’s what my work shows, what are some of these challenges and how do we overcome them. And how it will fundamentally alter our view on how spectrum is used.
So, what are some of the challenges facing India?
In this country and every country, our issue is about how do you connect more and more people to the Internet. There are a couple of different ways in which it can happen. One, you assume there is a fibre to every home. But we both know that’s prohibitively expensive to build and to maintain. It’s not easy. Second, everything will be through wireless, such as 4G. I think that’ll work to a certain extent, but as we have seen in other countries such as the UK and the US, if you try to wire up large portions of the country where you don’t have purchasing power, then the economic argument is not justified. In my mind, I’ve always felt there is some other way to connect these people. I think there is room to wire up large parts of the country with some sort of semi-unlicensed spectrum.
So, are we talking about the need for a wide-scale policy overhaul in India on this matter?
Separately, how do you view Reliance Jio’s entry and the current price wars that are brewing between telcos?
I’m a fan of anything where the consumer wins. If I can get 4G everywhere at a reasonable price, I’m happy.