- Interview with Dr. Ted Rappaport of Virginia Tech and Wireless Valley
- Interview with Dr. Ted Rappaport of Virginia Tech and Wireless Valley, Continued
Like this article? We recommend
Like this article? We recommend
Like this article? We recommend
Interview with Dr. Ted Rappaport of Virginia Tech and Wireless Valley, Continued
Question: Interesting stuff! Now, could you tell us some about the companies you've started that resulted from MPRG research?
In 1989, I started a company with a couple of great students, and we called it TSR Technologies, Inc. It was a cellular radio/PCS manufacturing firm, and in 1993 we sold it to Allen Telecom and it became Grayson Wireless, which is now a big test equipment maker.
Wireless Valley was a spinoff I did a couple of years ago with a PhD student, Roger Skidmore, who's now running it. It's pretty exciting because it's leading the way for in-building wireless measurement, deployment, and network management. We'd invented some new technology and had some new concepts that we thought could really be of value to the industry. So it's kind of a push technology, whereas our earlier company, TSR Technologies, was kind of a "me-too" product with a little bit easier-to-use interface.
Question: Where did you get the name Wireless Valley?
Answer: Well, Blacksburg sits in a beautiful part of the world, sort of a valley between the Appalachians and the Blue Ridge Mountains, and it's absolutely gorgeous here. When we started building the wireless program, I started calling the area Wireless Valley. Students really liked the nickname — in fact, we had sweatshirts we printed up in the early nineties that say, "Building the Wireless Valley." This reflected my hopes for the region, that academic core technologies from MPRG would lead to the start up of a number of wireless companies. So when we started Wireless Valley, the company, we trademarked the name.
Question: I see Wireless Valley offers a lot of hardware in addition to the software products that work with each other for doing measurement and channel sounding.
Answer: Exactly. What we've brought to the industry, and have protected by patents, is the novel idea of multi-faceted software that can integrate with many different hardware products for many applications. Wireless Valley's core competence is development of powerful and easy to use software along with great knowledge of the issues that people have to design and manage wireless networks. If we wanted to, we could do consulting morning, noon, and night, but we've chosen not to — instead, we've chosen to serve the industry by making products that can allow other people to be the experts. Our products, such as SitePlanner®, InFielder PDA®, LanFielder® and SiteSpy®, build in a ton of knowledge into expert programs that are easy to use. By the way, we've just issued new releases of these products this month. We have over 150 customers worldwide, and universitites are starting to use these products to build networks and teach new wireless courses.
We don't manufacture hardware ourselves, but we work with a lot of different hardware providers, we partner with them, such as Anritsu or ZK Celltest, Praxsym, Dynamic Telecommunications, Inc. (DTI), or Berkeley Varitronics, so their hardware will plug and play instantly with our software products. We also have a good agreement with Ciscoand the other Wireless LAN card vendors, and our products work instantly with every 802.11b WLAN modem, bluetooth, and even wired IP-based nodes, too. The idea here is you don't have to buy new hardware — our SitePlanner® program, for example, will help you design and measure, as well as manage and archive all of the information related to your network. Just plug it in to what you have already.
That's what's always been missing in the wireless industry, how to maintain and manage knowledge of many different people. And that's the key behind SitePlanner®. The way we learned to do this was from trying to maintain research knowledge gained by students who learn a core knowledge and then graduate and leave us. We'd have a great project going on a new frequency band, or an NII study at 5.8 GHz for residential wireless, for example — the students would make their measurements, generate their models, do their designs, learn all this great stuff and then graduate. Well, a new student coming in doesn't want to pick up the custom code or the custom knowledge that's been spread around in different spreadsheets, theses, papers, and a whole bunch of notebooks. The past knowledge base is lost, as it the measured datAnswer: To rectify this loss, we designed a system that will help us share, archive, manage, and grow knowledge and information.
And then we realized, that's exactly the same problem in the wireless industry! People change jobs, people move, people forget what they've done — and this is the same as the problem I was having as a Professor! But no one's ever solved this problem in the wireless networking industry — the cradle-to-grave information management problem. When you do a measurement, a bid, a deployment, a cost analysis, something breaks — you need all this information if you are a carrier or a manufacturer. And that is what SitePlanner® does — it handles all that. It does it in an easy-to-use, well-organized way so that we have the equivalent of a visual spreadsheet, where you can visualize the entire network in a campus or building or group of buildings. You can see it in 3D, see where the components were placed, right in the blueprint, see what it cost, see where it's been repaired, instantly see the measurements you made when you were on-site with that system, and it can update and archive all that information. So, you can pass this information from the consultant or installer to the person who owns the network, to the person who has to troubleshoot the network, and it doesn't matter if one of those people leave or forget what they've done. That's revolutionary! It's never existed! We have put this knowledge on palm pilots, on Ipaq computers, and on PCs, so that an enterprise or worldwide staff can now use the internet to manage this type of information.
Interestingly, we're getting interest in our products not only from wireless companies, but also from wired companies wanting to know where the wire was put in and where it was placed. And this all came from trying to grow a research program, and trying to keep and maintain the knowledge to try to grow the program.
Question: What do you see as the trends in wireless today?
Answer: We're in the convergence of wireless LAN, in-building wireless, broadband indoor, and the cellular systems trying to get indoor, and I think that's going to be the huge battleground over the next few years, who controls the wireless access in buildings.
Question: Do you think the 802.11 type stuff will be dominant there?
Answer: I think it's going to have a definite role. Whether it dominates or not remains to be seen, but I think it's definitely a contender. What I'm hearing in the industry is a convergence between 3G mobile and 802.11 wireless type products.
Question: I've heard that 3G is sort of getting pushed to the back burner because of the explosion in 802.11. Have you seen that?
Answer: I've heard that, but I think 3G is getting pushed back even more because of the 2.5G alternatives that have been coming out. The installed base for cellular is so big now, over 630 million phones, and they're a captive base of monthly subscribers, so anything that gives a data offering in cellular is in immediate demand and immediately keeps customers on the network. So I think the carriers are pushing 3G off a little because they can keep customers paying the monthly bill by giving them things like GPRS or a high data rate on Qualcomm's CDMA. So this forces the WLAN makers to figure out how to unify, and that's starting to happen. The problem with 802.11b is there's no worldwide structure trying to unify and standardize them all — it's starting to happen with BRAN in Europe and 802.11g and 15, but it's in the early stages. I think if the Wireless LAN folks get their act together they could dominate indoor. After all, indoor environments are the only ones where a user can actually digest huge datarates. You don't need them driving a car.
Question: What kind of advice would you give to people setting up large Wireless LAN systems? What would you tell people they need to do first?
Answer: Good question. First bit of advice: use a computer aided design software program to plan out coverage and to handle cost analysis and frequency planning for your network. This is something the cellular people learned they had to do, but the wireless LAN people haven't figured this out yet. Now granted, cellular base stations are more expensive than wireless LAN access points, so people will resist this argument, but there is tremendous value in doing some quick engineering analysis using a computer aided design tool before you set forth into the campus or buildings. It's like using a spreadsheet to estimate a large project — you can't really eyeball a complex system. What is happening today is that people will go out with rough rules of thumb, without really considering the buildings they're going into, and they will sort of randomly place access points where they think they belong, based on their intuition or some simple number. And that's OK if you have just a few access points, just like it's OK to use a hand calculator if you have just a few calculations, but the IT community really needs to move to a computer aided design environment like SitePlanner® because the networks are going to get a lot larger, and there will be more users, and more interference, and more management will eventually be needed — all the access points radiate energy, cause interference, and rely on power, and they will all break eventually. You need to know where they are, what they cost, how they work, and who put them in.
The value in doing strawman designs in a computer aided design platform is that your blueprint can give you so much information, and can get you to a better design before you just start plunking access points into the network. The cost savings for larger networks will be tremendously great.
My second bit of advice, and this is even more valuable than the first: use a computer aided design environment to archive your design, component placement, and network performance data. The advantage of this is that you build knowledge and informed intuition very quickly with the software bringing you up on the learning curve on network design, while at the same time the software can be used to archive where you've put the hardware. Knowing where you put the antennas above what raised ceiling is invaluable -- as indoor wireless networks proliferate, this is going to become the biggest problem of all. Imagine access points that are above raised ceilings that may start emitting spurious radiation. If you've got to go find that with a power meter, you're going to waste so much time walking around in a building. We can put that information on a palm pilot so you can go right to the spot. Also, suppose your user community builds, and you want to get more people on the network. Interference managing is going to become such a problem in high density WLANS that you're going to have to have a computer aided design environment to know where you put the components and to manage spectrum! But most indoor wireless networking companies don't understand that yet because there just haven't been enough large installations. Microsoft, by the way, has one of the largest WLAN installations to date, and they are getting SitePlanner® to manage their network.
Question: Do you think the move to 5.8 GHz and 802.11a will push even harder on that?
Answer: Yes. With 802.11a, you're going to decrease the range of access points because of the wider bandwidth. So you're going to have more access points per building, and initially the 802.11a equipment is going to be more expensive, so a product like SitePlanner® can instantly save you money on the deployment. As people start deploying 802.11a, I think you'll see the popularity of wireless LANs increase tremendously, so there will be more access points, more mobile users, more interference, and more issues to manage. So going out with a best guess and just throwing 'em up in a building isn't going to work anymore.
Question: How about UWB? Is any work going on at either MPRG or Wireless Valley on this subject?
Answer: We're in close contact with the UWB community, we're excited about it, and think it will be viable. The thing that holds it back commercially is the FCC and FAA regulatory issues. We think eventually UWB will be allowed.
We're not directly doing UWB research, but since our work at the University is fundamental — i.e,. when we're doing channel modeling it's generally of a fundamental nature with as much bandwidth as possible — a lot of our research is directly applicable to UWB. For example, at MPRG we have a 1 GHz bandwidth channel sounder that does full I and Q modulation and demodulation. This is 1 GHz on an RF carrier that we can vary from 2 GHz up to 60 GHz, so when we make channel measurements and channel models on path loss or for time delay or angle of arrival, all of that information is useful to UWB because we're able to resolve things to a 1 nanosecond resolution. So therefore, our work directly benefits UWB proponents.
At Wireless Valley, since SitePlanner® uses fundamental propagation models that are frequency independent, the good news is that it works for UWB already and we have some customers who are doing UWB and who are designing their systems and their coverage using SitePlanner®.
Question: Have you made any comments to the FCC in response to the Notices of Proposed Rulemaking or studies on UWB?
Answer: Not directly, no. But several companies have asked me to provide information on propagation and RF bands as part of their commenting process, and I have done that. I don't know if those companies have used my name in their comments or not.
Question: From looking at your website and seeing all the things you're involved in at MPRG and Wireless Valley, it's obvious that you have boundless energy. Are you involved in other things as well that we haven't talked about?
Answer: Well, my wife and I have 3 children, the oldest is in college and I have two girls at home that keep me on my feet. But I'm also involved in a barbershop quartet. It's called the Uncalled Four (as opposed to Uncalled For), and I sing lead tenor and sure love that. Also, I really enjoy the students, and try to help them out and find them opportunities to learn both in and outside of the classroom. And I really like interacting with engineers in the "real world," through Wireless Valley or through email conversations. That keeps me pretty busy!
Question: Well, you're certainly in the forefront of what's happening in the wireless field!
Answer: It's a lot of fun, it's a great career, and I've been tremendously blessed!
Dr. Rappaport, thank you very much for talking with us!