Airing Problems in Wireless

By: Ed Sperling - Electronic News
(Business News & Technology News, 15 Oct 2007)

Electronic News/Electronic Business sat down to talk about future wireless issues with Darren McCarthy, market development manager at Tektronix; Christopher Bergey, director of Broadcom's embedded WLAN wireless connectivity group; Zoltan Cendes, chairman and CEO of Ansoft; and Dan Karr, senior vice president of marketing and sales at Tzero Technologies. What follows are excerpts of that conversation:

Q: What are the big problems we're facing in wireless?
Bergey: If we look at the largest market, which is the cellular handset, the biggest issue is a huge desire to get smaller and smaller with these products while at the same time increasing the functionality. It¡¯s not just CPU MIPS (central processing unit millions of instructions per second) and more memory. It's wireless support, whether that's adding quad-band radios, Bluetooth, wireless LANs, NFC (near-field communication) and GPS. Everything comes with an antenna. This miniaturization and new features are the two vectors in the handset market.
McCarthy: Ease of use and convenience are really what it's about. Having wireless ear-buds, high-speed downloads when you take pictures on your camera—those devices are becoming more integrated. But the other problem is the mobility factor. Where you can take these devices—on airplanes and in hospitals and onto Air Force base—has become a problem. Between the integration and mobility, and the finite spectrum where you're trying to accomplish all this, that is really causing a pending crisis of spectrum.
Karr: We're focused on the delivery of HD video content. With wireless, there are a couple of problems. One is bandwidth. The second is quality of service. Now you have a new type of data that's being delivered. It requires a quality of service that ensures delivery and delivery on time. We use ultra-wideband to address that problem.
Cendes: We see a lot of these issues. We also see a lot of low-cost devices where they mix the digital together with analog and try to get better performance. There are a lot of problems with mixed signal design. With a cordless phone and a wireless LAN, we're trying to make the traffic aware of what else is on the network so it can be dealt with in the frequency domain or the time domain.

Q: Most people assume they don't get the same quality of service out of a wireless connection as a wired connection. Will that ever change?
Karr: It's different for voice, data and video. From the video perspective, one of the things you worry about is fading. As that channel fades, you don't the ability to carry data and you lose the signal. One mechanism for addressing that is a wideband technology. The other element of quality of wireless is the isochronous nature of the data—a frame of data every 16 seconds that must be there. That gets built in at the quality-of-service level. UP&P (universal plug and play) 3.0 is going to provide that level of service. WiMedia Alliance has an ultra-wideband spec. But it's new. It's an evolution of technology. Now we're delivering HD video. People don't want jitter as it plays. There are standards that are starting to address it. At the same time, it's problematic for existing technologies. That would require backward compatibility.
Bergey: People are willing to accept a slightly lower quality for wireless. But as these products evolve, the quality of service goes up. There are things we do in the digital domain, but there also are things being done in the wireless domain—802.11n for wireless. MIMO (multiple input, multiple output) systems are essentially an RF mechanism to provide more coverage and higher bandwidth. We're all trying to get more bandwidth and deal with the collisions in the analog and digital domains. The Nintendo Wii uses Bluetooth and wireless LAN technology from Broadcom, both at 2.4GHz, but they work together just fine.
McCarthy: As we go forward, new standards and new technologies are cognizant of each other. They can be used for mobility and brought into areas where other wireless technologies are going to try to use the same frequencies. Ultra-wideband technologies can sense their environment and avoid other transmissions. Cognition is starting to become part of the standards.
Bergey: With 802.11n, people were disappointed with how long it took to standardize, but one of the things we worked with in the last stage was ¡®good neighbor¡¯ technology. We certainly wanted to make sure it provided great wireless technology in a home where there was nothing else, but we also recognized that¡¯s not a real-world situation. It had to detect what else was out there and properly allow for an ecosystem of products. That¡¯s what the home is becoming.

Q: Is this problem the same in other countries, where cell phone connectivity and infrastructure is better?
Karr: There's licensed spectrum, which tends to be different in each country, and then there¡¯s unlicensed spectrum. A lot of the technologies we¡¯re talking about here are in the unlicensed spectrum. Anything goes. And those are common worldwide¡ªBluetooth, wireless LAN, Zigbee, RIFD¡ªthese are all broadcast in ISM (industrial, scientific and medical) band. Anyone can play here, as long it¡¯s at a low enough level. However, the mobility doesn¡¯t take into account how close in proximity these things can be used. Then there¡¯s ultra-wideband, which is trying to broadcast at a low enough level where it can make use of unused spectrum. It does vary from nation to nation, but that¡¯s only in the licensed spectrum.
Bergey: When you look at different technologies and ask, 'Is that technology going to be successful?', one of things we consider is whether it's in the licensed or unlicensed bands. One of the problems with ultra-wideband today is that it's very difficult to ship products around the world. There are only certain bands where it's not homogenous. That's going to change, but it's something to consider in market adoption.
Karr: Today, all of the chipsets on the market are 3GHz to 5GHz, which is band group one. Some are licensed, some are unlicensed. Probably the most common thing you'll see in that range soon will be WiMax, and over time, 4G. Depending on the level of investment various countries expect, they¡¯re using that to determine how they¡¯re going to detect and avoid. Above 6 GHz, there are more band groups, and that¡¯s wide open to anyone at this point.
Bergey: You also have to look at the temporal nature of bandwidth allocation. Each country has services. FAA landing radar gets deployed and sits on the ground for 30 or 40 years before it gets refreshed. Then there is the commercial aspect. The FCC likes to sell spectrum because it's a natural resource. They will want to sell more and more to get billions and billions of dollars of free money. In the wireless technologies, you might have a technology licensed for 10 to 15 years. Any technology, whether it's GSM, GPRS, Edge, wideband CDMA—any one of these might only last two or three years in its full maturity. So there's a rapidly changing consumer area, and then there are services and infrastructure, which do not change often. That's where some of the issues are starting to crop up.
McCarthy: In 1975, when the FAA was putting in its latest and greatest landing radar, it didn't conceive that people would be walking around with wireless transmitters a couple miles away. The problem is that radar can't tell when it¡¯s being jammed.

Q: There are a lot of end-run types of technologies hitting the market, including WiMax. Who manages all of this in terms of the interactions?
McCarthy: In the U.S. there is the FCC (Federal Communications Commission) and the NTIA (National Telecommunications and Information Administration). The NTIA owns the government spectrum and the FCC owns the commercial spectrum. They talk, but really it's an ongoing battle. The NTIA is in a defensive position because they have to defend their allocations. The FCC is basically saying, 'Prove it won't be a problem.' There are problems like Katrina, where the NTIA has mandated a common service that everyone can talk to in case of a disaster. At the same time, there is a commercial spectrum that Nextel had. The application of that can interfere with first-response services. You have commercial and federal groups fighting over the same spectrum, and at times having to be managed.
Cendes: My colleagues at Carnegie Mellon took some equipment on a commercial flight to see what happens. They discovered that people make phone calls—an average of one or two a flight—and they do interfere with the GPS system.
Bergey: From a product point of view, I don't think it's really an end run. But there needs to be a compelling reason for wireless video. If I hang a flat panel on my wall I have to get video to it. No one wants to see wires running up to it. We're trying to solve a problem in the unlicensed band. We don't just solve the problem when you buy a product from Company A. Consumer electronics companies care that you're happy with their product and it works in your environment and you don't call them. If they get a service call, that wipes out their profits.

Q: Consumers want the latest and greatest, which means upgrades on a regular basis. That also requires an upgrade in service and increases the likelihood of data collisions, right?
McCarthy: The iPhone when it first came out had an issue with one of the settings in the mode it was in. It was hogging bandwidth and shutting down networks. The user experience is really going to differentiate which technologies get adopted. If in the use of wireless technology people aren't being good citizens it's going to get public attention.

Q: With consumers now in the driver's seat, they're forcing changes faster than when the carriers were dictating what products hit the market. Can the infrastructure keep up with the user?
Bergey: What the consumer is buying is a point-to-point connection within their home. That's different than the cellular networks and what they're offering. If you look at Wi-Fi, globally that started out with a reasonably conservative approach. More recently, we have more spectrum and higher bandwidth. It was a marginal product initially, but the FCC has become more accepting and allowed it to be more widely used. I think we'll see the same in ultra-wideband and unlicensed technology.

Q: Is it really just the home that the consumer is looking at?
Karr: It's voice, video and data. You can have it over your cable system, over your phone, or on your laptop. People want untethered access to information wherever they go. It won't be one technology that will be a panacea for everything. You have a mixed set of technologies based on technology available and spectrum that's available, even in the home with cable. The question is what technology can be added that doesn't impact existing services such as television or triple plays for the cable operators.

Q: So who controls that? What limits who gets into this market?
Bergey: The carriers really are the gatekeeper. If you have a technology that is going to attach itself to the network, the carriers typically have a six-month cycle where they take the end product and ensure it's providing a certain quality of service. It's not that much different than an IT director in a company getting a product and trying it. The carriers are really selling to the consumer who wants the latest and greatest. The carriers also are trying to make their services more ubiquitous. There are issues with people in their homes, and companies like T-Mobile are trying to address that with their UMA strategy. There is the potential for femtocells, so that in your home you actually have a small cellular base station. Generally the service is becoming more ubiquitous.

Q: Is that the same around the globe?
Cendes: Yes. Japan is famous for wanting the latest and greatest. It's the same with Korea.

Q: Does that mean they'll see these collisions first?
Karr: Yes. The new adopters always see problems first.

Q: So what are Korea and Japan seeing?
Bergey: In the handset market, the carriers do a great job of qualifying the product. Generally you see delays in product introduction. By that point, all problems have been solved. The ones most impacted are the product solution provider because of the delays in getting the product out. This is very unique to handsets. If you want to sell a UWB gateway, the Best Buy buyer has not had a huge qualification process. That's where the consumer may be dissatisfied with the product.