T3 Webinar:

A Primer on Wireless Data Transport Systems: Wireless Trends, Tools and Tips

Question and Answer Transcript

August 14, 2007

Paul Olson: Q. Bob Tanner asks: Last week I just learned that the FCC auctions frequencies. Do transit agencies participate in these auctions? What frequencies are usually reserved for transit agencies? How does a consultant include this within the estimate? Where does one look for purchase prices?

A. Patrick Clair: I'll see if I can handle … all questions about the ability of transit agencies to participate in the spectrum auctions by the FCC. I have never heard of public sector organizations competing in an auction. I do not believe there is a mechanism for them to do that. I believe that the intent of the auction is to generate cash for the public treasury and that it is purely for private enterprise. Those frequencies that were being auctioned were associated with, I think, WiMAX if I remember the set of articles accurately. There was discussion over net neutrality and discussion by Google and AT&T about how to accomplish that. I do not believe it would be possible for public transit agencies to participate in the auction.

Q: Paul Olson: Q. Jim Bunch asks: Can you describe the layers in a DSRC network, physical, etc., how priorities are set for different messages?

A. Patrick Clair: It's ongoing right now. With regard to the description, the 802.11p working group is churning through that standard that describes at a low level how the standard is put together. All of that is, I would say, in process. We can capture information on that in a later discussion for you, if that's useful. That sounds like a good one.

Paul Olson: From Bob Rausch - I think it's more of a statement than a question - Note that timed relationships also work for remotes to central point rather than bold response. I think what he was getting to is the old Aloha network or slotted Aloha.

Patrick Clair: : That's some history right there.

Q: Paul Olson: Q. Benton Bonney asks: Is there a difference between backhaul and backbone?

A: Patrick Clair: Typically not. In so many of the things that we do, technically we kind of have poor definitions or the words mean different things to different people. I would say that backhaul and backbone may mean the same things in certain applications. It's just a question of when you hear it, just to ensure that you understand architecturally what the person is referring to.

Q: Paul Olson: Q. Mike Carroll asked: Some basic questions need answers by staff now for understanding of concepts and presentation.

A: Paul Olson: I guess that's to us, the presenters, that we need to have a little more work to do.

Patrick Clair: This was always a concern of ours in putting together the presentation that we had been too technical with regard to the information. So I think that comment reflects that notion, that we were not high level enough.

Q: Paul Olson: Q. R. Link asks: Please discuss the differences between the products and products designed for traffic and ITS applications.

A: Patrick Clair: Our biggest challenge in traffic, from the work that I've been doing with people, concerns environmental. This is a question of how robust, how hardened these devices that come out of the consumer products domain, how hardened are they for our applications? So we have a radio. We want to put up a pole. You cannot get it down without stopping traffic. We want the mean time between failure and the mean time to repair to be extreme. We want it to fail every 50 years. When it does fail, we want to be able to swap it out in 15 minutes. The second issue has to do with the environmental effects. We would like to see products that are specified for temperatures up to 60 or 80 degrees C. In terms of temperature resistance and certainly the condensation appropriate over the range of how it treats water in the enclosure and the degree with which the enclosure is watertight. There are standards published by the National Electronic Manufacturers Association, NEMA, that govern this. The standard is NEMA 4. There are also standards that individually describe the watertightness of an enclosure and how it deals with condensation and what not. When you read the manufacturers' data sheets, you look at are they using a standard. That's a good thing. Then look at how it handles the environmental aspects with regard to the job that you're asking it to do. You cannot take a consumer product that's made for the indoors and stick it up a pole or stick it in that traffic cabinet out in the bright sun. That will not work.

Q: Paul Olson: Q. Mike Pietrzyk asks: What atmospheric weather conditions, if any, influence the capability of wireless communications?

A: Patrick Clair: Well, great question. I'm thinking that how does rain affect the signal? Based on studies that have been done, it's not rain that we have to worry about in the frequency range that we're dealing with here. Typically, below 10 GHz, rain doesn't have too bad of an effect. You'll note that if you have Direct TV or Dish Network that your TV goes out every time it rains hard. That's because they're working at a higher frequency than the ones that we're usually talking about here in this unlicensed spectrum. What typically concerns us more is that when we have a lot of rain, we typically have a lot of wind. That means that the alignment and the stability of the mount are extremely important issues, because these devices are low power.

Q: Paul Olson: Q. Bill Lueck asks: How about having acceptable service today but in the future when there are so many users of spread spectrum and other modes in unlicensed spectrum, interference being so high that no one gets acceptable service?

A. Patrick Clair: You would evaluate the spectrum before you would buy the product. The FCC has provided a grant at whatever frequency and so you've decided to use a 5 or a 4.9, if you can. So the correct sequence of things is to evaluate the quality of the air, the spectrum in the place that you want to use it and then say, "It's clean. I'll get the equipment and we'll deploy it." To procure the equipment, mount it and then determine that the air is dirty is probably not a methodology that's going to create a lot of success.

Paul Olson: I might add, too, that the turnover in the technology is so quick that you may not get to that point. In other words, the technology you had chosen may be obsolete by the time we get to that point. That's another part of the crazy equation of having to deal with it.

Patrick Clair: Right.

Q. Paul Olson: Q. Do you consider unlicensed, i.e., unprotected frequencies adequate for public safety use?

A. Patrick Clair: Used carefully and judiciously, I think unlicensed can be part of public safety use. But you have to be careful. We have to have a certain degree of risk awareness here. If you were talking about something you absolutely couldn't afford to lose, I would say no. If you were talking about something that was supplemental to the mission that provided some other information, maybe a remote camera or something to help decision making, I would say yes.

Paul Olson: But ultimately, if you're really worried about it, you'd jump into the 4.9 GHz band, which is dedicated to public safety.

Q: Paul Olson: Q. Benton Bonney asks: The ISM and the 5.275 bands are shared. Is ISM or N-UII primary?

A. Patrick Clair: I'm not sure I understand the question. Typically, the U-NII is coded differently than ISM. So this becomes a question of how clean is the spectrum. The rules apply with regard to who got there first and who is using what. If the devices are deployed appropriately, then noise ought to be at a minimum. However, every now and again, you'll see somebody who's using an antenna that broadcasts widely when they don't really need to. When they broadcast widely from the antenna, the signal just goes everywhere, of course. So the radiation pattern is like a donut around the pole. This is essentially who gets there first, then just trying to work it out to make the system usable. Unlicensed means not much oversight from the FCC.

Q: Paul Olson: Q. Bill Leuck asks: What is the radio frequency for the 802.N standard?

Patrick Clair: A. 802.N is what we call MIMO. That sounds like such a simple question, but it means multiple in and multiple out, so there's not just one. It will choose among several.

Q: Paul Olson: Q. Robert Tanner Asks: I have read where the 802.11g plus is most for marketing hype and does not assist in increased range for users of PC MCIA cards. What current 802.11 version provides the best performance for cost and the most readily available for procurement and integration?

Patrick Clair: PA. 802.11g.

Q: Paul Olson: Q. Colleen Hill asks: Is the process for being granted a 10 year license for the 4.9 frequency as complicated and time consuming as obtaining a microwave frequency license?

A. Patrick Clair: No. The process for obtaining a license for 4.9 is pretty straightforward and simple. You can go to the universal licensing system at the FCC web page. You can do all the work on the web page. You will need to go there and let's talk about the mobile license first. When you have a mobile license, you will need to understand that latitude and the longitude of the center of your area of operations and what the operating radius around the center is. Or you can choose to describe the area of your operations by a rectangle and provide the FCC with the lat/long, the coordinates of the corners of the rectangle. If you are a public safety organization, a DOT, with that information, you would not have to involve a frequency coordinator and you can have the base license in about a week. That has been my experience.

Q: Paul Olson: Q. Bruce Tharp asks: What about some discussion of applications in the 900 MHz band, which although typically lower bandwidth, a significantly greater range, typically without the use of high gain antennas.

A. Patrick Clair: Absolutely correct. We were actually hoping to do a survey of the electromagnetic physics properties of the different frequencies at a later session. That would be a survey of why 900 MHz is better than 2.4 and 2.4 is better than 5. We'll see if we can bring that to you. It's certainly true. The lower the frequency, the better, typically.

Q: Paul Olson: Q. Robert Tanner asks: I have met cell phone tower technicians or subcontractors that have informed me they're having to move their Nextel frequency because of an FCC pending.

A. Patrick Clair: Right.

Q: Paul Olson: Q. What are the mandated frequency migrations and are there any secured for Homeland Security?

A. Patrick Clair: I believe what they are referring to was the ruling by the FCC with regard to activity around 700 and 800 MHz. I'm going by memory and I'm not exactly sure what the guys on the towers were referring to. I am sure that the information of the rulemaking is public domain and is probably available on either the FCC web page or the Cellular Telecommunications Industry Association's web page.

Q: Paul Olson: Q. Bob Brakman asked: I am working on a vertical lift bridge requiring a camera on each side. Would point to point be a good way to bring the data from one side of the bridge to the other?

A. Patrick Clair: Yes, it would. You probably have power that you get from one side to the other, so you probably have a conduit somehow. It might go under the water. But if you could use metallic or some kind of media-based path, that would be preferable to wireless. But if you're down to wireless, you could probably make it work.

Q: Paul Olson: Q. Roger Madden states: Nextel volunteered to move their frequencies that involve interference between Nextel frequencies and public safety frequencies. I think this is the 800 MHz.?

A. Patrick Clair: Yes, it is.

Paul Olson: The FCC agreed to permit the movement to other frequencies in the 800 MHz band consolidating Nextel at the high end of the band and public safety at the low end of the band. That's part of the answer to the previous question.

Q: Paul Olson: Q. Sudhir Murthy asks: Do IEEE standards also include requirements for extended temperature operations typical for any ITS environment?

A. Patrick Clair: No, they do not.

Q: Paul Olson: Q. Robert Tanner asks: I am preparing to perform a radio propagation signal power analysis for an agency. My current research has revealed that there appears to be more concern but interference with wireless data traffic frequencies than with voice, Motorola over data frequencies. What takes precedence, wireless over voice? How should this overlapping integration be approached?

A. Patrick Clair: I am not sure I understand the question. From the device perspective, I'm not sure if it's a systems administration issue or if it was a question of a separate frequency being allocated for a voice channel, as opposed to a data channel. If it's all on the same device, it seems to me it would be a question of internal operation. Without knowing the device, I just can't say. I don't know.

Q: Paul Olson: Q. Wayne Kurfees asked: You mentioned half duplex for 802.11. Do any of the 802.11 or 802.16 standards provide full duplex?

A. Patrick Clair: Yes, there are some that do. For these devices that are at 802.11 and 802.16 that are full duplex, you just have to search for them and read the data sheets about how they get there. In the wireless domain, there are more. Typically, what you will find is that they use separate frequencies for the paths. They talk downstream on one channel and they talk upstream on a separate channel, so that those communications can happen at the same time.

Q: Paul Olson: Q. Diane Silcox asks: Is there a standard to measure the radio throughput? Recently I have done some throughput tests with a couple of vendor's radios. One claims that the bandwidth is up to 208 Mbps, but I have gotten only 7.5 Mbps throughput in one direction test only with one mile apart and a very strong signal. Another one claims 14 Mbps usable throughput. I got 5.5 in one direction only.

A. Patrick Clair: Great question. What you'll find is typically that the numbers that are on the manufacturers' data sheets are correct and what you'll need to know about that is that they describe aggregate throughput. So the sum of all traffic that's possible, upstream and down, captured in a single number. The user throughput that the questioner would be referring to would be what she can get. That's separate from the processing power and time spent framing the data, overhead encryption, all of those other things that might or might not be going on within the device. So we start with the aggregate data and we start subtracting things until we get down to the possible user data. The number that the vendor provides you is going to be this clear air best possible number. Their numbers are typically right. It's important to know what they're describing.

Q: Paul Olson: Q. Roger Madden adds a comment here. There are no primary users in any unlicensed band. There is no priority accorded first user versus last user. You, as a user must find and resolve any interference you find unacceptable. Good point, Roger.

A. Patrick Clair: Yes, that's exactly right.

Q: Paul Olson: Q. Robert Brydia asks: What is the relationship of these 802.X standards to 802.2? My answer would be that the Ethernet frame remains the same in both of those and that's about where it ends.

A. Patrick Clair: Yes.

Q: Paul Olson: Bob Rausch makes a statement. You need to note that an analysis of your communications requirements need to be considered along with the overall network loading when deciding the actual network construct and equipment to be used. Remember that video can swamp out any other data requirements for ITS devices.

A. Patrick Clair: Good point, Bob.

A. Patrick Clair: Yes. What we find is that video is the hungry thing in the house. It will require the most expansive usage of the data rate for the devices. So whether we were talking that 2 or 4 Mbps for that 30 frame per second full resolution picture at MPEG 2 or MPEG 4, that dwarfs anything that's required for the low data rate 1,200 baud controller traffic.

Q: Paul Olson: Q. R. Link asks: There's a lot of discussion about the WiFi bands at 2.4 GHz. However, traffic and ITS applications should never use a 2.4 GHz band, unless there is no other choice, due to the amount of interference at that frequency. You might also note that 2.4 GHz is the operating frequency for microwave.

A. Patrick Clair: Yes, that's very good defensive engineering. 2.4 is very crowded. Because it's crowded, it tends to be noisy. It is the last resort.

Q: Paul Olson: Gabriel Gubash makes a statement that the Zigbee standard may not work in colder climates where having a constant power supply is necessary to keep it operable, and from Minnesota DOT experience. That might not be the best one to use.

Patrick Clair: Okay. I appreciate that note. I'm sitting in San Antonio.

Q: Paul Olson: Q. Ramvir Mashiana asks. What standards will apply to 900 MHz wireless?

A. Patrick Clair: We need to look at this in a layered approach. We can choose fiber or we can choose a metallic transport path, or we can choose the air. When we make the choice for the air, we're going to choose the frequency next. All of those choices can be independent of the standard that we're using to transmit the data. As a rule of thumb, you get the lowest possible frequency that you can that's the cleanest that you can. Then choose the standard to manage the transport of the data on top of that. You can view the choice of the frequency much like you would choose a wire. Different wires have different quality and capacity and different frequencies have the same.

Q: Paul Olson: Alex Power asks: What issue arrived from marrying high gain antenna to unlicensed wireless?

A. Patrick Clair: In some cases you may not be able to do it, because you may exceed the allocated FCC requirements for the unlicensed signal. Different antennae have different gains. They can focus the power better or worse, depending on your choice of antenna. The transmitter of the device is going to have a power that is set by the FCC, typically something less than one watt. You are not free to place any size antenna on that system that you would like. There are numbers that you cannot exceed.