Mobile WiMax explained for you: In this podcast, I interview Monica Paolini of Senza Fili Consulting. Paolini is an expert on the intersection of financial projections, technology, and market needs, and works with the WiMax Forum among other groups. Because of her wide-ranging interests, we focused on mobile WiMax in this podcast, as I think it's one of the coming technologies that's worst understood and most misrepresented because of the obscurity of parts of its operation.
We talk about the differences between fixed and mobile WiMax, along with 802.16-2004 and -2005, the underlying standards that are commonly associated with the two terms. We walk through the spectrum bands that might be used in the US and internationally for both fixed-only and fixed/portable/mobile services. And Clearwire's recent massive receipt of Intel and Motorola money is examined as it affects the future of mobile WiMax in the U.S. [39 min., 20 MB, MP3]
Glenn Fleishman: This is Glenn Fleishman, the editor of Wi-Fi networking news, and welcome to podcast #11 recorded July 25, 2006. In today's podcast, I am talking to Monica Paolini of Senza Fili Consulting. Monica is a wireless expert. She consults widely on WiMax and related wireless data services, unlicensed mobile access, all these voice related issues... One of the great secrets of your careers is that you studied with Umberto Eco also...
Monica Paolini: [laughs]
Monica: now, it's public...
Glenn: Well, it's all about science and semiotics. You know, the signaling and symbols that are involved in wireless data are transformative too.
Monica: It's all communications when it comes down to it, so... Humans are animals that like to communicate, that's my worldview.
Glenn: It comes through very clearly. You are an expert on a number of these topics. You consulted for and have written white papers for a number of associations that are dealing with wireless data issues, like the WiMax Forum, and I found out almost by accident that you were involved in request for information that's been put together, it's been answered now, for the Capital Corridor Joint Powers Authority in California for their train-based Wi-Fi test, Internet access test. So, we can talk about many things.
And also, another secret, you and I had an eight-hour conversation once, so we can have a podcast that will go on till the end of time. That was the ultimate conversation, a Boston-to-Seattle flight, so for this podcast, I think we decided we'll carve out an area of mobile WiMax because I think it's one of great confusion for a lot of people. We read a lot of problems in the mainstream press in terms of how they describe fixed-to-mobile WiMax. A lot of people even in the industry are who working with the technology seem to describe it wrong. So, let's start with the fundamentals. What's fixed WiMax and what's mobile WiMax? How do you define those two, or do you define it separately?
Monica: Yeah, well. That's sort of prevailing industry use, I personally don't like the mobile WiMax name, but I will get to that in a second. So, fixed WiMax is based on a technology that is based on the IEEE 802.16-2004 standard by the IEEE and uses OFDM as modulation and uses both TDD and FDD. It really is a great technology for fixed communication, fixed wireless communications. Actually, part of the confusions comes from the fact that initially what is now called fixed WiMax, was expected to become what is now known as mobile WiMax. So, initially there was a plan for just one technology, but as people were working on it and trying it out, they realized that fixed WiMax was really not well suited for mobility, that there were better technologies that were coming out and there was also some political issues because in Korea OFDMA was used. The WiMax Forum and the IEEE started working on a different version of the standard that is called IEEE 802.16-2005. Now, it has been approved and that is a different technology, so they are both called WiMax, but they are different because mobile WiMax uses scalable OFDMA and it can use both TDD and FDD, but right now the profiles only include TDD.
Glenn: I think we should define some terms because... So, OFDM is a orthagonal frequency division multiplexing, which is a widely used technology. It's used in 802.11 g and a. it's used in broadband over power line and it's a way of taking one swath of spectrum, dividing it into sub-channels, each of which can then carry data in different ways. It can be assigned from parties. And so then, TDD and FDD, time division duplexing and frequency division duplexing, how do they work separately in order to be used over the same set of spectrum? I know there is exclusive usage. You can use TDD or FDD.
Monica: Ok, yes. So, it's either FDD or TDD. In cellular networks, you tend to use FDD. And there you have two channels, one is for the uplink and one is for the downlink. And that's a solution works very well for data because data traffic is intrinsically symmetrical. So, it's efficient for that kind of a reason.
Glenn: So, it works well for voice? Monica: For voice, yes, because you have two channels. On the other hand, TDD has been traditionally used in BWA [broadband wireless access] technologies because it's better suited for data. You have only one channel and you have uplink and downlink traffic all in the same channel. So, basically what happens is that you have a time guard that separates the traffic--the downlink and uplink traffic--on a TDD network. On an FDD, instead, you have a spectrum band that separates the uplink from the downlink channel. So, with TDD, you have much more flexibility in structuring and managing your traffic, uplink and downlink. And also, it's less complex, so if you are a small wireless ISP, you probably want to go for TDD because it's easier to use. It's also that technologies like MIMO can be implemented more easily in the TDD network because you only have one channel.
Glenn: So, OFDMA, that's orthagonal frequency division multiple access--and so in OFDMA, you still have a swath of frequency divided into multiple sub-carriers or sub-channels, and then how each of these sub-channels can have its own characteristic, where each of these sub-channels can be assigned individual users?
Monica: Exactly. So, in OFDM, you have multiple carriers, but... For instance, if you look at it from the CPE side, there is only one CPE that transmits at any given time.
Glenn: That's Customer Premises Equipment, that's the gateway that separates...
Glenn: I know, there are so many of these TLAs, the Three Letter Acronyms in this business, we just have to be careful.
Monica: Well, the problem is that three-letter acronyms is good, the four-letters, five letters, it's...
[Glenn Fleishman laughs]
Glenn: So, the CPE, that's the piece of equipment that a customer would have in their home or it might be an automatic unit they take with them to gain good access.
Monica: Yeah. So, in that case, with OFDM, you have very short frames and for each frame there is one CPE that transmits. So, you alternate, one, two, three. With OFDMA, it's different because you basically allow different CPEs to transmit at the same time in separate parts of the spectrum. The network decides, CPE number one, two, three can transmit on different channels at the same time.
Glenn: This is coordinated by some central base station. Some central base station coordinates CPEs, so that they can transmit at the same time.
Monica: Exactly. What that means is that from the CPE end, you have a lot of limitations in terms of power. There is just so much power that you can have in a laptop. Basically, by focusing your transmission on a smaller channel in terms of spectrum, you have much better chances to transmit more data from the CPE end. This is a technology that allows you to transmit from the user to the network in a more efficient way.
Glenn: And my understanding also is that this reduces contention, so it increases spectral efficiency--because one of the big problems with Wi-Fi is that the more devices you get and the more high-power devices you get, you start to get contention where different devices have to back off because they start talking at the same time. It sounds like with OFDMA, because a base station is coordinating which sub-carriers different CPEs are using that dramatically reduces contention?
Monica: Yeah. The overhead is less. That's an issue, actually with Wi-Fi too because if you try to move away from this... With this approach you then tend to have a more WiMax type of solution that reduces the freedom in the network, but let's stay out of that.
Glenn: Well, that's often the contention, too, with Wi-Fi is that a lot of people claim that Wi-Fi is the wrong technology for mobile, for metro-scale mobile as opposed to, say, a company-wide service or a hot zone or someone's big house, because nowadays Wi-Fi has this issue of contention with more devices producing more overhead for any of them talking, but also because there is no way to assure any quality of service for any given device. There is no way to provision Wi-Fi; you'd have to change the Wi-Fi adaptor's MAC, its Media Access Control technology. So, in some ways OFDMA in some ways is descending from Wi-Fi. It's some aspect of what people wanted in Wi-Fi but couldn't get, put into a new form.
Monica: Yeah, the way I see it is, there are technology differences, but really in terms of the philosophy to two technologies... WiMax is a technology for carriers, for service providers. And I know that there are some companies that are developing products for the home, for instance, but that's not the strength of WiMax. And on the other hand, with Wi-Fi, the strength of Wi-Fi is that it's really simple, very flexible, and the network can be deployed in your home without you doing anything. The level of simplicity is such that you don't have to worry about pretty much anything. I guess that the guiding principle for these technologies is completely different.
Glenn: Let's get back to differences between what's been labeled fixed and what's been labeled mobile. What is 802.16-2004? That was called fixed WiMax and uses OFDM, so it's got the advantages of being more spectrally useful, it's diverse, it can deal with interference and it's robust, but it does not have this per-user efficiency. And then you get into what was 802.16e, now called 802.16-2005, just for simplicity's sake, right?
[Monica Paolini laughs]
Glenn: and then there's OFDMA, which offers better opportunities to schedule and provision service, so that one user might have several of these sub-carriers, because they are paying for it... Others might have one, that's possible. So, the 16e or the 2005 service, what's been called mobile WiMax, that can be used for fixed service also, right? This is where the confusion, I think, stems from.
Monica: Yeah, I think that what mobile WiMax gives you is the possibility for mobile services, too. But there is really no reason not to use it for fixed because if you have the very same network in the very same environment, and you have fixed and mobile side by side, performance is pretty much the same in the fixed environment. You might have some difference, but that's not that great. Basically, when you look at it from the general market point of view, and volumes are also important, what you end up realizing is that it's much better to use mobile WiMax for fixed as well, so we all use the same technology, have the big volumes of scale and no penalty, rather than use fixed WiMax. Now, this is a high-level view. But if you look in a little more detail, what you will see is that you will have different bands and one or the other will dominate. For instance, in the 5.8, the band, which is unlicensed in most countries, you are more likely to see fixed WiMax because in 5.8, you will not have mobility in the near future. And the type of services are necessarily fixed, so that you don't really gain any advantage.
Glenn: Now, you won't have mobility because the characteristics of propagation of 5.8 aren't good from that side.
Monica: Yeah. From a technical point of view, if money was not an issue and you could have the same density you have with Wi-Fi access points, the ones that you have at home, sure you could have mobility, but it just wouldn't make any sense from a cost perspective, it would be ridiculously expensive.
Glenn: Well, I know one of the advantages of 5.8 band is the way that the FCC rules--and in many other countries, it's the same thing--allows some very high gain. That's available for directional signals in 5.8GHz. So, if I am running fixed WiMax in 5.8, I can be putting out a huge amount of power, going enormous distances, going point-to-point, where I don't get that advantage in the mobile unidirectional world at all.
Monica: I think that it's the same thing you are seeing with some municipal Wi-Fi. You can tweak the technology to do something it was not designed for, but when you do that, you do have to deal with some other issues. So, from this point of view, with 5.8, you don't really, you don't want to go into mobility. 3.5 [GHz] is the same. It's not as bad, but you are still not going to have mobility. You will have some portability, but you will not have the type of coverage that you expect from a mobile service.
Glenn: Well, we are transitioning into the spectrum discussion, which is great and let's... the 3.5 GHz, that's a very difficult band. We should talk about where it's available. This frequency might be available in the U.S. on a quasi-licensed basis, but it's widely available in Europe where there are some limitations on how it could be used. How will that affect how WiMax is deployed there?
Monica: Yes, that's an issue in Europe. In some countries you can only use it for fixed services and it is not clear whether you can then use mobile WiMax to support fixed services because there's no reason why you shouldn't.
Glenn: So, you are saying--because this is where the technology becomes more complicated, because you've got one piece of technology--the -2005 standard, that can do all these things--what's he difference between a fixed service and a mobile service when it's the same technology. You are saying, it's not mobile, but it's still offering the same... it's still doing the same thing.
Monica: You are still portable. If you look at the mobile user with a laptop, most of the time you are stable and in one location, unless you are on a train or in a car. The reason why in 3.5 you have fixed services only... First of all, there's the spectrum, it's more suitable for fixed operation. At the same time, because it's limited to fixed services only, the price tag on the spectrum is lower. And the reason is to provide competition with DSL and cable modem. That's the reason why there is a limitation, but moving forward, this difference makes less and less sense because you have the same technology that can support both fixed and mobile and in my view this is the greatest advantage of WiMax, the fact that you can support both fixed and mobile on the same infrastructure. What that allows you to do is to have multiple revenue streams, you can bundle services in interesting new ways that you cannot do either with cellular or with DSL. That's really the key winning point and the distinction between the two is really coming down and not just in WiMax. I think that from a regulatory point of view, you need to start rethinking how you structure the licenses.
Glenn: That's interesting because it used to be dramatically different approaches for those two different kinds of technology--the way that spectrum had to be strictly allocated and divvied up. In fact, the frequency division duplexing made more sense for cellular mobile because it was the only to make sure that you had the right slots at the right time.
Let's move down the spectrum. The spectrum window here down to the left is 2.5 GHz. When you get below 3 GHz, I know it starts to become more and more in the sweet spot because of signal propagation and how it penetrates walls. 2.5GHz, we know who owns that in the U.S., that's Nextel and Clearwire. We can talk Clearwire now or in a little while, but 2.5GHz seems like an ideal technology for mobile WiMax.
Monica: Yeah. In fact, fixed WiMax will not work in the 2.5 or there is no current profile defined for it. So, that's an area where mobile WiMax clearly dominates. Well, that's the only option. It's a very good band and it's the best and that will also become available in Europe, but it's not clear whether WiMax will be able to be used in that band. In the U.S., once you have a spectrum license, you can deploy any technology you want as long as it meets some requirements. In Europe, historically, the tendency has been to mandate the use of specific technology and it's not clear what's going to happen now with a 2.5, 2.7, so WiMax proponents are usually pushing for a technology-neutral allocation.
Now, one interesting fact is that once you have different bands located next to each other, if everybody uses the same technology, you can be more spectrally efficient because you can synchronize the signal. The whole issue of technology neutrality, it's very attractive because this way, once you have the spectrum, you can keep up with whatever new technology emerges. So, it gives you more flexibility from a technology point of view, which I think is great, but on the other hand you might have more problems in terms of needing more bands, guard bands between channels.
Glenn: Oh, right.
Monica: But you know it might be a price that I think is probably worth paying for the flexibility. You know if you look at 3G, if the mobile operators were not required to deploy 3G, maybe things would have developed in a different way.
Glenn: Right because they got the licenses with the requirement in place... specific kind of service.
Monica: and a specific type of 3G, not just any 3G. UMTS, in Europe.
Glenn: Right, so that gave, right, that focused it in and it's been taking years to see how that's actually developed into a profitable market, which is the whole point of it.
Ok, so 2.5 GHz, 2.3 GHz, 1.1 GHz, those both seem to be options in the US and I know abroad as well, those are available in different countries too. Japan, South Korea with their Wi-Bro service which has substantial things in common with mobile WiMax... They're using...
Monica: It is mobile WiMax.
Glenn: It is mobile WiMax, just not under that name.
Well, hey actually let's talk about this a while too. So, you know, let's say the bands were 2.1, 2.3, 2.5, 2.7, 3.5, and 5.8. All these bands are available for different flavors of WiMax. The profile thing you talked about: this is the thing I found baffling for the longest time and I think I finally understand is that, you've got the WiMax Forum, which is going to certify devices compliant to a particular form of... particular standards they set that are based on the IEEE standards, same thing with Wi-Fi Alliance does, but where the Wi-Fi Alliance has a very simple task. They have interoperability using a specific set of standards, everything has to work together, has to meet certain framing tests and all this other stuff.
WiMax Forum's problem is that you've got all these different bands, you've got TDD and FDD as ways of handling duplexing of data going over the same frequencies. You've got, we haven't even talked about yet, you have bandwidths, literally bandwidths, so you have five MHz, 10 MHz, 20 MHz. How does this all get sorted out into profiles? Because the devices are going to be certified based on profiles, right? If I've got a mobile WiMax device, if I'm Samsung and I send it in to the forum, then I may be saying this device works with this bandwidths, with this kind of encoding, this duplexing, these bands. How does that all work and how is that going to be sensible in the marketplace?
Monica: well, it's not actually that... you know if you compare to Wi-Fi, Wi-Fi is much easier because you only have basically three options: A, B and G.
But if you compare it to mobile phones, it's actually not that different.
Monica: Now if I ask most people, you know, what kind of bands does your mobile phone support? What type of channel, you just don't know that because it's the mobile operator that sells you the phone. And I think that a lot of the confusion will be taken care by the fact that you go to Best Buy or whatever store, and you're going to buy a device that is supported by your service provider.
And also multi-mode devices will prevail because that's the only way you can have one device that serves a large market. So, from a consumer point of view, I think that it will not be too confusing, because there are other ways to guide the consumer.
But it is also true that you don't have the same type of experience that you have with Wi-Fi where, you know, you can travel worldwide, and your laptop Wi-Fi card will work everywhere. That will not happen with WiMax.
Glenn: This is the thing that I think gets confusing about certification. There is this long lead up to WiMax certification for fixed devices and now we've kind of got a timeline for when mobile WiMax devices will be certified. And there's this confusion because it seemed like a class thing initially. And those of us who didn't understand the intricacy of it, and now I believe I might, it's going to be, "This device is WiMax certified," it's like, "No," it's going to be, you know, everything not going to be interoperable because of these choices and in frequency band and duplexing. If I buy... and I noticed some of the announcements coming out from manufacturers releasing early CPEs for fixed or mobile, or getting their chips certified, they're saying, "Hey, here's the bands that we support, here's the frequency ranges." So a CPE might actually support eight profiles, six profiles, 10 profiles, whatever the manufacturer thinks makes sense given their base band and their chip and the market they're going for.
So, really as a consumer maybe I will get to a point where I can walk into a store like I buy a DSL modem today and buy a WiMax CPE, maybe with Rosedale2 chips built in or whatever it turns out to be, but what you're saying is that it's not going to be a consumer decision in that I'm going to be signing up with a particular service provider, I'm not going to be roaming with WiMax the way I'm roaming with Wi-Fi across hotspots? Or....
Monica: Well, it could work. Let's say that you buy a device in the US and there's just a given number of bands, you know, so probably all the products for the US market will support the frequencies that are available in the country.
The other issue being that you could not have, let's say, a CPE or say a PCI Card [that] supports [the] 3.5 band. It might not be legal in the US because you cannot use that spectrum. So there are also those issues. So I think that for most of the... It's really like you have on your cell phones. You just have several frequencies that you can use and then you just use the one that's available locally. And for roaming it's the same.
So, you can easily have a WiMax card that is sold to you say by Sprint, or a Clearwire card, but then you go into the Sprint network and you still are able to roam so from a technical point of view you are able to connect. Now the roaming bit is different because it implies that there is a business relationship between, say, Clearwire and Sprint, and if they don't have that in place you'd have to have two accounts. But it's possible. Let's say that you have an account with Clearwire and an account with Sprint and you might be able to use the same CPE to access both accounts.
Glenn: That's fascinating too because that's obviously one of Intel's goals. You know they just announced yesterday, finally made a formal announcement about the Rosedale2 chips which will be built into future laptops, and be used in many ways for doing mobile WiMax reception. And obviously Intel would like to sell a generic thing that could be used anywhere. But conceivably, just in the way, I mean... There's a lot of things you can say about the Centrino verification program that Intel started in April 2003 for certifying Wi-Fi hotspots. And maybe it was just a way for Wi-Fi hotspot operators to get some co-branding money and Intel's name behind it, but I also heard at the time that operators were really pleased because Intel gave them really valuable stress-testing on their systems. They were able to improve virtual private network connections; they were able to troubleshoot longstanding problems they couldn't work out on their own. So there really was some improvement in the inter-operability of Wi-Fi networks.
So could you see Intel coming along and saying that as part of our Rosedale2, whatever the shipping name is, we're going to actually help promote roaming across carriers because we want people to be able to use it more broadly. So I might subscribe to one carrier but be able to roam to another, maybe not within the same country but as I go around the world.
Monica: Yeah. They are working, they are doing a lot of work in terms of interoperability before certification because when you have the certification you have a limited time to sort out problems, so it's better if you start working on it before. And they are also very active in establishing roaming activities.
Now, Intel can push for this to happen, but really when you talk about roaming, it's really a business issue. That's the same it's been for Wi-Fi. You just need to have the carriers to work together, to agree to have a system by which they allow the users on each other's network.
With WiMax, I think that in several ways it's going to be easier. First of all because Wi-Fi taught a big lesson to everybody involved. With Wi-Fi there has been a lot of progress lately. That is all going to carry over to WiMax.
The other thing is that for WiMax you will see big operators. You will not have the same type of fragmentation you that have with Wi-Fi. Because with Wi-Fi we can just have one hotspot and you're in business, with WiMax it's not going to work that way. So you are going to have fewer service providers that need to work with each other. And if you have a mobile operator, they already are familiar with roaming.
And actually one of the reasons why some operators might want to adopt WiMax is actually for roaming. So for instance, let's look at Sprint. If they go with WiMax 2.5, they're going to have a better roaming than they currently have with CDMA that, for instance, is not used in Europe. So if I have a WiMax card here I might be able to go to Europe and use it there. Assuming that the 2.5 or some other band.
Glenn: Right. But they could have a card that does CDMA. They could have a 3G/WiMax card, for instance, with 3G built in and....
Monica: Yeah, WiMax card with multiple bands, for if you go to Europe. The 2.5 band might not be used for WiMax but....
Glenn: There's a point you raised there, that, I think, it's worth repeating, too. Which is that you were noting that--and you talked about this earlier--that WiMax is much more likely to be not an individual thing, someone's not going to set up a free WiMax hotspot the way you do a Wi-Fi hotspot, and part of it is all the different profiles and frequencies, but part of it is for mobility's sake, certainly all the frequencies in which mobility would make sense are all going to be licensed frequencies.
Monica: Absolutely. So you can have it in 5.8, but yeah.
The other thing is, for mobility...OK...let's say that you have spectrum in just one small part of the city. What does it do for you? I mean, the value of mobility is that I can use my card wherever I go. So unless you have a, I wouldn't say national coverage, but unless you have a wide enough coverage the value of mobility is greatly diminished.
Monica: So let's say in Seattle they only cover a two-square-mile area, well, it doesn't do you any good.
Glenn: For fixed that makes sense. But for mobile, I mean, fixed it's fine, this is what all the early pre-WiMax broadband replacement, you know, T1 replacement companies were doing.
Well, so let's finish up with this discussion with what the biggest news has been in recent, you know, I mean this has all been coming along, and different companies developing, releasing products, certifications happening... but then out of nowhere $900 million gets invested in Clearwire just before it had planned, ostensibly, to go public, file for an IPO and raise about $400 million. So Intel suddenly puts $600 million in, Motorola buys the NextNet division of Clearwire, which is the equipment manufacturing division, and what do we know about Clearwire--they own a lot of 2.5 GHz spectrum and they raised a lot of money as quickly as possible. How does this change the complexion of the future of mobile WiMax in the US with all this money having gone into Clearwire?
Monica: It's a major change. Almost how the US is really the most difficult market for WiMax. If WiMax has any success in the US, everything else is going to be so much easier.
The problem with the US is that the spectrums are already allocated and therefore if the few companies that own the spectrum go some other way, then obviously there is no other market for WiMax, other than 5.8, which is a small market anyway.
So I think that this changed things a lot. First of all because obviously before Clearwire was already a member of the WiMax Forum, there was some commitment to move towards WiMax but it was never really... you never see a timeline, there was nothing ever announced, so it was a little bit of uncertainty there. So that has been removed.
But I think that the interesting part is that once you have Clearwire clearly committed to WiMax this is going to be a big push for strength to adopt WiMax as well. Because if you think of it from Sprint's perspective, if they have a roaming agreement with Clearwire, and they operate in a largely different market, they basically have a good partner for roaming. So basically the value of the roam network would be higher because they have all this roaming opportunity.
If they were to go with the proprietary technology, that opportunity would just not be there. So I think that really helps. Also we need to remember that Clearwire is not just in the US. They also have service providers they work with in Mexico, in several European countries; the impact is going to go beyond the US. But in the US it's a major change for WiMax.
Glenn: Well, and there's a spectrum issue too there because Clearwire has some good holdings, they don't have great holdings at 2.5 GHz. And they say they can reach about 90 million households with the current [licenses].
They're estimated to have about 15% of the 2.5 GHz licenses, and Sprint has something like almost 80% and Bell South has a few. There's a lawsuit that, Clearwater just filed a lawsuit against the Peralta community college district and in that they reveal this very interesting thing. The allegations are that Peralta signed a contract with Clearwire; Clearwire says they did. Peralta says they didn't do the whole execution; there was a signing agreement that wasn't executed. So that's it, there'll be a contract dispute going on, but two interesting things came out. One is that Clearwater says that [this is the only] license [in] the entire San Francisco Bay area, which I think will be very important. There is this one with Peralta that would primarily serve Oakland, that they wouldn't have much reach beyond it.
And second, that the value of the license went from $1.8 million to $18 million in the last six months. Meaning that somebody else who could that be? Who might actually be interested in Peralta's license.
Monica: Yeah, we don't know that.
Glenn: We don't know, it might begin with "S", I don't know. So, clearly that seems to be another signal that things may be heating up in the band.
Monica: Absolutely. And I think there is some expectation that Sprint is going to make a decision in the next few months, so they have to look hard for something. I think that's going to be the real, a really big, whatever Sprint decides to do is going to have a huge impact on WiMax within the US and worldwide.
Again, I really think that US is really not the crucial market for WiMax in the sense that the worldwide opportunity is bigger than in the US, and it's really a technology that is best suited, in my view right now, for mobile, for developing countries, because there you just don't have the fixed infrastructure, you don't have 3G and therefore WiMax can really fit in the middle.
And that's also where mobile WiMax is particularly useful and that's really a problem with the name of mobile WiMax, because when you go and talk to people in the developing country they say "we need basic DSL, " they don't care about 3G.
Glenn: Right, we need a new marketing term for it.
Monica: I totally agree with you, but the reality is that mobile WiMax will give them the flexibility. So for instance in India they're now working on, they're incredibly active on the WiMax front but they are looking at 700, 2.3, 2.5 and 3.3 but they're only looking at mobile WiMax because they realize that you just gain in flexibility in a major way and you can even use WiMax and Wi-Fi and mobile WiMax basically for some backhaul, but that's fine, because you know...
Glenn: This is where I'm not making any friends because I've been saying that Wi-Fi is 1999 technology that's been spiffed up over the years.
Mobile WiMax is 2005 technology and if you were starting from scratch, if you went into a major city in Africa that had no 3G service, and not a great DSL infrastructure, what would you say? Should we use Wi-Fi because it's cheap or should we use mobile WiMax because it actually encompasses all these modern ideas, encompasses all these improvements that people talked about maybe building into future versions of Wi-Fi also, but just aren't there. Which would make, you know-- maybe cheap would make sense. But it seems like if you're playing for the next 20 or 30 years maybe using a newer standard that had better special ideas would make more sense.
Monica: But you can use both. The way I see this you can have WiMax spreading to the edge gradually as the price of the CPE goes down. Because right now when you have, you know, even at $100 each in a market like India where the DSL is $8 a month, a $200 CPE will just not take you anywhere. But let's say that you start building your WiMax infrastructure for business users and for some backhaul, that in itself is probably going to work out fine. And then what you can do is you can have some Wi-Fi at the edge where the CPE cost is very low.
Obviously the type of service is not as high, but it's still better than nothing if you don't have DSL and you are able to serve this market. And as the market requirements go up and the cost of equipment goes down, you will have more and more WiMax CPEs that will replace the Wi-Fi ones. But I think that if you can use the two technologies together, especially in developing countries to get going.
Glenn: That's what we're seeing I think in all the US deployments at the municipal scale really is WiMax or something very similar to it in the fixed point to multi-point version is the backhaul because it's much cheaper and less fiber is available, and fiber is cheap, then it's the cheapest way to pull the stuff back.
We'll talk more because we could talk about, we haven't talked about Flaring On [sp], 802.20, Qualcomm's role in this, patent holding, and there's all the issues with Voice Over IP, so we'll do another podcast and talk about some more issues but appreciate it. So I think hopefully the listeners have now completely understand mobile WiMax and will explain it to all their friends.
Thank you, Monica, for participating in this podcast.
Monica: Well thank you for having me online. It was great talking to you, as always.
Glenn: It's a pleasure. So this has been podcast number 11 in an ongoing endless series. Recorded on July 25th, I think it's the 25th, 25th, 2006. This is Glenn Fleishman, the editor of Wi-Fi Networking News.