Farpoint Group –
I've been working on another white paper on capacity management in wireless LAN systems. You can think of capacity management as kind of a system-wide extension to what happens in RF Spectrum Management (RFSM), a topic I covered in a white paper a while ago. Since then, I've been thinking about the need to control and manage wireless LAN systems from more of a macro perspective, something that becomes essential as (a) wireless LAN installations grow, and (b) dense deployments (see my last column on this) become the norm. Just like modern jet aircraft can't fly without the aid of dozens of microprocessors, the same is true for wireless LANs - manual control is no longer possible as an installation grows beyond a few APs. Anyway (I digress, as usual), the white paper on capacity management is almost done, and I'll have it for you shortly. I'm just checking one fact, and waiting for one last person to call me back. As an aside (here I go, more digressing), have you notice how busy everyone is today? It's funny, in a way, that wireless communications was supposed to make us more efficient and responsive, improving productivity, customer service, and even our quality of life. Now we find that many are essentially tied to their cell phones, and that workloads have grown to fill in all that free time we were supposed to have. We've pushed productivity, for many, to the breaking point, and I'm not sure we're still making progress. As I've often said, the sociology of wireless is much more important than the technology, and I'm sure you will be hearing more on this topic.
So, while I wait for my colleague to call me back (presumably on his cell phone), I have a couple more random thoughts for this week.
There have been a number of articles on the city of Philadelphia's plans to cover a 135-square-mile area with Wi-Fi, but my favorite was this one in the New York Times. I like the Times because it always refers to people in stories using honorifics like "Mr." or "Ms." or what have you. I'm fairly informal myself, but I do like the idea of showing respect to everyone, and this is a good way to do it in print. (What, more digression? Sorry...) Anyway, Philadelphia would like to provide broad Wi-Fi access to everyone in town, and a number of members of the press have called me for a comment. Two questions were typically posed. First, is it practical? Answer: yes. We have the technology. As I've written before, Wi-Fi meshes make deployment easy and are quite adaptable to growing user demands over time. Secondly, should the city itself be doing this, as a public utility, rather than simply licensing the service to an independent operator? I don't think the city has any plans to own and operate the network, nor should it. That's a job for the private sector. But there's another issue, and that is that the city also shouldn't be favoring one communications vehicle over another. Those who own and operate cable and DSL services shouldn't be disadvantaged unfairly as wireless services begin to compete with them. In other words, the city should remain neutral. Their interest in providing coverage to all parts of town, and not just the wealthy ones, can be preserved via a suitable contract.
One other matter - the wireless devices necessary to implement the infrastructure are likely to be attached to street-light poles, with the city owns. Given the potential for interference if lots of providers claimed the right to such pole attachment (or other suitable mountings), there is clearly a need to limit the number of broad-scale for-profit Wi-Fi services in any given area. I've advocated the licensing of the otherwise unlicensed spectrum for such for-profit activities, so as to limit the amount of broad-scale interference generated, and thus to protect us not-for-profit end-users for whom the unlicensed spectrum was originally intended. Regardless, local policy can be made to both encourage new services while not disadvantaging existing players (both wired and wireless), and policies should also be put in place to prevent the overbuilding of for-profit wireless services (of any form) in the unlicensed bands. We've certainly not heard the last of this, and I still expect metro-scale Wi-Fi to become common in many cities over the next five to ten years.
Next, we have the curious (and somewhat related) case of dorm-room Wi-Fi in Texas. It seems the University of Texas at Dallas for a time banned the use of 802.11b/g (i.e., 2.4 GHz.) personal access points in their dorms, citing interference with the already-in-place campus Wi-Fi network (see this Slashdot discussion for a lot on this). This has brought up a whole bunch of interesting arguments, and while the University later backed off, this issue is far from settled.
Can a landlord restrict Wi-Fi use, even under the terms of a lease? I'm not a lawyer, but it seems this is a matter of federal law, and only the FCC has the right to regulate spectrum use. Years ago, some landlords and condo associations tried to restrict the placement of satellite TV dishes, but the FCC said no dice. Could a landlord restrict other legal activities, just because the landlord wants to? I doubt it. But, as noted above, we've not heard the last of this issue. As metro-scale Wi-Fi networks become increasingly common, they might interfere with residential or business Wi-Fi nets that are already installed. Such interference would typically materialize as lower throughput, but, as more voice traffic starts to go over WLANs, it would be, well, like a bad cellular connection - still all too common. This problem at present has no solution, other than the use of 802.11a, which operates in the plentiful spectrum above 5 GHz. Most newer 802.11a implementations use capabilities added in 802.11h; these are Transmit Power Control (TPC) and Dynamic Frequency Selection (DFS), and are required for operation in Europe nonetheless. Both help deal with interference to a great degree, and are a further incentive for the eventual migration to 802.11a and its successor, 802.11n. I'd go so far as to say that, just like the 902-928 MHz band before it, the 2.4 GHz. band will be abandoned as WLAN traffic becomes more pervasive. There's only 84 MHz of spectrum at 2.4 GHz, but 555 MHz. available to 802.11a and n - plenty of room, and room (spectrum) is still the best way to avoid interference and the many detriments that come along with it.