more problems soon for US 2.4Ghz band

[hal]

The idea of urban hotspot networks in not new.  The usefulness of these 
networks in undeniable, however,
it is going to present serious problems for continued ham /secondary/ 
usage.   An urban hotspot network consists
of multiple Access Points (AP) spaced very close to one another in order 
to provide something like constant coverage within a geographic area.  
It is a common thing on a university campus or laboratory grounds. This 
basic
idea is now being planned to cover larger urban area such as a downtown 
district or an entire small town.
Planning is also underway to install urban hotspot networks in a section 
of a city, for example, Manhattan
between Battery Park and Midtown in NYC or downtown Philadelphia.  A 
wifi user might go anywhere within the wide coverage area and stay 
connected.
A primitive network design might not support true roaming, enabling a 
user to seamlessly move between APs,
however, more sophisticated single vendor systems do support this 
feature and their equipment might be
chosen for installation.  

The popularity of the 2.4Ghz band makes this frequency a good choice for 
an urban hotspot network. An architecture to support this kind of 
network will consist of a high speed fiber backbone connecting 
periodically spaced APs in a grid formation.  Some RF relaying, similar 
to packet repeating, might also be used. 

A bit of thinking should convince you that a well designed system will 
open up all kinds of new
commercial and personal use.  The old joke among Arpanet researchers 
talked about "toasternet"
a humorous mythical future network so pervasive that even a household 
toaster will have its own IP address.
 I do not see that as
so much a joke anymore than as a certainty.   So many  users who will 
discover so many uses of a new urban hotspot network will create fast 
growing traffic demands.  Among 802.11 designers it is well understood
that the 802.11 protocol was never designed nor intended to support this 
kind of common carrier access situation. The basic structure can easily 
become overloaded and there is no place to go but adjacent frequencies.

Now I want to say a few words about ham usage of the 2.4Ghz band.  There 
are two ways people use this
band, the first is the transitional ham usage in contests. satellite and 
some point-to-point experimental links. Then there is the ham wifi 
experimenters who, because of ham rules, can run higher power than their
non-ham co-occupants.  It will be very difficult for either of these 
groups of hams to use this band in any
location near an urban hotspot network because of interference.  The 
interference will show up as an increase
in the frequency of retransmissions and lower throughput. This might be 
mitigated by using very tight antennas
in a point-to-point link but broad beam widths near these urban centers 
will spell problems.

Although urban hotspot networks will happen and there is probably 
nothing hams can or ought
do about it, there is one
area that hams can have positive influence.  Commercial Wifi operators 
are also worried about interference
from ham usage of 2.4Ghz and try to stay informed by reading ham mailing 
lists.  Alas, hams seem intent
on mudding up the waters when asked a question about their operations.   
For their part, the commercial operators are not reassured at
the usual maneuvering on ham mailing lists and seriously talk about 
approaching the FCC (or their congressman or senator) for protection or 
at the least monitoring and opposing Part 97 NPR that they /think /will 
complicate their operations.   Gloating over their tertiary status as 
having to suffer through whatever interference
 secondary or primary users dish out will not lead to an understanding 
based on
confidence and trust.  This kind of thing could be headed off but we 
hams ought to
be better acquainted with the spectrum management issues on 2.4Ghz 
before offering up opinions to
concerned  commercial operators. 

Hal/wb3kdu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://amrad.org/pipermail/tacos/attachments/20060530/b5b62cc8/attachment.htm