SS and AR

[Dewayne Hendricks WA8DZP <75210.10@CompuServe.COM>]

Kevin, N6RCE writes:

>>     Maybe I should post the whole story of what we're about here soon to
>>clear up things a bit....
>
>Do as you see the need Dewayne.

     There are several efforts here in the SF Bay Area which are trying to
develop some sort of high-speed packet network.  The one I'm associated with is
the local Apple Macintosh crowd.  We are looking for some sort of hardware
which will give us at least a LocalTalk speed network (230.4 kbps) which we
will us with the System 7 based tcp/ip for the Macintosh we have in
development.  To that end, we have been looking at various commerical products
to see if they meet our requirements and if not determine if they could be
modified to do so.
     Wireless networking is a hot topic with a number of commercial firms these
days.  A number of companies have units which operate under the FCC Part 15
rules which don't require a license.  To make a long story short, after looking
at several of these products we selected a company called Proxim in Mt. View,
CA to examine further.  Proxim has several spread spectrum (SS) products which
operate in the 902-928 MHz band with output power of either 100 mw or 1 watt.
By "products" I mean they sell units on an OEM basis for companies to integrate
into their own products (note: they just announced an end-user product of their
own, a 241 kbps rf modem).
     We obtained an evaulation kit from them and have packaged them and
developed support for them in the Macintosh version of the KA9Q bits.  We have
been running various tests to see how well they function and the maximum range
we can get with the 1 watt units.  To date we have been able to get a range of
2.3 mi by using yagi antennas.  The units we have been using have a top speed
of 121 kbps.  We have been doing all of our testing at 19.2 kbps.  Our intent
right now is to mod the units in various ways in order to get the maximum range
possible out of them.  These mods will move them out of the Part 15 arena and
into the world of Part 97.  We have been working closely with Proxim and our
hope is that after all of this testing and evaulation effort, we will be able
to get them to produce some units which will meet our final set of
requirements.

>
>>LANs aren't WANs or even MANs.
>>
>>N6RCE
>>
>>>>Now I've got to make up some yagi antennas and see if I can make it
>>
>>At least people are hearing the message that omni-to-omni is wasteful.
>>
>>     You seem to be replying to both Russ Nelson's and my postings here.
>>
>I did cut a line from Russ's posting, but you also indicated you were planning
>on using directional antenna's.
>
>The comment about LANS, etc and Yagi's aren't directly related.  One of the
>things Glenn has been trying to point out for (ever) is how wasteful
>omni-to-omni is on spectrum usage within digital packet radio for AR.
>
>It seems like you and Russ are discovering experimentally what Glenn has been
>trying to explain therotically for some time.  Which I'm glad is happenning.

     I agree with Glenn's position in general.  However, I feel that there are
certain "mission" specific requirements where omni-to-omni is the best choice.
For those of you who are not familiar with Glenn N6GN's et al thoughts, I
suggest that you read their excellant paper in the proceedings of the 9th ARRL
Computer Networking Conference which is available from ARRL HQ.

>It's hard to explain the last years of tinkering and thinking, but some of the
>things worth noting include:
>
>1)  Radio receivers intended to work in LAN environment ( 50-200 ft)
>    don't have to be as sensitive as receivers for MANs or even WANS.
>    Those PROXIMs work to about -40dBm.  The units Glenn put together
>    so far work down to about -75dBm.  That HT on your belt goes down
>    to -110 dBm ( ignoring or lumping a few Ktb's in).

     Thanks for pointing this out.  You are correct that the Proxim units were
designed for the LAN environment.  The units which I am now playing with work
at -90 dBm.
     Proxim now realizes that they made a mistake in their product design.  It
seems that the prospective OEM's they encounter these days want more range out
of the product then they have been able to obtain.  OEM's seem to be wanting a
WAN product from Proxim and not a LAN.  This was one of the first things their
VP of Sales brought up when we started talking to them.  BTW, Proxim's stuff is
now being used by Nynex and NCR.

>2)  Ethernet ( CSMA/CD ) has a built-in slot time.  The major contributor
>    to the upper bound on slot-time is propogation delay.  What happens
>    with the proxim units when you get a large number of users going?

     That is another good issue.  That was the second problem out of the mouth
of the VP.  They haven't figured out a solution for that one either.  For now,
we will be ingnoring that issue.  It will have to be readdressed again later.

>3)  Frequency re-use and coordination.  Right now, 902 Mhz is pretty
>    much unused, so there is plenty of room for us all to experiment.
>    However, since SS isn't channelized, your either going to have to
>    develop a protocol to handle 500+ hams all on the same chipping
>    sequence, or issue spreading sequences per group.  Again,
>    coordination and cooperation is going to be the real problems,
>    rather than get the technology to work.

     I can't agree with you more.  We are really going to need better
coordination and cooperation in amateur packet radio then there has been (at
least in Northern CA) to get this type of technology into service in the
amateur community.  Proxim has addressed the channel problem somewhat by
implementing switchable channels as part of their design.  It won't however be
able to handle the case where there are 500+ hams using these units.

>Dewayne, these comments aren't mean't at flames at all.  I'm am very glad
>you and Russ are experimenting in this direction.  AR doesn't need just
>one solution or everyone going in the same direction.

     I didn't take this to be a 'flame' on your part.  You raised a number of
very good points.  I hope that what we are doing will complement the efforts of
you and Glenn, et al.  Right now, I am looking our for the needs of my
Macintosh  tcp/ip community.  With the code that we will be shipping this year,
we will need all of the speed we can get in order to run Macintosh applications
across the network.
     Two of the things that AR has been good at over the years is 1) finding
and applying commercial products to the ham radio environment and 2) cost
reducing those products in unique ways to suit the needs of the ham pocketbook.
     As I said earlier, there is a lot of new hardware out there these days to
do high-speed networking.  The commercial amateur packet firms are just now
starting to come out with products which will move us from 1200 bps to 19.2
kpbs.  If we can adapt/utilitize this other hardware which has been already
been designed for low-cost mass production to amateur packet radio, then it
will be a big "win" for everyone!

>With the advent of no-code, we will find the issues you present of
>higher power SS units an increasing reality.  If we aren't to become
>a complete computer-CB band then we must think about the issues of
>building a MAN or WAN ( depending if you live in the Bay Area or
>Wichita, Kansas).

     Right on!!!  It will be a real challenge and I feel that amateur radio can
make some real contributions here.

>However, I'm not sure technology designed for the LAN world can be
>adapted to the MAN world and not create a mess.

     You may be right of course.  We shall see.  In any event, could it be any
worse then the "mess" that exists now? :-)

>GL and I'm looking forward to hearing the results of your work.

     Likewise!!  I will be posting updates to the group as things develop.

73's
-- Dewayne WA8DZP
Internet: 75210.10@@compuserve.com