)This is a copy of a message from Dave Anderson, KG4YZY. There's some interesting information in here about some projects he is involved with to develop a 3.3GHz amateur broadband data system. It's an interesting concept, and one that might be the trend of the hobby in the future. If nothing else, it could bring some new interest into the hobby.
For those network-savvy folks, read on for some of the technical details. The message is long, but the system is explained in a lot of detail. You may want to check out www.ni4ce.org and look at the coverage map for the repeater system. It will give you some idea of how wide an area it would need to serve.
If we had more people coming to the table with ideas like this, maybe we can finally stop hearing all of the bitching about the hobby dying?
--------
>Well I was already wondering if D-STAR (or APCO25, or
>similar) could be used practically for the 220 backbone.
>But this 3.3 system just rendered that thought pointless.
That's the point of it. Plus other options along the way will be doable.
The goal is to have about 20-40meg of capacity between the sites.
(essentially 1/2 of a T3 in the sky)
>How would such a system perform over the 40+ miles
>distance between existing tower sites, summer rainfall
>and all? Would the signal need some help along the way?
I've got FAB customers doing 70km with 30dB (3ft) antennas in 5.8 with 30dB
of fade margin (more than rain/fog would induce by about 10d
. I'm notgoing to push us that far, the 8-10 node layout we've initially designed for
the repeater system linking provides some redundancy, and the longest
"normal" path would only be about 21 miles.
Still need the funds even for this once the POC (proof of concept) is done,
but we'll be setting up the first link between the new Pinellas county site
(in Largo) and Port Richey out of my pocket sometime in the next 30-60 days.
Once the proof of concept proves itself (which it will, I design/sell this
all day long), we'll have some interested folks at that point
Since it's being built as a mesh network, the shortest path always wins, but
all others are considered for redundancy for sites that can have more than
one path.
The need for a high speed fault tolerant network that's not reliant on
cable, satellite, or phone lines for disasters is something emergency
managers -want- now. As amateurs, we can step up to the plate and offer a
alternative for them that will be attractive for disaster response with no
commercial contracts or costs aside from the initial costs of the gear, so
we have a pretty attractive option for them.
Imagine, a hurricane like Charley. It only affected the southern and most
eastern parts of the system's coverage area. Now imagine the EOC's in the
affected areas that had no phone service having links to the state for
text/e-mails/voip over this network. Imagine a staging area with a
portable "ham hotspot" even on 802.11b/g if needed in a pinch that would
link into this backbone for the arriving agencies to use that would give
them data links for their laptops currently not even conceivable. The
options are really endless, when you have the amount of bandwidth we're
talking about having for this type of an application.
>Once this system is a few hundred feet up a tower, what
>would be considered an acceptable (safe) power output?
As hams we're allowed in this band 100 watts TPO (transmitter power output).
That's cost wise absolutely rediculous. A 4 watt amp is 1500.00 alone.
However, 1 watt TPO is easy to achieve at a reasonable price, with a
25-30dBi antenna, it gives a EIRP of around 100-200 watts for the -really-
long hauls.
I have part 15 users that have to obey much lower power limits and still get
incredible distances. The advantage we'll have is we are not going to be
sharing the spectrum we use with other consumers. It'll be only other part
97 users, so the noise floor will be incredibly lower.
Most of our sites will use 400mW radios on the 5ghz side, and 700mW radios
on the 3.3ghz side. These radios are very reasonable. (100 bucks for the
5ghz radio, 249 for the 3.3 ghz radio) with EIRP's down in the 50-80 watt
range that will provide near line of sight communications (3.3 and 5.6 are
not like 2.4, they penetrate water absorbing objects, where was 2.4 will
not).
Well, honestly, 100watts at 3.3ghz will only cause slight, and I mean slight
(.2-.4 degree) heating in human tissue. Eyes are the most succeptable to
this, but that's still not enough to cause any damage, even over long term
exposure. Now the debate of cancer/whatnot related to RF, I'll not get
into, that's not part of the RF safety training I've ever had, it's all
still speculation at this point.
Let's just say an antenna we'd be using for this application would be far
enough from a user say at 4ft it'd be 8 times less SAR than holding a cell
phone, so very safe. Think of it this way, at 70cm, the SAR at 8 inches
from your face is almost 20 times what most cell phones have with the HT at
5 watts
At heights of the high power stuff, the 800-1100 watts EIRPthat the repeaters put off on 70cm with the 110 watt PA's is more of a
concern than the microwave stuff for tower climbers.
For end hams that want to participate in the 5.6ghz side of the data network
(3.3 is the backhaul, the 5.6-5.7 portion of the band is also part 97 only,
and most 802.11a radios will go down that far with special firmware or
drivers) there will be AP's running at each of the sites as well with EMWIN,
Radar, APRS, and a e-mail gateway and AIM/ICW chat gateway to the internet
for end user hams, the EOC's, hospitals, etc that want to partcipate in our
network as we build it out.
The consideration for putting a proxy for web browsing was talked about but
that's not really what the purpose of the network is going to be (a
replacement for a ISP). If a home user wants to setup his own proxy on his
own bandwidth and use it remotely on the network for himself, we'll not stop
them, but the network itself will not "inherantly" have native internet on
it. Clearly, during a disaster this probably would change to meet the
needs of agencies cut off, but the intention here is to -not- create a ham
wireless ISP.
>Either I'm over-estimating this concept, or it's really
>got a ton of potential for additional services down the line.
Ton is really a under estimate. The sky is really the limit. Video QSO's,
data sharing, it's really going to be nice.
NI4CE's goal up front is to link all of our APRS assets and get the linking
off of 220 and over to VOIP (220 will be maintained, just only used as a
backup). We'll be using Cisco routers with special voice cards made for
linking radio sites together called "E&M" ports. At 64kB/sec, the audio
quality is superb, and latency is only 10ms (80ms less than the anti-squelch
crash on the 220 link repeater now). Audio quality on the pair of routers I
have tested this on will actually end up improving audio quality site to
site, plus give us the ability to create a web interface showing real time
what site / frequency is "receiving" the audio that's bringing the system
up, and also provide some web based telemetry (temperature in equipment
cabinet, AC power status, generator running or not, etc). Basically this
will be my toy box for quite some time
This is the purpose of the group we're forming nationwide called ARBA
(amateur radio broadband alliance). ARBA's goals is to promote the use of
the SHF bands for broadband data (bands in the gigahertz range 99% of hams
don't even in most cases know exist). The Tampa roll out is one of 4 we're
working on right now, pending funding from some DHS grants in the pipeline.
Some are already starting to call the concept of ARBA the "ARRL for above
1ghz". We're still hamering out the details of the organization right now,
we've got 7 folks involved, many whom are scientists and engineers, this
group will be very serious at promoting and devoping HAM use of this type of
technology with the technical expertise to make it happen.
>I'll stay tuned...
Absolutlely.
I'll keep you up to date. I'll tell you up front my goal in doing this also has a alterior motive.
I'm wanting to get some of the hams in the area that are IT professionals or
just computer savvy / geeks to take a new look and generate new interest in
the hobby. We've literally got 250+mhz of spectrum that we can take off the
shelf radios and modify with special firmware/drivers to work in our bands
for this application. I think, honestly, once we start doing this, we'll
start seeing a new breed of amateur come out of the woodwork.
