[nos-bbs] JNOS needs D-STAR abilities

[Jay Nugent]

Greetings,

On Sun, 31 Aug 2008, Mark Phillips wrote:

> Hi all,
> 
> I've been playing with D-RATS all day and have decided that whilst it is
> a very good application it still doesn't address the needs of the
> average data user.

   I concur.

 
> As you may know, D-STAR has a "low speed" data ability on its radio's of
> roughly 4800bd. Why can't we tap into this?

   Actually, the data rate is only 950 *BITS* per second, async.  No
protocol, no handshaking, which is Okay.... see below...

 
> The problem with the current crop of D-STAR data packages is that they
> are all point-to-point apps. Except that is for text messaging which
> appears to be a sort of multicast or at least broadcast.

   ALL data networking is point-to-point.  Hence "routers" are placed at 
every junction.

 
> With the modem already embedded into the D-STAR appliance all we need to
> to is attach the radio to the serial port as though it was a TNC. I'm
> guessing that JNOS would need a driver for D-STAR to accomplish this.

   Gads NO!  No new application or driver needs to be written.  

 
> The problem arises (I think) at the user end. They either have to run
> JNOS as a client or have some sort of "shim" app that captures their IP
> traffic and pushes it through the radio. This sort of already exists
> with D-RATS.

   By simply placing a JNOS box on the 950 bps serial port of each end of
a point-to-point radio link, the JNOS boxes can then run "SLIP" across
that link.  SLIP is the Serial Line Internet Protocol (already built into
JNOS). 

   This *HAS* been tested by members of the ARRL Michigan Section "Digital
Radio Group" (DRG)  [www.MI-DRG.org] and it works quite well.  But it DOES
NOT work ANY better than placing cheap 2-meter rigs and conventional TNC's
on each end for 1/3 the $$price$$. 

   The layout looks something like this:

 +------+   SLIP    +--------+        +--------+   SLIP   +------+
 | JNOS |-----------| D-Star |   RF   | D-Star |----------| JNOS |
 +------+   link    +--------+        +--------+   link   +------+


   You get 1200 bps throughput with conventional KISS TNC's, while you
only get 950 bps throughput with D-Star.  But when using TNC's we use the
AX.25 protocol as the Link Layer".  While with a SLIP link we save the
Link Layer overhead.  So this 1200 bps verses 950 bps is pretty much a
wash between the two systems and throughput is comparabale.  However, 
D-Star when not used purely for a data network costs you dearly.  See more 
below...


> It occurs to me that as the data mode can use the repeaters as well as
> the voice mode this could be quite a useful tool.

   The scenario:  Jay has data to pass to Glenn, while Ken and Mark are 
chatting away on the repeater (that we all share for both voice and data).  
Now while Ken is blathering away about the fine Elderberry wine he just 
made in his basement, can Jay's data be transmitted through to Glenn????

   Nope :(   The *RF* channel is occupied.  The DATA burst will wait until
the channel is clear before being transmitted.  And the ACK coming back
the other way may have to wait several MINUTES before Glenn can
acknowledge Jay's packet.  By then the link will have timed out :(

   So though you *can* use a D-Star repeater system for both voice and 
data. Do you REALLY want to?   Do you WANT your data streams held up for 
two or three *MINUTES* while you wait for the RF channel to become 
available so that data burst can be sent?  I don't think you do.

   But if you DEDICATE two D-Star radios to moving DATA only, you can 
indeed build a dependable and relatively speedy network.  But for 1/3 the 
cost you can use old 1980's vintage radios and old TNC's to do the same 
thing and at the same throughput rate. 

   Or you can dedicate two D-Star ID-1's to the same duty and up your
throughput to 128,000 bps!!!  They are ethernet ready and are configured
with their own IP addresses.  But keep in mind they are *NOT* routers and
still ONLY provide a point-to-point (not multipoint) link.  So the
importance of moving data fast from Point-A to Point-B has to be worth it
to invest nearly $2000.

   In any case, it's your wallet and your choice.


> Am I barking mad here or do I have an idea?

   No, not at all.  This is a great question and one that everyone asks
themselves.  This is a good venue to discuss this technical matter.


> Mark Phillips, G7LTT/NI2O
> Randolph, NJ


      --- Jay Nugent  WB8TKL
          o Chair, ARRL Michigan Section "Digital Radio Group" (DRG)
            [www.MI-DRG.org]
          o Michigan AMPRnet IP Address Coordinator

In response to the issue about dumbing-down Amateur Radio, a fellow Ham 
said:    "Would you like fries with that emergency communication?"
+------------------------------------------------------------------------+
| Jay Nugent   jjn at nuge.com    (734)484-5105    (734)544-4326/Fax        |
| Nugent Telecommunications  [www.nuge.com]     (734)649-0850/Cell       |
|   Internet Consulting/Linux SysAdmin/Engineering & Design/ISP Reseller |
| ISP Monitoring [www.ispmonitor.net] ISP & Modem Performance Monitoring |
| Web-Pegasus    [www.webpegasus.com] Web Hosting/DNS Hosting/Shell Accts|
| LinuxNIC, Inc. [www.linuxnic.net]   Registrar of the .linux TLD        |
+------------------------------------------------------------------------+
  5:01pm  up 11 days, 14:07,  4 users,  load average: 0.72, 0.69, 0.58