[drats_users] AX.25 Support
compuvet at aol.com
Mon Jun 1 23:27:22 PDT 2009
Dan:
Having spent some time digging under the covers of AX.25, I can see why
you would not want to build in support. AX.25 is intended for
multi-hop networks like the Internet we know and love today. True
AX.25 support would require both a command mode (control plane) and
converse mode (data plane). D-STAR data doesn't work that way. The
user controls RF routing by setting the radio parameters such urcall,
rpt1 and rpt2. For true AX.25 support in D-RATS, you would have to
build in a command mode to set TNC parameters. As I think you realize,
this may be too much effort for too little gain. AX.25 doesn't offer
us that much.
However, if you stick to KISS mode and only support Unnumbered
Information frames (UI), you could greatly simplify the command mode
structure. You could provide a simple form for configuring the initial
TNC parameters and the rest would operate as D-RATS does today on
D-STAR. A few initial TNC terminal commands to enter KISS mode and you
are done.
To send data to a KISS TNC, you only have to frame the data with the
Frame End (FEND) character xC0. There are escape charater sequences
for data that contains the FEND character. FEND in the data is
replaced by FESC TFEND (xDB xDC). FESC in the data is replaced by FESC
TFESC (xDB xDD). No CRC or checksum is provided and no RS-232C
handshaking signals employed.
The first byte of each frame is a type indicator and port indicator.
low-order nibble is command number and high-order nibble is port
number. Only five commands are required to be supported: 0,1,2,3 and 5.
Command '0' indicates data frame
Command '1' sets transmitter delay
Command '2' sets persisitance
Command '3' sets slot interval
Command '5' sets full or half duplex
There are no TNC based acknowledgements, but that's ok because you
handle them in D-RATS when necessary. At this point the only
difference between a D-STAR data port and the KISS TNC is the frame
charaters and escape characters.
Phil Karn KA9Q was one of the authors of KISS TNC. He could provide
all the protocol details.
I think this was originally your idea.
de k4lsu, Craig
More information about the drats_users
mailing list