home *** CD-ROM | disk | FTP | other *** search
-
-
-
- The MS programs seem not to work at all with AEA PK232 and PK87
- TNC.
-
- The reason is that those TNC do not protect their buffers from
- an overrun of characters. The result is an emission of wrong
- packets and finally the total crash of the TNCs.
-
- This phenomena happens also in TCP/IP, but is very difficult to
- reproduce.
-
- MS programs implement an Hardware Handshake on the RS232
- interface, so RTS and CTS wires must be on the cable.
-
- When running MS a TNC in KISS mode must be connected to the
- serial interface because a sequence of commands is sent to the
- TNC just at the beginning.
-
- MS version 3 has been developed on an original IBM PS2/30 mod.21
- using Interrupt 15 for handling the keyboard. I have discovered
- that few imperfect clones do hang with this program. Please
- report to me the problems you might find.
-
- The Noise Blanker is very bad with packets so, if the idle noise
- in your QTH reqires it, forget about Meteor Scatter in PR.
-
- The real trick for receiving packets with S0 signals is the
- demodulator. The right demodulator is "coherent ", i.e. it can
- recover the carrier. The best demodulator for 2PSK is the
- Costa's Loop. The right Costas's Loop is analog and has one
- single decision point. Forget about digital implementations of
- the Costa's Loop !! Do not attempt to use FSK on FM...you would
- waste your time...and band too.
-
- Room for improvements: Yes.. ther'is a lot of room... the
- speed can increase to about 4000 bit/sec, but I am afraid that
- the most of the normal transceivers will not handle the band.
- It might take to design a transceiver ad hoc.
-
- 4PSK or 4QAM could be the right answer as for the same band they
- carry twice the bits...and, as they are two 2PSK ortogonal
- signal, ther is no loss in performance respect to a single 2PSK.
- Some work is in progress on the subject.
-
- Happy DX in MS ! de I2KFX Pino
-
-