MAC sublayer protocol

[Glenn Elmore <glenne@hpnmdlc1.hp.com>]

Steve Johnson writes (in part):
> Agreed.  Am I wrong in assuming that 56k and microwave packet systems
> are purpose-built radios, and do exhibit superior turn-on/off delays as
> compared to low-speed modems connected to standard voice radios?

I'll try to answer a couple of hardware specific questions here.  Look
elsewhere for a more general answer.

  Both the 256 kbps 900/1200 MHz radios and the uwave stuff  I did was
targetted specifically for amateur datacomm.  The 10 GHz stuff primarily
exhibits the propagation delay associated with the path since rx/tx both
run full time and propagation delay dominates over delay of the bandpass
filters.  At 1 bit/microsecond it is by no means less than a bit time
but it is small compared with most current simplex radios which have to turn
transmitters&receivers on and off.

  The 904 radios achieve valid Tx RF/data at their antenna connector
about 10 microseconds after RTS is asserted.  The receivers come alive
at full sensitivity about 40 microseconds after RTS drops.  As mentioned
in a radio status update last year this allows the receiver to be ready
before the end of the data has reached the other end of the link. This
was the essence of the "packet radar" capability.

> This seems the crux of the matter.  Controlling and limiting the reconfig
> overhead is crucial.

Yes.  With the above hardware it does seem to dominate.  See the CNC
hubmaster paper for details and analysis.

Glenn n6gn