Andy,
What I don't understand is if you already have a suite of modes, Pactor,
Pactor 2, and Pactor 3, then why create another mode like they did?
This is not compatible with existing packet, right? So you would have to
have SCS products on both ends? Then why not use Pactor modes,
especially the Pactor 2 mode which is of a similar bandwidth and throughput?
73,
Rick, KV9U
Andrew O'Brien wrote:
> I found the item (below) on the SCS web site. Anyone use this "new
> class" of packet ?
>
>
> Robust Packet-Radio (RPR)
>
> Up to now Packet-Radio over shortwave has been basically a
> non-starter, it has even been heavily criticized because of the low
> effective throughput and repeats. AX.25 is for shortwave not an ideal
> protocol, but with automatic FRack-setting and a small MAXFrame value
> the protocol should, however, function much better on a shortwave
> channel than has previously been the case generally.
>
> One cannot of course expect an asynchrone protocol to reach the same
> efficiency as a tight synchrone ARQ protocol (e.g. PACTOR), but for
> some applications a multi-user service, with very uncritical
> transmit/receive switching, as well as almost zero power holding up a
> connection when no data passing, brings a real advantage that
> outweighs the lower data throughput.
>
> What finally are the reasons that up to now HF-PR works so poorly, and
> apart from "forwarding" is hardly ever used? One finds a simple
> answer: The current modulation type for HF-PR namely uncoded 300 Bd
> FSK is really unsuitable for normal HF channels. The symbols are much
> too short even with moderate "Multi-Path effect" ("delay spread") to
> work. Additionally, because no sort of error correction code is used,
> even short troughs or "static" will destroy a many seconds long
> Packet. Just one missing bit leads to a repeat of the whole packet.
>
> To help cure this problem, SCS has developed a new class of robust
> modulations types especially for Packet-Radio. As a special feature
> for all the variants of this "Robust PR", a completely new
> synchronizations algorithm with "catch" properties that were not
> possible before has been realized. Frequency deviations of ±250 Hz are
> immediately recognized and without any loss of sensitivity
> compensated, and this with signals that are buried deep in the noise.
> Because of this it's possible to remove a tuning display. One can say
> with good conscience that this is "Plug and Play" for shortwave.
>
> The currently available "Robust PR" modulation types have the
> following properties:
> Bandwidth: 500 Hz @ -30 dB
> Modulation: Pulse-Shaped OFDM (BPSK, QPSK), similar to PACTOR-III
> Average Throughput: 200 or 600 Bit/sec
> Crestfaktor: 3.0 or 4.2 dB
> Delay-Spread: up to ±8 msec is tolerated
> Coding: High performance convolutional code, "full-frame interleaved",
> rate/2 or rate3/4
>
>
