Bulletins from the Pacific Packet Radio Society - page 109

The original petition by the ARRL requested increase of permissible speed from 1 200 to 4800 baud between 50 and 225 MHz. The FCC's notice of proposed rule making proposed that sending speeds up to 19.6 kb/s may be used on ASCII (unspecified speed on additional digital codes) so long as the bandwidth did not exceed 16 kHz. Comments filed by AMRAD suggested that the speed (particularly in the 220-MHz band) be raised to 56 kb/s and that bandwidth be increased to 100 kHz. The League filed comments in support of this and a similar request by Paul Newland, AD71 to raise the limits above 420 MHz. So, at QEX press time, we have to say that we don't know the speed limits in the new rules. But we should know within a few days.

The maximum bandwidth and/or speed is particularly important in the development of a packet radio backbone network (internet) which will connect between cities. If it is too slow, the network trunks will not have enough capacity to handle peak traffic loads.

At this point, if the new rules were effective tomorrow, we would have a difficult time finding high-speed modems. Someone's rule of thumb said that modems above 1200 cost something like a buck a baud . Today's competition seems to be driving the commercial ones down to about half that. Even at that, the price is awful and will drive us to home-brewing. Among those of us who have been discussing high-speed modems, there is no agreement (yet) on the best type of modulation to be used. Leading candidates are: msk (minimum-shift keying where the shift in Hz is half the baud rate) and qpsk (quadinary phase-shift keying).

Packet radio experimenters with Vancouver terminal node controller (TNC) boards have been using non-return-to-zero inverted (NRZI) encoding. NRZI is no longer considered to be state of the art in the computer industry and is being replaced by others such as Manchester and Miller encoding for magnetic storage media. One characteristic of some of the more sophisticated encoding schemes is ensuring of changes of state often enough that the dc component is eliminated. Bob Carpenter, W30TC, raised the possibility that the right encoding scheme could permit transmission of data without a modem by feeding the data stream directly into the microphone input of a transceiver.

The FCC action clears the air for us to experiment, at least above 50 MHz, with forward error correcting (FEC) codes. There hasn't been a pressing need for them to date in the handful of amateur packet radio networks in North America which are mostly local in coverage. High-speed inter-city paths which are long enough to encounter fading on vhf could be good test beds for certain types of error-correcting codes.

Docket 81-699 applies only to 50 MHz and above. It will not change the situation on hf where we are currently limited to Baudot and ASCII codes. The League has petitioned the FCC to permit "AMTOR" transmissions in the hf bands. That petition has been designated RM-4122. (For details see QST August, 1982.) At

Click for Original Graphic Image of this page.