OpenBCM V1.07b12 (Linux)

Packet Radio Mailbox

IW8PGT

[Mendicino(CS)-Italy]

 Login: GUEST





  
EI2GYB > PACKET   05.12.18 07:45l 81 Lines 4812 Bytes #999 (0) @ WW
BID : 3248_EI2GYB
Read: GUEST
Subj: INTRODUCTION TO PACKET RADIO - Part 15
Path: IW8PGT<CX2SA<ZL2BAU<EI2GYB
Sent: 181205/0537Z 3248@EI2GYB.DGL.IRL.EURO BPQ6.0.16

INTRODUCTION TO PACKET RADIO - Part 15 - by Larry Kenney, WB9LOZ

Here are some tips to help make your packet operating more enjoyable. 
Whether it's while making local QSOs, checking into a BBS or mailbox, or
working DX, there are a few things you should take into consideration that
will help eliminate problems and waiting time, increase your "throughput"
and make packet a lot more fun.  ("Throughput" is a word that has come
into common usage by packet operators and means the amount of usable
packet information transmitted or received.)

When connecting to another station, don't use a digipeater or node unless
you have to.  Each digipeater you add to the path increases the time
required to get your signal to its destination and to get an acknowledgement
returned.  It also increases the chance for interference and for collisions
with other packets.  You'll be amazed at the difference in throughput when
comparing a direct connect to one with just one digipeater in the path.

The packet node network, as discussed in previous articles in this series,
does a great deal to help you get your packets through, but you must
remember that throughput there, too, is affected by the number of nodes used
and by the conditions between you and the destination station.  The big
advantage of the nodes is that the acknowledgements do not have to return
all the way from the destination station before your TNC is satisfied.
Packets are acknowledged from node to node, so that eliminates a large part
of the problems encountered.  Getting the original packet through, however,
remains to be as much of a problem for the nodes as it is for you when using
digipeaters.  It can take several minutes to get a packet through when you're
working a station some distance away.  Have patience!

Dr. Tom Clark, W3IWI, has determined that for EACH HOP in a packet path the
loss of packets can vary anywhere from 5% to 50% depending on the amount of
traffic.  Remember, each digipeater and node adds a hop, compounding the
problem, and you have twice as many hops as you might think, because of the
acknowledgements.  You can see how quickly the path deteriorates as traffic
increases and digipeaters and nodes are added to it.

If you have a choice, use a frequency that doesn't have a lot of other
traffic on it.  It makes sense that the more stations there are on a
frequency, the more chances there are for collisions and retries.  A path
that will work perfectly without a lot of traffic, can become totally
useless under heavy traffic conditions.  Just one additional station on
the frequency can decrease throughput by about half in many cases.

Another consideration, especially if working over a long distance, is
atmospheric conditions.  You might not have experienced this before on VHF,
but with packet's high sensitivity to noise, a slight change in signal
strength can mean the difference between getting your packets through or
not getting them through.  Long paths between nodes are very susceptible
to these changing conditions.  There are times, especially on a hot summer
day, when it's impossible to get a packet from one node to the other on 
what is normally a good path.  At other times, "thermals" can increase
your range dramatically and you're able to use node paths that normally
don't exist.  In the San Francisco Bay Area, the fog has a significant
affect on VHF signals.  When a fog bank is moving in off the Pacific, it
can act as an excellent reflector.  Signals that normally aren't heard or
are very weak can reach signal strengths of 40 over S9.

Multipath is another problem that can greatly affect your packet signal. 
Multipath is the term used to describe the receipt of multiple signals from
one source due to reflections off of buildings, hills or mountains.  The
"ghost" in a television picture is a form of multipath.  A station with a
very strong signal into a digipeater or node often cannot use that path if
multipath causes the signal to be distorted.  Each packet is checked for
100% accuracy and is not acknowledged unless it is.  Multipath reflection
can cause occasional bits to be lost so you can end up with multiple retries
and a poor path even with strong signals.

To sum up, for best results on VHF use the least number of digipeaters and
nodes as possible, use a frequency with low activity, and be aware of
atmospheric conditions and multipath problems.  Remember, by decreasing
PACLEN and MAXFRAME in your TNC, you improve your chances of getting packets
through under poor conditions.

If you use packet on HF, remember to change your transmit baud rate to 300
and to use a short PACLEN (a value of 40 seems to work quite well) and a
MAXFRAME of 1.  The chances of getting a short packet through the noise and
QRM are much better than for a long one.

_ _ _



Read previous mail | Read next mail


 11.05.2024 09:47:48lGo back Go up