Monday, April 13, 2015

TX-enabled IGATES not harmful

.. food for thought....


http://blog.nwaprs.info/2010/10/summary-on-uiview-and-igates-by-wa8lmf.html
note 1 :  I don't know why this isn't valid advise for any APRS software. 
note 2 : see point 2 below, you have to beacon out and be heard by the igate before they start transmitting you traffic. 


Sunday, October 3, 2010


Summary on UIView and IGates by WA8LMF

1) UIview does an excellent job as an igate. Contrary to the previous posting on this thread that two-way igates are difficult to set up, the default settings in UIview make it almost trivial to place a two-way igate on the air with next-to-zero impact on the RF channel.

2) Two-way igates DO NOT indiscriminately flood the RF channel with traffic from the APRS Internet System. Unless you go out of your way to explicitly edit a control file in UIview, the program only "reverse-gates" (i.e. Internet-to-RF) messages addressed to specific callsigns --- NOT generic APRS traffic such as WX beacons and position reports that are "broadcast" to everyone, and not reverse-gate to stations that are not in the immediate area.

3) On the "Setup, APRS Server Setup" dialog, locate the area in the upper right of the dialog that says "Gate RF To Interenet". Check "Open the gateway" and "Insert station callsign". In the area just below this ("Gate Internet To RF"), check "Gate local messages" and "Use reverse digi path" .

4) The option "Use reverse digi path" causes UIview to override the normal "Unproto Address" digi path set in "Setup, Station Setup" for the station's own beacons, when reverse-gating messages. Instead, messages sent to stations heard by the igate directly (i.e. not via digipeaters) will be transmitted to RF with NO DIGIPEATER PATH AT ALL.

5) Messages will be reverse-gated IF and ONLY IF the destination station has been heard on RF locally (either direct or within one digi hop) --- and --- if the destination station has been heard on RF by the igate recently (within the last half hour).

6) This process of selectively reverse-gating on a minimal path works both for messages originated by the RF (mobile) station, and for messages originated by the Internet party.

7) It even works for a station on RF in one part of the world (say on the East Coast or in Europe) working a mobile on RF in another part of the world (say the West Coast). That is a message path can go RF<-->Internet<-->RF in a manner similar to a station on RF in California talking to a mobile in Europe via EchoLink or IRLP.

The same reverse-gating rules apply: If the RF stations on each end are both local to their respective igates, and have both been heard by their respective igates recently, then two-way messaging will work. Otherwise, reverse-gating will not happen.

8) As a mobile travels cross-country, it will move out of the local range of a given igate, and into the range of another one. When an igate stops hearing the mobile LOCALLY (it might still hear the mobile via several digi hops), the digi will then age the mobile's call out of it's table of "recently heard stations" and will stop reverse-gating any messages to/from that mobile. Hopefully by this time, the mobile is now within the "local" range of another igate which will take up the task of reverse-gating. [This process is very similar to a traveler with a cellular phone moving from one cell site to another.]

Note that the stations in question don't have to be messaging for these decisions to be made. The igates make the determination of "local" and "recently heard" based on the periodic "one-way" position beacon transmissions the station is normally making anyway.

8) One-way igates break this symmetry of send/receive. A mobile originating a message will get INTO the Internet system through a one-way igate, but the reply (ack) can't get BACK to him. The originating mobile responds by cluttering the RF channel with endless automatic repetitions of the message in a vain attempt to get an ack back. In other words, one-way igates can create MORE congestion than two-way ones!

Further the Internet (or remote RF station) being messaged can't reply to the incoming message he just received.

Stephen H. Smith wa8lmf (at) aol.com
Home Page: http://wa8lmf.net

No comments:

Post a Comment