APRS.fi scaling for graphics display and suggestions on intrepretation on monthly receiver graphic

[Richard Driscoll]

Hello Hessu and everyone else on APRS.fi google group.

Thank you for providing an excellent tool and some very interesting articles.

I am working to analyze APRS performance in the Houston, Texas area after having TEAC (.net) club members indicate they could not access the local digitpeater/igate W5SI-1.

I applied Bob KB5ARS (sk) solution suggest in his 2005 article "Fixing LA".   That has addressed the immediate problem.

Now, I am using APRS.fi to review the general performance of APRS digipeaters/igate in Houston area. 

So three questions:

1. Is there a missing multiplier on the packets transmitted graphic (x100 perphaps)?  On that note, what is the auto vertical scaling range?

2. Am I correct in assuming a basic straight line display for packets transmitted represents a 100% duty cycle for that particular site?

3. For the Monthly receiver performance graphic is an explanation on how to use and what the horizontal scaling represents?

Thank you for any suggestions.  If you need additional clarification on information or questions, please let me know.

Sincerely,

Richard

KD5URB

[Heikki Hannikainen]

On Sat, 5 Mar 2022, Richard Driscoll wrote:

> I am working to analyze APRS performance in the Houston, Texas area after having TEAC (.net) club members indicate they could not
> access the local digitpeater/igate W5SI-1.
>

> So three questions:
>
> 1. Is there a missing multiplier on the packets transmitted graphic
> (x100 perphaps)?  On that note, what is the auto vertical scaling range?

No, I don't think so. Note that it is packets transmitted by that station,
not packets forwarded by an igate. It will scale up to any number that is
necessary, 0 to infinity.

> 2. Am I correct in assuming a basic straight line display for packets
> transmitted represents a 100% duty cycle for that particular site?

I'm not sure I understand the question. If a station sends a packet once
per hour, it will show a straight line at Y == 1, but it is quite far from
100% duty cycle, as it could probaly technically transmit a thousand
packets per hour, or two thousand.

> 3. For the Monthly receiver performance graphic is an explanation on how to use and what the horizontal scaling represents?

Horizontal scale is in kilometers. Try to make that line extend to longer
distances by improving receiver performance, antenna, etc. The "heard" map
may be useful too.

https://aprs.fi/doc/guide/aprsfi-network-tools.html

- Hessu

[Richard Driscoll]

Hello Hessu

Thank you for the prompt and detailed replied.  I figured out km after I had sent the message. I missed the obvious initially.

If I have done the calculations correctly @ 1200 buad, there is a potential for 9600 each 50 character long messages per hour.  I realize this calculation ignores many variables and assumes data transfer in only one direction but it does indicate potential magnitude of signal rate. 

Looking at W5SI-1 (UIView-32 software), I see 4-5 transmissions a minute or ~240 APRS messages an hour.  I do recognize that aprs.fi data display has to accumulate and then process data, so there is some lag by necessity in the system.  When I look at https://aprs.fi/info/graphs/a/W5SI-1, I am unable to correlate that direct information to what the graphics are indicating.  I am assuming a local time display not GMT and I am not concerned about +/- one hour due to daylight savings time.

I hope I am not missing the obvious here as well.....

The second chapter of this is reviewing the APRS data for other local digipeater/igate combinations on APRS.fi.

If you look at N5LUY-2, -3, -10, KI5GJM-10, W0MAC-1, KC5SQD-1,  K5WH-8 and others all show an almost straight line packet transmission rate regardless of time of day.  It this a real indication of continuous data rate and or is there something in between that limits data?

Most Sincerely,

Richard

KD5URB

[Richard Driscoll]

Hessu,

You're correct, The heard map indicates is also a good tool to use.

Thank you,

Richard

KD5URB

[Heikki Hannikainen]

On Sun, 6 Mar 2022, Richard Driscoll wrote:

> Looking at W5SI-1 (UIView-32 software), I see 4-5 transmissions a minute or ~240 APRS messages an hour.

Looking at the raw, packets, it only sends out packets every 10 minutes or
so, "643 seconds between packets on average during 31500 seconds":

https://aprs.fi/?c=raw&call=W5SI-1

and the graph does indicate 6-8 packets per hour:

https://aprs.fi/info/graphs/a/W5SI-1

So it's probably calculating it right.

Note that this is packets *originated* by that station itself, not packets
it retransmitted or otherwise forwarded.

>  I do recognize that aprs.fi data display has to accumulate and then
> process data, so there is some lag by necessity in the system.

Those graphs, except for the receiver performance graph, update
immediately.

>  When I look at
> https://aprs.fi/info/graphs/a/W5SI-1, I am unable to correlate that direct information to what the graphics are indicating.  I am
> assuming a local time display not GMT and I am not concerned about +/- one hour due to daylight savings time.

The graph headers indicate time zone or UTC, and you can adjust that in
Preferences -> Units and time.

https://aprs.fi/doc/guide/aprsfi-preferences.html

> The second chapter of this is reviewing the APRS data for other local
> digipeater/igate combinations on APRS.fi.
>
> If you look at N5LUY-2, -3, -10, KI5GJM-10, W0MAC-1, KC5SQD-1,  K5WH-8
> and others all show an almost straight line packet transmission rate
> regardless of time of day.  It this a real indication of continuous data
> rate and or is there something in between that limits data?

This is the number of packets received from that station at aprs.fi, check
the raw packets for each callsign. Most igates and digipeaters and other
non-moving stations transmit their position at a fairly regular fixed
rate, once in 10 minutes or 20 minutes or so, resulting in a rather
straight line.

- Hessu

[Richard Driscoll]

Ah, thank you Hessu,  I apologize for taking up so much of your time.  I appreciate your patience and detailed explanations.

Sincerely,

Richard

KD5URB