Previously, I wrote about my modest APRS gateway, KD0TLS-1. It's a 10W Leixen transceiver (that actually puts out approximately 8W), connected to a monoband 5/8 whip mag-mounted to my third-story balcony. It generally has a ten-mile coverage radius, and was originally intended to leverage the wide coverage of the K0YTH-13 digipeater in Robbinsdale to reach the entire metro area. Unfortunately, that digipeater has become unreliable, often disappearing for months.
In other words, it's a minor gateway with a small footprint. It supplements other (larger) elements of the local infrastructure and has found a niche in handling relatively large numbers of RF-based APRS messages.
About once a month, something strange happens to this gateway. Suddenly, it starts gating huge amounts of digipeated traffic. However, the direct coverage (stations that my gateway hears without the help of a digipeater) remains unchanged.
 |
| Click to enlarge |
Normally, in a good month my modest gateway handles about 400 digipeated packets, and that's when the Robbinsdale digipeater is active to funnel traffic to my I-Gate. Currently, I'm well over 18k digipeated packets, and we're not even a full week into this month.
It also includes traffic that is quite distant, from the Upper Peninsula of Michigan, across the whole of Wisconsin, out to the border of South Dakota, north to Canada, and deep into Iowa. It's very difficult for me to believe that there are no other gateways that would handle this traffic more locally. Why is any digipeated packet from, say, Fort Dodge IA passing up local gateways, the Faribault gateway, the large metro gateways, etc. to find its ultimate home at a minor gateway in Plymouth?
 |
| Green Bay - click to enlarge |
 |
| Iowa - click to enlarge |
 |
| Michigan - click to enlarge |
To be clear, I'm not complaining about this phenomenon. I just find it completely baffling. It makes no sense at all. I'm happy that my modest gateway can stand in as virtually the only I-Gate for multiple States. But it's not the only I-Gate; the network is just behaving as if it was.
Now, before others start handing out glib and wise 'explanations' for this phenomenon, let me just say that nothing has changed in my gateway to account for this. If my RX capability had suddenly and mysteriously improved by a thousand-fold, why am I not gating large quantities of local traffic directly? My direct coverage remains as small as it ever was. My station stats show that W0YC-5 is the most distant station that I receive without the aid of a digipeater, and W0YC-5 is only ten miles away.
If this was just one DX packet, it would be easy to dismiss it as an anomaly produced by tropo propagation. But we're talking hundreds of DX packets each day, mixed in with thousands of digipeated packets from a 100-mile radius. Hundreds are flooding in each hour.
To put the scale of this traffic into perspective, my minor suburban gateway has (at this moment) handled more digipeated packets in four days than the big Medina gateway W0PZT-1 handles over one month. And that number rises continually.
One other thing deepens this mystery. I have now gated hundreds of ID packets from W0PZT-1. In the time it's taken me write this post, that count has risen to 280 from 273. The strange thing is, W0PZT-1 doesn't transmit ID packets over RF. Technically it does, but those packets are internally gated by N0AGI-1. No other metro gateway is reporting RF ID packets from W0PZT-1, and that includes stations that are far more capable than my minor I-Gate. Isn't it strange that the big Medina gateway is transmitting RF packets that can only be heard by my small I-Gate? Isn't it strange that this only started less than a week ago?
W0PZT-1 is a station whose transmissions commonly reach north of Duluth. Yet, a modest station eight miles away is the only gateway able to pick up these RF transmissions from W0PZT-1? Not weird enough? Try this: nothing in the raw packet data from W0PZT-1 shows any RF packets being gated by KD0TLS-1. Instead, it shows up in the data for KD0TLS-1.
 |
| Click to enlarge |
Again, I'm not complaining about the need to gate RF packets from W0PZT-1. The point is that nothing about this makes sense, and my reception of packets that otherwise don't exist to the wider world are simply another data point in the very strange data plot that is emerging.
As I mentioned before, this isn't the first time this has happened. When it happens, it's always just after I reboot the Windows 10 tablet running APRSIS32 -- usually because of a Windows update. It ends a week or two later as abruptly and mysteriously as it started. The RF packets from W0PZT-1 stop coming in. The digipeated packets from far-flung locations are gated by local stations. Nothing has changed in my radio, antenna, internet connection, or software. Everything returns to "normal".
While merely anecdotal evidence, I have seen this same thing happening to other local gateways over the years: small RX-only gateways suddenly start gating a flood of packets from all over the metro and well beyond it. It ends after a week or so.
I don't have an explanation for this phenomenon. One could dismiss it as some kind of "glitch" in the APRS servers, but that's facile. How would packets from Canada, Iowa, Michigan, and South Dakota even reach my gateway on two or three hops? Again, I have a proven 10-mile RX coverage radius. When I track down these distant stations, they all exist on aprs.fi and they all have RF capability -- so these aren't "ghost" packets from non-existent stations.
One of the things that initially attracted me to APRS is that everything was so clear: every packet sent is documented on aprs.fi, you can see the path each packet took to reach its destination in the raw packet data, which digipeaters heard my packets was visible, and everything was tracked and recorded on gateway station stats. Later, I discovered that not everything is recorded, but the vast majority of traffic is. Still, there was nothing mysterious or baffling about this. Some RF traffic never gets gated. In other cases, stations designate their transmissions as NOGATE or RFONLY to deliberately avoid being gated. Perfectly fine, but hardly a mystery.
To come face-to-face with something so incomprehensible, but undeniable, makes me wonder what other mysteries are out there to discover with APRS.
In another week or so, KD0TLS-1 will return to obscurity. Operators in Green Bay, Two Harbors, Mason City, etc. will find other gateways to handle their traffic. That's as it should be. For the moment, however, I'm glad that I went to the trouble to set up a gateway years ago. Otherwise, I never would have experienced this.
EDIT: Ultimately, this strange event ended with a Windows update, forcing the Win10 tablet to restart on July 13 at 9 PM. Well over 63k packets from stations as far away as Kansas and Illinois (and a huge number of local packets) made their way to my modest I-Gate. Nothing changed with the radio, TNC, or antenna, yet the flood of traffic (both distant and local) suddenly stopped. KD0TLS-1 has returned to "normal" operation -- though handling no RF messages since the restart.
EDIT: It's unbelievable to me that some people have turned this post completely upside-down. Ignoring everything I wrote here, they seem to believe that what I am really saying is, "Duuhh...my station is very powerful. It's the best there is!"
That's the exact opposite of what I wrote, of course. But, having invented a completely different post out of whole cloth, it's not a far step to imagine that I am really begging for advice on how to improve my APRS gateway. These wise mentors revealed two deep secrets that (apparently) I am unaware of: more powerful rigs exist, and a tower will improve the range of my station. Gosh, who knew?
Please stop emailing these ridiculous attempts at 'help'.
No comments:
Post a Comment