Is Packet Radio Dead?

Discussion in 'Working Different Modes' started by K0LTZ, Dec 11, 2019.

?

Is packet radio dead?

  1. Yes

    36.0%
  2. No

    64.0%
ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Left-2
ad: Left-3
ad: HRDLLC-2
ad: abrind-2
ad: MessiPaoloni-1
ad: L-Geochron
  1. W4EAE

    W4EAE XML Subscriber QRZ Page

    If ARES has a future in aiding with comms in emergency situations, then packet is essential. ICS forms are becoming an absolute requirement. Sending critical information by voice just does not fit into the overall response structure anymore.
     
  2. W4EAE

    W4EAE XML Subscriber QRZ Page

    I meant to add more...

    You are not longer asked to 'send this message,' you are asked to 'sent this exact form.'
     
  3. KA2DEW

    KA2DEW Subscriber QRZ Page

    I've been playing with packet radio stuff for a while, though not continuously. I was not really playing with it from 2002 until about 2013. I was disenchanted by how it had all become Internet based while still looking like ham radio. I finally took it upon myself to see if I could change that, at least locally.

    I think the packet radio tools, TNCs, software, etc.. are very useful for creating one's own social group, much like putting up a repeater and equipping locals with radios can be a thing.

    There is a ham-thing which is held by many hams in which they will want to build a station and then go operate. I think if this is the way we look at packet radio, it's pretty low in function, except for APRS. APRS seems to be a thing still.
    But... if one is willing to work with other people and build stuff, using packet radio tools is a bit of fun. About 6 years ago I started something in Raleigh NC. At the time, the only daily use of packet radio around here was APRS. There was infrastructure held by ARES groups that were not for public consumption as well. Anybody else in my area using packet was doing it on HF.

    What I did was to build several packet stations and loan them out to get people to start a social network. The several of us went to a few club meetings to demonstrate the equipment and attracted a few more people. Soon the newcomers were buying their own equipment. 6 years later there are 30 or so equipped people and a 22 station network across 60 miles. We wrote a few rules of behavior which turn this into a ham-radio-exclusive project, and I think that helped with attracting many of the newcomers. It also makes it rewarding to the participants in that they are truly responsible for making the network operate.
    Check out http://ncpacket.org and TARPN There are also several videos on youtube under the search-term "TARPN".

    Tadd - KA2DEW - Raleigh NC
     
    NQ1B, AH2AP, KX4O and 3 others like this.
  4. KA2DEW

    KA2DEW Subscriber QRZ Page

    I just had another discussion about the topic of packet dead vs packet opportunity. Here is the summary

    Once upon a time we had packet radio in most of the country. It was implemented well in about 1/10th of the country. The rest was really bad. People put single port nodes on top of huge towers and huge mountains and then expected everybody who could hear it would be able to make use of it and be happy. It doesn't work that way. Why didn't anybody tell us about Aloha and the affect of collisions on connections?

    The total throughput on a channel, in bits per second per person, is, at best, ((baud-rate/3)*shared_channel_factor)/(num-of-users -1))

    The /3 is because of channel overhead, acknowledgements, TXdelay, callsigns sent both ways, procedural bytes in each packet.
    The shared-channel factor is 1 if there are never more than 2 stations on the channel, or 0.18 if there are more than 2 stations. This is related to the Aloha results about how much channel time must be free to avoid collisions (82%).
    (num-of-users-1) is the PPERSISTANCE value. PPERSIST is 256 divided by the number of colliders minus 1.

    The answer for best case channel performance, if there are 10 users on the channel, is (0.18*400)/9 = 8 bits per second per person, i.e. 1-character per second.

    Ok. that's best case.

    The worst case is that some or all of the people don't pay attention to SLOTTIME or PPERSIST and go wild. The throughput at that point goes to as near zero as practical to guess because everybody will get disconnected over and over.

    Does this seem familiar?
    It's worse if there are more than one digipeater on the frequency.

    Ok. So the crappy packet networks were crappy and people eventually moved over to Internet or gave up.

    What does this mean for packet? It means there is an opportunity. We get to build networks which are good and which are fun to play with.
    But why bother?
    It's educational. Most of the world uses packet messaging for communications. People should know about this stuff.
    It's social. Build a good network and you'll have more time to chat with other hams in the area and, eventually, the region.
    It's attractive to younger hams. Now why would we want younger hams?
    It can be done entirely on our own networks, built using hobby radio. This means we have a chance if the Internet is shut down.
    It is a whole lot cheaper than it used to be. A TNC is $40 and a computer is $35? No kidding. And we have access to 25watt FM rigs on 2m and UHF for $50 or $60 each. Really.

    I have spent most of my hobby time for 6 years creating a formula for building your own "good" network. It's been tested by firemen, lawyers, English professors, carpenters, machinists, teenagers, old-farts like me, and, god help us, ENGINEERS!

    We're not building packet, we're using packet networks to build a fun lively social network.

    We use G8BPQ software on a Raspberry PI as the root of each station. We have our own software to provide a pretty and easy to use User Interface. We even have our own TNC. It's a a $40 USB KISS 1200 and 9600 baud microprocessor powered TNC with both AX.25 and it's own IL2P protocol. IL2P has forward error correction and is still as efficient as AX.25.

    The expected throughput between any two 1200 baud neighboring stations using junky radios is 25 characters per second continuously, even under load. A station can have several of those links fully saturated at the same time. With 9600 baud we get 250 characters per second continuously. We don't do collisions. Every station has a BBS. Every station has a local CHAT engine. Networking CHAT engines has been amazingly successful and results in a fun to use, very social, real-time chat environment spread over dozens of stations at the same time. It's busy and interesting.
    Here's a video of the chat system over New Year's eve 2019/2020 in the NCPACKET ham radio social network.
    http://ncpacket.net/tarpn-home-chat-2019-12-31-at-11.23pm.mp4
    The video is a 100 megabytes so give it 10 minutes to load or so. Then you can mouse it around and find the busy parts. It starts at about 11:30pm and there is a time-stamp on every message typed and a clock down at the bottom of the display.

    You are invited to use packet radio to build a fun project. The other 230 people in our email reflector will also help.
    http://tarpn.net

    No Internet required. No Internet desired.
    Off The Grid Social Network.

    Tadd - KA2DEW - Raleigh NC
     
    NQ1B and W4KYR like this.
  5. K4DBZ

    K4DBZ Ham Member QRZ Page

    I learned about this discussion and poll on a packet radio network, so... I think no, it is not dead.
     
    KA2DEW likes this.
  6. KN4ORB

    KN4ORB Ham Member QRZ Page

    Totally dead. ;) I'm QSY back to packet radio now. 73!
     
    KA2DEW likes this.
  7. KG7CSS

    KG7CSS Ham Member QRZ Page

    Except for APRS should we be moving on the Ham Broadband?
     
  8. KA2DEW

    KA2DEW Subscriber QRZ Page

    Ham Broadband works much better in dry land with big mountains. The advantage packet has is how trivial it is to understand, how imprecise the antennas can be and still function, and how well it works in wet lands with no line of site. 25w 440 rigs and 2m rigs have very different performance than 2.4ghz 70mW transceivers.
     
    KN4ORB and KX4O like this.
  9. KA2DEW

    KA2DEW Subscriber QRZ Page

    The TARPN group has produced a TNC for their own purposes. They buy the parts for around $25 per unit and use them in the local packet network. It does 1200, 2400, 4800, 9600 in AX.25 and a home-grown protocol called Improved Layer 2 Protocol. IL2P adds forward error correction. This is handy for making the point to point network links TARPN uses.
    The group has decided to sell the board and CPU via ETSY for cost at about $8+shipping. The TARPN web page at http://tarpn.net/d has links to an assembly page which gives you a ready-for-upload CSV file with the rest of the parts to upload to Digikey and buy the rest for under $20. Save shipping by buying more, like the network groups do.

    The group has already shipped over 300 boards to not-TARPN buyers. This costs a barely more than an audio card, does pretty fair at packet reception, and is trivial to configure for.

    http://tarpn.net/d
    We also have the best kit building instructions. Check out the A3 assembly page.
    [​IMG]
     
    W4KYR, KN4ORB and KC9ONY like this.
  10. KC2SIZ

    KC2SIZ Ham Member QRZ Page

    What I really miss is the VHF packet bulletin board. There is one in NJ but it's just far enough away that I can't count on a reliable connection.
     

Share This Page