4 ms·
Delay-Tolerant Networks Might Be the Killer App Ham Radio Needs
- phaedrus 9y agoI looked into this in 2007 - 2008, when I was in college. I was doing a NASA internship, working on an engineering study of IP communication over deep space links. However my directive concerned IP protocol specifically. DTN was an interesting case study to me, but we weren't trying to use the DTN protocol so I never tried to actually set it up. I thought it was a neat idea, but I hadn't kept up with developments in that field since then.
- calgoo 9y agoI think of this as well, especially when looking towards the future with Mars colonies etc. However, I see it as a mix of data dumps of the internet as we know it and some sort of email system as well.
- jvanderbot 9y agoI actually think IoT is a great application, as devices become producers, but also relays for messaging. The use case for this functionality is orbiter relay, but there's nothing to prevent any moving agent to be a "router" in DTN, no matter how long it's offline.
- idoescompooters 9y agoIt's definitely the key to the networking link to Mars. Whether it will involve lasers or RF, I don't know, but an IP system won't work.
- zeckalpha 9y agoCuriously, the “relay” mentioned in ARRL, is a form of a DTN. They were doing this a hundred years ago, just manually!
- jvanderbot 9y agoI work at JPL and have begun using the ION implementation of DTN as part of a distributed computing research project. The development of this stack is very active, and momentum is pretty solid internally. It wouldn't surprise me to see more use of DTN in the near future. The lead devs are active and responsive despite being promoted away from front line work most of the time. Also, this is TRL 8/9, so ION / DTN is pretty stable.
- Steltek 9y agoMirror link please? (Oddly Google on Firefox Mobile doesn't appear to offer cache links) And IMO, the shot of adrenaline that ham radio needs is for the FCC to take the gloves off of data modes on HF. 300 baud, are you kidding me? The same RF bandwidth that analog voice gets to use could deliver magnitudes more as a digital signal.
- vvanders 9y agoYup, floored me when I came across that on the exam. Tons of other awesome stuff in countries that do a proper bandwidth limit as opposed to symbol limit.
- brian-armstrong 9y agoI thought this wasn't so much the FCC's doing as it was older hams who shun anything digital
- vvanders 9y agoI think that tide is turning, esp given the recent explosion of FT8.
- kb1lqc 9y agoTry again, website is back up. May still be slow.
- VLM 9y agoIts a weird case of legislating good engineering into political law. Its not needed; should be able to just quote FCC part 97.101(a) "In all respects not specifically covered by FCC Rules each amateur station must be operated in accordance with good engineering and good amateur practice" Google up "digital transmission on fading multipath channels" or similar and for example there's a nice whitepaper from National http://www.ni.com/white-paper/14931/en/ http://www.ni.com/white-paper/14931/en/ The ionosphere makes massive incredible multipath distortion causing intersymbol interference making anything over 100 or so symbols per second basically broadband noise aka jamming. Even if by some miracle a (short NVIS?) hop were decodable between hams X and Y, the ionosphere being what it is means everyone else on the band would be hearing distorted broadband noise. Most of the work recently (like last 30 years) has been in very low baud rate systems. From memory, doesn't FT8 run at 5.86 baud? It uses multi channel fsk and some FEC weirdness to push a remarkable number of bits per symbol. Like a decade ago I used to talk to people using Olivia mode which runs at 31.25 baud which frankly was too fast for conditions. Olivia also encoded quite a few bits per symbol. Generally symbol rates (baud rates) have been dropping while bits/sec has been increasing for a very long time. In the really old days like RTTY the baud rate was the bits/sec rate so the ratio of bits/baud was 1, it hasn't been that low in a long time. It was very frustrating and weird (an annoying!) to watch a waterfall display as multipath fading drifted thru a band and destroyed just the mark or just the space carrier of a RTTY signal. 45 baud rtty is probably the highest baud rate you'll see on HF digital ham radio.
- macintux 9y agoTried to comment on the page before it died. There's at least one vaguely plausible chain of events that could lead to DTN being ubiquitous: if/as vehicles start communicating with each other as we approach full autonomy, incorporating DTN into those communications would seem a possibility.
- kb1lqc 9y agoWoops, try again. Network issues this morning with the website. (I am one of the co-founders of FaradayRF ;)
- Animats 9y agoHe's re-inventing UUCP.[1] [1] https://en.wikipedia.org/wiki/UUCP https://en.wikipedia.org/wiki/UUCP
- vvanders 9y agoYeah, yeah everything old is new again. The one thing I like about faradayrf is they're at least trying. Right now the state of packet radio is either 1200bps bell tones with no FEC or modern modulation and on the other end something like FT8 which is incredible but has almost zero bandwidth and good for only communicating "I heard you". I'd love to see a more modern packet infrastructure that isn't stuck in the 80s.
- kb1lqc 9y agoSpot on ;) In fact this post was meant to start the discussion you are alluding to. Let's not reinvent the wheel as so much as take what we have available to us now and use it combined with some new technology to propel ham radio into modern communications.
- vvanders 9y agoYeah, I would love to see some modern N-QAM, FEC spec make it's way into the APRS space. The number of times I see 144.390 key up at S8+ and no packet has to be at least 50% of the time. Some of the stuff that happens in DMR space is pretty cool, even simple TDMA opens up so many possibilities. You could do Voice+APRS at the same time over the modern repeater network. For instance we've got BawFaw that covers most of SW Washington, would be awesome to be able to run something like that for packet radio.
- tlrobinson 9y agoI wonder if there's opportunity for collaboration with Secure Scuttlebutt, which is essentially a DTN: https://www.scuttlebutt.nz/ https://www.scuttlebutt.nz/