[HamWAN PSDR] Mesh backbone

[Bart Kus]

Hi Craig,

First, let's speak the same language.

There's a lot of confusion around the word "MESH".  I'm also not sure 
why it's always capitalized.  It's not an acronym as far as I know.  The 
term "mesh network" describes nothing more than the logical topology of 
a network.  There's a really good write-up on this at the wikipedia page 
<http://en.wikipedia.org/wiki/Mesh_networking>.  I'm pretty sure that 
when you (and others on this email) use the phrase "mesh network" it is 
not the wikipedia definition you're intending.  It means something 
different, including:

1) Using a common RF channel
2) Promiscuous neighbor discovery + association
3) Automatic IP configuration based on MAC
4) Nearly open node authentication
5) Omnidirectional operation
6) WDS <http://en.wikipedia.org/wiki/Wireless_distribution_system>-style 
operation

These are the typical traits of a SeattleWireless 
<http://seattlewireless.net/> style mesh network (which, BTW, has been 
attempting to bootstrap itself for the last 13 years).  This type of 
definition of "mesh network" is quite a different animal from the 
canonical mesh network definition (wikipedia's, derived from network 
theory).

The reason I want to make the distinction clear is that nearly all large 
networks in the world are indeed mesh networks, but nearly none of them 
possess the qualities of 1-6.  So when I say something like "HamWAN 
<https://www.hamwan.org/t/tiki-index.php> is a mesh network", I want it 
to be clear that I'm referring to the network topology only (the 
wikipedia definition).  I'm not sure what is the right word to describe 
the other type of network; perhaps capitalizing all the letters is a 
good differentiator after all.  :)

So, with the definitions out of the way, let me address your actual 
email now that I can speak to it unambiguously.

I concur that we all want to implement a mesh network, but I don't think 
everyone wants to implement a MESH network.  Phrasing the problem of 
"how do we provide modern digital communications to the ham community" 
in terms of "how do we implement a MESH network" is putting the cart 
before the horse.  Over the last 6 months, I've been leading the HamWAN 
effort to create solutions to the first question.  We've made 
respectable progress on both the RF engineering and networking fronts.  
We had a fully functional cell site setup @ last weekend's Flea Market.  
This site design is about to start rolling out to the real world.

This transition in the project's status has allowed me to start thinking 
about how HamWAN might integrate with other ham networking efforts.  
We've had a good relationship with BCWARN.net 
<http://www.bcwarn.net/intermapper/rf-map.html> for the last few months, 
and integration with that network will be simple.  The physical links 
are already planned in fact!  We're both excited to make it happen and 
start exchanging traffic internationally!

The integration with NW-MESH efforts is far more challenging.  In fact, 
it may be outright impossible unless changes are made in the NW-MESH 
design.  From what I've seen, this difficulty of peering will be present 
in all MESH networks.  The problems range from simple route exchange, to 
address space conflicts, to policy propagation (access, QoS, filtering, 
etc).  I don't even wanna think about DNS.  :P

Having said all that, there is some value to HamWAN using a MeSh (hybrid 
of mesh and MESH) layer.  Traffic between nodes may flow more optimally 
on the ground than through the mountain sites.  A nearby ham who doesn't 
want to invest in a dish might get on HamWAN by MeShing with his 
neighbor.  I'd be good for the health of HamWAN to make use of these 
optimizations.  But like I said, in order for these routing decision to 
be made correctly and automatically, NW-MESH designs will need to change.

I'd like to invite you (in fact, all of you) to join the HamWAN weekly 
meeting today @ 7PM.  I'll re-send the connectivity details on the 
mailing list (email: psdr-join at hamwan.org) an hour before the meeting, 
but basically install Mumble <http://mumble.sourceforge.net/> 1.2.4 
<http://mumble.info/snapshot/mumble-1.2.4-rc1-8-gb115a29.msi>+ 
(currently beta), and connect to BartK.us.  Please use a headset to 
avoid generating echoes.

Craig, can you give me an idea of your skills?  Perhaps you would enjoy 
solving these types of problems as part of the HamWAN development team?  
We run a tight ship with specific assignments and weekly reporting.  I 
believe this is the "small group of experts" approach you were 
proposing.  :)

--Bart


On 03/12/2013 12:09 PM, Craig B wrote:
> When I first heard about the MESH project from Daniel Stevens (KL7WM) 
> back in late fall 2012, the first question I had for him was "what 
> will it connect to?"  Since then, as I have become more involved I 
> have started to formulate what I think it could be connected to and 
> how it could be used.
>
> Based on what I have learned and seen to date, I see 3 tiers of 
> network involved here.  The backbone, which I see as a long-haul that 
> would be based on a region that is defined by terrain and distance. 
>  The middle tier would be smaller and could be between HAM towers or 
> other "secondary" sites.  The 3rd tier would be for the "neighborhood" 
> or "home" MESHing with WRT's and other low-power devices.  In this 
> type of configuration, I see the backbone as being the one common 
> piece across regions while the secondary and tertiary tiers could be 
> specific to the 'regional' implementation.  Each tier would have to 
> bridge from itself to the next level, which seems to be reasonable 
> where a given site could choose to bridge by adding necessary hardware 
> or remain remote.
>
> What I would like to do is see if we can't get a written network plan 
> for a regional backbone and then any additional tiers that need to be 
> included in a good network design document.  I am a firm believer that 
> it should be hardware agnostic for the most part, although could 
> provide a list of acceptable components that have been shown or 
> believe to be the best hardware based on application.  It would also 
> dictate how traffic might be handled moving up/down through the tiers, 
> possibly allowing for QoS or other transport methods.
>
> As I am sure we all want to see a MESH network available to all HAMs; 
> given the area NW-MESH has been getting feedback on, I think we need 
> to start looking at how we connect them all together.  As such, in 
> talking with Bob Rutherford, it seems like the first step is to build 
> out a plan that could be presented to FWARC, Tukwila Radio Club, and 
> EMCOMM (and other clubs/groups).
>
> Since this is all great in theory, it seems the next step is to secure 
> funding and move it from paper to reality.  Given the real application 
> of this MESH network for EMCOMM, and their generally deep pockets, it 
> seems like a great way to get a backbone built.
>
> If we are going to design this, I believe it needs to be initially 
> designed by a small group of network and radio experts.  Once an 
> initial plan is cobbled together it could be released to the larger 
> MESH community for comments and additions/subtractions, etc.
>
> I am willing to take on leading this charge; however, I will need a 
> team of experts behind me helping lay the ground work.  My initial 
> thought is to have something ready by mid or late summer, given that 
> we all have other priorities as well, not adverse to taking longer if 
> it means having a complete and well planned out design.
>
> Any thoughts on this?
>
> Thanks,
> Craig
> KF7LLA


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.hamwan.net/pipermail/psdr/attachments/20130312/9df4128a/attachment.html>