A VIDEO

On Saturday April 6, 2013, the first PittMesh node’s hardware was mounted. Along with help from numerous individuals from Hack Pittsburgh and Meta Mesh LLC, we were able to send and receive data from the 5GHz directional antenna across the Monongahela river to a Macbook Pro at street level with no antenna at a speed of 13Mb/s at a range of nearly 1 mile. We also mounted a 12 dBi 2.4GHz omnidirectional antenna servicing a block or more of the Hill District. This is the culmination of 2 years of hobbyist-turned-part-time obsession of mine. In the following weeks and months, we will be redesigning www.MetaMesh.org, refining our networking protocols and designing the PittMesh network along with preparing useful and interesting services that can be reached via the mesh network. The network range can be even greater than initially anticipated and we are now ready to begin looking at new locations and investors in hardware for the next set of nodes. Many thanks to Colin, Pat C and Pat D, Jeff, Gene, Jon, and Chad for their assistance. I look forward to more networking adventures with you in the near future.

A PHOTO

PittMesh node 1 pre mounting! After a long hiatus, the first set of antennae are ready to be mounted atop Hack Pittsburgh! Pictured is the 30 dbi antenna operating on 5 GHz and the 12 dbi omni directional antenna on 2.4 GHz. In a few weeks time we should make the first link to the South Side— crossing the Monongahela river and creating a link about 1 mile away!

A TEXT POST

PittMesh LLC Founded!

As of September 1, 2012, PittMesh LLC has been registered with the state of Pennsylvania as a provider of Computer Systems Design Services. I will be going on a tour de meshes from Sept 8 through the Sept 12 in Philadelphia, New York, and Washington D.C. to get a hands-on understanding of how other mesh networks are being set up. 

[EDIT: I have changed the name of the company to Meta Mesh LLC. It’s cooler sounding. like how bow ties are cool.]

A PHOTO

Announcing PittMesh!

I am happy to finally announce the “official” launch of our little meshnet project here in Pittsburgh. I’m seriously excited about this and want to ake a few points about the goals of this project.

The primary goal at this time is to show that an inexpensive mesh net is feasible from off-the-shelf computer components IN THE REAL WORLD. There seems to be an urge with other mesh net projects around the country to create networks from higher end devices, power boosting hardware, and fancy antennae. I mean no disrespect to these other projects and hope that they can find success with their own methods. However, it is my belief that for a mesh network to become useful, usable, and ubiquitous that the point of entry needs to be low enough so that nearly anyone can obtain the proper parts and jump on the network in a matter of hours. The primary goal is bigger than setting up a meshnet in Pittsburgh but rather to invent a documented process to easily implement a wireless mesh network that is resistant to natural disasters, censorship, and promotes anonymity by allowing for some daily connectivity tasks to be done away from the Internet.

In the coming months I will gather capable individuals from the area and from around the world to work on this project. Initially, we will work on making a hardware and protocol standard for the network and begin setting up nodes, implementing security measures and advertising the use of locally hosted services on the network. I openly and enthusiastically ask for help with this project. I am by no means capable of doing this work alone and am actively looking for SKILLED individuals such as programmers, business analysts, interface designers, project managers, and, of course, network admins to assist in the development of Pittmesh.

While nothing has yet to be done with the new website, please keep an eye out for the launch of www.PittMesh.net in the coming weeks. I will be posting detailed instructions on the set up of my routers this weekend as well as a super cheap outdoor housing prototype I made for about 5$ this weekend.

Thank you for your time.
Now let’s do this!

A PHOTO

IT WORKS.


I have succeeded in navigating the confusion about meshnetworking with DD-WRT and now have a 3-router meshnet running in my house. I have been able to ping between two clients and as you can see in the photo above, RDP from the right screen’s pc to the left’s. Although, to be fair, it timed out. But hey, it’s a start. And you want to know the kicker? It’s really easy to set up. I don’t understand why there is no straightforward description that is readily available.

More details and instructions to come.

-A

A VIDEO

So I found the dream program for doing quick Visibility Analysis to help with the placement of meshnet routers: GeoWeb3d. While it’s a bit confusing to set up (installation is easy but you need to go find elevation maps on the internet yourself), the tool that makes it all worth while is great: the analysis tool. As you can see above, I drew out a few ‘cones’ of vision from my a site. The program takes the elevation data and shows you what areas around the origin are visible to the site itself. You can even set the distance of the ‘cone’ to simulate the distance of your (presumably suped up) router set up.

For me, for example, with the hardware I’ve been developing, the physical width of the usable signal is reduced because of the Yagi antenna. Since the beam is more directional, I can simulate that narrowness with the program or i can widen it to show 360 degree visibility coverage.

GeoWeb3D is free on a trial basis for 15 days. Buying it will require $3,995. Yeah. The program is not so robust that it is even remotely worth that. I’d pay 100$ for it. But still. It’s exactly what I needed. So try it out.

A TEXT POST

ArcGIS and visualization

So I got it in my head to make a large layer model of Pittsburgh to get a sense of what kind of visibility and range is possible when setting up routers. Originally I just wanted to build a big cardboard map in my basement with each layer represeting a 20-40 ft elevation change. This led me down a path using Google Earth and then ArcGIS which my father helped me out with because he’s studying GIS at Pitt.

Long story short: I used the SQL I learned at the University of Pittsburgh this past semester to weed out a lot of unhelpful countour lines from some data I got about elevation levels in Allegheny County. That’s what the previous post’s picture is of. 

Then my ever-helpful father recommended I check out “visibility analysis” with this elevation information. Basically, you choose a point within ArcGIS or another visualization program and tell it to show you what other points on the map are visible from that point. This would be AMAZING because if I could set this up I could instantly visualize visibility of a specific point around town without having to drive there and, say, take a lot of photos or play around with Google Earth’s crap-ass controls.

Suddenly, a cardboard model doesn’t seem like a good use of time.