Like any entrepreneur, I have more business ideas than time to pursue them. One idea that I strongly considered before committing to Integrated Roadways was building a mesh cellphone. I actually studied this exact topic at MS&T Rolla for the NSF in 2009, “Voice over multi-hop wireless mesh networks” with Dr. Maciej Zawodniok. I’m intrigued that my concept recurs nearly a decade later on a popular show.
A primary obstacle is routing. The more hops you have from point to point, the more latency in the system, and the more system capacity is taken up by routing, trying to figure out the most efficient means to hop your message to its destination. Admittedly my day-to-day familiarity with routing protocols in mesh networking is nearly a decade out of date, but I have kept a casual watch on the topic and haven’t noticed any revolutionary new routing methods emerge.
I have some ideas myself about how to address routing using a combination of physical position (geo address) and internet position (IP address), along with a netmap that correlates geo position to internet position, but I’ve been too busy with IR to really advance this in any real way.
Eventually I realized we don’t need a fully wireless mesh network, we only need mesh at the edge. If the “first-mile/last-mile” were mesh, it would eliminate the perception of a natural monopoly on internet access by expanding the radius of available connections, greatly reducing the ability to monopolize the connection at one’s physical location. Instead of only being able to buy internet access from whoever has the wire at your home or business, you could buy internet access from anyone within a radius of your home or business, without the cost to drag a wire.
Mesh is brilliant for local networking, but not a great way to send an email from Iowa to China. The eventual future state will reveal a combination of mesh and traditional access options, as no method fully excludes or replaces each other with improved performance in all areas.