September 2005

From the Editor: Mesh Networking Matters

Continued from page 1

By Jason Pontin

What does all this mean? A few, early applications of mesh networks are already emerging. Meshes will allow municipalities to create cheap or free urban Wi-Fi networks (we will be writing about Philadelphia's effort in our November issue). Meshes have obvious advantages for military and security personnel who want networks that are unbreakable and "horizontal" (see "Instant Networks," June 2005).

Environmental scientists like meshes because they can provide continuous data from large geographical areas over many years (see "Casting the Wireless Sensor Net," July/August 2003). But the most important application of meshes will be in what technologists once called "pervasive computing": embedding sensors and processors in things like clothes, electronics, and buildings and connecting them into smart networks.

Mesh networks will be big business. There are billions of networked devices and embedded processors in the world; many more will be built. The best way to connect all of them will be through mesh networks. But the most disruptive business impact of meshes will be this: telecommunications companies do not own them. Meshes profoundly diminish the organizations that own and manage communications backbones.

But I believe that the most intriguing aspect of mesh networks is their cybernetic qualities. That is, mesh networks are adaptive systems that resemble biological systems (we recently wrote about MIT mathematics professor Norbert Wiener, the founder of cybernetics: see "Cybernought," June 2005). Many meshies like to say that they draw their inspiration from the behavior of swarming bees or ants.

Some go even further. In "AntHocNet: An Adaptive Nature-Inspired Algorithm for Routing in Mobile Ad-Hoc Networks," published this year by the Dalle Molle Institute for Artificial Intelligence in Manno, Switzerland, Gianni di Caro and colleagues describe how ants from the same colony will converge to discover the shortest path from their nest to food; he proposes an algorithm for routing on mesh networks that explicitly imitates ant behavior. Ant colonies suggest how apparently intelligent behavior can emerge from a few fairly simple rules. Maybe mesh networks will promote new technologies that possess some of the properties of emergent intelligence? Write and tell me at jason.pontin@technologyreview.com.