|Monster Mesh: Decentralized Wireless Broadband|
by Kevin Werbach
Originally published on The Feature (http://www.thefeature.com) - January 01, 2002
Just when you thought the peer-to-peer (P2P) craze was over, a new application for P2P architectures is emerging. So-called meshed wireless systems are poised to attack the seemingly intractable problem of last-mile broadband connectivity.
First, some definitions. Unlike traditional arrangements such as client-server, P2P networks use direct connections between autonomous nodes at the edges. Data flows from one user to another, rather than through a central server. The most famous P2P application is Napster. Instead of transferring files from a master repository down to individual users, Napster allows those users to communicate and share files between each other’s hard drives. Other P2P applications offer decentralized collaboration, distributed content delivery for large media files, and cheap computing power for complex calculations by dividing problems among thousands of inexpensive machines.
The P2P structure makes it easy to exchange music in violation of copyright restrictions, which is why the recording industry forced Napster to shut down through legal action. However, the P2P approach has unrelated benefits. A P2P network doesn’t depend on a central authority to manage connections and store necessary data. It grows with each new node. Those nodes gain from the connection to a large network, but the network gains as well from what each node brings to the collective. There are no single points of failure, and no requirements that a complete core network be built before the users at the edges can connect.
The P2P architecture makes great sense for wireless networks. Direct connections between handheld wireless devices and other local transmitters could replace traditional systems composed of towers and long-distance network backbones.
Wireless P2P Gets Real