Even more on meshes

By C.J. Mathias, Farpoint Group |  Mobile & Wireless

Regular readers of this column know how bullish I am on mesh-based wireless networks. I still consider meshes to be the most important architectural trend in wireless today, and for one simple reason: no other approach can match the flexibility and scalability of meshes, whether we're talking about the micromeshes used in sensing and telemetry applications, or the metro-scale meshes based on Wi-Fi now becoming fixtures in many cities around the world.

The ubiquity of metro-scale Wi-Fi meshes hinges, as is always the case in high tech, on performance. Performance, when applied to wireless, implies not just throughput, but also range and reliability. More throughput is always desirable, of course, and it's usually a function of available capacity in wireless networks. Time-bounded performance, the ability to expedite high-priority traffic such as voice (or any data that is consumed in real time by carbon-based units like you and me), is also of increasing interest, and is perhaps an obvious challenge in the case of meshes where traffic is by default routed among infrastructure nodes. Range is more problematic, given the use of the unlicensed bands and their limitations on transmit power. Similarly, the unlicensed bands (often called the free-for-all bands) are inherently prone to interference.

All of these issues conspire to create one sticky metro-scale challenge. But meshes are uniquely suited to this challenge, and are performing admirably in many venues today. The key to performance is in the protocols used to implement meshes. And this is the subject of our latest white paper, "Metro-Scale Wireless Meshes: Key Issues and Requirements." We've covered a lot of territory in this publication, and I'd be most interested in your comments.

Regardless, metro-scale meshes are going to be one of the great success stories of the wireless era, and one that we will eventually take for granted. It's hard to imagine life without Wi-Fi, and that's OK, because, thanks to meshes, it's going to be almost everywhere over the next few years.

Join us:
Facebook

Twitter

Pinterest

Tumblr

LinkedIn

Google+

Answers - Powered by ITworld

ITworld Answers helps you solve problems and share expertise. Ask a question or take a crack at answering the new questions below.

Join us:
Facebook

Twitter

Pinterest

Tumblr

LinkedIn

Google+

Ask a Question
randomness