Feature
News
8/22/2005
12:50 PM
Dave Molta
Dave Molta
Features
Connect Directly
RSS
E-Mail
50%
50%

Primer: Making A Mesh Of Your WLAN

Wireless mesh isn't just for city-wide deployments. It's an increasingly popular option when wired connections aren't feasible. Here are the basics.

   

Outdoor mesh gets the most attention and arguably offers the most compelling value proposition, but indoor and hybrid indoor/outdoor meshes also are useful, especially in hard-to-wire environments. Today's wireless mesh systems are proprietary, but that's not necessarily a reason to postpone deploying one. After all, most large organizations try to standardize on their critical infrastructure elements to minimize support costs, so multivendor interoperability isn't critical as long as each system supports open standards like Ethernet and Wi-Fi.

Unfortunately, mesh capabilities aren't integrated with modern enterprise Wi-Fi systems, so if you want to extend your existing WLAN with mesh backhaul connections, you must integrate it with a mesh vendor's technology.


Conventional WLAN With Wired Distribution System
Click to Enlarge

The Basics

Most WLAN APs are deployed in a modified star topology, as part of a broader wired Ethernet system (see "Conventional WLAN With Wired Distribution System" left). Although clients can roam from one access point to the next, traffic to and from clients at the edge of the wireless network follows a static path from client to AP to Ethernet switch and into the backbone network.


Municipal Network Using mesh Architecture
Click to Enlarge

In a mesh architecture, access traffic at the edge of the network is dynamically routed across a wireless mesh that interconnects the APs and possibly other Ethernet devices. "Municipal Network Using Mesh Architecture" (right), shows a small municipal wireless network that uses wireless links between APs rather than Ethernet cabling to move traffic from wireless clients to the backbone. Client traffic at the edge that's destined for systems on the wired network can take alternate paths through the mesh. Mesh nodes optimize those paths.

This wireless mesh architecture is similar to the wired mesh used on the Internet, where routers make forwarding decisions using dynamic routing protocols. In both cases, the specific path that packets take through intermediate nodes is transparent to clients. Many factors, including traffic levels, link capacity, routing-protocol efficiency and overhead, can impact overall performance. Networks with small diameters (small hop counts) generally will have better throughput and latency characteristics than those with large diameters, which experience a performance hit for every intermediate hop. To overcome this, you'll probably want fast and dedicated mesh backhaul connections.

Design Options

Although some mesh systems limit themselves to providing wireless backbone services, most provide a combination of backbone/infrastructure and client-access services. Thus, a Wi-Fi client can connect to a node that's simultaneously acting as an infrastructure device for the mesh backbone. In these systems, the mesh node must handle standard Wi-Fi access (usually 802.11b/g but sometimes 802.11a as well), ingress traffic from other mesh nodes, egress traffic to other mesh nodes and, in some cases, an Ethernet connection to the wired network.

The simplest wireless mesh design uses a single-radio for access, ingress and egress. Its distinguishing characteristics are low cost and simplicity. In a single-radio design, both client access and communication between mesh nodes take place over one radio, which dynamically switches its function from AP-mode to mesh-node (see "Solo Radio" left).

This system, which typically uses 2.4-GHz 802.11b/g radios, is the least expensive system to deploy, but offers limited performance and capacity. That's because the single radio in each mesh node must time-slice between client access, ingress and egress. To overcome these limitations, you must design your network to minimize hop counts. So about one-third to one-half of all mesh nodes also should have connections into the wired network, directly over Ethernet or over a dedicated point-to-point, or point-to-multipoint, fixed wireless backhaul system. Large, single-radio mesh networks are typically deployed in conjunction with 5-GHz multipoint wireless backhaul systems from Alvarion or Motorola. That can mean higher costs as well as network-management complications.

More sophisticated wireless mesh networks use a multiradio design, separating the access, ingress and egress functions. In a two-radio design (see "The Power of Two" right), client access traffic takes place on one radio channel (usually 2.4-GHz 802.11b/g) while mesh ingress/egress traffic uses a different channel (usually 5-GHz 802.11a).


The Power Of Two
Click to Enlarge

By separating access and mesh functions, the dual-radio design offers some performance and design benefits. You get an increase in the overall system capacity and more flexible allocation of your available RF channels. The trade-off, however, is that dual-radio systems tend to cost more than single-radio systems, because hardware for dual-radio setups is more expensive and they generally rely on 5-GHz radios for mesh communications. These signals are heavily attenuated by buildings and foliage, so you'll need more mesh nodes for this kind of a system than for a single-radio, 2.4-GHz design. And deployment may require more complex radio-link engineering to ensure line of sight between mesh nodes.

The last and arguably most sophisticated mesh network design uses three or more radios per node. These additional radios can be used for one of two purposes. First, you can create a multisector and/or multichannel access system using directional antennas to extend range and provide more client access capacity. Second, you can optimize mesh traffic by separating ingress and egress traffic onto different radios/channels. This multiradio approach offers excellent performance, but is more costly and complex to install.

Keep in mind that adding lots of radios to a mesh node doesn't always guarantee superior performance. That's because other factors, including radio efficiency, routing protocols and mesh diameter, also contribute to performance. As a general rule, however, more radios translates to better performance and capacity--albeit at a higher cost.

Previous
1 of 3
Next
Comment  | 
Print  | 
More Insights
The Business of Going Digital
The Business of Going Digital
Digital business isn't about changing code; it's about changing what legacy sales, distribution, customer service, and product groups do in the new digital age. It's about bringing big data analytics, mobile, social, marketing automation, cloud computing, and the app economy together to launch new products and services. We're seeing new titles in this digital revolution, new responsibilities, new business models, and major shifts in technology spending.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest - July 22, 2014
Sophisticated attacks demand real-time risk management and continuous monitoring. Here's how federal agencies are meeting that challenge.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.