Wireless Networks

How do wireless networks work?

Wireless data is predominately transferred over two kinds of networks: wide area networks (WANs) and local area networks (LANs). These networks are similar to their wired counterparts, but they just use radio waves instead of copper or fiber.

WANs can cover areas as large as several countries. AT&T Wireless, Cingular Wireless, Sprint and Verizon and are among the carriers that use wireless WANs.

Wireless LANs (WLANs), already popular in airports, coffee shops and hotels, are often used to replace or enhance wired LANs. WLANs can cover 1.25 miles indoors and up to 4.35 miles outdoors in extreme cases, but work best in the 500-foot range. They may service a smaller area than their WAN cousins, but LANs can transfer data much faster, with speeds of 54Mbps now possible. Many companies are switching to WLANs for voice over IP.

More on LANs

Wireless LANs all use some standard in the 802.11 family to communicate. There's an 802.11 standard for just about every letter in the alphabet. You don't really need to worry about most of them - they are security and clarity standards that mean only that wireless LANs are getting even safer and easier to use. There are only three standards you really need to worry about:

802.11b—This protocol, which uses the 2.4GHz spectrum, dominated the market first, earning it the nickname Wi-Fi. It's a busy spectrum, full of baby monitors, cordless phones and microwave ovens, and the more traffic builds the greater the chance for interference and drops in speed. Today, service is good and speeds remain fast. An 802.11b user can get connection rates of up to 11Mbps from up to three different information sources at a time - for maximum throughput availability of 33Mbps - before interference becomes a problem.

802.11a—A competing wireless LAN standard that operates in the 5GHz spectrum allows for speeds of up to 54Mbps from up to eight different access points at a time, giving it a greater range and max speeds up to 13 times faster. However, in order to use 802.11a you need all of those access points. The primary reason to choose one is a need for lots of bandwidth, for example, for transmitting voice or video over your network. Since b is ubiquitous - and a and b aren't compatible - 802.11a equipment isn't good for portability.

802.11g—This is the next generation of 802.11b, promising the price and range of b (it operates on the same 2.4GHz spectrum) with the speed of a. It is compatible with existing b infrastructure and is generally believed to be the heir to most LAN connections.

More on WANs

The first-generation wireless WAN was analog voice (the earliest cell phones). The second generation was digital - more efficient cell phones that could move data at rates of 9.6Kbps to 14.4Kbps. Most U.S. carriers are now at 2.5 generation, or 2.5G transmission rates, which will carry data up to 114Kbps, but will most likely perform similar to a dial-up modem. Carriers have been pushing to 3G, which will include theoretical transmission rates of up to 2Mbps to 5Mbps, advanced roaming capabilities, as well as the sought after (by some) "always-on" potential.

Like LANs, WANs have many acronyms, mainly because carriers use different standards. Each works, but they don't always work with each other.

What have been the traditional hurdles to wireless adoption?

Real wireless projects depend on three elements: the device, the network (i.e. the WAN) and the application. If one of those elements isn't up to par, then the project won't work. No one uses cumbersome devices; people give up if they can't connect to the network, and there's no point in doing a project if you can't deliver the data. Until recently, most devices had small screens that made it hard to view data, they ran out of batteries quickly-sometimes wiping out all the information in the process-and they were expensive. Networks, meanwhile, were proprietary and expensive.

Those that succeeded fell into predictable categories. They were the companies that had large mobile workforces, depended on data from those workforces and, most importantly, could afford to invest in custom devices, proprietary coverage plans and homegrown applications. UPS is a great example. More common examples were trucking companies that tracked their drivers with GPS devices, shipping companies offering delivery confirmation, and utility companies whose repair crews collected large amounts of data about problems and fixes in the field.

Is the landscape any different today?

It is changing. Both devices and network technologies have improved by leaps and bounds in the past year or so. Devices now have color screens, more memory and faster processors, which means that people can actually use them. Also, 2.5G networks now support IP packets, meaning data can pass over existing voice networks. The result is better, more reliable coverage.

Prices for both devices and network time are dropping between 15 percent and 20 percent a year. There is still work to be done on the application side, but many vendors, such as Microsoft, Oracle and SAP are building wireless functionality into new versions of their software.

What it all boils down to is that wireless technology is now available off the shelf, and even companies with long-standing investments in custom-developed wireless systems are turning to commercial products and services.

What are the limitations of wireless?

When it comes to WANs, bandwidth is still limited. When transmitting data, users must sometimes send smaller bits of data so the information moves more quickly. The size of the device that's accessing the information is also still an issue. Even the most recent phones and PDAs have small screens - often only a couple of inches in diameter - and it is hard to read large documents on them.

Many applications need to be reconfigured if they are going to be used through wireless connections. Most client/server applications rely on a persistent connection, which is not the case with wireless. Transactional systems require safeguards for dropped wireless connections. Remedies for all of these shortcomings cost money.

Do I really need wireless?

Just because your company can go wireless doesn't mean it should—not every company needs wireless.

Critical, time-sensitive applications are the best candidates for WAN projects. If getting information in real-time makes or breaks a sale, give your salespeople access to that data. But remember that WANs are best suited for accessing small pieces of information because of bandwidth constraints.

Wireless LANs are often installed for convenience, such as when an enterprise doesn't want to wire the building, or when an IT staff is dispersed throughout the building. They are often used in hospitals, where doctors and clinicians can check in while on rounds or on the floor. Wireless LANs are faster and more reliable than WANs.

What about ROI?

Pinning down return on investment for wireless projects is hard. ROI can more easily be figured if a wireless project raises revenue or decreases cost, but most wireless projects improve productivity and lower stress, and that is hard to quantify. For example, a businessperson using a BlackBerry device can download e-mail throughout the day rather than spend an hour at night in his hotel room. That extra hour could be used to phone home or make more sales calls.

Wireless gives people who spend most of their time in the field vital information that can create efficiencies, and save time and money, but office workers can also benefit. An executive with real-time access to e-mail from a mobile device can get to the next meeting and still get that important message instead of waiting at a desk.

What about security?

Security is one of the biggest barriers to any kind of wireless initiative—whether it's implementing a WLAN in a remote office or rolling out wireless handhelds with application capabilities such as e-mail or CRM data. The mobile security vendors are a force to be reckoned with in this space. The key for IT executives is to cut through the FUD and determine the real risks to a wireless initiative and the intended benefits. Then the decision becomes a simple risk management, cost-benefit equation.