Unless you have been living in a shoebox or still work in one of those computing centers with controlled environments and floating floors, you probably know that 802.11 wireless is spreading like wildfire throughout the enterprise market. Shipments of 802.11b and the faster 802.11g networks increased by about 40% last year, and will most likely hit the 95 million mark by 2009, according to some industry number crunchers.
One of the more popular standards right now is 802.11g, since it is theoretically about five times faster than 802.11b systems and incorporates more of the latest security protocols. In fact, prices of 802.11g systems have plummeted recently, reaching a point where they are just as affordable as 802.11b systems were six months or so ago. Acceptance of 802.11g systems is so strong that sales have impacted the saturation of 802.11a technology, which is every bit as fast but offers a much broader bandwidth and more security and reliability since this technology operates on the 5.4 GHz spectrum instead of the 2.4 GHz frequencies populated by 802.11b, 802.11g, and assorted microwave ovens and cordless telephones (which creates some degree of collision and interference).
We use both 802.11b and 802.11g systems here at Shoreline Research, basically because we do not yet have a strong need for the increased security levels offered by 802.11a and cannot in good conscience rationalize the increased cost for these systems. Also, 802.11a technology is essentially incompatible with 802.11b/g (since it transmits and receives on a different radio spectrum), so in order to use it to its full extent we would have to throw our current wireless baby out with the bath water.
We have seen 802.11a systems in action, and are quite aware of how well thee systems can handle multimedia-rich files and multiple streams of complicated content (at least in the demonstrations we have witnessed by companies like Atheros Communications Inc.). We are also very aware that enterprises continue to adopt 802.11a technology (along with 802.11b/g and some evolving Wi-Fi flavors), and that multi-based 802.11 b/a/g chipsets are fast becoming all the rage in corporate applications.
So, why are we still more gung-ho on 802.11g as a primary alternative than 802.11a technology? The main reason, as we stated earlier, is the compatibility issue. Since speed is a major factor why most people upgrade their current wireless technologies, there is no compelling reason not to opt for 802.11g or wait for faster compatible versions to arrive within the next year or two. But, there are other reasons why 802.11a may not be the best option, event though the technology entered the scene a few years ago with great promise and was essentially the valedictorian of the emerging wireless class.
Some of these reasons concern the technology's basic reliability, scalability and performance. We have heard a number of complaints from enterprise users who were dissatisfied about the performance of 802.11a, as compared with current 802.11g and even 802.11b systems. Most of these concerns have focused on difficulties in effectively mapping out and deploying 802.11a wireless access points (APs), and the technologies inability to dance around physical obstructions in an office as quickly and adroitly as god old 802.11b/g technology.
Don't take just our word for it, though. Many major companies have performed independent studies of 802.11a and 802.11b/g technologies that highlight the apparently inherent problems of 802.11a in enterprise settings. One of these studies was conducted by a little company called Intel Corp., which released the findings of the research in January 2002. The study consisted of some 2100 measurements taken over a period of three months in both residential and office environments, and involving single and multi-floor wireless installations. The findings:
- Wireless operation in the 5 GHz band presents a number of unique challenges for radio design because of multi-path and additional path loss
- The propagation losses created by normal obstructions -- like walls, flooring and elevator shafts -- are measurably greater for 5 GHz devices than for 2.4 GHz systems. 802.11a systems are also more susceptible to interference from blockage losses created by people walking through line of sight (LOS) pathways (since people are made up of about 75% water and water tends to block and distort signals)
- Since 802.11a systems are more susceptible to physical obstructions, and the performance of these systems of these systems is subsequently impaired, wireless networks operating in the 5 GHz range may also suffer from a reduced transmission window, meaning there may be some question of a reliable and full signal getting through to its intended destination
- Because of its inherent limitations, the potential speed and reduced signal interference gains offered by 802.11a systems may be significantly offset by its limitations when it comes to path loss caused by physical obstacles
Now, before you go running down to purchasing to pull that order for 802.11a systems, remember that this is but one study on the matter. Also, there are ways to avoid reliability and performance issues if you do your homework and adequately map out your wireless AP locations within an organization. Also, there are some very fine tools and add-on technologies that provide strong management and security functions, and can be used to pinpoint weak spots in a wireless network and automatically hand users off to stronger wireless APs to avoid signal loss and distortion.
Our main point is that while speed and limited interference are important factors in selecting a wireless system, these are not the only criteria you should consider -- especially since every wireless network is unique and different. Our advice is to examine your needs very carefully, and then implement a system that performs the best in your particular environment. You may also note that many companies, including 802.11a pioneer Atheros, are investing more time and effort into hybrid 802.11g systems, which offer multi-network compatibility, increased management capabilities and improved reliability. These systems offer the most benefits for companies looking to have the least wireless headaches today and suffer little pain moving into the future.
Tim Scannell is the president and chief analyst with Shoreline Research, a Quincy, Mass.-based consulting company specializing in mobile and wireless technology and initiatives. Shoreline works with end users, looking to implement mobile solutions, and vendors, developing new products and seeking business and customer opportunities. The company also specializes in training and strategic planning projects. For more information on Shoreline Research and the company's strategic services please go to http://www.shorelineresearch.com.