By Jim Geier
If your wireless LAN applications require high performance, then you’re probably facing a decision on whether to use 802.11a or wait for 802.11g. Before making the choice, you need to fully understand what both of these standards have to offer. Let’s compare and contrast these two competing technologies and then see which one best fits your needs.
The 802.11g standard is still under development, with a final standard likely available by the end of 2002. With pre-standard chipsets just becoming available now, product vendors will probably release 802.11g radio cards and access points in late 2002 or early 2003.
802.11g is an extension to 802.11b, the basis of the majority of wireless LANs in existence today. 802.11g will broaden 802.11b’s data rates to 54 Mbps within the 2.4 GHz band using OFDM (orthogonal frequency division multiplexing) technology. Because of backward compatibility, an 802.11b radio card will interface directly with an 802.11g access point (and vice versa) at 11 Mbps or lower depending on range. You should be able to upgrade the newer 802.11b access points to be 802.11g compliant via relatively easy firmware upgrades.
Range at 54 Mbps will likely be less than existing 802.11b access points operating at 11 Mbps. As a result, don’t count on upgrading your existing access points that currently provide 11 Mbps throughout all areas. You’ll probably need to move the access points closer together and include additional ones to accommodate higher data rates.
Similar to 802.11b, 802.11g operates in the 2.4GHz band, and the transmitted signal uses approximately 30MHz, which is one third of the band. This limits the number of non-overlapping 802.11g access points to three, which is the same as 802.11b. This means that you’ll have the same difficulty with 802.11g channel assignment as you do with 802.11b when covering a large area where there is a high density of users. The solution of course is to lower the power of each access point, which enables you to place access points closer.
A big issue with 802.11g, which also applies to 802.11b, is considerable RF interference from other 2.4 GHz devices, such as the newer cordless phones. Companies often complain about limited wireless LAN performance when people in the facility operate cordless telephones. It is possible to manage the problem by limiting sources of RF interference; however, you can’t always eliminate the problem.
The 802.11a standard and FCC spectrum regulatory status is firmly in place. Chipsets have been available for nearly a year, and several product vendors are now shipping 802.11a access points and radio NICs. This places 802.11a ahead of 802.11g in the market by about six months.
A big difference with 802.11a is that it operates in the 5GHz frequency band with twelve separate non-overlapping channels. As a result, you can have up to twelve access points set to different channels in the same area without them interfering with each other. This makes access point channel assignment much easier and significantly increases the throughput the wireless LAN can deliver within a given area. In addition, RF interference is much less likely because of the less-crowded 5 GHz band.
Similar to 802.11g, 802.11a delivers up to 54 Mbps, with extensions to even higher data rates possible by combining channels. Due to higher frequency, however, range (around 80 feet) is somewhat less than lower frequency systems (i.e., 802.11b and 802.11g). This increases the cost of the overall system because it requires a greater number of access points, but the shorter range enables a much greater capacity in smaller areas via a higher degree of channel reuse.
A huge problem with 802.11a is that it’s not directly compatible with 802.11b or 802.11g networks. In other words, a user equipped with an 802.11b or 802.11g radio card will not be able to interface directly to an 802.11a access point. In applications where you have little or not control over what radio NICs, you’ll run into interoperability issues. The cure for this will come eventually, however, as multimode NICs become the norm.
So, which one should you use: 802.11a or 802.11g? Both of them operate at 54 Mbps using OFDM. 802.11a provides greater total capacity and is less likely to encounter RF interference, and it’s relatively easy and cost effective. The multimode 802.11 NICs will eventually help eliminate the interoperability issues, though, at about the time you will be able to purchase fully-compliant 802.11g access points.
Choose 802.11a if you’re implementing a wireless LAN from scratch and need to install it today. You can purchase 802.11a products now, but 802.11g products won’t be available for another six months or so. Consider waiting for 802.11g products if you have a relatively large installed base of 802.11b. You’ll be able to upgrade to 802.11g rather easily.
If you don’t have high performance needs today but might in the future, consider deploying dual-slot access points. That will enable you to install 802.11b access points now, and you can add an 802.11a NIC later. Of course your users will need to have 802.11a/b NICs to ensure interoperability.
Implement 802.11a now within existing 802.11b networks if you have specific areas needing high performance (e.g., conference and computer rooms). As with the point above, however, end-users will need 802.11a/b NICs to ensure interoperability.
Keep in mind that you should base the decision for implementing either 802.11a or 802.11g on existing or future needs for higher performance. Many applications can get by with what 802.11b offers, so be sure that you need the extra bandwidth. If a year or two from now you need higher performance, then you can easily upgrade to 802.11g or plug-in 802.11a where needed.
Jim Geier provides independent consulting services to companies developing and deploying wireless network solutions. He is the author of the book, Wireless LANs (2nd Edition), and regularly instructs workshops on wireless LANs.