Exploiting a problem
As described in Part 1, all 802.11 devices transmit radio waves, using antennas to distribute (propagate) them through the air. A basic dipole antenna spreads out those waves in all directions, just as dropping a pebble into a pool generates ripples. Now picture those ripples hitting the sides of the pool. Some waves actually bounce back from the wall, colliding with other waves still moving in the opposite direction. When waves bump into one another, some ripples get higher while others get smaller or disappear.
A similar thing happens to radio waves that encounter doors, windows and other objects between the transmitter and the receiver. This phenomenon, known as multi-path, causes many reflections of the same transmission to reach the receiver at slightly different times. In the process, some of those reflections will add to, subtract from, or garble one another.
You have probably experienced multi-path on analog TV as "ghosting." Your mind is a very sophisticated input processor -- it can ignore a little ghosting, but a lot of ghosting renders a TV show unwatchable. Similarly, with a high degree of multi-path fading and delay, an 802.11 receiver can have trouble making sense of reflections -- the signal may even be so degraded as to prevent meaningful communication.
When vendors started working on 802.11n to increase WLAN speed and capacity, they used multi-path to their advantage. Specifically, each 2.4- or 5-GHz channel carries only so much information. However, if you split an 802.11 frame into multiple pieces and transmit each of those pieces over a different path, that frame can actually reach the receiver in a shorter time -- and more frames can be sent over a fixed period of time. Of course, the receiver must know how to recombine pieces into the original frame. This technique to improve speed and capacity is called spatial multiplexing.
To tap the power of spatial multiplexing, 802.11n devices must be able to send and receive more than one signal simultaneously; 802.11a/b/g devices lack this capability. Even APs with diversity antennas transmit or receive through only a single antenna at any time. However, all 802.11n devices use Multiple Input Multiple Output (MIMO) antennas to make more of each available channel.
MIMO devices are described by the number of transmit (M) and receive (N) antennas they use simultaneously. For example, a 2x1 MIMO AP transmits through two antennas and receives through one antenna, while a 3x3 MIMO AP transmits and receives through three antennas. Although there are further subtleties at work here, two transmit antennas generally provide less bandwidth than three transmit antennas. MIMO antenna configuration is therefore an important factor when purchasing 802.11n products.
MIMO antennas not only allow different information to be sent along M spatial paths, they can also be used to transmit the same information on M paths. Why would you do this? To improve signal quality at the receiver. If you have an urgent message for a colleague, you probably deliver it in multiple ways -- office voicemail, cell voicemail, email -- because doing so increases the odds of successful delivery. Some 802.11n APs can use MIMO in an analogous way to improve signal strength through redundant transmissions. Better signal strength can increase a WLAN's speed and/or reach.
Thus far, we have seen how multiple antennas and more powerful signal processing work together to improve 802.11n performance. The 802.11n standard also employs many other broadly supported techniques, such as bonding two regular (20 MHz) 802.11a/b/g channels into one double-wide (40 MHz) 802.11n channel to support higher-throughput applications. However, there's another lesser-known option that can affect 802.11n: transmit beamforming.
Without beamforming, each factory-issued 802.11n antenna radiates a signal in all directions, more or less like those omni antennas found on older 802.11a/b/g APs. With beamforming, a transmitter can actually adjust the signal emitted from each individual MIMO antenna to improve reception. This lets an AP aim a given transmission in the direction of a particular receiver, with the result being something akin to that achieved with a directional antenna. Be aware that beamforming requires close coordination between transmitter and receiver and is not included in the 802.11n draft 2.0 snapshot.
Finally, aftermarket omni or directional antennas can still be used with some 802.11n APs to better focus output power on a desired coverage area. For example, you might still connect patch antennas to an 802.11n AP mounted on the back wall of a retail store. Like hearing aids, 802.11n signal processors are better at making sense of what they hear, but a bullhorn (directional antenna) can still boost sound (signal) at the source.
By achieving speeds in excess of 100 Mbps and covering larger "cells" more reliably, 802.11n products illustrate the positive impact of better antenna design and more advanced signal processing.
Nonstandard "Pre-n" WLAN products have been available for quite some time. Last summer, the Wi-Fi Alliance launched an interoperability test and certification program, based on draft 2.0 of the emerging 802.11n standard. Many of the speed- and range-boosting features described here can now be found in certified 802.11n draft 2.0 products. To learn more, read this Wi-Fi Alliance paper or search this list of certified products.
About the author:
Lisa A. Phifer is vice president of Core Competence Inc. She has been involved in the design, implementation and evaluation of data communications, internetworking, security and network management products for more than 20 years and has advised companies large and small regarding security needs, product assessment and the use of emerging technologies and best practices.
This was first published in January 2008