Connect802 is a nationwide wireless data equipment reseller providing system design consulting, equipment configuration, and installation services.

CSS Mega Menu



August 1, 2005

Essential Wi-Fi:
For those who are new to Wi-Fi networking...
Technology and Engineering:
For the engineer and Wi-Fi network administrator...
To Infinity... and Beyond!
News from the wireless marketplace...

Essential Wi-Fi

2 dB Gain Omnidirectional Antenna
Note how Elevation (right graph) shows an
almost circular cross section.

10 dB Gain Omnidirectional Antenna
Note how Elevation (right graph) shows a
flattened cross section.

 Technology and Engineering

This month, we continue our discussion of RF signal propagation behavior. Last month, we covered the relatively intuitive absorption and reflection. This month, we'll cover refraction.

Of these three behaviors, refraction is probably the one we encounter the most. Refraction occurs when a wave travels from a medium of one density to a medium of a different density. The effect is that the wave front is bent. We experience refraction in visual light when looking at an object that is underwater. The light rays passing from the air to the water are bent because of the density difference between the air and the water, and the object appears to be displaced from its actual location. If, on the other hand, the human places his or her head underwater or raises the object out of the water, the light rays stay in a single medium as they travel from the object to the eyes, and the object appears to be where it really is.

Refraction occurs because the wave travels slower in the more-dense medium and faster in the less-dense medium. When the wave hits the boundary between the mediums , part of the wave front will enter the more dense medium first and slow down. The part of the wave that is still in the less-dense medium continues moving at its faster speed. Since the two parts of the wave front are moving at different speeds, the wave front turns.

At a basic level, refraction in microwaves (such as those used by 802.11) and light waves is identical, but because of their different frequencies, there are some practical differences. The first practical difference comes into play when a wave passes from one medium, into another medium, and then back into the first medium again--for example, if an 802.11 signal were to pass from the air into a gypsum board wall, and then back into air on the other side of the wall. When this happens, the refraction that occurs when the signal passes into the second medium is exactly reversed as it passes back into the first medium. The effect is that the direction of the signal doesn't change, but the signal is displaced slightly.

Of course, light waves and radio waves are both affected in the same way when they pass from a medium of one density, into a medium of another density, and then back into the original medium. The difference is that radio waves are almost always transmitted and received in mediums of the same density--air. Therefore, radio waves are more often displaced by refraction than bent.

Second, microwaves are highly absorbed by some substances that pass light--for example, water. Like light waves, microwaves are readily reflected by smooth water, but because the water is so absorptive of the microwaves, and, more importantly, because 802.11 receivers are usually not underwater, the fact that the microwaves are bent by refraction as they pass into the water is usually not important in designing an 802.11 network.

What conclusions should RF network designers make about refraction? First, if the transmitter and the receiver are both in a medium of roughly equal density, then refraction can probably be ignored, even if the two communicators are separated by a medium of a different density (e.g. the receivers are in the air and are separated by one or more walls). Typically, displacing the signal by a few inches won't affect the ability to receive it by too much, especially in an indoor, omnidirectional situation, where the network is designed to provide coverage over a wide area. In fact, you can probably ignore refraction when low-gain antennas are used, and especially in indoor environments where the receivers are usually relatively close to the transmitters and reflection provides redundant coverage to most areas.

Refraction must be considered in two main cases: first, where the transmitter and receiver are in mediums of different density, and second, when alignment of the antennas must be very precise, and even a few inches of variance may make a difference.

As we discussed, the first scenario is somewhat rare--it's unusual to find an 802.11 transmitter underwater. The major place where a density difference between transmitter and receiver can arise is in a long-distance point to point link. In this case, the density of the air can differ enough between the two locations to refract the signal enough to degrade the link. Remember, when meteorologists talk about a "high-pressure" or "low-pressure" front, they're referring to the density of the air in the area! Differences in density cause refraction.

Frankly, the second scenario is even rarer than the first. It typically occurs when signal passes from air through a wall or window and then back into air again, but in these cases, alignment of antennas is rarely critical.It's relatively easy to compensate for refraction. Simply use the antenna alignment tools provided by the vendor to find the antenna positions that maximize signal strength.

Because of refraction, the best position might NOT be one where the two antennas are pointed visually at one another, and that's okay. That only seems unusual because we can't "see" microwaves. This technique can be compromised when the density of the air in a location changes, as it inevitably will. You might align the antennas perfectly, only to find that when a high-pressure front rolls in, the link goes down. Remember that this will only be a problem when the transmitter and receiver are in areas of different density. If the same high-pressure front covers both locations, refraction won't occur.

One method of compensating for the variability caused by refraction is to use antennas with lower gain and wider beamwidths. A 24 dBi dish antenna with a three-degree beamwidth might seem optimal for a point-to-point link, but maybe a 15 dBi yagi with a wider beamwidth will be more tolerant of imperfect alignment and may be more robust even though it ends up with lower signal strength.

Next month, we'll discuss scattering and diffraction.

To Infinity... and Beyond!

We say MUST because there is a potential liability for the business or building owner when their network is the source of an Internet virus or worm attack, or when an unauthorized person uses the network for illegal activities. Consider a completely open Wi-Fi network where a spammer sits down and sips coffee while they send hundreds of spam emails through the unsecured portal. OK, so the CAN-SPAM police probably won't come running, but your ISP may block email from your IP address when it gets spam reports. It only takes ten or twelve spam reports to an ISP (with copies of the obviously spam email) to get an ISP to block your IP address for email sending. This means that visitors to your public access network will not be able to send email - and that's a bad thing for you. Now you're going to have to document your security corrections and submit paperwork to the ISP to get them to unblock your IP address. So, whether it's a legal concern over liability, or an administrative concern to protect against IP blocking, (or just good networking practice!) you need to properly secure your Wi-Fi network.