This document answers frequently asked questions about wireless systems, and covers areas such as antennas, polarization, interference, and safety.
A. Use any antenna that is:
Specified to work at the chosen or assigned carrier frequency.
Specified to operate over at least the 6 or 12 MHz bandwidth, as appropriate.
All antennas must have a 50-ohm impedance specification, and almost all do. For the most part, your antenna choice(s) are based on gain and directivity pattern characteristics required, which in turn are based on the range (path length) of the link and the topology (point-to-point or multipoint).
A. No. For example, there are cases where the antenna-mounting arrangements at one end of a link is only able to physically support relatively small antennas, such as a one- or two-foot dish. Yet the link requires a larger antenna at the other end to provide the necessary antenna gain for the path length in question. Sometimes, a high-gain, narrow pattern antenna is necessary at one end to avert an interference problem, which is probably not a concern at the other end.
Remember that the total antenna gain for a link is commutative—if the two antennas have different gains, you do not need to consider which antenna is at which end (except in consideration of mounting/interference issues).
Warning: Even though the two antennas for a link can look very different from each other, they must have the same polarization in order for the link to work properly.
A. The gain of any antenna is essentially a specification that quantifies how well that antenna is able to direct the radiated radio frequency (RF) energy into a particular direction. Thus, high-gain antennas direct energy more narrowly and precisely, and low-gain antennas direct energy more broadly. With dish-type antennas, for example, the operation is exactly analogous to the operation of the reflector on a flashlight. The reflector concentrates the output of the flashlight bulb into one predominant direction in order to maximize the brightness of the light output. This principle applies equally to any gain antenna, because there is always a trade-off between gain (brightness in a particular direction) and beam width (narrowness of the beam). Therefore, the gain and pattern of an antenna are fundamentally related. They are actually the same thing. Higher gain antennas always have narrower beamwidths (patterns), and low gain antennas always have wider beam widths.
A. Polarization is a physical phenomenon of radio signal propagation. In general, any two antennas that are to form a link with each other must be set for the same polarization. Typically, you set polarization through the way you mount the antenna (or just the feedhorn). As such, polarization is almost always adjustable at the time of antenna installation, or later.
There are two types of polarization, namely, linear and circular. Each has two subcategories within: for , and right- or left-handed for .
Linear polarization is categorized as vertical or horizontal.
Circular polarization is categorized as right-handed or left-handed.
Polarization Category Polarization Subcategory Notes Linear Vertical or Horizontal The vast majority of microwave or dish-type antennas are linearly polarized. Circular Right Handed or Left Handed Not encountered much in the commercial data communications realm. If, for example, the two antennas for a link are linearly polarized, they must both be either vertically polarized or horizontally polarized. If both antennas do not have the same polarization the link either works poorly or does not work at all. The situation where one antenna is vertically polarized and the other is horizontally polarized is known as cross-polarization.
For licensed links, the terms of the license can specifically dictate the polarization. For unlicensed links, you are typically free to choose, and the choice can be crucial to avert or correct an interference problem. See the interference resolution section for more information. Note that for most microwave (dish) antennas, you cannot determine the exact type of polarization the antenna is set up for through observation from a distance (such as when you view a tower-mounted antenna from the ground).
A. When two antennas do not have the same polarization the condition is called cross-polarization.
For example, if two antennas both had linear polarization, but one had vertical polarization and the other had horizontal polarization, the antennas are cross-polarized. The term cross-polarization (or "cross-pol") also generally describes any two antennas with opposite polarization.
Cross-polarization is sometimes beneficial. An example of this is a situation in which the antennas of link A are cross-polarized to the antennas of link B, where links A and B are two different but nearby links that are not meant to communicate with each other. In this case, the fact that links A and B are cross-polarized is beneficial because the cross-polarization prevents or reduces any possible interference between the links.
A. First of all, be sure that the two antennas for the link are not cross-polarized. After that, you need to be sure that each antenna is pointed or aligned to maximize the received signal level. A tool is commonly provided on the radio equipment to help determine this, in the form of an indicator or alignment port (use the Find function on your browser to locate this term) for a meter that gives a voltage reading proportional to the received signal level. At one end of the link at a time, the antenna pointing direction is carefully adjusted to maximize (or "peak") the reading on the indicator tool.
After this is done for both ends, you must obtain the actual received signal level in dBm in order to verify that it is within 0 to 4 dB of the value obtained from the link budget calculation. If the measured and calculated values differ by more than about 8 dB, you can suspect either that the antenna alignment is still not correct or that there is another defect in the antenna/transmission line system (or both).
Note: You can get a "peak" reading during the antenna alignment process if one or both of the antennas is aligned on a "side lobe," in which case the measured receive level may be 20 dB (or more) lower than the calculated value would indicate it should be. Be aware that the link may still work under these circumstances. If you get agreement to within 0 to 4 dB between the measured and calculated receive signal levels, you can be confident that the antennas are properly aligned with no other problems.
A. No. Any type of radio (or other electromagnetic) signal that propagates through space (or air) remains unaffected by any other signal that happens to cross the same point in space. In order to prove this, get two flashlights, and shine one onto a wall. Hold the other flashlight a distance away from the first, but point the second flashlight so that the two light beams cross. You notice that the beam from the second flashlight has no effect on the spot on the wall from the first. This same principle is true for radio signals of any frequency. Of course, in the flashlight example, if you shine the second light onto the same point on the wall, the spot appears brighter. If the beams were radio signals of the same frequency, and the spot on the wall was a receive antenna for one of the links, the second beam is indeed likely to cause interference. However, this is a different situation from when the beams cross in space.
A. No. Problems are unlikely in this situation. At the radio frequencies at which the links operate, the wires appear to be infinitely long conductors. As such, there is bound to be some slight diffraction effect on the signal that propagates across them. However, because the wires are thin, this effect is very slight, so much so that you can not even measure the effect. There must be no adverse impact on the operation of the link.
A. Probably not. First of all, the intermediate frequency (IF) cable (and RF cable) must have a 50-ohm impedance specification. Some types of coax cables that are/were used with LANs can have other impedance specifications, and thus you cannot use such cables.
If you verify that the existing cable is a 50-ohm type, the cable still must meet two other specification requirements before you can use the cable:
The total loss at 400 MHz for the entire run length must be 12 dB or less.
The center conductor size of the coax must be #14 AWG or larger.
If these requirements are met, you can use the existing cable. If there is any doubt, do not use the cable. Also remember that someone stopped using the existing cable for a reason, and that reason can that the cable has some invisible internal damage that caused the previous user expensive and frustrating problems. Coaxial cable, and even its installation, is relatively inexpensive, so do not take chances with your important link.
A. For your own single link, polarization does not really matter. However, there are two situations in which polarization is important:
(a) There are other nearby links that you do not control.
(b) You plan to install, or have already installed, other links to one of the end points of the new link.
For (a), determine whether the other nearby links are on a frequency that can possibly cause you an interference problem. Then attempt to determine the polarization of those links. If you can, you must set up your new link to be cross-polarized to the nearby links.
For (b), the same applies as for (a), except that now you can easily determine the frequency and polarization, because you deal with links that you control. A site with multiple links is known as a hub, and any two links to that hub that are on the same frequency (or a close enough frequency that they could interfere with each other) must be cross-polarized to each other to avoid potential interference problems.
A. You must seal the connections as soon as possible, as long as the system is functional and has not yet suffered any moisture-related damage. Some types of sealing products, such as Coax-Seal, enable you to seal the connections without the need to disconnect the connections or take an operational link off-line.
A. Unfortunately, this common question does not have a quick or simple answer. Here are the factors that govern the maximum link distance:
Maximum available transmit power.
Receiver sensitivity.
Availability of an unobstructed path for the radio signal.
Maximum available gain for the antenna(s).
System losses (such as loss through coax cable runs, connectors, and so forth).
Desired reliability level (availability) of link.
Some product literature or application tables quote figures, such as "20 miles." In general, these quoted single values are optimum, with all of the above variables optimized. Also, remember that the availability requirement has a drastic affect on the maximum range. That is, the link distance can perhaps be double, or more, than the quoted value if you are willing to accept consistently higher error rates, which can be appropriate in an example where you use the link only for digitized voice applications.
The best way to get a useful answer is to do a physical site survey, which involves examination of the radio path environment (terrain and man-made obstructions) at the proposed link location. The results of such a survey can yield valuable information on:
The radio path loss.
Any issues that can further compromise link performance, for example, potential interference.
When you obtain this information, you can choose and know the other variables, such as antenna gain, and you can obtain a very definitive answer for the maximum range.
A. In short, the duplexer is a device that allows a transmitter and a receiver to be connected simultaneously to the same antenna.
Any two-way wireless communication requires both a transmitter and a receiver. If you want to transmit and receive at the same time (also known as full-duplex operation), clearly the transmitter and receiver must both operate at the same time. Even if each had its own antenna, full-duplex operation can present a problem because the power output of the transmitter is millions of times greater than the power level of signals the receiver tries to receive. If these two devices operate at the same time in close proximity (which they typically are), some of the energy from the transmitter is bound to find its way into the receiver, where the energy is more powerful in comparison to the signals the receiver wants to receive. When the transmitter and receiver are connected to the same antenna, the problem becomes even more acute.
In order for full-duplex to work at all, there has to be some scheme to separate the transmit and receive signals. One common technique to do this, which Cisco broadband wireless products employ, is to transmit and receive on different frequencies. This system is called frequency-division duplex. The idea is that the receiver will not be able to "hear" the transmitted signal because the receiver is selective. The receiver only receives a frequency (or a small range of frequencies) to which the receiver is tuned, and does not receive the transmitted signal if the frequency is outside of the tuning range of the receiver (called the receive passband).
Although this fundamental idea is quite sound, you can still face a problem. The receiver obtains the selectivity characteristic through filters, which pass certain frequencies and reject others. However, the types of filters that are practical to incorporate into the internal circuitry design of the receiver are not selective enough to prevent the relatively powerful transmit signal from adversely affecting the operation of the receiver, even if the transmit frequency is well outside the passband range of the receiver filter. In this situation, add more filtering.
Think of the duplexer as just a pair of bandpass filters incorporated together in one box. It has three connection ports:
The transmit (TX) port.
The receive (RX) port.
The antenna port.
The TX and RX ports are usually interchangeable. In most implementations (including Cisco's broadband wireless solutions), the duplexer is a passive device. The duplexer neither requires nor consumes any power. Consequently, you cannot configure the duplexer, either through software control or other means.
In fact, some mechanical adjustments are made at the time of manufacture, but after that time there must never be any need to readjust these, and so any adjustment or calibration access points are typically sealed and you must not tamper with them. The two passband filters that make up the duplexer are very steep-skirted, which means they easily pass frequencies within the passband, but then greatly attenuate signals that are outside of the passband frequency range by only a small amount. This characteristic is important to enable the duplexer to keep powerful transmit signals out of the receiver. The requirements of steep-skirted selectivity and high out-of-band attenuation are what make the duplexer unique. The duplexer must also be able to handle the power level of the transmitted signal that passes through.
The duplexer has two non-overlapping passband frequency ranges, and thus one is naturally higher than the other. You can set up a system to transmit through the higher frequency passband filter and receive through the lower frequency one, or vice-versa. These two scenarios are usually described as transmit-high or transmit-low. The duplexer is not concerned with how this is done. The only real requirement, as far as the duplexer is concerned, is to make sure that the transmit frequency falls within the passband range of one of the filters of the duplexer, and the receive frequency falls within the other. This requires that you know the passband frequency ranges of the duplexer, and the TX and RX operating frequencies when you install or operate the duplexer.
In practice, you must first determine, to at least some rough degree, what the transmit and receive frequencies must be. Then, choose a duplexer with appropriate TX and RX passband ranges to accommodate the necessary operation frequencies. This does not require an infinite range of offerings of duplexers. Rather, they are provided in a relatively few choices, one of which fulfills the requirement. If you try to operate on a TX or RX frequency (or both) that falls outside of the passband range(s) of the duplexer, the system does not work. After you install or order the system, if you want to alter either the TX or RX frequencies (or both), you can do so as long as any new frequencies that you choose fall within the passbands of the duplexer. Otherwise, you must obtain a different duplexer (for each end of the link).
Finally, note that you cannot reverse the existing TX/RX split (change TX high to TX low, or vice-versa) unless you also physically reverse the connections to the duplexer. Otherwise, the system cannot work after the split is reversed in the setup configuration, because now neither the TX nor RX frequencies fall within the duplexer passbands. For the Cisco Systems solution, in order to reverse the duplexer connections, you must remove the duplexer from the transverter, "flip" it around, and re-install it.
A. Yes. Aside from the obvious concerns, such as safety when you climb structures or when you work with dangerous AC line voltage, you must also be aware of the issue of exposure to RF radiation.
There is still a lot that is unknown, so there is much debate about the safe limits of human exposure to RF radiation.
Remember that the use of the word "radiation" here does not necessarily connote any linkage to or issue with nuclear fission or other radioactive processes.
The best general rule is to avoid unnecessary exposure to radiated RF energy. Do not stand in front of, or in close proximity to, any antenna that radiates a transmitted signal. Antennas that are only used to receive signals do not pose any danger or problem. For dish-type antennas, you can safely be near an operating transmit antenna if you are to the back or sides of the antenna, because these antennas are directional and potentially hazardous emission levels are only present at the front of the antenna. For more details, refer to the radiation hazard calculation table. Use the Find function on your browser to locate this term.
Always assume that any antenna transmits RF energy, especially because most antennas are used in duplex systems. Be particularly wary of small-sized dishes (one foot or less), because these dish antennas often radiate RF energy in the tens-of-gigahertz frequency range. As a general rule, the higher the frequency, the more potentially hazardous the radiation. If you look into the open (unterminated) end of waveguide that carries RF energy at 10 or more GHz, you can suffer from retinal damage if the exposure lasts only tens of seconds and the transmit power level is only a few watts. There is no known danger if you look at the unterminated end of coaxial cables that carry such energy. In any case, be careful to ensure that the transmitter is not operational before you remove or replace any antenna connections.
If you are on a rooftop and near an installation of microwave antennas, do not walk, and especially do not stand, in front of any of the equipment. If you must traverse a path in front of any such antennas, there is typically a very low safety concern if you move briskly across an antenna's path axis.
A. In general, the diversity option is not necessary if the link is unobstructed. In other words, you do not require the diversity option if the link is a "radio line-of-sight" link.
The diversity feature of Cisco's broadband wireless solutions is designed to allow reliable link operation in installations where you cannot achieve line-of-sight, and where establishment of a usable radio link would not be possible otherwise. The diversity transverter, when installed, is used only to receive signals. The diversity transverter does not transmit.
Note that the diversity option is not effective if the obstruction to the path is severe, for example, obstruction due to a mountain. The option is most effective in urban installations where the path might be line-of-sight except for one or two buildings in the path, for instance. In such cases, the best way to know the degree of effective performance gain that the diversity option provides is the empirical approach—install and see.
There is a way to run a test on an installed non-diversity link to get a fairly good idea of how much such a link can benefit from the addition of the diversity feature. Refer to the wireless line card documentation for information about throughput setting. Use the find function on your browser to locate this term.
In general, the antenna of the diversity transverter must be the same as the antenna you use for the main transverter, but this is not an absolute requirement. However, the polarization of the diversity antenna must be the same as the main antenna.
A. When you consider the possibility of interference problems, there are some "common sense" items to know and watch out for. Here is the list:
Understand that operation in unlicensed bands carries an inherently higher risk of interference, because the controls and protections of a license are not afforded to you.
In the United States, for example, the Federal Communications Commission (FCC) does not have any rule that specifically prohibits a new user from installing a new unlicensed-band radio link in your area and on "your" frequency. In such a case, you can experience interference. However, there are two issues to consider in such a situation.
If someone installs a link that interferes with you, chances are that you also interfere with them. The other party can note the problem during system installation, and choose another frequency or channel.
With point-to-point links that employ directional antennas, any signal source (of a comparable power level to yours) that can cause you any interference would have to be closely aligned along your own path axis. The higher the gain of the antennas you use, the more precisely the interfering signal would have to be aligned with your path in order to cause a problem. That is why, Cisco recommends that you use the highest-gain antennas for point-to-point links as is practical. Thus, in unlicensed bands, the potential for interference from another unlicensed user, as a practical matter, is not much greater than for licensed bands, where you essentially "own" your frequency.
Remember that some licensed users sometimes operate in the unlicensed bands as well. The unlicensed bands are allocated on a shared basis, and while there is no requirement for you to obtain a license to operate for low-power datacom applications with approved equipment, other licensed users can be allowed to operate with significantly higher power. A specifically important example of this is operation of U.S. government radar equipment in the U.S. U-NII band at 5.725 to 5.825 GHz. These radars often operate at peak power levels of millions of watts, which can cause significant interference problems to other nearby users in this band. Therefore, look around your site to determine whether there are any airports or military bases, where such radars can exist. If so, you must be prepared to experience periods of interference.
If you are a licensed user and you operate in a licensed band, you do not have to worry about interference. If you experience problems, there are legal statutes that provide for resolution of the matter.