Antenna Selection Guide

When Selecting an antenna for a wireless system, designers need to consider the future needs of the cell site. 

By Robert Wilson, TESSCO

A communications system has many component parts, each of which contributes to the overall performance and ultimately affects operator revenues. Well-designed components that are complementary with each other will keep overall performance within acceptable limits (see Figure 1a).

One of the first considerations in a communications system is antennas. Although, for example, the base station antenna rep-resents only 2% to 4% of the overall cost of a communications site, its performance impact is enormous. These antennas represent the critical piece of the puzzle that either initiates the transition of RF energy into free space to communicate with remote users or pulls the remote user's signal out of the air and allows it to be passed on to the communications system. Using the wrong antenna for the job will degrade the overall performance of an otherwise well engineered system, resulting in customer dissatisfaction.

Another consideration is future needs. Cell site antenna products that initially perform well when communications sites are not tightly spaced or heavily loaded may have significant problems as site density increases and traffic loading peaks. It can cost upwards of $2,000 to rent a crane and $3,000 a day for a crew of tower climbers and riggers to go to a communications site and change or adjust installed antennas. The site may have to be taken off the air or traffic rerouted during this process. Obviously, if the system owner is to maximize both returns and customer satisfaction, base station antenna selection deserves the consideration any far-reaching decision should command. Similar consideration is needed when choosing a point-to-point microwave, earth station, or in-building antenna system.

Base Station Antennas
Typical PCS, wireless, and cellular base station antenna sites are sectorized to increase the number of frequencies available for carrying traffic. In the simplest form, each sector will have three antennas, one transmit (Tx) and two receive (Rx) (see Figure lb). The two receive antennas are spaced a number of feet apart to provide improvement of the received signal through space diversity. In this configuration, three runs of transmission line will be needed on the tower, one for each antenna. A refinement of this approach is to use a duplexer on one antenna so it can do double duty for both Tx and Rx. This approach offers the advantages of reduced weight and wind loading on the mounting structure and only two runs of transmission line.

Figure 1.  Typical site configurations for services anywhere in the world that use cell-based wireless systems. A) A typical space diversity base station antenna site with duplexed transmit antenna. B) A typical space diversity antenna configuration with separate transmit and receive uses three antennas. This platform is installed on a monopole tower, which is often more acceptable to planners in densely populated areas than the standard tower. C)  newer configuration that uses polarization diversity antennas mounted on a tripod rooftop pedestal. These antennas can also be mounted on monopoles as shown in b. The units shown in b and c can also have a microwave point-to-point antenna as illustrated on the tower in a.

There are now new choices of technology available and the most sophisticated applications are switching from the traditional space diversity to polarization diversity, which uses only one antenna with two transmission lines. In this configuration, slant 45-degree polarization diversity is used. One polarization diversity antenna will replace two vertically polarized antennas spaced several feet apart. The polarization diversity antenna, typically no larger than a standard vertically polarized antenna, has one slant polarization duplexed for Tx and Rx and the other slant polarization for Rx only. The structure loading is reduced to only one antenna and two transmission lines with this system (see Figure 1c). In situations where the monthly rent is paid by the number of antennas used, the use of polarization diversity can save a substantial amount of money over the life of the system.

Selection Criteria for Base Station Antennas
When applying base station antennas in a PCS/wireless/cellular environment, there are a number of pattern characteristics that will have an impact on performance. These are bandwidth, gain, elevation beamwidth, azimuth beamwidth, upper sidelobe level, null-fill, front-to-back ratio, VSWR (Voltage Standing Wave Ratio), and intermodulation distortion.

Figure 2a is a representative polar plot of the elevation pattern of a directional PCS/wireless/cellular antenna that illustrates the relationship between these factors. The plot shows a 40 dB range of relative power. This is used to accurately portray the characteristics of the antenna. Use of a lesser scale can mask the true character of an antenna.

	Figure 2a.  This polar coordinate plot of the elevation cut of a directional antenna pattern labels the most important technical features of the pattern. This pattern illustrates the use of null-fill and upper sidelobe suppression.
	Figure 2b.  This is a polar coordinate plot of a standard "uniform illumination" antenna without null-fill. The graph shows how nulls from that antenna affect the energy on the surface of the earth in reference to its pattern.

The antenna's gain can be shown by the following formula:
Gain = Directivity
- loss from resistance of antenna/feed line conducting elements
- dielectric losses of radome
- impedance mismatch to external feed line
- radiation outside of intended polarization

Antenna gain increases as its aperture size increases. For PCS/wireless/cellular panel or omni antennas, this can be equated to the antenna length. As a general rule, gain doubles (3 dB increase) when the antenna's length doubles or when the beamwidth is decreased by one half. As the gain increases, the internal feed network increases in internal losses start to increase faster than the increase in
gain. At this point, substantial increases in antenna length are required for relatively small increases in gain. To make matters worse, the elevation beamwidth will become very small, which makes it difficult to achieve uniform cell illumination and makes the antenna very sensitive to movement of the tower it is mounted on. These factors establish practical limits on gain of approximately 18 dBi for 2 GHz panel antennas and 11 dBi for 2 GHz omni antennas.

Mechanical vs. Electrical Beamtilt. The footprint of an antenna pattern behaves quite differently with mechanical versus electrical tilting. When a panel-type antenna is mechanically tilted, only the peak of the main beam is at the specified angle. A way to visualize this is to cut a flat piece of paper into a circular disk to represent the energy from the antenna. The antenna would be at the center of the paper, and a line representing the peak of the beam is drawn from the center to the edge of the paper. When the paper is tilted to tilt the beam, it can be seen that there is no tilt at the angle 90 degrees from the peak, but at the opposite peak at 180 degrees, the beam is pointed upwards. 

Thus, a mechanically tilted panel antenna gives a reduced coverage footprint at the peak of the beam, but as the angle increases from this point, the amount of beamtilt decreases. The effect produces a pattern for a smaller tilt or no tilt at all.

A reduction in energy is seen at the horizon at cell sites off boresight, and the front-to-back ratio of the mechanically beamtilted antenna may actually degrade with the tilt angle.

With electrical beamtilt, the energy is phased between elements of the antenna, such that the radiation from the top of the antenna is slightly ahead of that from lower elements. This effectively tilts the beam down. This can be visualized by taking a conical coffee filter and placing it upside down to invert the cone. The angle of tilt is now equal at all horizontal tilt angles.

Clearing Obstacles. When an antenna's energy is reflected from an obstacle, such as the edge of a rooftop, there is a 180-degree phase shift. This reflected signal adds to the direct beam in a destructive manner, reducing received signal levels. There are two guidelines for compensating for this effect. The first is a field-derived rule of thumb that states that the -10 dB angle of the antenna's elevation pattern must clear any obstructions. Another more scientific approach is to calculate the Fresnel Zone clearance and position the antenna such that the main beam has at least 0.6 first Fresnel clearance. The first Fresnel Zone is a locus of points along the main beam that will produce cancellation of the main and reflected signals to a user. The object is to make sure that there are no obstacles that intrude into this first Fresnel Zone for optimum performance.

RF Exposure Guidelines. When adding equipment to existing communications sites and developing new sites, a user in the United States must determine compliance to the FCC's OET (Office of Engineering and Technology) Bulletin No. 65 "Evaluating Compliance with FCC-Specified Guidelines for Human Exposure to Radio Frequency Radiation." This bulletin contains many tables and figures to help an applicant make a fairly quick determination that a facility is in compliance with the new limits. Note should be taken of section 4 of this bulletin, dealing with controlling exposure situations. In uncontrolled situations, people in the general population are exposed to RF energy in their workplace and are not fully aware of the potential for exposure and/or can not exercise control over their exposure. There are no worldwide guidelines or mandates regarding RF exposure so when working internationally, the local PTT authorities need to be consulted. 

Figure 2c.  This a polar coordinate plot of an antenna with null fill. The graph shows the energy on the surface of the earth from that antenna in reference to its pattern and how null-fill improves the signal.

Moving the Signal Between Base Stations
Supporting base station antennas are terrestrial microwave (TMW) "dish" type point-to-point antennas or earth station antennas (ESAs). Both of these choices can be more economical than local copper or fiber connections. TMW antennas are typically used where a link can be established to a switch or hub within a line-of-sight radius. The ESA is used where line-of-sight links can not be established.

Terrestrial Microwave Connections
Terrestrial microwave point-to-point antennas are used primarily for connecting a cell site to the switch (shown in Figure la). This is known as backhaul. Microwave is an excellent economical alternative to leased land-based telephone lines because it often offers higher signal quality and less maintenance. A microwave system can be deployed quickly and has much higher reliability than land lines. Microwave antennas are matched to the application, whether it is fixed short-haul base station-to-microcell site links or long-haul repeater communications. These antennas are offered in single and dual polarized as well as standard and low profile versions to meet the needs of the system operator and standards for local environmental impact.

As frequencies are increased, there are two factors that are often overlooked in selecting an antenna for a given application. The first is the twist and sway of the structure that the antenna is mounted to. For example, if the frequency of an 8-ft diameter antenna is increased from 2 GHz to 6.5 GHz, the maximum allowable tower twist and sway decreases from 3.5 degrees to 1.0 degrees - an increase in tower rigidity by 3.5 times. A 2-ft diameter dish at 22 GHz will have a beamwidth of approximately 2 degrees; therefore, the movement of the antenna must be limited to a fraction of this value to assure optimal performance under windload conditions.

The second is rainfall. For microwave frequencies above about 12 GHz, the absorption of RF energy by rainfall becomes an important factor in determining path length and antenna gain. The amount of annual rainfall is not as important as the intensity of the expected rain. Areas such as Washington State with its frequent rains are not as much of a problem as areas such as Florida or those along the Gulf Coast.

Earth Station Antenna Links
Earth station antennas can provide higher bandwidth and capacity than their TMW counterparts. Their selection and design is more difficult, and these systems are designed around the carrier-to-noise density ratio required, government regulations, frequency band, coordination requirements, satellite system utilization, site location, and implementation parameters.

The satellite manufacturer will provide the effective isotropic radiated power (EIRP) towards the earth that is known as the satellite's "footprint." In multicarrier applications where there is a mix of voice data and video, EIRP is replaced by EIRP/carrier, as there is normally a back-off in power associated with multicarrier transponders to avoid high levels of inter-modulation. This is due to the limitation of the total power available as a composite to all carriers. If this power limit is exceeded, unwanted spurious intermodulation interference is generated, degrading signal quality.

The final carrier-to-noise ratio that is delivered to the demodulation equipment includes degradation from the uplink and downlink paths and terrestrial, adjacent satellite, cross polarized, and adjacent transponder interferences. In this calculation, it is assumed that these sources of interference are noncoherent and the downlink path will normally dominate the calculation.

As can be seen in Figure 3, the earth station is the gateway to a large number of communications services that are complimentary to PCS/wireless/cellular communications systems.

Figure 3.  An ESA system in support of PCS/wireless/cellular applications typically connects the services shown. The ESA system is used for very long distance interconnections and can concentrate large amounts of information.

Transmission Line Basics
When communications systems are designed, the concept of a "link energy budget" or just simply link budget is important. This link budget takes into account the energy from the transmitter, the path loss, antenna gains, and losses to the receiver at the opposite end. To over-come losses from propagation by choosing the proper antenna gain and pattern, it is important to reduce the losses from the antenna to the base station or switch as well.

Meeting this calculated loss figure can be quite a challenge and requires diligent selection of everything that configures a wireless system. The first place to start is wherever there are junctions or connections, and that means the transmission line offers opportunities to keep system loss within acceptable limits.

The transmission line system incorporates the main transmission line cable, jumpers, connectors, and accessories that are part of all site configurations. These components can be important contributors of loss in the system if not properly selected and matched. Foam-dielectric coaxial cable transmission line, most commonly used in PCS/wireless/cellular installations, should have copper inner conductors, low dielectric foam with closed cells, and a solid copper outer conductor. Copper inner and outer conductors provide low loss and high shielding efficiency, and the closed-cell foam dielectric will prevent moisture accumulation and migration. Moisture can rapidly increase attenuation and return loss.

The connector should electronically appear as an extension of the transmission line. It should offer good shielding, low VSWR, easy attachment, and low intermodulation levels. This is especially true for the jumper assemblies used at the top of the tower between the main feeder and the antenna and at the lower end of the feeder to the RF equipment. These assemblies and the grounding and attachment accessories associated with them are also critical items for proper system performance.

Impedance. Impedance is a characteristic that has been standardized in the industry at 50 ohms and represents a compromise between optimum attenuation (about 75 ohms) and optimum power handling (about 35 ohms). It is primarily determined by a ratio of the inner and outer conductor dimensions and the dielectric foam density, which determines the velocity of the cable. For best system performance, impedance should be controlled within close limits, typically to within 1 ohm from 50 ohms nominal.

Shielding.  Excellent shielding is one of the primary reasons for using transmission line with a continuous outer conductor. In practice, shielding is limited by the connectors but should be at least 120 dB.

Attenuation.  Attenuation is principally determined by overall cable diameter and will vary approximately inversely with cable diameter. In other words, the larger the cable the lower the attenuation. A typical system will allow between 1 and 2 dB of loss due to the attenuation of the transmission line and cable combination.

Power Ratings. Peak and average power ratings of transmission lines are rarely a limitation for typical wireless communications systems. A more important factor is VSWR. For a transmission line system, measured VSWR consists of three components. First is the reflection from the connector nearest the measuring equipment. Second is the contribution from the other connectors, appropriately reduced by the attenuation of the cable. Third is the contribution from the cable itself and arises chiefly from very small dimensional variations that are periodic at constant intervals of cable length. The periodicity means that a large number of these very small reflections are in phase and will combine at certain frequencies where the periodic spacing is an integral number of half wave-lengths. The result is that the cable con-tributes a number of narrow band "spikes" of higher VSWR.

Intermodulation. Intermodulation (IM) is another critical specification. IM is the mixing of two or more signals at a nonlinear mechanical junction that will produce spurious signals at multiples of the signals that are present at this junction. A principal cause of the nonlinearity and consequent IM generation is the use of ferromagnetic materials, such as stainless steel and nickel, in the RF path. The second major cause of IM is low contact pressures at connector inner conductor contacts on cable connections or inside of the antennas. These generating mechanisms are minimized by eliminating ferromagnetic materials in RF current paths, minimizing the number of junctions, ensuring good metal-to-metal contacts at all junctions, and soldering or high mechanical contact pressures. This is true for all antenna types mentioned. 

Figure 4.  An off-air building communication extensions system accepts signals from an existing cell site and extends it into a building. Conversely, this system will take a signal from within the building that is normally blocked and transmit it back to the outside existing PCS/wireless/cellular site.

In-Building Antennas
The last frontier in wireless communications systems is bringing PCS/wireless/cellular capabilities into office buildings, parking garages, hospitals, shopping malls, airports, convention centers, amusement parks, hotels, and small underground areas such as train or subway stations. To give wireless users the greatest mobility would be to allow them to move seamlessly between outdoors and indoors with no disruptions in call quality and service, but designing an in-building communications system can be the most complex of all sys-tem designs.

There are two basic types of in-building systems. The first extends an existing cell site by borrowing its signal and bringing it indoors. This is an off-air type of site. The second locates a full cell site or a portion of a cell site in the building. This is a service extension type of site.

The off-air site is used where the additional capacity from in-building communications will not load the existing cell site beyond capacity, and the in-building traffic is not expected to grow to the point where it will decrease availability for the normal external subscribers. The service extension type of site is used where there is enough subscriber traffic to justify on-site cellular PCS/wireless/cellular equipment. For the off-an type site (see Figure 4), the cell site signal is captured by an antenna located with a clear view of the closest cell site and coaxial cable is used to route the signal to the in-building system. The in-building system will often include one or more amplifiers to overcome path loss to the building and the internal cable losses. 

The signal is then distributed through a combination of radiating coaxial cable and point source antennas. Point source antennas work best in large unobstructed areas, such as the interiors of shopping malls and convention centers, and the radiating coaxial cable works best where coverage is restricted or needs to be contained such as hospitals, elevator shafts, corridors, subway tunnels, and office spaces. Four primary factors that must be taken into account when designing in-building communications system are 1) insertion loss of the cable, 2) coupling loss of the antenna or radiating coaxial cable, 3) operating margin of the system, and 4) fire rating of the cable.

Insertion loss is a measure of the RF attenuation per unit length of the cable and increases with frequency. Coupling loss is a measure of the RF path attenuation between the antenna or radiating coaxial cable and the intended coverage area.

Operating margin is the additional margin needed to assure a high probability of coverage and allow for system growth. When selecting components for in-building use, it is very important to know the building codes for the municipality and the specific building that will receive the installation. A "riser" fire rated cable can usually be used in vertical shafts and open wall mounting, but "plenum" fire rated cable should be used above suspended ceilings or compartments where air ducts are connected to form part of the building. 

This overview of what makes up a typical PCS/wireless/cellular antenna system is the first in a series of three articles. The objective has been to show how the components interact with each other and give the reader an understanding of what factors are important to the application of these products. The second article in this series will give specific technical criteria for evaluating and selecting each of these types of antennas.