For optimal performance of your wireless network, you must understand the purpose behind proper AP and antenna selection. Choose the correct AP and antenna type to ensure that application and band requirements are met and that RF energy is directed to the correct coverage areas.
The flow chart in Figure 10 shows the high-level process to select the appropriate AP for the deployment. Some choices, such as using only 2.4 GHz-capable APs, have substantial impact on the flexibility of future deployments and should not be made lightly. The following sections describe each of the decision points and the trade-offs that are made. It is important to fully understand the implications of the chart before you select an AP. The AP selection process uses the flow chart in Figure 10.
Figure 10 AP selection process
Starting at the top of the decision tree, the first selection to make is whether to use single- or dual-radio APs. Consider over the lifetime of this deployment, typically 4-7 years, what types of devices and device counts the AP will be expected to handle. In most cases, device counts are increasing, with a mix of clients that are often dual-band capable.
In a capacity-based deployment, you should always choose a dual-radio AP. The additional radio allows you to balance clients between 2.4 GHz and 5 GHz bands with features such as band steering, and it doubles the density of clients that can be served in a given location.
Single-radio APs are appropriate for low-density deployments and to act as air monitors. If you are selecting a single-radio AP as an air monitor, you should only select a dual-band capable AP because they are capable of scanning the 2.4 GHz and 5 GHz bands.
APs that support only 2.4 GHz should be selected only after very careful evaluation. They are limited by both their ability to access only a single band and the fact that bonded 40 MHz channels are not available, which limits the maximum PHY data rate to 135 Mb/s. Use these APs primarily for countries in the Asia-Pacific and Middle-East regions, which do not allow the use of the 5 GHz band. 2.4 GHz only APs are not appropriate for any environment where 5 GHz scanning is required, such as to meet the payment card industry (PCI) scanning requirement. Before you consider using these APs, take some time to evaluate the capacity you are planning and the use cases for the network.
Each AP in Figure 10 also lists its radio and spatial stream count in the format of:
Figure 11 Radio characteristics
In general terms, more radios lead to more spatial streams being available, though it is not always a 1-1 mapping as in the case of the AP-125. The more spatial streams you have, the higher the data throughput that can be transmitted to client devices, assuming they have the same capability. As newer laptops enter the market, 3x3:3 is becoming common in mid-range models. Older laptops will vary in their radio and spatial stream count. Most tablets today are 1x1:1 devices.
If the 802.11n data rates and spatial stream terminology is new to you, it is strongly recommended that you also read the Aruba 802.11n Networks Validated Reference Design (Version 8). This guide covers indoor 802.11n WLANs and is considered part of the foundation guides within the VRD core technologies series. This guide describes 802.11n technology and differences in 802.11n and 802.11a/b/g functionality.
For many deployments, ceiling mounted down-tilt antennas are the appropriate choice. These APs can be mounted on ceiling rails and they cover the floor below. For deployments where the APs need to be above the plenum for aesthetic reasons, external antennas may be an appropriate choice. This is especially true if the ceiling material is an RF absorber or reflector. External antennas also allow for the use of directional antennas in a challenging environment such as a warehouse.
Application usage and RF band requirements are the most important factors to consider when you select the proper AP for the deployment. It is critically important to understand the application and the types of devices that will be used to connect to the network.
You must consider current and future applications that may be deployed. Today, the network may need to support only data applications that are used to run the business. However, in the future the network may need to support voice or multicast video delivery. To begin to understand the data requirements, you must understand the application requirements and define the expected use cases. Consider these items:
• Application type and average bandwidth
• Tolerance to latency, jitter, and delay
• Requirements for traffic prioritization
• Data streaming rates
• Average file transfer sizes and frequency
Some application requirements will be easy to deal with, such as email and web browsing. However, others, such as voice and multicast video, require special planning. Closely examine any custom applications to ensure a smooth transition.
RF channels are another major decision point and they are related to the decision of whether to use single- or dual-radio APs. Understanding which RF bands are available in a particular country and the applications in use often influences that decision. RF bands matter most where a decision needs to be made between single- or dual-radio APs. In a single-radio model, each AP can serve clients on only one RF band (2.4 or 5 GHz) at a time. In a dual-radio model, both bands can be used.
Single-radio operation can be appropriate for some organizations that will only use devices that operate on one band. Examples include scanner guns, game consoles, digital video recorders, or single-mode voice handsets that are capable of supporting only the 2.4 GHz band. Dual-band capable single-radio APs also make excellent air monitors, where the AP can scan both channels. It is not uncommon to deploy dual-radio APs to serve clients and also deploy single-radio APs as full-time AMs to scan the network environment.
In other cases, user density and application types require that as much bandwidth as possible should be available to devices and applications. Some plans call for explicitly separating applications on different bands, such as having voice run on 2.4 GHz while data and video run on 5 GHz. These cases are the most common in a campus, dense user environments, or large branch deployment, where dual-radio APs are deployed.
Whether you are using a coverage or capacity approach, the rules for AP spacing are very different for 2.4 GHz vs. 5 GHz frequency radios. As the frequency increases, the distance at which a signal can be heard decreases, assuming a constant output power. On average, for any given data rate, 2.4 GHz signals travel twice as far as 5 GHz signals. This means that many more APs are required in 5 GHz to provide a level of service comparable to that experienced in a 2.4 GHz system.
In exchange for the higher density requirement, the 5 GHz spectrum offers many more channels and generally less congestion than the 2.4 GHz airspace. Bonded 40MHz channels that allow the full 300 Mb/s or 450 Mb/s data rates are possible only in 5 GHz. The 2.4 GHz airspace is shared with Bluetooth headsets, frequency-hopping (FH) devices, APs in neighboring locations, and wireless hotspots for customers. By moving to the 5 GHz band, users may need more APs, but the quality and reliability of voice and data communications increases significantly. Figure 12 shows the difference in speeds and range in an idealized model. Real world performance and 802.11n modulation and coding scheme (MCS) rates may vary for reasons beyond signal strength.
Figure 12 Cell radius varies with data rate and transmission frequency
When dual-band APs will be used, the coverage area must be planned for a 5 GHz AP density. This will result in an oversubscription in the 2.4 GHz band, which ARM can account for by adjusting transmit power. In general, each non-overlapping AP serves no more than 5,000 square feet, and usually much less if a capacity model is being deployed.
For optimal performance of the wireless network, it is critical to understand the purpose behind proper antenna selection. The correct antenna type will ensure that RF energy is being directed to the correct coverage areas. In some instances, the built-in omnidirectional or down-tilt antennas are not the correct choice for a deployment. For aesthetic reasons, the AP may need to be hidden from view. Or the nature of the connection may dictate that an omnidirectional antenna is not the correct choice. Aruba sells a wide variety of antennas to suite various deployment needs. For a complete, detailed view of the Aruba antenna line, see the antenna line matrix that is available on the VRD page: http://www.arubanetworks.com/vrd.
Omnidirectional antennas provide equal coverage in all directions, and directional antennas are engineered to deliver RF energy in a beam with a specific width and height measured in degrees. Integral, omnidirectional antennas provide low-gain performance over the entire frequency band that the AP supports (typically between 3 dBi and 5 dBi). Detachable antennas, omnidirectional or directional, are selected when a higher level of gain is required in a specific direction or for more aesthetic reasons. In Figure 13, an omnidirectional antenna has coverage that radiates in all directions compared with a directional antenna that radiates only in a narrow field.
Figure 13 Omnidirectional vs. directional antennas
Aruba has standardized on a special type of omnidirectional antenna that is used in our indoor product models that feature integrated antennas. It has what is called a “squint” or “down-tilt” pattern, which is a combination of a directional and an omnidirectional antenna. It is omnidirectional in the horizontal plane, providing a full 360 degrees of coverage. And it is directional in the vertical plane, optimized to provide coverage from a ceiling mount with very little energy directed behind the AP. You can visualize this in Figure 14. The AP-93H is built to be mounted into a wall box and provides coverage for a room using a similar antenna pattern as shown in Figure 15.
Figure 14 Ceiling coverage from Aruba AP with integrated down-tilt antenna
Figure 15 Wall-mounted AP with integrated directional antenna
Antenna gain is a relative measure of how the antenna compares to an ideal isotropic radiator. An ideal isotropic radiator would radiate power in all directions equally over a sphere, such as the sun. The relationship between gain, power, and propagation distance is detailed already in textbooks and those expressions are not repeated here. Aruba recommends the CWNP series of books for engineers looking for a solid foundation in Wi-Fi networking concepts, http://www.cwnp.com.
Antenna gain often is confused with power gain in amplifiers. However, it is important to note that antenna gain makes the power of a transmitter greater than would be predicted by calculation of the power fed to the antenna and then spread equally over a sphere. Antenna gain itself is a completely passive and bidirectional property that is determined only by the shape and construction of the antenna.
Gain is only a comparison of the apparent power to the power that would be required if fed to an ideal isotropic antenna. Gain can only be created by distorting the antenna pattern from the ideal spherical pattern. Think of this as focusing the same power that would normally distribute evenly over a sphere into a tighter region of band. Thus, the higher the gain, the more concentrated the antenna pattern must be to achieve that gain.
To visualize the concept of gain, picture a rubber ball. The surface area of the ball represents the total available power radiated by an ideal isotropic antenna over its sphere of radiation. (See Figure 16.)
Figure 16 Equal signal strength radiated in all directions
The ball is not the most efficient way to provide network access on a horizontal floor. One way is to press down on the top of the ball and squash it down vertically. The same basic shape is kept in the horizontal plane (round), but the ball is forced to compress in the vertical plane to create a donut shape. (See Figure 17.) This example represents the concept of the high-gain, omnidirectional antenna, which achieves a greater coverage distance in the horizontal direction at the expense of coverage in the vertical areas of the radiating sphere.
Figure 17 High-gain omnidirectional antenna
To stretch the ball primarily in one direction (instead of in all directions), force must be used to push the ball vertically and horizontally, on the sides and on the back, to force the ball to deform in a single direction. This action significantly distorts the shape of the original ball horizontally and vertically, allowing it to stretch in a single direction. (See Figure 18.) This example represents the concept of the high-gain directional antenna, which is designed to compress the entire radiating sphere into a single predominate direction.
Figure 18 High-gain directional antenna
Gain is created by forcing transmitted power to radiate in a preferred direction rather than radiating in all directions of an ideal sphere. Therefore, a high-gain signal always is accompanied by loss of available signal in some other portion of the ideal sphere. High-gain directional antennas are ideal for sites that require directed coverage in a specific area or in an extended range for bridging applications. They are not suited for sites that require uniform coverage in large areas. Remember that vertical and horizontal coverage can be affected by the use of a higher-gain antenna, and beamwidth (a measure of coverage) is always inversely related to gain.
Traditional two-dimensional (2-D) pattern plots and beamwidth specifications require one to use some imagination because they provide only a snapshot of the information in two planes. These two planes are often referred to as the azimuth (H-plane or horizontal) and elevation (E-plane or vertical) planes. The azimuth view is considered to be the view from directly above, which views the antenna pattern on the horizontal plane. The elevation view is considered to be a side view, which views the antenna pattern on the vertical plane. It is helpful to think of these planes as “slices” of the real antenna pattern, which is actually three dimensional (3-D). Figure 19 illustrates where these “cuts” are located for a typical omnidirectional antenna pattern.
Figure 19 Antenna pattern conventions (omnidirectional pattern shown)
The antenna shown in Figure 19 is commonly referred to as the dipole pattern because it is produced by an ideal dipole antenna. The gain of this antenna is 2.17 dB, which is achieved by compression in the vertical plane (elevation) compared to the ideal sphere. To refer to the true 3-D pattern, this compression is sometimes called the donut shape. (See Figure 20.)
Figure 20 Donut shape compression of an omnidirectional antenna
It is evident from these figures that the 2-D pattern plots that are typically provided in antenna specifications are a simplification of the real 3-D situation. Often, 2-D plots are reduced even further to a set of simple specifications based on the antenna gain and 3 dB beamwidth.
If the AP supports detachable antennas, select the correct antenna type to support the required frequency band (2.4 GHz or 5 GHz) and desired coverage pattern.
To select the correct antenna type for the deployment, download and read the antenna product matrix. A link is available from the VRD site at http://www.arubanetworks.com/vrd.
If you are new to external antennas, some of the terminology used in the Aruba antenna specifications may be imposing at first. But you will become comfortable with the terms soon because many of them are determined by the vertical or horizontal beamwidth. These are some of the terms used to describe the Aruba detachable antenna offerings:
• down-tilt: An omnidirectional antenna that focuses its energy downward (perpendicular to ground). Used for ceiling deployments in dense areas or in areas with high ceilings, such as manufacturing or warehouse environments.
• sector/patch/panel: A directional antenna that provides coverage of a focused area (or “sector”), typically parallel to the ground. Common indoor examples are 60 degree, 90 degree, and 120 degrees).
• If you want omnidirectional coverage but you want a higher gain than the integral antenna supports, select one of the Aruba detachable antennas with high-gain, omnidirectional coverage.
• If the application type is air monitoring with a single-radio AP, the detachable antenna must be able to support the required frequency bands (2.4 GHz and/or 5 GHz). If multiband support is required, the antenna must be multiband-capable. (AP-ANT-1B and AP-ANT-19 are the only detachable antennas capable of omnidirectional, multiband support.)
• If a directional antenna is needed to direct RF coverage, it must be able to support all of the frequency bands that require support (2.4 GHz and/or 5 GHz).
Table 1 summarizes the available Aruba indoor APs. More detailed information is available in the access point product line matrix available at: http://www.arubanetworks.com/vrd. The AP-134 and AP-135 feature dual 10/100/1000BASE-T Ethernet interfaces and operate from standard 802.3af and 802.3at power-over-Ethernet (PoE) sources. The secondary Ethernet interface (active only when supplying 802.3at PoE or DC power to the AP) enables secure authorized backhaul for wired network-attached devices.
Table 1 AP Features and Functions