Cisco Cisco Catalyst 6000 Multilayer Switch Feature Card MSFC2 White Paper

Page 30 of 41

Shadows

Features in a room such as support columns or balconies can represent natural shadows within a room. Under

normal densities, such features would require an additional AP to ensure coverage for users placed in such a

shadow. In a high-density installation, these features can be used to advantage, increasing the channel reuse by

designing to maximize this effect and then filling with another channel.

Under Seat Mounting

One of the optimal ways to cover a large, dense area is from underneath the users. This provides two advantages.

First, the users themselves attenuate the signals (so more APs can be accommodated with channel reuse than

other more open methods would allow). Second, this can generally provide a great way to hide APs. To do this

successfully, a few experiments must be performed to evaluate the propagation characteristics of the mounting

location. Under seat or under desk mounting can provide from 6 dB to 20 dB of attenuation to the cell, depending

on the openness of the mounting options.

It is important to note that metal chair legs and desk components will interact with the antenna of the AP and

change the pattern of the radiation. Surveying the results of placement decisions with a good tool is necessary

before making permanent mounting decisions. This can be done by selecting a representative section of the room

and placing a minimum of 4 APs (even more is better) temporarily in place. Analyzing more than one mounting

option will yield comparative data that will be helpful in making a final decision.

The solution chosen should provide decent attenuation and a usable pattern. Having shadows in this type of

mounting is not a problem and may be useful if the shadows are consistent and filled in by other cells. This may

sound like considerable work but the results can be well worth the effort. Generally, use of APs with internal

antennas is recommended for this type of installation. Experiment with the AP antennas pointed both up and

down. Do not be overly concerned about the power levels; in an empty room they will generally look very high

when compared to a normal installation. As a final test, select a portion of the room-perhaps 25 percent of the

seating area-and measure the site with users present. This can be done using a non production service set

identifier (SSID) to prevent attracting live users. What we are evaluating at this stage is placement, not user

throughput. Look at CCI and the coverage pattern to evaluate the effectiveness of the design.

Under Floor Mounting

Mounting APs under the floor with the intent of radiating through to the users can provide a great deal of

attenuation. This in theory will work and will allow the creation of more pico cells within a given environment.

However, based on experience, the number of facilities where this has proven practical remains small. Mounting

beneath the floor should not be ignored as an option and it should be investigated. The occasional corner case

where this is a viable option can be quite useful to the design. The RF density of the floor and access available will

be the limiting factors. Pre-stressed concrete floors or floors containing a lot of reinforcing materials are generally

not suitable. It is also important to consider the reflective potential of the floor material and where this will reflect to

in the rest of the infrastructure.