Wireless Network Planning: Conducting a Site Survey
Engineers conduct site surveys of a location while planning a road, and an IT professional should conduct a wireless survey before installing a wireless network. Factors such as absorptions, reflection, and scattering can cause degradation in the quality of a wireless signal. When conducting a site survey, you goal is to measure the quality of the signal as well as identify potential sources of interference.
A site survey allows you to do the following:
Identify other wireless networks in the area that might conflict with your network as well as other sources of interference.
Place a temporary AP and measure the signal strength from different areas.
When planning your access point (AP) locations, remember to
Plan deployments in multiple floor buildings because signals will span floors. Remember that you are working in three dimensions.
Ensure that all areas where roaming is expected will be connected to a common Layer 2 data link.
Many methods of conducting a site survey and many tools are available for you to use. In most cases, people use 802.11 network devices and measure signal strength with some piece of software; but more expensive tools are available to measure signals across the RF spectrum, thereby showing you interference that may exist which is outside the 802.11 wireless network ranges.
Because these tools are more expensive, most people who offer to do a site survey do not use this technology.
You can use tools such as Network Stumbler or MetaGeek’s inSSIDer for free to collect statistics or pay for a tool from a company such as AirMagnet. The following illustration shows the interfaces for Network Stumbler. The AirMagnet toolsets are very nice and will make creating a polished report very easy.
Cisco recommends that, even after conducting a site survey with these tools, that a pure performance testing using different devices be performed, because you can get a wide range of performance differences based on the antenna gain and application limitations when switching between devices.
You can conduct the site survey using two methods:
Survey surround signals: This method includes temporary APs that have been set up for your testing purpose. You place these test APs (often when testing you will have just one AP that you move around) close to your expected production locations. Then you test that signal from different locations to get a real-world rating for signal strength from various locations and to identify construction in areas that unexpectedly impact performance.
Performance testing: This method is similar to the first method, but in addition to checking signal strength and sources of interference, you also do some performance testing. The closer to real world this test can be, the better.
To perform this testing, you require another device connected to the AP, such as a laptop, running software. At a minimum, the device will be a Windows computer with a shared folder, with which you will copy files to and from (while documenting the transfer times). These times will be compared with results from other locations, providing you with not only signal strength, but also real-world performance.