Understanding the term 'Site Survey' - correctly...
12-03-2014 09:17 AM
A Site Survey is a component of the WLAN Design process. It is NOT the actual design!
We need to call things how they really are and not rely on silly terms like 'Site Survey' - it means too many things to too many different groups. Some use it incorrectly to mean the WLAN design process, others a visit on site post installation to validate the Wireless LAN - others in the construction industry use if for entirely different meanings.
If we want to talk about Wireless LAN Design, lets just call it Wirelss LAN Design.
First - Define Requirements: this should take a bulk of the design time. It need to be detailed and complete.
Second - Design: this is the process of attempting to meet all the requirements while staying within the design constraints. Coverage is very easy to obtain... but keeping the CCI/CCC to a minimum is the most difficult part. An on-site assessment of the RF characteristics of the buildings will help with accuracy.
Third - Deploy: the easist bits... mounting the APs, configuring the controller, etc. One tip - be sure to ALWAYS test the wired drops for PoE, DHCP, DNS, Default Gateway, VLAN assignments, etc. BEFORE attaching any AP to the wired network.
Fourth - NOW we do a Validiation Site Survey. This is a required step in the WLAN deploment process. You MUST test and verify the installed WLAN meets all the design requirements. If you leave this step out... you are bound to have another truck roll to return and fix the WLAN in the future.
Coverage is easy!
Anyone can get Coverage. Coverage is the easiest of all design requirements to meet. The hard part of WLAN design is working to ensure you can achieve Frequency Reuse - without it. Adding more AP's does NOT give you more capacity - it just adds to Co-Channel Contention and lower overall throughputs.
One more time... merely adding more access points does NOT ensure more capacity - you ONLY achieve additional capacity when you achieve frequency reuse.
Some Survey Terms -- sometimes people confuse these as well.
Predictive Survey: A software tool based on RF mathematics and specific algorithms to simulate RF in your environment and to allow for an itterative approach to design. Design to meet requirements while staying within constraints. Should be used with actual on-site measured RF attenuation values from the walls, and floors.
Passive Survey: Software tool that collects RF data from all AP's in the area. Plots heat-maps to show where coveragea is and is not at certain levels. Can also be used to validate many of the design requirements - Primary RSSI, Secondary RSSI, SNR, Noise, and most importantl Co-Channel Interference.
Active Survey: much like the Passive Survey - only collecting data from a single Access Point at a time. VERY difficult to do correctly. You have to walk the site multiple times to ensure getting Active data from each and every AP being tested. And be sure to collect the data clear to the edge of each AP's coverage. Else you'll have false data and not collect the CCI areas that are the most difficult to find/fix.
Active surveys contain some extra information not available in the Passive version. Like actual connected data rates, error rates, etc. But remember you are only collecting data for ONE AP at a time... takes a very long time to do an Active Survey properly and with complete and accurate data.
AP-on-a-Stick Survey: Using only a single AP, this technique was used in the past to confirm coverage. Very difficult to obtain any of the other requirements other than Primary RSSI. Does not give you Secondary RSSI, or Co-Channel Interference at all. It may be used prior to deployment, but must always be followed up with a post install Validation Survey. Always!
Post Validation Survey: This is one of the most important bits in any Wireless LAN design. It is the same as testing Cat6 cable... you do it AFTER it is installed. If it meets detailed specs, like near-side cross-talk, far-side cross-talk, pinouts, twist ratios, etc. Then you can confidently use the cable knowing it will achieve the designed gigabit Ethernet speeds.
The same is true for Wireless LANs. AFTER - you install, you need to test and measure to make sure the RF characteristics meet the design requirements. These include:
- Primary RSSI
- Secondary RSSI
- Co-Channel Interference
- Data Rates
- Device to Radio Ratios
- Jitter, Latency, and QoS
- High Density Areas
Only after a validation survey can you know for sure your Wireless LAN meets your design requirements and can be confident it will perform as expected.
In the future, try to use the correct term for 'Site Survey' - say what you mean and mean what you say. If you want to talk about Wireless LAN Design, just say so. If you are refering to a validation survey, or a passive survey, or an AP-on-a-Stick survey - just be a bit more precise in how you use this term.