Introduction Into Wi-fi Spectrum Analysis
Wi-Fi as most people know it is designed around two parameters: as much speed as cheaply as possible. What is missing is the “reliability” requirement.
Now, for most people, Wi-Fi seems to be reliable “enough”. Drop outs are rare and the equipment just works when you turn on your laptop. This is not always the case once you start to get multiple clients connected within the same area. This is not always the case in corporate environments. This is not always the case once you start to populate your premises with the multitude of electronic devices that can spew noise into the airspace.
What do I mean by the last point? Wi-Fi (802.11b, 802.11g and soon to be 802.11n) operates on the 2.4GHz Industry Scientific and Medical (ISM) band. This particular range of frequencies is unlicensed. This means that every company is free to use this spectrum in any way they see fit (provided they meet the emission standards for that country!!). This means that the band tends to get crowded. It is not just Wi-Fi. Microwaves, Bluetooth devices and some cordless phones, A/V senders, baby monitors, and wireless cameras etc. can emit noise into this band.
Many installers try to optimise their client’s WiFi installs by strategically choosing a subset of the 2.4Ghz band to communicate over. So how can you do the same?
Let’s start off with some background info. WiFi is broken up into 11 channels along the 2.4Ghz spectrum. As an installer, you can specify that a wireless router send all traffic over channel 6 for example. All people that connect to that router will also communicate on the same channel. What is not commonly known is that Wi-Fi will spread out communications over a few channels – centred on the chosen channel. In the previous example, I chose channel 6. In reality, the wireless router and connected clients will also be using channels 4, 5, 7 and 8 (2 channels on either side of 6). A general rule of thumb is to try and stick to channels 1, 6 and 11 to ensure the least interference between adjoining networks while still maximising the number of potential networks on the one site.
Now that we have that out of the way, how does an installer go about determining what channel to place his equipment on? There are two types of tools that need to be consulted.
- Spectrum analyzer to determine where the sources of noise are
- Wi-Fi scanner to determine what other Wi-Fi networks are running within reception range.
A Spectrum Analyzer basically “listens” to a particular spectrum and displays the results for you to interpret. It is there to tell you how much “power” it sees on each frequency. As an installer, you are trying to identify patterns within the result, and over time you will become skilled in identifying these patterns and associating them with particular devices. Based on the patterns and power on each frequency, you can identify gaps or areas of lower power to place your equipment on. A really affordable spectrum analyzer is Wi-Spy. You can track it down in Australia at Wi-Spy Australia, or from the US at MetaGeek. Chanalyzer is the main software package used with Wi-Spy, and you can view videos of Chanalyzer in action from the link.
The most (in)famous Wi-Fi scanner for Windows is NetStumbler. When running, the software will display the various SSIDs that it finds, and their associated channel number and instantaneous Signal-to-Noise ratio (SNR). Try and place your equipment into a place where no other Wi-Fi channels are being used, or overlap the ones with the lowest SNR if you cannot locate a free channel. Once again, try to stick to channel 1, 6 and 11 if possible. Remember to keep an eye on where the sources of noise are located based on information from the spectrum analyser.
No doubt with these two tools will become a mainstay of your toolbox.
Samer Shami is the distributor of Wi-Spy products in Australia. He has helped clients in Australia with troubleshooting and analysis of their 2.4Ghz spectrums. Feel free to contact him and ask him about spectrum optimisation.
Tags: 2.4ghz, Analysis, chanalyzer, metageek, samer, shami, spectrum, wi-fi, wi-spy, wifi, wispy, zigbee