Michael Nancarrow's Blog
Understanding Co-Channel Interference
2019-08-25 (Last Update: Sun, 25 Aug 2019) Michael Nancarrow 0 Linux
In this post, we will briefly touch on the topics of Channels, and why having an Access Point in the same channel as another can have a negative impact on your Wireless LAN. For more in-depth analysis of Channels and what to use, you will need to perform your own investigation - the CWNA and CWNP books are a great resource.
Brief Touch on Wireless
In a wireless network, our access points broadcast their SSIDs on a wireless medium (we will call it "the atmosphere"). Each access point has (at a minimum) 1 radio that is used to broadcast the energy for a receiver, and this is performed over a "channel".
Some links that may help you get a better understanding of WiFi channels, and Wireless in general:
Some terminology and theory to get a better understanding of are: DFS Channels, Attenuation and Signal to Noise. These will also help you in understanding what to take into consideration when designing a wireless network.
What are channels?
A wireless channel is a logical identifier us humans use to categorize frequencies - they denote a signal frequency for data to be transmitted over. These signals are not just limited to WiFi - technologies like Microwaves, Baby Monitors and Bluetooth occupy this space as well.
Right, back to WiFi. Within the 2.4Ghz spectrum there are 11 logical channels that we can transmit data over, as follows:
What you will notice is that for Channel 1, the band starts at 2.400 and ends 2.422 - each wireless channel is 20MHz wide with a 2Mhz spread. When a Wireless Network Engineer tells you they are using Channel 1 for an iAP, what that means is an access point is broadcasting between 2.400 and 2.420Mhz.
If you want to get any more in-depth than this, go checkout Radio Frequencies!
Why do we not just use the same channel everywhere?
Using the same Channels on multiple iAPs has a serious impact to Wireless Health, for many reasons. The main item to be concerned with is Clear Channel Assignment ("CCA") - where access points will listen to the spectrum and defer all transmissions if the medium is busy.
Clear Channel Assignment
CCA is a method for Access Points (below mentioned as stations) to send out a probe request to see if there is activity on the channel they operate in. If they note there is activity, they will defer traffic for a time (which can be worked out - but I'll let you locate the formula) until such a time as they are next in line to transfer data.
4 Way Hand Shake
Furthermore, having multiple access points on the same channel can have a large impact when hosts are trying to authenticate and join the network. Once you understand what the 4 Way Hand-Shake is, and the frequency in which it occurs, you can get a better understanding of how it can saturate a network.
For example, if we were to use 802.11b at 1Mbps (yes, I opted for the worst-case scenario to emphasize this) and had 6 iAPs utilising the same channel, each with 2 SSIDs we have already saturated the networks usage for 38.7% - over a third of the time!
Over-utilization of a channel can have large adverse effects on client capacities. In a wireless environment, there is only 1 collision domain per channel. When planning to build your network, make sure you calculate appropriately to give as much airtime to clients as possible, by utilizing more than 1 channel .
Where should we use the same channels for Wireless?
Regardless of whether you're using 2.5 or 5Ghz channels, you will need to re-use channels at some point. This is where the design of a network really comes into play.
Remember when we stated that channel 1, 6 and 11 did not overlap? That's important because in our design, we can rotate through these 3 channels to ensure we have enough coverage at locations, whilst also mitigating collisions. Because Channel 1, 6 and 11 are seperated on the frequency range, they do not impact one another and are all able to transmit at the same time.
Correctly assessing your campuses and planning your channel usage will allow you to cover the area, whilst also allowing for 3 collision domains to be in use at any point in time (not taking into account neighbouring WiFi and their channel usage).
Should we not use the 2.4 spectrum?
We have touched on 2.4 and how we can sort of mitigate issues with smart channel allocations - but, should we still use 2.4? My opinion is that you should not be planning a primary network design on a 2.4Ghz design. The channel is over-utilised by WiFi and non-WiFi devices, and has too limited of a channel range to be used for large-scale deployment.
So, why is 5Ghz a better option? I thought you'd never ask. Introducing, channels. Channels as far as the eye can see!
5Ghz offers more channels (and also DFS - which I will not cover here) with higher throughput capacities, and theoretically less contention on the spectrum. For the short answer, always try to make 5Ghz your primary design, and fall back to 2.4 to cover range.
Attenuation on the Antenna of an iAP will impact the RF propagation. As the rate of attenuation increases, the communication weakens or degrades. Attenuation occurs on computer networks because of:
- Range – over longer distances both wired and wireless transmissions gradually dissipate in strength
- Interference – radio interference or physical obstructions, such as walls, dampen communication signals on wireless networks
- Wire size – thinner wires suffer from more attenuation than thicker wires on wired networks
Attenuation is important in telecommunications and ultrasound applications because it’s critical to determining signal strength as a function of distance. Minimizing the loss of attenuation is important in microwave, wireless and cellular applications because to function correctly an optical data link depends on modulated light reaching the receiver with enough power to be correctly demodulated. This power is reduced through attenuation, resulting in a loss of the light signal that’s being transmitted.
Dynamic Frequency Selection (DFS) , refers to a mechanism to allow unlicensed devices, especially those operating out-door to share the 5GHz frequency bands which have been allocated to radar systems without causing interference to those radars.
When support for DFS is enabled, it will be necessary for WiFi access points to verify that any radar in proximity is not using DFS frequencies. This process is called Channel Availability Check, and it’s executed during the boot process of an access point (AP) as well as during its normal operations.
If the AP detects that a radar is using a particular DFS channel, then it will exclude that channel from the list of available channels. This state will last for 30 minutes, after which the AP will check again if the channel can be used for WiFi transmissions.
The Channel Availability Check performed during the boot process can take anywhere between 1 and 10 minutes, depending on which country you’re in. For this reason, DFS channels are not immediately available when an AP boots.
No comments yet