The idea was proposed on Twitter of wireless “Contention Zones”. A wireless “Contention Zone” is the purposeful channel design grouping access points (AP) configured for the same channel in close proximity. The 802.11 (WiFi) protocol uses several mechanisms to determine if the frequency is clear for a transmission opportunity (TxOp) or if the AP station or client station should wait for a period of time. If a wireless station receives a transmission and can decode it, then the station has to wait until that transmission is complete for a TxOp. If the wireless station doesn’t hear another transmission and other criteria are met then the station can transmit. This effectively makes all 802.11 protocol wireless transmissions half-duplex. In short only one station can transmit at a time. All other stations must wait for an opportunity to transmit until the first station has completed sending its data.
Co-Channel Interference is when two or more wireless cells are Tx on the same channel and can receive and demodulate the 802.11 preamble. The separate wireless cells effectively become one large airtime contention domain.
This creates one of the challenges of wireless design, Channel re-use. Designing wireless cells in such a way that a station transmitting on channel 1 in one wireless cell won’t interfere with a different station in a different wireless cell transmitting on channel 1 by buffering wireless cells in-between with AP(s) transmitting on different channels.
This can be easily depicted in a graphic.
Please note that I borrowed this graphic from the WWW. It is representative of many such graphics.
The problem with this graphic is that it doesn’t take into account wireless design of a real world building with walls, desks, doors, and people. All physical objects impact the radio frequency that is being transmitted by the wireless stations. In addition, modern businesses need a robust wireless design with primary and secondary coverage to help ensure that in the event of a failure of an AP won’t mean a loss of wireless service. Three channels is simply not enough to have enough separation between wireless cells transmitting on the same channel.
This is a problem with the 2.4GHz frequency in particular which only has three non-overlapping frequencies. Cisco and Apple both have recommended on not using the 2.4GHz frequency for mission critical wireless service. The reality is that many enterprises rely upon 2.4GHz for mission critical services. This is especially true in the Medical Vertical. So as a wireless engineer, you provide the best design possible minimizing co-channel interference (CCI).
So what about grouping the channels into “Zones”? Creating larger wireless cells that share the same transmit opportunities. Wouldn’t this create a more deterministic experience for the stations? If it was more deterministic and perhaps decreasing collisions speed up the ability of the cell to send and receive radio data? The answer is I don’t know. Not really.
I modeled this idea during the wireless design of a large medical practice. I will go into the design in a separate blog post, but I wanted to show with a computer model the difference between a “honey-comb” wireless design and a “contention zone” wireless design on the 2.4GHz frequency.
The building contains patient rooms with sound deadening material. These special walls attenuated the strength of the wireless signal by 4dBm.
The following png shows the channel coverage with the standard “honey-comb” design.
The following png shows the channel overlap with the standard “honey-comb” design. A wireless station in an orange area can hear four AP(s) transmitting on the same channel.
The following png shows the channel coverage with the “contention zones” design. It sure makes a pretty png!
The following png shows the channel overlap with the “contention zone” design. A wireless station in an orange area can hear four AP(s) transmitting on the same channel.
So by looking at the computer models, it would seem that the “contention zone” design idea would create larger areas of co-channel interference. It would be a less desirable solution at least for this wireless deployment.
It may work in certain small business wireless deployments. What do you think about it?