When designing a new wireless network, RF power levels are often overlooked or misunderstood. Depending on the type of deployment, power levels can have a great impact on client performance and reliability.
The strange thing is that the allowed power levels on different channels are not all the same. This seems pretty strange to the newcomer; and different regulatory domains enforce different power caps on different channels.
Cisco Prime displays the APs from power 1 (maximum) to their minimum powers – its just that one AP might be running at a much higher power than the AP next to it, which is on a different channel. Taking this into account it means that if you don’t use some sort of power caps you’ll have APs which are running at a much higher power than the clients can transmit back at – and that leads to one way transmission problems.
Understanding that there are different power levels on different channels means that you should plan your network to run at the weakest power across your APs. Further, if you’re looking to take advantage of adaptive algorithms such as Transmit Power Control (TPC), you should allow for some spare capacity again.
In practice, yes this will mean you have more APs. It does however mean that your RF powers on your APs will be smoother, and you can support a larger number of clients.
If you don’t take into account that powers are different on different channels, you’ll have a disparity of powers running on your APs and find that you have sticky client issues and clients dropping out.