1. The RAP sends a message to the controller that the channel is infected with radar. The channel is marked as infected on the RAP and on the controller.
2. The RAP blocks the channel for 30 minutes. This 30-minute period is called the nonoccupancy period.
3. The controller sends a TRAP, which indicates that the radar has been detected on the channel. A TRAP remains until the nonoccupancy period expires.
4. The RAP has 10 seconds to move away from the channel. This period is called the channel move time, which is defined as the time for the system to clear the channel and is measured from the end of the radar burst to the end of the final transmission on the channel.
5. The RAP enters the quiet mode. In the quiet mode, the RAP stops data transmissions. Beacons are still generated and probe responses are still delivered. The quiet mode exists until the channel move time is over (10 seconds).
6. The controller picks up a new random channel and sends the channel information to the RAP.
7. The RAP receives the new channel information and sends channel change frames (unicast, encrypted) to the MAP, and each MAP sends the same information to its lower children down the sector. Each mesh access point sends the channel change frames once every 100 msecs for a total of five times.
8. The RAP tunes to the new channel and enters into the silent mode. During the silent mode, only the receiver is ON. The RAP keeps scanning the new channel for any radar presence for 60 seconds. This process is called channel availability check (CAC).
9. The MAP tunes to the new channel and enters into the silent mode. During the silent mode, only the receiver is ON. The MAP keeps scanning the new channel for any radar presence for 60 seconds.
10. If radar is not detected, the RAP resumes full functionality on this new channel and the whole sector tunes to this new channel.

1. The MAP sends a radar seen indication to the parent and ultimately to the RAP indicating that the channel is infected. The RAP sends this message to the controller. The message appears to be coming from the RAP. The MAP, RAP, and controller mark the channel as infected for 30 minutes.
2. The MAP blocks the channel for 30 minutes. This 30-minute period is called the nonoccupancy period.
3. The controller sends a TRAP, which indicates that the radar has been detected on the channel. The TRAP remains until the nonoccupancy period expires.
4. The MAP has 10 seconds to move away from the channel. This is called the channel move time, which is defined as the time for the system to clear the channel and is measured from the end of the radar burst to the end of the final transmission on the channel.
5. The MAP enters the quiet mode. In the quiet mode, the MAP stops data transmissions. Beacons are still generated and probe responses are still delivered. The quiet mode exists until the channel move time is over (10 seconds).
6. The controller picks up a new random channel and sends the channel to the RAP.
7. The RAP receives the new channel information and sends channel change frames (unicast, encrypted) to a MAP, and each MAP sends the same information to its lower children down the sector. Each mesh access point sends the channel change frames once every 100 msecs for a total of five times.
8. Each mesh access point tunes to the new channel and enters into the silent mode. During the silent mode, only the receiver is ON. There is no packet transmission. An AP keeps scanning the new channel for any radar presence for 60 seconds. This process is called the channel availability check (CAC). The MAP should not disconnect from the controller. The network should remain stable during this one-minute period.

DFS functionality allows a MAP that detects a radar signal to transmit that up to the RAP, which then acts as if it has experienced radar and moves the sector. This process is called the coordinated channel change. This functionally can be turned on or off on the controller. The coordinated channel change is enabled by default.

To enable DFS, enter the following command: