I am confused by the use of IFS (interframe spacing) in a wireless network. Assuming each station on a network using DCF mode, is required to wait until DIFS has expired before any station can contend for the network. During the CP (Contention Period), which immediately follows DIFS, stations calculate their random back off time and count down the random time. The station with the shortest time gains control of medium first. And here are questions:
1. Suppose the IFS is not use for collision avoidance, why does a station need to wait?
2. In the period (IFS) - what does a station and an AP do?
Thank you for your help.
Interframe spaces are the minimum number of microseconds that a channel must remain clear after the previous transmission ends. In other words, this is how a station can tell when one frame ends and thus another frame can (potentially) start. The channel is idle during this interval.
SIFS (short interface spaces) separate high priority control frames. PIFS (point coordination function interframe spaces) separate management and data frames in PCF mode. DIFS (distributed coordination function interframe spaces) separate management and data frames in DCF mode. These spaces do not by themselves provide collision avoidance ? they are just frame separators.
Collision avoidance depends upon mode. In DCF mode, every station shares responsibility for arbitrating access to the channel. If every station with frames to send were to start transmitting immediately after the DIFS expired, collisions would occur. Therefore, a contention period is applied after DIFS. Each station sets a counter to a random number and starts counting down. It then checks to see if the channel is clear. The station that picks the smallest number will seize and start using the clear channel first. Other stations will continue to wait while the entire cycle repeats itself. In this fashion, stations randomly seize control of the channel. This reduces the chance of collision, while giving all stations their fair share of the channel.