If the listening station is able to synchronize with one of the colliding frames, and thus initiate a receiving process,
an EIFS will be used (see the reason in Section 4.2). On the other hand, if this is not the case (e.g., comparable
received power level for the two colliding frame preambles), the listening station will just see a busy channel:
without initializing the reception of any frame, and a DIFS will be used. In all the simulation (as well as
analytical) models we are aware of do not enter into this technical issue for simplicity, but consistently use either
DIFS or EIFS for all collisions (for example, the 802.11 ns-2 implementation uses EIFS, and this leads to different
results with respect to other models that use DIFS after a collision).
Performance Study of IEEE 802.11 DCF and IEEE 802.11e EDCA 72
Hence, a discrete and integer time scale can be defined. Note that this discrete time
scale does not directly relate to the system time, being a ???slot??? on the channel either an
empty slot (in which case, the slot will last exactly one slot-time ??), or a busy slot, in
which case the slot duration will depend on the events occurring (a transmission or a
collision).
Pages:
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233