We conclude that a slot
immediately following a successful transmission cannot be used for transmissions by any
other station, except the transmitting one. Hence, in ideal channel conditions, such a
transmission is granted to be successful as no collision may occur.
The described effect can be accounted in the model by redefining the notion of
successful transmission slot, and specifically by including either i) the extra slot-time at the
end of a transmission, as well as ii) the possible extra frames transmitted in the ???reserved???
slot. In formulae:
, (39) ?? ?? +
+
=
+
??· ??·??? ??¶
??¬ ??¬??
??«
+
+ = ??‘???
= min
min
1 min
1
1
1
CW
CW T T
CW
T T s
k
s
k
s s
where Ts is the successful slot size given in Eqs. (19) and (20) for the Basic and RTS/CTS
cases respectively, which accounts for a single frame transmission plus a DIFS. To be
consistent, in the throughput computation, the amount of information transmitted into a
successful slot shall also include the MPDU payload due to extra frames transmitted in
such ???reserved??? slots:
Performance Study of IEEE 802.11 DCF and IEEE 802.11e EDCA 82
min
min
CW
CW ] P [ E ] P [ E 1 +
= (40)
Since a successful ???transmission slot??? now includes an extra idle slot, a further model
detail is that, in the slot immediately after such a transmission slot, the backoff counter will
be found in the range (0, CWmin-1) and not in the range (0, CWmin).
Pages:
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254