As a result, it seems that adaptive framed slotted ALOHA protocol is little better
than binary tree protocol. However, as this result gives us only asymptotic property, it is
needed to con?¬?rm that in reality. We conducted a Monte Carlo simulation to make sure
what we have analyzed. Figure 9.8 depicts the measurement of the average slot delay of
the two protocols. The adaptive framed slotted ALOHA protocol with DFSA is assumed as
the optimal one in our simulation. The difference between two protocols is not obvious
when the number of tags is low, but it clearly increases as the number of tags increases. The
reason for this observation is that they converge to the asymptotic average delay values as
Comparative Performance Analysis of Anticollision Protocols in RFID Networks 173
the number of tags is getting bigger. Nevertheless, we cannot determine which protocol is
the best in reality because, in the RFID standards, the frame sizes of probabilistic protocols
are formatted to the powers of two, not integer values. Of course, the protocols with the
powers of two frames will show the same average slot delay asymptotically, but its
converging speed will be fairly degraded.
9.6 Performance Evaluation and Analysis
In this section, we evaluate the performance of tag anticollision protocols examined so far.
We make following assumptions for re?¬‚ecting only the effect on the operational principle
of tag anticollision protocols.
Pages:
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342