Thus it is not unreasonable to expect that the tags would lie in one of the three
Cartesian planes but with an arbitrary orientation within that plane, although this would
certainly not be true if cases are not stacked in some uniform fashion as they move through
the portal. On the other hand, in item-level applications it is not possible to accurately
predict how the product will lie and the tags can thus be expected to randomly take on
many different orientations. Even in some pallet-level retailer applications, cases with
RFID tags are sometimes carried through the portal by staff, therefore strictly limiting
the orientations of tags is not a good idea. To determine the 100a% read accuracy region
it is necessary to be able to represent and evaluate the readability of all of these different
possible tag orientations. If we discretize all possible orientations into M discrete unit
vectors from the location of the tag center, then if there is no bias toward a speci?¬?c
orientation each of these unit vectors should be uniformly distributed on the surface of a
unit sphere whose center is coincident with the location of the tag center.
The conventional approach is to discretize uniformly around the latitude and the
longitude (e.g., every 38 from 08 to 3608); however, as shown in Figure 10.6, this approach
Maximizing Read Accuracy by Optimally Locating RFID Interrogators 191
leads to a biased sample that is highly anisotropic and has a stronger concentration of
directions pointing toward the poles and relatively few directions pointing toward positions
on the equator.
Pages:
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378