On the other hand, from a computational viewpoint, if a coarse resolution
can lead to the same optimal reader placement, then it would be ideal to use such a
resolution for determining the actual placement of the readers in the enumeration procedure
but a ?¬?ner resolution can then be used at the end to obtain a more precise estimate of
the actual coverage percentage obtained by the placement scheme.
Similar to the tag space search resolution, for every point, the coverage for various
orientations should also ideally be calculated in continuous three-dimensional space;
therefore a larger value of M is always preferred. A less-than-ideal value of M could give
rise to two types of errors. Type 1 error occurs when a point for which more than 100a% of
the orientations can be covered with the ?¬?nest resolution is (mistakenly) classi?¬?ed as not
being covered with the smaller value forM. Conversely, a Type 2 error occurs when a point
that does not achieve the minimum coverage of 100a% with the ?¬?nest resolution is
classi?¬?ed as being covered with the coarser resolution. Increasing the value of M reduces
both types of errors; however, to reduce the computational time, the value of M should be
chosen as small as possible without leading to suboptimal or erroneous solutions.
10.3.4 Numerical Examples
In the example shown in Figure 10.8, a portal with dimensions 33333 m has 18
candidate antenna positions on three walls spaced at 0.
Pages:
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383