The key institution behind this approach is that in most applications
employing RFID tags, the set of objects encountered in successive readings from a particular
reader does not change substantially and information from one reading can be used for
the next.
8.5 Adaptive Query Splitting
AQS uses reader??™s queries and tag IDs analogous to the query tree protocol. The reader
transmits a query including a bit string. The tag responds to a query with its ID, if the
pre?¬?x of its ID is equal to the bit string of the query, that is, r1r2 . . . rx??q1q2 . . . qx where
the tag ID is r1r2 . . . rb (ri is the ith binary value of the ID and b is the total number of bits
of the ID) and the bit string of the query is q1q2 . . . qx (qi is the ith binary value of the
query, 1 x b). Tags are memoryless because they do not maintain any information
except their own IDs.
The reader has queue Q and candidate queue CQ to make queries. Queue Q maintains
bit strings for queries in the current identi?¬?cation frame. Candidate queue CQ
maintains bit strings for queries in the next identi?¬?cation frame. The reader uses the bit
strings stored in CQ as the starting point of tag identi?¬?cation in the next frame. The
starting point of tag identi?¬?cation moves downward (toward descendants in the tree) by
the query insertion procedure and moves upward (toward the root of the tree) by the
query deletion procedure.
Pages:
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295