Appeal No. 95-1955
Application 08/109,982
meter operation, it is extremely important that such
events be detected and the READ or WRITE operation be
retried. If the incorrect READ or WRITE is determined
to be a result of a problem with the memory device, the
data must be reconstructed, for instance in the case of
a failed read, and then relocated to a different part
of the non-volatile memory.
Page 2, lines 10-13
Also, as brought out in the teaching of [another]
application, each data item stored in the block
requires its own CRC for determination of whether the
register data is corrupt.
The foregoing sets up the context in which memory fault
corruptions are detected or determined as a part of each memory
READ and WRITE operation, and in which corruption is determined
on the basis of calculating the corresponding CRC.
As is described on page 15 and shown in the flow chart on
Figure 10A, the READ is "tested" at decision block 1280, after
the READ operation takes place in block 1270. Similarly, as is
described on page 17 and shown in the flow chart in Figure 11,
the WRITE is tested at decision block 1470, after the WRITE
operation takes place in block 1450. One with ordinary skill in
the art would understand that the testing referred to concerns
the detection of memory fault or corruption and thus relates to
the generation of the CRC. Evidently, the appellant argues the
same. (Br. page 6, line 30 to page 7, line 2).
Figure 10A reveals that the READ operation is not performed
-7-
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Next
Last modified: November 3, 2007