Appeal No. 2006-1729
Application No. 10/107,628
compaction occurs. In the commonly-used signature
analysis [8] time compaction technique, the time
compaction ratio is typically several (> 3) orders.
The row index of the test data matrix. D is referred
to as the space dimension since it corresponds to the
output bits from different circuit nodes resulting from
the application of a single input test pattern. Thus,
if Φ is such that C has its space dimension p < m, then
space compaction occurs. Using the quotient bit of
MISRs [12], [37], parity [7], [27], and MINSRs [1], are
known structures for performing space compaction.
It is possible to have both time and space
compaction performed concurrently. An example of such
a compactor is the well-known MISR.
With the above discussion in mind, we find that with regard
to representative claim 1, the Ivanov reference teaches a built-
in self-test (BIST) space compaction scheme wherein wide multi-
bit input streams are reduced to narrower output bit streams.
Particularly, as depicted in figure 4, Ivanov teaches a general
compaction case where an input test data matrix Dmxn is fed into a
programmable space compactor (PSC) to yield an output matrix Cpxq
where p and q are less than m and n respectively. One of
ordinary skill in the art would have duly recognized that
Ivanov’s teaching of a PSC to transform the input matrix Dmxn into
an output matrix Cpxq of a reduced dimension amounts to the PSC
being characterized as a vector or a matrix (PSCxy) itself, as
required by the claim. It is self-evident from the principles of
9
Page: Previous 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Next
Last modified: November 3, 2007