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 9Page: Previous 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 NextLast modified: November 3, 2007