Failure bitmaps of manufactured memory arrays may contain the information associated to some systematic defects and have hence been used to monitor the process and improve the memory yield. It is very important to develop a method to extract and classify the fault signatures in the failure bitmaps. The fault signatures can be classified into two categories: local fault signatures and global fault signatures. Focusing on the local fault signatures, this paper introduces a supervised one-layer ANN method to solve the signature classification problem. The method is efficient for recognizing the local fault signatures in the failure bitmaps, and more importantly, it has the ability to find new signatures unseen before.
[1]
Larg Weiland,et al.
Advanced Semiconductor Manufacturing Conference
,
2002,
ASMC 2002.
[2]
J. P. Card,et al.
SRAM bitmap shape recognition and sorting using neural networks
,
1995
.
[3]
畠山 一実.
国際会議報告-International Test Conference に見るテスト技術の動向
,
2004
.
[4]
Wojciech Maly,et al.
Test response compression and bitmap encoding for embedded memories in manufacturing process monitoring
,
2001,
Proceedings International Test Conference 2001 (Cat. No.01CH37260).
[5]
Teuvo Kohonen,et al.
The self-organizing map
,
1990
.