Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 6-9 and 11-12 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding Claim 6, it recites “wherein the raw data includes the same size of wafer bin maps” (lines 1-2). It is unclear what “the same size” means in this context. The claim only recites “wafer bin maps,” but it is not clear what these are being compared to in order to determine that the size is the same. The present claim further recites “in the case of the first tuning parameter and the second tuning parameter . . .” However, there is no condition or value placed upon the first tuning parameter and the second tuning parameter. So it cannot be determined what “case” is being considered. Is it simply the “case” in which both parameters exist? The term “in the case of” implies that the parameters must meet certain criteria or take on certain values for the case to be true. Furthermore, the term “the second tuning parameter” lacks antecedent basis. A second tuning parameter is introduced in claim 4, but the present claim depends on claim 2 (and claim 1); claim 4 is not in the scope of the present claim. Finally, the claim recites “a single tuning parameter is determined for all wafers through grid search using at least a portion of the raw data.” It is unclear how a grid search can determine a tuning parameter. For all of these reasons, the examiner is unable to make a meaningful interpretation of the present claim for examination under prior art.
Regarding Claim 7, it depends on claim 6, so it is indefinite and cannot be examined with respect to prior art for the same reasons. Claim 7 also recites “the second tuning parameter,” which lacks antecedent basis as described for claim 6.
Regarding Claim 8, it depends on claim 6, so it is indefinite and cannot be examined with respect to prior art for the same reasons. Claim 8 also recites “the second tuning parameter” (line 3), which lacks antecedent basis as described for claim 6.
Regarding Claim 9, it recites elements substantially similar to those of claim 6, so it is indefinite for the same reasons, and it too cannot be examiner under prior art because it cannot be interpreted in a meaningful way.
Regarding Claim 11, it recites “wherein the method for semiconductor wafer pattern detection and classification uses an orthogonal and polar coordinate system in a two-dimensional space to quantify a feature of a single wafer pattern and uses the same . . .” It is unclear what “the same” refers to. Is it the orthogonal coordinate system? Or the polar coordinate system” Or both? The claim further recites “ . . . by including a feature . . .” It is unclear what the feature is included in. The wording suggests that the feature is included in the “orthogonal and polar coordinate system,” but it does not make sense for a feature to be included in a coordinate system. A feature may be used for classification learning and prediction, but it doesn’t really make sense for a feature to be “included” in a learning process. Because of the many indefinite terms and concepts, the examiner is only able to interpret the present claim to mean “wherein the method for semiconductor wafer pattern detection and classification uses an orthogonal and polar coordinate system in a two-dimensional space.”
Regarding Claim 12, it recites “ . . . and uses the same for failure classification learning and prediction.” However, the claim previously recites many elements, so it is unclear what “the same” refers to, and therefore which element(s) are used for failure classification learning and prediction. For the purposes of examination under prior art, the examiner is unable to make a meaningful interpretation of the phrase “and uses the same for failure classification learning and prediction.”
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 1 is rejected under 35 U.S.C. 103 as being unpatentable over M. B. Alawieh, F. Wang and X. Li, (“Identifying Wafer-Level Systematic Failure Patterns via Unsupervised Learning,” in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 37, no. 4, pp. 832-844, April 2018; hereinafter “Alawieh”) in view of Rundo et al. (U.S. 2024/0202908, hereinafter “Rundo”).
Regarding Claim 1, Alawieh teaches a method for semiconductor wafer pattern detection and classification using hierarchical clustering (HC), performed by a computing device comprising at least a processor (Abstract), the method comprising:
acquiring raw data (sections III and III. A—raw data are received to perform clustering. Section II further describes previous work that receives raw data);
detecting a failure pattern of a target semiconductor wafer (section II, first two paragraphs—a failure pattern is detected by comparing a target wafer to clusters of failure patterns); and
classifying the failure pattern of the target semiconductor wafer (section II, first two paragraphs—the failure pattern is classified by comparing a target wafer to clusters of failure patterns), wherein the detecting of the failure pattern comprises:
removing a cluster that does not satisfy a first tuning parameter through the hierarchical clustering (HC) (section III. B—hierarchical clustering merges clusters that are within a defined distance, effectively removing a cluster by merging it with another cluster. The distance is a first tuning parameter).
Alawieh does not specifically teach determining that the target semiconductor wafer is normal when the cluster that satisfies the first tuning parameter is absent.
However, Rundo teaches:
detecting a failure pattern of a target semiconductor wafer (fig. 4; ¶ [0110]—classifier circuitry detects a failure pattern of a target semiconductor wafer to classify it as one of the defect pattern types); and
classifying the failure pattern of the target semiconductor wafer (¶ [0111] – [0115]); and
determining that the target semiconductor wafer is normal when the cluster that satisfies the first tuning parameter is absent (¶ [0084] – [0085]—wafers that do not have a defect pattern that matches one of the clusters is determined to be normal).
All of the claimed elements were known in Alawieh and Rundo and could have been combined by known methods with no change in their respective functions. It therefore would have been obvious to a person of ordinary skill in the art at the time of filing of the applicant’s invention to combine the classifying and determining that a wafer is normal of Rundo with the clustering and classifying of Alawieh to yield the predictable result of detecting a failure pattern of a target semiconductor wafer; and classifying the failure pattern of the target semiconductor wafer, wherein the detecting of the failure pattern comprises: removing a cluster that does not satisfy a first tuning parameter through the hierarchical clustering (HC); and determining that the target semiconductor wafer is normal when the cluster that satisfies the first tuning parameter is absent. One would be motivated to make this combination for the purpose of improving wafer manufacturing by detecting production process drift and other problems that cause defects in wafers (Rundo, ¶ [0059]).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Alawieh in view of Rundo, as applied to claim 1, above, and further in view of C. H. Jin, H. J. Na, M. Piao, G. Pok and K. H. Ryu, (“A Novel DBSCAN-Based Defect Pattern Detection and Classification Framework for Wafer Bin Map,” in IEEE Transactions on Semiconductor Manufacturing, vol. 32, no. 3, pp. 286-292, Aug. 2019; hereinafter “Jin”).
Regarding Claim 2, Alawieh/Rundo teaches wherein a first tuning parameter includes a first minimum allowable distance between clusters (Alawieh, section III. B), but does not specifically teach wherein the first tuning parameter includes a first height that represents a first minimum allowable distance between clusters and a first minimum number of individuals in a cluster (minPts) to use a hierarchical clustering method in which a density-based de-noising function of spatial data is added.
However, Jin teaches wherein a first tuning parameter includes a first height that represents a first minimum allowable distance between clusters and a first minimum number of individuals in a cluster (minPts) to use a hierarchical clustering method in which a density-based de-noising function of spatial data is added (section III and fig. 5—clusters are based on a distance L between points and a minimum MinPts as a minimum number of individuals in a cluster).
All of the claimed elements were known in Alawieh/Rundo and Jin and could have been combined by known methods with no change in their respective functions. It therefore would have been obvious to a person of ordinary skill in the art at the time of filing of the applicant’s invention to combine the height and minimum number of individuals of Jin with the distance of Alawieh/Rundo to yield the predictable result of wherein the first tuning parameter includes a first height that represents a first minimum allowable distance between clusters and a first minimum number of individuals in a cluster (minPts) to use a hierarchical clustering method in which a density-based de-noising function of spatial data is added. One would be motivated to make this combination for the purpose of improving processing speed by performing outlier detection and defect cluster pattern extraction at the same time (Jin, Abstract and section II. D).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Alawieh in view of Rundo, as applied to claim 1, above, and further in view of Murtagh, Fionn, and Pedro Contreras (“Methods of hierarchical clustering,” arXiv preprint arXiv:1105.0121 (2011); hereinafter “Murtagh”).
Regarding Claim 3, Alawieh/Rundo does not specifically teach wherein the hierarchical clustering (HC) uses a single linkage and a Minkowski (p=3) distance for density-based cluster construction. However, Murtagh teaches wherein hierarchical clustering (HC) uses a single linkage and a Minkowski (p=3) distance for density-based cluster construction (section 2—a traditional way to measure distances is a Minkowski distance. Sections 3 and 4 describe linkages used in hierarchical clustering, including a single linkage).
All of the claimed elements were known in Alawieh/Rundo and Murtagh and could have been combined by known methods with no change in their respective functions. It therefore would have been obvious to a person of ordinary skill in the art at the time of filing of the applicant’s invention to combine the single linkage and Minkowski distance of Murtagh with the distance of Alawieh/Rundo to yield the predictable result of wherein the hierarchical clustering (HC) uses a single linkage and a Minkowski (p=3) distance for density-based cluster construction. One would be motivated to make this combination for the purpose of applying a distance measurement that is necessary in order to decide which elements belong to a group (Murtagh, section 2, first two paragraphs).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Alawieh in view of Rundo, as applied to claim 1, above, and further in view of Geng, Hao, et al. (“When wafer failure pattern classification meets few-shot learning and self-supervised learning,” 2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD). IEEE, 2021; hereinafter “Geng”).
Regarding Claim 10, Alawieh/Rundo does not specifically teach wherein the method for semiconductor wafer pattern detection and classification learns a failure detection status using five or less samples for each failure pattern and 100 or more normal samples. However, Geng teaches a method for semiconductor wafer pattern detection and classification that learns a failure detection status using five or less samples for each failure pattern and 100 or more normal samples (section III—few-shot learning resolves the issue of imbalanced training data by learning from a few failure samples and many normal samples. The recited details of five or less samples for each failure pattern and 100 or more normal samples is an obvious version of the parameters of Geng).
All of the claimed elements were known in Alawieh/Rundo and Geng and could have been combined by known methods with no change in their respective functions. It therefore would have been obvious to a person of ordinary skill in the art at the time of filing of the applicant’s invention to combine the few-shot learning of Geng with the wafer pattern detection and classification of Alawieh/Rundo to yield the predictable result of wherein the method for semiconductor wafer pattern detection and classification learns a failure detection status using five or less samples for each failure pattern and 100 or more normal samples. One would be motivated to make this combination for the purpose of enabling wafer failure detecting when only a few wafer defect images are available (Geng, Abstract).
Allowable Subject Matter
Claims 4-5 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. None of the prior art of record teaches all of the limitations of claim 4, which recites:
4. The method of claim 1, wherein the classifying of the failure pattern comprises:
removing a cluster that does not satisfy a second tuning parameter through the hierarchical clustering (HC) and isolating the cluster into a meaningful individual pattern through the second tuning parameter;
extracting a feature of a cluster that satisfies the second tuning parameter; and
classifying the failure pattern using a pretrained random forest model based on the feature.
Alawieh performs hierarchical clustering using a first tuning parameter, as detailed for claim 1, above, but does not perform an additional step of removing a cluster that removing a cluster that does not satisfy a second tuning parameter through the hierarchical clustering (HC) and isolating the cluster into a meaningful individual pattern through the second tuning parameter; and extracting a feature of a cluster that satisfies the second tuning parameter. None of the prior art of record remedies the deficiencies of Alawieh.
Claim 5 recites allowable subject matter at least by virtue of its dependence on Claim 4.
Claims 9 and 11-12 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims. The present claims recite allowable subject matter by virtue of their dependence on claim 4, and would be allowable if claim 4 were incorporated into claim 1 and if claims 9, 11, and 12 were amended to overcome the rejections under 35 U.S.C. 112 detailed above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure. M. Abdullah, M. H. Rahman and S. Akhter (“Pattern Recognition in Analog Wafermaps with Multiple Ensemble Approaches,” 2021 2nd International Conference on Robotics, Electrical and Signal Processing Techniques (ICREST), DHAKA, Bangladesh, 2021, pp. 587-591) teaches testing of several random forest methods for wafer map classification
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HAL W SCHNEE whose telephone number is (571) 270-1918. The examiner can normally be reached M-F 7:30 a.m. - 6:00 p.m.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michael Huntley can be reached at 303-297-4307. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HAL SCHNEE/Primary Examiner, Art Unit 2129
Prosecution Insights