DETAILED ACTION
This communication is in response to the application filed 11/10/23 in which claims 1-20 were presented for examination.
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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/10/23 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or non-obviousness.
Claims 1, 4-6, 8-12, 15-20 are rejected under 35 U.S.C. 103 as being unpatentable over Etter (US 5,673,210; published Sep. 30, 1997) in view of Cui, Wenqi, Weiwei Yang, and Baosen Zhang. "A frequency domain approach to predict power system transients." IEEE Transactions on Power Systems 39.1 (2023): 465-477 (“Cui”) and Tang (US 2023/0298611 A1; published Sep. 21, 2023).
Regarding claim 1, Etter discloses [a] computer-implemented method, comprising:
monitoring, by a processor set, a target system to collect at least one data metric; (Etter 11:18-22 (“FIG. 4 is a hardware block diagram showing a preferred embodiment of a signal interpolator equipped to perform the interpolation methods disclosed herein. An A/D converter 201 converts an analog input signal into a digitized source signal.”))
pre-processing, by the processor set, the at least one data metric as a seed based on a predetermined policy; (Etter 6:16-30 (“A first embodiment disclosed herein provides a novel iterative signal restoration technique that performs interpolation based upon first and second autoregressive models. The first autoregressive model represents a first known set of samples preceding a missing set of samples, and therefore may be conceptualized as a left-sided autoregressive model. The second autoregressive model represents a second known set of samples succeeding a missing set of samples, and therefore may be conceptualized as a right-sided autoregressive model. Autoregressive models are advantageously employed in the context of signal restoration to provide restored signals having smaller interpolation errors than exist using prior art techniques. In general, autoregressive models utilize a plurality of autoregressive parameters. These parameters are defined as including any set of mathematical variables and/or constants employed for the purpose of modeling a time-dependent function.”)).
Although Etter 6:47-50 teaches that “Autoregressive parameters contain information for computing various characteristics of the original time-dependent function (i.e., the original signal), including, for example, the frequency spectrum of this function,” Etter does not expressly disclose:
encoding, by the processor set, the pre-processed seed using a transform; (but see Cui Section IV B (“For the input of each layer, we conduct discrete Fourier transform F to convert the input trajectory into the frequency domain [28].”); see also FIG. 6:
PNG
media_image1.png
532
708
media_image1.png
Greyscale
)
post-processing, by the processor set, the encoded seed in a frequency domain; (but see Cui Section IV B (“Inspired by the work in [22], we use neural networks parameterized by θj to learn in the frequency domain in each layer j, and then recover the time-domain sequences by inverse Fourier transform F−1. This process is defined as Fourier neural operator Kθj (·) represented by:
PNG
media_image2.png
56
544
media_image2.png
Greyscale
where the function ψ(·) is a low-pass filter that truncates the Fourier series at a maximum number of modes kmax for efficient computation [22]. Then θj is the weight tensor that conducts a linear combination of the modes in the frequency domain.”))
generating, by the processor set, synthetic metrics data by applying an inverse transform to the post-processed seed; and (but see Cui Section IV B (“The output of the j-th layer adds up Fourier neural operator with the initial time-domain sequence weighted by Wj to recover aperiodic and high-frequency components
PNG
media_image3.png
58
572
media_image3.png
Greyscale
where σ is a nonlinear activation function whose action is defined component-wise.”))
training, by the processor set, an artificial intelligence (AI) model using the generated synthetic metrics data (but see Cui FIG. 6 (output of the j-th layer is used as input to the j+1-th layer and updates the trainable weights Wj+1 in the j+1-th layer:
[AltContent: oval]
PNG
media_image4.png
507
675
media_image4.png
Greyscale
).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter to incorporate the teachings of Cui to apply learning to power system transients in the frequency domain, at least because doing so would enable leveraging the fact that system behavior is complex in the time domain but there are relatively few dominant modes in the frequency domain. See Cui Abstract.
Etter and Cui do not expressly disclose a processor set (but see Tang ¶ 119 (“The functionality described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), and the like.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Tang to execute the signal reconstruction method on hardware logic at least because doing so would enable the use of the method in conjunction with an instruction execution system.
Claim 12 is a computer-readable medium claim corresponding to claim 1 and, therefore, is similarly rejected. Etter and Cui do not expressly disclose [a] computer program product comprising one or more computer readable storage media having program instructions collectively stored on the one or more computer readable storage media (but see Tang ¶ 121 (“In the context of this subject matter described herein, a machine-readable medium may be any tangible medium that may contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Tang to provide the signal reconstruction method on a computer readable medium, at least because doing so would enable the use of the method in conjunction with an instruction execution system.
Regarding claim 4, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter does not expressly disclose wherein the at least one data metric comprises system behavior and characteristics of the target system (but see Cui Abstract (“The dynamics of power grids are governed by a large number of nonlinear differential and algebraic equations (DAEs). To safely operate the system, operators need to check that the states described by these DAEs stay within prescribed limits after various potential faults.”)).
The rationale for combining Etter and Cui is the same as set forth above.
Claim 15 is a CRM claim corresponding to claim 4 and, therefore, is similarly rejected.
Regarding claim 5, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter further discloses wherein the pre-processing the at least one data metric as the seed comprises filling a plurality of gaps between a plurality of metric datasets captured in different time windows of the at least one data metric (Etter 6:16-30 (“A first embodiment disclosed herein provides a novel iterative signal restoration technique that performs interpolation based upon first and second autoregressive models. The first autoregressive model represents a first known set of samples preceding a missing set of samples, and therefore may be conceptualized as a left-sided autoregressive model. The second autoregressive model represents a second known set of samples succeeding a missing set of samples, and therefore may be conceptualized as a right-sided autoregressive model. Autoregressive models are advantageously employed in the context of signal restoration to provide restored signals having smaller interpolation errors than exist using prior art techniques. In general, autoregressive models utilize a plurality of autoregressive parameters. These parameters are defined as including any set of mathematical variables and/or constants employed for the purpose of modeling a time-dependent function.”)).
Claim 16 is a CRM claim corresponding to claim 5 and, therefore, is similarly rejected.
Regarding claim 6, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter does not expressly disclose wherein the encoding the pre-processed seed using the transform comprises encoding the pre-processed seed using a Fourier transform (but see Cui Section IV B (“For the input of each layer, we conduct discrete Fourier transform F to convert the input trajectory into the frequency domain [28].”).
Claim 17 is a CRM claim corresponding to claim 6 and, therefore, is similarly rejected.
Regarding claim 8, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter does not expressly disclose: wherein the post-processing the encoded seed in the frequency domain comprises predicting at least one frequency component based on at least one existing frequency component to determine potential periodic patterns and predict future trends of the at least one data metric (but see Cui Section IV B (“The output of the j-th layer adds up Fourier neural operator with the initial time-domain sequence weighted by Wj to recover aperiodic and high-frequency components
PNG
media_image3.png
58
572
media_image3.png
Greyscale
where σ is a nonlinear activation function whose action is defined component-wise.”)).
Claim 18 is a CRM claim corresponding to claim 8 and, therefore, is similarly rejected.
Regarding claim 9, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter does not expressly disclose evaluating a result of the pre-processed seed including the at least one data metric (but see Cui Section IV A (“We construct the structure of neural network shown in Fig. 6, which consists of several Fourier layers for learning in the frequency domain. The input trajectory is first passed through an encoder to integrate the spatial-temporal relationships and the fault information. The encoded data is then passed through l level of Fourier Layers [22], where the input of the j-th layer is ggj and the output is ggj+1 for j=1,…,l. Each Fourier layer consists of one path with trainable weights θθj that learns periodic components in the frequency domain, and another path with trainable weights WWj that directly operate on the time-domain data. Intuitively, the second path (often called the pass-through layer in Machine Learning literature) with weights WWj helps to keep the track of aperiodic and high-frequency components.”)).
Etter is combinable with Cui for the same reasons as set forth above.
Claim 19 is a CRM claim corresponding to claim 9 and, therefore, is similarly rejected.
Regarding claim 10, Etter, in view of Cui, discloses the invention of claim 9 as discussed above. Etter does not expressly disclose wherein the evaluating the result of the pre-processed seed including the at least one data metric comprises evaluating the result of the pre-processed seed including the at least one data metric based on a frequency contribution (but see Cui Section IV A (“We construct the structure of neural network shown in Fig. 6, which consists of several Fourier layers for learning in the frequency domain. The input trajectory is first passed through an encoder to integrate the spatial-temporal relationships and the fault information. The encoded data is then passed through l level of Fourier Layers [22], where the input of the j-th layer is ggj and the output is ggj+1 for j=1,…,l. Each Fourier layer consists of one path with trainable weights θθj that learns periodic components in the frequency domain, and another path with trainable weights WWj that directly operate on the time-domain data. Intuitively, the second path (often called the pass-through layer in Machine Learning literature) with weights WWj helps to keep the track of aperiodic and high-frequency components.”)).
Etter is combinable with Cui for the same reasons as set forth above.
Regarding claim 11, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter does not expressly disclose defining a metric payload template from the at least one data metric (but see Cui Section V Encoding Spatial-Temporal Relationships (“We construct 3D tensors to encode the input trajectories such that the spatial-temporal relationships in the power system can be included.”)).
Etter is combinable with Cui for the same reasons as set forth above.
Claims 2, 3, 13, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Etter, Cui, and Tang as applied to claims 1 and 12 above, and further in view of Gupta (US 10,534,799 B1; published Jan. 14, 2020).
Regarding claim 2, Etter, in view of Cui and Tang, discloses the invention of claim 1 as discussed above. Etter and Cui do not expressly disclose: capturing a plurality of labels and values in the pre-processed seed; and applying the captured labels and values to the generated synthetic metrics data by including the captured labels and values in the generated synthetic metrics data (but see Gupta 3:8-17 (“FIG. 1A is a block diagram of a behavior detection module 110 according to one embodiment. The behavior detection module 110 receives a training database, applies transformations to attributes of the training database, interpolates missing values in the transformed training database, and constructs a classifier that labels a data entry into one among a set of two or more classification labels based on the interpolated training database, and is one means for doing so. The training database is a collection of data entries, in which each data entry is labeled with a classification label.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Gupta to label a data entry obtained by interpolation, at least because doing so would enable using the interpolated data as training data.
Claim 13 is a CRM claim corresponding to claim 2 and, therefore, is similarly rejected.
Regarding claim 3, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter and Cui do not expressly disclose:
capturing a plurality of logs and traces in the pre-processed seed; and applying the captured logs and traces to the generated synthetic metrics data by including the captured logs and traces in the generated synthetic metrics data (but see Gupta (US 10,534,799 B1; published Jan. 14, 2020) 3:8-17 (“FIG. 1A is a block diagram of a behavior detection module 110 according to one embodiment. The behavior detection module 110 receives a training database, applies transformations to attributes of the training database, interpolates missing values in the transformed training database, and constructs a classifier that labels a data entry into one among a set of two or more classification labels based on the interpolated training database, and is one means for doing so. The training database is a collection of data entries, in which each data entry is labeled with a classification label.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Gupta to label a data entry obtained by interpolation, at least because doing so would enable using the interpolated data as training data.
Claim 14 is a CRM claim corresponding to claim 3 and, therefore, is similarly rejected.
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Etter, Cui, and Tang as applied to claim 1 above, and further in view of Dubuc (US 7,251,291 B1; published Jul. 31, 2007).
Regarding claim 7, Etter, in view of Cui, discloses the invention of claim 1 as discussed above. Etter does not expressly disclose wherein the encoding the pre-processed seed using the transform comprises encoding the pre-processed seed using a Wavelet transform (but see Dubuc 19:39-53 (“The time to frequency domain converter 420 is electrically coupled to the first analog to digital (A/D) converter 418 and the second analog to digital (A/D) converter 416 and is adapted to generate a receive signal stream 421, representing data as a digital signal in the frequency domain, responsive to receiving the first time-based receive signal 419 and the second time-based receive signal 417, as is well known to those skilled in the relevant art. Preferably, the time to frequency domain converter 420 employs a Fourier transform, but may alternatively employ an discrete cosine transform, an wavelet transform, and the like, each being well known to those skilled in the relevant art. More particularly, the time to frequency domain converter 420 employs a fast Fourier transform (FFT), as is well know to those skilled in the relevant art.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Dubue to use a wavelet transform instead of a Fourier transform for the time to frequency domain transformation, at least because a wavelet transform also converts the signal to the frequency domain.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Etter, in view of Cui, Gupta, and Tang.
Regarding claim 20, Etter discloses [a] system comprising:
monitor a target system to collect at least one data metric; (Etter 11:18-22 (“FIG. 4 is a hardware block diagram showing a preferred embodiment of a signal interpolator equipped to perform the interpolation methods disclosed herein. An A/D converter 201 converts an analog input signal into a digitized source signal.”))
pre-process the at least one data metric as a seed based on a predetermined policy; (Etter 6:16-30 (“A first embodiment disclosed herein provides a novel iterative signal restoration technique that performs interpolation based upon first and second autoregressive models. The first autoregressive model represents a first known set of samples preceding a missing set of samples, and therefore may be conceptualized as a left-sided autoregressive model. The second autoregressive model represents a second known set of samples succeeding a missing set of samples, and therefore may be conceptualized as a right-sided autoregressive model. Autoregressive models are advantageously employed in the context of signal restoration to provide restored signals having smaller interpolation errors than exist using prior art techniques. In general, autoregressive models utilize a plurality of autoregressive parameters. These parameters are defined as including any set of mathematical variables and/or constants employed for the purpose of modeling a time-dependent function.”)).
Although Etter 6:47-50 teaches that “Autoregressive parameters contain information for computing various characteristics of the original time-dependent function (i.e., the original signal), including, for example, the frequency spectrum of this function,” Etter does not expressly disclose:
encode the pre-processed seed using a transform; (but see Cui Section IV B (“For the input of each layer, we conduct discrete Fourier transform F to convert the input trajectory into the frequency domain [28].”); see also FIG. 6:
PNG
media_image1.png
532
708
media_image1.png
Greyscale
)
post-process the encoded seed in a frequency domain; (but see Cui Section IV B (“Inspired by the work in [22], we use neural networks parameterized by θj to learn in the frequency domain in each layer j, and then recover the time-domain sequences by inverse Fourier transform F−1. This process is defined as Fourier neural operator Kθj (·) represented by:
PNG
media_image2.png
56
544
media_image2.png
Greyscale
where the function ψ(·) is a low-pass filter that truncates the Fourier series at a maximum number of modes kmax for efficient computation [22]. Then θj is the weight tensor that conducts a linear combination of the modes in the frequency domain.”))
generate synthetic metrics data by applying an inverse transform to the post-processed seed; (but see Cui Section IV B (“The output of the j-th layer adds up Fourier neural operator with the initial time-domain sequence weighted by Wj to recover aperiodic and high-frequency components
PNG
media_image3.png
58
572
media_image3.png
Greyscale
where σ is a nonlinear activation function whose action is defined component-wise.”)).
train an artificial intelligence (AI) model using the generated synthetic metrics data (but see Cui FIG. 6 (output of the j-th layer is used as input to the j+1-th layer and updates the trainable weights Wj+1 in the j+1-th layer:
[AltContent: oval]
PNG
media_image4.png
507
675
media_image4.png
Greyscale
).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter to incorporate the teachings of Cui to apply learning to power system transients in the frequency domain, at least because doing so would enable leveraging the fact that system behavior is complex in the time domain but there are relatively few dominant modes in the frequency domain. See Cui Abstract.
Etter and Cui do not expressly disclose: capture a plurality of labels and values in the pre-processed seed; capture a plurality of logs and traces in the pre-processed seed; apply the plurality of labels, values, logs, and traces to the generated synthetic metrics data; and (but see Gupta 3:8-17 (“FIG. 1A is a block diagram of a behavior detection module 110 according to one embodiment. The behavior detection module 110 receives a training database, applies transformations to attributes of the training database, interpolates missing values in the transformed training database, and constructs a classifier that labels a data entry into one among a set of two or more classification labels based on the interpolated training database, and is one means for doing so. The training database is a collection of data entries, in which each data entry is labeled with a classification label.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Gupta to label a data entry obtained by interpolation, at least because doing so would enable using the interpolated data as training data.
Etter does not expressly disclose a processor set, one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to: (but see Tang ¶ 119 (“The functionality described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), and the like.”), ¶ 121 (“In the context of this subject matter described herein, a machine-readable medium may be any tangible medium that may contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.”)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Etter and Cui to incorporate the teachings of Tang to provide the signal reconstruction method on a computer readable medium, at least because doing so would enable the use of the method in conjunction with an instruction execution system.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Godsill, Simon J., and Peter JW Rayner. "Frequency-based interpolation of sampled signals with applications in audio restoration." 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol. 1. IEEE, 1993
C. T. Leondes and D. D. Rivers, "Frequency Domain Interpolation," in IEEE Transactions on Aerospace and Electronic Systems, vol. AES-13, no. 3, pp. 323-327, May 1977
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHAHID KHAN whose telephone number is (571)270-0419. The examiner can normally be reached M-F, 9-5 est.
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, Usmaan Saeed can be reached at (571)272-4046. 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.
/SHAHID K KHAN/Primary Examiner, Art Unit 2146
Prosecution Insights