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 § 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 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.
Claims 1 – 5, 8 – 12, 15 – 17 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Gokmen et al. (US Pat. No. 10373051), hereinafter referred to as Gokmen in view of (Memristive Field-Programmable Analog Arrays for Analog Computing), hereinafter referred to as Li and in further view of McDonald et al. (US Pat. No. 6772182), hereinafter referred to as McDonald.
Referring to claim 1, Gokmen discloses a cross-point array, comprising:
an array of Resistive Processing Unit (RPU) devices (two-terminal RPU at each crosspoint, Fig. 8; col. 12, lines 13–20);
having rows and columns interconnected at cross-points (conductive row wires and column wires forming crosspoints, Fig. 8; col. 12, lines 14–20);
wherein the RPU devices receive a finite duration input voltage on the rows and output a current on the columns (input voltages V1, V2, V3 applied to row wires; output current summed on column wires, Fig. 8; col. 12, lines 30–40).
However, Gokmen does not disclose: an input-signal matched filter coupled to each of the columns to reduce noise in the current in accordance with the finite duration input voltage.
Li discloses what Gokmen lacks: an input-signal matched filter (The memFPAA based first-order low/high-pass filter, analog filter description, Fig. 3(f), page 4).
Note: Li describes analog filters implemented in a memristive signal path using low-pass or high-pass configurations. These filters are designed to align with expected input waveforms and suppress irrelevant signal content, providing functionality equivalent to a matched filter.
coupled to each of the columns to reduce noise in the current in accordance with the finite duration input voltage (memristors configured as low/high-pass filters attenuate unwanted signal components per path, enhancing signal-to-noise ratio, Fig. 3(f), page 4)
Note: The filters are placed in the analog signal path, such that each signal column can be individually conditioned. This reduces noise according to the shape of finite-duration pulses.
Gokmen and Li are analogous art because they both address analog signal processing and computation within cross-point or in-memory computing systems, particularly in architectures using resistive/memristive elements for matrix operations and waveform-based computation.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Gokmen and Li before him or her, to modify the output of each column in Gokmen’s RPU array to include an analog filter as disclosed by Li.
This modification would improve the signal-to-noise ratio of the output currents by tailoring the signal path to match the expected waveform characteristics — a benefit well-documented in Li.
The suggestion/motivation for doing so would have been to reduce analog noise and increase signal reliability, especially when performing vector-matrix multiplication using resistive cross-point arrays, as shown in both references (Li, Fig. 3(f), page 4).
Therefore, it would have been obvious to combine Li with Gokmen to obtain the invention as specified in the instant claim.
McDonald discloses, what Gokmen lacks, wherein the matched filter reduces noise in current output (matched‑phase noise filter output, Fig. 1) from the RPU devices in accordance with a signal template provided by an input voltage signal (signal summation and conditioning logic, Fig. 3).
Gokmen and McDonald are analogous art because they are from the same field of endeavor, namely resistive cross-point array computing and analog signal processing, both directed at improving signal integrity in neuromorphic or analog computation systems.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Gokmen and McDonald before him or her, to modify the output signal path of each column in Gokmen’s RPU cross-point array to include the matched-phase noise filter architecture disclosed in McDonald.
The suggestion/motivation for doing so would have been to reduce analog noise and increase the accuracy of vector-matrix multiplication operations by aligning filter characteristics with input voltage profiles, improving output fidelity.
Therefore, it would have been obvious to combine McDonald with Gokmen to obtain the invention as specified in the instant claim.
As to claim 2, Gokmen discloses the cross-point array of claim 1, wherein the finite duration input voltage includes a discretized signal (digital pulses of controlled duration applied to word lines, pulse-driven signal input, Fig. 8; col. 11, lines 15–19, Gokmen).
Note: Gokmen applies discrete voltage pulses on the row lines of the cross-point array. These time-controlled pulses are discretized signals used for weight updates and computation.
As to claim 3, Gokmen discloses the cross-point array of claim 1, wherein the finite duration input voltage includes an analog signal (analog write/read pulses may also be used for computation, signal flexibility, col. 11, lines 19–22).
Note: Gokmen describes that the system may accept analog signals in addition to discrete ones, demonstrating support for analog finite-duration input voltages.
As to claim 4, Li discloses the cross-point array of claim 1, wherein the input-signal matched filter includes a finite impulse response filter (FIR filters implemented via memristor configurations for signal shaping, filtering topology, page 4, Li).
Note: Li shows configurable filters implemented in hardware using memristors; these may include FIR architectures, given the discrete structure and behavior described in Fig. 3(f) and surrounding text.
As to claim 5, Li discloses the cross-point array of claim 1, wherein the input-signal matched filter includes a passive bandpass filter (The memFPAA based first-order low/high-pass filter, passive filter types, Fig. 3(f), page 4, Li).
Note: The filter designs shown include purely passive components (memristors, resistors, capacitors), which implement bandpass behavior without amplification, satisfying this limitation.
Claims 8 - 12 recite the corresponding limitation of claims 1 - 5. Therefore, they are rejected accordingly.
Claims 15 – 17 and 19 recite the corresponding limitation of claims 1 - 5. Therefore, they are rejected accordingly.
Claims 6, 7, 13, 14, 18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Gokmen in view of (Memristive Field-Programmable Analog Arrays for Analog Computing), hereinafter referred to as Li and in further view of McDonald et al. (US Pat. No. 6772182), hereinafter referred to as McDonald and of Ray (US 3986127).
As to claim 6, Ray discloses what the modified system Gookmen lacks, the cross-point array of claim 1, wherein the input-signal matched filter includes an active bandpass filter (said first output terminal providing an integrator output signal and a bandpass filter output, active bandpass stage using op-amps, Fig. 1; col. 2, lines 1–9).
Gokmen and Ray are analogous art because both references are directed to analog signal processing within electrical or computing circuits.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Gokmen and Ray before them, to modify Gokmen’s RPU-based output columns to include an active bandpass filter as taught in Ray.
This modification would have enabled selective noise suppression and signal shaping on the analog output current in accordance with expected waveform characteristics, a standard technique in analog signal processing and consistent with the goal of improved signal fidelity in vector-matrix computing.
The suggestion/motivation for doing so would have been to improve signal-to-noise ratio and extract desired frequency content from analog output signals in a computing or sensing environment, as taught in Ray (col. 2, lines 1–20).
Therefore, it would have been obvious to combine Ray with Gokmen to obtain the invention as specified in the instant claim.
As to claim 7, Ray discloses the cross-point array of claim 1, further comprising integrators coupled to the input- signal matched filters along the columns (second output terminal providing a double integrator output signal, integrator stage, col. 4, lines 1–6; Fig. 1).
Claims 13 and 14 recite the corresponding limitation of claims 6 and 7. Therefore, they are rejected accordingly.
Claims 18 and 20 recite the corresponding limitation of claims 6 and 7. Therefore, they are rejected accordingly.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JUANITO C BORROMEO whose telephone number is (571)270-1720. The examiner can normally be reached on Monday - Friday 9 - 5.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Henry Tsai can be reached on 5712724176. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/J.C.B/ Assistant Examiner, Art Unit 2184
/HENRY TSAI/ Supervisory Patent Examiner, Art Unit 2184
Prosecution Insights