DETAILED ACTION
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 . Applicant's response filed 6/2/26 canceled claims 11 and 13, leaving claims 1-10, 12 and 14-20 pending.
Response to Arguments
Applicant’s amendment of claim 9 overcame the 112(b) rejection raised in the preceding office action.
Applicant’s arguments with respect to claim 1, 10 and 16 have been considered but are moot because the new ground of rejection rely upon an additional reference to address the newly added claim limitations directed toward the variety of frequencies output by the microring resonator at higher input power.
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.
Claims 1-10, 12 and 14-15 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 Claims 1 & 10, it is unclear what the limitations primary optical mode and primary lasing mode show up nowhere in the specification. Even the word primary on its own shows up nowhere in the specification. For purposes of compact prosecution, the term primary optical mode or primary lasing mode is being interpreted as having the frequency of the input signal.
Regarding Claims 2-9, they are also rejected under 35 USC 112(b) for depending from rejected base claim 1.
Regarding Claims 12 and 14-15, they are also rejected under 35 USC 112(b) for depending from rejected base claim 10.
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 text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
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 nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-9 and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over US PG PUB 20230045938 (hereinafter Carolan) in view of US Patent 6,351,311 (hereinafter Numai) and further in view of Wu et al, “Low-threshold all-optical nonlinear activation function based on a Ge/Si hybrid structure in a microring resonator”.
Regarding Claim 1, Carolan teaches a method of operating a nonlinear activation device (optical non-linearity unit 126, 700, 924), the method comprising:
adjusting a bias applied to an optical source that includes a microring resonator ([0068] describes changing an optical response of a tunable ring resonator, see FIG. 7), the microring resonator comprising
(Carolan is silent as to the internal structure of the tunable ring resonator) and wherein the bias is adjusted to a threshold current to cause the optically active region to transition to stimulated emission operation;
while the optically active region is in stimulated emission operation, receiving a first optical signal at a first wavelength (FIG. 9B shows a first optical signal input to the left of the tunable ring oscillator and a second optical signal different from the first being output), wherein the first optical signal is provided at an optical power equal to or above a threshold optical power to cause formation of one or more secondary optical modes in the optically active region, and wherein the one or more secondary optical modes correspond to gain peaks formed at wavelengths different from a wavelength of a primary optical mode (Carolan is silent as to what specific effect(s) there would be on frequency shift caused by specific changes in optical power);
providing an activation function as at least one of the one or more secondary optical modes (FIG. 9D shows production of a range of nonlinear activation functions and [0072] describes how different functions can be selected by varying the bias voltage being applied to the tunable ring resonator).
As indicated by the strikeout above, Carolan fails to teach claimed internal features of a tunable ring resonator (i.e. optical source) and does not specifically mention injection locking function of the tunable ring resonator.
However, Numai teaches limitations regarding claimed internal features of a tunable ring resonator absent from Carolan. In particular, Numai teaches a ring resonator type laser including “an optically active region (active layer 48, see FIG. 14) positioned between semiconductor layers (cladding layer 46 and Semiconductor substrate 54), wherein the semiconductor layers comprise a Group III-V semiconductor material (46 and 54 are described as being formed from InP at col 17 lines 52-55, which is a Group III-V semiconductor material)”. Numai also teaches how the first optical signal injection locks the optical source (see col 17 lines 55-58).
Carolan and Numai both describe applications of tunable laser ring resonators. Given Carolan’s silence with regards to how a ring resonator is formed a person having ordinary skill in the art at the time of filing would have found it obvious to apply the teachings of Numai to form a ring resonator capable of injection locking based on a received optical signal.
As noted above, Carolan is also silent as to what specific effect(s) there would be on frequency shift caused by specific changes in optical power. However, Wu in FIGS. 6A – 6B and 6C – 6D show how increasing input power causes an identifiable shift in frequency with multiple secondary optical modes appearing as peaks in FIGS. 6B and 6D, corresponding to the two different types of ring resonators being tested. The beginning of section 4.2 at the bottom of p976 makes clear that the frequency/wavelength shift is a result of increasing optical power.
Wu and the combination of Carolan and Numai both describe the construction of neural network using ring resonators. A person having ordinary skill in the art at the time of filing reading Wu would have found it obvious to modify the teachings of the combination of Carolan and Numai to generate an output with one or more secondary optical modes at different frequencies as shown is possible in FIGS. 6A-6D as indicated above. This would be obvious in light of Section III on p973 that describes how increasing the input power above a threshold amount of 1mW allows for shifting of the output frequency as shown in FIG. 2C on p963 and thereby makes a Relu nonlinear activation function possible. Examiner notes [0072] of Carolan in reference to FIGS. 9C and 9D also discusses the system being used to implement ReLU type nonlinear activation functions further demonstrating the similarities between the teachings of Carolan and Wu.
Regarding Claim 2, the combination of Carolan, Numai and Wu teaches the method of claim 1, wherein the optically active region comprises at least one of quantum dots, quantum wells, and quantum-dash structures (Col 17 lines 46-48 of Carolan teaches formation of the optically active region from InGaAsP positioned between layers of InP, which creates a Quantum well).
Regarding Claim 3, the combination of Carolan, Numai and Wu teaches the method of claim 1, further comprising operating the optical source in a stimulated emission region of an optical response based on the bias ([0083] of Carolan describes the disclosed optical neural network as running on coherent light and consequently the optical source would operate in a stimulated emission region to generate coherent light).
Regarding Claim 4, the combination of Carolan, Numai and Wu teaches the method of claim 1, wherein the second wavelength differs from the first wavelength ([0068] of Carolan describes how the resonant frequency of the tunable ring resonator can be shifted to change the output frequency of the tunable ring resonator).
Regarding Claim 5, the combination of Carolan, Numai and Wu teaches the method of claim 1, wherein the one or more secondary optical signals propagate at one or more wavelengths that are different from the first and second wavelengths ([0071] describes application of detuning to vary wavelength outputs of the tunable ring resonator and as shown by FIG. 9C capable of operating at multiple different wavelengths).
Regarding Claim 6, the combination of Carolan, Numai and Wu teaches the method of claim 1, wherein the one or more secondary optical signals comprises a first secondary optical signal of the first optical signal and a second secondary optical signal of the second optical signal (the secondary optical signals are based on both the first optical signal and resonant frequency (sometimes the same as the second optical signal) of the tunable ring resonator and so secondary signals can be said to be of the first and second optical signals, examiner has construed this limitation broadly as the instant specification does not detail what it means for the secondary signals to be of one or another optical signal ).
Regarding Claim 7, the combination of Carolan, Numai and Wu teaches the method of claim 1, further comprising driving the optically active region in stimulated emission operation responsive to adjusting the bias applied equal to or above a threshold bias ([0093] of Carolan describes the need to maintain the power above a threshold power, [0041] of the instant specification describes how operation of the optically active region above the threshold bias is the current at which operation of the laser moves into stimulated (i.e. coherent) emission, and [0083] of Carolan describes the disclosed optical neural network as running on coherent light and consequently the optical source would operate in a stimulated emission region to generate coherent light and be operating equal to or above the threshold bias).
Regarding Claim 8, the combination of Carolan, Numai and Wu teaches the method of claim 1, wherein generating the one or more secondary optical signals is responsive to the optical power of the first optical signal being equal to or above a threshold optical power ([0093] of Carolan describes the need to maintain the power above a threshold power).
Regarding Claim 9, the combination of Carolan, Numai and Wu teaches the method of claim 1, wherein the first optical signal is received on a bus waveguide that is evanescently coupled to a resonator structure comprising an optical gain mechanism ([0068] and FIG. 7 of Carolan teach an evanescent coupling between a waveguide 710 and ring resonator 720), wherein the optical gain mechanism comprises: a cathode formed on the bus waveguide (FIG. 7 of Carolan shows electrical routing between detector 740 and tunable resonator 720 for optical nonlinearity unit 700, which is contained in a single photonic chip as shown in FIG. 1. FIGS. 13-14 of Numai describes a specific location for anode 43 and cathode 55 on a ring resonator, which when incorporated into the teachings of Carolan as described above would result in cathode 55 extending below optical waveguide 31 (i.e. bus waveguide) and the mesa structure); a mesa structure (see the structure formed by elements 43 and 45-49 in FIG. 14 of Numai) formed on the cathode (55), the mesa structure comprising the optically active region (active layer 48 of Numai); and an anode (anode 43 of Numai) formed on the mesa structure, wherein the optically active region emits the second optical signal based on the bias applied between the cathode and the anode ([0072] of Carolan describes applying the bias to shift the output (i.e. second optical signal) of the resonator) and generates the one or more secondary optical signals responsive to the optical power of the first optical signal being equal to or above a threshold optical power ([0093] of Carolan describes the need to maintain the power above a threshold power).
Regarding Claim 16, the combination of Carolan, Numai and Wu as applied to claim 1 teaches a nonlinear activation device comprising:
a bus waveguide (710, see FIG. 7 of Carolan) configured to receive a first lasing mode having an optical power that is based on a weighted summation of a neuron ([0044] and [0047] of Carolan teaches the use of interconnected Mach-Zehnder Interferometers to apply weight parameters via matrix transformation to the optical signals before the arrive at the nonlinear activation device by way of a waveguide);
a resonator structure optically coupled to the bus waveguide (FIG. 7 of Carolan shows tunable resonator 720 optically coupled to waveguide 710) and comprising an optical gain mechanism (active layer 48, see FIG. 14 of Numai) configured to emit a second lasing mode based on injection locking to the first lasing mode (see col 17 lines 55-58 of Numai, which describes the injection locking function of a ring resonator structure) and one or more secondary lasing modes (FIG. 9B shows a second lasing mode different from the first lasing mode following interaction between the ring resonator and the waveguide) based on injection locking to the first lasing mode and based on the optical power of the first lasing mode, wherein the one or more secondary lasing modes correspond to gain peaks formed at wavelengths different from wavelengths of the first and second lasing modes (Wu in FIGS. 6A – 6B and 6C – 6D show how increasing input power causes an identifiable shift in frequency with multiple secondary optical modes appearing as peaks in FIGS. 6B and 6D, corresponding to the two different types of ring resonators being tested. The beginning of section 4.2 at the bottom of p976 makes clear that the frequency/wavelength shift is a result of increasing optical power); and
the bus waveguide configured to output an activation function based on at least one of the one or more secondary lasing modes (FIG. 9D of Carolan shows production of a range of nonlinear activation functions and [0072] describes how different functions can be selected by varying the bias voltage being applied to the tunable ring resonator).
Regarding Claim 17, the combination of Carolan, Numai and Wu teaches the nonlinear activation device of claim 16, wherein the resonator structures is a microring resonator structure (Carolan teaches a tunable ring resonator 720, given that Carolan is implemented on a photonic chip the tunable ring resonator 720 is interpreted as a microring resonator).
Regarding Claim 18, the combination of Carolan, Numai and Wu teaches the nonlinear activation device of claim 16, wherein the optical gain mechanism comprises: a cathode formed on a waveguide of the resonator structure (FIG. 14 of Numai shows cathode 55 formed on the same structure as waveguide 48); a mesa structure formed on the cathode (FIG. 14 also shows a mesa structure formed of elements 43 – 54 formed on cathode 55), the mesa structure comprising an optically active region (48 of Numai) formed between a first Group III-V semiconductor material layer and a second Group III-V semiconductor material layer (cladding layer 46 and Semiconductor substrate 54 of Numai are described at col 17 lines 52-55 as being formed from InP, which is a Group III-V semiconductor); and an anode (anode 14 as shown in FIG. 14 of Numai) formed on the mesa structure, wherein the optically active region generates the second lasing mode based on injection locking (see col 17 lines 55-58, describing injection locking operation of a ring resonator) to the first lasing mode responsive to a bias applied to the cathode and the anode being equal to or above a threshold bias ([0041] of the instant specification describes how operation of the optically active region above the threshold bias is the current at which operation of the laser moves into stimulated (i.e. coherent) emission, and [0083] of Carolan describes the disclosed optical neural network as running on coherent light and consequently the optical source would operate in a stimulated emission region to generate coherent light and be operating equal to or above the threshold bias) and generates the one or more secondary lasing responsive to the optical power of the first lasing mode equal to or above a threshold optical power ([0093] of Carolan describes the need to maintain the power above a threshold power).
Regarding Claim 19, the combination of Carolan, Numai and Wu teaches the nonlinear activation device of claim 18, wherein the optically active region comprises one or more of quantum dots, quantum wells, and quantum-dash structures (Col 17 lines 46-48 of Carolan teaches formation of the optically active region from InGaAsP positioned between layers of InP, which creates a Quantum well).
Regarding Claim 20, the combination of Carolan, Numai and Wu teaches the nonlinear activation device of claim 16, wherein the nonlinear activation device is formed as part of the neuron on a common substrate of a semiconductor platform (FIG. 1 of Carolan shows optical interference unit 124 and optical nonlinearity unit 126 formed on a common photonic integrated circuit 120).
Claims 10, 12 and 14-15 are rejected under 35 U.S.C. 103 as being unpatentable over Carolan in view of Wu and further in view of Numai.
Regarding Claim 10, Carolan teaches a neuromorphic computation system (Neural Network 100 of Carolan), comprising:
an optical interference unit configured to
(Carolan teaches encoding of input data onto the optical signals via electrical means instead of optical means at electronic interface 110 as shown in FIG. 1),
apply a weight matrix to the plurality of input optical signals by tuning optical interference ([0044] and [0047] of Carolan teaches the use of interconnected Mach-Zehnder Interferometers to apply weight parameters via matrix transformation to the optical signals), and
output a plurality of weighted optical signals (FIG. 1 of Carolan shows and [0033] describes optical signals 105c being guided by an array of waveguides 128 to optical nonlinearity unit 126); and
an optical nonlinearity unit (Optical Nonlinearity Unit 126) configured to provide a nonlinear activation function based on a weighted summation output from the optical interference unit (FIG. 9D shows production of a range of nonlinear activation functions & FIG. 1 shows weighted optical signals 105c being the input for optical nonlinearity unit 126), the optical nonlinearity unit comprising a plurality of ([0070] describes the optical nonlinearity unit being made up of ring resonators) configured to generate one or more secondary lasing modes (FIG. 9D shows secondary lasing modes) responsive to optical power of the plurality of weighted optical signals being at or above a threshold optical power ([0093] of Carolan describes the need to maintain the power above a threshold power).
wherein generating the one or more secondary lasing modes includes:
adjusting a bias to a threshold current to cause injection locking of the plurality of injection locked resonator cavity lasers and transition to stimulated emission operation (FIGS. 9A – 9D of Carolan shows how the bias applied results in stimulated emission operations and is capable of changing the output of the ring resonator, but does not specifically discuss the injection locking aspect of resonant cavity laser); and
while in the stimulated emission operation, providing the plurality of weighted optical signals at or above the threshold optical power to cause formation of the one or more secondary lasing modes, wherein the one or more secondary lasing modes correspond to gain peaks formed at wavelengths different from a wavelength of a primary lasing mode (FIG. 9D of Carolan shows production of a range of nonlinear activation functions and [0072] describes how different functions can be selected by varying the bias voltage being applied to the tunable ring resonator but Carolan does not address the specifics of the frequency response generated by increasing optical power above a threshold sufficient to generate secondary optical modes).
As indicated by strikeout above, Carolan fails to teach that the encoding is performed optically and that the lasers are injection locked resonator cavity lasers.
However, Wu at FIG. 4 describes an all-optical nonlinear activation function that uses optical modulators instead of a digital electrical signals to encode information on light entering the neural network.
Carolan and Wu both describe or suggest the use of ring resonators in optical neural networks. A person having ordinary skill in the art at the time of filing would have found it obvious to improve the configuration taught by Carolan to incorporate optical modulators for encoding optical signals prior to the signals running through a weighting matrix. The person having ordinary skill in the art would have found this obvious since Wu so would allow more of the neural network to operate in the optical domain.
Numai teaches the limitations absent from the combination of Carolan and Wu. In particular, Numai teaches how ring resonators are operable in a way such that a first optical signal can be used to injection lock an optical source (see col 17 lines 55-58 of Numai).
Numai and the combination of Carolan and Wu both describe applications of tunable laser ring resonators. Given Carolan’s silence with regards to how a ring resonator is formed or operates a person having ordinary skill in the art at the time of filing would have found it obvious to apply the teachings of Numai to form a ring resonator capable of injection locking a received optical signal.
As noted above, Carolan is also silent as to what specific effect(s) there would be on frequency shift caused by specific changes in optical power. However, Wu in FIGS. 6A – 6B and 6C – 6D show how increasing input power causes an identifiable shift in frequency with multiple secondary optical modes appearing as peaks in FIGS. 6B and 6D, corresponding to the two different types of ring resonators being tested. The beginning of section 4.2 at the bottom of p976 makes clear that the frequency/wavelength shift is a result of increasing optical power.
Wu and the combination of Carolan and Numai both describe the construction of neural network using ring resonators. A person having ordinary skill in the art at the time of filing reading Wu would have found it obvious to modify the teachings of the combination of Carolan and Numai to generate an output with one or more secondary optical modes at different frequencies as shown is possible in FIGS. 6A-6D as indicated above. This would be obvious in light of Section III on p973 that describes how increasing the input power above a threshold amount of 1mW allows for shifting of the output frequency as shown in FIG. 2C on p963 and thereby makes a Relu nonlinear activation function possible. Examiner notes [0072] of Carolan in reference to FIGS. 9C and 9D also discusses the system being able to implement ReLU type nonlinear activation functions further demonstrating the similarities between the teachings of Carolan and Wu.
Regarding Claim 12, the combination of Carolan, Wu and Numai teaches the neuromorphic computation system of claim 10, wherein the optical interference unit and the optical nonlinearity unit are formed on a common substrate of a semiconductor platform (FIG. 1 of Carolan shows optical interference unit 124 and optical nonlinearity unit 126 formed on a common photonic integrated circuit 120).
Regarding Claim 14, the combination of Carolan, Wu and Numai teaches the neuromorphic computation system of claim 10, wherein each of the injection locked resonator cavity lasers comprises: a bus waveguide configured to receive a weighted optical signal of the plurality of weighted optical signals (see input waveguide 710 as shown in FIG. 7 of Carolan, which receives weighted optical signals from optical interference unit 124) comprising at least a first lasing mode; and a resonator structure optically coupled to the bus waveguide (FIG. 7 shows tunable ring resonator 720 optically coupled to input waveguide 710) and comprising an optical gain mechanism configured to emit a second lasing mode and the one or more secondary lasing modes (FIG. 9B of Carolan shows a difference between an input and output of nonlinearity unit 924).
Regarding Claim 15, the combination of Carolan, Wu and Numai teaches the neuromorphic computation system of claim 10, further comprising a plurality of neuron, at least one neuron of the plurality of neuron comprising the optical interference unit and the optical nonlinearity unit (FIG. 4A and [0051] of Carolan describes the optical interference unit including multiple neurons, FIG. 4B and [0055] describes how optical interference units and optical nonlinearity unit cooperatively form multiple neurons in each layer of the neural network), wherein the plurality of neurons are formed on a common substrate of a semiconductor platform (FIG. 1 of Carolan shows optical interference unit 124 and optical nonlinearity unit 126 formed on a common substrate 120).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BENJAMIN WIGGER whose telephone number is (571)272-4208. The examiner can normally be reached 9:30am to 7:00pm.
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, Helal Algahaim can be reached at (571)270-5227. 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.
/BENJAMIN DAVID WIGGER/Examiner, Art Unit 3645
/HELAL A ALGAHAIM/SPE , Art Unit 3645
Prosecution Insights