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 .
This is action is in response to the Request for Continued Examination filed 13 November 2025. Claims 1, 3, 10, 12, 16, 18-19, 21 have been amended. Claim 4-5, 20 has been canceled. Claims 22-23 are new. Claims 1-3, 6-19, 21-23 are pending and have been considered below.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 13 November 2025 has been entered.
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.
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.
Claim(s) 1-2, 7-19, 21, 23 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ramfjord et al. (US 2020/0301036 A1) in view of O’Shea et al. (US 2019/0274108 A1) and further in view of Polehn (US 2017/0155476 A1).
Claim 1. Ramfjord discloses a telemetry system, a telemetry system (P 0084), comprising:
a subsea transmitter communicatively coupled to oil and gas equipment disposed in a subsea environment, equipment includes communication circuitry to receive and to transmit information with respect to one or more networks (P 0040) in an offshore geological environment below level (P 0071) which may be land-based, sea-based (e.g., vessel, ocean bottom, etc.) or land and sea-based (P 0110) for example, a marine survey may be an ocean bottom survey, which may employ ocean bottom streamers or nodes using one or more types of receivers (P 0159) in a tomographic process a source transmits a signal for oil and gas explorations (P 0164), …
… representative of oil and gas operations performed by the oil and gas equipment, providing petroleum systems modeling via input of various data (P 0048) formation evaluation is performed for interpreting data acquired from a drilled borehole to provide information about the geological formations and/or in-situ fluid(s) (P 0120) a model provides more accurate guidance as to one or more field operations in the region, including a drilling operation to drill to a layer of a reservoir construct a well to produce fluid from the reservoir (P 0230),
a surface receiver disposed in a region above the subsea environment, wherein the surface receiver is configured to receive the analog signal and to convert the analog signal into output digital bits, energy received may be discretized by an analog-to-digital converter that operates at a sampling rate (P 0069) in an offshore geological environment below level (P 0071) which may be land-based, sea-based (e.g., vessel, ocean bottom, etc.) or land and sea-based (P 0110) for example, a marine survey may be an ocean bottom survey, which may employ ocean bottom streamers or nodes using one or more types of receivers (P 0159), wherein the surface receiver comprises one or more receiver components, a transmitter emits a signal and the signal is received by a receiver (P 0069) samples of seismic energy as acquired by one or more seismic energy receivers (P 0143),
comprising a neural network trained via machine learning, in embodiments, a computational imaging framework uses trained deep convolutional neural networks (CNN) to detect stratigraphic units in images of seismic sections (P 0148) seismic data from an ongoing seismic survey is received and interpreted via an interpreter, a ML system is trained using the interpreted seismic data to generate a trained ML system, and the trained ML system is applied to additional seismic data acquired by the ongoing seismic survey (P 0149).
Ramfjord does not disclose the subsea transmitter is configured to convert digital bits … into an analog signal, and to transmit the analog signal via a communications channel present in the subsea environment, as disclosed in the claims. However, in the same field of invention, O’Shea discloses a machine-learning encoder and a machine-language decoder including a digital to analog converter and an analog to digital converter (P 0103). Therefore, considering the teachings of Ramfjord and O’Shea, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the subsea transmitter is configured to convert digital bits … into an analog signal, and to transmit the analog signal via a communications channel present in the subsea environment with the teachings of Ramfjord with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Ramfjord does not disclose receive observables comprising intermediately processed data by at least one receiver component of the one or more receiver components based upon the digital bits, as disclosed in the claims. New Claim 23 defines “observables” as “time traces of a constellation phase shift and/or soft symbols before and after error correcting codes. In the same field of invention, Polehn discloses binary data is determined based on signal constellation data corresponding to a modulation scheme, and symbols of a signal constellation are decoded (P 0016) where constellations pertaining to various modulation schemes, e.g. quadrature phase shift keying or binary phase shift key are used (P 0048) for the claimed observables. Polehn further discloses In Figure 5A discloses channel information is received and demodulated (P 0072) if the data is not error free, corrective service data that includes error is stored that includes bits with error (P 0074) and then retransmission of the segment of the data is requested and received (P 0075) and then a corrective signal constellation matrix is generated based on channel information and error vector magnitude data (P 0080). The initially received is analogous to the claimed intermediately processed data. Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine receive observables comprising intermediately processed data by at least one receiver component of the one or more receiver components based upon the digital bits with the teachings of Ramfjord and O’Shea with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide better error correction techniques (Polehn: P 0001).
Ramfjord does not disclose wherein the neural network comprises a machine learning agent configured to analyze the observables … and to adaptively tune the surface receiver based on the observables, as disclosed in the claims. However, Ramfjord discloses a workflow outputs rock properties based at least in part on processing of seismic data (P 0035) where a method executes during drilling, LWD measurements may be utilized to output formation parameters (e.g., formation parameter values) that may be utilized to guide the drilling (P 0119) the trained CNNS outputs output (stratigraphic) data (P 0193) once the neural network has been trained, it can be utilized for prediction on a desired seismic section and output segmented stratigraphic units from the input seismic section (P 0208) labeled stratigraphic units can be returned as output, which enhance a seismic image, overlaid on a seismic image to allow for visualization of stratigraphic units (P 0211) output from a solver can be implicit function values (P 0220) the accuracy of the output of the trained CNN as to stratigraphy can be enhanced (P 0221). O’Shea discloses a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Polehn discloses the correction service may “learn” the interference patterns and compensate for errors that may be caused by various conditions, such as multipath fading, Doppler effect, interference, etc. (P 0018) an instance-based learning algorithm may be implemented, including a locally weighted learning algorithm (P 0040) the communication interface may adaptively change the modulation and demodulation scheme used (P 0068). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the neural network comprises a machine learning agent configured to analyze the observables … and to adaptively tune the surface receiver based on the observables with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide better error correction techniques (Polehn: P 0001).
Ramfjord does not disclose determine incorrect recovery of the digital bits by the one or more receiver components; transmit hyperparameters to adjust operating parameters of the one or more receiver components to adaptively tune the surface receiver based on the observables to improve recovery of the digital bits when incorrect recovery of the digital bits by the one or more receiver components is determined, as disclosed in the claims. However, Polehn discloses an attempt is made to recover bits from the symbols based on a signal constellation matrix (P 0049) without error (P 0052) to adapt the modulation and demodulation scheme of the signal constellation matrix (P 0068) a determination is made that a segment is not error free (P 0073, 0074) a request to retransmit the segment data is made and the data is received (P 0075) a determination is made that the received retransmitted data is error free (P 0077) a corrective signal constellation matrix is generated (P 0080). The adaptation of the signal constellation matrix is analogous to the tuning of the surface receiver. Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine determine incorrect recovery of the digital bits by the one or more receiver components; transmit hyperparameters to adjust operating parameters of the one or more receiver components to adaptively tune the surface receiver based on the observables to improve recovery of the digital bits when incorrect recovery of the digital bits by the one or more receiver components is determined with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide better error correction techniques (Polehn: P 0001).
Claim 2. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and Ramfjord discloses wherein the neural network, received seismic image data is processed to generate stratigraphic information using a trained convolution neural network (P 0003). O’Shea discloses processed information may be in any suitable form that be communicated over a channel such as packets (P 0113) a preamble is imbedded into encoded information (P 0169) and the signal is converted from a digital signal to an analog signal and from an analog signal to a digital signal (P 0170). The encoded preamble in O’Shea need not serve the same function as the transmitted signals in Ramfjord because transmitting a preamble with a signal is well known in the art. Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine configured to detect a data packet preamble transmitted via the communications channel or to provide a receiver pulse shape to filter the analog signal with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and the Supreme Court in KSR International Co. v. Teleflex Inc. identified applying a known technique to a known device (method, or product) ready for improvement to yield predictable results as a rationale to support a conclusion of obviousness which is consistent with the proper “functional approach” to the determination of obviousness as laid down in Graham.
Claims 4-5 canceled.
Claim 7. Ramfjord, O’Shea and Polehn disclose the system of claim 1, but Ramfjord does not disclose one or more transmitter components trained via machine learning to process transmitter information, wherein the one or more transmitter components, the one or more receiver components, or any combination thereof, are trained via a dataset created by a Generative Adversarial Network (GAN), as disclosed in the claims. However, O’Shea discloses the encoder and decoder implement encoding and decoding techniques that are learned from one or more machine-learning networks that have been trained to learn suitable input-output encoding and decoding mappings based on one or more objective criteria wherein the machine-learning networks may be artificial neural networks (P 0096) impairments may be in a transmitted signal (P 0116) the channel machine-learning network of the approximated channel may be trained (P 0117) the system includes a real-world channel an approximated channel, and a discriminator, the approximated channel has a machine-learning network (“channel network”) and the discriminator has a machine-learning network (“discriminator network”), the approximated channel and the discriminator may form a generative an adversarial network (GAN) (P 0207). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine one or more transmitter components trained via machine learning to process transmitter information, wherein the one or more transmitter components, the one or more receiver components, or any combination thereof, are trained via a dataset created by a Generative Adversarial Network (GAN) with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Claim 8. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and Ramfjord discloses a loop may be created whereby a model is iteratively revised based on output of stratigraphic units from a trained machine model and where synthetic seismic data are generated from the model and compared to actual seismic data and/or input to the trained machine model to compare stratigraphic units based on synthetic seismic data to stratigraphic units based at least in part on actual seismic data (P 0229). O’Shea discloses the channel machine-learning network training may be designed to account for communications channel impairment effects, wherein the encoder machine-learning network and/or decoder machine-learning network may be trained to encode information as a signal that is transmitted over a radio transmission channel (P 0089) the approximated channel includes effects of transmitter and receiver components, a simulated channel is used for training, an analytic channel impairment model may be utilized that fits a specific set of hardware/software and wireless deployment conditions, the encoder network and the decoder network may be trained to operate under different channel conditions, as well as for different real-world transmitter and receiver scenarios (P 0227). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine comprising one or more transmitter components trained via machine learning to process transmitter information, wherein the one or more transmitter components, the one or more receiver components, or any combination thereof, are trained via an autoencoder neural network that accounts for system-in-the- loop data transmissions with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Claim 9. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and O’Shea discloses the training of the encoder machine-learning network and/or decoder machine-learning network achieves particular spectral properties, to perform well in particular regimes such as at a low signal to noise (SNR) ratio (P 0089) the first operations and the second operations may include filtering, tuning, etc. in order to account for noise (P 0107) performance metrics may be incorporated into training, as part of the distance loss functions, including bit error rate (BER) as a function of the signal-to-noise ratio (SNR) and spectral efficiency (P 0165). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the one or more receiver components are trained to provide for spectrum sensing that classifies noise and provide indications noise-free regions in the communications channel with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Claim(s) 10 is/are directed to method for telemetry claim(s) similar to the telemetry system claim(s) of Claim(s) 1, and Ramfjord further discloses the medium can be water (P 0046, 0064, 0163), and Claim 10 is/are rejected with the same rationale.
Claim(s) 11 is/are directed to method for telemetry claim(s) similar to the telemetry system claim(s) of Claim(s) 2 and is/are rejected with the same rationale.
Claim 12. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and Ramfjord discloses, a workflow outputs rock properties based at least in part on processing of seismic data (P 0035) where a method executes during drilling, LWD measurements may be utilized to output formation parameters (e.g., formation parameter values) that may be utilized to guide the drilling (P 0119) the trained CNNS outputs output (stratigraphic) data (P 0193) once the neural network has been trained, it can be utilized for prediction on a desired seismic section and output segmented stratigraphic units from the input seismic section (P 0208) labeled stratigraphic units can be returned as output, which enhance a seismic image, overlaid on a seismic image to allow for visualization of stratigraphic units (P 0211) output from a solver can be implicit function values (P 0220) the accuracy of the output of the trained CNN as to stratigraphy can be enhanced (P 0221). O’Shea discloses a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the machine learning agent is configured to generate the hyperparameters to adaptively tune the surface receiver with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Claim(s) 13 is/are directed to method for telemetry claim(s) similar to the telemetry system claim(s) of Claim(s) 7, and Ramfjord discloses , a signal may be discretized by an analog-to-digital converter (P 0069) the U-Net can be applied as part of a deep network training method where annotated (e.g., labeled) training samples are utilized to train an NNS (P 0172), that is, the neural network is trained and then discretizes an analog signal with an analog-to-digital converter, and Claim 13 is rejected with rejected with the same rationale.
Claim 14. Ramfjord, O’Shea and Polehn disclose the method of claim 13, and Ramfjord discloses wherein the one or more transmitter components, the one or more receiver components, or a combination thereof, are trained via supervised training, via semi-supervised training, via unsupervised training, or a combination thereof, the neural network may be trained using supervised learning (P 0146, 0208).
Claim 15. Ramfjord, O’Shea and Polehn disclose the method of claim 11, and Ramfjord discloses generating a training dataset via machine learning for training of the one or more receiver components, training a NNS or NNSs on available training data (P 0153) training data is selected (P 0161) additional training data is generated (P 0172).
Claim(s) 16, 17, 18, is/are directed to non-transitory computer readable media claim(s) similar to the method claim(s) of Claim(s) 10, 11, 12 and is/are rejected with the same rationale.
Claim 19. Ramfjord, O’Shea and Polehn disclose the non-transitory computer readable medium of claim 16, Ramfjord discloses the neural network may be trained using supervised learning (P 0146, 0208) a neural network such as a convolution neural network (CNN) may be trained by passing in seismic sections (e.g., as tiles) along with labeled stratigraphic units (e.g., training information) (P 0208).
O’Shea discloses the approximated channel and the discriminator may form a generative adversarial network (GAN) (P 0207). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein one or more receiver components comprising the machine learning agent, one or more transmitter components of the subsea transmitter, or a combination thereof, are trained via supervised training, via semi-supervised training, via unsupervised training, or a combination thereof, and wherein unsupervised training comprises executing an autoencoder neural network, a Generative Adversarial Network (GAN), or a combination thereof with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Claim 20. canceled.
Claim 21. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and Ramfjort discloses wherein the communications channel comprises a pipe, sensors are operatively coupled to portion of a standpipe (P 0101).
Claim 23. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and Ramfjord discloses the model may optionally be updated in response to model output, changes in time (P 0045) modeling may also model geometry with respect to time (P 0047) a period of time (P 0058) time information, as associated with sensed energy, may allow for understanding spatial relations of layers, interfaces, structures, etc. in a geologic environment (P 0065). Polehn discloses binary data is determined based on signal constellation data corresponding to a modulation scheme, and symbols of a signal constellation are decoded (P 0016) where constellations pertaining to various modulation schemes, e.g. quadrature phase shift keying or binary phase shift key are used (P 0048) Figure 5A discloses channel information is received and demodulated (P 0072) if the data is not error free, corrective service data that includes error is stored that includes bits with error (P 0074) and then retransmission of the segment of the data is requested and received (P 0075) and then a corrective signal constellation matrix is generated based on channel information and error vector magnitude data (P 0080). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the observables comprise time traces of a constellation phase shift, soft symbols before and after error correcting codes, or both with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Claim(s) 3, 6, 22 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ramfjord et al. (US 2020/0301036 A1) in view of O’Shea et al. (US 2019/0274108 A1) and Polehn (US 2017/0155476 A1) and further in view of Sumi et al. (US 2019/0129026 A1).
Claim 3. Ramfjord, O’Shea and Polehn disclose the system of claim 1, and Ramfjord discloses, a workflow outputs rock properties based at least in part on processing of seismic data (P 0035) where a method executes during drilling, LWD measurements may be utilized to output formation parameters (e.g., formation parameter values) that may be utilized to guide the drilling (P 0119) the trained CNNS outputs output (stratigraphic) data (P 0193) once the neural network has been trained, it can be utilized for prediction on a desired seismic section and output segmented stratigraphic units from the input seismic section (P 0208) labeled stratigraphic units can be returned as output, which enhance a seismic image, overlaid on a seismic image to allow for visualization of stratigraphic units (P 0211) output from a solver can be implicit function values (P 0220) the accuracy of the output of the trained CNN as to stratigraphy can be enhanced (P 0221). O’Shea discloses a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the machine learning agent is configured to generate the hyperparameters to adaptively tune the surface receiver with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003).
Ramfjord does not disclose wherein the hyperparameters comprise parameters of an allocation of feedforward and feedback filters to compensate the communications channel, parameters of tracking loops to compensate for variation in propagation speed, or a combination thereof, as disclosed in the claims. However, Ramfjord discloses as survey data becomes available, interpretation tasks may be performed for building, adjusting, etc., one or more models (P 0130) the CNNS is trained by determining values of weights using backpropagation (P 0206) a model is iteratively revised based on output of stratigraphic units from a trained machine model and where synthetic seismic data are generated from the model and compared to actual seismic data and/or input to the trained machine model to compare stratigraphic units based on synthetic seismic data to stratigraphic units based at least in part on actual seismic data (P 0229). O’Shea discloses the channel network models the impairments which occur to a radio signal sent over a channel based radio propagation effects of the channel (P 0108) the encoder and/or decoder is updated during deployment based on real-time performance results such as propagation characteristics, delay, etc (P 0111) the machine learning model includes collections of operations arranged in a feed-forward or in a manner with feedback (P 0123). Both Ramfjord and O’Shea disclose adjusting models based on at least feedback of output data and training data, and O’Shea uses both feedforward and feedback operations. Furthermore, O’Shea discloses that impairments of communications channels are modeled based on propagation characteristics, such as delay, and the encoder/decoder are updated based on the propagation characteristics of the channel. That is, the combination of Ramfjord and O’Shea disclose updating a model based on propagation characteristics of a communications channel using feedforward and feedback of data. In the same field of invention, Sumi discloses command signals from a control unit to transmitters can be expressed as the information of wave shapes of drive signals to be generated by the transmitters (P 0245) a wave shape can be optimized for driving signals that generate waves (P 0616) and the wave geometry can be used as a target for an optimization (linear or nonlinear optimizations, linear programming) (P 0617) a signal generator generates a drive signal according to a trigger signal provided by the control unit that controls the wave shape or geometry of a pulse wave (P 0927) spectra of the reception signals are obtained on the basis of the spectral analysis, the properties of filterings, or beams or waves can be controlled such as a shape, a beam geometry (P 0941). Sumi further discloses for superpostions, phase aberrations are corrected by compensating for an inhomogeneity in a speed or a wave propagation speed (P 0640, 0768). Therefore, considering the teachings of Ramfjord, O’Shea, Polehn and Sumi, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the hyperparameters comprise parameters of an allocation of feedforward and feedback filters to compensate the communications channel, parameters of tracking loops to compensate for variation in propagation speed, or a combination thereof with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide highly accurate signal processing (Sumi: P 0012).
Claim 6. Ramfjord, O’Shea, and Polehn disclose the system of claim 2, but Ramfjord does not disclose a transmitter component trained via machine learning to provide a transmitter pulse shape for transmitting the analog signal, wherein the transmitter pulse shape and the receiver pulse shape cooperate to improve receipt of the analog signal, as disclosed in the claims. However, Ramfjord discloses seismic data from an ongoing seismic survey is received and interpreted via an interpreter, a ML system is trained using the interpreted seismic data to generate a trained ML system, and the trained ML system is applied to additional seismic data acquired by the ongoing seismic survey (P 0149) and O’Shea discloses the encoder and decoder implement encoding and decoding techniques that are learned from one or more machine-learning networks that have been trained to learn suitable input-output encoding and decoding mappings based on one or more objective criteria wherein the machine-learning networks may be artificial neural networks (P 0096) impairments may be in a transmitted signal (P 0116) the channel machine-learning network of the approximated channel may be trained (P 0117). In the same field of invention, Sumi discloses command signals from a control unit to transmitters can be expressed as the information of wave shapes of drive signals to be generated by the transmitters (P 0245) a wave shape can be optimized for driving signals that generate waves (P 0616) and the wave geometry can be used as a target for an optimization (linear or nonlinear optimizations, linear programming) (P 0617) a signal generator generates a drive signal according to a trigger signal provided by the control unit that controls the wave shape or geometry of a pulse wave (P 0927) spectra of the reception signals are obtained on the basis of the spectral analysis, the properties of filterings, or beams or waves can be controlled such as a shape, a beam geometry (P 0941). Therefore, considering the teachings of Ramfjord, O’Shea, and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine a transmitter component trained via machine learning to provide a transmitter pulse shape for transmitting the analog signal, wherein the transmitter pulse shape and the receiver pulse shape cooperate to improve receipt of the analog signal with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide highly accurate signal processing (Sumi: P 0012).
Claim 22. Ramfjord, O’Shea and Polehn disclose the system of claim 1, but Ramfjord does not disclose wherein the machine learning agent is configured to generate the hyperparameters to adaptively tune the surface receiver, as disclosed in the claims. However, However, Ramfjord discloses a workflow outputs rock properties based at least in part on processing of seismic data (P 0035) where a method executes during drilling, LWD measurements may be utilized to output formation parameters (e.g., formation parameter values) that may be utilized to guide the drilling (P 0119) the trained CNNS outputs output (stratigraphic) data (P 0193) once the neural network has been trained, it can be utilized for prediction on a desired seismic section and output segmented stratigraphic units from the input seismic section (P 0208) labeled stratigraphic units can be returned as output, which enhance a seismic image, overlaid on a seismic image to allow for visualization of stratigraphic units (P 0211) output from a solver can be implicit function values (P 0220) the accuracy of the output of the trained CNN as to stratigraphy can be enhanced (P 0221). O’Shea discloses a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Polehn discloses the correction service may “learn” the interference patterns and compensate for errors that may be caused by various conditions, such as multipath fading, Doppler effect, interference, etc. (P 0018) an instance-based learning algorithm may be implemented, including a locally weighted learning algorithm (P 0040) the communication interface may adaptively change the modulation and demodulation scheme used (P 0068). Polehn further discloses an attempt is made to recover bits from the symbols based on a signal constellation matrix (P 0049) without error (P 0052) to adapt the modulation and demodulation scheme of the signal constellation matrix (P 0068) a determination is made that a segment is not error free (P 0073, 0074) a request to retransmit the segment data is made and the data is received (P 0075) a determination is made that the received retransmitted data is error free (P 0077) a corrective signal constellation matrix is generated (P 0080). The adaptation of the signal constellation matrix is analogous to the tuning of the surface receiver. Therefore, considering the teachings of Ramfjord, O’Shea and Polehn, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the machine learning agent is configured to generate the hyperparameters to adaptively tune the surface receiver with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide better error correction techniques (Polehn: P 0001).
Ramfjord does not disclose wherein the hyperparameters comprise a syncword detection configured to adjust to the background noise level, an allocation of feedforward filters and feedback filters, parameters of tracking loops configured to compensate for a variation in propagation speed, or a combination thereof, as disclosed in the claims. However, Ramfjord discloses as survey data becomes available, interpretation tasks may be performed for building, adjusting, etc., one or more models (P 0130) the CNNS is trained by determining values of weights using backpropagation (P 0206) a model is iteratively revised based on output of stratigraphic units from a trained machine model and where synthetic seismic data are generated from the model and compared to actual seismic data and/or input to the trained machine model to compare stratigraphic units based on synthetic seismic data to stratigraphic units based at least in part on actual seismic data (P 0229). O’Shea discloses the channel network models the impairments which occur to a radio signal sent over a channel based radio propagation effects of the channel (P 0108) the encoder and/or decoder is updated during deployment based on real-time performance results such as propagation characteristics, delay, etc (P 0111) the machine learning model includes collections of operations arranged in a feed-forward or in a manner with feedback (P 0123) a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Both Ramfjord and O’Shea disclose adjusting models based on at least feedback of output data and training data, and O’Shea uses both feedforward and feedback operations. Furthermore, O’Shea discloses that impairments of communications channels are modeled based on propagation characteristics, such as delay, and the encoder/decoder are updated based on the propagation characteristics of the channel. That is, the combination of Ramfjord and O’Shea disclose updating a model based on propagation characteristics of a communications channel using feedforward and feedback of data. In the same field of invention, Sumi discloses command signals from a control unit to transmitters can be expressed as the information of wave shapes of drive signals to be generated by the transmitters (P 0245) a wave shape can be optimized for driving signals that generate waves (P 0616) and the wave geometry can be used as a target for an optimization (linear or nonlinear optimizations, linear programming) (P 0617) a signal generator generates a drive signal according to a trigger signal provided by the control unit that controls the wave shape or geometry of a pulse wave (P 0927) spectra of the reception signals are obtained on the basis of the spectral analysis, the properties of filterings, or beams or waves can be controlled such as a shape, a beam geometry (P 0941). Sumi further discloses for superpostions, phase aberrations are corrected by compensating for an inhomogeneity in a speed or a wave propagation speed (P 0640, 0768). Therefore, considering the teachings of Ramfjord, O’Shea, Polehn and Sumi, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the hyperparameters comprise a syncword detection configured to adjust to the background noise level, an allocation of feedforward filters and feedback filters, parameters of tracking loops configured to compensate for a variation in propagation speed, or a combination thereof with the teachings of Ramfjord, O’Shea and Polehn with the motivation to provide an efficient system for analyzing real-world communications (O’Shea: P 0003) and provide highly accurate signal processing (Sumi: P 0012).
Response to Arguments
The following response to arguments were presented in the Advisory Action mailed 29 October 2025 in response to the Response After Final Action filed 17 October 2025.
The applicant argues: Independent claims 1, 10, and 16 recite, inter alia, wherein the surface receiver comprises one or more receiver components comprising a neural network trained via machine learning, wherein the neural network comprises a machine learning agent configured to; receive observables comprising intermediately processed data by at least one receiver component of the one or more receiver components based upon the digital bits; analyze the observables to determine incorrect recovery of the digital bits by the one or more receiver components; and transmit hyperparameters to adjust operating parameters of the one or more receiver components to adaptively tune the surface receiver based on the observables to improve recovery of the digital bits when incorrect recovery of the digital bits by the one or more receiver components is determined Emphasis added. Applicant submits that Ramfjord, O'Shea, and Soni, taken alone or in hypothetical combination, fail to teach or suggest at least the foregoing aspects of independent claims 1, 10, and 16.
The amendments directed to the following or similar limitations "observables comprising immediately processed data", "analyze the observables to determine incorrect recovery of the digital bits by the one or more receiver components", and "adaptively tune the surface receiver to improve recovery of the digital bits when incorrect of the digital bits by the one or more receiver components is determined", in addition to new dependent claims 22 and 23 will require further search and consideration by the examiner. The applicant argues: the Examiner does not appear to provide an analogous feature in the prior art of record for "transmit the analog signal via a communications channel present in the subsea environment." Additionally, the Examiner appears to assert Ramfjord teaches or suggests the "surface receiver disposed in a region above the subsea environment" of the present claims. While Ramfjord does teach "a seismic survey and/or other data acquisition may be for onshore and/or offshore geologic environments,' Ramfjord does not specifically teach or suggest a surface receiver that receives from a subsea transmitter, as recited in the present claims.
Ramfjord discloses equipment includes communication circuitry to receive and to transmit information with respect to one or more networks (P 0040) in an offshore geological environment below level (P 0071) which may be land-based, sea-based (e.g., vessel, ocean bottom, etc.) or land and sea-based (P 0110) for example, a marine survey may be an ocean bottom survey, which may employ ocean bottom streamers or nodes using one or more types of receivers (P 0159) in a tomographic process a source transmits a signal for oil and gas explorations (P 0164). That is, it is clear that Ramfjord discloses transmitting signals from an undersea transmitter to an above-surface receiver. The examiner combined O 'Shea for limitations directed to "convert digital bits into an analog signal". O'Shea discloses a machine-learning encoder and a machine-language decoder including a digital to analog converter and an analog to digital converter (P 0103).
Applicant's arguments filed 17 October 2025 have been fully considered but they are not persuasive.
Claim 19 has been rejected over Ramfjord in view of O’Shea. Ramfjord discloses the neural network may be trained using supervised learning (P 0146, 0208) a neural network such as a convolution neural network (CNN) may be trained by passing in seismic sections (e.g., as tiles) along with labeled stratigraphic units (e.g., training information) (P 0208). O’Shea discloses the approximated channel and the discriminator may form a generative adversarial network (GAN) (P 0207).
Claim 21 has been rejected over Ramfjord. Ramfjort discloses wherein the communications channel comprises a pipe, sensors are operatively coupled to portion of a standpipe (P 0101).
Applicant’s arguments with respect to claim(s) 1, 10, 16 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
The examiner combined new prior art reference Polehn with Ramfjord to reject limitations directed to:
the neural network comprises a machine learning agent configured to;
receive observables comprising intermediately processed data by at least one receiver component of the one or more receiver components based upon the digital bits; analyze the observables to determine incorrect recovery of the digital bits by the one or more receiver components; and
transmit hyperparameters to adjust operating parameters of the one or more receiver components to adaptively tune the surface receiver based on the observables to improve recovery of the digital bits when incorrect recovery of the digital bits by the one or more receiver components is determined.
New Claim 23 defines “observables” as “time traces of a constellation phase shift and/or soft symbols before and after error correcting codes. In the same field of invention, Polehn discloses binary data is determined based on signal constellation data corresponding to a modulation scheme, and symbols of a signal constellation are decoded (P 0016) where constellations pertaining to various modulation schemes, e.g. quadrature phase shift keying or binary phase shift key are used (P 0048) for the claimed observables. Polehn further discloses an attempt is made to recover bits from the symbols based on a signal constellation matrix (P 0049) without error (P 0052) to adapt the modulation and demodulation scheme of the signal constellation matrix (P 0068) In Figure 5A discloses channel information is received and demodulated (P 0072) a determination is made that a segment is not error free (P 0073, 0074) if the data is not error free, corrective service data that includes error is stored that includes bits with error (P 0074) and then retransmission of the segment of the data is requested and received (P 0075) a determination is made that the received retransmitted data is error free (P 0077) and then a corrective signal constellation matrix is generated based on channel information and error vector magnitude data (P 0080).
Regarding the limitations:
wherein the neural network comprises a machine learning agent configured to analyze the observables … and to adaptively tune the surface receiver based on the observables,
Polehn discloses the correction service may “learn” the interference patterns and compensate for errors that may be caused by various conditions, such as multipath fading, Doppler effect, interference, etc. (P 0018) an instance-based learning algorithm may be implemented, including a locally weighted learning algorithm (P 0040) the communication interface may adaptively change the modulation and demodulation scheme used (P 0068).
New Claim 22 has been rejected over Ramfjord in view of O’Shea and Polehn. Ramfjord discloses a workflow outputs rock properties based at least in part on processing of seismic data (P 0035) where a method executes during drilling, LWD measurements may be utilized to output formation parameters (e.g., formation parameter values) that may be utilized to guide the drilling (P 0119) the trained CNNS outputs output (stratigraphic) data (P 0193) once the neural network has been trained, it can be utilized for prediction on a desired seismic section and output segmented stratigraphic units from the input seismic section (P 0208) labeled stratigraphic units can be returned as output, which enhance a seismic image, overlaid on a seismic image to allow for visualization of stratigraphic units (P 0211) output from a solver can be implicit function values (P 0220) the accuracy of the output of the trained CNN as to stratigraphy can be enhanced (P 0221). O’Shea discloses a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Polehn discloses the correction service may “learn” the interference patterns and compensate for errors that may be caused by various conditions, such as multipath fading, Doppler effect, interference, etc. (P 0018) an instance-based learning algorithm may be implemented, including a locally weighted learning algorithm (P 0040) the communication interface may adaptively change the modulation and demodulation scheme used (P 0068). Polehn further discloses an attempt is made to recover bits from the symbols based on a signal constellation matrix (P 0049) without error (P 0052) to adapt the modulation and demodulation scheme of the signal constellation matrix (P 0068) a determination is made that a segment is not error free (P 0073, 0074) a request to retransmit the segment data is made and the data is received (P 0075) a determination is made that the received retransmitted data is error free (P 0077) a corrective signal constellation matrix is generated (P 0080). The adaptation of the signal constellation matrix is analogous to the tuning of the surface receiver.
Furthermore, Ramfjord discloses as survey data becomes available, interpretation tasks may be performed for building, adjusting, etc., one or more models (P 0130) the CNNS is trained by determining values of weights using backpropagation (P 0206) a model is iteratively revised based on output of stratigraphic units from a trained machine model and where synthetic seismic data are generated from the model and compared to actual seismic data and/or input to the trained machine model to compare stratigraphic units based on synthetic seismic data to stratigraphic units based at least in part on actual seismic data (P 0229). O’Shea discloses the channel network models the impairments which occur to a radio signal sent over a channel based radio propagation effects of the channel (P 0108) the encoder and/or decoder is updated during deployment based on real-time performance results such as propagation characteristics, delay, etc (P 0111) the machine learning model includes collections of operations arranged in a feed-forward or in a manner with feedback (P 0123) a machine-learning model’s hyperparameters may be updated (P 0152) input information is chosen from a training set limited to a particular class of information so the system can be trained to learn communication near-optimal encoding and decoding techniques that are tuned to communicate that particular class of information for a particular scenario (P 0163). Both Ramfjord and O’Shea disclose adjusting models based on at least feedback of output data and training data, and O’Shea uses both feedforward and feedback operations. Furthermore, O’Shea discloses that impairments of communications channels are modeled based on propagation characteristics, such as delay, and the encoder/decoder are updated based on the propagation characteristics of the channel. That is, the combination of Ramfjord and O’Shea disclose updating a model based on propagation characteristics of a communications channel using feedforward and feedback of data. Sumi discloses command signals from a control unit to transmitters can be expressed as the information of wave shapes of drive signals to be generated by the transmitters (P 0245) a wave shape can be optimized for driving signals that generate waves (P 0616) and the wave geometry can be used as a target for an optimization (linear or nonlinear optimizations, linear programming) (P 0617) a signal generator generates a drive signal according to a trigger signal provided by the control unit that controls the wave shape or geometry of a pulse wave (P 0927) spectra of the reception signals are obtained on the basis of the spectral analysis, the properties of filterings, or beams or waves can be controlled such as a shape, a beam geometry (P 0941). Sumi further discloses for superpostions, phase aberrations are corrected by compensating for an inhomogeneity in a speed or a wave propagation speed (P 0640, 0768).
New Claim 23 has been rejected over Ramfjord in view of Polehn. Ramfjord discloses the model may optionally be updated in response to model output, changes in time (P 0045) modeling may also model geometry with respect to time (P 0047) a period of time (P 0058) time information, as associated with sensed energy, may allow for understanding spatial relations of layers, interfaces, structures, etc. in a geologic environment (P 0065). Polehn discloses binary data is determined based on signal constellation data corresponding to a modulation scheme, and symbols of a signal constellation are decoded (P 0016) where constellations pertaining to various modulation schemes, e.g. quadrature phase shift keying or binary phase shift key are used (P 0048) Figure 5A discloses channel information is received and demodulated (P 0072) if the data is not error free, corrective service data that includes error is stored that includes bits with error (P 0074) and then retransmission of the segment of the data is requested and received (P 0075) and then a corrective signal constellation matrix is generated based on channel information and error vector magnitude data (P 0080).
Conclusion
Any inquiry concerning this communication should be directed to JOHN M HEFFINGTON at telephone number (571)270-1696.
Examiner interviews are available via a variety of formats. See MPEP § 713.01. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN M HEFFINGTON whose telephone number is (571)270-1696. The examiner can normally be reached on Monday through Friday from 9:30 am to 5:30 pm Eastern.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Cesar B Paula, can be reached at telephone number 571-272-4128. 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 Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center to authorized users only. Should you have questions about access to the USPTO patent electronic filing system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Examiner interviews are available via a variety of formats. See MPEP § 713.01. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) Form at https://www.uspto.gov/InterviewPractice.
/J.M.H/Examiner, Art Unit 2145 2/25/2026
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145