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 action is responsive to the application filed 03/06/2023. Claims 1-20 are presented for examination.
Priority
Applicant’s claim for the benefit of a prior filed application US 16/233576, filed 12/27/2018, is acknowledged.
Information Disclosure Statement
The information disclosure statements (IDS) submitted 06/16/2026, 12/23/2025, 09/03/2025, 06/02/2025, 02/27/2025, 01/24/2025, 01/02/2025, 11/05/2024, 08/06/2024, 03/05/2024, 07/26/2023, 03/31/2023, 03/06/2023 have been considered by the examiner.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1, 3-5, 7-13, 15-16, & 18-19 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1, 3-7, 11, 13, 15, & 17 of U.S. Patent No. 11416735. Although the claims at issue are not identical, they are not patentably distinct from each other because the claims of the present invention are similar in scope to the claims of U.S. Patent No. 11416735. For example, the table below shows similarities and differences between the instant application and the U.S. Patent No. 11416735.
18/179,317 (Instant Application)
US 11416735 B2
1. An apparatus comprising: a first stage of combiners configured to receive encoded input data and further configured to implement a first function to provide first intermediate data; and a second stage of combiners configured to receive the first intermediate data and further configured to combine the first intermediate data using a set of predetermined weights to provide the encoded data with reduced noise.
1. An apparatus comprising: an encoder configured to encode input data using encoded bits in accordance with an encoding technique and to provide encoded input data; a memory configured to receive the encoded input data from the encoder and configured to store the encoded input data; and combiners configured to receive the encoded input data from the memory and further configured to combine the encoded input data among a set of predetermined weights, wherein the combiners are further configured to provide encoded data with reduced noise, the noise introduced by the memory.
3. The apparatus of claim 1, wherein the combiners comprise: a first stage of combiners configured to receive the encoded input data and evaluate at least one non-linear function using combinations of the encoded input data to provide intermediate data; and at least a second stage of combiners configured to receive the intermediate data and combine the intermediate data using a set of predetermined set of weights to provide the encoded data with reduced noise.
3. The apparatus of claim 1, wherein the first function is a nonlinear function.
3. The apparatus of claim 1, wherein the combiners comprise: a first stage of combiners configured to receive the encoded input data and evaluate at least one non-linear function using combinations of the encoded input data to provide intermediate data; and at least a second stage of combiners configured to receive the intermediate data and combine the intermediate data using a set of predetermined set of weights to provide the encoded data with reduced noise.
4. The apparatus of claim 1, wherein the first stage of combiners and second stage of combiners comprises a first plurality of multiplication/accumulation units, the first plurality of multiplication/accumulation units each configured to multiply at least one bit of the encoded input data with at least one of the set of predetermined weights and sum multiple weighted bits of the encoded input data.
1. An apparatus comprising: an encoder configured to encode input data using encoded bits in accordance with an encoding technique and to provide encoded input data; a memory configured to receive the encoded input data from the encoder and configured to store the encoded input data; and combiners configured to receive the encoded input data from the memory and further configured to combine the encoded input data among a set of predetermined weights, wherein the combiners are further configured to provide encoded data with reduced noise, the noise introduced by the memory.
3. The apparatus of claim 1, wherein the combiners comprise: a first stage of combiners configured to receive the encoded input data and evaluate at least one non-linear function using combinations of the encoded input data to provide intermediate data; and at least a second stage of combiners configured to receive the intermediate data and combine the intermediate data using a set of predetermined set of weights to provide the encoded data with reduced noise.
4. The apparatus of claim 3, wherein the first stage of combiners and second stage of combiners comprises a first plurality of multiplication/accumulation units, the first plurality of multiplication/accumulation units each configured to multiply at least one bit of the encoded input data with at least one of the set of predetermined weights and sum multiple weighted bits of the encoded input data.
5. The apparatus of claim 4, wherein the first stage of combiners further comprises a first plurality of table look-ups, the first plurality of table look-ups each configured to look-up at least one intermediate data value corresponding to an output of a respective one of the first plurality of multiplication/accumulation units based on at least one non-linear function.
4. The apparatus of claim 3, wherein the first stage of combiners and second stage of combiners comprises a first plurality of multiplication/accumulation units, the first plurality of multiplication/accumulation units each configured to multiply at least one bit of the encoded input data with at least one of the set of predetermined weights and sum multiple weighted bits of the encoded input data.
5. The apparatus of claim 4, wherein the first stage of combiners further comprises a first plurality of table look-ups, the first plurality of table look-ups each configured to look-up at least one intermediate data value corresponding to an output of a respective one of the first plurality of multiplication/accumulation units based on at least one non-linear function.
7. The apparatus of claim 1, wherein the set of predetermined weights is based at least in part on an encoding technique associated with the encoded input data.
6. The apparatus of claim 1, wherein the set of predetermined weights based at least in part on the encoding technique associated with the encoded input data.
8. The apparatus of claim 7, wherein the encoding technique comprises Reed-Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof.
8. The apparatus of claim 1, wherein the encoding technique comprises Reed-Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof.
9. The apparatus of claim 1, wherein the set of predetermined weights are based on training of a neural network using known noisy encoded data and encoded data pairs.
7. The apparatus of claim 1, wherein the set of predetermined weights is based at least in part on an encoding technique associated with the encoded input data.
10. The apparatus of claim 1, further comprising: an encoder configured to encode the input data using encoded bits in accordance with an encoding technique and to provide the encoded input data; and a memory configured to receive the encoded input data from the encoder and configured to store the encoded input data, wherein, in storing the encoded input data, noise is introduced into the encoded input data.
1. An apparatus comprising: an encoder configured to encode input data using encoded bits in accordance with an encoding technique and to provide encoded input data; a memory configured to receive the encoded input data from the encoder and configured to store the encoded input data; and combiners configured to receive the encoded input data from the memory and further configured to combine the encoded input data among a set of predetermined weights, wherein the combiners are further configured to provide encoded data with reduced noise, the noise introduced by the memory.
11. A method comprising: transmitting signaling, from a memory of a computing device, indicative of data encoded an encoding technique; and modifying, at a neural network of the computing device, the data encoded with an encoding technique using a set of weights to provide encoded data with reduced noise.
11. A method comprising: passing known encoded data through noisy channels to provide known noisy encoded data, wherein the known encoded data is encoded with a particular encoding technique; receiving, at a computing device that comprises a neural network, signaling indicative of a set of encoded data pairs comprising the known encoded data; determining, for the neural network, a set of weights that modifies the known noisy encoded data using the signaling indicative of the set of encoded data pairs; transmitting signaling, from a memory of the computing device, indicative of data encoded with the particular encoding technique; modifying, at the neural network, the data encoded with the particular encoding technique using the set of weights to provide encoded data with reduced noise; and writing the encoded data with reduced noise to or reading the encoded data with reduced noise from a memory or storage medium of the computing device.
12. The method of claim 11, further comprising: receiving, at the computing device, signaling indicative of a set of encoded data pairs comprising encoded data; and determining, for the neural network, the set of weights that modifies the encoded data using the signaling indicative of the set of encoded data pairs.
11. A method comprising: passing known encoded data through noisy channels to provide known noisy encoded data, wherein the known encoded data is encoded with a particular encoding technique; receiving, at a computing device that comprises a neural network, signaling indicative of a set of encoded data pairs comprising the known encoded data; determining, for the neural network, a set of weights that modifies the known noisy encoded data using the signaling indicative of the set of encoded data pairs; transmitting signaling, from a memory of the computing device, indicative of data encoded with the particular encoding technique; modifying, at the neural network, the data encoded with the particular encoding technique using the set of weights to provide encoded data with reduced noise; and writing the encoded data with reduced noise to or reading the encoded data with reduced noise from a memory or storage medium of the computing device.
13. The method of claim 12, wherein determining the set of weights comprises selecting weights resulting in a minimized value of an error function between an output of the neural network and known noisy encoded data.
16. The method of claim 11, wherein determining the set of weights comprises selecting weights resulting in a minimized value of an error function between an output of the neural network and the known noisy encoded data.
15. The method of claim 11, wherein the encoded data with reduced noise is an estimate of the encoded data relative to output of an encoder associated with the encoding technique.
13. The method of claim 11, wherein the encoded data with reduced noise is an estimate of the encoded data relative to output of an encoder associated with the encoding particular technique.
16. The method of claim 11, wherein the encoding technique comprises Reed- Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof.
15. The method of claim 11, wherein the particular encoding technique comprises Reed-Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof.
18. A memory system comprising: one or more output buffers configured to transmit noisy output data; and a neural network configured to receive the noisy output data, and configured to utilize initial weights selected based on encoded data pairs to provide an estimate of encoded data with reduced noise.
17. A memory system comprising: memory cells configured to store encoded data associated with an encoding technique; output buffers configured to read selected portions of the encoded data from the memory cells and transmit corresponding output data, wherein, in transmitting the corresponding output data, noise is introduced into the output data; and a neural network configured to utilize initial weights selected based on encoded data pairs, the neural network configured to receive the output data and provide an estimate of encoded data with reduced noise, the noise introduced in transmitting the output data from the output buffers.
19. The memory system of claim 18, wherein the noise is introduced in transmitting the output data from the output buffers.
17. A memory system comprising: memory cells configured to store encoded data associated with an encoding technique; output buffers configured to read selected portions of the encoded data from the memory cells and transmit corresponding output data, wherein, in transmitting the corresponding output data, noise is introduced into the output data; and a neural network configured to utilize initial weights selected based on encoded data pairs, the neural network configured to receive the output data and provide an estimate of encoded data with reduced noise, the noise introduced in transmitting the output data from the output buffers.
Claims 14 & 20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 12 & 17 of U.S. Patent No. 11416735 in view of Nachmani et al. ("Learning to Decode Linear Codes Using Deep Learning", Fifty-fourth Annual Allerton Conference, IEEE Xplore) (Year: 2016), hereafter Nachmani. Although the claims at issue are not identical, they are not patentably distinct from each other because the claims of the present invention are similar in scope to the claims of U.S. Patent No. 11416735. For example, the table below shows similarities and differences between the instant application and the U.S. Patent No. 11416735.
18/179,317 (Instant Application)
US 11416735 B2
14. The method of claim 11, wherein modifying the data using the neural network using the set of weights to provide encoded data with reduced noise comprises: combining the data encoded with the encoding technique among the set of weights to provide the encoded data with reduced noise using a plurality of layers of the neural network, comprising an input layer, a hidden layer, an output layer, or combinations thereof.
12. The method of claim 11, wherein modifying the data using the neural network using the set of weights to provide encoded data with reduced noise comprises: combining the data encoded with the particular encoding technique among the set of weights to provide the encoded data with reduced noise.
20. The memory system of claim 18, wherein the neural network is configured to use multiple stages of nodes to provide the estimate of the encoded data, the multiple stages of nodes comprising an input stage, a hidden stage, an output stage, or combinations thereof.
17. A memory system comprising: memory cells configured to store encoded data associated with an encoding technique; output buffers configured to read selected portions of the encoded data from the memory cells and transmit corresponding output data, wherein, in transmitting the corresponding output data, noise is introduced into the output data; and a neural network configured to utilize initial weights selected based on encoded data pairs, the neural network configured to receive the output data and provide an estimate of encoded data with reduced noise, the noise introduced in transmitting the output data from the output buffers.
The U.S. Patent No. 11416735 did not teach using a plurality of layers of the neural network, comprising an input layer, a hidden layer, an output layer, or combinations thereof; wherein the neural network is configured to use multiple stages of nodes; the multiple stages of nodes comprising an input stage, a hidden stage, an output stage, or combinations thereof as recited in claims 14 and 20 of the instant application.
However, Nachmani teaches: using a plurality of layers of the neural network ([Sec. 2] discusses the use of an input layer and hidden layer); the neural network is configured to use multiple stages of nodes…comprising an input stage, a hidden stage, an output stage, or combinations thereof ([Sec. 2] discusses the use of an input layer and hidden layer, which constitute stages of nodes; [Sec. 1 & 3, Eq. 6] discusses the output layer modifies data to compute a weighted sum of hidden-layer outputs passed through a sigmoid to produce the estimate of the encoded data).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the concept of using a plurality of layers of the neural network, and multiple stages of nodes acting as layers as suggested by Nachmani into the instant application because both of these systems are addressing decoding encoded data. Doing so would be motivated by the desire to employ a Tanner graph allowing for improved decoding, wherein the edges of the Tanner graph are corresponding to the hidden layers (Nachmani [Sec. 1]).
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 11-17 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claim 11
Step 1: The claim recites “A method comprising:”; therefore, it is directed to the statutory category of a process.
Step 2A Prong 1: The claim recites, inter alia:
and modifying, at a neural network of the computing device, the data encoded with an encoding technique using a set of weights to provide encoded data with reduced noise: These limitations recite mathematical relationships similar to organizing information and manipulating information through mathematical correlations per MPEP 2106.04(a)(2)(I)(A)(iv).
Thus, the claim recites a judicial exception.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
transmitting signaling, from a memory of a computing device, indicative of data encoded an encoding technique: These additional elements amount to insignificant extra-solution activity in the form of mere data gathering per MPEP § 2106.05(g).
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include insignificant extra-solution activity of data gathering recited by “transmitting signaling, from a memory of a computing device, indicative of data encoded an encoding technique” which are well-understood routine and conventional activities similar to storing and retrieving information in memory per MPEP 2106.05(d)(II). Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 12
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 11, as well as, inter alia:
and determining, for the neural network, the set of weights that modifies the encoded data using the signaling indicative of the set of encoded data pairs: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to determine the set of weights that modifies the encoded data using the signaling indicative of the set of encoded data pairs.
Thus, the claim recites a judicial exception.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
receiving, at the computing device, signaling indicative of a set of encoded data pairs comprising encoded data: These additional elements amount to insignificant extra-solution activity in the form of mere data gathering per MPEP § 2106.05(g).
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include insignificant extra-solution activity of data gathering recited by “receiving, at the computing device, signaling indicative of a set of encoded data pairs comprising encoded data” which are well-understood routine and conventional activities similar to presenting offers and gathering statistics per MPEP 2106.05(d)(II). Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 13
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 12, as well as, inter alia:
wherein determining the set of weights comprises selecting weights resulting in a minimized value of an error function between an output of the neural network and known noisy encoded data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to select weights that result in a minimized value of an error function between an output of the neural network and known noisy encoded data.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a
practical application and is not considered to be significantly more. As such, the claim is patent
ineligible.
Claim 14
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 11.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
combining the data encoded with the encoding technique among the set of weights to provide the encoded data with reduced noise using a plurality of layers of the neural network, comprising an input layer, a hidden layer, an output layer, or combinations thereof: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. combining the data encoded with the encoding technique among the set of weights to provide the encoded data with reduced noise, to a particular technological environment or field of use, e.g. uses a plurality of layers of the neural network, comprising an input layer, a hidden layer, an output layer, or combinations thereof. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 15
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 11.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
wherein the encoded data with reduced noise is an estimate of the encoded data relative to output of an encoder associated with the encoding technique: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. the encoded data with reduced noise, to a particular technological environment or field of use, e.g. is an estimate of the encoded data relative to output of an encoder associated with the encoding technique. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 16
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 11.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
wherein the encoding technique comprises Reed- Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. the encoding technique, to a particular technological environment or field of use, e.g. comprises Reed- Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 17
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 16.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows: wherein the neural network is trained multiple times, once for each encoding technique used: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. the neural network, to a particular technological environment or field of use, e.g. is trained multiple times, once for each encoding technique used. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
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.
Claims 1-4 & 9 are rejected under 35 U.S.C. 103 as being unpatentable over Dodgson et al. (US 7502766 B2, published 03/10/2009), hereafter Dodgson, in view of Nachmani et al. ("Learning to Decode Linear Codes Using Deep Learning", Fifty-fourth Annual Allerton Conference, IEEE Xplore) (Year: 2016), hereafter Nachmani.
Regarding independent claim 1, Dodgson teaches a system for a neural network decoder comprising:
a first stage of combiners and further configured to implement a first function to provide first intermediate data ([Col. 4, Lines 23-49] discusses a set of layers, including an input layer, functioning as combiners receiving data, then applying a non-linear transfer function to output the first data to another stage);
and a second stage of combiners configured to receive the first intermediate data and further configured to combine the first intermediate data using a set of predetermined weights ([Col. 1, Lines 51-53] discusses the second stage is connected to receive the output of the first stage; [Col. 4, Lines 23-49] discusses weights are initialized and are combined with the data).
Dodgson does not explicitly teach a first stage of combiners configured to receive encoded input data; using a set of predetermined weights to provide the encoded data with reduced noise.
However, in the same field of endeavor, Nachmani teaches a system for decoding encoded data, by taking encoded data ([Sec. 2] discusses the input composing encoded input data); and outputting encoded data with reduced noise ([Sec. 1 & 3, Eq. 6] discusses the output layer computes a weighted sum of hidden-layer outputs passed through a sigmoid to produce a transmitted codeword with reduced noise).
Because Dodgson teaches a first stage of combiners configured to receive data, and implement a first function to provide first data, and a second stage of combiners configured to receive the first data and combine the data with predetermined weights; and Nachmani teaches the system receiving encoded data and providing the encoded data with reduced noise, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate receiving encoded data and providing the encoded data with reduced noise as taught by Nachmani into Dodgson’s system, with a reasonable expectation of success, to teach a first stage of combiners configured to receive encoded input data and further configured to implement a first function to provide first intermediate data; and a second stage of combiners configured to receive the first intermediate data and further configured to combine the first intermediate data using a set of predetermined weights to provide the encoded data with reduced noise. This combination would have been motivated by the desire to reduce the amount of codewords required for the proper training process (Nachmani [Sec. 1]).
Regarding dependent claim 2, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 1, including a third stage of combiners configured to receive the first intermediate data and implement a second function to provide second intermediate data to the second stage of combiners (Nachmani [Sec. 2-3, Eq. 4-5] discusses that a separate stage receives first input and implements a second function to provide more data to the second stage of combiners; Nachmani [Sec. 4] discusses a neural network with 10 layers will have 5 full iterations and thus, there are two stages configured to provide intermediate data through a second function to the second stage of combiners).
Regarding dependent claim 3, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 1, including wherein the first function is a nonlinear function (Dodgson [Col. 4, Lines 47-49] discusses the first function applied is a non-linear function).
Regarding dependent claim 4, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 1, including wherein the first stage of combiners and second stage of combiners comprises a first plurality of multiplication/accumulation units, the first plurality of multiplication/accumulation units each configured to multiply at least one bit of the encoded input data with at least one of the set of predetermined weights and sum multiple weighted bits of the encoded input data (Dodgson [Col. 4, Lines 32-38] discusses a multiplier bank feeding a summation node, which are multiplication and accumulation units; Dodgson [Col. 1, Lines 44-48] discusses multiplying input data by their corresponding weights and summing the weighted values).
Regarding dependent claim 9, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 1, including wherein the set of predetermined weights are based on training of a neural network using known noisy encoded data and encoded data pairs (Nachmani [Sec. 1 & 3] discusses the goal is to train parameters to achieve an output close to the zero codeword; the input being noisy encoded data and data pairs; thus, the weights are based on the training process of the neural network rather than being established beforehand).
Claims 5-6 are rejected under 35 U.S.C. 103 as being unpatentable over Dodgson, in view of Nachmani as applied in claim 4, and further in view of Shen et al. (US 2016/0148078, published 05/26/2016), hereafter Shen.
Regarding dependent claim 5, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 4, including:
one of the first plurality of multiplication/accumulation units (Dodgson [Col. 4, Lines 32-38] discusses a multiplier bank feeding a summation node, which are multiplication and accumulation units);
based on at least one nonlinear function (Dodgson [Col. 4, Lines 47-49] discusses the first function applied is a non-linear function).
The combination of Dodgson and Nachmani does not explicitly teach wherein the first stage of combiners further comprises a first plurality of table look-ups, the first plurality of table look-ups each configured to look-up at least one intermediate data value corresponding to an output.
However, in a similar field of endeavor, Shen teaches a system for analyzing neural network inputs and storing values within a table look-up ([0013] discusses the outputs, or possible outputs, can be stored in a look-up table; then when operation of the neural network occurs, the results can be obtained from the look-up table).
Because the combination of Dodgson and Nachmani teaches multiplication/accumulation units and the use of a nonlinear function; and Shen teaches the use of table look-ups, wherein the intermediate data values correspond to an output, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate a plurality of table look-ups as taught by Shen into the combination of Dodgson & Nachmani’s system, with a reasonable expectation of success, to teach wherein the first stage of combiners further comprises a first plurality of table look-ups, the first plurality of table look-ups each configured to look-up at least one intermediate data value corresponding to an output of a respective one of the first plurality of multiplication/accumulation units based on at least one non-linear function. This combination would have been motivated by the desire to increase the speed of operation by pre-filling the look-up table (Shen [0013]).
Regarding dependent claim 6, the combination of Dodgson, Nachmani, & Shen teaches the invention as claimed in claim 5, including wherein the at least one non-linear function comprises a Gaussian function, a piece-wise linear function, a sigmoid function, a thin- plate-spline function, a multiquadratic function, a cubic approximation, an inverse multi-quadratic function, or combinations thereof (Dodgson [Col. 1, Lines 49-53] discusses the non-linear function is typically a sigmoid function).
Claims 7-8 are rejected under 35 U.S.C. 103 as being unpatentable over Dodgson, in view of Nachmani as applied in claim 1, and further in view of Lugosch et al. ("Learning from the Syndrome", McGill University, arXiv) (Year: 2018), hereafter Lugosch.
Regarding dependent claim 7, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 1, including a second stage of combiners configured to receive the first intermediate data and further configured to combine the first intermediate data using a set of predetermined weights ([Col. 1, Lines 51-53] discusses the second stage is connected to receive the output of the first stage; [Col. 4, Lines 23-49] discusses weights are initialized and are combined with the data).
The combination of Dodgson and Nachmani does not explicitly teach wherein the set of predetermined weights is based at least in part on an encoding technique associated with the encoded input data.
However, in the same field of endeavor, Lugosch teaches a system for decoders that determines weights based on an encoding technique ([Sec. 4] discusses the use of four codes which comprise techniques for coding and encoding; each of these codes necessarily has its own trained weight set, due to the fact that each code has its own parity-check structure).
Because the combination of Dodgson and Nachmani teaches the use of predetermined weights; and Lugosch teaches the predetermined weights are based in part on an encoding technique associated with the encoded input data, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate predetermined weights being determined based in part on an encoding technique associated with the encoded input data as taught by Lugosch into the combination of Dodgson & Nachmani’s system, with a reasonable expectation of success, to teach wherein the set of predetermined weights is based at least in part on an encoding technique associated with the encoded input data. This combination would have been motivated by the desire to compare performance for decoders with learning where weights are different and without learning where weights are all equal (Lugosch [Sec. 4]).
Regarding dependent claim 8, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 7, including wherein the encoding technique comprises Reed-Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof (Lugosch [Sec. 4] discusses the encoding technique comprising BCH, LDPC, and polar coding; Nachmani [Sec. 1] discusses the use of BCH and LDPC encoding techniques).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Dodgson, in view of Nachmani as applied in claim 1, and further in view of Kumar et al. (US 2018/0343017 A1, published 11/29/2018), hereafter Kumar.
Regarding dependent claim 10, the combination of Dodgson and Nachmani teaches the invention as claimed in claim 1, including receiving the encoded input data (Dodgson [Col. 4, Lines 23-49] discusses a set of layers, including an input layer, functioning as combiners receiving data, then applying a non-linear transfer function to output the first data to another stage; Nachmani [Sec. 2] discusses the input composing encoded input data).
The combination of Dodgson and Nachmani does not explicitly teach an encoder configured to encode the input data using encoded bits in accordance with an encoding technique and to provide the encoded input data; and a memory configured to receive the encoded input data from the encoder and configured to store the encoded input data, wherein, in storing the encoded input data, noise is introduced into the encoded input data.
However, in the same field of endeavor, Kumar teaches an encoder that encodes input data and stores it in memory ([0032-0034] discusses encoding input data using bits and a LDPC technique and storing the encoded input data in memory, and during this, noise is introduced into the encoded input data).
Because the combination of Dodgson and Nachmani teaches receiving encoded input data; and Kumar teaches an encoder configured to encode input data using bits and provide the encoded data to a memory while introducing noise, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate an encoder configured to encode input data using bits and provide the encoded data to a memory while introducing noise as taught by Kumar into the combination of Dodgson & Nachmani’s system, with a reasonable expectation of success, to teach an encoder configured to encode the input data using encoded bits in accordance with an encoding technique and to provide the encoded input data; and a memory configured to receive the encoded input data from the encoder and configured to store the encoded input data, wherein, in storing the encoded input data, noise is introduced into the encoded input data. This combination would have been motivated by the desire to implement the system into hardware with memory (Kumar [0001-0007]).
Claims 11-14 & 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Nachmani et al. ("Learning to Decode Linear Codes Using Deep Learning", Fifty-fourth Annual Allerton Conference, IEEE Xplore) (Year: 2016), hereafter Nachmani, in view of Kumar et al. (US 2018/0343017 A1, published 11/29/2018), hereafter Kumar.
Regarding independent claim 11, Nachmani teaches a method for decoding encoded input data comprising modifying, at a neural network of the computing device, the data encoded with an encoding technique using a set of weights to provide encoded data with reduced noise ([Sec. 1 & 3, Eq. 6] discusses the output layer modifies data using an encoding technique and a set of weights to compute a weighted sum of hidden-layer outputs passed through a sigmoid to produce a transmitted codeword with reduced noise).
Nachmani does not explicitly teach transmitting signaling, from a memory of a computing device, indicative of data encoded an encoding technique.
However, in a similar field of endeavor, Kumar teaches a method for encoding information and transmitting it from memory ([0032-0034] discusses storage will transmit data that has been encoded by a LDPC encoder to the receiver).
Because Nachmani teaches modifying data encoded using a set of weights to provide encoded data with reduced noise; and Kumar teaches transmitting signaling indicative of data encoded by an encoding technique, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate transmitting signaling, from memory, indicative of data encoded as taught by Kumar into Nachmani’s system, with a reasonable expectation of success, to teach transmitting signaling, from a memory of a computing device, indicative of data encoded an encoding technique; and modifying, at a neural network of the computing device, the data encoded with an encoding technique using a set of weights to provide encoded data with reduced noise. This combination would have been motivated by the desire to implement Nachmani’s method into hardware with memory (Kumar [0001-0007]).
Regarding dependent claim 12, the combination of Nachmani and Kumar teaches the invention as claimed in claim 11, including receiving, at the computing device, signaling indicative of a set of encoded data pairs comprising encoded data; and determining, for the neural network, the set of weights that modifies the encoded data using the signaling indicative of the set of encoded data pairs (Nachmani [Sec. 3] discusses using a set of weights as parameters to achieve N dimensional output, wherein the weights are determined using the training data which corresponds to signaling indicative of encoded data pairs; Kumar [0059] discusses the weights are derived from the training of the neural network; Kumar [0032-0034] discusses storage will transmit data that has been encoded by a LDPC encoder to the receiver; Kumar [0024] discusses training bits are separated between error bits and correct bits, which represent the encoded data pairs; thus, the receiver receives signaling indicative of a set of encoded data pairs).
Regarding dependent claim 13, the combination of Nachmani and Kumar teaches the invention as claimed in claim 12, including wherein determining the set of weights comprises selecting weights resulting in a minimized value of an error function between an output of the neural network and known noisy encoded data (Nachmani [Sec. 3] discusses using stochastic gradient descent which adjusts weights to minimize a loss function, as well as selecting weights in the parameters; Nachmani [Sec. 4] discusses applying a cross-entropy loss using the gradient descent to minimize the loss between network output and the known codeword).
Regarding dependent claim 14, the combination of Nachmani and Kumar teaches the invention as claimed in claim 11, including wherein modifying the data using the neural network using the set of weights to provide encoded data with reduced noise comprises: combining the data encoded with the encoding technique among the set of weights to provide the encoded data with reduced noise using a plurality of layers of the neural network, comprising an input layer, a hidden layer, an output layer, or combinations thereof (Nachmani [Sec. 2] discusses the use of an input layer and hidden layer; Nachmani [Sec. 1 & 3, Eq. 6] discusses the output layer modifies data using an encoding technique and a set of weights to compute a weighted sum of hidden-layer outputs passed through a sigmoid to produce a transmitted codeword with reduced noise).
Regarding dependent claim 16, the combination of Nachmani, Kumar teaches the invention as claimed in claim 11, including wherein the encoding technique comprises Reed- Solomon coding, Bose-Chaudhuri-Hocquenghem (BCH) coding, low-density parity check (LDPC) coding, Polar coding, or combinations thereof (Nachmani [Sec. 1] discusses the use of BCH and LDPC encoding techniques).
Regarding independent claim 18, Nachmani teaches a system for decoding encoded input data comprising:
one or more output buffers configured to utilize initial weights selected based on encoded data pairs to provide an estimate of encoded data with reduced noise ([Sec. 1 & 3, Eq. 6] discusses the output layer modifies data using an encoding technique and a set of weights that has been initialized to compute a weighted sum of hidden-layer outputs passed through a sigmoid to produce an estimate of a transmitted codeword with reduced noise);
a neural network configured to receive the noisy output data ([Sec. 2] discusses the neural network input composing noisy encoded input data).
Nachmani does not explicitly teach one or more output buffers configured to transmit noisy output data.
However, in a similar field of endeavor, Kumar teaches a method for encoding information and transmitting the noisy information from memory ([0032-0034] discusses a storage location feeds a detector with the output data, and when the detector receives this data, it comprises noise and errors; and discusses storage will transmit data that has been encoded by a LDPC encoder to the receiver).
Because Nachmani teaches a neural network configured to receive noisy output data, and output buffers configured to utilize initial weights selected to provide an estimate of encoded data with reduced noise; and Kumar teaches transmitting noisy output data, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate transmitting noisy output data as taught by Kumar into Nachmani’s system, with a reasonable expectation of success, to teach one or more output buffers configured to transmit noisy output data; and a neural network configured to receive the noisy output data, and configured to utilize initial weights selected based on encoded data pairs to provide an estimate of encoded data with reduced noise. This combination would have been motivated by the desire to implement Nachmani’s method into hardware with memory and transmit encoded information to the system (Kumar [0001-0007 & 0030-0034]).
Regarding dependent claim 19, the combination of Nachmani and Kumar teaches the invention as claimed in claim 18, including wherein the noise is introduced in transmitting the output data from the output buffers (Kumar [0032-0034] discusses a storage location feeds a detector with the output data, and when the detector receives this data, it comprises noise and errors that was introduced during transmission).
Regarding dependent claim 20, the combination of Nachmani and Kumar teaches the invention as claimed in claim 18, including wherein the neural network is configured to use multiple stages of nodes to provide the estimate of the encoded data, the multiple stages of nodes comprising an input stage, a hidden stage, an output stage, or combinations thereof (Nachmani [Sec. 2] discusses the use of an input layer and hidden layer, which constitute stages of nodes; Nachmani [Sec. 1 & 3, Eq. 6] discusses the output layer modifies data to compute a weighted sum of hidden-layer outputs passed through a sigmoid to produce the estimate of the encoded data).
Claims 15 & 17 are rejected under 35 U.S.C. 103 as being unpatentable over Nachmani, in view of Kumar as applied in claim 11, and further in view of Lugosch et al. ("Learning from the Syndrome", McGill University, arXiv) (Year: 2018), hereafter Lugosch.
Regarding dependent claim 15, the combination of Nachmani and Kumar teaches the invention as claimed in claim 11, including providing the encoded data with reduced noise ([Sec. 1 & 3, Eq. 6] discusses the output layer modifies data using an encoding technique and a set of weights to compute a weighted sum of hidden-layer outputs passed through a sigmoid to produce a transmitted codeword with reduced noise).
The combination of Nachmani and Kumar does not explicitly teach wherein the encoded data with reduced noise is an estimate of the encoded data relative to output of an encoder associated with the encoding technique.
However, in a similar field of endeavor, Lugosch teaches a system for error-correcting decoders wherein the encoded data is an estimate of the encoded data relative to output of an encoder associated with the encoding technique ([Sec. 3] discusses the output is designed to be a soft estimate of the transmitted codeword; thus, the encoded data with reduced noise is an estimate of the encoded data relative to output of an encoder associated with the encoding technique).
Because the combination of Nachmani and Kumar teaches providing encoded data with reduced noise; and Lugosch teaches the encoded data with reduced noise is designed to be an estimate of the encoded data relative to output, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the encoded data with reduced noise is designed to be an estimate of the encoded data relative to output as taught by Lugosch into the combination of Nachmani and Kumar’s system, with a reasonable expectation of success, to teach wherein the encoded data with reduced noise is an estimate of the encoded data relative to output of an encoder associated with the encoding technique. This combination would have been motivated by the desire to implement the decoder in multiple formats and forms as Lugosch argues if the output is a soft estimate of the transmitted codeword, multiple tasks can be achieved by the decoder (Lugosch [Sec. 3]).
Regarding dependent claim 17, the combination of Nachmani and Kumar teaches the invention as claimed in claim 16, including the use of multiple encoding techniques (Nachmani [Sec. 1] discusses the use of BCH and LDPC encoding techniques; and tests the BCH encoding technique).
The combination of Nachmani and Kumar does not explicitly teach wherein the neural network is trained multiple times, once for each encoding technique used.
However, in a similar field of endeavor, Lugosch teaches a system for training a neural network based on encoding techniques used ([Sec. 4] discusses the use of four codes for training which comprise techniques for coding and encoding; each of these codes represents a different encoding technique).
Because the combination of Nachmani and Kumar teaches use of multiple encoding techniques; and Lugosch teaches training the neural network multiple times, once for each encoding technique used, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate training the neural network for each encoding technique as taught by Lugosch into the combination of Nachmani and Kumar’s system, with a reasonable expectation of success, to teach wherein the neural network is trained multiple times, once for each encoding technique used. This combination would have been motivated by the desire to study the effect of each decoder (Lugosch [Sec. 4]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Sharon et al. (US 2018/0159553 A1, published 06/07/2018) ([Abstract] A device includes a low density parity check (LDPC) decoder that configured to receive a representation of a codeword. The LDPC decoder includes a message memory configured to store decoding messages, multiple data processing units (DPUs), a control circuit, and a reording circuit. The control circuit is configured to enable a first number of the DPUs to decode the representation of the codeword in response to a decoding mode indicator indicating a first decoding mode and to enable a second number of the DPUs to decode the representation of the codeword in response to the decoding mode indicator indicating a second decoding mode. The reordering circuit is configured to selectively reorder at least one of the decoding messages based on the decoding mode indicator).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RILEY S ACOSTA whose telephone number is (571)272-8714. The examiner can normally be reached Monday-Thursday 6am-4pm.
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, Jennifer N Welch can be reached at (571)272-7212. 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.
/RILEY S ACOSTA/Examiner, Art Unit 2143
/JENNIFER N WELCH/Supervisory Patent Examiner, Art Unit 2143
Prosecution Insights