DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 1-20 are presented for examination.
The claims and only the claims form the metes and bounds of the invention. “Office personnel are to give claims their broadest reasonable interpretation in light of the supporting disclosure. In re Morris, 127 F.3d 1048, 1054-55, 44 USPQ2d 1023, 1027-28 (Fed. Cir. 1997). Limitations appearing in the specification but not recited in the claim are not read into the claim. In re Prater, 415 F.2d 1393, 1404-05, 162 USPQ 541, 550-551 (CCPA 1969)” (MPEP p 2100-8, c 2, I 45-48; p 2100-9, c 1, l 1-4). The Examiner has full latitude to interpret each claim in the broadest reasonable sense. The Examiner will reference prior art using terminology familiar to one of ordinary skill in the art. Such an approach is broad in concept and can be either explicit or implicit in meaning.
Response to Argument
Applicant's arguments have been considered but they are not persuasive. However, the Examiner welcomes any suggestion(s) Applicant may have on moving prosecution forward. The Examiner's contact information is in the Conclusion of this office action.
Applicant argues:
As detailed in the Office Action and as elaborated by the Examiner during the telephone interview, the Examiner's position is that Wang et al. teaches "summing" because Wang et al. performs a cross-point calculation operation. Applicant respectfully disagrees. As explained by the undersigned attorney during the telephone interview, a cross-point calculation does NOT involve summing, as it merely identifies the intersection point between two curves. More specifically, Figure 13 and the accompanying text in Wang et al. teach that the cross-point calculation module receives probability density functions (PDFs) of two candidate read threshold voltages and determines the optimal read threshold voltage (Vt __ opt) by finding the intersection (the cross-point) of those two curves. Figure 15 shows a graph of this cross-point calculation operation, where the optimal read threshold voltage is depicted at the intersection of the two curves. There is no summation involved in this cross-point calculation, as it only involves finding the intersection of two curves and not the summation of values. This is shown in the graph of Figure 15, where the optimal read threshold voltage is stated to be the point of intersection (0.29151)- not the summation of those two points (0.29151 plus 0.29151).
In response, the Examiner submits:
Independent claim 1 recites “obtain an inferred read threshold by summing the inference results of the plurality of trees”.
Sum can mean an aggregate or combination. The recited “summing” does not necessarily mean summation that is the result of addition operation.
At least Fig. 16 of Wang et al. shows inferred ⊖R* and ⊖L* that combine to result in an inferred read threshold (an optimal read threshold voltage). Wang’s cross-point voltage (Wang: at least ¶0111) results from aggregate or combination of voltage ranges (Wang: at least ¶0128).
Applicant further argues:
“Independent Claims 13 and 20 were rejected under 35 U.S.C. § 102 as being anticipated by Wood et al. Without acquiescing to those rejections and merely to expedite prosecution of this application, Applicant is amending independent Claims 13 and 20 to recite an element that is clearly not present in Wood et al. More specifically, amended independent Claims 13 and 20 now recite determining whether a magnitude of a correction between the read threshold calibration result and a previous read threshold calibration result is below a threshold. If the magnitude of the correction is below the threshold, the memory is read using the read threshold calibration result. However, if the magnitude of the correction is not below the threshold, the read threshold calibration unit is operated at least one additional time until the magnitude of the correction is below the threshold”.
In response, the Examiner submits:
Contrary to Applicant’s allegation above, Wood does teach the recited limitations of determining whether a magnitude of a correction between the read threshold calibration result and a previous read threshold calibration result is below a threshold (Wood: at least ¶¶0104, 0106; “a deviation from the known bias caused by errors in the data set is an error bias” and “determine whether the re-read data set has a read bias that deviates from the known bias, and may iteratively adjust the read voltage threshold to a new read voltage threshold until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”);
in response to determining that the magnitude of the correction is below the threshold (Wood: at least ¶0106; “… until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”; ¶0159 further discloses “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias”), reading the memory using the read threshold calibration result (Wood: at least ¶0099; “three example read voltage thresholds may be x volts, y volts, and z volts, described in greater detail below with regard to the read voltage thresholds 662 of FIG. 6C. If the voltage read from a storage cell falls between Vmin and x volts, a binary 11 state is indicated. In certain embodiments, Vmin may be a negative voltage. If the voltage read from a storage cell falls between x volts and y volts, a binary 01 state is indicated. If the voltage read from a storage cell falls between y volts and z volts, a binary 00 state is indicated”); and
in response to determining that the magnitude of the correction is not below the threshold, operating the read threshold calibration unit at least one additional time until the magnitude of the correction is below the threshold (Wood: at least ¶0106; “… may iteratively adjust the read voltage threshold to a new read voltage threshold until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”; ¶0159 further discloses “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias”).
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 13-20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claims 13 and 20 recite “in response to determining that the magnitude of the correction is below the threshold, reading the memory using the read threshold calibration result; and
in response to determining that the magnitude of the correction is not below the
threshold operating the read threshold calibration unit at least one additional time until the magnitude of the correction is below the threshold”.
Applicant’s original disclosure discloses:
[0048] “In another embodiment, the read threshold activation control unit 400 can either decide on the number of activation times before the entire calibration operation or on-the-fly based on the feedback from the read threshold calibration unit 410. For example, if the change magnitude of the read threshold is large (i.e., above a certain threshold), there may be more merit to additional activations of the calibration unit to refine the results as the optimal read threshold are far from the current read threshold. More specifically, the tree-based model can be trained such that each tree corrects for the residual error of the inference model based on the previous trees. In this case, the correction applied by each consecutive tree is expected to be of lower magnitude. Hence, the decision on the number of trees applied in a specific inferencing operation may be dynamic based on the magnitude of correction introduced by the last set of k trees that was calculated in the previous round. Once the magnitude of correction drops below a predefined threshold, the inferencing operation can be deemed as accurate enough and terminated”.
Applicant’s original specification appears to teach “in response to determining that the magnitude of the correction is above the threshold operating the read threshold calibration unit at least one additional time until the magnitude of the correction is below the threshold” instead of the recited “in response to determining that the magnitude of the correction is not below the threshold operating the read threshold calibration unit at least one additional time until the magnitude of the correction is below the threshold”.
Claims 14-19 depend from claim 13 and are rejected for the same reason(s) under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless -
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-2, 9 and 11 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by USPGPUB 2023/0035983 by Wang et al. (“Wang”).
As to Claim 1, Wang teaches a data storage device comprising: a memory (Wang: at least ¶0027; “a system; a computer program product embodied on a computer-readable storage medium; and/or a processor, such as a processor suitable for executing instructions stored on and/or provided by a memory coupled to the processor”); and one or more processors, individually or in combination (Wang: at least ¶0027; “… a processor, such as a processor suitable for executing instructions stored on and/or provided by a memory coupled to the processor”), configured to: use a tree-based inferencing model to generate a plurality of trees, wherein each tree produces an inference result based on a plurality of read thresholds (Wang: at least ¶¶0083-0084; “provide a scheme for estimating a read threshold voltage in order to overcome the weakness of all existing algorithms” and “use deep learning and provide a parametric framework for program voltage or program verify (PV)-level modeling and optimal read threshold voltage (Vt) estimation”; ¶0091 further discloses “optimal read threshold determiner 1030 may provide a parametric framework for program voltage or program verify (PV)-level modeling and optimal read threshold voltage (Vt) estimation. The optimal read threshold determiner 1030 may be implemented with one or more deep neural networks (DNNs)”; ¶0137 further discloses “IDNN3 1620 may receive the connection vectors ⊖.sub.L* and ⊖.sub.R* and generate an optimal read threshold voltage”; ¶0094-0095 further disclose “knowledge may be exchanged between nodes through node-to-node interconnections. Input to the neural network 1100 may activate a set of nodes. In turn, this set of nodes may activate other nodes, thereby propagating knowledge about the input. This activation process may be repeated across other nodes until nodes in the output layer 1130 are selected and activated” and “hierarchy of nodes interconnected in a feed-forward way. The input layer 1110 may exist at the lowest hierarchy level. The input layer 1110 may include a set of nodes that are referred to herein as input nodes. When the feature map 1102 is input to the neural network 1100, each of the input nodes of the input layer 1110 may be connected to each feature of the feature map 1102. Each of the connections may have a weight. These weights may be one set of parameters that are derived from the training of the neural network 1100. The input nodes may transform the features by applying an activation function to these features. The information derived from the transformation may be passed to the nodes at a higher level of the hierarchy”; ¶0099 further discloses “generally, the hidden layer(s) 1120 may allow knowledge about the input nodes of the input layer 1110 to be shared among the output nodes of the output layer 1130. To do so, a transformation ƒ may be applied to the input nodes through the hidden layer 1120. In an example, the transformation ƒ is non-linear. Different non-linear transformations ƒ are available including, for instance, a rectifier function ƒ(x)=max(0, x). In an example, a particular non-linear transformation ƒ is selected based on cross-validation. For example, given known example pairs (x,y), where xεX and yεY, a function ƒ: X.fwdarw.Y is selected when such a function results in the best matches”); obtain an inferred read threshold by summing the inference results of the plurality of trees (Wang: at least ¶0005; “combined neural network generates first and second connection vectors based on the first and second CDF values and first weight values, and estimates an optimal read threshold voltage based on the first and second connection vectors and second weight values”; ¶0091 also discloses “optimal read threshold determiner 1030 may be implemented with one or more deep neural networks (DNNs)”; ¶0133 further discloses “optimal read threshold determination apparatus 1600 may be implemented with a combined deep neural network (CDNN). CDNN 1600 may include a first internal deep neural networks (IDNN1) 1610A, a second internal deep neural networks (IDNN2) 1610B and a third internal deep neural network (IDNN3) 1620”; ¶¶0116 & 0118 further disclose “each DNN2 1300A, 1300B may determine PDF values of the distributions for various candidate read threshold voltages based on the estimated probability distribution parameters p. For example, DNN2 1300A may receive the estimated probability distribution parameters p.sub.A and determine PDF values PDF.sub.A based on the estimated probability distribution parameters p.sub.A. DNN2 1300B may receive the estimated probability distribution parameters p.sub.B and determine PDF values PDF.sub.B based on the estimated probability distribution parameters p.sub.B. In some embodiments, each DNN2 1300A, 1300B may determine PDF values of the distributions for various candidate read threshold voltages based on the estimated probability distribution parameters p” and ““the cross-point calculation component 1310 may determine the cross-point of the two candidate read threshold voltages as the optimal read threshold voltage Vt_opt”); and
read the memory using the inferred read threshold (Wang: at least ¶0075; “a read operation is performed on the memory array using a certain reference voltage such as a read threshold voltage (also called “read voltage level” or “read threshold”)”; ¶0122 also discloses “read operation using each read threshold voltage”; claim 2 further disclose “perform a next read operation on the cells using the optimal read threshold voltage”).
As to Claim 2, Wang teaches the data storage device of Claim 1, wherein the one or more processors, individually or in combination, are further configured to use the tree-based inferencing model to generate additional trees, wherein the inferred read threshold is obtained by summing the inference results of the plurality of trees and the additional trees (Wang: at least ¶0005; “combined neural network generates first and second connection vectors based on the first and second CDF values and first weight values, and estimates an optimal read threshold voltage based on the first and second connection vectors and second weight values”; ¶0091 also discloses “optimal read threshold determiner 1030 may be implemented with one or more deep neural networks (DNNs)”; ¶¶0133, 0137 further disclose “optimal read threshold determination apparatus 1600 may be implemented with a combined deep neural network (CDNN). CDNN 1600 may include a first internal deep neural networks (IDNN1) 1610A, a second internal deep neural networks (IDNN2) 1610B and a third internal deep neural network (IDNN3) 1620” and “IDNN3 1620 may receive the connection vectors ⊖.sub.L* and ⊖.sub.R* and generate an optimal read threshold voltage”; ¶¶0116 & 0118 further disclose “each DNN2 1300A, 1300B may determine PDF values of the distributions for various candidate read threshold voltages based on the estimated probability distribution parameters p. For example, DNN2 1300A may receive the estimated probability distribution parameters p.sub.A and determine PDF values PDF.sub.A based on the estimated probability distribution parameters p.sub.A. DNN2 1300B may receive the estimated probability distribution parameters p.sub.B and determine PDF values PDF.sub.B based on the estimated probability distribution parameters p.sub.B. In some embodiments, each DNN2 1300A, 1300B may determine PDF values of the distributions for various candidate read threshold voltages based on the estimated probability distribution parameters p” and ““the cross-point calculation component 1310 may determine the cross-point of the two candidate read threshold voltages as the optimal read threshold voltage Vt_opt”), which provides greater accuracy (Wang: at least ¶0133; “CDNN 1600 may have an architecture by combining IDNN1, IDNN2 and IDNN3 into a single network that can be trained in an end-to-end manner for better accuracy”).
As to Claim 9, Wang teaches the data storage device of Claim 1, wherein the one or more processors, individually or in combination, are further configured to apply performance throttling to allow additional bandwidth for obtaining the inferred read threshold using a full tree-based model (Wang: at least ¶0077; “one or more read retry operations for the memory cells using one or more read threshold voltages applied in a set order (S100). For example, the read threshold voltages may include N (e.g., N is 5 or 10) read threshold voltages (or read voltage levels) including a first read threshold voltage to an Nth read threshold voltage. The first read threshold voltage may be a previously used read threshold voltage (i.e., history read threshold voltage). The history read threshold voltage may be the read threshold voltage used in the last successful decoding, that is, a read voltage used in a read-passed read operation performed before the read retry operations. The controller 120 may perform the read retry operations until it is determined that decoding associated with a corresponding read retry operation is successful”).
As to Claim 11, Wang teaches the data storage device of Claim 1, wherein the one or more processors are implemented purely in hardware (Wang: at least ¶0027; “a system; a computer program product embodied on a computer-readable storage medium; and/or a processor, such as a processor suitable for executing instructions stored on and/or provided by a memory coupled to the processor”).
Claims 13-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by US PGPUB 2013/0132652 by Wood et al. (“Wood”).
As to Claim 13, Wood teaches in a data storage device comprising a memory, a method comprising:
analyzing operating conditions and/or available resources of the data storage device (Wood: at least ¶0169; “… adjust the read voltage threshold as described above until the ECC module 412 determines that the ECC decoder 322 can correct the error”);
based on the analyzing, determining a number of single read threshold operations (Wood: at least ¶0099; “three example read voltage thresholds may be x volts, y volts, and z volts, described in greater detail below with regard to the read voltage thresholds 662 of FIG. 6C. If the voltage read from a storage cell falls between Vmin and x volts, a binary 11 state is indicated. In certain embodiments, Vmin may be a negative voltage. If the voltage read from a storage cell falls between x volts and y volts, a binary 01 state is indicated. If the voltage read from a storage cell falls between y volts and z volts, a binary 00 state is indicated”);
operating a read threshold calibration unit the determined number of times to
provide a read threshold calibration result (Wood: at least ¶0169; “adjustment module 408 may repeatedly read the data set, determine that the read bias for the data set deviates from the known bias, determine the direction of deviation for the data set, and adjust the read voltage threshold as described above until the ECC module 412 determines that the ECC decoder 322 can correct the error”; ¶0159 also discloses iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias, or until each of the read voltage threshold level has been tested, and/or until the data set can be corrected using ECC checkbits, or the like; note: adjust as calibrate);
determining whether a magnitude of a correction between the read threshold calibration result and a previous read threshold calibration result is below a threshold (Wood: at least ¶¶0104, 0106; “a deviation from the known bias caused by errors in the data set is an error bias” and “determine whether the re-read data set has a read bias that deviates from the known bias, and may iteratively adjust the read voltage threshold to a new read voltage threshold until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”);
in response to determining that the magnitude of the correction is below the threshold (Wood: at least ¶0106; “… until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”; ¶0159 further discloses “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias”), reading the memory using the read threshold calibration result (Wood: at least ¶0099; “three example read voltage thresholds may be x volts, y volts, and z volts, described in greater detail below with regard to the read voltage thresholds 662 of FIG. 6C. If the voltage read from a storage cell falls between Vmin and x volts, a binary 11 state is indicated. In certain embodiments, Vmin may be a negative voltage. If the voltage read from a storage cell falls between x volts and y volts, a binary 01 state is indicated. If the voltage read from a storage cell falls between y volts and z volts, a binary 00 state is indicated”); and
in response to determining that the magnitude of the correction is not below the threshold, operating the read threshold calibration unit at least one additional time until the magnitude of the correction is below the threshold (Wood: at least ¶0106; “… may iteratively adjust the read voltage threshold to a new read voltage threshold until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”; ¶0159 further discloses “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias”).
As to Claim 14, Wood teaches the method of Claim 13, wherein the determining of the number of single read threshold operations is performed before performing a calibration operation (Wood: at least ¶0159; “the adjustment module 408 may iteratively readjust the read voltage threshold”; note: adjustment comes after determining).
As to Claim 15, Wood teaches the method of Claim 13, wherein the determining of the number of single read threshold operations is performed on-the-fly based on feedback from the read threshold calibration unit (Wood: at least ¶0159; “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias, or until each of the read voltage threshold level has been tested, and/or until the data set can be corrected using ECC checkbits, or the like”).
As to Claim 16, Wood teaches the method of Claim 13, further comprising:
after each iteration of the read threshold calibration unit, determining whether to
terminate operations or perform another calibration unit iteration (Wood: at least ¶0159; “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias, or until each of the read voltage threshold level has been tested, and/or until the data set can be corrected using ECC checkbits, or the like”; ¶0169 further discloses “adjustment module 408 may repeatedly read the data set, determine that the read bias for the data set deviates from the known bias, determine the direction of deviation for the data set, and adjust the read voltage threshold as described above until the ECC module 412 determines that the ECC decoder 322 can correct the error”), wherein the read threshold calibration result is provided in response to determining to terminate operations (Wood: at least ¶¶0159, 0169; “… until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias, or until each of the read voltage threshold level has been tested, and/or until the data set can be corrected using ECC checkbits, or the like” and “until the ECC module 412 determines that the ECC decoder 322 can correct the error”).
As to Claim 17, Wood teaches the method of Claim 16, wherein determining whether to terminate operations or perform another calibration unit iteration is based on an operating condition (Wood: at least ¶0159; “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias, or until each of the read voltage threshold level has been tested, and/or until the data set can be corrected using ECC checkbits, or the like”; ¶0169 further discloses “adjustment module 408 may repeatedly read the data set, determine that the read bias for the data set deviates from the known bias, determine the direction of deviation for the data set, and adjust the read voltage threshold as described above until the ECC module 412 determines that the ECC decoder 322 can correct the error”), an available resource, and/or feedback from the read threshold calibration unit.
As to Claim 18, Wood teaches the method of Claim 13, wherein the read threshold calibration unit comprises a single read threshold calibration unit (Wood: at least ¶0130; “configuration module 352 includes a data set read module 402, a deviation module 404, a direction module 406, an adjustment module 408, a persistence module 410, an ECC module 412, a distribution module 414, a data set source module 422”).
As to Claim 19, Wood teaches the method of Claim 13, wherein the method is performed in a dedicated hardware module in the data storage device (Wood: at least ¶0095; “the non-volatile memory controller 104 includes a configuration module 352 that sets and adjusts configuration parameters for the non-volatile”; ¶0130 further discloses “the configuration module 352 includes a data set read module 402, a deviation module 404, a direction module 406, an adjustment module 408, a persistence module 410, an ECC module 412, a distribution module 414, a data set source module 422, a proactive configuration module 424, and a write voltage module 416”).
As to Claim 20, Wood teaches a data storage device comprising:
a memory (Wood: at least ¶0038; “a non-volatile memory device 102, a non-volatile memory controller 104, a write data pipeline 106, a read data pipeline 108, non-volatile memory media 110”); and means for: operating a read threshold calibration unit of the data storage device a
plurality of times to provide a read threshold calibration result (Wood: at least ¶0169; “adjustment module 408 may repeatedly read the data set, determine that the read bias for the data set deviates from the known bias, determine the direction of deviation for the data set, and adjust the read voltage threshold as described above until the ECC module 412 determines that the ECC decoder 322 can correct the error”), wherein the plurality of times is based on operating conditions (Wood: at least ¶0169; “… adjust the read voltage threshold as described above until the ECC module 412 determines that the ECC decoder 322 can correct the error”) and/or available resources of the data storage device; and
determining whether a magnitude of a correction between the read threshold calibration result and a previous read threshold calibration result is below a threshold (Wood: at least ¶¶0104, 0106; “a deviation from the known bias caused by errors in the data set is an error bias” and “determine whether the re-read data set has a read bias that deviates from the known bias, and may iteratively adjust the read voltage threshold to a new read voltage threshold until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”);
in response to determining that the magnitude of the correction is below the threshold (Wood: at least ¶0106; “… until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”; ¶0159 further discloses “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias”), reading the memory using the read threshold calibration result (Wood: at least ¶0099; “three example read voltage thresholds may be x volts, y volts, and z volts, described in greater detail below with regard to the read voltage thresholds 662 of FIG. 6C. If the voltage read from a storage cell falls between Vmin and x volts, a binary 11 state is indicated. In certain embodiments, Vmin may be a negative voltage. If the voltage read from a storage cell falls between x volts and y volts, a binary 01 state is indicated. If the voltage read from a storage cell falls between y volts and z volts, a binary 00 state is indicated”); and
in response to determining that the magnitude of the correction is not below the threshold, operating the read threshold calibration unit at least one additional time until the magnitude of the correction is below the threshold (Wood: at least ¶0106; “… may iteratively adjust the read voltage threshold to a new read voltage threshold until the read bias of the data set no longer deviates from the known bias more than a threshold amount (which may be zero)”; ¶0159 further discloses “iteratively readjust the read voltage threshold based on the re-determined direction of deviation until the deviation module 404 determines that the read bias of a re-read data set does not deviate from the known bias”).
Claim Rejections - 35 USC § 103
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.
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over USPGPUB 2023/0035983 by Wang et al. (“Wang”) in view of US PGPUB 2025/0165148 Shukla et al. (“Shukla”).
As to Claim 3, Wang teaches the data storage device of Claim 2.
Wang does not explicitly disclose, but Shukla discloses wherein a number of trees in the additional trees is based on a program-erase count (Shukla: at least ¶0082; “selecting a neural network model, of the plurality of neural network models, based on a number of program/erase cycles of the one or more blocks”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Shukla before him/her at a time before the effective filing date of the claimed invention to incorporate Shukla’s feature of wherein a number of trees in the additional trees is based on a program-erase count (Shukla: at least ¶0082) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to perform “read operations using pre-determined threshold voltages associated with two overlapped charge states” (Shukla: at least ¶0081).
Claims 4-6 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over USPGPUB 2023/0035983 by Wang et al. (“Wang”) in view of US PGPUB 2022/0351019 Chen et al. (“Chen”).
As to Claim 4, Wang teaches the data storage device of Claim 2.
Wang does not explicitly disclose, but Chen discloses wherein a number of trees in the additional trees is based on an environment condition (Chen: at least ¶¶0155-0156; “one or more evaluation metrics include any one or more of a precision evaluation metric of the neural network model, a time overhead evaluation metric of the neural network model, a storage space evaluation metric of the neural network model, a power consumption evaluation metric of the neural network model, a utilization evaluation metric of a tensor calculation unit of the neural network model, and a memory read/write speed evaluation metric of the neural network model” and “the power consumption evaluation metric of the neural network model includes that power consumption of the target neural network is not greater than a power consumption threshold” and “the memory read/write speed evaluation metric of the neural network model includes that a memory read/write speed of the target neural network is not less than a memory read/write speed threshold”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Chen before him/her at a time before the effective filing date of the claimed invention to incorporate Chen’s feature of wherein a number of trees in the additional trees is based on an environment condition (Chen: at least ¶¶0155-0156) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to implement an “adaptive search method for a neural network” (Chen: at least ¶0009).
As to Claim 5, Wang teaches the data storage device of Claim 2.
Wang does not explicitly disclose, but Chen discloses wherein a number of trees in the additional trees is based on a performance requirement (Chen: at least ¶¶0155-0156; “one or more evaluation metrics include any one or more of a precision evaluation metric of the neural network model, a time overhead evaluation metric of the neural network model, a storage space evaluation metric of the neural network model, a power consumption evaluation metric of the neural network model, a utilization evaluation metric of a tensor calculation unit of the neural network model, and a memory read/write speed evaluation metric of the neural network model” and “the power consumption evaluation metric of the neural network model includes that power consumption of the target neural network is not greater than a power consumption threshold” and “the memory read/write speed evaluation metric of the neural network model includes that a memory read/write speed of the target neural network is not less than a memory read/write speed threshold”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Chen before him/her at a time before the effective filing date of the claimed invention to incorporate Chen’s feature of wherein a number of trees in the additional trees is based on a performance requirement (Chen: at least ¶¶0155-0156) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to implement an “adaptive search method for a neural network” (Chen: at least ¶0009).
As to Claim 6, Wang teaches the data storage device of Claim 2.
Wang does not explicitly disclose, but Chen discloses wherein a number of trees in the additional trees is based on available resources (Chen: at least ¶¶0155-0156; “one or more evaluation metrics include any one or more of a precision evaluation metric of the neural network model, a time overhead evaluation metric of the neural network model, a storage space evaluation metric of the neural network model, a power consumption evaluation metric of the neural network model, a utilization evaluation metric of a tensor calculation unit of the neural network model, and a memory read/write speed evaluation metric of the neural network model” and “the power consumption evaluation metric of the neural network model includes that power consumption of the target neural network is not greater than a power consumption threshold” and “the memory read/write speed evaluation metric of the neural network model includes that a memory read/write speed of the target neural network is not less than a memory read/write speed threshold”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Chen before him/her at a time before the effective filing date of the claimed invention to incorporate Chen’s feature of wherein a number of trees in the additional trees is based on available resources (Chen: at least ¶¶0155-0156) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to implement an “adaptive search method for a neural network” (Chen: at least ¶0009).
As to Claim 8, Wang teaches the data storage device of Claim 2.
Wang does not explicitly disclose, but Chen discloses wherein a number of trees in the additional trees is based on a power consumption (Chen: at least ¶¶0155-0156; “one or more evaluation metrics include any one or more of a precision evaluation metric of the neural network model, a time overhead evaluation metric of the neural network model, a storage space evaluation metric of the neural network model, a power consumption evaluation metric of the neural network model” and “the power consumption evaluation metric of the neural network model includes that power consumption of the target neural network is not greater than a power consumption threshold”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Chen before him/her at a time before the effective filing date of the claimed invention to incorporate Chen’s feature of wherein a number of trees in the additional trees is based on a power consumption (Chen: at least ¶¶0155-0156) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to implement an “adaptive search method for a neural network” (Chen: at least ¶0009).
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over USPGPUB 2023/0035983 by Wang et al. (“Wang”) in view of US PGPUB 2018/0260007 by Ping.
As to Claim 7, Wang teaches the data storage device of Claim 2.
Wang does not explicitly disclose, but Ping discloses wherein a number of trees in the additional trees is based on a temperature (Ping: at least ¶0049; “during training of the neural network, various combinations of flash operations (or “nonvolatile memory operations”) may be performed, and the temperature of the SSD may be monitored, to allow the neural network to form a model of the relationship between different kinds of flash operations and the temperature”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Pasha before him/her at a time before the effective filing date of the claimed invention to incorporate Ping’s feature of wherein a number of trees in the additional trees is based on a temperature (Ping: at least ¶0049) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to implement “temperature control” for “storage devices such as solid state drives” (Ping: at least ¶0005).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over USPGPUB 2023/0035983 by Wang et al. (“Wang”) in view of US PGPUB 2024/0265327 by Pasha et al. (“Pasha”).
As to Claim 10, Wang teaches the data storage device of Claim 1.
Wang does not explicitly disclose, but Pasha discloses wherein the tree-based inferencing model comprises a random-forest gradient boosting prediction model (Pasha: at least ¶0053; “ensemble Methods: Techniques like Random Forests and Gradient Boosting can be utilized to improve prediction accuracy. These methods combine multiple learning algorithms to obtain better predictive performance than could be obtained from any of the constituent learning algorithms alone”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Pasha before him/her at a time before the effective filing date of the claimed invention to incorporate Pasha’s feature of wherein the tree-based inferencing model comprises a random-forest gradient boosting prediction model (Pasha: at least ¶0053) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to “obtain better predictive performance” using artificial intelligence and machine learning (Pasha: at least ¶¶0049, 0053).
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over USPGPUB 2023/0035983 by Wang et al. (“Wang”) in view of US PGPUB 2013/0015519 Fujii et al. (“Fujii”).
As to Claim 12, Wang teaches the data storage device of Claim 1.
Wang does not explicitly disclose, but Fujii discloses wherein the memory comprises a three-dimensional memory (Fujii: at least ¶¶0060, 0087; “the reliability of the three-dimensional stacked layer type semiconductor memory can be improved” and “highly reliable three-dimensional NAND flash memory”).
It would have been obvious to one having ordinary skill in the art and the teachings of Wang and Fujii before him/her at a time before the effective filing date of the claimed invention to incorporate Fujii’s feature of wherein the memory comprises a three-dimensional memory (Fujii: at least ¶¶0060, 0087) with the data storage device disclosed by Wang.
The suggestion/motivation of doing so would have been to provide a memory structure that has high capacity and high reliability (Fujii: at least ¶¶0004, 0209; “the advantage of this three-dimensional stacked layer type semiconductor memory is that memory cells can be formed into a three-dimensional configuration without a substantial increase of processes and that a high memory capacity can be obtained at low cost” and “high reliability of the three-dimensional stacked layer type semiconductor memory”).
Relevant Prior Art
US PGPUB 2012/0083241 by Annamalai et al. teaches [0067] “another type of mathematical combination is selecting a point at the intersection of two or more ranges or areas that represent two or more location estimates”.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the Examiner should be directed to Huen Wong whose telephone number is (571) 270-3426. The examiner can normally be reached on Monday - Friday (10:30AM EST - 6:30PM EST).
If attempts to reach the examiner by telephone are unsuccessful, the Examiner's supervisor, Charles Rones can be reached on (571) 272-4086. The fax phone number for the organization where this application or proceeding is assigned is (571) 273-8300 for regular communications and after final communications.
Information regarding the status of an application may be obtained from thePatent Application Information Retrieval (PAIR) system. Status information forpublished applications may be obtained from either Private PAIR or Public PAIR.Status information for unpublished applications is available through Private PAIR only.For more information about the PAIR system, see http://pair-direct.uspto.gov. Shouldyou have questions on access to the Private PAIR system, contact the ElectronicBusiness Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from aUSPTO Customer Service Representative or access to the automated informationsystem, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/H .W./
Examiner, AU 2168
03 May 2026
/CHARLES RONES/Supervisory Patent Examiner, Art Unit 2168
Prosecution Insights