DETAILED ACTION
The action is in response to the original filing on April 21, 2022 and the Remarks and Amendments filed on 12/8/2025. Claims 1-20 are pending and have been considered below. Claims 1, 8 and 15 are independent claims.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Objections
The objections to claims 1, 8 and 15 have been withdrawn as necessitated by the amendment.
Claim Rejections - 35 USC § 101
The rejection of claims 1-20 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more, been withdrawn as necessitated by the amendment and Applicant’s supportive statements.
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, 2, and 5-7 are rejected under 35 U.S.C. 103 as being unpatentable over Jain et al. (“A Novel Data Prediction Technique Based on Correlation for Data Reduction in Sensor Networks,” hereinafter Jain) in view of Cui et al. (“Classification of data aggregation functions in wireless sensor networks,” hereinafter Cui) in further view of Yilmaz et al. (“Sequential Decentralized Parameter Estimation Under Randomly Observed Fisher Information,” hereinafter Yilmaz) further in view of Zhang (US-20220108150-A1).
Regarding claim 1, Jain teaches a method for managing data collection process between a data collector and a data aggregator, comprising: (Jain, Fig. 1; Here, “Data Collection” denotes obtaining, by the data aggregator and, from a data collector. Jain, Abstract; “We have used real sensor dataset of 54 SN [sensor nodes] that was deployed in the Intel Berkeley Research laboratory,” wherein the “[sensor nodes]” are the data collector[s].).
Jain does not explicitly teach a data statistic, the data statistic being based on a series of measurements performed by the data collector. However, Cui, in the area of comparing aggregation functions for use in optimizing data transmission within wireless sensor networks,
teaches this limitation (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Typical example is the monitoring of the average temperature or the average pressure over a given period. The report of the whole raw data is thus not required by the application. The average reported by the sink to the application is set to the most recent mean sent by the sensor. The update occurs whenever the sensor collects a raw data whose value deviates above the threshold from the current average value,” wherein “the most recent mean sent by the sensor” encompasses a data statistic being based on a series of measurements performed by the data collector, in this case “the sensor,” in accordance with the example embodiment provided at paragraph [0050] of the specification of the claimed invention, “For example, the data statistic obtained from a distributed system may be an average temperature over a period of time based on temperature measurements collected by one or more temperature sensors.”).
Cui is analogous to the claimed invention as both are from the same field of endeavor, that is, methods of optimizing data transmission in wireless sensor networks. Jain teaches a group of sensors that collect data but does not explicitly reduce said data into a representative statistic as is required by the definition of the term data statistic provided at paragraph [0016] of the specification of the claimed invention, “the data aggregator may obtain a data statistic from a data collector, a data statistic being any reduced-size representation of a series of measurements.” Cui teaches this step. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the data collection step of Jain to replace the raw data values with a reduced set of average or mean values, as taught by Cui. The motivation to do so is to save memory and reduce data transmission costs by compressing large quantities of sensor data into meaningful representations (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Typical example is the
monitoring of the average temperature or the average pressure over a given period. The report of the whole raw data is thus not required by the application.”).
Jain does not explicitly teach method for managing a data collection process between a data collector and a data aggregator; bits, wherein quantizing the data statistic comprises rounding the data statistic to a predetermined decimal point; the data statistic also being quantized by the data collector prior to the data collector transmitting the data statistic to a data aggregator such that the data statistic is reduced from a first number of bits to a second number of bits smaller than the first number of bits. However, Yilmaz, in the area of decentralized parameter estimation in sensor networks, teaches this limitation (Yilmaz, Abstract; “In the proposed scheme, each sensor computes its local random processes, and sends a single bit to the fusion center (FC) whenever the local random processes passes certain predefined levels. The FC, upon receiving a bit from a sensor, updates its approximation to the corresponding global random process and, accordingly, its estimate,” wherein “each sensor” corresponds to the data collector, and “the fusion center (FC)” corresponds to a data aggregator. Yilmaz, IV. Decentralized Estimators, A. AWGN Channels, pp. 1285, col. 1, paragraph 4; “From Corollary 1, and (7), we see that 𝑉𝑡𝐼 is a sufficient statistic for optimally estimating x, hence sensors should report {𝑉𝑡𝐼}𝑘 to the FC,” wherein “𝑉𝑡𝐼” encompasses the data statistic. Yilmaz, IV. Decentralized Estimators, A. AWGN Channels, 1) DMLE, pp. 1285, col. 1, paragraph 7; “Each sensor k following the fixed-time approach at time 𝑡𝐼, quantizes 𝑉𝑡𝐼 into 𝑉̃𝑡𝐼 using a traditional mid-riser uniform quantizer with the step size 𝑡𝐼𝜙𝑘2𝑅𝑘−1, and transmits 𝑅𝑘 bits to the FC,” thereby transmitting the data statistic to a data aggregator such that the data statistic is reduced from a first number of bits to a second number of bits smaller than the first number of bits.). Zhang teaches determining the decimal point in quantized data (76--wherein quantizing the data statistic comprises rounding the data statistic to a predetermined decimal point.
Yilmaz, and Zhang are analogous to the claimed invention as both are from the same field of endeavor, that is, reducing the bandwidth of data transmission in distributed sensor networks. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to add the quantization steps of Yilmaz, and Zhang to the combined data collection and reduction method of Jain and Cui. The motivation to do so is evident in that quantization provides an efficient solution to bandwidth constraints that is easier to implement than its alternatives (Yilmaz, I. Introduction, pp. 1281-82; “In order to conform to the low bandwidth requirement sensors either quantize their observations with a small number of bits, such as 1 bit, (e.g., [1], [5], [23]) or appropriately pulse-shape their analog transmissions (e.g., [10], [11], [17]). Quantization with a small number of bits causes the observations to be recovered in a coarse resolution at the FC, although it is much easier to implement than analog transmission.”). Zhang teaches reducing the complexity from high to low complexity(76).
Jain further teaches generating, by the data aggregator and while the data aggregator does not have and does not know the series of measurements performed by the data collector, a complementary data statistic, the complementary data statistic being generated using an inference model hosted by the data aggregator that also does not have and does not know the series of measurements performed by the data collector; determining, by the data aggregator, whether the complementary data statistic matches the data statistic obtained from the data collector, (Jain, 1 Introduction, pp. 596, paragraph 4; “Data prediction technique focuses on reducing the number of transmission from SNs [sensor nodes] to the BS [basestation] during the continuous monitoring of an application,”, wherein “the BS [basestation]” is the data aggregator does not have and does not know the series of measurements performed by the data collector, a complementary data statistic, the complementary data statistic being generated using an inference model hosted by the data aggregator that also does not have and does not know the series of measurements performed by the data collector that the “SNs [sensor nodes],” or data collectors, collected when the complementary statistic is obtained. In other words, the “transmission” of the series of measurements has yet to happen and is initiated by the match[ing] process that follows. Jain, 3 Proposed Approach, pp. 598, paragraph 5; “The SN will only transmit the data reading to the BS when the prediction fails that means the error threshold is higher than required. Both the prediction models will update actual data reading for synchronization in case of a failed prediction. However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value, which is also equal to the SN’s predicted value,” wherein “when the prediction fails [because]…the error threshold is higher than required” is equivalent to when determining, by the data aggregator, whether the complementary data statistic matches the data statistic obtained from the data collector in accordance with the explanation provided at paragraph [0016] of the specification of the claimed invention, “If the complementary data statistic does not match the data statistic within some threshold, the inferences may be determined inaccurate.” Jain, 3 Proposed Approach, p-, 600, paragraph 3; “The data values collected from N SNs will be aggregated and three quantitative attributes have been determined, i.e. minimum value, maximum value and average value. The data aggregation will be accomplished by these simple aggregations functions. These three values will be represented as 𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔, respectively,” wherein each of these “aggregation functions” produces a complementary data statistic, “𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔.”) the complementary statistic being based on a series of inferences generated by the data aggregator; (Jain, Abstract; “The purpose of the proposed model is to exempt the sensor nodes (SN) from sending huge volumes of data for a specific duration during which the BS will predict the future data values and thus minimize the energy utilization of WSN,” wherein the “BS [basestation] is the data aggregator to which the “sensor nodes (SN) [are] sending huge volumes of data” and to “predict the future data values” is to produce the complementary statistic being based on a series of inferences. Jain, 3 Proposed Approach, pp. 600, paragraph 3; The data aggregation will be accomplished by these simple aggregations functions. These three values will be represented as 𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔, respectively,” wherein the complementary data statistic[s] “𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔” are necessarily based on a series of inferences for the values used to calculate a minimum, maximum, or average.).
Jain teaches “The SN will only transmit the data reading to the BS when the prediction fails that means the error threshold is higher than required. Both the prediction models will update actual data reading for synchronization in case of a failed prediction. However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value, which is also equal to the SN’s predicted value,” (3 Proposed Approach, pp. 598, paragraph 5), wherein “only transmit[ting] the data reading to the BS when the prediction fails” is equivalent to the determination being made to determine whether the data aggregator can accurately infer the series of measurements without actually accessing an actual copy of the series of measurements such that the series of measurements performed by the data collector need not be transmitted to the data aggregator in order to reduce an amount of data transmission between the data collector and the data aggregator during the data collection process. and wherein a size of the data statistic is smaller than a size of the series of measurements; (Jain, 3 Proposed Approach, pp. 600, paragraph 3; The data aggregation will be accomplished by these simple aggregations functions. These three values will be represented as 𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔, respectively,” wherein each of these data statistic[s] is smaller than a size of the series of measurements they correspond to. In other words, a minimum, maximum and average all serve to summarize a series of measurements in a singular value.).
in a first instance of the determination where the complementary data statistic is different from the data statistic, and by the data aggregator: treating the inference model as being inaccurate; and obtaining at least a portion of the actual copy of the series of measurements from the data collector to use by the data aggregator to provide computer- implemented services to a user for which the series of measurements was performed. Jain teaches , Fig. 3; Here, the evaluation of “Is threshold error accepted” as “No” is equivalent to treating the inference model as being inaccurate. Upon this evaluation, there is a “data transmission step,” which is equivalent to obtaining at least a portion of the actual copy of the series of measurements from the data collector to use by the data aggregator to provide computer- implemented services to a user for which the series of measurements was performed. This is done to provide environmental monitoring-Jain, 3 Proposed Approach, pp. 598, paragraph 5, abstract; “The SN will only transmit the data reading to the BS when the prediction fails that means the error threshold is higher than required,” wherein to evaluate “the error threshold” is to the determination and to “transmit the data reading” is equivalent to obtaining [at “the BS”] at least a portion of the actual copy [the data reading] from the data collector [“the SN”] according to the explanation provided at paragraph [0017] of the specification of the claimed invention, “If the inferences are determined inaccurate, the data aggregator may take corrective action to increase the accuracy of future inferences and collect the actual data from the corresponding data collectors.”).
Furthermore, Jain teaches “The SN will only transmit the data reading to the BS when the prediction fails that means the error threshold is higher than required. Both the prediction models will update actual data reading for synchronization in case of a failed prediction. However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value, which is also equal to the SN’s predicted value,” (3 Proposed Approach, pp. 598, paragraph 5, abstract). This is done to provide environmental monitoring---and in a second instance of the determination where the complementary data statistic matches the data statistic, and by the data aggregator: treating the inference model as being accurate; and using the complementary data statistic to provide the computer-implemented services to the user without the data aggregator ever accessing the actual copy of the series of measurements from the data collector.
Regarding claim 2, the combination of Jain, Cui and Yilmaz teaches the method of claim 1, further comprising (and thus the rejection of claim 1 is incorporated).
Jain further teaches obtaining, from a data collector (Jain, Fig. 1; Here, “Data Collection” denotes obtaining, from a data collector. Jain, Abstract; “We have used real sensor dataset of 54 SN [sensor nodes] that was deployed in the Intel Berkeley Research laboratory,” wherein the “[sensor nodes]” are the data collector[s].).
Jain does not explicitly teach a second complementary data statistic, the second data statistic being based on a second series of measurements performed by the data collector. However, Cui, in the area of comparing aggregation functions for use in optimizing data transmission within wireless sensor networks, teaches this limitation (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Typical example is the monitoring of the average temperature or the average pressure over a given period. The report of the whole raw data is thus not required by the application. The average reported by the sink to the application is set to the most recent mean sent by the sensor. The update occurs whenever the sensor collects a raw data whose value deviates above the threshold from the current average value,” wherein “the most recent mean sent by the sensor” encompasses a second complementary data statistic being based on a second series of measurements performed by the data collector, in this case “the sensor,” in accordance with the example embodiment provided at paragraph [0050] of the specification of the claimed invention, “For example, the data statistic obtained from a distributed system may be an average temperature over a period of time based on temperature measurements collected by one or more temperature sensors.” Note that “the most recent” indicates that “the sensor” is recording the data and calculating the averages perpetually. Thus, there must necessarily be a second complementary data statistic. Additionally, the term second could be interpreted as alternative rather than temporal, in which case there could be a first “current average value” that “deviates above the threshold” and a second that does not. This interpretation is reinforced by the language at paragraph [0079] of the specification of the claimed invention, “In a second scenario, a second 𝑇𝑎𝑣𝑔 (𝑖𝑛𝑓𝑒𝑟𝑒𝑛𝑐𝑒) may be based on the following second set of inferences…,” wherein “second scenario” implies an alternative to a first scenario rather than a subsequent event. This interpretation is applied to all instances of the term second in the present claim.)
Cui is analogous to the claimed invention as both are from the same field of endeavor, that is, methods of optimizing data transmission in wireless sensor networks. Jain teaches a group of sensors that collect data but does not explicitly reduce said data into a representative statistic as is required by the definition of the term data statistic provided at paragraph [0016] of the specification of the claimed invention, “the data aggregator may obtain a data statistic from a data collector, a data statistic being any reduced-size representation of a series of measurements.” Cui teaches this step. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the data collection step of Jain to replace the raw data values with a reduced set of average or mean
values, as taught by Cui. The motivation to do so is to save memory and reduce data transmission costs by compressing large quantities of sensor data into meaningful representation (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Typical example is the monitoring of the average temperature or the average pressure over a given period. The report of the whole raw data is thus not required by the application.”).
Jain further teaches making a second determination that the second data statistic matches a second complementary data statistic obtained by a data aggregator that does not have access to the second series of measurements when the second complementary data statistic is obtained, (Jain, 1 Introduction, pp. 596, paragraph 4; “Data prediction technique focuses on reducing the number of transmission from SNs [sensor nodes] to the BS [basestation] during the continuous monitoring of an application,” wherein “the BS [basestation]” is a data aggregator that does not have access to the second series of measurements that the “SNs [sensor nodes],” or data collectors, collected when the second complementary statistic is obtained. In other words, the “transmission” of the series of measurements has yet to happen and is initiated by the match[ing] process that follows. Jain, 3 Proposed Approach, pp. 598, paragraph 5; “The SN will only transmit the data reading to the BS when the prediction fails that means the error threshold is higher than required. Both the prediction models will update actual data reading for synchronization in case of a failed prediction. However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value, which is also equal to the SN’s predicted value,” wherein “when the prediction is successful” is equivalent to when the second data statistic matches a second complementary data statistic in accordance with the explanation provided at paragraph [0016] of the specification of the claimed invention, “If the complementary data statistic matches the data statistic within some
threshold, the inferences may be determined accurate.” Jain, 3 Proposed Approach, p-, 600, paragraph 3; “The data values collected from N SNs will be aggregated and three quantitative attributes have been determined, i.e. minimum value, maximum value and average value. The data aggregation will be accomplished by these simple aggregations functions. These
three values will be represented as 𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔, respectively,” wherein each of these “aggregation functions” produces a complementary data statistic, “𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔.”) the second complementary statistic being based on a second series of inferences generated by the data aggregator; (Jain, Abstract; “The purpose of the proposed model is to exempt the sensor nodes (SN) from sending huge volumes of data for a specific duration during which the BS will predict the future data values and thus minimize the energy utilization of WSN,” wherein the “BS [basestation] is the data aggregator to which the “sensor nodes (SN) [are] sending huge volumes of data” and to “predict the future data values” is to produce the second complementary statistic being based on a series of inferences. Jain, 3 Proposed Approach, pp. 600, paragraph 3; The data aggregation will be accomplished by these simple aggregations functions. These three values will be represented as 𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔, respectively,” wherein the second complementary data statistic[s] “𝑋𝑚𝑖𝑛, 𝑋𝑚𝑎𝑥, and 𝑋𝑎𝑣𝑔” are necessarily based on a second series of inferences for the values used to calculate a minimum, maximum, or average.)
based on the second determination: treating the second series of inferences as being accurate; and allowing the data collector to discard the second series of measurements without providing the data aggregator with the second series of measurements. (Jain, Fig. 3; Here, the evaluation of “Is threshold error accepted” as “Yes” is equivalent to treating the second series of inferences as being accurate. Conversely, the “store past data value” step is only taken when the “threshold error” is not accepted. Therefore, when the “transmission cycle” does not take place and the data is not stored, the second series of measurements is necessarily discard[ed]. Jain, 3 Proposed Approach, pp. 598, paragraph 5; “However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value,” wherein to “not transmit data” is equivalent to not providing the data aggregator with the second series of measurements.).
Regarding claim 5, the combination of Jain, Cui and Yilmaz teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated).
Jain does not explicitly teach wherein the data statistic comprises one selected from a group consisting of an average of the series of measurements performed by the data collector, a mode of the series of measurements performed by the data collector, and a median of the series of measurements performed by the data collector. However, Cui, in the area of comparing aggregation functions for use in optimizing data transmission within wireless sensor networks, teaches this limitation (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Typical example is the monitoring of the average temperature or the average pressure over a given period. The report of the whole raw data is thus not required by the application. The average reported by the sink to the application is set to the most recent mean sent by the sensor. The update occurs whenever the sensor collects a raw data whose value deviates above the threshold from the current average value,” wherein “the most recent mean sent by the sensor” encompasses the data statistic and is equivalent to an average of the series of measurements performed by the data collector, in this case “the sensor.”).
Cui is analogous to the claimed invention as both are from the same field of endeavor, that is, methods of optimizing data transmission in wireless sensor networks. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the data collection step of Jain to replace the raw data values with a reduced set of average or mean values, as taught by Cui. The motivation to do so is to save memory and reduce data transmission costs by compressing large quantities of sensor data into meaningful representation (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Typical example is the monitoring of the average temperature or the average pressure over a given period. The report of the whole raw data is thus not required by the application.”). Furthermore, using an average specifically is a technique well-known in the art of data aggregation (Cui, 4.2. Forecasting aggregation functions, pp. 4, col. 1, paragraph 4; “Average is a simple aggregation function, which has been widely investigated in works on data aggregation for WSNs.”).
Regarding claim 6, the combination of Jain, Cui and Yilmaz teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated).
Jain further teaches wherein the series of measurements are obtained using a sensor that measures a characteristic of an ambient environment (Jain, 1 Introduction, pp. 596, paragraph 1; “WSNs can be periodically used in any environment to sense hydrological and meteorological parameters in the vicinity, like temperature, light intensity, humidity, traffic, wind speed and direction, voltage, air and water quality and many others.” Jain, 4 Simulation and Result Analysis, pp. 601, paragraph 1; “The data values are continuously sensed by 54 SNs which are randomly deployed in the sensing area as illustrated in Fig. 2. The data values
comprise of four different parameters namely temperature, humidity, light intensity and voltage,” wherein “data values” representing “temperature” and “humidity” constitute a series of measurements…characteristic of an ambient environment in accordance with the examples given at paragraph [00130] of the specification of the claimed invention, “The series of measurements may represent some characteristic of an ambient environment. For example, the series of measurements may include temperature data, pH data, humidity data, etc.”).
Regarding claim 7, the combination of Jain, Cui and Yilmaz teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated).
Jain further teaches wherein the inference model hosted by the data aggregator is trained to duplicate the series of measurements (Jain, 1 Introduction, pp. 596, paragraph 4; “Data prediction is a most effective and efficient way to reduce data in WSNs that makes use of the predicted data instead of the real sensor’s readings, hence restricting the data transmission. Data prediction technique focuses on reducing the number of transmission from SNs to the BS during the continuous monitoring of an application,” wherein to “make use of the predicted data instead of the real sensor’s readings” is to attempt to duplicate the series of measurements produced by the data collector[s] or “SNs.” Jain, 3 Proposed Approach, pp. 598, paragraph 5; “the same prediction model is also deployed in the BS…However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value, which is also equal to the SN’s predicted value,” wherein “the prediction model” is equivalent to the inference model and “deployed in the BS” indicates it is hosted by the data aggregator. Jain, 3 Proposed Approach, pp. 600, paragraph 1; “if the prediction error is within the user-defined threshold, there is no need to send data to BS, and just wait for the next data sampling period; otherwise, update the prediction model and send sampled data to BS,” wherein to “update the prediction” encompassing train[ing].)
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Jain in view of Cui, Yilmaz, Zhang, and Liazid et al. (“An improved adaptive dual prediction scheme for reducing data transmission in wireless sensor networks,” hereinafter Liazid).
Regarding claim 3, the combination of Jain, Cui and Yilmaz teaches the method of claim 1, (and thus the rejection of claim 1 is incorporated).
Jain further teaches wherein in the first instance of the determination where the complementary data statistic is different from the data statistic, the method further comprises, by the data aggregator: retraining, by the data aggregator and after treating the inference model as being inaccurate, the inference model that was used to obtain the complementary data statistic, the inference model being retrained using a training data set
comprising, at least in in part …the portion of the series of measurements from the data collector --“The SN will only transmit the data reading to the BS when the prediction fails that means the error threshold is higher than required. Both the prediction models will update actual data reading for synchronization in case of a failed prediction. However, when the prediction is successful, the SN will not transmit data reading to the BS and the BS will save its own predicted value, which is also equal to the SN’s predicted value,” (3 Proposed Approach, pp. 598, paragraph 5, abstract), “if the prediction error is within the user-defined threshold, there is no need to send data to BS, and just wait for the next data sampling period; otherwise, update the prediction model and send sampled data to BS,” (Jain, 3 Proposed Approach, pp. 600, paragraph 1), wherein “the prediction model” is equivalent to an inference model and the “BS” is equivalent to the data aggregator. Furthermore, to update both prediction models is to retrain, which itself is a form of training; therefore, to update both prediction models in the BS with “sampled data” encompasses retraining, by the data aggregator and after treating the inference model as being inaccurate, this “sampled data” encompassing a portion of a series of measurements from the data collector or “sensor nodes.”.
Jain does not explicitly teach a portion of a series of inferences generated by the inference model that make up the complementary data statistic . However, Liazid, in the area of optimizing data transmission within wireless sensor networks using prediction-based data aggregation, teaches this limitation (Liazid, 3 Contribution, pp. 3546, col. 2, paragraph 3; “The main contribution of this work tries to avoid the usage of the data history table when updating model. This further reduces the data transmission from sensor node to the sink. The idea is to exploit the collection of the previous models used during the past prediction sequences to predict the parameters of the new one instead of using the data history table,” wherein “updating [the] model” in the context of predictive models encompasses updating an inference model thereby making “past prediction sequences” equivalent to a portion of a series of inferences.
Liazid is analogous to the claimed invention as both are from the same field of endeavor, that is, methods of optimizing data transmission in wireless sensor networks. Jain teaches updating a prediction model with data from a sensor (Jain, 3 Proposed Approach, pp. 600, paragraph 1; “if the prediction error is within the user-defined threshold, there is no need to send data to BS, and just wait for the next data sampling period; otherwise, update the prediction model and send sampled data to BS.”) but does not explicitly include past inference data generated by an aggregator. Liazid teaches this limitation. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the sensor measurements of Jain with the sink prediction sequences of Liazid in creating a dataset to update an inference model. The motivation to do combine the two rather than to rely solely on past sensor measurements is to cut down on transmission costs between the collector and aggregator nodes in a wireless sensor network (Liazid, 3 Contribution, pp. 3546, col. 2, paragraph 3; “The main contribution of this work tries to avoid the usage of the data
history table when updating model. This further reduces the data transmission from sensor node to the sink.”).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Jain in view of Cui, Yilmaz, Zhang and Lewandowski et al. (“Classifier-Based Data Transmission Reduction in Wearable Sensor Network for Human Activity Monitoring,” hereinafter Lewandowski).
Regarding claim 4, the combination of Jan and Cui teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated).
Jain does not explicitly teach wherein the series of inferences is generated by the inference model after the inference model was trained using a training data set comprising a second series of measurements performed by the data collector, the second series of measurements being performed by the data collector before the data collector performed the series of measurements. However, Lewandowski, in the area of optimizing data transmission within wireless sensor networks using classification models, teaches these limitations.
wherein the series of inferences is generated by the inference model after the inference model was trained using a training data set, (Lewandowski, 3.1. Overview of the Method, pp. 7-8, paragraphs 4 and 1; “Application of this proposed method involves the following steps:…2. Divide the training data into two samples. 3. Train recognition model M using the first data sample…A training data set, collected at the first step, has to include preprocessed sensor readings from all sensor nodes (S) and information about activities of the monitored persons (A) for a representative period,” wherein the “recognition model M” constitutes an inference model trained using “the first data sample.” Lewandowski, Fig. 1; Here, the “[first] sample data set” is a training dataset used to train the “recognition model M.” Lewandowski, Fig. 2;
Here, the “cluster head,” or data aggregator “receives data from sensor nodes,” or data collectors and uses the “recognition model M,” or inference model, to generate a series of inferences regarding the type of “activity.”) a training data set comprising a second series of measurements performed by the data collector, the second series of measurements being performed by the data collector before the data collector performed the series of measurements (Lewandowski, 3.1. Overview of the Method, pp. 7-8, paragraphs 4 and 1; “Application of this proposed method involves the following steps:…2. Divide the training data into two samples. 3. Train recognition model M using the first data sample…A training data set, collected at the first step, has to include preprocessed sensor readings from all sensor nodes (S) and information about activities of the monitored persons (A) for a representative period,” wherein “the first data sample” used to train the “recognition model M” is necessarily a second series of measurements, collected by the data collector[s] or “sensor nodes (S),” at a “training stage,” represented in figure 1, that occurs prior to the series of measurements produced at the “network operation stage,” represented by the left half of figure 2.).
Lewandowski is analogous to the claimed invention as both are from the same field of endeavor, that is, methods of optimizing data transmission in wireless sensor networks. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to add the training stage of Lewandowski to the combined transmission reduction method of Jain, Cui and Yilmaz. The motivation to do so is to have a final inference model that is robust to missing data from the data collectors, which is a practical drawback of wireless sensible networks or IoT (Internet of Things) systems comprising a multitude of sensors that should be accounted for (Lewandowski, 3.1 Objective of the Method, pp. 7, paragraph 2; “The available data are then used to recognize current activity of the monitored person. Since the proposed method assumes that only selected data are sent to the cluster head node, an activity recognition algorithm R is needed, which can deal with incomplete data sets. In general, the activity recognition task, performed by cluster head, is expressed as follows: 𝑎̂𝑡=𝑅(𝑆𝑡,𝑀), where M is an activity recognition model trained with use of a machine
learning algorithm and 𝑆𝑡 denotes a set of data transmitted to the cluster head from sensor nodes a time step t.”).
Claims 8, 9, 12-16, 19 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Jain in view of Cui, Yilmaz, Zhang and Wouhaybi et al. (US 2019/0044786 A1, hereinafter Wouhaybi).
Regarding claim 8, Jain does not explicitly teach a non-transitory machine-readable medium having instructions stored therein, which when executed by a processor of a data aggregator, cause the processor to perform operations for managing a data collection process between a data collector and the data aggregator, the operations comprising. However, Wouhaybi, in the area of managing data distribution among control devices, teaches these limitations (Wouhaybi, Abstract; “Methods, apparatus, systems and articles of manufacture to dynamically control devices based on distributed data are disclosed,” wherein “control devices based on distributed data” necessarily perform operations for managing a data collection process. See Figs. 3-8 of Wouhaybi for examples steps regulating the reception, transmission, forwarding, etc. of data within a “device topology.” Wouhaybi, [0039]; “The machine readable instructions may be a program or portion of a program for execution by a processor such as the processor 1012 shown in the example processor platform 1000 discussed below in connection with FIG. 10. The program may be embodied in software stored on a non-transitory computer readable storage medium such as a CD-ROM, a floppy disk, a hard drive, a DVD, a Blu-ray disk, or a memory associated with the processor 1012”).
Wouhaybi is analogous to the claimed invention as both are from the same field of endeavor, that is, managing the communication of data between devices in a network. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to implement the combined transmission reduction method of Jain, Cui and Yilmaz on the hardware of Wouhaybi. The motivation to do so is inherent as any computer-implemented method necessarily comprises instructions that require a process for execution and memory for storage.
The following limitations correspond to the steps of claim 1 and are thus rejected for the same reasons as claim 1.
Claims 9 and 12-14 are non-transitory computer-readable medium claims corresponding to the steps of claims 2 and 5-7 and are thus rejected for the same reasons as claims 2 and 5-7.
Regarding claim 15, Jain does not explicitly teach a data aggregator, comprising: a processor; and a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing a data collection process between the data aggregator, and a data collector, the operations comprising. However, Wouhaybi, in the area of managing data distribution among control devices, teaches these limitations (Wouhaybi, Abstract; “Methods, apparatus, systems and articles of manufacture to dynamically control devices based on distributed data are disclosed,” wherein “control devices based on distributed data” necessarily perform operations for managing data. See Figs. 3-8 of Wouhaybi for examples steps regulating the reception, transmission, forwarding, etc. of data within a “device topology.” Wouhaybi, [0039]; “The machine readable instructions may be a program or portion of a program for execution by a processor such as the processor 1012 shown in the example processor platform 1000 discussed below in connection with FIG. 10. The program may be embodied in software stored on a non-transitory computer readable storage medium such as a CD-ROM, a floppy disk, a hard drive, a DVD, a Blu-ray disk, or a memory associated with the processor 1012”).
Wouhaybi is analogous to the claimed invention as both are from the same field of endeavor, that is, managing the communication of data between devices in a network. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to implement the data collection and comparison steps of Jain, modified by Cui on the hardware of Wouhaybi. The motivation to do so is inherent as any computer-implemented method necessarily comprises instructions that require a process for execution and memory for storage.
The following limitations correspond to the steps of claim 1 and are thus rejected for the same reasons as claim 1.
Claims 16, 19 and 20 are data aggregator or system claims corresponding to the steps of claims 2, 5 and 6 and are thus rejected for the same reasons as claims 2, 5 and 6.
Claims 10 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Jain in view of Cui, Yilmaz, Liazid and Wouhaybi.
Claim 10 is a non-transitory computer-readable medium claim corresponding to the steps of claim 3 implemented using the hardware elements of Wouhaybi inherited from claim 8. Claim 10 is thus rejected for the same reasons as claim 4.
Claim 17 is a data aggregator or system claim corresponding to the steps of claim 3 implemented using the hardware elements of Wouhaybi inherited from claim 15. Claim 17 is thus rejected for the same reasons as claim 3.
Claims 11 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Jain in view of Cui, Yilmaz, Zhang, Lewandowski and Wouhaybi.
Claim 11 is a non-transitory computer-readable medium claim corresponding to the steps of claim 4 implemented using the hardware elements of Wouhaybi inherited from claim 8. Claim 11 is thus rejected for the same reasons as claim 4.
Claim 18 is a data aggregator or system claim corresponding to the steps of claim 4 implemented using the hardware elements of Wouhaybi inherited from claim 15. Claim 18 is thus rejected for the same reasons as claim 4.
Response to Arguments
In the response filed 12/8/2025, the Applicant states that “Applicant submits that the amended independent claims now recite limitations (e.g., generating the complementary data statistic using a hosted inference model while not having and not knowing the series of measurements performed by the data collector, using the portions of the actual copy of the series of measurements to provide computer-implemented service, using the complementary data statistic to provide the computer-implemented services, and the like) that cannot be practically performed within the human mind using just the aid of pen and paper…. the amended claims now also recite all of the steps and processes associated the technological improvement described in the specification. In particular, paragraphs [0014]-[0018] and [0034]-[0036] of the Original Specification describe the technical improvements associated with configuring a data aggregator to predict the data that data collectors collect instead of having the data collectors send actual collected data to the data aggregator. As a result, the number of data transmissions between these devices can be reduced, which directly improves multiple aspects of the overall system as well as improves sensor data collection and management technology as a whole. One of ordinary skill in the art would also appreciate that such reduction in the number of data transmissions and the amount of data being transmitted also directly improves the functionality of both the data collectors and data aggregator (namely, where limited computing resources of the data collectors and data aggregators, both of which are implemented using computing devices, can be used for other, more critical, processes besides data transmission).”. The examiner agrees that the amendment overcomes the abstract idea rejection which have been withdrawn as indicated above.
Additionally, Applicant argues that “ Without acquiescing to the Office's position, Applicant has amended independent claim 1 to recite, in part, "wherein quantizing the data statistic comprises rounding the data statistic to a predetermined decimal point." (Emphasis Added).Amended independent claims 8 and 15 recite substantially similar limitations. Applicant respectfully submits that all of the cited prior art references (i.e., Jain, Cui, Yilmaz Liazid, Lewandowski, and Wouhaybi) fail to disclose or suggest at least these limitations of the amended independent claims.”. Applicant is directed towards the rejection of this limitation above in light of Zhang.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CESAR PAULA whose telephone number is (571)272-4128. The examiner can normally be reached Monday - Friday, 6.30am- 4:30 pm ET.
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, David Wiley can be reached at (571)272-3923. 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.
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights