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 Rejections - 35 USC § 112
The following is a quotation of the second paragraph of 35 U.S.C. 112:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 38 is rejected under 35 U.S.C. 112, second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which applicant regards as the invention.
In particular, claim 38 recites in the preamble, “A second node in a communications network for determining whether a channel between a first node and a target node is in use..” the body of the claim does not contain any limitations indicating the structure of the device. A system claim should always claim the structure or the hardware that performs the function. Applicant’s claimed limitations consist of nodes that do not describe the structure of the communication network. Appropriate correction is required.
Claim Rejections - 35 USC § 102
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 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)(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 35-36, and 38 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Novlan et al. (Patent No: US 2019/0261412 A1), hereinafter, Novlan.
Regarding Claim 35, Novlan teaches,
A method performed in a second node for determining whether a channel between a first node and a target node is in use, the method comprising: -Fig. 1; Paragraph [0029] ([0029] recites, “FIG. 1 illustrates an example wireless communication system 100 in accordance with various aspects and embodiments of the subject disclosure. In one or more embodiments, system 100 can comprise one or more user equipment UEs 104 and 102, which can have one or more antenna panels having vertical and horizontal elements…… a UE 102 can be communicatively coupled to the wireless communication network via a network node 106.” Node 1 here is Network node 106 in communication with Node 2 (target node) UE (102/104))
receiving a message from the first node comprising an indication of whether the second node should obtain channel information for the channel for use by the first node in determining whether the channel is in use. -Paragraph [0015] ([0015] recites, “The embodiments disclosed herein enable a base station to coordinate the LBT process at both the base station and a receiver in order to avoid hidden node interference where the interfering nodes are outside the sensing range of the transmitting node. The base station device can send a LBT trigger to the receiver to synchronize the clear channel assessments that are performed at each device to determine if there is any activity on the channel. The receiving device can then send back a report to the base station device, and if both devices detect no activity on the channel, the base station device can schedule a transmission on the channel.” As explained above, the second node UE receives LBT trigger from the first node (Network) and perform LBT operation at the second node (UE). [0050] recites, “In an embodiment, the base station device 304 can coordinate LBT between both devices 306 and 308 by sending LBT triggers 310 and 314 to the devices 306 and 308 respectively requesting CCAs to be performed at the same time. Devices 306 and 308 can then send back their reports 312 and 316 to the base station device 304 for the base station device 304 to determine whether to facilitate scheduling a transmission to either device 306 or 308.”)
Regarding Claim 36, Novlan teaches the limitations of Claim 35.
Novlan further teaches,
The method as in claim 35, wherein the message further indicates one or more of a type of channel information to obtain, a type of sensing to perform in order to obtain the channel information and a periodicity with which the channel information should be obtained. -Paragraph [0043] ([0043] recites, “In an embodiment, the LBT trigger can be sent on a downlink control channel (e.g., PDCCH or PDSCH), or on a new dedicated physical channel. The trigger can include a request to perform the clear channel assessment as well as providing parameters indicating how the mobile device 204 should perform the CCA. The indicated LBT parameters may include a starting time location/offset for carrier sensing as well as a duration in symbols/slots, energy detection threshold, LBT type, priority, etc.” duration dictates the periodicity. [0046] recites, “In one example the LBT feedback is provided in the uplink portion of a self-contained subframe/slot, whether the LBT trigger is carried in the downlink portion of the same subframe/slot. In another example the LBT trigger and/or feedback may be carried on mini-slots within the duration of a slot. This is beneficial to reduce the delay between LBT trigger and feedback in case the LBT duration is short. In case of longer LBT durations which are more than one slot in length, the LBT trigger DCI may indicate the timing of the LBT feedback message, for example the starting slot or symbol offset. In another example the LBT feedback timing is implicitly determined based on the duration of the LBT sensing period.”)
Regarding Claim 38, Novlan teaches,
A second node in a communications network for determining whether a channel between a first node and a target node is in use, wherein the first node is configured to: receive a message from the first node comprising an instruction to cause the second node to obtain channel information for the channel. -Paragraph [0015, 0050] ([0015] recites, “ The base station device can send a LBT trigger to the receiver to synchronize the clear channel assessments that are performed at each device to determine if there is any activity on the channel. The receiving device can then send back a report to the base station device, and if both devices detect no activity on the channel, the base station device can schedule a transmission on the channel.” As explained above, the second node UE receives LBT trigger from the first node (Network) and perform LBT operation at the second node (UE). [0050] recites, “In an embodiment, the base station device 304 can coordinate LBT between both devices 306 and 308 by sending LBT triggers 310 and 314 to the devices 306 and 308 respectively requesting CCAs to be performed at the same time. Devices 306 and 308 can then send back their reports 312 and 316 to the base station device 304 for the base station device 304 to determine whether to facilitate scheduling a transmission to either device 306 or 308.”)
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.
Claims 1-6, 12, 15, 20, 22-24, 26, 33, and 34 are rejected under 35 U.S.C. 103 as being unpatentable over Wenli Ning et al. (JOURNAL OF COMMUNICATIONS AND NETWORKS, VOL. 22, NO. 1, FEBRUARY 2020, Page 12-22, “Reinforcement Learning Enabled Cooperative Spectrum Sensing in Cognitive Radio Networks”), hereinafter, Ning, in view of Rahil Sarikhani et al. (IEEE COMMUNICATIONS LETTERS, VOL. 24, NO. 7, JULY 2020, page 1459-1462, “Cooperative Spectrum Sensing Meets Machine Learning: Deep Reinforcement Learning Approach”), hereinafter, Sarikhani.
Regarding Claim 1, Ning teaches,
the selection being performed using a first model trained using a first machine learning process to select the subset of other nodes based on accuracy of the resulting determination of whether the channel is in use; -Section I, IV-B; Page 12 (It recites,” We propose a channel selection algorithm based on Q-learning to determine the scanning order of the channels, so as to reduce the scanning overhead and access delay. Specifically, each SU learns the occupancy pattern of the primary channels, and updates a dynamic scanning preference list of the channels based on the predicted channel status. A novel reward function is devised to improve the accuracy of channel status prediction during the learning process. We propose a cooperation partner selection algorithm based on discounted upper confidence bound (D-UCB) algorithm to improve the detection efficiency.” Section II, Page 14 recites, “the authors compared the performance
of different machine learning approaches in terms of spectrum classification accuracy and computational time.”)
and sending a message to cause the subset of other nodes to obtain the channel information. -Page 17, Section IV, (recites, “ Moreover, by applying the selection
rule in (6), SUk chooses SUf as the cooperation partner, and sends a notification message to inform SUf to detect the selected primary channel. Once SUf finishes the detection, it reports the result to SUk. Then, SUk makes the final decision based on the received detection result. In case that multiple cooperation partners are selected, SUk will send a notification message to each partner. The partners detect the selected primary channel individually, and report the results to SUk. Then, SUk fuses the detection results of the partners based on a certain fusion rule, such as the majority rule or the weighted rule [25], to make the final decision.”)
Although implicit, Ning does not explicitly mention,
A computer implemented method performed by a first node in a communications network for use in determining whether a channel between the first node and a target node is in use, the method comprising: selecting, from a plurality of other nodes that are suitable for making measurements on the channel, a subset of the other nodes from which to obtain channel information in order to determine whether the channel is in use
However, in an analogous invention, Sarikhani teaches,
A computer implemented method performed by a first node in a communications network for use in determining whether a channel between the first node and a target node is in use, the method comprising: -Page 1460, Section II (Recites, “the SU broadcasts the request of cooperative sensing, and all the one-hop neighbors will respond to its appeal by their local sensing results, intermittently. The initiating SU, which requests for cooperative sensing, is called the agent in this phase. The agent will combine the local sensing results and take the role
of Fusion Center (FC) in the network.” Initiating SU is first node in here and the target node is one among the one-hop neighbor.)
selecting, from a plurality of other nodes that are suitable for making measurements on the channel, a subset of the other nodes from which to obtain channel information in order to determine whether the channel is in use – Page 1460, Section II (Recites, “The FC, which is called the agent from now on in this letter, begins the DRL-based cooperative spectrum sensing and detects the presence or absence of the PU based on the surrounding one-hop neighbors. The number of one-hop neighbors surrounding the agent is N. In each period of cooperative sensing, C < N nodes are selected based on the reliance of the cooperation, which is estimated by the location-based correlation metric _ji, and it would change based on the conditions of the channel and location of SUs. These C nodes are informed step-by-step to update their local sensing results.” As explained above a sub-set C<N is chosen to obtain channel information (sensing)”)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the “Reinforcement Learning Enabled Cooperative Spectrum Sensing in Cognitive Radio Networks “ proposed by Ning to include the concept of “A computer implemented method performed by a first node in a communications network for use in determining whether a channel between the first node and a target node is in use, the method comprising: selecting, from a plurality of other nodes that are suitable for making measurements on the channel, a subset of the other nodes from which to obtain channel information in order to determine whether the channel is in use ” of Sarikhani. One of ordinary skill in the art would have been motivated to make this modification in order to improve the sensing quality (page 1459, Section I)
Regarding Claim 2, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 1 wherein the first model is trained to select the subset of other nodes so as to optimize accuracy of the resulting determination of whether the channel is in use. -Page 18; Section V-B (it recites, “In the proposed channel selection algorithm, parameter α is the learning rate, which determines the accuracy of the prediction of channel status thus the effectiveness of the resulting scanning preference list. It should be noted that the optimal value of α may vary with the network settings.”)
Regarding Claim 3, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 1 wherein the first model is further trained to select the subset of other nodes based on values of one or more other parameters; -Page 15-16; Section IV-B ( Recites. “Therefore, we propose that each SU selects the neighbor with the strongest detection ability as the cooperation partner to detect the primary channel of interest, so as to strike a balance between efficiency and overhead….. To this end, we model the partner selection algorithm as a D-MAB problem, wherein each SU estimates the detection probability of its neighbors in a dynamic situation, and learns about the cooperation partner selection strategies to maximize the detection efficiency…. The resulting reward Rt relies on whether the selected cooperation partner can detect the channel correctly. …..To maximize the total reward, the partner selection strategy should take into account both the estimated detection probability and the exploration degree of the neighbors.” Equation 5-8 and Table 2 explains the recursive learning (RL) scheme and how the partner selection strategy is based on many parameters including estimated detection probability and the exploration degree of the neighbors.)
wherein the first model is trained to optimize the accuracy of the resulting determination of whether the channel is in use and the values of the one or more other parameters. -Page 18; Section V-B (it recites, “In the proposed channel selection algorithm, parameter α is the learning rate, which determines the accuracy of the prediction of channel status thus the effectiveness of the resulting scanning preference list. It should be noted that the optimal value of α may vary with the network settings….In the proposed cooperation partner selection algorithm, parameter γ controls the weight of the recent reward when learning the detection probability of the neighbors, thereby determining the validity of the selection of cooperation partner.”)
Regarding Claim 4, Ning and Sarikhani teach the limitations of Claim 3.
Ning further teaches,
The method as in claim 4, wherein the one or more other parameters comprises a measure of overhead associated with making the determination. -Page 15, Section IV-B (It recites, “In order to improve the detection efficiency, each SU should select the ones with stronger detection ability as partners. But the more partners participant in cooperation, the higher overhead and complexity. Therefore, we propose that each SU selects the neighbor with the strongest detection ability as the cooperation partner to detect the primary channel of interest, so as to strike a balance between efficiency and overhead.”)
Regarding Claim 5, Ning and Sarikhani teach the limitations of Claim 4.
Ning further teaches,
The method as in claim 4 wherein the measure of overhead is one or more of: signalling overhead associated with making the determination; volume of traffic flow through the communications network associated with making the determination; computational energy used by the subset of nodes associated with making the determination; and energy efficiency associated with making the determination. -Page 13, Section II (Recites, “The resulting user group which participates in cooperative sensing procedures is safe, less redundant, or the optimized user group, leading to better performance in terms of security, energy consumption,
and sensing efficiency.”)
Regarding Claim 6, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 1, wherein the first model is a reinforcement learning agent; wherein the step of selecting is performed by the reinforcement learning agent as an action, a; -Page 14; section IV-A ( It recites, “Q-learning is a widely used reinforcement learning algorithm. In Q-learning model, an agent in state s ϵ S interacts with the environment by taking an action a ϵ A”)
and wherein the reinforcement learning agent is rewarded for the action based on the accuracy of the resulting determination of whether the channel is in use. -Section I/II, Page 12-13 (Page 13 recites, “In RL framework, the action-taking agent interacts with the external environment through reward mechanisms, and adjusts its action according to the reward values obtained in the environment. The aim of the agent is to learn the optimal action to maximize the total reward.” Page 12, Section I recites, “A novel reward function is devised to improve the accuracy of channel status prediction during the learning process.”)
Regarding Claim 12, Ning and Sarikhani teach the limitations of Claim 6.
Ning further teaches,
The method as in claim 6, wherein the first model is further trained to select the subset of other nodes based on values of one or more other parameters; -Page 16; Section IV (Recites, “It’s worth mentioning that the proposed partner selection algorithm is not limited to the scenario of selecting one partner, but also can be extended to the scenario of selecting multiple
partners. Let Ncoop be the size of cooperation cluster for spectrum sensing, i.e., the number of SUs participating in channel detection. In case that multiple partners are considered, i.e., the
case of Ncoop > 1, SUk can apply the selection rule in (6) to select the first Ncoop neighbors (itself may be included) as the cooperation partners.” Equation 6 shows selection process is dependent on at least two parameters detection probability and exploration degree.)
and wherein the first model is trained to optimize the accuracy of the resulting determination of whether the channel is in use and the values of the one or more other parameters, -Page 18; Section V-B (it recites, “In the proposed channel selection algorithm, parameter α is the learning rate, which determines the accuracy of the prediction of channel status thus the effectiveness of the resulting scanning preference list. It should be noted that the optimal value of α may vary with the network settings.”)
wherein the reinforcement learning agent receives a reward based on a reward function that rewards the reinforcement learning agent based on relative priorities of the accuracy and the values of the one or more other parameters, so as to apply a trade-off between the accuracy and the one or more parameters according to the relative priority of each parameter. -Page 16; Section IV (Recites, “Since the detection probability of SUs varies with time, the recent rewards play a more important role than the previous rewards in the estimation of detection probability. We can use a discount factor to give different weight to the reward obtained at different time.” Equation 5-8 show the selection strategy. Priorities are controlled by assigning different weights in the second term of Equation 6. Section IV also recites, “The benefit of introducing the second term is that the cooperation partner selection strategy can fully explore all the possible options. c is a system parameter, which controls the degree of exploration versus exploitation. If c is set properly, a good balance between exploration and exploitation can be achieved”)
Regarding Claim 15, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 1,wherein the first model is a classification model, wherein the first model is further trained to select the subset of other nodes based on values of one or more other parameters; -Page 13; Section II/Section IV-B (Section II Recites, “the authors compared the performance of different machine learning approaches in terms of spectrum classification accuracy and computational time.” Sec IV-B describes the node (Cooperation partner selection algorithm) based on multiple parameters including detection probability and degree of exploration (equation 6))
and wherein the first model is trained to optimize the accuracy of the resulting determination of whether the channel is in use and the values of the one or more other parameters, wherein the first model was trained by minimizing a loss function that comprises a first term to encourage the classification model to select a subset of nodes so as to optimize accuracy of the resulting determination of whether the channel is in use and one or more subsequent terms to optimize the one or more other parameters. -Page 12, Section I; Page 18, Section V-B (section I “Before transmitting, SUs are required to sense the available channels which are not occupied by PUs so as to minimize the interference caused to the PUs…. A novel reward function is devised to improve the accuracy of channel status prediction during the learning process.” Minimizing interference is minimizing the loss function. Section V-B recites, “In the proposed channel selection algorithm, parameter α is the learning rate, which determines the accuracy of the prediction of channel status thus the effectiveness of the resulting scanning preference list. It should be noted that the optimal value of α may vary with the network settings.” As shown in equation 6 parameter α is optimized based on detection probability and system parameter c.)
Regarding Claim 20, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 1, wherein the first model is further trained to output a type of channel information that is to be obtained by the subset of other nodes. -Page 14, 15,17; Section IV-A. ( Section IV-A describes channel selection Algorithm. Recites, “….To accelerate the learning procedure with a better view of the recent usage of the primary channels, the computation of reward is based on the channel status maintained by both SUk and its neighbors…” Page 17 recites, “Moreover, it also collects the channel status maintained by its neighbors, which is received via the dedicated control channel. When there is a demand at SUk, it selects a primary channel to detect and attempt to access by ϵ-greedy strategy based on the scanning preference list. Moreover, by applying the selection rule in (6), SUk chooses SUf as the cooperation partner, and sends a notification message to inform SUf to detect the selected primary channel. Once SUf finishes the detection, it reports the result to SUk. Then, SUk makes the final decision based on the received detection result. In case that multiple cooperation partners are selected, SUk will send a notification message to each partner. The partners detect the selected primary channel individually, and report the results to SUk. Then, SUk fuses the detection results of the partners based on a certain fusion rule, such as the majority rule or the weighted rule [25], to make the final decision.”)
Regarding Claim 22, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 20 z wherein the type of channel information comprises: an indication of whether the channel is in use, as determined by a respective other node; or measurements of the channel quality as determined by a respective other node. -Page 14; Section IV-A (recites, “When a call arises at SUk, it takes an action by scanning a particular primary channel e.g. channel ci, and gets a real-valued reward evaluating the choice of the action” Equation 2 shows the reward calculation….. “where Nk represents the set of indexes of the neighbors of SUk. sj (ci) is the status of channel ci maintained by neighbor SUj , where sj (ci) = 1 if channel ci was detected being idle and SUj accessed it successfully,” From above description it is understandable that channel information of each neighbor sj (ci) i.e., whether the channel is in use or idle is taken into account to calculate the reward at the action agent)
Regarding Claim 23, Ning and Sarikhani teach the limitations of Claim 1.
Ning further teaches,
The method as in claim 1, comprising: receiving the obtained channel information transmitted from the subset of other nodes; -Page 14; Section IV-A (recites, “When a call arises at SUk, it takes an action by scanning a particular primary channel e.g. channel ci, and gets a real-valued reward evaluating the choice of the action” Equation 2 shows the reward calculation….. “where Nk represents the set of indexes of the neighbors of SUk. sj (ci) is the status of channel ci maintained by neighbor SUj , where sj (ci) = 1 if channel ci was detected being idle and SUj accessed it successfully,”)
and determining whether the channel between the first node and the target node is in use based on the obtained channel information. -Page 15; Section IV-A (recites,” According to (2), the reward rk (st; ci) is calculated in different ways depending on the value of channel status sk (ci). Precisely, SUk will get a positive reward if sk (ci) = 1 indicating it is able to access channel ci successfully, otherwise a negative reward.” Positive reward means channel is accessible and negative rewards means channel is occupied and not accessible)
Regarding Claim 24, Ning and Sarikhani teach the limitations of Claim 23.
Ning further teaches,
The method as in claim 23 wherein the first model is further trained to determine a manner in which to combine the obtained channel information in order to determine whether the channel is in use. -Page 14; Section IV-A (Equation 2 shows evaluation of reward by combining (summation) of channel information (status) of neighboring nodes sj, wheresj (ci) is the status of channel ci maintained by neighbor SUj )
Regarding Claim 26, Ning and Sarikhani teach the limitations of Claim 23.
Ning further teaches,
The method as in claim 23 further comprising using a third model trained using a third machine learning process to determine a manner in which to combine the obtained channel information in order to determine whether the channel is in use. -Page 17, Section IV (Although the analysis is done by Q-learning, however, it could use “hidden-mode MDP based varying environment modeling [26], to model-based method for detecting changes in environment models [27], to context detection based RL algorithm [28] and its extension [29] “ as second or third machine learning process to determine a manner in which to combine the obtained channel information in order to determine whether the channel is in use)
Claim 33 is the apparatus claim corresponding to the method Claim 1 which is rejected above. Applicant’s attention is drawn towards Claim 1. Claim 33 is rejected under the same rational as Claim 1.
Ning further teaches,
the first node comprising: a memory comprising instruction data representing a set of instructions; and a processor configured to communicate with the memory and to execute the set of instructions, the set of instructions, when executed by the processor, causing the processor to: -Fig. 1 (Cognitive radio network is shown in Fig. 1 consisting of SUs and Pus where each of the SUs and Pus consists of one or more processors, memory and other components)
Claim 34 is the apparatus claim corresponding to the method Claim 2 which is rejected above. Applicant’s attention is drawn towards Claim 1. Claim 33 is rejected under the same rational as Claim 2.
Claims 18, 21 are rejected under 35 U.S.C. 103 as being unpatentable over Ning, in view of Rahil Sarikhani and further in view of Montlary et al. (Patent No: US 12189643 B1), hereinafter, Montlary.
Regarding Claim 18, Ning and Sarikhani teach the limitations of Claim 15.
Although implicit, Ning does not explicitly mention,
The method as in claim 15 wherein the first model is a classification model, and wherein the first model was trained using a training dataset comprising example inputs and ground truth subsets of the other nodes from which to obtain channel information.
However, in an analogous invention Montlary teaches,
The method as in claim 15 wherein the first model is a classification model, and wherein the first model was trained using a training dataset comprising example inputs. -col. 6, line 24-29; (col. 6, line 24-29 recites, “The user classification model may include a machine learning model, the machine learning model being trained based at least in part on: (i) user data of a plurality of users, the service facility being one of a plurality of units, and (ii) previous interactions with the computer system by each of the plurality of users.“)
and ground truth subsets of the other nodes from which to obtain channel information -col. 40, line 41-67 (col. 40, line 41-67 recites, “For example, the prediction model management engine 902 may perform supervised or unsupervised learning to generate prediction models. Typically, especially in the case of supervised learning, as part of the training and validation processes, ground truth labels may be created for data samples and included in (or alongside) one or more of the subsets of data determined by the data preparation engine 914. A ground truth label may refer to information that is provided by direct observation, as opposed to information provided by inference. The ground truth label may be used to measure the accuracy of a training data set's classification.….The prediction that is output by the prediction model may be compared against the ground truth label to determine the accuracy of the prediction, and the comparison results may be used to adjust (e.g., learn) weights and/or parameters of the model accordingly.” It is easily understandable that in the context of this invention training dataset will be channel information from the cooperating nodes)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the “Reinforcement Learning Enabled Cooperative Spectrum Sensing in Cognitive Radio Networks “ proposed by Ning to include the concept of “first model is a classification model, and wherein the first model was trained using a training dataset comprising example inputs and ground truth subsets of the other nodes from which to obtain channel information” of Montlary. One of ordinary skill in the art would have been motivated to make this modification in order to improve the classification accuracy for a large number of users (nodes) (col. 62, line 61-62).
Regarding Claim 21, Ning and Sarikhani teach the limitations of Claim 1.
Although implicit, Ning does not explicitly mention,
The method as in claim 1, further comprising using a second model trained using a second machine learning process to output a type of channel information that is to be obtained by the subset of the other nodes.
However, in an analogous invention Montlary teaches,
The method as in claim 1, further comprising using a second model trained using a second machine learning process to output a type of channel information that is to be obtained by the subset of the other nodes. -col. 5; line 24-28 (recites, “receive second training data samples corresponding to a second subset of user data records of users of the service facility, the second subset being a subset of the first subset; and train, in a second training round, the user classification model utilizing the second training data samples.” It is easily understandable that in the context of the disclosure the output of the machine learning process will be the channel information (occupancy))
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the “Reinforcement Learning Enabled Cooperative Spectrum Sensing in Cognitive Radio Networks “ proposed by Ning to include the concept of “a second model trained using a second machine learning process to output a type of channel information that is to be obtained by the subset of the other nodes.” of Montlary. One of ordinary skill in the art would have been motivated to make this modification in order to improve the classification accuracy for a large number of users (nodes) (col. 62, line 61-62).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AHMED SAIFUDDIN whose telephone number is (703)756-4581. The examiner can normally be reached Monday-Friday 8:30am-6:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, KHALED M KASSIM can be reached on 571-270-3770. 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.
/AHMED SAIFUDDIN/Examiner, Art Unit 2472
/KHALED M KASSIM/supervisory patent examiner, Art Unit 2475