DETAILED ACTION
Response to Amendment
The amendment filed on 16 March 2026 has been entered.
Claims 1-20 are pending.
Claims 1-4, 7, 9-12, 15, 17-19 are amended.
Applicant’s amendments to the Claims have overcome each and every rejection under 35 USC 101 previously set forth in the Non-Final Office Action mailed 12 January 2026.
Response to Arguments
Applicant’s remarks, regarding the rejections of claims under 35 USC 103, have been fully considered.
While Applicant respectfully disagrees with the rejection, and solely in order to expedite prosecution, Claim 1 has been amended to further differentiate the claim from the cited references. Particularly, Claim 1 has been amended to require:
deploying a policy to edge devices in an edge computing environment,
wherein the edge computing environment includes an edge hub that is
not able to directly query the edge devices; receiving, at the edge hub
from the edge devices, information from at least some of the edge
devices, wherein the information comprises performance metrics of the
edge devices; causing an edge device policy analysis module to analyze,
using an artificial intelligence (AI) reasoning model, the policy to
understand an intent of deploying the policy, wherein the analyzing includes
discounting a weight value assigned to data points that are determined to
not apply to a current decision of a first of the edge devices, wherein the
analysis comprises inputting the information into the AI reasoning
model, wherein an output of the AI reasoning model includes relative
weight assignments for data points associated with the edge devices; and
causing the policy to be altered based on the analysis.
Applicant submits none of the art of record (singularly or in any combination) teach or suggest an edge computing environment arranged and that operates as now recited in Claim 1 of the present application.
Applicant’s arguments have been considered, but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
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.
Claims 3, 7, 11, 15 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 3 and analogous claim 11 recites the limitation "the weight values" in line 2. There is insufficient antecedent basis for this limitation in the claim. For examination purposes, the term "the weight values" has been construed to be “weight values”.
Claim 7 and analogous claim 15 recites the limitation "the rule" in line 6. There is insufficient antecedent basis for this limitation in the claim. For examination purposes, the term "the rule" has been construed to be “the first rule”.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Latapie et al. (U.S. Pre-Grant Publication No. 20230196063, hereinafter ‘Latapie'), in view of Pick et al. (U.S. Pre-Grant Publication No. 20230188408, hereinafter 'Pick'), Xu et al. (U.S. Pre-Grant Publication No. 20230087821, hereinafter 'Xu'), and further in view of Wang et al. (NPL: "Attention-Weighted Federated Deep Reinforcement Learning for Device-to-Device Assisted Heterogeneous Collaborative Edge Caching", hereinafter 'Wang').
Regarding claim 1 and analogous claims 9, 17, Latapie teaches A computer-implemented method, comprising: wherein the information comprises performance metrics of the edge devices;
causing an edge device policy analysis module to analyze, using an artificial intelligence (AI) reasoning model, the policy to understand an intent of deploying the policy ([0042] According to various embodiments, FIG. 3 illustrates an example hierarchy 300 for a deep fusion reasoning engine (DFRE). For example, DFRE process 248 shown in FIG. 2 may execute a DFRE for any number of purposes. In particular, DFRE process 248 may be configured to analyze sensor data in an IoT deployment (e.g., video data, etc.), to analyze networking data for purposes of network assurance, control, enforcing security policies and detecting threats, facilitating collaboration, or, as described in greater detail below, to aid in the development of a collaborative knowledge generation and learning system for visual programming.; [0054] Thus, a analyzing, using an artificial intelligence (AI) reasoning model, the policy to understand an intent of deploying the policy DFRE generally refers to a cognitive engine capable of taking sub-symbolic data as input (e.g., raw or processed sensor data regarding a monitored system), recognizing symbolic concepts from that data, and applying symbolic reasoning to the concepts, to draw conclusions about the monitored system.; [0058] More specifically, in various embodiments, DFRE middleware 402 may obtain sub-symbolic data 408. In turn, DFRE middleware 402 may leverage various ontologies, programs, rules, and/or structured text 410 to translate sub-symbolic data 408 into symbolic data 412 for consumption by DFRE agent 404. This allows DFRE agent 404 to apply symbolic reasoning to symbolic data 412, to populate and update a DFRE knowledge base (KB) 416 with knowledge 414 regarding the problem space (e.g., the network under observation, etc.). In addition, DFRE agent 404 can leverage the stored knowledge 414 in DFRE KB 416 to make assessments/inferences.),
Latapie fails to teach deploying a policy to edge devices in an edge computing environment, wherein the edge computing environment includes an edge hub that is not able to directly query the edge devices; receiving, at the edge hub from the edge devices, information from at least some of the edge devices, wherein the information comprises performance metrics of the edge devices; causing an edge device policy analysis module to analyze, using an artificial intelligence (AI) reasoning model, the policy to understand an intent of deploying the policy, wherein the analyzing includes discounting a weight value assigned to data points that are determined to not apply to a current decision of a first of the edge devices, wherein the analysis comprises inputting the information into the AI reasoning model, wherein an output of the AI reasoning model includes relative weight assignments for data points associated with the edge devices; and causing the policy to be altered based on the analysis.
Pick teaches deploying a policy to edge devices in an edge computing environment ([Abstract] This disclosure describes systems, methods, and devices related to identification and assessment of security threats to a computer system using zero trust security. A method may include receiving, at a policy enforcement device of a computer network, first data from a first subsystem of the computer network; receiving, at the policy enforcement device, second data from a second subsystem of the computer network, the first subsystem different than the first subsystem; identifying, by the policy enforcement device, based on a comparison of at least one of the first data or the second data to a security policy, a security threat to the computer network; and causing, by the policy enforcement device, a threat intelligence device of the computer network to determine a risk of the security threat.; [0017] In one or more embodiments, a centralized networks analytics model may execute on a cloud-based server or edge device of a network, and may function as software as a solution (SaaS). The centralized networks analytics model may use analytic accelerators to deploying a policy to edge devices in an edge computing environment evaluate a variety of network performance data from a variety of sources to identify and predict performance issues, correlate related network performance issues, identify root causes of network performance issues, remediate the network performance issues, and recommend remediation actions in response to the network performance issues.);
Latapie and Pick are considered to be analogous to the claimed invention because they are in the same field of edge computing. In view of the teachings of Latapie, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Pick to Latapie before the effective filing date of the claimed invention in order to evaluate a variety of network performance data from a variety of sources to identify and predict performance issues, correlate related network performance issues, identify root causes of network performance issues, remediate the network performance issues, and recommend remediation actions in response to the network performance issues (cf. Pick, [0016] There is therefore a need for enhanced analysis and remediation of network performance.; [0017] In one or more embodiments, a centralized networks analytics model may execute on a cloud-based server or edge device of a network, and may function as software as a solution (SaaS). The centralized networks analytics model may use analytic accelerators to evaluate a variety of network performance data from a variety of sources to identify and predict performance issues, correlate related network performance issues, identify root causes of network performance issues, remediate the network performance issues, and recommend remediation actions in response to the network performance issues.).
Xu teaches wherein the analyzing includes discounting a weight value assigned to data points that are determined to not apply to a current decision of a first of the edge devices; and causing the policy to be altered based on the analysis ([0284] The first performance tracking module may monitor a long-term key performance indicator (key performance indicator, KPI) of the system, and the KPI may indicate the first data collection and training module to generate the reward function R(θ,ϕ). R represents a reward, and a target parameter θ is performance data obtained by the terminal device by executing AI decision information, for example, a throughput or a packet loss rate. A weight value ϕ of the target parameter is determined by a first AI entity based on performance data of one or more terminal devices, and indicates weights of different short-term KPIs. In other words, the weight value ϕ of the target parameter may be obtained by the first performance tracking module in the first AI entity by performing long-term monitoring on performance of all terminal devices in the system.; [0288] It is that the reward function is R(θ,ϕ)=α×thp+β×jfi+γ×pdr. The target parameter θ={thp, jfi, pdr} includes three types of performance data: a throughput, a fairness parameter, and a packet loss rate. ϕ—{α,β,γ} includes weights of the three types of performance data. It is assumed that an initial value is ϕ={1,1,1}. If the PMF detects that fairness deteriorates due to an emergency after the system runs for a period of time, update of the reward function is triggered, and the weights of the three types of performance data are updated to ϕ={1,2,1}.; [0310] The following uses a DRL scheduling process as an example to describe the decision early stopping solution shown in FIG. 13A, FIG. 13B, and FIG. 13C.; [0311] For example, if the system makes a scheduling decision for five users, a decision weight generated by the DRL may be {1.5, 1.1, 1.2, 0.2, 0}. However, in a possible case, an estimated throughput of a user 0 and a user 4 is 0. In this case, scheduling the user 0 and/or the user 4 inevitably causes a waste of system resources.; [0312] After predicting this case, the performance tracking module may generate a decision mask. For example, the decision masks of the five users are respectively {0, 1, 1, 1, 0}. Based on the decision mask, the performance tracking module may obtain masked analyzing includes discounting a weight value assigned to data points that are determined to not apply to a current decision of a first of the edge devices decision weights are respectively {0, 1.1, 1.2, 0.2, 0}. In this case, the system schedules a user 2 based on the decision weight information. It can be learned that the causing the policy to be altered based on the analysis scheduling helps reduce a waste of system resources and optimize overall performance of the system.; [0307] The decision mask information is used for performing mask processing on the AI decision information. In this way, a part that reduces system performance is processed. For example, if one or more users accessing the system significantly reduce system performance, the performance tracking module may minimize a weight of an AI decision of the one or more users, and the one or more users no longer execute a corresponding AI decision. The decision mask information may be directly obtained based on a prediction result, or may be obtained by using a backup algorithm in the performance tracking module.).
Latapie, Pick, and Xu are considered to be analogous to the claimed invention because they are in the same field of edge computing. In view of the teachings of Latapie and Pick, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Xu to Latapie before the effective filing date of the claimed invention in order to help improve a processing capability of the radio access network (cf. Xu, [0003] An artificial intelligence (AI) technology is successfully applied in the field of image processing and natural language processing. For example, the AI technology is applied to a network layer (such as network optimization, mobility management, and resource allocation), or the AI technology is applied to a physical layer (such as channel coding/decoding, channel prediction, and a receiving device). An AI entity may be deployed in an access network to improve a processing capability of the access network (for example, improve resource allocation efficiency). However, currently a basic interaction mode between the AI entity in the access network and user equipment (UE) is not defined, and the AI technology cannot be efficiently applied to a radio access network.; [0004] Embodiments of this application provide an information processing method and a related device. In the information processing method, an AI technology may be applied to a radio access network, to help improve a processing capability of the radio access network.).
Wang teaches wherein the edge computing environment includes an edge hub that is not able to directly query the edge devices ([1. Introduction, pg. 155] Thus, in this article, we are motivated to exploit the frame work design of heterogeneous collaborative edge caching by jointly optimizing the node selection and cache replacement in D2D assisted mobile networks, and consider the flexible wherein the edge computing environment trilateral collaboration among UEs (user equipment), includes an edge hub that is not able to directly query the edge devices BSs (base stations) and the cloud server. We formulate the joint optimization problem as a Markov decision process (MDP), and propose an attention weighted federated deep reinforcement learning (AWFDRL) framework to address the problem.);
receiving, at the edge hub from the edge devices, information from at least some of the edge devices, wherein the information comprises performance metrics of the edge devices
([A. Whole Process, pg. 159] 3) Aggregation Phase: There are two steps in this phase which is the right part of round i in Fig. 4. Firstly, after E local training rounds, each information from at least some of the edge devices, wherein the information comprises performance metrics of the edge devices UE collects its training evaluation indicators (e.g., average reward, average loss, and hit rate) during the training phase, and receiving, at the edge hub from the edge devices sends them to the local BS with the local model parameter θu t which is step 4. Then, in step 5, the BS calculates each agent’s aggregation weights based on the training evaluation indicators. Particularly, we employ the attention mechanism to provide different devices with different aggregation weights. Then we aggregate different local DQN parameters with the attention weights, instead of aggregating them equally or calculating weights simply based on the data size.);
causing an edge device policy analysis module to analyze, using an artificial intelligence (AI) reasoning model, the policy to understand an intent of deploying the policy ([IV. PROBLEM FORMULATION, pg. 158] In this section, we formulate the joint problem of node selection and cache replacement in a UE as a Markov decision process (MDP). Particularly, the UE’s state and action are introduced in Sec. IV-A and Sec. IV-B, respectively. Sec. IV-C calculates the system reward. Finally in Sec. IV-D, the objective of UEs is to find an optimal policy to optimize the expected long-term reward which is defined as a value function.; [A. Whole Process, pg. 159] Then, in step 5, the causing an edge device policy analysis module to analyze BS calculates each agent’s aggregation weights based on the training evaluation indicators. Particularly, we employ the attention mechanism to provide different devices with different aggregation weights. Then we aggregate different local DQN parameters with the attention weights, instead of aggregating them equally or calculating weights simply based on the data size.),
wherein the analysis comprises inputting the information into the AI reasoning model, wherein an output of the AI reasoning model includes relative weight assignments for data points associated with the edge devices ([A. Whole Process, pg. 159] Then, in step 5, the BS calculates each agent’s aggregation weights based on the training evaluation indicators. Particularly, we employ the attention mechanism to wherein an output of the AI reasoning model includes relative weight assignments for data points associated with the edge devices provide different devices with different aggregation weights. Then we aggregate different local DQN parameters with the attention weights, instead of aggregating them equally or calculating weights simply based on the data size.; [C. Weighted Federated Aggregation, pg. 160-161] wherein the analysis comprises inputting the information into the AI reasoning model In this model, we use the reward C(sut,aut) and some device-related indicators as the measurement to evaluate the local model’s contribution to the global model. The update process of weighted federated aggregation is shown in Fig. 6.); and
Latapie, Pick, Xu, and Wang are considered to be analogous to the claimed invention because they are in the same field of machine learning. In view of the teachings of Latapie, Pick, and Xu, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Wang to Latapie before the effective filing date of the claimed invention in order to optimize the aggregation weights to avoid the imbalance of local model quality (cf. Wang, [Abstract, pg. 154] We further design an attention-weighted federated deep reinforcement learning (AWFDRL) model that uses federated learning to improve the training efficiency of the Q-learning network by considering the limited computing and storage capacity, and incorporates an attention mechanism to optimize the aggregation weights to avoid the imbalance of local model quality. We prove the convergence of the corresponding algorithm, and present simulation results to show the effectiveness of the proposed AWFDRL framework in reducing average delay of content access, improving hit rate and offloading traffic.).
Regarding claim 2 and analogous claims 10, 18, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 1, The computer program product of claim 9, and The system of claim 17, respectively.
Latapie teaches wherein the information is based on selected from the group consisting
of: memory resources of the edge devices, types of memory that the edge devices use, an amount of processing operations being performed and/or capable of being concurrently performed by the edge devices, central processing unit (CPU) resources of the edge devices, and geographical locations of the edge devices ([0028] FIG. 1B illustrates an example of network 100 in greater detail, according to various embodiments. As shown, network backbone 130 may provide connectivity between devices located in different geographical areas and/or different types of local networks. For example, network 100 may comprise local/ branch networks 160, 162 that include devices/nodes 10-16 and devices/nodes 18-20, respectively, as well as a data center/cloud environment 150 that includes servers 152-154. Notably, local networks 160-162 and data center/cloud environment 150 may be located in different geographic locations.; [0156] In some embodiments, a user may also be able to review and adjust the concepts and relationships relied upon by DFRE metamodel 700. For instance, FIG. 9 illustrates an example user interface 900 allowing a user to select which concepts are to be used by metamodel 700 when assessing the generation of artificial intelligence models, according to various embodiments. As shown, user interface 900 may include various inputs 908 that allow a user to interact with the system. For instance, user interface 900 may include a button or other input 1008 b that allows the user to select a target node/device to which an artificial intelligence model is to be deployed. Once a node is selected, the user may interact with input 1008 c, to review wherein the information is based on selected from the group consisting of: details about that node, such as its type, geographical locations of the edge devices location in the network, types of memory that the edge devices use capabilities, other an amount of processing operations being performed and/or capable of being concurrently performed by the edge devices functions, memory resources of the edge devices, central processing unit (CPU) resources of the edge devices resource usage, or the like. In other instances, this information may be obtained through other means, such as from system 802 in FIG. 8.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 3 and analogous claim 11, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 2, The computer program product of claim 10, respectively.
Wang teaches comprising: adding to a first of the weight values assigned by the AI reasoning model to data points that are determined to apply to the current decision of the first of the edge devices ([D. Convergence Analysis, pg. 161] For those who have more computing capacity, experience pool, training batch and better training data, the difference between F∗ u and F∗ is theoretically less than other UEs, and the variance of stochastic gradients σ2 u is also smaller than others. By considering these divergences, we model a more complicated weight factor wu in (16). The UEs with better optimal value will adding to a first of the weight values assigned by the AI reasoning model to data points that are determined to apply to the current decision of the first of the edge devices get more weight in the aggregation process, so the term Γ will be less than those who just average local parameter or those who just consider the data size of local data. With a less Γ, we will get a better convergence bound. Actually, we can consider Γ as a function of the aggregation weight wu.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 4 and analogous claims 12, 19, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 1, The computer program product of claim 9, and The system of claim 17, respectively.
Pick teaches wherein the analysis is based on a first rule of the policy, wherein the first rule specifies that the first edge device is to have a key performance indicator (KPI) value within the performance metrics that exceeds a predetermined threshold ([0019] The centralized networks analytics model may prioritize the data to analyze based on its source and/or type, which may adjust over time due to training data and/or feedback. When performance data are trending in a manner that indicates a future performance issue is plausible, the centralized networks analytics model may predict the performance issue and search for its root cause. For example, when one network wherein the first rule specifies that the first edge device is to have a key performance indicator (KPI) value within the performance metrics that exceeds a predetermined threshold device's performance metrics experience poor performance (e.g., based on comparisons of performance metrics to respective thresholds, such as utilization thresholds, packet loss thresholds, signal-to-noise thresholds, latency thresholds, jitter thresholds, packet error thresholds, and the like, indicating a metric is too high or low for intended performance), the centralized networks analytics model may search for other devices with which the network device may communicate, and may analyze the other devices for possible performance issues or trends toward possible performance issues.; [0031] In one or more embodiments, the centralized networks analytics model 106 may rely on baseline metrics for different types of performance data and devices to use in comparison (e.g., to identify when performance data are unusual and/or indicative of a performance issue). For example, the baseline metrics may define latency and/or transmission rate in a geographic region. The centralized networks analytics model 106 wherein the analysis is based on a first rule of the policy may identify reusable use cases to apply for different users and situations, for example, to detect performance issues and anomalies.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 5 and analogous claim 13, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 4, The computer program product of claim 12, respectively.
Pick teaches wherein the KPI of the first edge device is selected from the group consisting of: available memory resources, computer processing unit resources, access to a predetermined physical component, an amount of processing being performed, and physical movement ([0019] In one or more embodiments, using one or more application programming interfaces (APIs), the centralized networks analytics model may receive network data, such as syslog data, infrastructure management data, Ethernet/Internet Protocol data, and network configuration settings. Text data, such as from the syslog data, may be analyzed using natural language processing, for example, to generate performance metrics for analysis. The centralized networks analytics model may prioritize the data to analyze based on its source and/or type, which may adjust over time due to training data and/or feedback. When performance data are trending in a manner that indicates a future performance issue is plausible, the centralized networks analytics model may predict the performance issue and search for its root cause. For example, when one network device's performance metrics experience poor performance (e.g., based on comparisons of performance metrics to respective thresholds, such as available memory resources utilization thresholds, packet loss thresholds, signal-to-noise thresholds, computer processing unit resources latency thresholds, jitter thresholds, an amount of processing being performed packet error thresholds, and the like, indicating a metric is too high or low for intended performance), the centralized networks analytics model may search for other devices with which the network device may communicate, and may analyze the other devices for possible performance issues or trends toward possible performance issues.; [0021] In one or more embodiments, the centralized networks analytics model may use a data manager to collect and curate performance data from different sources. The centralized networks analytics model may rely on baseline metrics for different types of performance data and devices to use in comparison (e.g., to identify when performance data are unusual and/or indicative of a performance issue). For example, the baseline metrics may define latency and/or transmission rate physical movement in a geographic region. The centralized networks analytics model may identify reusable use cases to apply for different users and situations, for example, to detect performance issues and anomalies.; [0045] A device operational status 376 may indicate whether a access to a predetermined physical component network device is active or inactive, disconnected, in standby mode, or the like. For example, an inactive or disconnected device may indicate a performance issue. Incident management data 373 may identify performance issues that have occurred and any remedies applied.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 6 and analogous claims 14, 20, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 1, The computer program product of claim 9, and The system of claim 17, respectively.
Latapie teaches wherein the Al reasoning model is a neuro-symbolic Al model ([0136] Thus, DFRE metamodel 700 is a neuro-symbolic metamodel that leverages both sub-symbolic processing (e.g., using deep/neural networks) and symbolic reasoning.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 7 and analogous claim 15, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 1, The computer program product of claim 9, respectively.
Pick teaches wherein causing the policy to be altered based on the analysis includes modifying a first rule of the policy and modifying a second rule of the policy, wherein the modification of the first rule comprises decreasing a relative strictness of predetermined thresholds of the first rule ([0053] At block 408, the device optionally may evaluate the network performance data and/or metrics for a match with an existing use case. Use cases may be generated to define criteria (e.g., threshold values) for certain performance metrics, emphasis of certain metrics over others (e.g., based on which performance metrics most strongly correlate to a performance issue, etc.), and remediation options (e.g., based on selected remediation options and feedback regarding the outcome of the remediation options). When the performance data and/or metrics match those of a known performance issue, the use case for the known performance issue may be selected at block 410, resulting in an identification of the root cause and/or a selection of a remediation action based on the remediation applied to the known performance issue (e.g., what has worked in the past to solve a similar issue). When the performance data and/or metrics do not match any existing use cases, the device at block 412 may generate a new use case to test and implement for future analysis of network performance data.; [0059] At block 424, optionally, the device may modify machine learning models used to identify performance issues, root causes, correlations, use cases, and remediation actions. The wherein causing the policy to be altered based on the analysis includes modifying a first rule of the policy and modifying a second rule of the policy modifications may include wherein the modification of the first rule comprises decreasing a relative strictness of predetermined thresholds of the first rule adjusting performance thresholds, changing the emphasis (e.g., weighting) of certain metrics (e.g., higher emphasis indicating a higher correlation between a metric and its relationship/indication of a performance issue), updating the remediation actions to recommend or select for a performance issue, updating relationships between metrics to use in identifying correlations (e.g., recognizing that high usage and increased latency may be related), and the like.),
wherein the rule is used to restrict which edge devices are allowed to operate in
the edge computing environment ([0057] At block 420, the device may select or recommend remediation actions for the respective network performance issues. The remediation actions may be selected based on previously selected and effective remediation actions, such as those included in an existing use case, training data (e.g., select action X for issue Y), and/or learned outcomes of remediation actions (e.g., whether a remediation action resolved the issue). For example, remediation may include re-routing traffic, prioritizing certain traffic, wherein the rule is used to restrict which edge devices are allowed to operate in the edge computing environment activating or de-activating network resources, modifying user roles and responsibilities, modifying network paths, and the like.),
wherein the modification of the second rule includes granting permission for edge devices associated predetermined weight values to be allowed to apply a predetermined application
([0059] At block 424, optionally, the device may modify machine learning models used to identify performance issues, root causes, correlations, use cases, and remediation actions. The modifications may include adjusting performance thresholds, changing the emphasis (e.g., weighting) of certain metrics (e.g., higher emphasis indicating a higher correlation between a metric and its relationship/indication of a performance issue), updating the remediation actions to recommend or select for a performance issue, updating relationships between metrics to use in identifying correlations (e.g., recognizing that high usage and increased latency may be related), and the like.; [0057] At block 420, the device may select or recommend remediation actions for the respective network performance issues. The remediation actions may be selected based on previously selected and effective remediation actions, such as those included in an existing use case, training data (e.g., select action X for issue Y), and/or learned outcomes of remediation actions (e.g., whether a remediation action resolved the issue). For example, remediation may include re-routing traffic, prioritizing certain traffic, wherein the modification of the second rule includes granting permission for edge devices associated predetermined weight values to be allowed to apply a predetermined application activating or de-activating network resources, modifying user roles and responsibilities, modifying network paths, and the like.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 8 and analogous claim 16, Latapie, as modified by Pick, Xu, and Wang, teaches The computer-implemented method of claim 1, The computer program product of claim 9, respectively.
Pick teaches wherein causing the policy to be altered based on the analysis includes: determining a supplement to the policy, and outputting an indication of the determined supplement ([0021] In one or more embodiments, the centralized networks analytics model may use a data manager to collect and curate performance data from different sources. The centralized networks analytics model may rely on baseline metrics for different types of performance data and devices to use in comparison (e.g., to identify when performance data are unusual and/or indicative of a performance issue). For example, the baseline metrics may define latency and/or transmission rate in a geographic region. The centralized networks analytics model may identify reusable use cases to apply for different users and situations, for example, to detect performance issues and anomalies.; [0022] In one or more embodiments, the centralized networks analytics model may determining a supplement to the policy identify remediation options. For example, the centralized networks analytics model may determine that packet re-routing is an option to avoid a performance issue, may select a different network circuit for use, update packet priorities, and the like. The centralized networks analytics model automatically may outputting an indication of the determined supplement implement the remediation and report to a user that the remediation has been implemented, or may assess and report remediation options to a user for selection.; [0032] In one or more embodiments, the centralized networks analytics model 106 may identify remediation options. For example, the centralized networks analytics model 106 may determine that packet re-routing is an option to avoid a performance issue, may select a different network circuit for use, update packet priorities, and the like. The centralized networks analytics model 106 automatically may implement the remediation and report to a user (e.g., using the notification service 118) that the remediation has been implemented, or may assess and report remediation options to a user for selection.).
Latapie, Pick, Xu, and Wang are combinable for the same rationale as set forth above with respect to claim 1.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 MAGGIE MAIDO whose telephone number is (703) 756-1953. The examiner can normally be reached M-Th: 6am - 4pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michael Huntley can be reached on (303) 297-4307. 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.
/MM/Examiner, Art Unit 2129
/MICHAEL J HUNTLEY/Supervisory Patent Examiner, Art Unit 2129
Prosecution Insights