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 § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barron et al. (NPL: Towards Self-Adaptive Network Management for a Recursive Network Architecture, published July 2016, hereinafter “Barron”) in view of Sanchez-Navarro et al. (NPL: Advanced spatial network metrics for cognitive management of 5G networks, published Aug. 2020, hereinafter “Sanchez-Navarro”).
Regarding claim 1, Barron teaches a non-transitory computer-readable medium having instructions stored thereon for programming a device for performing steps of:
receiving information and data from a network having resources (Barron, Section III, Paragraph 1 – “RINA takes as a starting point the basic premise that networking is inter-process communication (IPC) and only IPC [20], and the network is modelled as the interconnection of applications over different scopes, known as Distributed Application Facilities (DAFs). RINA considers one single layer, recursively repeated as needed, implementing specialized DAFs (i.e. IPC processes on each system that work together to form a Distributed IPC Facility (DIF)), and two protocols at each layer: one for data transfer, with a consistent QoS model (termed a QoS cube) [19], and another for application (layer) management.” – teaches receiving information and data from a network having resources (takes inter-process communication as starting point, network is modelled as interconnection of applications));
implementing feature selection on one or more network Machine Learning (ML) models (Barron, Section IV, Subsection A “Observe”, Paragraph 1 – “Pertinent network events observed from systems/devices are identified and used as input to the knowledge model (i.e. pattern matching). Due to the sheer volume of events generated from systems/devices, events need to be aggregated/correlated. This function identifies specific event triggers from deployed DIF configurations and ensures event collection from network systems and devices that involves pre-processing events to aggregate the events generated and identify the most important and irregular events for submission to the knowledge model and machine learning algorithm while filtering routine and regular events.” – teaches implementing feature selection on one or more network Machine Learning models (autonomic control loop performs feature selection in the Observe function by preprocessing events and identifying the most important and irregular events)), such that each comprises a multi-stage ML processing pipeline to control the resources and with specified interfaces to other control applications (Barron, Section IV, Paragraph 1 – “The self-adaption mechanisms enable self-control (i.e. self-FCAPS [21]) of underlying network operations, functions and state minimising expensive, manual system administrator intervention. The self-adaptive network management framework uses a closed autonomic control loop [22] referred to as a decision cycle originally designed by John Boyd, a military strategist, and includes the sub-processes of Observe, Orient, Decide and Act (OODA). In our system the sub-processes perform the following tasks.” – teaches a pipeline of a plurality of functions to control the resources and with specified interfaces to control other applications (teaches observe, orient, decide, and act as a plurality of functions to control the resources and interfaces to control applications));
Barron fails to explicitly teach utilizing one or more feature graph engines (FGEs) that constructs one or more feature graphs, from the information and data, as a functional component to derive a design-time set of feature vector for a specific context, each feature graph explicitly encoding network-layer relationships across multiple network layers including service-layer, control-layer, and resource-layer constructs; and implementing changes to the one or more feature graphs based on any run-time updates from the pipeline including changes in configuration state, operational state, or feature-policy constraints.
However, analogous to the field of the claimed invention, Sanchez-Navarro teaches:
utilizing one or more feature graph engines (FGEs) that constructs one or more feature graphs, from the information and data, as a functional component to derive a design-time set of feature vector for a specific context, each feature graph explicitly encoding network-layer relationships across multiple network layers including service-layer, control-layer, and resource-layer constructs (Sanchez-Navarro, Figs. 1 & 3, and in Section 4 Paragraph 1 – “All the segments of the 5G network including Radio Access Network(RAN), Edge, Core network as well as mobile users are continuously monitored by the topology discovery sensors, which allow modelling dynamically the 5G network as a connected directed graph G = (N, E), consisting of a set N nodes or vertices, corresponding to the 5G network devices, and set E of edges corresponding to the network connections between devices.” and Section 4.1.2 Paragraph 5 & Algorithm 2 – “The algorithm returns the vector with the 5G Closeness Centrality values for each node that is worth calculating the 5G spatial metrics, i.e. those physical nodes that might allocate a VNF” – teaches utilizing one or more feature graph engines (modelling dynamically the 5G network as a connected directed graph), from the information and data (corresponding to network devices and connections between devices), as a functional component to derive a design-time set of feature vector for a specific context (derives feature vector with closeness centrality values for nodes that might allocate a VNF), each feature graph explicitly encoding network-layer relationships across multiple network layers including service-layer, control-layer, and resource-layer constructs (Figs. 1 and 3 show the feature graph explicitly encoding network-layer relationships across multiple network layers); and
implementing changes to the one or more feature graphs based on [[any]] run-time updates from the pipeline including changes in configuration state, operational state, or feature-policy constraints (Sanchez-Navarro, Algorithm 3 and Section 4.2 Paragraph 1 – “5G networks are subject to continuous topology changes that require autonomous and dynamic management of the associated evolving network graph. The topology changes are motivated by different factors. On one hand, due to 5G mobility, users are continuously performing handovers from one DU to another. Besides, UE devices are continuously connecting/disconnecting from the network, e.g. user devices restart or simply lose their coverage. These kind of changes are minimal for the topology but sufficient to force a recalculation of the closeness centrality in the whole graph. On the other hand, the virtualization and on-demand provisioning of VNFs lead to continuous and dynamic changes in the network topology beyond the leaves nodes in the graph that requires recalculating our 5GCC indexes, as this quantification depends on the overall status of the graph.” – teaches implementing changes to the one or more feature graphs based on any run-time updates from the pipeline including changes in configuration state, operational state, or feature policy constraints (continuous and dynamic management of evolving network graph such as adding/removing nodes/edges or provisioning of virtual network functions, changes feature graph based on updates including changes in configuration state or operational state)).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the feature graphs of Sanchez-Navarro to the network Machine Learning model pipeline and functions of Barron in order to use feature graphs for network representations as input to machine learning pipeline. Doing so would make use of spatial network metrics to identify critical locations in the network topology, represented as evolving graphs with hardware and software resources, and use the metrics to make valuable decisions that optimize the allocation and management of VNFs and traffic (Sanchez-Navarro, Introduction).
Claims 10 and 19 incorporate substantively all the limitations of claim 1 in a method and apparatus, and are rejected on the same grounds as above. Sanchez-Navarro teaches the one or more processors of Claim 19 in Section 3.3 Paragraph 7.
Regarding claim 2, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the plurality of functions includes a sequence of pre-processing, inference, and decision functions configured to consume the feature (Barron, Section IV, Paragraph 1 – “The self-adaptive network management frame-work uses a closed autonomic control loop [22] referred to as a decision cycle originally designed by John Boyd, a military strategist, and includes the sub-processes of Observe, Orient, Decide and Act (OODA). In our system the sub-processes perform the following tasks”, Section IV, Subsection A “Observe”, Paragraph 1 – “Pertinent network events observed from systems/devices are identified and used as input to the knowledge model (i.e. pattern matching). Due to the sheer volume of events generated from systems/devices, events need to be aggregated/correlated. This function identifies specific event triggers from deployed DIF configurations and ensures event collection from network systems and devices that involves pre-processing events to aggregate the events generated and identify the most important and irregular events for submission to the knowledge model and machine learning algorithm while filtering routine and regular events.”, and in Section IV, Subsection B “Orient”, Paragraph 1 – “The self-adaptive framework correlates generated events with their associated DIF and calculates the effective network management impact caused by the event and specifically application impact. Semantic web rules are used to update and query the knowledge model to accurately reflect the current runtime state of the system.” – teaches wherein the plurality of functions include a sequence of pre-processing, inference, and decision functions configured to consume the features (in OODA Observe, Orient, Decide, and Act perform the functions of pre-processing, inference, and decision)).
Barron fails to explicitly teach the feature vectors derived from the feature graphs.
However, analogous to the field of the claimed invention, Sanchez-Navarro teaches:
the feature vectors derived from the feature graphs (Sanchez-Navarro, Section 4.1.2 Paragraph 5 & Algorithm 2 – “The algorithm returns the vector with the 5G Closeness Centrality values for each node that is worth calculating the 5G spatial metrics, i.e. those physical nodes that might allocate a VNF” – teaches the feature vectors derived from the feature graphs).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the feature vectors derived from the feature graph of Sanchez-Navarro to the plurality of functions of Barron in order to use the feature vectors in the pre-processing, inference, and decision functions. Doing so would make use of spatial network metrics to make valuable decisions that optimize the allocation and management of VNFs and traffic (Sanchez-Navarro, Introduction).
Claims 11 and 20 are similar to claim 2, hence similarly rejected.
Regarding claim 3, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 2, wherein the ML model is utilized in the pipeline embedded within a controller application (Barron, Fig. 1 and Section IV, Paragraph 1 – “The self-adaption mechanisms enable self-control (i.e. self-FCAPS [21]) of underlying network operations, functions and state minimising expensive, manual system administrator intervention. The self-adaptive network management framework uses a closed autonomic control loop [22] referred to as a decision cycle originally designed by John Boyd, a military strategist, and includes the sub-processes of Observe, Orient, Decide and Act (OODA). In our system the sub-processes perform the following tasks” – teaches wherein the ML model is utilized in the SDIDA (in Barron, OODA) pipeline embedded within a controller application (Fig. 1 shows the ML model utilized within the self-adaptive network management framework using a closed autonomic control loop with the subprocesses of OODA, and the network management system has self-control of underlying network operations)).
Claim 12 is similar to claim 3, hence similarly rejected.
Regarding claim 4, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the steps further include optimizing feature selection to include policies including limits for those features when mapped to a ML model for analytics on data related to network services (Sanchez-Navarro, Section 4.1.1 Paragraph 1 – “This optimization aims to create a subset of nodes P ={p ∈ N | p.type == PhysicalServer} which includes only those nodes that are subject to serve as possible candidates v to allocate VNFs, i.e. only physical nodes p in the graph G(N, E) will be considered for calculating their CC(v). It significantly reduces the number of nodes v required to calculate the 5GCC(v)” and in Section 4.4 Paragraph 1 – “When a service request reaches the cognitive management framework, the policy framework component must decide how to allocate optimally certain VNFs in the substrate physical nodes, depending on the service request necessities and the current status of the entire 5G management network, including physical and virtual appliances.” – teaches optimizing feature selection (optimizes selection of nodes to calculate vector) to include policies (policy framework component) including limits for those features when mapped to a ML model for analytics on data related to network services (optimizes allocation of VNFs in nodes based on policies and current status of network)), the optimizing further comprising applying dependency constraints and conditional relationships represented in the feature graphs to prune or augment the selected features (Sanchez-Navarro, Section 4.1.2 Paragraph 1 – “The second optimization is the pruning of the leaves nodes of the 5G graph, i.e. the user mobiles attached to DU nodes while maintaining the same resultant closeness centrality index values when computing CC(v) for each n ∈ P.” and in Section 4.4 Paragraph 2 – “We formulate the placement problem as an optimization procedure that starts firstly by checking hard constraints, either, placement constraint (e.g. nodes resource restrictions and current status in terms of computing, memory, etc.), as well as network constraints (e.g. resource’s connections bandwidth, latency constraints, or flows). This hard constraints phase ends up with a subset of potential candidates to allocate the VNF, as defined in Eq. 3, where Vp denotes the subset of V that are already allocated in a physical p machine and Rv is the set of resource requirements for a given node v. Thus, an r ∈ R, might refer, for instance, to resource storage requirement or a memory requirement of the VNF v.” – teaches the optimizing further comprising applying dependency constraints and conditional relationships represented in the feature graphs to prune the selected features (applies constraints and conditional relationships represented in feature graphs to prune the selected features or nodes)).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the policies including limits, dependency constraints, and conditional relationships for the selected features of Sanchez-Navarro to the Machine Learning model of Barron in order to optimize feature selection when those features are mapped to a ML model for analytics. Doing so would significantly reduce the number of features to include in further calculations (Sanchez-Navarro, Section 4.1.1), thus minimizing the time needed to compute the feature vector, and optimizes by only processing worthwhile features (Sanchez-Navarro, Section 4.1.2).
Claim 13 is similar to claim 4, hence similarly rejected.
Regarding claim 5, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the steps further include leveraging the information and data for network services and resources including business and network policies to create and use the one or more feature graphs for feature selection (Sanchez-Navarro, Section 4.1.1 Paragraph 1 – “This optimization aims to create a subset of nodes P ={p ∈ N | p.type == PhysicalServer} which includes only those nodes that are subject to serve as possible candidates v to allocate VNFs, i.e. only physical nodes p in the graph G(N, E) will be considered for calculating their CC(v). It significantly reduces the number of nodes v required to calculate the 5GCC(v)” and in Section 4.4 Paragraph 1 – “When a service request reaches the cognitive management framework, the policy framework component must decide how to allocate optimally certain VNFs in the substrate physical nodes, depending on the service request necessities and the current status of the entire 5G management network, including physical and virtual appliances.” – teaches wherein the steps further include leveraging the information and data for network services and resources including business and network policies to create and use the one or more feature graphs for feature selection (utilizes business and network policies to create and use the feature graph for feature selection, leveraging the service request and current status of the network)).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the use of business and network policies in feature selection of Sanchez-Navarro to the Machine Learning model pipelines of Barron in order to leverage the business and network policies during feature selection. Doing so would significantly reduce the number of features to include in further calculations (Sanchez-Navarro, Section 4.1.1), thus minimizing the time needed to compute the feature vector, and optimizes by only processing worthwhile features (Sanchez-Navarro, Section 4.1.2).
Claim 14 is similar to claim 5, hence similarly rejected.
Regarding claim 6, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the network includes a plurality of virtual or physical network function elements with links to form various types of network topologies (Sanchez-Navarro, Fig. 3 – shows the network including a plurality of virtual or physical network function elements with links to form various types of network topologies).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the network including a plurality of virtual or physical network function elements with links of Sanchez-Navarro to the Machine Learning model pipelines of Barron in order to model networks of various types of network topologies. Doing so would allow dynamically modeling the network as connected directed graph with nodes corresponding to network devices and edges modeling relationships between devices (Sanchez-Navarro, Section 4).
Claim 15 is similar to claim 6, hence similarly rejected.
Regarding claim 7, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the steps further include performing the utilizing with the information and data for a given service intent and its resources (Sanchez-Navarro, Section 4.1 Paragraph 2 – “In our case, the weight will be embodied by additional 5G network performance metrics calculated by our sensing components, such as latency or bandwidth, depending on the specific use case being addressed (e.g. cache allocation, load balancing, computation offloading, etc., as shown in the next section).” – teaches wherein the steps further include the utilizing with the information and data for a given service intent (use case) and its resources (network performance metrics calculated by sensing components)).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the utilization of information and data for a given service intent and its resources of Sanchez-Navarro to the Machine Learning model pipelines of Barron in order to utilize information for a given service intent. Doing so would allow higher network resource utilization efficiency, reduced latency, increased availability, optimal VNF allocation, migrations of VNFs, balancing the network traffic, and balancing the computational load among edge nodes (Sanchez-Navarro, Introduction).
Claim 16 is similar to claim 7, hence similarly rejected.
Regarding claim 8, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the one or more feature graphs include the relative weights to indicate the importance of the features in the ML model, the relative weights being assigned by the feature graph engine based on relationships encoded in the feature graph and persisted for use in generating the design-time feature vector (Sanchez-Navarro, Eq. (1) & (2) and Section 4.1 Paragraph 2 – “Our proposal extends and adapts the Closeness Centrality metric by Sabidussi (1966), defined as how close a node is to all the other nodes of the network. Nodes are more central as they are closer to the majority of the other nodes. Closeness Centrality (CC(v)) is mathematically defined in Eq. 1, where dG(v,n) is the function that calculates the distance of the shortest path from node v to node n in the Graph G… The distance is the sum of the weights of its edges, where the weight measures the strength of a link. In our case, the weight will be embodied by additional 5G network performance metrics calculated by our sensing components, such as latency or bandwidth, depending on the specific use case being addressed (e.g. cache allocation, load balancing, computation offloading, etc., as shown in the next section).” – teaches wherein the one or more feature graphs include the relative weights to indicate the importance of the features in the ML model, the relative weights assigned by the feature graph engine based on relationships encoded in the feature graph and persisted for use in generating the design-time feature vector (feature graph assigns relative weights to indicate strength of link or relationships between nodes, the weights persisted for use in generating the design-time feature vector 5GCC(v), as in Eq. 2)).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the feature graphs including relative weights of Sanchez-Navarro to the Machine Learning model pipelines of Barron in order to construct feature graphs including relative weights indicating the importance of features. Doing so would measure the strengths of links between nodes as weights and utilize the weight to embody metrics depending on use case (Sanchez-Navarro, Section 4.1)
Claim 17 is similar to claim 8, hence similarly rejected.
Regarding claim 9, the combination of Barron and Sanchez-Navarro teaches the non-transitory computer-readable medium of claim 1, wherein the steps further include one or more of receiving governance change updates by the plurality of functions (Barron, Section IV, Subsection C “Decide”, Paragraph 2 – “The function analyses changes in network service demand from specific users or user groups and harnesses a supervised machine learning algorithm to facilitate domain adaptation by developing a system of network service demand prediction and provisioning which allows the underlying network to resize and resource itself to serve predicted network service demand according to various parameters such as bandwidth, undetected bit error rate, delay and jitter.” and in Section IV, Subsection D “Act”, Paragraph 1 – “The framework considers QoS cube configurations deployed for provisioning currently active network services, and those requested for newly deployed applications. Should there be a mis-match and the application or is deemed important enough, a QoS cube (tailored to the requirements of the new application) can be created in the DIF and the resource allocation policy adjusted to the relative ”importances” of the QoS cubes. Thus, it learns when an application’s network requirements are not being met and where appropriate, takes remedial actions.” – teaches receiving governance change updates (analyses changes in network service demand from specific users of user groups and uses a ML algorithm to facilitate adaption and changes) by the plurality of functions (by the Decide and Act functions of OODA)).
Barron fails to explicitly teach receiving feature vector updates.
However, analogous to the field of the claimed invention, Sanchez-Navarro teaches:
receiving feature vector updates (Sanchez-Navarro, Section 4.3 Paragraph 2 – “Thus this optimization avoids computing the shortest path dG(ps,d) each time a UE node is added to the graph, reducing to the minimum the complexity to keep up-to-date the 5GCC metric in the entire 5G network topology. Thus, for these cases, which might represent more than 95% of changes in the 5G network, Algorithm 2 is given as input the set D empty, meaning the loop of lines number 6–8, which calculates distance function dG(ps,d) are not exe cuted. In addition, for this case, we have defined another 5G distance function 5GdG(v,r) similar to the one defined in Algorithm 1, but slightly changed, without code of line number 3, meaning the distance function dG(v,r) is obtained from another structure DistanceV R kept in memory-cache that holds the distance values for each pair physical server and DU. The rest of the function description is kept, where lines 4–5 in Algorithm 1 update the weights, and therefore the resultant 5GCC, according to the added/removed UE in the graph.” – teaches receiving feature vector updates (updates feature vector based on update to graph))
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate the feature vector updates of Sanchez-Navarro to the governance change updates by a plurality of functions of Barron. Doing so would provide continuous and dynamic management of the associated evolving network graph (Sanchez-Navarro, Section 4.2).
Claim 18 is similar to claim 9, hence similarly rejected.
Response to Arguments
Applicant’s arguments, see pp. 5-8, filed 4 December 2025, with respect to the rejection(s) of claim(s) 1-20 under Barron in view of Tizghadam have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made over Barron in view of Sanchez-Navarro et al. (NPL: Advanced spatial network metrics for cognitive management of 5G networks, published Aug. 2020, hereinafter “Sanchez-Navarro”). Barron teaches “receiving information and data…;” and “implementing feature selection on one or more network Machine Learning (ML) models…”. Sanchez-Navarro teaches “utilizing one or more feature graph engines (FGEs) that constructs one or more feature graphs…” and “implementing changes to the one or more feature graphs based on…”.
Applicant argues on pp. 6-7 that Barron does not disclose feature selection for ML models. Examiner respectfully disagrees, and points to Barron at Section IV, Subsection A “Observe”, Paragraph 1 – “Due to the sheer volume of events generated from systems/devices, events need to be aggregated/correlated. This function identifies specific event triggers from deployed DIF configurations and ensures event collection from network systems and devices that involves pre-processing events to aggregate the events generated and identify the most important and irregular events for submission to the knowledge model and machine learning algorithm while filtering routine and regular events.” – which teaches performing feature selection for machine learning models (pre-processing events to identify most important and irregular events for submission to machine learning algorithm, thus identifying and selecting the most important features to be input to the machine learning algorithm). The examiner notes that “feature” is a broad term that can be defined as any piece of input data that helps the model make predictions or decisions.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Shen et al. (NPL: Accurate Decentralized Application Identification via Encrypted Traffic Analysis Using Graph Neural Networks, published Jan. 2021) teaches utilizing graph neural networks to generate Traffic Interaction Graphs as information-rich representations of encrypted Decentralized Application flows. Teaches adding/removes edges/nodes of the graph representing the network. Teaches calculating a feature vector from the graph to be input into a multi-layer perceptron for classification.
Rusek et al. (NPL: Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN, published Oct. 2019) teaches modeling networks using a graph neural network to understand the complex relationships in the topology, routing and input traffic. Teaches wherein a computer network can be represented by a set of links and generates a routing scheme within the network by a set of paths.
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 LOUIS C NYE whose telephone number is 571-272-0636. The examiner can normally be reached Monday - Friday 9:00AM - 5:00PM.
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, MATT ELL can be reached at 571-270-3264. 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.
/LOUIS CHRISTOPHER NYE/Examiner, Art Unit 2141
/MATTHEW ELL/Supervisory Patent Examiner, Art Unit 2141