# US 12,009,983 B1 · Automatically configuring network elements in multi-vendor and multi-domain topologies > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,009,983 B1 - Document ID: US12009983B1 - Status: Granted - Kind: B1 - Application number: 18/065,324 - Filed: 2022-12-13 - Published: 2024-06-13 - Granted: 2024-06-11 - Earliest claimed: 2022-12-13 - Assignees: AT&T Intellectual Property I, L.P.; AT&T Communications Services India Private Limited; AT&T Global Network Services (UK) B.V. - Applicants: AT&T Intellectual Property I, L.P.; AT&T Communications Services India Private Limited; AT&T Global Network Services (UK) B.V. - Inventors: Mritunjay Pandey; Subhash Kapoor; Carl Thomas; Somnath Mallick; Lalena Aria; Donald Charles Jeffery; Saurav Paira - Web page: https://patentgrep.com/patent/US12009983B1 ## CPC - H04L 41/0895: Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements - H04L 41/0823: characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability (for optimising operational conditions of wireless networks H04W24/02) - H04L 41/12: Discovery or management of network topologies - H04L 41/122: of virtualised topologies, e.g. software-defined networks [SDN] or network function virtualisation [NFV] - H04L 41/145: {involving simulating, designing, planning or modelling of a network} - H04L 41/147: for predicting network behaviour - H04L 41/16: using machine learning or artificial intelligence - H04L 41/40: using virtualisation of network functions or resources, e.g. SDN or NFV entities - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract Devices, computer-readable media, and methods for automatically configuring network elements in multi-vendor and multi-domain topologies. In one example, a method includes determining a need of a communications network, where a topology of the communications network includes a plurality of network functions from at least two different vendors, predicting a subset of the plurality of network functions and respective configuration parameter values for network functions in the subset, that are expected to support the need of the communications network, and modifying the configurations of the network functions in the subset to reflect the respective configuration parameter values. ## Claims ### Claim 1 A method comprising: determining, by a processing system including at least one processor, a need of a communications network, where a topology of the communications network includes a plurality of network functions from at least two different vendors; predicting, by the processing system, a subset of the plurality of network functions and respective configuration parameter values for network functions in the subset, that are expected to support the need of the communications network; and modifying, by the processing system, the configurations of the network functions in the subset to reflect the respective configuration parameter values, wherein at least one of: the determining or the predicting is performed using a machine learning model that is trained to take as an input the need of the communications network and to generate as an output a prediction as to the subset of the plurality of network functions and the respective configuration parameter values for the network functions in the subset, wherein training of the machine learning model comprises: discovering, by the processing system, the topology of the communications network, including the plurality of network functions and a plurality of communication links between the plurality of network functions; collecting, by the processing system, data from each network function of the plurality of network functions, wherein the collecting comprises: selecting, by the processing system, a route in the communications network, a source network function of the plurality of network functions for the route, and a destination network function of the plurality of network functions for the route; testing, by the processing system, various combinations of a plurality of combinations of configuration parameter settings for the route; and repeating, by the processing system, the selecting and the testing for other routes in the communications network; and training, by the processing system, the machine learning model, using the data, to predict the need of the communications network and the respective configurations of the plurality of network functions that will support the need. ### Claim 2 (depends on 1) The method of claim 1, wherein the plurality of network functions includes a plurality of virtual network functions, a plurality of physical network functions, and a plurality of cloud-native network functions. ### Claim 3 (depends on 1) The method of claim 1, wherein the plurality of network functions includes network functions having different features and different parameters that are configurable. ### Claim 4 (depends on 1) The method of claim 1, wherein the plurality of network functions includes network functions that utilize different proprietary data models and schema definitions. ### Claim 5 (depends on 1) The method of claim 1, wherein the communications network further includes at least two domains. ### Claim 6 (depends on 1) The method of claim 1, wherein the machine learning model is a generalized net model. ### Claim 7 (depends on 1) The method of claim 1, wherein the discovering, the collecting, and the training are repeated in response to a change in the topology of the communications network. ### Claim 8 (depends on 1) The method of claim 1, wherein the plurality of combinations of configuration parameter settings for the route is predefined. ### Claim 9 (depends on 1) The method of claim 1, wherein the data is stored in a database for use in identifying at least one trend in usage of the communications network. ### Claim 10 (depends on 1) The method of claim 1, wherein the subset includes network functions that originate with at least two different vendors of the at least two different vendors. ### Claim 11 (depends on 1) The method of claim 1, wherein the processing system obtains an approval from a human administrator prior to the modifying. ### Claim 12 (depends on 11) The method of claim 11, wherein the processing system presents an estimated monetary cost of the modifying to the human administrator prior to obtaining the approval. ### Claim 13 (depends on 11) The method of claim 11, wherein the processing system presents an estimated impact of the modifying on operation of the communications network to the human administrator prior to obtaining the approval. ### Claim 14 (depends on 1) The method of claim 1, wherein the modifying is performed during a scheduled maintenance window for the communications network. ### Claim 15 A non-transitory computer-readable medium storing instructions which, when executed by a processing system including at least one processor, cause the processing system to perform operations, the operations comprising: determining a need of a communications network, where a topology of the communications network includes a plurality of network functions from at least two different vendors; predicting a subset of the plurality of network functions and respective configuration parameter values for network functions in the subset, that are expected to support the need of the communications network; and modifying the configurations of the network functions in the subset to reflect the respective configuration parameter values, wherein at least one of: the determining or the predicting is performed using a machine learning model that is trained to take as an input the need of the communications network and to generate as an output a prediction as to the subset of the plurality of network functions and the respective configuration parameter values for the network functions in the subset, wherein training of the machine learning model comprises: discovering the topology of the communications network, including the plurality of network functions and a plurality of communication links between the plurality of network functions; collecting data from each network function of the plurality of network functions, wherein the collecting comprises: selecting a route in the communications network, a source network function of the plurality of network functions for the route, and a destination network function of the plurality of network functions for the route; testing various combinations of a plurality of combinations of configuration parameter settings for the route; and repeating the selecting and the testing for other routes in the communications network; and training the machine learning model, using the data, to predict the need of the communications network and the respective configurations of the plurality of network functions that will support the need. ### Claim 16 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the plurality of combinations of configuration parameter settings for the route is predefined. ### Claim 17 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the data is stored in a database for use in identifying at least one trend in usage of the communications network. ### Claim 18 A system comprising: a processing system including at least one processor; and a computer-readable medium storing instructions which, when executed by the processing system, cause the processing system to perform operations, the operations comprising: determining a need of a communications network, where a topology of the communications network includes a plurality of network functions from at least two different vendors; predicting a subset of the plurality of network functions and respective configuration parameter values for network functions in the subset, that are expected to support the need of the communications network; and modifying the configurations of the network functions in the subset to reflect the respective configuration parameter values, wherein at least one of: the determining or the predicting is performed using a machine learning model that is trained to take as an input the need of the communications network and to generate as an output a prediction as to the subset of the plurality of network functions and the respective configuration parameter values for the network functions in the subset, wherein training of the machine learning model comprises: discovering the topology of the communications network, including the plurality of network functions and a plurality of communication links between the plurality of network functions; collecting data from each network function of the plurality of network functions, wherein the collecting comprises: selecting a route in the communications network, a source network function of the plurality of network functions for the route, and a destination network function of the plurality of network functions for the route; testing various combinations of a plurality of combinations of configuration parameter settings for the route; and repeating the selecting and the testing for other routes in the communications network; and training the machine learning model, using the data, to predict the need of the communications network and the respective configurations of the plurality of network functions that will support the need. ### Claim 19 (depends on 18) The system of claim 18, wherein the plurality of combinations of configuration parameter settings for the route is predefined. ### Claim 20 (depends on 18) The system of claim 18, wherein the data is stored in a database for use in identifying at least one trend in usage of the communications network.