Enabling intent-based network management with generative ai and digital twins
Aspects of the subject disclosure may include, for example, an intent-based network (IBN) management system that effects changes in a communication network based on high-level intents. High-level intents are translated into operator-level intents by a large language model (LLM). Conflicts between operator-level intents are resolved, and the operator-level intents are mapped to intent functions. The intent functions are then mapped to policies that may effect changes in the network in accordance with the high-level intents. Other embodiments are disclosed.
1 . A device, comprising:
a processing system including a processor; and
a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:
receiving network operator-level intents from a large language model that translates high level intents into the network operator-level intents, wherein the network operator-level intents are specific to network operators and comprise a first operator-level intent associated with a first network operator having a first number of radio access network (RAN) nodes and a second operator-level intent associated with a second network operator having a second number of RAN nodes;
interpreting the network operator-level intents, extracting intent functions, and mapping the network operator-level intents into the intent functions, wherein the intent functions comprise a function to subscribe to measurements, a function to update cell configurations, and a function to apply control policies in the RAN;
performing a function call to the mapped intent functions;
performing two-level conflict resolutions comprising performing network operator-level intent conflict resolution on the network operator-level intents and performing application level intent conflict resolution, wherein the network operator-level intent conflict resolution is performed by identifying a network operator-level conflict configured to have overlapping requirements or mutually exclusive configuration between the first operator-level intent and the second operator-level intent, and wherein the application level intent conflict resolution is performed regarding an application level conflict that affect a same key performance indicator (KPI); and
applying the intent functions to modify RAN policies to effect RAN changes in accordance with the high level intents.
2 . The device of claim 1 , wherein the operations further comprise application-level conflict resolution of the intent functions.
3 . The device of claim 1 , wherein the mapping the network operator-level intents into the intent functions comprises accessing the intent functions via application programming interface (API) calls.
4 . The device of claim 1 , wherein the performing network operator-level intent conflict resolution is in accordance with requirements provided by a policy auditor.
5 . The device of claim 1 , wherein the intent functions comprise updating cell configurations.
6 . The device of claim 1 , wherein the applying the intent functions comprises applying the intent functions to a digital twin of a physical RAN.
7 . The device of claim 6 , wherein the operations further comprise applying the intent functions to the physical RAN.
8 . The device of claim 1 , wherein the high level intents specify an expected increase in network users.
9 . The device of claim 8 , wherein the network operator-level intents specify an increased bandwidth requirement to support the expected increase in network users.
10 . A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:
receiving network operator-level intents from a large language model that translates high level intents into the network operator-level intents, wherein the network operator-level intents are specific to network operators and comprise a first operator-level intent associated with a first network operator having a first number of radio access network (RAN) nodes and a second operator-level intent associated with a second network operator having a second number of RAN nodes;
interpreting the network operator-level intents, extracting intent functions, and mapping the network operator-level intents into the intent functions, wherein the intent functions comprise a function to subscribe to measurements, a function to update cell configurations, and a function to apply control policies in the RAN;
performing a function call to the mapped intent functions;
performing two-level conflict resolutions comprising performing network operator-level intent conflict resolution on the network operator-level intents and performing application level intent conflict resolution, wherein the network operator-level intent conflict resolution is performed by identifying a network operator-level conflict configured to have overlapping requirements or mutually exclusive configuration between the first operator-level intent and the second operator-level intent, and wherein the application level intent conflict resolution is performed regarding an application level conflict that affect a same key performance indicator (KPI); and
applying the intent functions to modify RAN policies to effect RAN changes in accordance with the high level intents.
11 . The non-transitory machine-readable medium of claim 10 , wherein the operations further comprise application-level conflict resolution of the intent functions.
12 . The non-transitory machine-readable medium of claim 10 , wherein the mapping the network operator-level intents into the intent functions comprises accessing the intent functions via application programming interface (API) calls.
13 . The non-transitory machine-readable medium of claim 10 , wherein the performing network operator-level intent conflict resolution is in accordance with requirements provided by a policy auditor.
14 . The non-transitory machine-readable medium of claim 10 , wherein the intent functions comprise updating cell configurations.
15 . The non-transitory machine-readable medium of claim 10 , wherein the applying the intent functions comprises applying the intent functions to a digital twin of a physical RAN.
16 . The non-transitory machine-readable medium of claim 15 , wherein the operations further comprise applying the intent functions to the physical RAN.
17 . A method, comprising:
receiving, by a processing system including a processor, network operator-level intents from a large language model that translates high level intents into the network operator-level intents, wherein the network operator-level intents are specific to network operators and comprise a first operator-level intent associated with a first network operator having a first number of radio access network (RAN) nodes and a second operator-level intent associated with a second network operator having a second number of RAN nodes;
interpreting, by the processing system, the network operator-level intents, extracting intent functions, and mapping, by the processing system, the network operator-level intents into the intent functions, wherein the intent functions comprise a function to subscribe to measurements, a function to update cell configurations, and a function to apply control policies in the RAN;
performing, by the processing system, a function call to the mapped intent functions;
performing, by the processing system, two-level conflict resolutions comprising performing network operator-level intent conflict resolution on the network operator-level intents and performing application level intent conflict resolution, wherein the network operator-level intent conflict resolution is performed by identifying a network operator-level conflict configured to have overlapping requirements or mutually exclusive configuration between the first operator-level intent and the second operator-level intent, and wherein the application level intent conflict resolution is performed regarding an application level conflict that affect a same key performance indicator (KPI); and
applying, by the processing system, the intent functions to modify RAN policies to effect RAN changes in accordance with the high level intents.
18 . The method of claim 17 , wherein the applying the intent functions comprises applying the intent functions to a digital twin of a physical RAN.
19 . The method of claim 18 , wherein the operations further comprise applying, by the processing system, the intent functions to the physical RAN.
20 . The method of claim 18 , wherein the operations further comprise iterating changes to the RAN policies in the digital twin before applying the changes to the physical RAN.