Protocol and signaling framework enabling machine learning models in wireless communication networks
Aspects of the subject disclosure may be directed to, for example, a method including determining a set of data collection parameters that are configurable to collect data indicative of network events occurring in real time or near real time in the wireless communication networks, receiving the collected data based on the set of data collection parameters from a group of network entities operating in the wireless communication networks, based on the received collected data, and generating training data for a machine learning model deployed in the wireless communication networks. Other embodiments are disclosed.
1 . A method, comprising:
formulating, by a processing system including a processor, a set of data collection parameters that are configurable to collect data corresponding to one or more network events, wherein the one or more network events occur in real time or near real time in wireless communication networks;
receiving, by the processing system, first data collected based on the set of data collection parameters from a group of network entities including a first network entity operating in the wireless communication networks;
receiving, by the processing system, second data collected based on the set of data collection parameters from the group of network entities including a second network entity operating in the wireless communication networks;
generating a training data based on the first collected data, the second collected data or both;
training, by the processing system, a machine learning model based on the generated training data to learn the one or more network events; and
broadcasting a message signaling life cycle management procedures of the machine learning model to at least one of the group of network entities, wherein the machine learning model operates in the at least one of the group of network entities.
2 . The method of claim 1 , wherein the formulating the set of data collection parameters comprises configuring the set of data collection parameters to specify a predetermined validity time duration.
3 . The method of claim 1 , wherein the formulating the set of data collection parameters comprises configuring the set of data collection parameters to maintain a predetermined time interval for periodically collected information until the predetermined time interval is overridden by an updated time interval.
4 . The method of claim 1 , further comprising receiving, by the processing system, an on-demand request for data collection from a third network entity, the on-demand request specifying a predetermined time duration and target measurement information.
5 . The method of claim 4 , wherein the formulating the set of data collection parameters comprises configuring the set of data collection parameters to specify the predetermined time duration specified in the on-demand request and the target measurement information.
6 . The method of claim 1 , wherein the formulating the set of data collection parameters comprises configuring the set of data collection parameters to specify a data type, a data format or both.
7 . The method of claim 1 , further comprising broadcasting, by the processing system, a message indicative of support capability of the machine learning model to the first network entity, wherein the first network entity serves as a user device.
8 . The method of claim 1 , further comprising broadcasting, by the processing system, a message indicative of validity of the machine learning model to the first network entity, wherein the first network entity serves as a user device.
9 . The method of claim 1 , further comprising broadcasting, by the processing system, periodically, on an on-demand basis, or both, a message indicative of information relevant to the machine learning model to the first network entity, wherein the first network entity serves as a user device.
10 . 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:
determining a set of data collection parameters that are configurable to generate relevant data corresponding to network events, wherein the network events occur in real time or near real time in wireless communication networks;
collecting data based on the set of data collection parameters from a group of network entities operating in the wireless communication networks;
generating a training data based on the collected data;
training a machine learning model based on the generated training data to learn the network events; and
broadcasting a message signaling life cycle management procedures of the machine learning model to at least one of the group of network entities, wherein the machine learning model operates in the at least one of the group of network entities.
11 . The device of claim 10 , wherein the broadcasting the message signaling the life cycle management procedures further comprises facilitating an activation or deactivation, a switching, a fallback of the machine learning model or a combination thereof, based on a predetermined trigger.
12 . The device of claim 10 , wherein the operations further comprise signaling, to at least one of the group of network entities, a set of machine learning model information including availability, update information, supported functionality, validity, or a combination thereof.
13 . The device of claim 12 , wherein the signaling further comprise signaling the set of machine learning model information periodically, in response to an on-demand request, or both.
14 . The device of claim 10 , wherein the operations further comprise performing an activation or a deactivation of the machine learning model based on a predetermined trigger.
15 . The device of claim 10 , wherein the operations further comprise:
receiving an on-demand request for data collection, the on-demand request specifying a predetermined time duration and target measurement; and
collecting another data based on the predetermined time duration and the target measurement specified in the on-demand request.
16 . 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:
determining a set of data collection parameters that are configurable to collect data indicative of network events occurring in real time or near real time in wireless communication networks;
receiving the collected data based on the set of data collection parameters from a group of network entities operating in the wireless communication networks;
based on the received collected data, generating training data for a machine learning model deployed in the wireless communication networks; and
broadcasting a message signaling life cycle management procedures of the machine learning model to at least one of the group of network entities, wherein the machine learning model is deployed in the at least one of the group of network entities.
17 . The non-transitory machine-readable medium of claim 16 , wherein the operations further comprise training the machine learning model based on the generated training data to learn the network events.
18 . The non-transitory machine-readable medium of claim 16 , wherein the broadcasting the message further comprises broadcasting the message periodically or on an on-demand basis.
19 . The non-transitory machine-readable medium of claim 16 , wherein the broadcasting the message signaling the life cycle management procedures further comprises facilitating an activation or deactivation, a switching, a fallback of the machine learning model or a combination thereof, based on a predetermined trigger.
20 . The non-transitory machine-readable medium of claim 16 , wherein the operations further comprise providing a condition-based life cycle management procedure to a user equipment by using a broadcast or a dedicated signaling.