Back to results
US 11,388,719 B2Granted
Adaptive beam sweeping for 5G or other next generation network
AssigneeAT&T Intellectual Property I, L.P.
InventorsDeva-Datta Sharma, Shraboni Jana, Mostafa Tofighbakhsh, Bogdan Ungureanu, Rittwik Jana
Life of this patent
Estimating the term…
Abstract
Given real-time user equipment (UE) measurements from an open radio access network (O-RAN) infrastructure, a radio access network intelligent controller (RIC) can compute a UE distribution context space. The O-RAN can comprise gNode-Bs, centralized units, and distributed units. The UE distribution context space computations can be performed by a UE context correlator module of the RIC. The UE context correlator module can also utilize a pre-defined UE context model, which contains definitions and values for various UE context attributes to generate adaptive beam-forming patterns.
Claims20
A method, comprising:
- based on attribute data representative of attributes of a first number of user equipment, modeling, by network equipment comprising a processor, a distribution of the first number of user equipment at a first time, resulting in model, wherein the attribute data comprises a time associated with utilization of the first number of user equipment, a location associated with the first number of user equipment, and a social event associated with the first number of user equipment;
- correlating, by the network equipment, the attribute data to defined parameters, wherein correlating the attribute data to the defined parameters comprises partitioning the attribute data based on defined context attributes comprising a time context attribute, a location context attribute, and a social event context attribute;
- generating a machine-learning model based on the correlating of the attribute data to the defined parameters;
- based on the machine-learning model and the model, estimating, by the network equipment, a distribution model, resulting in an estimated distribution model, to be used to determine a beam sweep process to be applied by the processor, at a second time later than the first time, to a second number of user equipment, less than the first number of user equipment, that have previously utilized the beam sweep process;
- and
- based on the estimated distribution model, determining, by the network equipment, a beam sweep periodicity to be applied to the beam sweep process.
3 more elements
A system, comprising:
- a processor;
- and
- a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising:
- based on attribute data representative of attributes of a first number of user equipment at a first time, modeling a distribution of the first number of user equipment at the first time, resulting in a distribution model, wherein the attribute data comprises a time of utilization of the first number of user equipment, a geographic area comprising locations of the first number of user equipment, and a social event associated with the utilization of the first number of user equipment;
- correlating the attribute data to defined parameters, wherein correlating the attribute data to the defined parameters comprises partitioning the attribute data based on defined context attributes comprising a time context attribute, a geographic area context attribute, and a social event context attribute;
- generating a machine-learned distribution model based on the correlating of the attribute data to the defined parameters;
- based on the distribution model and the machine-learned distribution model, estimating a distribution model for a second time that is later than the first time, resulting in an estimated distribution model to be applied to a second number of user equipment, wherein the second number of user equipment is less than the first number of user equipment;
- and
- based on the estimated distribution model, determining a beam sweeping control parameter, wherein the beam sweeping control parameter comprises a scanning periodicity value to be applied, by the processor, to a beam sweep;
- and
- in response to the determining, applying the beam sweeping control parameter to the second number of user equipment at the second time.
8 more elements
A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
- based on correlating mobile device data of a first number of mobile devices at a first time to a group of defined parameters, generating a distribution model currently applicable to the first number of mobile devices, wherein the distribution model is generated based on a time associated with utilization of the first number of mobile devices, a representative location of the first number of mobile devices, and a social event associated with the first number of mobile devices, and wherein correlating the mobile device data to the group of defined parameters comprises partitioning attribute data of the first number of mobile devices based on a time context attribute, a location context attribute, and a social event context attribute;
- generating a machine-learned distribution model based on the correlating of the mobile device data to the group of defined parameters;
- based on the distribution model and the machine-learned distribution model, generating an estimated distribution model to be applied to a second number of mobile devices less than the first number of mobile devices;
- using the estimated distribution model, obtaining a beam sweeping control parameter, wherein the beam sweeping control parameter comprises a beam sweep periodicity to be applied to a beam sweeping process;
- and
- in response to obtaining the beam sweeping control parameter, applying the beam sweeping control parameter to the second number of mobile devices at a second time after the first time and according to the beam sweep periodicity.
3 more elements
Dependencies are inferred from the claim text.