Real-time ML-supported radio propagation computation for RAN planning
AssigneeAT&T Intellectual Property I, L.P.
InventorsYaron Kanza, Arun Jotshi, Velin Kounev, Krystian Czapiga
Life of this patent
Estimating the term…
Abstract
Aspects of the subject disclosure may include, for example, network deployment or radio-propagation computation based on a combination of photon mapping and machine learning including supporting near-real-time computation of the radio transmissions for different layouts of antennas and allowing examination of a large variety of antenna locations and layouts, changing configuration details, e.g., tilting antennas or optimally selecting the sector that each antenna covers, and so on. Other embodiments are disclosed.
Claims20
A device, comprising:
- a processing system including a processor;
- and
- a memory that stores executable instructions that, when executed by the processing system, facilitate a performance of operations, the operations comprising:
- obtaining a first plurality of images at a first resolution via a use of at least one electromagnetic source, wherein the first plurality of images is associated with an environment comprising a plurality of objects;
- obtaining a second plurality of images at a second resolution via a use of the at least one electromagnetic source, wherein the second plurality of images is associated with the environment, and wherein the second resolution is different from the first resolution;
- applying respective pairs of the first plurality of images and the second plurality of images to a neural network to generate a trained model;
- applying the trained model to a photon mapping at a third resolution to generate an illumination map at a fourth resolution;
- obtaining a second model of the environment, wherein the second model comprises an indication of radio propagation conditions within the environment;
- and
- processing the illumination map and the second model to generate a presentation of a network coverage associated with the environment.
A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate a performance of operations, the operations comprising:
- applying respective pairs of a first plurality of images associated with an environment and a second plurality of images associated with the environment to a network to generate a first model, wherein each of the first plurality of images is at a first resolution and each of the second plurality of images is at a second resolution that is different from the first resolution;
- applying the first model to a photon mapping at a third resolution to generate an illumination map at a fourth resolution;
- obtaining a second model of the environment, wherein the second model comprises an indication of radio propagation conditions within the environment;
- and
- processing the illumination map and the second model to generate a presentation of a network coverage associated with the environment.
A method, comprising:
- applying, by a processing system including a processor, at least a first image at a first resolution and at least a second image at a second resolution to a network to generate a trained model;
- applying, by the processing system, the trained model to a photon mapping at a third resolution to generate an illumination map at a fourth resolution, the fourth resolution being greater than the third resolution;
- and
- processing, by the processing system, the illumination map based on an identification of radio propagation conditions within an environment to generate an identification of a network coverage associated with the environment.
Dependencies are inferred from the claim text.