Granted Patent B2

US 12,328,598 · App. 17/797,412 · Granted Jun 10, 2025

Inventors: Taesuh Park (Santa Clara, CA); Jianfu Zhang (Santa Clara, CA); Joji Philip (Milpitas, CA)

Assignee: Telefonaktiebolaget LM Ericsson (publ)

H04W16/20H04B17/27H04W64/003

Abstract

Claims (30)

1. A method to estimate a count of indoor radio transmitters for a building, wherein the indoor radio transmitters are to be planned to build an indoor cellular network within the building, the method comprising:

obtaining locational information of the building in a database based on a query;

extracting features external to the building based on the locational information of the building, wherein the features external to the building capture characteristics about the building that are observable from outside of the building; and

estimating the count of the indoor radio transmitters for the building using the extracted features and a pre-trained model that has been trained using historical building dataset and corresponding numbers of indoor radio transmitters.

2. The method of claim 1 , wherein the count of the indoor radio transmitters includes a number of radio transmitters to be implemented per floor for the building.

3. The method of claim 1 , wherein the query indicates a geo-coordinate including latitude and longitude of the building.

4. The method of claim 1 , wherein the query indicates at least one of a property name and a street address, and wherein the at least one of the property name and street address is to be converted to a geo-coordinate in a digital map.

5. The method of claim 1 , wherein the features external to the building include an area size of a footprint polygon of the building in a digital map.

6. The method of claim 1 , wherein the features external to the building include metadata showing a building supply-demand relationship in an area in which the building is located.

7. The method of claim 6 , wherein the metadata showing the building supply-demand relationship includes a number of neighbor buildings that is calculated from a list of footprint polygons of neighbor buildings within a pre-determined range from the building.

8. The method of claim 6 , wherein the metadata showing the building supply-demand relationship includes a population density of a town or neighborhood in which the building is located.

9. The method of claim 6 , wherein the metadata showing the building supply-demand relationship includes average or median of rental prices of buildings in a neighborhood of the building.

10. The method of claim 6 , wherein the metadata showing the building supply-demand relationship includes a list of rental prices of nearby buildings within a predetermined range from the building.

11. The method of claim 1 , wherein the features external to the building include information about floor density of the building.

12. The method of claim 11 , wherein the information about floor density of the building includes a building coverage ratio of the building.

13. The method of claim 1 , wherein the features external to the building includes a list of footprint polygons of neighbor buildings within a pre-determined range from the building.

14. The method of claim 13 , wherein the list of footprint polygons is further used to calculate gaps between the building and its neighbor buildings.

15. The method of claim 13 , wherein the list of footprint polygons is further used to calculate angles of blockage to the building by surrounding buildings.

16. The method of claim 1 , wherein the pre-trained model uses one or more of a Random Forest Regressor model, an adaptive boosting (AdaBoost) model, and an extreme Gradient Boost (XGBoost) model.

17. An electronic device to estimate a count of indoor radio transmitters for a building, wherein the indoor radio transmitters are to be planned to build an indoor cellular network within the building, the electronic device comprising:

a processor and non-transitory machine-readable storage medium storing instructions, which when executed by the processor, are capable of causing the electronic device to perform:

obtaining locational information of the building in a database based on a query;

extracting features external to the building based on the locational information of the building, wherein the features external to the building capture characteristics about the building that are observable from outside of the building; and

estimating the count of the indoor radio transmitters for the building using the extracted features and a pre-trained model that has been trained using historical building dataset and corresponding numbers of indoor radio transmitters.

18. The electronic device of claim 17 , wherein the features external to the building include metadata showing a building supply-demand relationship in an area in which the building is located.

19. A non-transitory machine-readable storage medium that stores instructions, which when executed by a processor, are capable of causing an electronic device to estimate a count of indoor radio transmitters for a building, wherein the indoor radio transmitters are to be planned to build an indoor cellular network within the building, the electronic device is to be caused to perform:

obtaining locational information of the building in a database based on a query;

extracting features external to the building based on the locational information of the building, wherein the features external to the building capture characteristics about the building that are observable from outside of the building; and

estimating the count of the indoor radio transmitters for the building using the extracted features and a pre-trained model that has been trained using historical building dataset and corresponding numbers of indoor radio transmitters.

20. The non-transitory machine-readable storage medium of claim 19 , wherein the pre-trained model uses one or more of a random forest regressor model, an adaptive boosting (AdaBoost) model, and an extreme Gradient Boost (XGBoost) model.

Assignments (1)

Continuity (2)

References Cited (42)