Predicting small cell capacity and coverage to facilitate offloading of macrocell capacity

Abstract

Predicting small cell capacity and coverage to facilitate offloading of macrocell capacity is presented herein. A system selects a group of candidate locations for placement of respective small cells to facilitate offloading, via the respective small cells, of traffic from respective macrocells corresponding to the candidate locations—the respective small cells including first transmission powers that are less than second transmission powers of the respective macrocells. Further, for each candidate location of the group of candidate locations, the system determines an estimated amount of traffic capacity of a small cell of the respective small cells that has been presumed to have been placed at the candidate location, and determines estimated signal strengths of respective signals that have been predicted to have been received from the small cell at respective portions of a grid of a defined signal coverage area corresponding to the candidate location.

What is claimed is: 
     
         1 . A system, comprising:
 a processor; and   a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations by the processor, comprising:
 in response to a machine learning model being selected from a group of machine learning models comprising a decision tree based learning model, a linear regression based learning model, and a Bayesian based learning model, selecting, using the machine learning model, a group of candidate locations for placement of respective small cells, wherein the respective small cells comprise first transmission powers that are less than second transmission powers of respective macrocells corresponding to respective candidate locations of the group of candidate locations; and 
 for the respective candidate locations of the group of candidate locations, determining, via the machine learning model, respective estimated amounts of traffic capacity of a small cell of the respective small cells determined to have been placed at the respective candidate locations, wherein a number of hyperparameters for the machine learning model has been determined based on a defined accuracy with respect to the respective estimated amounts of traffic capacity. 
   
     
     
         2 . The system of  claim 1 , wherein the operations further comprise:
 performing a binary search with respect to determining a number of nodes of a decision tree corresponding to the decision tree machine learning model.   
     
     
         3 . The system of  claim 1 , wherein the operations further comprise:
 updating the machine learning model using signal strength measurements that have been obtained from user equipment that have been communicatively coupled to the respective macrocells.   
     
     
         4 . The system of  claim 1 , wherein selecting the group of candidate locations comprises:
 selecting a candidate location of the group of candidate locations based on a signal quality of a macrocell of the respective macrocells corresponding to the candidate location.   
     
     
         5 . The system of  claim 1 , wherein determining the respective estimated amounts of traffic capacity of the small cell comprises:
 determining a predicted amount of megabytes to be transmitted via a downlink channel of the small cell over a defined period of time.   
     
     
         6 . The system of  claim 1 , wherein determining the respective estimated amounts of traffic capacity of the small cell comprises:
 determining a group of macrocell characteristics of a macrocell of the respective macrocells corresponding to a candidate location of the group of candidate locations;   determining a group of small cell characteristics of the small cell;   determining a group of location characteristics corresponding to the candidate location;   determining a traffic distribution of user equipment corresponding to the candidate location; and   based on the group of macrocell characteristics, the group of small cell characteristics, the group of location characteristics, and the traffic distribution of user equipment, determining the respective estimated amounts of traffic capacity of the small cell.   
     
     
         7 . The system of  claim 6 , wherein the group of macrocell characteristics comprises:
 at least one of a reference signal receive power corresponding to the macrocell, a reference signal received quality corresponding to the macrocell, or an amount of megabytes that have been transmitted via a downlink channel of the macrocell over a defined period of time.   
     
     
         8 . The system of  claim 6 , wherein the group of small cell characteristics comprises:
 at least one of an installation height of the small cell, a transmission power of the small cell, an antenna model of the small cell, or a period of deployment of the small cell.   
     
     
         9 . The system of  claim 6 , wherein the group of location characteristics comprises:
 at least one of a line of sight percentage of the defined signal coverage area that has been determined to be within a line of site of the small cell, an indoor location percentage of the defined signal coverage area that has been determined to be located within a building or structure, or a pedestrian traffic percentage of pedestrian traffic that has been determined to correspond to the defined signal coverage area.   
     
     
         10 . The system of  claim 6 , wherein the traffic distribution of user equipment comprises a number of the user equipment that have been determined to have been located in the defined signal coverage area over a defined period of time. 
     
     
         11 . The system of  claim 1 , wherein the operations further comprise:
 determining a group of small characteristics of the small cell;   determining a group of location features corresponding to the small cell and the respective portions of the defined signal coverage area; and   based on the group of small characteristics and the group of location features, determining estimated signal strengths of respective signals that have been predicted to have been received from the small cell at respective portions of a grid of a defined signal coverage area corresponding to the candidate location.   
     
     
         12 . The system of  claim 11 , wherein the group of small cell characteristics comprises:
 at least one of an installation height of the small cell, a transmission power of the small cell, an antenna tilt of the small cell, a width of transmission of a beam from the small cell, or a frequency band of the transmission of the beam.   
     
     
         13 . The system of  claim 11 , wherein the group of location features comprises:
 at least one of a distance between the small cell and a portion of the respective portions of the defined signal coverage area, a difference between a bearing angle of a predicted transmission of a beam that has been predicted to have been transmitted by the small cell from the candidate location and a reception angle of a predicted reception of the beam at the portion of the defined signal coverage area, or a path characteristic of a transmission path between the small cell and the portion of the defined signal coverage area.   
     
     
         14 . The system of  claim 1 , wherein selecting the group of candidate locations comprises:
 selecting the candidate location based on an amount of traffic communicated by the macrocell over a defined period of time.   
     
     
         15 . The system of  claim 1 , wherein selecting the group of candidate locations comprises:
 selecting the candidate location based on a number of user equipment that have been determined to be communicatively coupled to the macrocell over a defined period of time.   
     
     
         16 . A method, comprising:
 selecting, by a system comprising a processor, a machine learning model from a group of machine learning models comprising a decision tree based learning model, a linear regression based learning model, and a Bayesian based learning model;   based on the machine learning model, determining, by the system, a group of candidate locations for placement of respective small cells comprising first respective transmission powers that are less than second respective transmission powers of respective macrocells;   determining, by the system using the machine learning model based on a number of hyperparameters of the machine learning model, an estimated amount of traffic capacity of a small cell of the respective small cells that has been determined to have been placed at a candidate location of the group of candidate locations; and   based on a defined accuracy with respect to the estimated amount of traffic capacity, determining, by the system, the number of hyperparameters for the machine learning model.   
     
     
         17 . The method of  claim 16 , further comprising:
 performing, by the system, a binary search with respect to determining a number of nodes of a decision tree corresponding to the decision tree based machine learning model.   
     
     
         18 . The method of  claim 16 , further comprising:
 updating, by the system, the machine learning model using signal strength measurements that have been obtained from user equipment that have been communicatively coupled to the respective macrocells.   
     
     
         19 . A non-transitory machine-readable medium, comprising executable instructions that, when executed by network equipment comprising a processor, facilitate performance of operations, comprising:
 in response to selecting a machine learning model from a group of machine learning models comprising a decision tree based learning model, a linear regression based learning model, and a Bayesian based learning model, determining, via the machine learning model, candidate locations for placement of respective small cells; and   for respective candidate locations of the candidate locations, determining, via the group of machine learning models based on a number of hyperparameters of the machine learning models, respective estimated amounts of traffic capacity of a small cell of the respective small cells that has been determined to have been placed at the respective candidate locations, wherein the number of hyperparameters has been determined based on a defined accuracy with respect to the respective estimated amounts of traffic capacity of the small cell.   
     
     
         20 . The non-transitory machine-readable medium of  claim 19 , wherein determining the respective estimated amounts of traffic capacity of the small cell comprises:
 determining a predicted amount of megabytes to be transmitted via a downlink channel of the small cell over a defined period of time.