Method and Apparatus for Obtaining Emission Probability, Method and Apparatus for Obtaining Transition Probability, and Sequence Positioning Method and Apparatus

Patent metadata

Field	Value
Publication number	US-2020187149-A1
Application number	US-202016791795-A
Country	US
Kind code	A1
Filing date	Feb 14, 2020
Priority date	Aug 15, 2017
Publication date	Jun 11, 2020
Grant date	—

Abstract

A method for obtaining an emission probability includes obtaining a plurality of measurement reports (MRs) of a terminal in a target region and an engineering parameter of at least one base station in the target region, obtaining, based on parameter information in each of the plurality of MRs and the engineering parameter of the at least one base station, a feature vector corresponding to each of the plurality of MRs, processing, using a regression model, location information in each of the plurality of MRs and the feature vector corresponding to each of the plurality of MRs, to obtain a single-point positioning model, calculating, based on the single-point positioning model, the location information in each of the plurality of MRs, and the feature vector corresponding to each of the plurality of MRs, an emission probability of the feature vector corresponding to each of the plurality of MRs.

First claim

1 . A method for obtaining an emission probability for sequence positioning, comprising:
obtaining a plurality of measurement reports (MRs) of a terminal in a target region and an engineering parameter of at least one base station in the target region, wherein the target region is a predetermined geographical region, wherein each of the MRs comprises location information and parameter information, wherein the location information indicates a location that is in the target region and that is of a terminal corresponding to the MRs comprising the location information, wherein the parameter information comprises an environment parameter, and wherein the environment parameter indicates an environment in which the terminal corresponding to the MRs comprising the environment parameter is located; obtaining a feature vector corresponding to each of the MRs based on the parameter information in each of the MRs and the engineering parameter of the at least one base station; processing the location information in each of the MRs and the feature vector corresponding to each of the MRs using a regression model to obtain a single-point positioning model; and calculating the emission probability of the feature vector corresponding to each of the MRs based on the single-point positioning model, the location information in each of the MRs, and the feature vector corresponding to each of the MRs, wherein the emission probability comprises an emission probability value, and wherein the emission probability value indicates a probability that the feature vector corresponds to a piece of location information.

2 . The method of claim 1 , wherein the parameter information in each of the MRs comprises at least one base station identifier (ID) corresponding to the at least one base station of a plurality of base stations, wherein the at least one base station ID indicates a base station to which the terminal is connected, and wherein the method further comprises:
matching the MRs with the engineering parameter of the at least one base station based on the at least one base station ID to obtain an associated engineering parameter of each of the MRs, wherein the associated engineering parameter of each of the MRs comprises the engineering parameter of the base station indicated by each of the base station IDs; and obtaining the feature vector corresponding to each of the MRs based on the associated engineering parameter and the parameter information of each of the MRs, wherein the feature vector comprises the associated engineering parameter and parameter information of one of the MRs.

3 . The method of claim 1 , wherein processing the location information in each of the MRs and the feature vector corresponding to each of the MRs to obtain the single-point positioning model comprises:
obtaining a plurality of training sets wherein each of the training sets corresponds to each of the MRs based on the location information in each of the MRs and the feature vector corresponding to each of the MRs, wherein any of the training sets comprises the feature vector and the location information that correspond to one of the MRs; and inputting the training set corresponding to each of the MRs to obtain the single-point positioning model into a machine learning model for training.

4 . The method of claim 1 , wherein calculating the emission probability of the feature vector corresponding to each of the MRs comprises:
inputting the location information in each of the MRs and the feature vector corresponding to each of the MRs into the single-point positioning model to obtain a mapping relationship, wherein the mapping relationship indicates a correspondence between the feature vector and the location information; and calculating the emission probability of the feature vector corresponding to each of the MRs based on the mapping relationship.

5 . The method of claim 1 , wherein the environment parameter comprises at least one of time period information, weather information, or event information.

6 . A method for obtaining a transition probability, comprising:
obtaining a plurality of pieces of track data of a terminal in a target region from a third-party platform, wherein the target region is a predetermined geographical region, wherein each of the pieces of the track data comprises a same environment parameter and at least two pieces of location information, wherein the environment parameter indicates an environment in which the terminal is located, wherein the at least two pieces of the location information indicate a location of the terminal that is in the target region wherein each of the at least two pieces of the location information corresponds to a time stamp; and calculating the transition probability based on the pieces of the track data, wherein the transition probability comprises a transition probability value, wherein the transition probability value indicates a probability that the terminal moves from one of the at least two pieces of the location information to another of the at least two pieces of the location information after a time interval T.

7 . The method of claim 6 , wherein calculating the transition probability based on the pieces of the track data comprises:
processing the pieces of the track data to obtain a combination sequence of each of the pieces of the track data, wherein the combination sequence comprises any two of the at least two pieces of the location information in one of the pieces of the track data and a time interval between the any two of the at least two pieces of the location information; and obtaining the transition probability corresponding to a first preset condition based on a first preset condition and the combination sequence of each of the pieces of the track data, wherein the first preset condition is any one of a plurality of preset conditions, wherein each of the preset conditions comprises a preset time interval and preset location information, wherein the preset time interval corresponds to the time interval T, and wherein the preset location information corresponds to one of the at least two pieces of the pieces of the location information.

8 . The method of claim 7 , wherein obtaining, the transition probability corresponding to the first preset condition comprises:
determining combination sequences that comprise the preset time interval and the preset location information in the first preset condition and that are in the combination sequences comprised in the pieces of the track data; collecting statistics about the combination sequences that comprise the preset time interval and the preset location information in the preset condition; and calculating the transition probability corresponding to the first preset condition.

9 . The method of claim 6 , wherein before calculating the transition probability, the method further comprised removing defective track data from the pieces of the track data, wherein the defective track data is the track data in which one of the at least two pieces of the location information deviates from a road in the target region by a distance greater than a first threshold or is the track data in which a distance between two pieces of adjacent location information is greater than a second threshold.

10 . The method of claim 6 , wherein before calculating the transition probability based on the pieces of the track data, the method further comprises:
determining sparse track data in the pieces of the track data, wherein the sparse track data is the track data in which a distance between any two pieces of adjacent location information in the at least two pieces of the location information comprised in the track data is greater than a third threshold; and inserting

Assignees

• Huawei Tech Co Ltd

Inventors

• Zhu Fangzhou
• Yuan Mingxuan
• Zeng Jia
• Rao Weixiong

Classifications

• H04W64/00 Primary

  Locating users or terminals {or network equipment} for network management purposes, e.g. mobility management · CPC title

• H04W4/029

  Location-based management or tracking services · CPC title

• H04W4/023

  using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds · CPC title

• H04W24/08 Primary

  Testing, {supervising or monitoring} using real traffic · CPC title

• H04W64/003 Primary

  locating network equipment · CPC title