公开(公告)号：US09747610B2

公开(公告)日：2017-08-29

申请号：US14087199

申请日：2013-11-22

Applicant: AT&T Intellectual Property I, LP

Inventor： Rong Duan ,  Guang-Qin Ma ,  Olivia Hong

IPC: G06Q10/00 ,  G06Q30/00 ,  G06Q30/02

CPC classification number: G06Q30/0202 ,  G06Q10/04

Abstract: A system that incorporates teachings of the present disclosure may include, for example, determining a first linear model and a first log-linear model according to first explanatory variables and a first plurality of occupancy estimates, determining a second linear model and a second log-linear model according to second explanatory variables and a second plurality of occupancy estimates, evaluating, according to a location of a building, the first linear model, the first log-linear model, the second linear model, and the second log-linear model according to first values of the first explanatory variables and second values of the second explanatory variables, and combining results from the evaluating to generate a mixed occupancy estimate. Other embodiments are disclosed.

公开(公告)号：US09544785B2

公开(公告)日：2017-01-10

申请号：US14511562

申请日：2014-10-10

Applicant: AT&T Intellectual Property I, LP

Inventor： Rong Duan ,  Siu-Tong Au ,  Heeyoung Kim ,  Guang-Qin Ma

IPC: G01R31/08 ,  H04M3/00 ,  H04W24/02 ,  H04L12/24 ,  H04L12/26 ,  H04W24/08 ,  H04W16/24

CPC classification number: H04W24/02 ,  H04L41/0816 ,  H04L43/067 ,  H04L43/16 ,  H04W16/24 ,  H04W24/08

Abstract: A method collects data and subjects it to statistical analysis to detect localized events, which assists in network provisioning. Illustratively, the data employed is hourly network traffic counts that are collected at cell sites. By taking advantage of the additive property of a Poisson process, the method integrates spatial neighbor information by aggregating temporal data in various areas, and iteratively estimating the event location and the radius of event impact by examining the posterior probability based on the aggregated data.

Abstract translation: 一种方法收集数据并对其进行统计分析，以检测本地化事件，这有助于网络配置。 说明性地，所采用的数据是在小区站点收集的小时网络流量计数。 通过利用泊松过程的附加属性，该方法通过聚合各个区域的时间数据来集成空间邻居信息，并且通过基于聚合数据检查后验概率来迭代地估计事件位置和事件影响的半径。

公开(公告)号：US20150149245A1

公开(公告)日：2015-05-28

申请号：US14087199

申请日：2013-11-22

Applicant: AT&T Intellectual Property I, LP

Inventor： Rong Duan ,  Guang-Qin Ma ,  Olivia Hong

IPC: G06Q30/02

CPC classification number: G06Q30/0202 ,  G06Q10/04

Abstract: A system that incorporates teachings of the present disclosure may include, for example, determining a first linear model and a first log-linear model according to first explanatory variables and a first plurality of occupancy estimates, determining a second linear model and a second log-linear model according to second explanatory variables and a second plurality of occupancy estimates, evaluating, according to a location of a building, the first linear model, the first log-linear model, the second linear model, and the second log-linear model according to first values of the first explanatory variables and second values of the second explanatory variables, and combining results from the evaluating to generate a mixed occupancy estimate. Other embodiments are disclosed.

Abstract translation: 结合本公开的教导的系统可以包括例如根据第一解释变量和第一多个占用估计来确定第一线性模型和第一对数线性模型，确定第二线性模型和第二对数线性模型， 根据第二解释变量的线性模型和第二多个占用估计，根据建筑物的位置，根据第一线性模型，第一对数线性模型，第二线性模型和第二对数线性模型来评估 对第一解释变量的第一值和第二解释变量的第二值，并将来自评估的结果组合以产生混合占用估计。 公开了其他实施例。

公开(公告)号：US20150024768A1

公开(公告)日：2015-01-22

申请号：US14511562

申请日：2014-10-10

Applicant: AT&T Intellectual Property I, LP

Inventor： Rong Duan ,  Siu-Tong Au ,  Heeyoung Kim ,  Guang-Qin Ma

IPC: H04W24/02 ,  H04W24/08 ,  H04W16/24

CPC classification number: H04W24/02 ,  H04L41/0816 ,  H04L43/067 ,  H04L43/16 ,  H04W16/24 ,  H04W24/08

Abstract: A method that collects data and subjects it to statistical analysis to detect localized events, which assists in network provisioning. illustratively, the data employed is hourly network traffic count that is collected at cell sites. By taking the advantage of additive property of Poisson process, the method integrates spatial neighbor information by aggregating temporal data in various areas, and iteratively estimating the event location and the radius of event impact by examining the posterior probability base on the aggregated data.

Abstract translation: 收集数据并对其进行统计分析以检测本地化事件的方法，这有助于网络配置。 例如，采用的数据是在小区站点收集的小时网络流量计数。 通过利用泊松过程的附加属性，该方法通过聚合各个区域的时间数据来集成空间邻居信息，并且通过基于聚合数据检查后验概率来迭代地估计事件位置和事件影响的半径。