# US 12,004,114 B2 · Geolocation of wireless network users

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,004,114 B2
- Document ID: US12004114B2
- Status: Granted
- Kind: B2
- Application number: 17/346,179
- Filed: 2021-06-11
- Published: 2022-12-15
- Granted: 2024-06-04
- Earliest claimed: 2021-06-11
- Assignees: AT&T Intellectual Propety I, L.P.
- Applicants: AT&T Intellectual Property I, L.P.
- Inventors: Abhijeet Bhorkar; Baofeng Jiang; Peter Wong
- Web page: https://patentgrep.com/patent/US12004114B2

## CPC

- G01S 5/0294: {Trajectory determination or predictive filtering, e.g. target tracking or Kalman filtering}
- H04W 64/00: Locating users or terminals {or network equipment} for network management purposes, e.g. mobility management
- G01S 5/0278: {involving statistical or probabilistic considerations (G01S5/0252, G01S5/0294 take precedence)}
- G06F 16/211: {Schema design and management}
- G06F 16/29: Geographical information databases
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- H04W 16/22: Traffic simulation tools or models
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A method includes selecting a first machine learning model from a plurality of machine learning models that are trained for use in performing geolocation, wherein the first machine learning model is selected to perform geolocation within a first cell of a plurality of cells of a wireless network, acquiring event data from a plurality of wireless devices within the first cell, grouping the event data into a plurality of records, wherein each record of the plurality of records contains event data that indicates a common wireless device of the plurality of wireless devices, a common cell of the plurality of cells, and a common timestamp, and generating a predicted location of a first wireless device of the plurality of wireless devices, using the first machine learning model, wherein the first machine learning model outputs the predicted location in response to an input of a record of the plurality of records.

## Claims

### Claim 1

A method comprising:
selecting, by a processing system including at least one processor, a first machine learning model from among a plurality of machine learning models that are trained for use in performing geolocation, wherein the first machine learning model is selected to perform geolocation within a first cell of a plurality of cells of a wireless network;
acquiring, by the processing system, event data from a plurality of wireless devices operating within the first cell;
grouping, by the processing system, the event data into a plurality of records, wherein each record of the plurality of records contains event data that indicates a common wireless device of the plurality of wireless devices, a common cell of the plurality of cells, and a common timestamp; and
generating, by the processing system, a predicted location of a first wireless device of the plurality of wireless devices, using the first machine learning model, wherein the first machine learning model outputs the predicted location in response to an input of a record of the plurality of records.

### Claim 2 (depends on 1)

The method of claim 1, wherein the first machine learning model comprises one selected from a group of: a recurrent neural network with Long Short-term Memory (LSTM) units, a probabilistic model, and a lookup table.

### Claim 3 (depends on 1)

The method of claim 1, wherein the selecting is based at least in part on a set of computational resources available to the processing system.

### Claim 4 (depends on 1)

The method of claim 1, wherein the selecting is based at least in part on a desired throughput for the first cell, and wherein the desired throughput indicates a desired number of predicted locations to be generated per unit of time.

### Claim 5 (depends on 1)

The method of claim 1, wherein the event data comprises signal strength data for the plurality of wireless devices and timing events for the plurality of wireless devices.

### Claim 6 (depends on 5)

The method of claim 5, wherein the signal strength data comprises at least one selected from a group of: reference signals received power, serving cell, and neighbor cell.

### Claim 7 (depends on 5)

The method of claim 5, wherein the timing events comprise timing advance.

### Claim 8 (depends on 1)

The method of claim 1, wherein the grouping comprises:
selecting, by the processing system, a first event from the event data to function as a pivot event;
identifying, by the processing system, a second event which corresponds to the pivot event;
deriving, by the processing system, a key from a wireless device of the plurality of wireless devices associated with the second event, a cell of the plurality of cells associated with the second event, and a timestamp associated with the second event; and
using, by the processing system, the key to correlate the wireless device of the plurality of wireless devices associated with the second event, the cell of the plurality of cells associated with the second event, and the timestamp associated with the second event with other events from the event data.

### Claim 9 (depends on 8)

The method of claim 8, wherein the pivot event comprises an event which contains a timing advance feature.

### Claim 10 (depends on 8)

The method of claim 8, wherein the grouping is performed using a chain correlator comprising a plurality of data mappers and a plurality of data correlators.

### Claim 11 (depends on 1)

The method of claim 1, wherein the generating comprises refining the predicted location with global positioning system data from the first wireless device to generate an updated predicted location.

### Claim 12 (depends on 11)

The method of claim 11, wherein the updated predicted location comprises a weighted combination of the predicted location and the global positioning system data.

### Claim 13 (depends on 1)

The method of claim 1, wherein the generating comprises refining the predicted location using a filtering technique.

### Claim 14 (depends on 13)

The method of claim 13, wherein the filtering technique comprises at least one selected from a group of: an extended Kalman filtering, a particle filtering, and a smoothing function.

### Claim 15 (depends on 1)

The method of claim 1, further comprising:
augmenting, by the processing system, the plurality of records with static data from an external data source.

### Claim 16 (depends on 15)

The method of claim 15, wherein the augmenting comprises replacing a value in a record of the plurality of records with a value from the static data.

### Claim 17 (depends on 1)

The method of claim 1, further comprising:
creating, by the processing system, a time-series model of the plurality of records.

### Claim 18 (depends on 17)

The method of claim 17, further comprising:
batching, by the processing system, the time-series model with another time-series model, prior to the generating the predicted location.

### Claim 19

A non-transitory computer-readable storage device storing a plurality of instructions which, when executed by a processing system including at least one processor, cause the processing system to perform operations, the operations comprising:
selecting a first machine learning model from among a plurality of machine learning models that are trained for use in performing geolocation, wherein the first machine learning model is selected to perform geolocation within a first cell of a plurality of cells of a wireless network;
acquiring event data from a plurality of wireless devices operating within the first cell;
grouping the event data into a plurality of records, wherein each record of the plurality of records contains event data that indicates a common wireless device of the plurality of wireless devices, a common cell of the plurality of cells, and a common timestamp; and
generating a predicted location of a first wireless device of the plurality of wireless devices, using the first machine learning model, wherein the first machine learning model outputs the predicted location in response to an input of the a record of the plurality of records.

### Claim 20

An apparatus comprising:
a processing system including at least one processor; and
a non-transitory computer-readable storage device storing a plurality of instructions which, when executed by the processing system cause the processing system to perform operations, the operations comprising:
selecting a first machine learning model from among a plurality of machine learning models that are trained for use in performing geolocation, wherein the first machine learning model is selected to perform geolocation within a first cell of a plurality of cells of a wireless network;
acquiring event data from a plurality of wireless devices operating within the first cell;
grouping the event data into a plurality of records, wherein each record of the plurality of records contains event data that indicates a common wireless device of the plurality of wireless devices, a common cell of the plurality of cells, and a common timestamp; and
generating a predicted location of a first wireless device of the plurality of wireless devices, using the first machine learning model, wherein the first machine learning model outputs the predicted location in response to an input of a record of the plurality of records.