# US 10,484,105 B2 · Method and apparatus for constructing wireless positioning feature library

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

- Number: US 10,484,105 B2
- Document ID: US10484105B2
- Status: Granted
- Kind: B2
- Application number: 16/287,738
- Filed: 2019-02-27
- Published: 2019-08-22
- Granted: 2019-11-19
- Earliest claimed: 2016-08-31
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Xiaofeng Hu; Jiehua Chen
- Web page: https://patentgrep.com/patent/US10484105B2

## CPC

- H04W 4/029: Location-based management or tracking services
- G01S 19/47: the supplementary measurement being an inertial measurement, e.g. tightly coupled inertial
- H04B 17/11: for calibration
- H04B 17/27: for locating or positioning the transmitter
- H04B 17/327: Received signal code power [RSCP]
- H04W 4/02: Services making use of location information
- H04W 64/00: Locating users or terminals {or network equipment} for network management purposes, e.g. mobility management
- H04W 64/003: {locating network equipment}
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)
- 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
- H04B: TRANSMISSION

## Abstract

A method for constructing a wireless positioning feature library includes: obtaining a wireless positioning training data set, where the wireless training data set includes a plurality of pieces of training data; performing clustering and merging on the plurality of pieces of training data one by one; and determining wireless positioning feature data based on at least one piece of sample data obtained after the merging.

## Claims

### Claim 1

A method for constructing a wireless positioning feature library, comprising:
obtaining, by an apparatus, a wireless positioning training data set, wherein the wireless positioning training data set comprises a plurality of pieces of training data;
performing, by the apparatus, clustering and merging on the plurality of pieces of training data one by one, to obtain at least one new piece of sample data; and
determining, by the apparatus, wireless positioning feature data based on the at least one new piece of sample data;
wherein performing clustering and merging on a piece of training data A of the plurality of pieces of training data comprises:
determining, from at least one piece of sample data of a sample data set, sample data that falls within a preset range of the piece of training data A;
determining, from the sample data that falls within the preset range of the piece of training data A, sample data B with highest similarity to the piece of training data A;
merging the piece of training data A and the sample data B to form new sample data C; and
replacing the sample data B with the new sample data C in the sample data set.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the training data and the sample data each comprise location data, at least one cell identifier, and a received signal strength indicator that is in a one-to-one correspondence with the at least one cell identifier; and
wherein the sample data further comprises an accumulation count of location data and an accumulation count of received signal strength indicators of each cell.

### Claim 3 (depends on 2)

The method according to claim 2, wherein determining the sample data that falls within the preset range of the piece of training data A is based on location data in the at least one piece of sample data and location data in the piece of training data A.

### Claim 4 (depends on 2)

The method according to claim 2, wherein after determining the sample data that falls within the preset range of the piece of training data A, the method further comprises:
determining to-be-merged sample data from the sample data that falls within the preset range of the piece of training data A, wherein a ratio of a quantity of same cell identifiers comprised in the to-be-merged sample data and the piece of training data A to a total quantity of cell identifiers comprised in the piece of training data A is greater than or equal to a first preset threshold.

### Claim 5 (depends on 4)

The method according to claim 4, wherein the sample data B with highest similarity to the piece of training data A
is determined from the to-be-merged sample data.

### Claim 6 (depends on 2)

The method according to claim 2, wherein determining similarity between sample data and training data comprises:
calculating a root mean square of the training data and the sample data according to
RMS
=
∑
i
=
1
m
⁢
(
sumRSSI
i
sumM
i
-
Rssi
i
)
2
m
,
wherein a smaller root mean square indicates higher similarity between the training data and the sample data;
wherein Rssii is a received signal strength indicator of an ith cell in the training data, sumRSSi is a received signal strength indicator of an ith cell in the sample data, sumMi is an accumulation count in the sample data, and m is a quantity of same cell identifiers comprised in the training data and the sample data.

### Claim 7 (depends on 2)

The method according to claim 2, wherein merging the piece of training data A and the sample data B to form the new sample data C comprises:
adding location data in the piece of training data A and location data in the sample data B together, to obtain location data in the new sample data C;
adding received signal strength indicators corresponding to a same cell identifier in the piece of training data A and the sample data B together, and keeping a received signal strength indicator corresponding to a different cell identifier in the piece of training data A and the sample data B unchanged, to obtain a received signal strength indicator of each cell of the new sample data C; and
increasing, by 1, an accumulation count of the received signal strength indicators corresponding to the same cell identifier comprised in the piece of training data A and the sample data B, keeping an accumulation count of a received signal strength indicator corresponding to a cell identifier that is comprised only in the sample data B unchanged, recording, as 1, an accumulation count of a received signal strength indicator corresponding to a cell identifier that is comprised only in the piece of training data A, and increasing, by 1, an accumulation count corresponding to the location data, to separately obtain an accumulation count of the received signal strength indicator of each cell and an accumulation count of the location data in the new sample data C.

### Claim 8 (depends on 2)

The method according to claim 2, wherein when the sample data set is empty, or when the sample data that falls within the preset range of the piece of training data A is empty, or when all cell identifiers comprised in the sample data that falls within the preset range of the piece of training data A are different from all cell identifiers in the piece of training data A, or when a ratio of a quantity of same cell identifiers comprised in the sample data that falls within the preset range of the piece of training data A and the piece of training data A to the quantity of cell identifiers comprised in the piece of training data A is less than the first preset threshold, the method further comprises:
determining the piece of training data A as new sample data D, wherein location data, a cell identifier, and a received signal strength indicator corresponding to the cell identifier in the new sample data D are all the same as those in the piece of training data A, and both an accumulation count of the location data and an accumulation count of a received signal strength indicator of each cell in the new sample data D are recorded as 1.

### Claim 9 (depends on 2)

The method according to claim 2, wherein the determining the wireless positioning feature data based on the at least one new piece of sample data comprises:
for each piece of sample data, dividing location data in the sample data by an accumulation count corresponding to the location data, to obtain location data in wireless positioning feature data corresponding to the sample data; and
for each piece of sample data, dividing a received signal strength indicator corresponding to each cell identifier in the sample data by an accumulation count of the received signal strength indicator corresponding to each cell identifier, to obtain a received signal strength indicator of each cell in the wireless positioning feature data corresponding to the sample data.

### Claim 10 (depends on 1)

The method according to claim 1, wherein merging the piece of training data A and the sample data B to form the new sample data C comprises:
determining whether the similarity between the piece of training data A and the sample data B is greater than or equal to a second preset threshold; and
in response to the similarity between the piece of training data A and the sample data B being greater than or equal to the second preset threshold, merging the piece of training data A and the sample data B to form the new sample data C.

### Claim 11

An apparatus for constructing a wireless positioning feature library, comprising:
a transceiver;
at least one processor, and
a non-transitory computer-readable storage medium having processor-executable instructions stored thereon;
wherein the processor-executable instructions, when executed by the at least one processor, facilitate:
obtaining, via the transceiver, a wireless positioning training data set, wherein the wireless positioning training data set comprises a plurality of pieces of training data;
performing clustering and merging on the plurality of pieces of training data one by one, to obtain at least one new piece of sample data; and
determining wireless positioning feature data based on the at least one new piece of sample data;
wherein performing clustering and merging on a piece of training data A of the plurality of pieces of training data comprises:
determining, from at least one piece of sample data of a sample data set, sample data that falls within a preset range of the piece of training data A;
determining, from the sample data that falls within the preset range of the piece of training data A, sample data B with highest similarity to the piece of training data A;
merging the piece of training data A and the sample data B to form new sample data C; and
replacing the sample data B with the new sample data C in the sample data set.

### Claim 12 (depends on 11)

The apparatus according to claim 11, wherein the training data and the sample data each comprise location data, at least one cell identifier, and a received signal strength indicator that is in a one-to-one correspondence with the at least one cell identifier; and
wherein the sample data further comprises an accumulation count of location data and an accumulation count of received signal strength indicators of each cell.

### Claim 13 (depends on 12)

The apparatus according to claim 12, wherein determining the sample data that falls within the preset range of the piece of training data A
is based on location data in the at least one piece of sample data and location data in the piece of training data A.

### Claim 14 (depends on 12)

The apparatus according to claim 12, wherein the processor-executable instructions, when executed, further facilitate, after determining the sample data that falls within the preset range of the piece of training data A:
determining to-be-merged sample data from the sample data that falls within the preset range of the piece of training data A, wherein a ratio of a quantity of same cell identifiers comprised in the to-be-merged sample data and the piece of training data A to a total quantity of cell identifiers comprised in the piece of training data A is greater than or equal to a first preset threshold.

### Claim 15 (depends on 14)

The apparatus according to claim 14, wherein the sample data B with highest similarity to the piece of training data A is determined
from the to-be-merged sample data.

### Claim 16 (depends on 12)

The apparatus according to claim 12, wherein determining similarity between sample data and training data comprises:
calculating a root mean square of the training data and the sample data according to
RMS
=
∑
i
=
1
m
⁢
(
sumRSSI
i
sumM
i
-
Rssi
i
)
2
m
,
wherein a smaller root mean square value indicates higher similarity between the training data and the sample data;
wherein Rssii is a received signal strength indicator of an ith cell in the training data, sumRSSIi is a received signal strength indicator of an ith cell in the sample data, sumMi is an accumulation count in the sample data, and m is a quantity of same cell identifiers comprised in the training data and the sample data.

### Claim 17 (depends on 12)

The apparatus according to claim 12, wherein merging the piece of training data A and the sample data B to form the new sample data C comprises:
adding location data in the piece of training data A and location data in the sample data B together, to obtain location data in the new sample data C;
correspondingly adding received signal strength indicators corresponding to a same cell identifier in the piece of training data A and the sample data B together, and keeping a received signal strength indicator corresponding to a different cell identifier in the piece of training data A and the sample data B unchanged, to obtain a received signal strength indicator of each cell of the new sample data C; and
increasing, by 1, an accumulation count of the received signal strength indicators corresponding to the same cell identifier comprised in the piece of training data A and the sample data B, keeping an accumulation count of a received signal strength indicator corresponding to a cell identifier that is comprised only in the sample data B unchanged, recording, as 1, an accumulation count of a received signal strength indicator corresponding to a cell identifier that is comprised only in the piece of training data A, and increasing, by 1, an accumulation count corresponding to the location data, to separately obtain an accumulation count of the received signal strength indicator of each cell and an accumulation count of the location data in the new sample data C.

### Claim 18 (depends on 12)

The apparatus according to claim 12, wherein the processor-executable instructions, when executed, further facilitate:
when the sample data set is empty, or when the sample data that falls within the preset range of the piece of training data A is empty, or when all cell identifiers comprised in the sample data that falls within the preset range of the piece of training data A are different from all cell identifiers in the piece of training data A, or when a ratio of a quantity of same cell identifiers comprised in the sample data that falls within the preset range of the piece of training data A and the piece of training data A to the quantity of cell identifiers comprised in the piece of training data A is less than the first preset threshold,
determining the piece of training data A as new sample data D, wherein location data, a cell identifier, and a received signal strength indicator corresponding to the cell identifier in the new sample data D are all the same as those in the piece of training data A, and both an accumulation count of the location data and an accumulation count of a received signal strength indicator of each cell in the new sample data D are recorded as 1.

### Claim 19 (depends on 11)

The apparatus according to claim 11, wherein merging the piece of training data A and the sample data B to form the new sample data C comprises:
determining whether the similarity between the piece of training data A and the sample data B is greater than or equal to a second preset threshold; and
in response to the similarity between the piece of training data A and the sample data B being greater than or equal to the second preset threshold, merging the piece of training data A and the sample data B to form the new sample data C.

### Claim 20

A non-transitory computer-readable medium having processor-executable instructions stored thereon, wherein the processor-executable instructions, when executed, facilitate:
obtaining a wireless positioning training data set, wherein the wireless positioning training data set comprises a plurality of pieces of training data;
performing clustering and merging on the plurality of pieces of training data one by one, to obtain at least one new piece of sample data; and
determining wireless positioning feature data based on the at least one new piece of sample data;
wherein performing clustering and merging on a piece of training data A in the plurality of pieces of training data comprises:
determining, from at least one piece of sample data of a sample data set, sample data that falls within a preset range of the piece of training data A;
determining, from the sample data that falls within the preset range of the piece of training data A, sample data B with highest similarity to the piece of training data A;
merging the piece of training data A and the sample data B to form new sample data C; and
replacing the sample data B with the new sample data C in the sample data set.