U.S. patent application number 15/721596 was filed with the patent office on 2018-12-06 for compressing radio maps.
This patent application is currently assigned to Apple Inc.. The applicant listed for this patent is Apple Inc.. Invention is credited to Anders M. Holtsberg, David Benjamin Millman, Jasvinder Singh, Darin Tay.
Application Number | 20180348334 15/721596 |
Document ID | / |
Family ID | 64459552 |
Filed Date | 2018-12-06 |
United States Patent
Application |
20180348334 |
Kind Code |
A1 |
Millman; David Benjamin ; et
al. |
December 6, 2018 |
COMPRESSING RADIO MAPS
Abstract
Embodiments are disclosed for compressing radio maps of
fingerprint-based positioning systems. In an embodiment, a method
comprises: receiving access point (AP) data from a plurality of
mobile devices operating in a geographic region, the AP data
including signal strength measurements of AP signals received at a
plurality of reference locations in the geographic region;
filtering the AP data to remove outlier AP data; fitting a surface
to the AP data; projecting AP data at surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the AP data at the surface control points; encoding
the boundary; encoding the AP data at the surface control points
included within the boundary; generating compressed radio maps from
the encoded AP data; and responsive to a request from a mobile
device operating in the geographic region, sending a data packet
including the compressed radio maps to the mobile device.
Inventors: |
Millman; David Benjamin;
(Mountain View, CA) ; Singh; Jasvinder; (Santa
Clara, CA) ; Holtsberg; Anders M.; (Los Gatos,
CA) ; Tay; Darin; (Sunnyvale, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Apple Inc. |
Cupertino |
CA |
US |
|
|
Assignee: |
Apple Inc.
Cupertino
CA
|
Family ID: |
64459552 |
Appl. No.: |
15/721596 |
Filed: |
September 29, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62514759 |
Jun 2, 2017 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04W 4/026 20130101;
H04W 24/08 20130101; H04W 64/003 20130101; H04W 48/20 20130101;
G01S 5/0252 20130101 |
International
Class: |
G01S 5/02 20060101
G01S005/02; H04W 4/02 20060101 H04W004/02 |
Claims
1. A method comprising: receiving, by a computing device, access
point (AP) data from a plurality of mobile devices operating in a
geographic region, the AP data including signal strength
measurements of AP signals received at a plurality of reference
locations in the geographic region; filtering the AP data to remove
outlier AP data; fitting a surface to the AP data; projecting AP
data at surface control points onto a two-dimensional image grid;
determining a boundary surrounding locations of the AP data at the
surface control points in the image grid; encoding the boundary;
encoding the AP data at the surface control points included within
the boundary; generating compressed radio maps from the encoded AP
data; and responsive to a request from a mobile device operating in
the geographic region, sending a data packet including the
compressed radio maps to the mobile device.
2. The method of claim 1, wherein filtering AP data further
comprises: identifying non-servable APs in the AP data; and
excluding the non-servable APs from further processing.
3. The method of claim 1, wherein filtering AP data further
comprises: clustering the AP data; identifying outlier AP signal
strength measurements based on the clustering; and excluding
outlier AP signal strength measurements from further
processing.
4. The method of claim 1, wherein fitting the surface to the AP
data further comprises: using bilinear interpolation to fit the
surface to the AP data.
5. The method of claim 1, wherein prior to fitting the surface to
the AP data, the AP data is quantized.
6. The method of claim 1, wherein determining the boundary further
comprises: applying contour detection to the projected signal
strength measurements at the surface control points; and encoding
the boundary using an encoding process.
7. The method of claim 1, wherein the geographic region is divided
into a grid of contiguous cells having a first size and the
plurality of reference locations are in a plurality of the
contiguous cells.
8. The method of claim 7, further comprising: adjusting the first
cell size to a second cell size based on the AP data.
9. The method of claim 8, wherein adjusting the first cell size to
a second cell size further comprises: (a) fitting the surface to
the AP data corresponding to a set of contiguous cells; (b) for the
set of contiguous cells, calculating a joint likelihood of signal
strength observations falling between the cells; (c) determining
that the likelihood is less than a threshold; (d) dividing the set
of cells in half; and repeating steps (a) through (d) until the
likelihood is greater than or equal to the threshold.
10. The method of claim 1, wherein the AP data includes AP signal
strength measurements and uncertainty values corresponding to the
AP signal strength measurements, and the method further comprises:
fitting surfaces to the signal strength measurements and the
uncertainty values; projecting signal strength measurements and
uncertainty values at the surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the signal strength measurements and corresponding
uncertainty values at the surface control points in the image grid;
encoding the boundary; encoding the signal strength measurements
and corresponding uncertainty values at the surface control points
that are included within the boundary; generating compressed radio
maps from the encoded signal strength measurements and
corresponding uncertainty values and the encoded boundary.
11. The method of claim 10, wherein the encoding scheme is at least
one of predictive differential encoding, run length encoding or
Rice encoding.
12. A system comprising: one or more processors; memory coupled to
the one or more processors and configured to store instructions,
which, when executed by the one or more processors, cause the one
or more processors to perform operations comprising: receiving
access point (AP) data from a plurality of mobile devices operating
in a geographic region, the AP data including signal strength
measurements of AP signals received at a plurality of reference
locations in the geographic region; filtering the AP data to remove
outlier AP data; fitting a surface to the AP data; projecting AP
data at surface control points onto a two-dimensional image grid;
determining a boundary surrounding locations of the AP data at the
surface control points in the image grid; encoding the boundary;
encoding the AP data at the surface control points included within
the boundary; generating compressed radio maps from the encoded AP
data; and responsive to a request from a mobile device operating in
the geographic region, sending a data packet including the
compressed radio maps to the mobile device.
13. The system of claim 12, wherein filtering AP data further
comprises: identifying non-servable APs in the AP data; and
excluding the non-servable APs from further processing.
14. The system of claim 12, wherein filtering AP data further
comprises: clustering the AP data; identifying outlier AP signal
strength measurements based on the clustering; and excluding
outlier AP signal strength measurements from further
processing.
15. The system of claim 12, wherein fitting the surface to the AP
data further comprises: using bilinear interpolation to fit the
surface to the AP data.
16. The system of claim 12, wherein prior to fitting the surface to
the AP data, the AP data is quantized.
17. The system of claim 12, wherein determining the boundary
further comprises: applying contour detection to the projected
signal strength measurements at the surface control points; and
encoding the boundary using an encoding process.
18. The system of claim 12, wherein the geographic region is
divided into a grid of contiguous cells having a first size and the
plurality of reference locations are in a plurality of the
contiguous cells.
19. The system of claim 18, wherein the operations further
comprise: adjusting the first cell size to a second cell size based
on the AP data.
20. The system of claim 19, wherein adjusting the first cell size
to a second cell size further comprises: (a) fitting the surface to
the AP data corresponding to a set of contiguous cells; (b) for the
set of contiguous cells, calculating a joint likelihood of signal
strength observations falling between the cells; (c) determining
that the likelihood is less than a threshold; (d) dividing the set
of cells in half; and repeating steps (a) through (d) until the
likelihood is greater than or equal to the threshold.
21. The system of claim 12, wherein the AP data includes AP signal
strength measurements and uncertainty values corresponding to the
AP signal strength measurements, and the operations further
comprise: fitting surfaces to the signal strength measurements and
the uncertainty values; projecting signal strength measurements and
uncertainty values at the surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the signal strength measurements and corresponding
uncertainty values at the surface control points in the image grid;
encoding the boundary; encoding the signal strength measurements
and corresponding uncertainty values at the surface control points
that are included within the boundary; generating compressed radio
maps from the encoded signal strength measurements and
corresponding uncertainty values and the encoded boundary.
22. The system of claim 21, wherein the encoding scheme is at least
one of predictive differential encoding, run length encoding or
Rice encoding.
23. A method comprising: receiving, by a computing device, access
point (AP) data from a plurality of mobile devices operating in a
geographic region, the AP data including signal strength
measurements of AP signals received at a plurality of reference
locations in the geographic region; fitting a surface to the AP
data; projecting AP data at surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the AP data at the surface control points in the image
grid; encoding the AP data at the surface control points included
within the boundary; generating compressed radio maps from the
encoded AP data; and responsive to a request from a mobile device
operating in the geographic region, sending a data packet including
the compressed radio maps to the mobile device.
24. The method of claim 23, further comprising: identifying
non-servable APs in the AP data and excluding the non-servable APs
from further processing.
25. The method of claim 23, wherein filtering AP data further
comprises: clustering the AP data; identifying outlier AP signal
strength measurements based on the clustering; and excluding
outlier AP signal strength measurements from further
processing.
26. The method of claim 23, wherein fitting the surface to the AP
data further comprises: using bilinear interpolation to fit the
surface to the AP data.
27. The method of claim 23, wherein prior to fitting the surface to
the AP data, the AP data is quantized.
28. The method of claim 23, wherein the AP data includes AP signal
strength measurements and uncertainty values corresponding to the
AP signal strength measurements, and the method further comprises:
fitting surfaces to the signal strength measurements and the
uncertainty values; projecting signal strength measurements and
uncertainty values at the surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the signal strength measurements and corresponding
uncertainty values at the surface control points in the image grid;
encoding the boundary; encoding the signal strength measurements
and corresponding uncertainty values at the surface control points
that are included within the boundary; generating compressed radio
maps from the encoded signal strength measurements and
corresponding uncertainty values and the encoded boundary.
29. A system comprising: one or more processors; memory coupled to
the one or more processors and configured to store instructions
that when executed by the one or more processors, cause the one or
more processors to perform operations comprising: receiving access
point (AP) data from a plurality of mobile devices operating in a
geographic region, the AP data including signal strength
measurements of AP signals received at a plurality of reference
locations in the geographic region; fitting a surface to the AP
data; projecting AP data at surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the AP data at the surface control points in the image
grid; encoding the AP data at the surface control points included
within the boundary; generating compressed radio maps from the
encoded AP data; and responsive to a request from a mobile device
operating in the geographic region, sending a data packet including
the compressed radio maps to the mobile device.
30. The system of claim 29, the operations further comprising:
identifying non-servable APs in the AP data; and excluding the
non-servable APs from further processing.
31. The system of claim 29, the operations further comprising:
clustering the AP data; identifying outlier AP signal strength
measurements based on the clustering; and excluding outlier AP
signal strength measurements from further processing.
32. The system of claim 29, wherein fitting the surface to the AP
data further comprises: using bilinear interpolation to fit the
surface to the AP data.
33. The system of claim 29, wherein prior to fitting the surface to
the AP data, the AP data is quantized.
34. The system of claim 29, wherein the AP data includes AP signal
strength measurements and uncertainty values corresponding to the
AP signal strength measurements, and the operations further
comprise: fitting surfaces to the signal strength measurements and
the uncertainty values; projecting signal strength measurements and
uncertainty values at the surface control points onto a
two-dimensional image grid; determining a boundary surrounding
locations of the signal strength measurements and corresponding
uncertainty values at the surface control points in the image grid;
encoding the boundary; encoding the signal strength measurements
and corresponding uncertainty values at the surface control points
that are included within the boundary; generating compressed radio
maps from the encoded signal strength measurements and
corresponding uncertainty values and the encoded boundary.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims priority to U.S. Provisional Patent
Application No. 62/514,759, filed Jun. 2, 2017, the entire contents
of which are incorporated herein by reference.
TECHNICAL FIELD
[0002] This disclosure relates generally to fingerprint-based
positioning systems.
BACKGROUND
[0003] Outdoor positioning systems typically cluster harvested
access point (AP) signal measurements from a large number of mobile
devices. Because outdoor positioning systems rely on satellite
signal visibility, the clustering approach does not work indoors or
in outdoor spaces where satellite signals are blocked or where the
quality of satellite signals is poor due to multipath. One solution
for improving outdoor localization in spaces with poor signal
reception is to use indoor positioning techniques to estimate
location. For example, an outdoor space, such as an "urban canyon,"
is divided into a grid. The grid can be irregularly shaped and
include cells that have no signal data or only sparse signal data.
In this case, the grid includes a large number of radio maps, where
the radio maps include "fingerprints" collected at reference
locations within the outdoor space, and each fingerprint includes
an identification and received signal strength of the APs
observable at the reference location. Due to the high cost of
cellular data service and the limited storage capacity of the
typical mobile device, it is desirable to compress the radio maps
before serving the radio maps to mobile devices for use in location
estimation.
SUMMARY
[0004] Embodiments are disclosed for compressing radio maps of
fingerprint-based positioning systems.
[0005] In an embodiment, a method comprises: receiving, by a
computing device, AP data from a plurality of mobile devices
operating in a geographic region, the AP data including signal
strength measurements of AP signals received at a plurality of
reference locations in the geographic region; filtering the AP data
to remove outlier AP data; fitting a surface to the AP data;
projecting AP data at surface control points onto a two-dimensional
image grid; determining a boundary surrounding locations of the AP
data at the surface control points in the image grid; encoding the
boundary; encoding the AP data at the surface control points
included within the boundary; generating compressed radio maps from
the encoded AP data; and responsive to a request from a mobile
device operating in the geographic region, sending a data packet
including the compressed radio maps to the mobile device.
[0006] In an embodiment, a system comprises: one or more
processors; memory coupled to the one or more processors and
configured to store instructions, which, when executed by the one
or more processors, causes the one or more processors to perform
operations comprising: receiving AP data from a plurality of mobile
devices operating in a geographic region, the AP data including
signal strength measurements of AP signals received at a plurality
of reference locations in the geographic region; filtering the AP
data to remove outlier AP data; fitting a surface to the AP data;
projecting AP data at surface control points onto a two-dimensional
image grid; determining a boundary surrounding locations of the AP
data at the surface control points in the image grid; encoding the
boundary; encoding the AP data at the surface control points
included within the boundary; generating compressed radio maps from
the encoded AP data; and responsive to a request from a mobile
device operating in the geographic region, sending a data packet
including the compressed radio maps to the mobile device.
[0007] Particular embodiments disclosed herein provide one or more
of the following advantages. The disclosed embodiments allow
compression of radio maps delivered in tiles to mobile devices for
use in location estimation. The compressed radio maps reduce the
cost of cellular data service for users and also reduce the amount
of data stored on mobile devices which often have limited storage
capacity. Additionally, the multiple versions of tiles with
different resolutions can be generated to mobile devices based on
memory and processing constraints of the mobile device and a
desired location estimation accuracy.
[0008] The details of one or more implementations of the subject
matter are set forth in the accompanying drawings and the
description below. Other features, aspects and advantages of the
subject matter will become apparent from the description, the
drawings and the claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] FIG. 1 illustrates an example fingerprint-based positioning
system (FPS), according to an embodiment.
[0010] FIG. 2 is a flow diagram of a server-side FPS processing
pipeline, according to an embodiment.
[0011] FIG. 3 is a flow diagram of a client-side FPS processing
pipeline, according to an embodiment.
[0012] FIGS. 4A-4C illustrate surface fitting to AP data, according
to an embodiment.
[0013] FIGS. 5A-5C illustrate contour detection and blob boundary
encoding, according to an embodiment.
[0014] FIG. 6 is a flow diagram of a process of generating
compressed AP radio maps, according to an embodiment.
[0015] FIG. 7 is a flow diagram of a process of surface fitting and
encoding, according to an embodiment.
[0016] FIG. 8 illustrates an example server architecture,
implementing the service-side FPS features and operations described
in reference to FIGS. 1-7.
[0017] FIG. 9 illustrates an example device architecture of a
mobile device implementing client-side FPS features and operations
described in reference to FIGS. 1-7.
DETAILED DESCRIPTION
Example Fingerprint-Based Positioning System
[0018] FIG. 1 illustrates an example FPS system 100, according to
an embodiment.
[0019] System 100 includes FPS server computers 102, mobile device
103, network 104, access points (APs) 105 and radio map database
106. RF signals transmitted by APs 105 are measured by mobile
devices 103 operating in various geographic regions. An RF signal
measurement is typically in the form of a received signal strength
indicator (RSSI), which is the metric used by the embodiments
described herein. It should be noted, however, that other
measurements of RF signal strength can be used with the disclosed
embodiments.
[0020] FPS 100 harvests AP data packets from a large number of
mobile devices 103 to build FP database 106. The AP data packets
include AP data for a particular reference harvest or survey
location in the geographic region, also referred to herein as
"reference location." An AP can be any wireless network AP,
including a WiFi router or a cell tower transmitter. The AP data
typically includes position data for the reference location (e.g.,
latitude, longitude, altitude, position uncertainty) provided by a
survey map, global navigation satellite system (GNSS) receiver,
WiFi position system or cell tower trilateration. The AP data also
includes RSSI values and media access control (MAC) addresses for
APs that are observed (scanned) at each reference location. If
available, the AP data also includes GNSS velocity and/or
pedestrian dead reckoning (PDR) data generated from equations of
motion on mobile devices 103, such as position, velocity (speed)
and heading. The PDR data can be generated from inertial sensor
data provided by inertial sensors on the mobile device (e.g.,
accelerometer, gyro, magnetometer). Some AP data may also include
barometric pressure data (e.g., provided by a pressure sensor on
the mobile device) that can be used to indicate the altitude of the
reference location.
[0021] FPS servers 102 calculate probability distributions of the
RSSI values (e.g., Rayleigh, Rician, Gaussian distributions, etc.)
for each AP observed at each reference location. The radio maps are
stored in radio map database 106. In response to a request from a
mobile device, FPS server computers 102 download a data packet or
"tile" to the mobile device that includes compressed radio maps as
described in further detail below in reference to FIG. 2
[0022] FIG. 2 is a flow diagram of a server-side FPS processing
pipeline 200, according to an embodiment. Pipeline 200 includes AP
filter 202, AP signal filter 203 and radio map generator 204. AP
data 201 is collected from a large number of mobile devices and
initially stored in a database separate from the FPS production
system.
[0023] AP filter 202 determines whether AP locations in AP data 201
are servable or not-servable to mobile devices. AP locations that
are servable are sent to mobile devices for use in location
estimation. AP locations that are not servable are not sent to
mobile devices for use in location estimation. An AP location is
not servable if, for example, the AP location has moved over a
specified time period, or a portion of outlier AP data is too high
even if the estimated position of an AP has not moved. In an
embodiment, AP data 201 can be clustered and APs that have moved
between clusters are moving APs. Some examples of suitable
clustering algorithms include but are not limited to: DBSCAN,
k-means and hierarchical cluster analysis (HCA). A technique for
identifying moving APs is described in co-pending U.S. Patent
Application No. 62/509,562, filed on May 22, 2017, which
provisional patent application is incorporated by reference herein
in its entirety.
[0024] In an embodiment, an AP may be excluded from further
processing based on memory cost, sighting probabilities and/or
coverage. For example, each AP can be ranked based on a score that
is a function of memory cost, cumulative sighting and probability
over all cells. The APs can then be included in, or excluded from,
further processing based on their scores. In an embodiment, a
greedy algorithm can be used to determine scores.
[0025] AP signal filter 203 receives the filtered AP data from AP
filter 202 and clusters the RSSI values in the AP data to identify
outlier RSSI values. For example, RSSI values that are less than a
minimum threshold and greater than a maximum threshold may be
considered outlier RSSI values and excluded from further
processing. Some examples of suitable clustering algorithms include
but are not limited to: DBSCAN, k-means and HCA.
[0026] Radio map generator 204 receives the AP data output by AP
signal filter 203 and performs surface fitting to generate RSSI and
uncertainty maps, as described in reference to FIGS. 4 and 5. In an
embodiment, radio map generator 204 also generates tiles 205
including radio maps 206 having different resolutions based on
tradeoffs between mobile device memory and processing constraints
and location estimation accuracy. To achieve different resolutions,
some of tiles 205 may include more radio maps 206 than other tiles
205, or have different cell sizes. For example, a tile created for
a dense urban environment may have more radio maps than a tile for
a rural environment. Or, in another example, an open parking lot
adjacent to a mall may need a lower cell resolution than the cell
resolution used in the mall.
[0027] FIG. 3 is a flow diagram of a client-side FPS pipeline 300,
according to an embodiment. Tiles 301 for a geographic region,
which were generated using pipeline 200, are pre-fetched or
downloaded on-the-fly to a mobile device operating at a location
(x, y) in a geographic region. In an embodiment, selected ones of
tiles 301 that match the memory or processing constraints of the
mobile device and desired location accuracy for the location (x, y)
are used for localization.
[0028] When an AP is observed at location (x, y), RSSI lookup
module 302 decodes and decompresses radio maps 206 that are
included in tiles 301 and stores them in local cache memory. An
RSSI prediction at location (x, y) is then computed using an
interpolation function f (x, y) using the RSSIs at the surface
control points. In an embodiment, the function f (x, y) is bilinear
and the RSSI values at the four surface control points (Q11, Q12,
Q21, Q22) that are nearest to location (x, y) are used in the
bilinear interpolation function f (x, y) as shown in Equations [1]
and [2].
f ( x , y ) .apprxeq. b 11 f ( Q 11 ) + b 12 f ( Q 12 ) + b 21 f (
Q 21 ) + b 22 f ( Q 22 ) , [ 1 ] [ b 11 b 12 b 21 b 22 ] = ( [ 1 x
1 y 1 x 1 y 1 1 x 1 y 2 x 1 y 2 1 x 2 y 1 x 2 y 1 1 x 2 y 2 x 2 y 2
] - 1 ) T [ 1 x y xy ] . [ 2 ] ##EQU00001##
where (x.sub.1, y.sub.1), (x.sub.1, y.sub.2), (x.sub.2, y.sub.1),
(x.sub.2, y.sub.2) are the locations of the surface control points
Q11, Q12, Q21, Q22, respectively, in a blob of RSSI values
projected on a two-dimensional (2D) grid, as described in reference
to FIGS. 4 and 5. Using Equations [1] and [2] and the RSSI values
at the surface control points and their (x, y) locations in the 2D
blob, the RSSI prediction at location (x, y) can be computed by
localizer 304 of the mobile device, as described below. Because
only the encoded RSSI values at the surface control points and
their respective locations in the 2D blob are sent to the mobile
device (rather than all of the RSSI values in neighboring cells),
compression is achieved.
[0029] Other interpolation functions can also be used for RSSI
prediction, including but not limited to: bicubic, trilinear or
spline interpolation. In an embodiment, region quadtrees can be
used to lookup radio maps 206 of arbitrary resolutions. A quadtree
is a tree data structure in which each internal node has exactly
four children. The quadtree can partition a given geographic region
recursively by subdividing it into four quadrants or regions. The
regions may be square or rectangular or may have arbitrary shapes.
In an embodiment RSSI lookup module 302 selects the nearest four
(or less) surface control points Q.sub.11, Q.sub.12, Q.sub.21,
Q.sub.22 in the quadtree whose convex hull covers the location (x,
y).
[0030] Next, probability distribution generator 303 calculates a
probability distribution over location (x, y), where the
probability depends on a difference between predicted and observed
RSSI values plus uncertainty values. The probability distribution
can then be propagated using localizer 304 to generate location
estimates 305. Localizer 304 can be any suitable localizer,
including but not limited to: a particle filter, Kalman filter or
least squares estimator.
[0031] In an embodiment, the predicted RSSI values represent a mode
parameter of a Rayleigh distribution, e.g., if we take multiple
Rayleigh distributions with a single parameter (predicted_rssi). A
Rayleigh distribution can be represented mathematically by Equation
[3]:
f ( x ; .sigma. ) = x .sigma. 2 e - x 2 ( 2 .sigma. 2 ) , x
.gtoreq. 0. [ 3 ] ##EQU00002##
where sigma a is the predicted RSSI (predicted_rssi) and x is the
observed RSSI (observed_rssi) to get the conditional probability
P(x,y|observed_rssi, predicted_rssi). The predicted uncertainty
value .alpha. (predicted_uncertainty) is combined with
P(x,y|observed_rssi, predicted_rssi) as shown in Equation [4]:
final_probability=.alpha.*p(x,y| . . .
)+(1-.alpha.)*prior_probability, [4]
where prior_probability is a fixed tunable parameter.
[0032] In another embodiment, it is assumed that predicted_rssi and
predicted_uncertainty represent the mean and variance of a Gaussian
distribution. In this embodiment, the probability can be found by
substituting the parameters x, .mu., .sigma., with observed_rssi,
predicted_rssi and predicted_uncertainty, respectively, in Equation
[5]:
f ( x .mu. , .sigma. 2 ) = 1 2 .pi. .sigma. 2 e - ( x - .mu. ) 2 2
.sigma. 2 . [ 5 ] ##EQU00003##
[0033] FIGS. 4A-4C illustrate surface fitting to RSSI values,
according to an embodiment. As shown in FIG. 4A, surface fitting is
performed by surface fitting module 402, which takes as input RSSI
values and the associated uncertainty values for a given AP, and
outputs RSSI and uncertainty maps 206 (collectively, referred to as
"compressed radio maps") for the given AP. Any suitable
interpolation function can be used for surface fitting, including
but not limited to: bilinear, bicubic, trilinear and spline
interpolation. In the example shown, bilinear interpolation is
used.
[0034] FIG. 4B illustrates RSSI values 405 for a given AP that are
taken from neighboring or contiguous cells (e.g., 7 cells). Any
number of neighboring or contiguous cells can be selected based on
memory/processing and accuracy tradeoffs. RSSI values 405, when
projected onto a 2D image plane, define a 2D blob 404, which bounds
the geographic area where RSSI values 405 were harvested/surveyed.
Blob 404 can include all of the neighboring cells and/or portions
of the neighboring cells and can be any shape and size.
[0035] FIG. 4C shows surface 406 fitted to RSSI values 405,
resulting from the application of bilinear interpolation and four
surface control points. In an embodiment, surface control points
are at locations (x.sub.1, y.sub.1), (x.sub.1, y.sub.2), (x.sub.2,
y.sub.1), (x.sub.2, y.sub.2) in blob 404. Any suitable locations
for the surface control points can be used. In the example, corners
of grid cells are used. RSSI values 407a-407d at the surface
control locations (the values f(Q.sub.11), f(Q.sub.12),
f(Q.sub.21), f(Q.sub.22) in Equation [1]) and their respective
locations ((x.sub.1, y.sub.1), (x.sub.1, y.sub.2), (x.sub.2,
y.sub.1), (x.sub.2, y.sub.2)) can then be encoded, as described in
reference to FIGS. 5A and 5B, to achieve compression gain.
[0036] In an embodiment, the surface control points can be selected
as points on a regular 2D grid, as shown in FIG. 4C. In other
embodiments, surface control points are selected using an iterative
end-point fit algorithm (e.g., the Ramer-Douglas-Peucker algorithm)
or any other known algorithm that can be used for selecting surface
control points (e.g., Visvalingam-Whyatt, Reumann-Witkam, Opheim
simplification, Lang simplification, Zhao-Saalfeld).
[0037] In an embodiment, bilinear interpolation can be achieved
more efficiently using a cell adjustment process. First, a course
grid is selected and bilinear fitting is performed on RSSI values
in the coarse grid. For each set of N neighboring cells in the
coarse grid (e.g., four neighboring cells), a joint likelihood of
the RSSI values falling between the N cells is computed. If the
joint likelihood is below a threshold, the N cells are divided in
half to create a finer grid and bilinear fitting is performed again
on the finer grid. This dividing and surface fitting process
continues until the joint likelihood of the N neighboring cells is
greater than or equal to the threshold, at which point the process
stops.
[0038] In an embodiment, additional compression gain is achieved by
quantizing the RSSI values to a number of quantization levels
before surface fitting. For example, the RSSI values can be placed
in bins, where each been spans a range of RSSI values. Each bin can
be assigned a signal RSSI value (an RSSI quantization level). Each
RSSI value that falls within a given bin is replaced by the RSSI
quantization level. The quantization levels can be uniform or not
uniform. For example, the RSSI range [-100, 0] can be divided
non-uniformly into the quantization levels (-100, -50), (-49, -25),
(-25, -5), (-5, 0). The quantization levels need not be uniform
because the RSSI levels do not necessarily carry the same amount of
information.
[0039] FIGS. 5A-5C illustrate contour detection and blob boundary
encoding, according to an embodiment. For each AP, the RSSI values
at surface control points and their associated uncertainties, are
input to image processing module 502. Image processing module 502
places the RSSI values and uncertainty values on a 2D image grid
and performs contour detection to define blob boundaries. The blob
boundaries and the RSSI values and uncertainty values contained in
the blobs are input into encoder 503. Encoder 503 encodes the
boundaries and RSSI and uncertainty values, resulting in compressed
radio maps 504.
[0040] Referring to FIGS. 5BA and 5C, RSSI values at surface
control points are projected onto image grid 505 (not shown to
scale) and processed. Each RSSI value can be represented by one or
more pixels 506 in image grid 505. In an embodiment, image grid 505
can be a binary image grid. For example, pixels 506 representing
RSSI position data are given a value of 0 (black) and background
pixels are given values of 1 (white). To assist the contour
detection algorithm, in an embodiment the binary image is morphed
using flat dilation (e.g., a Minkowski sum) using, for example, a
disk as a structuring element with a specified radius (e.g., 3
pixels).
[0041] A contour detection algorithm is applied to the binary image
to detect one or more contours from the binary image grid 505,
where each contour represents a blob boundary. In the example
shown, blob boundaries 507, 508 are detected. Some examples of
image contour algorithms, include but are not limited to: square
tracing, Moore-Neighbor tracing and radial sweep. In an embodiment,
the input into a contour algorithm can be a square tessellation
with at least one continuous blob formed from black pixels and the
output is a sequence of boundary pixels for the blob. In some
cases, there may be more than one blob in the binary image
grid.
[0042] Blob boundaries 507, 508 are encoded using any suitable
encoding scheme, such as predictive differential encoding.
Additionally, the RSSI values at surface control points inside blob
boundaries 507, 508 are encoded using any suitable encoding scheme,
and selection of the encoding scheme can be based on the knowledge
of the range of values to be encoded. For example, in an embodiment
predictive differential encoding is used to encode the RSSI values
in blobs 507, 508, which are typically in dBm and are negative
double digit integers (e.g., -70 dBm). Because of the uncertainties
associated with the locations of the RSSI values in blobs 507, 508,
there are uncertainty regions 509, 510 around blobs 507, 508,
respectively. In an embodiment, the uncertainty values associated
with locations (x, y) of the RSSI values in blobs 507, 508 are
encoded using the same or different encoding scheme. For example,
in an embodiment the uncertainty values can range from 0 to 1 (1=no
uncertainty) and are encoded using predictive differential encoding
or run length encoding. In some implementations, variable Rice
coding can be used for encoding RSSI values and/or uncertainty
values within blobs. The encoded RSSI values and uncertainties are
stored as the aforementioned radio maps 206, which can be served to
mobile devices in tiles.
Example Processes
[0043] FIG. 6 is a flow diagram of a process of generating
compressed AP radio maps based on filtered AP data, according to an
embodiment. Process 600 can be implemented by architectures 800 and
900, as described in reference to FIGS. 8 and 9, respectively.
[0044] Process 600 can begin by receiving AP data (601). The AP
data is received from a plurality of mobile devices operating in a
geographic region. In an embodiment, the AP data includes a
reference location, RSSI values, RSSI uncertainty values and MAC
addresses for each AP observed (e.g., obtained in a wireless scan)
at the reference location, a timestamp and optionally other values
(e.g., PDR data, barometric sensor data). The reference location
can be provided by at least one of survey map data, a GNSS
receiver, such a Global Positioning System (GPS) receiver and WiFi
or cell tower position data. For example, a GPS receiver can
provide the position coordinates (e.g., latitude and longitude) of
the reference location in a local-level (geodetic) coordinate
system.
[0045] Process 600 continues by filtering APs (602). For example,
moving APs can be excluded from further processing. In an
embodiment, AP locations can be clustered to identify moving APs,
as described in reference to FIG. 2.
[0046] Process 600 continues by filtering RSSI values (603). For
example, RSSI values for an AP that are outside a range define by
minimum and maximum thresholds can be excluded from further
processing. Also, RSSI values associated with reference locations
that have large position uncertainty (e.g., a horizontal position
uncertainty greater than a threshold) can be excluded from further
processing.
[0047] Process 600 continues by generating compressed radio maps
(604) as described in reference to FIGS. 4 and 5. The compressed
radio maps include AP RSSI and uncertainty maps generated by
fitting a surface to the RSSI and uncertainty values. Because only
quantized RSSI values at surface control points, and their
respective locations in neighboring cells of the 2D grid, are
included in the AP RSSI map (rather than all of the RSSI values in
the neighboring cells), compression is achieved. The compressed
RSSI and uncertainty maps, collectively referred to as compressed
radio maps, are stored in a radio map database. The maps can be
served to mobile devices in tiles. In an embodiment multiple
versions of the radio maps having different resolutions can be
stored in a quad-tree to facilitate decoding and decompression on
the mobile device.
[0048] FIG. 7 is a flow diagram of a process of surface fitting and
blob boundary encoding, according to an embodiment. Process 700 can
be implemented by architectures 800 and 900, as described in
reference to FIGS. 8 and 9, respectively.
[0049] Process 700 can begin by receiving RSSI values and
associated uncertainty values for an AP (701). The RSSI values can
have reference locations from one cell or a set of neighboring
cells (e.g., four cells) in a virtual grid dividing a geographic
area into cells. In an embodiment, bilinear interpolation can be
achieved more efficiently using a cell adjustment process. First, a
course grid is selected and bilinear fitting is performed on RSSI
values in the coarse grid. For each set of N neighboring cells in
the coarse grid (e.g., four neighboring cells), a joint likelihood
of the RSSI values falling between the N cells is computed. If the
joint likelihood is below a threshold, the N cells are divided in
half to create a finer grid and bilinear fitting is performed again
on the finer grid. This dividing and surface fitting process
continues until the joint likelihood of the N neighboring cells is
greater than or equal to the threshold, at which point the process
stops.
[0050] Process 700 continues by fitting a surface to the RSSI
values and the uncertainty values (702). Prior to surface fitting,
the RSSI values can be quantized to increase compression gain. In
an embodiment, bilinear interpolation is used for surface fitting.
Surface control points (e.g., four or less for bilinear) can be
selected from the RSSI values based on a desired cell
resolution.
[0051] Process 700 continues by projecting RSSI values at the
selected surface control points to a binary image grid and
determining and encoding a blob boundary (703). For example, the
RSSI values and associated uncertainties at the surface control
points that are projected to the binary image grid can be processed
by a contour detection algorithm to determine blob boundaries,
which define a geographic region where the RSSI values were
harvested/surveyed.
[0052] Process 700 continues by encoding RSSI values and associated
uncertainty values for an AP using one or more encoding schemes
(704). The encoding results in compressed RSSI and uncertainty maps
for the AP (collectively, compressed radio maps), which are stored
in FP database 106 (See FIG. 1). For example, blob boundaries, RSSI
values and their associated uncertainties at the surface control
points can be encoded using one or more encoding schemes, including
but not limited to: predictive differential encoding, run length
encoding and Rice coding.
Example Server Architecture
[0053] FIG. 8 is a block diagram of example server architecture 800
for implementing the server-side features and processes described
in reference to FIGS. 1-7, according to an embodiment. Other
architectures are possible, including architectures with more or
fewer components. In some implementations, architecture 800
includes one or more processor(s) 802 (e.g., dual-core Intel.RTM.
Xeon.RTM. processors), one or more network interface(s) 806, one or
more storage device(s) 804 (e.g., hard disk, optical disk, flash
memory) and one or more computer-readable medium(s) 808 (e.g., hard
disk, optical disk, flash memory, etc.). These components can
exchange communications and data over one or more communication
channel(s) 810 (e.g., buses), which can utilize various hardware
and software for facilitating the transfer of data and control
signals between components.
[0054] The term "computer-readable medium" refers to any storage
medium that stores and provides instructions to processor(s) 802
for execution, including without limitation, non-volatile media
(e.g., optical or magnetic disks, ROM) and volatile media (e.g.,
memory, RAM). Computer-readable medium(s) 808 can further include
computer program instructions for implementing operating system 812
(e.g., Mac OS.RTM. server, Windows.RTM. NT server), network
communication stack 814, FPS module 816 and tile distribution
manager 818 for performing the server-side processes described in
reference to FIGS. 1-7. Computer program instructions can be based
on any suitable computer language (e.g., C++, Java).
[0055] Operating system 812 can be multi-user, multiprocessing,
multitasking, multithreading, real time, etc. Operating system 812
performs basic tasks, including but not limited to: recognizing
input from and providing output to devices 802, 804, 806 and 808;
keeping track and managing files and directories on
computer-readable medium(s) 808 (e.g., memory or a storage device);
controlling peripheral devices; and managing traffic on the one or
more communication channel(s) 810. Network communications stack 813
includes various components for establishing and maintaining
network connections (e.g., software for implementing communication
protocols, such as TCP/IP, HTTP, etc.).
[0056] Architecture 800 can be included in any computer device,
including one or more server computers in a local or distributed
network each having one or more processing cores. Architecture 800
can be implemented in a parallel processing or peer-to-peer
infrastructure or on a single device with one or more processors.
Software can include multiple software components or can be a
single body of code.
Example Mobile Device Architecture
[0057] FIG. 9 illustrates an example device architecture 900 of a
mobile device implementing client-side features and operations
described in reference to FIGS. 1-7. Architecture 900 can include
memory interface 902, one or more data processors, image processors
and/or processors 904 and peripherals interface 906. Memory
interface 902, one or more processors 904 and/or peripherals
interface 906 can be separate components or can be integrated in
one or more integrated circuits. The various components in
architecture 900 can be coupled by one or more communication buses
or signal lines.
[0058] Sensors, devices and subsystems can be coupled to
peripherals interface 906 to facilitate multiple functionalities.
For example, one or more motion sensors 910, light sensor 912 and
proximity sensor 914 can be coupled to peripherals interface 906 to
facilitate motion sensing (e.g., acceleration, rotation rates),
lighting and proximity functions of the mobile device. Location
processor 915 can be connected to peripherals interface 906 to
provide geopositioning. In some implementations, location processor
915 can be a GNSS receiver, such as the Global Positioning System
(GPS). Electronic magnetometer 916 (e.g., an integrated circuit
chip) can also be connected to peripherals interface 906 to provide
data that can be used to determine the direction of magnetic North.
Electronic magnetometer 916 can provide data to an electronic
compass application. Motion sensor(s) 910 can include one or more
accelerometers and/or gyros configured to determine change of speed
and direction of movement of the mobile device. Barometer 917 can
be configured to measure atmospheric pressure around the mobile
device.
[0059] Camera subsystem 920 and an optical sensor 922, e.g., a
charged coupled device (CCD) or a complementary metal-oxide
semiconductor (CMOS) optical sensor, can be utilized to facilitate
camera functions, such as capturing photographs and recording video
clips.
[0060] Communication functions can be facilitated through one or
more wireless communication subsystems 924, which can include radio
frequency (RF) receivers and transmitters (or transceivers) and/or
optical (e.g., infrared) receivers and transmitters. The specific
design and implementation of the communication subsystem 924 can
depend on the communication network(s) over which a mobile device
is intended to operate. For example, architecture 900 can include
communication subsystems 924 designed to operate over a GSM
network, a GPRS network, an EDGE network, a Wi-Fi.TM. or WiMax.TM.
network and a Bluetooth.TM. network. In particular, the wireless
communication subsystems 924 can include hosting protocols, such
that the mobile device can be configured as a base station for
other wireless devices.
[0061] Audio subsystem 926 can be coupled to a speaker 928 and a
microphone 930 to facilitate voice-enabled functions, such as voice
recognition, voice replication, digital recording and telephony
functions. Audio subsystem 926 can be configured to receive voice
commands from the user.
[0062] I/O subsystem 940 can include touch surface controller 942
and/or other input controller(s) 944. Touch surface controller 942
can be coupled to a touch surface 946 or pad. Touch surface 946 and
touch surface controller 942 can, for example, detect contact and
movement or break thereof using any of a plurality of touch
sensitivity technologies, including but not limited to capacitive,
resistive, infrared and surface acoustic wave technologies, as well
as other proximity sensor arrays or other elements for determining
one or more points of contact with touch surface 946. Touch surface
946 can include, for example, a touch screen. I/O subsystem 940 can
include a haptic engine or device for providing haptic feedback
(e.g., vibration) in response to commands from a processor.
[0063] Other input controller(s) 944 can be coupled to other
input/control devices 948, such as one or more buttons, rocker
switches, thumb-wheel, infrared port, USB port and/or a pointer
device such as a stylus. The one or more buttons (not shown) can
include an up/down button for volume control of speaker 928 and/or
microphone 930. Touch surface 946 or other controllers 944 (e.g., a
button) can include, or be coupled to, fingerprint identification
circuitry for use with a fingerprint authentication application to
authenticate a user based on their fingerprint(s).
[0064] In one implementation, a pressing of the button for a first
duration may disengage a lock of the touch surface 946; and a
pressing of the button for a second duration that is longer than
the first duration may turn power to the mobile device on or off.
The user may be able to customize a functionality of one or more of
the buttons. The touch surface 946 can, for example, also be used
to implement virtual or soft buttons and/or a virtual touch
keyboard.
[0065] In some implementations, the mobile device can present
recorded audio and/or video files, such as MP3, AAC and MPEG files.
In some implementations, the mobile device can include the
functionality of an MP3 player. Other input/output and control
devices can also be used.
[0066] Memory interface 902 can be coupled to memory 950. Memory
950 can include high-speed random access memory and/or non-volatile
memory, such as one or more magnetic disk storage devices, one or
more optical storage devices and/or flash memory (e.g., NAND, NOR).
Memory 950 can store operating system 952, such as iOS, Darwin,
RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system
such as VxWorks. Operating system 952 may include instructions for
handling basic system services and for performing hardware
dependent tasks. In some implementations, operating system 952 can
include a kernel (e.g., UNIX kernel).
[0067] Memory 950 may also store communication instructions 954 to
facilitate communicating with one or more additional devices, one
or more computers and/or one or more servers, such as, for example,
instructions for implementing a software stack for wired or
wireless communications with other devices. Memory 950 may include
graphical user interface instructions 956 to facilitate graphic
user interface processing; sensor processing instructions 958 to
facilitate sensor-related processing and functions; phone
instructions 960 to facilitate phone-related processes and
functions; electronic messaging instructions 962 to facilitate
electronic-messaging related processes and functions; web browsing
instructions 964 to facilitate web browsing-related processes and
functions; media processing instructions 966 to facilitate media
processing-related processes and functions; GNSS/Location
instructions 968 to facilitate generic GNSS and location-related
processes and instructions, including processed described in
reference to FIGS. 1-7; and camera instructions 970 to facilitate
camera-related processes and functions. Memory 950 further includes
radio map decoding/decompression instructions 972 for decompressing
compressed radio maps. The memory 950 may also store other software
instructions (not shown), such as security instructions, web video
instructions to facilitate web video-related processes and
functions and/or web shopping instructions to facilitate web
shopping-related processes and functions. In some implementations,
the media processing instructions 966 are divided into audio
processing instructions and video processing instructions to
facilitate audio processing-related processes and functions and
video processing-related processes and functions, respectively.
[0068] Each of the above identified instructions and applications
can correspond to a set of instructions for performing one or more
functions described above. These instructions need not be
implemented as separate software programs, procedures, or modules.
Memory 950 can include additional instructions or fewer
instructions. Furthermore, various functions of the mobile device
may be implemented in hardware and/or in software, including in one
or more signal processing and/or application specific integrated
circuits.
[0069] One or more features or steps of the disclosed embodiments
may be implemented using an Application Programming Interface
(API). An API may define on or more parameters that are passed
between a calling application and other software code (e.g., an
operating system, library routine, function) that provides a
service, that provides data, or that performs an operation or a
computation. The API may be implemented as one or more calls in
program code that send or receive one or more parameters through a
parameter list or other structure based on a call convention
defined in an API specification document. A parameter may be a
constant, a key, a data structure, an object, an object class, a
variable, a data type, a pointer, an array, a list, or another
call. API calls and parameters may be implemented in any
programming language. The programming language may define the
vocabulary and calling convention that a programmer will employ to
access functions supporting the API. In some implementations, an
API call may report to an application the capabilities of a device
running the application, such as input capability, output
capability, processing capability, power capability, communications
capability, etc.
[0070] While this specification contains many specific
implementation details, these should not be construed as
limitations on the scope of any inventions or of what may be
claimed, but rather as descriptions of features specific to
particular embodiments of particular inventions. Certain features
that are described in this specification in the context of separate
embodiments can also be implemented in combination in a single
embodiment. Conversely, various features that are described in the
context of a single embodiment can also be implemented in multiple
embodiments separately or in any suitable sub combination.
Moreover, although features may be described above as acting in
certain combinations and even initially claimed as such, one or
more features from a claimed combination can in some cases be
excised from the combination, and the claimed combination may be
directed to a sub combination or variation of a sub
combination.
[0071] Similarly, while operations are depicted in the drawings in
a particular order, this should not be understood as requiring that
such operations be performed in the particular order shown or in
sequential order, or that all illustrated operations be performed,
to achieve desirable results. In certain circumstances,
multitasking and parallel processing may be advantageous. Moreover,
the separation of various system components in the embodiments
described above should not be understood as requiring such
separation in all embodiments, and it should be understood that the
described program components and systems can generally be
integrated together in a single software product or packaged into
multiple software products.
[0072] Thus, particular embodiments of the subject matter have been
described. Other embodiments are within the scope of the following
claims. In some cases, the actions recited in the claims can be
performed in a different order and still achieve desirable results.
In addition, the processes depicted in the accompanying figures do
not necessarily require the particular order shown, or sequential
order, to achieve desirable results. In certain implementations,
multitasking and parallel processing may be advantageous.
[0073] As described above, some aspects of the subject matter of
this specification include gathering and use of data available from
various sources to improve services a mobile device can provide to
a user. The present disclosure contemplates that in some instances,
this gathered data may identify a particular location or an address
based on device usage. Such personal information data can include
location-based data, addresses, subscriber account identifiers, or
other identifying information.
[0074] The present disclosure further contemplates that the
entities responsible for the collection, analysis, disclosure,
transfer, storage, or other use of such personal information data
will comply with well-established privacy policies and/or privacy
practices. In particular, such entities should implement and
consistently use privacy policies and practices that are generally
recognized as meeting or exceeding industry or governmental
requirements for maintaining personal information data private and
secure. For example, personal information from users should be
collected for legitimate and reasonable uses of the entity and not
shared or sold outside of those legitimate uses. Further, such
collection should occur only after receiving the informed consent
of the users. Additionally, such entities would take any needed
steps for safeguarding and securing access to such personal
information data and ensuring that others with access to the
personal information data adhere to their privacy policies and
procedures. Further, such entities can subject themselves to
evaluation by third parties to certify their adherence to widely
accepted privacy policies and practices.
[0075] In the case of advertisement delivery services, the present
disclosure also contemplates embodiments in which users selectively
block the use of, or access to, personal information data. That is,
the present disclosure contemplates that hardware and/or software
elements can be provided to prevent or block access to such
personal information data. For example, in the case of
advertisement delivery services, the present technology can be
configured to allow users to select to "opt in" or "opt out" of
participation in the collection of personal information data during
registration for services.
[0076] Therefore, although the present disclosure broadly covers
use of personal information data to implement one or more various
disclosed embodiments, the present disclosure also contemplates
that the various embodiments can also be implemented without the
need for accessing such personal information data. That is, the
various embodiments of the present technology are not rendered
inoperable due to the lack of all or a portion of such personal
information data. For example, content can be selected and
delivered to users by inferring preferences based on non-personal
information data or a bare minimum amount of personal information,
such as the content being requested by the device associated with a
user, other non-personal information available to the content
delivery services, or publically available information.
* * * * *