U.S. patent application number 14/338419 was filed with the patent office on 2014-12-11 for tdoa based positioning with calculation of correction factors for compensating the clock offsets of unsynchronized network stations.
The applicant listed for this patent is ROCKSTAR CONSORTIUM US LP. Invention is credited to David Bevan, Simon Gale.
Application Number | 20140364142 14/338419 |
Document ID | / |
Family ID | 41697953 |
Filed Date | 2014-12-11 |
United States Patent
Application |
20140364142 |
Kind Code |
A1 |
Bevan; David ; et
al. |
December 11, 2014 |
TDOA BASED POSITIONING WITH CALCULATION OF CORRECTION FACTORS FOR
COMPENSATING THE CLOCK OFFSETS OF UNSYNCHRONIZED NETWORK
STATIONS
Abstract
The present invention presents a method, arrangement and
computer program product for clocking exploiting the relative
behavior of clocks of individual receiving stations as well as a
corresponding modeling to derive a time difference of arrival of a
signal from a user device which can be used to correct the time
difference of arrival based on the modeled clock behavior and leads
to a correct clocking of received user signals without the need of
synchronization of the clocks in the various receiving stations.
This principle is applicable to a plurality of pairs of receiving
stations and beacon signals transmitted amongst them and allows for
a correct location estimation of a user device.
Inventors: |
Bevan; David; (Bishops
Stortford, GB) ; Gale; Simon; (Bishops Stortford,
GB) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
ROCKSTAR CONSORTIUM US LP |
Plano |
TX |
US |
|
|
Family ID: |
41697953 |
Appl. No.: |
14/338419 |
Filed: |
July 23, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13513875 |
Aug 14, 2012 |
8818406 |
|
|
PCT/EP2010/050747 |
Jan 22, 2010 |
|
|
|
14338419 |
|
|
|
|
Current U.S.
Class: |
455/456.1 |
Current CPC
Class: |
G01S 5/06 20130101; H04W
64/003 20130101; H04L 7/0054 20130101; G01S 5/021 20130101; H04W
24/08 20130101 |
Class at
Publication: |
455/456.1 |
International
Class: |
G01S 5/06 20060101
G01S005/06; H04L 7/00 20060101 H04L007/00 |
Foreign Application Data
Date |
Code |
Application Number |
Dec 10, 2009 |
EP |
PCT/EP2009/066858 |
Claims
1.-17. (canceled)
18. A method of determining a position of a first device relative
to a second device, a third device and a fourth device, the second,
third and fourth devices being in a fixed positional relationship
and having second, third and fourth local clocks respectively, the
method comprising: estimating respective offsets of the second
clock and the third clock relative to the fourth clock as a
function of time; transmitting a signal from the first device;
receiving the signal transmitted from the first device at the
second and third devices; determining respective reception times of
the signal at the second and third devices using the second and
third local clocks respectively; and estimating the position of the
first device using the respective reception times of the signal at
the second and third devices and the respective offsets of the
second clock and the third clock as a function of time.
19. The method of claim 18, wherein estimating the position of the
first device using the respective reception times of the signal at
the second and third devices and the respective offsets of the
second clock and the third clock as a function of time comprises
estimating the position of the first device using the respective
reception times of the signal at the second and third devices and
respective offsets of the second clock and the third clock at times
bracketing the respective reception times of the signal at the
second and third devices.
20. The method of claim 18 wherein estimating the position of the
first device using the respective reception times of the signal at
the second and third devices and the respective offsets of the
second clock and the third clock as a function of time comprises
using time difference of arrival (TDOA).
21. The method of claim 18, wherein the first device is a mobile
wireless communication terminal and the second and third devices
are wireless access points deployed at fixed locations.
22. The method of claim 18, wherein the fourth device is a beacon
deployed at a fixed location.
23. The method of claim 18, wherein estimating respective offsets
of the second clock and the third clock relative to the fourth
clock as a function of time comprises: transmitting a sequence of
individually identifiable signals from the fourth device; receiving
the sequence of individually identifiable signals at each of the
second device and the third device; determining respective
reception times of each of the individually identifiable signals at
the second device using the second clock; determining respective
reception times of each of the individually identifiable signals at
the third device using the third clock; and estimating the
respective offsets of the second clock and the third clock relative
to the fourth clock from the reception times of the individually
identifiable signals at the second and third devices.
24. The method of claim 23, wherein the sequence of individually
identifiable signals comprises a sequence of beacon signals.
25. The method of claim 24, wherein each beacon signal comprises an
identifier which is unique to that beacon signal.
26. The method of claim 25, wherein each beacon signal is a frame,
and the identifier which is unique to that beacon signal is a frame
number.
27. The method of claim 26, wherein each frame is a frame compliant
with IEEE 802.11.
28. The method of claim 23, wherein: transmitting the sequence of
individually identifiable signals from the fourth device comprises
transmitting at least a first beacon signal and a second beacon
signal from the fourth device; determining the respective reception
times of each of the individually identifiable signals at the
second device comprises determining reception times for the first
and second beacon signals at the second device using the second
clock; determining respective reception times of each of the
individually identifiable signals at the third device comprises
determining reception times for the first and second beacon signals
at the third device using the third clock; and estimating the
respective offsets of the second clock and the third clock relative
to the fourth clock from the reception times of the individually
identifiable signals at the second and third devices comprises
estimating the respective offsets based on the reception times of
the first and second beacon signals at the second device determined
using the second clock and the reception times of the first and
second beacon signals at the third device determined using the
third clock.
29. The method of claim 28, wherein estimating the position of the
first device using the respective reception times of the signal at
the second and third devices and the respective offsets of the
second clock and the third clock as a function of time comprises
estimating the position of the first device using the respective
reception times of the signal at the second and third devices and
the reception times of the first and second beacon signals at the
second device determined using the second clock and the reception
times of the first and second beacon signals at the third device
determined using the third clock.
30. The method of claim 23, wherein: transmitting the sequence of
individually identifiable signals from the fourth device comprises
transmitting more than two successive beacon signals from the
fourth device; determining the respective reception times of each
of the individually identifiable signals at the second device
comprises determining reception times for each of the beacon
signals at the second device using the second clock; determining
respective reception times of each of the individually identifiable
signals at the third device comprises determining reception times
for each of the beacon signals at the third device using the third
clock; and estimating the respective offsets of the second clock
and the third clock relative to the fourth clock from the reception
times of the individually identifiable signals at the second and
third devices comprises estimating the respective offsets based on
the reception times of the first and second beacon signals at the
second device determined using the second clock and the reception
times of the beacon signals at the third device determined using
the third clock.
31. The method of claim 30, wherein estimating the position of the
first device using the respective reception times of the signal at
the second and third devices and the respective offsets of the
second clock and the third clock as a function of time comprises
estimating the position of the first device using the respective
reception times of the signal at the second and third devices and
the reception times of the beacon signals at the second device
determined using the second clock and the reception times of the
beacon signals at the third device determined using the third
clock.
32. The method of claim 30, wherein estimating the respective
offsets based on the reception times of the first and second beacon
signals at the second device determined using the second clock and
the reception times of the beacon signals at the third device
determined using the third clock comprises modelling a function
that relates the reception times of the beacon signals at the
second device determined using the second clock to the reception
times of the beacon signals at the third device using the third
clock.
33. The method of claim 32, wherein modelling a function that
relates the reception times of the beacon signals at the second
device determined using the second clock to the reception times of
the beacon signals at the third device using the third clock
comprises interpolating the function between the determined times
of the beacon signals.
34. The method of claim 33, wherein interpolating the function
between the determined times of the beacon signals comprises
modelling the function as a polynomial.
35. The method of claim 34, wherein modelling the function as a
polynomial comprises using least squares fitting.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a Continuation of copending U.S.
application Ser. No. 13/513,875, filed Aug. 14, 2012, which is a
U.S. nationalization under 35 U.S.C. .sctn.371 of International
Application No. PCT/EP2010/050747, filed Jan. 22, 2010, which
claims priority to International Patent Application No.
PCT/EP2009/066858, filed Dec. 10, 2009. The disclosures set forth
in the referenced applications are incorporated herein by reference
in their entireties.
TECHNICAL BACKGROUND
[0002] The present invention relates to wireless mobile networks or
access points of a wireless local network having mobile user
location functions, associated methods and computer program
products.
[0003] In recent times it has become increasingly important to
determine the location of a user using a mobile device in order to
provide appropriate services to users, such as restaurant or shop
recommendation or to locate the user in order to provide emergency
medical services. On the other hand it has also become important to
locate a user to provide appropriate law enforcement regarding
criminal subjects.
[0004] For determining the location of a mobile device commonly
satellite supported methods have been established, such as using
the Global Positioning System which is widely used to determine the
position of cars and used for route planning and navigation. Such
methods however have the disadvantage, that they require special
receivers and transmitters which are only suitable for navigational
purposes and require an additional technical effort as well in the
user device and in terms of infrastructure to supply the
corresponding satellites emitting proper navigation signals.
[0005] Due to the high competition in the mobile infrastructure and
devices market there is a strong tendency among the competitors to
keep mobile devices and their infrastructure technically simple, in
order to remain cost-competitive on the market or to gain a
competitive advantage. Therefore, there is a strong need to provide
location services for mobile devices which are technically simple
while at the same time being reliable and efficient as well as
sufficiently accurate to determine the position of a user in order
to be able to provide to him or her suitable services.
[0006] Accordingly methods have been established to determine the
position of a user, by receiving a signal emitted by the user's
mobile device at fixed receiver stations, and based on the
different signal propagation from the user device to the different
fixed receivers to calculate a position of the user device. In such
an environment it is crucial that all the devices involved in the
location determination are basing their corresponding analysis on a
common clock. The positional accuracy of such a system is directly
related to the accuracy of the clocks and the corresponding
synchronicity of the clocks which are used to determine the
propagation delays involved in the location determination.
[0007] One such method is based on a time difference of arrival
TDOA and uses the time difference it takes for a signal to travel
to two destinations as an indirect method of calculating a
distance. With a minimum of three base stations, for instance,
receiving the signal from a handset, the difference in time it
takes for the signal to reach each tower of a base station can be
used to triangulate the position of the mobile unit. TDOA systems
do not need any specialized antennas and as such the infrastructure
is kept simple. When a mobile that has to be located transmits, the
arrival time of the target mobile signal is recorded by a TDOA
location measuring unit at each base station or access point which
is able to receive the signal. Since the mobile's signal travels at
a constant speed (the speed of light), comparison of the arrival
time of the signal for any two sides allows a straightforward
calculation to determine the mobile's relative position to each
side. When plotted, this relationship describes an imaginary
hyperbola in space. The target mobile is located somewhere on this
curve although additional information is required to determine
precisely where. When the same calculation is made involving
measurements from a third base station or access point side,
calculating a difference of arrival time from sites A, B, C, e.g.
between either sites A and C or between sites B and C, an
independent positional hyperbola can be described. The point at
which the two hyperbolas AB and BC intersect is the location of the
target mobile. Commonly, TDOA requires accurate time
synchronization at the base station or access points, but not
necessarily at the target mobile device. It is immediately
apparent, that inaccuracies in the clock measurements may lead to
large location errors. A high clock accuracy and the corresponding
master clock and an associated synchronization procedure require
however technically complicated solutions which on the other hand
require adaptations at the infrastructure devices such as base
stations and for instance access points in order to provide them
with a suitable clock reference.
SUMMARY OF THE INVENTION
[0008] Therefore, there exists a need to provide a location
estimation of a mobile device without the requirement of clock
synchronization or expensive master clocks.
[0009] This problem is solved by a method for clocking according to
claim 1, by an arrangement for clocking according to claim 13 or
15, and by a computer program product for clocking according to
claim 17.
[0010] Further advantageous developments of the invention are given
in the dependent claims.
[0011] Advantageously the method according to the present invention
exploits the fact that base stations of a wireless mobile network
or access points of a wireless local network are located at known
positions and thus are situated in a fixed positional relationship.
This allows to calculate signal propagation delays of beacon
signals transmitted amongst those stations and received
individually. These are calculated based on the known signal
propagation speed and the respective distance between the
stations.
[0012] Additionally, determining the time difference of arrival
allows determination of the relation of local clocks of two
individual receiving stations to each other and requires no
absolute values. Thus, by measuring the reception times of uniquely
identified signals at individual receiving stations, a relative
clock behavior of pairs of individual receiving stations can be
modeled and exploited together with the absolute known time
difference of arrival to calculate a correction value for a signal
received from a mobile device of a user by the respective base
stations or wireless access points to correct the corresponding
time difference of arrival for the user signal. Consequently, the
method according to the present invention solves the problem of the
present invention with using no additional hardware only by
relating suitable calculations of suitable measurements at the
access points respectively base stations.
[0013] Expediently according to a further development of an example
of the method according to the present invention a transmission and
a time stamping of a plurality of signals is analyzed which
advantageously provides for a higher accuracy in the model
curve.
[0014] Beneficially according to a further development of the
method of the present invention the signals are transmitted
wirelessly over the air, which advantageously allows for a very
simple infrastructure as no cables and wires need to be
deployed.
[0015] Beneficially according to a further development of the
method of the present invention the signal is embodied as a beacon
signal, because this allows using normal wireless access points or
base stations which already transmit beacon signals to be used in
the context of the present invention.
[0016] Advantageously, according to a further development of the
method according to the present invention a signal is a frame, as
frames by nature provide the advantage to possess a unique
identification property in their frame number and thus further ease
the implementation of the method of the present invention in
presently used transmission systems.
[0017] Further beneficially according to a development of the
present invention the frame can be embodied as a frame according to
a local area network protocol such as the IEEE 802.11 standard, as
in this manner standards can be easily adapted by the present
invention and commonly used transmission standards are suitable for
incorporation in the method according to the present invention.
[0018] Advantageously for the modeling of the time difference of
arrival of the method according to the present invention according
to an embodiment, a polynomial with least squares may be used
because such a polynomial is simple in its mathematical structure
while at the same time satisfying the descriptive needs of the
dependency of two clocks according to the method of the present
invention without sacrificing any accuracy in the modeling
process.
[0019] Expediently according to a further development of an
embodiment of the method according to the present invention at
least one signal is being transmitted before the arbitrary point in
time and one is being transmitted after the arbitrary point in time
in order to improve the accuracy of the clocking at the arbitrary
point in time according to the method of the present invention.
[0020] Beneficially according to a further development of the
method according to the present invention it is ensured that the
same number of signals has been transmitted before the arbitrary
point in time and after the arbitrary point in time to provide the
utmost accuracy in the determination of the clocking according to a
further embodiment of the method according to the present
invention.
[0021] Expediently according to a further development of the method
of the present invention the timing difference of arrival at a
different pair of receiving stations is used to be able to
accurately determine the location of a mobile station by a method
of time difference of arrival based location determination. This
allows the determination of two hyperbolas and their corresponding
intersection point as the location of the mobile device.
[0022] The present invention provides an arrangement for clocking
comprising:
[0023] at least a first, a second and a third device;
[0024] the first device having transmitting means for transmitting
at least a first and a second signal with a unique
identification;
[0025] the second and third devices having receiving means and a
clock for receiving the at least first and second signals and
processing means for associating a respective measured second and
third local clock at the reception time of the respective signal to
it, leading to at least two value pairs of local clocks of the
second and third device;
[0026] modeling means for modeling based on the value pairs of
local clocks a time function of a dependency of the second and
third local clock over time as a first model curve;
[0027] the second and third device further being adapted to
receiving at an arbitrary point in time a user signal from a user
device at the second and third device and associating a reception
time measured by the respective second and third local clock of the
second and third device to the user signal upon its reception,
leading to a user value pair of local clocks of the second and
third device,
[0028] processing means for calculating a reference time difference
of arrival RTDOA for a signal from the first device received at the
second and third device from the fixed positional relationship,
based on the respective distance between the first and the second
device and the first and the third device and the known signal
propagation speed; and for calculating a user time difference of
arrival UTDOA from the user value pair;
[0029] means for determining at the arbitrary point in time the
value form the first model curve and based on this value
calculating a determined time difference of arrival DTDOA;
[0030] wherein the processing means is adapted to relate the RTDOA
and DTDOA to determine a current correction factor; and
[0031] for clocking to use the current correction factor to correct
the UTDOA.
[0032] Beneficially the arrangement according to the present
invention comprises a minimum number of receiving devices to
achieve the clocking according to the present invention which
provides a simple infrastructure in a competitive manner capable to
solve the problem of the present invention.
[0033] Advantageously, a further development of the arrangement
according to the present invention allows employing a server for
the computational intensive tasks that need to be performed, which
on the other hand allows a further simplification of the receiving
stations, and only requires corresponding communications of the
respective identification of the received signals together with
their associated time information to the computation server for
calculating a location of the mobile device.
[0034] Advantageously the computer program product of the invention
provides a simple means to implement the method of the present
invention at respective base stations and access points by
providing a means of storage and transport.
BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Subsequently the invention will further be explained by
means of examples and embodiments shown in drawings, wherein
[0036] FIG. 1 shows a simple transmitter and receiver arrangement
according to an embodiment of the present invention;
[0037] FIG. 2 gives an example of a relative clock behaviour of two
receiving stations;
[0038] FIG. 3 shows an example of signal transmission according to
an embodiment of the present invention;
[0039] FIG. 4 shows a flow-chart explaining the estimation of a
user location based on a time difference of arrival according to an
embodiment of the present invention;
[0040] FIG. 5 gives an example of a computer program product
according to the present invention; and
[0041] FIG. 6 gives an example of a device of an arrangement
according to the present invention.
DESCRIPTION OF THE ILLUSTRATIVE EMBODIMENTS
[0042] The present invention will be described with respect to
particular embodiments and with reference to certain drawings but
the invention is not limited thereto but only by the claims. The
drawings described are only schematic and are non-limiting. In the
drawings, the size of some of the elements may be exaggerated and
not drawn on scale for illustrative purposes. Where the term
"comprising" is used in the present description and claims, it does
not exclude other elements or steps. Where an indefinite or
definite article is used when referring to a singular noun e.g. "a"
or "an", "the", this includes a plural of that noun unless
something else is specifically stated.
[0043] The term "comprising", used in the claims, should not be
interpreted as being restricted to the means listed thereafter; it
does not exclude other elements or steps. Thus, the scope of the
expression "a device comprising means A and B" should not be
limited to devices consisting only of components A and B. It means
that with respect to the present invention, the only relevant
components of the device are A and B.
[0044] Furthermore, the terms first, second, third and the like in
the description and in the claims, are used for distinguishing
between similar elements and not necessarily for describing a
sequential or chronological order. It is to be understood that the
terms so used are interchangeable under appropriate circumstances
and that the embodiments of the invention described herein are
capable of operation in other sequences than described or
illustrated herein.
[0045] Moreover, the terms top, bottom, over, under and the like in
the description and the claims are used for descriptive purposes
and not necessarily for describing relative positions. It is to be
understood that the terms so used are interchangeable under
appropriate circumstances and that the embodiments of the invention
described herein are capable of operation in other orientations
than described or illustrated herein.
[0046] As an example given in FIG. 1, a basic configuration to
embody the method of the present invention comprises or consists of
three access points AP1, AP2 and AP4. Without limiting the
invention these access points can be any stations that are capable
of transmitting respectively receiving signals, e.g. wireless
signals. The potential to be a transceiver, i.e. as well a
transmitter as a receiver has certain advantages in terms of not
having to have separate transmitting and receiving stations and
thus providing an optimum use of resources. In this case the access
points may be access points of a wireless LAN, e.g. according to
the IEEE Standard 802.11. But without any limitation the access
points may be stations that are capable of transmitting
respectively receiving a signal, respectively a number of signals,
especially wireless signals which can be provided with a unique
identifier or at least an element that enables a signal to be
identifiable as a unique signal. It is for instance conceivable
that each signal has a different format and thus the unique
identifier of the signal to be a unique signal is the format.
[0047] In this case the station AP4 transmits a unique signal,
respectively a sequence of unique signals that is/are received by
stations AP2 and AP1. Upon reception of the unique signal or
signals, by means of their local clock, the respective stations AP1
and AP2 associate the unique signal with a time measured by their
local clock that has been measured at the time of reception of the
unique signal. Without any limitation such an association may be
performed for one, two or a plurality of unique signals in order to
protocol a dependency of the local clocks of the stations AP2 and
AP1 in association to a joint reception of the same individual
unique signal from the station AP4.
[0048] Often at wireless access points, such as the stations AP1,
AP2 and AP4, these have a simple construction and thus only low
cost clocking devices are employed, which has the disadvantage,
that these clocks are subject to a temperature drift to a clock
inaccuracy drift, e.g. to a phase drift, and/or to a frequency
drift of the clock, which imposes a certain inaccuracy of the
clocking of the individual stations AP1 to AP4. The absolute
offsets and rate differences of the various clocks of the stations
AP1 to AP4 may be quantified in microseconds of drift per second,
or equivalently, parts per million PPM. These offsets can be
accumulated or built up and maintained over long periods of many
seconds or minutes when they are measured. This provides accurate
measurement. The accumulated offsets or any related value thereto
can be stored, e.g. in an optional location engine. The location
engine may be implemented in the form of a server performing the
calculation of the time difference of arrival at the various
stations.
[0049] For the time difference of arrival measurement, a time
difference of the signal flight times between a user device (not
shown), which may be implemented as a mobile device, and two or
more receiving stations, for instance AP2, AP1, and AP4 may be
calculated.
[0050] In FIG. 1 hyperbolas marked from 2.410 to 2.480 indicate
lines of constant delay differences between AP2 and AP4 whereas
hyperbolas marked from 1.410 to 1.480 mark lines of constant delay
difference between stations AP1 and AP4. As indicated by 110 in the
given example the user's position should be somewhere on hyperbola
120. On the other hand as also indicated by 140 the user's position
should also be somewhere on the hyperbola marked by reference sign
130. Therefore, at the intersection of hyperbolas 110 and 130 the
user's position also indicated by reference sign 140, respectively
the position of the mobile device of the user, can be
identified.
[0051] The embodiment of the present invention exploits the fact,
that the distances between the known stations AP1 to AP4 are known,
as they are in a fixed positional relationship and that based on
the separation distances of the stations between each other, an
accurate propagation time of a correspondingly transmitted signal
can be calculated by taking the distance and the propagation speed.
By establishing a time difference of arrival based on an accurate
calculation allows the establishment of a relationship of this
accurate time difference of arrival to a time difference of arrival
that has been measured by the inaccurate clocks of the two stations
in question. Based on a model of the time dependency of the time
difference of arrival at the two receiving stations, an analysis
can be made to identify a corresponding error value in the time
difference of arrival measured by the inaccurate clocks of the two
receiving stations. Further, based on the accurate calculation the
system is able to compute a correction value and thus to derive an
accurate time difference of arrival at an arbitrary point in time
regarding these two receiving stations. For instance, this can be
achieved by relating the correct values at an arbitrary point in
time to the modeled value at this point in time to calculate the
current correction value. Mathematical operations in this context
are preferably division to relate the values and multiplication to
calculate the corrected time difference of arrival for a signal
received from a user device with the current correction value.
[0052] Such a process may be performed for any two receiving
stations that are part of the configuration shown in the embodiment
of FIG. 1.
[0053] Consequently, according to the present invention no absolute
values are required to be determined and instead only the relative
clock behavior between any two receiving stations needs to be
determined. Of course the time difference of arrival can also be
determined by a user device, which on the other hand requires a
number of further provisions for receptions and forwarding of the
corresponding data to and/or from the user device. In order to
perform an accurate location estimation of a user device a timing
accuracy in the order of 1 ns which corresponds to one third of a
meter is preferably required.
[0054] FIG. 2 gives an example of a clock behavior at two stations
AP1 and AP2 indicated by reference sign 290. On the vertical axis
the local time of the clock at station AP2 is marked by 275 and on
the horizontal axis the time at station AP1 is marked by 265. The
triangles in the diagram indicated by reference signs 215, 220,
230, 235, 240, 245 and 255 represent the reception times of unique
signals at the station AP1 respectively AP2. If the clocks at AP1
and AP2 were performing correctly and very accurately, there would
be no curve connecting the triangles and they would all lie on a
straight line.
[0055] In this case the known propagation time of the signal based
on a calculation of the signal propagation speed and the distance
between respectively AP4 and AP2 and AP4 and AP1 have been
subtracted from the reception times of the signal, which may be a
beacon signal from AP4. As the locations of the station are known
such a calculation can be easily performed by a skilled person. It
is also conceivable, that the curve represented by reference sign
210 connecting the various triangles can be modeled by any suitable
regression or interpolation method, e.g. by a simple polynomial.
However, other suitable modeling techniques are conceivable like
value approximation by a neural network trained with the value
pairs of measured local clocks. Reference sign 250 indicates the
time as an arbitrary point in time at which a signal from a mobile
device of a user is received at the respective stations AP1 and
AP2. A smooth curve between measurements can be obtained by
polynomial curve fit which yields a time difference between AP1 and
AP2 at a user observation time.
[0056] Such a dependency of the inaccurate local clocks is for
instance further explained in the configuration shown in FIG. 3. In
particular here a local clock of station AP2 referenced by 321 and
a local clock of station AP1 referenced by 311 is shown. It is
further indicated, that a signal propagation time between AP4 and
AP2 marked by reference sign T.sub.p4,2 is known based on a
calculation of the distance and the propagation speed, which also
holds true for the propagation time of a signal transmitted between
AP4 and AP1 indicated by T.sub.p4,1. The measurements taken for a
signal transmitted from station AP4 and the associated timing can
for instance be transmitted to a server indicated by reference sign
380 performing the required calculations as well as a suitable
modeling function such as a regression analysis of which a
polynomial curve fit is one example whereas the transmission is
exemplified by an arrow marked by reference sign 350. For instance,
a relative clock behavior may be obtained by over air
measurements.
[0057] For instance, a propagation time correction may be performed
in the following manner. It may be assumed, that there are two
receiving stations AP1 and AP2 at known locations. The clocks of
AP1 and AP2 are ticking in nano seconds and are not synchronized
with each other. Thus they are drifting in time with respect to
each other. On the other hand both AP1 and AP2 are each able to
accurately time stamp the same received signal for instance in form
of a frame from a single transmitting station AP4. It is known
based on the signal propagation speed and the distance between the
stations that a frame transmitted by AP4 propagates 0.3 m in 1 ns
and thus AP4 being 30 m apart from AP1 and 18 m apart from AP2
leads to a propagation delay from AP4 to AP1 which is 30 m/0.3
m/ns=100 ns whereas the propagation delay from AP4 to AP2 is 18
m/0.3 m/ns=60 ns. For instance, AP1 time stamps a frame received
from AP4 with the reading of 629,154,927 ns whereas AP2 time stamps
the same frame received from AP4 with the reading of 402,549,572
ns. In this case AP1 for instance calculates that its clock reading
at the same time that AP4 transmitted the frame was
629,154,927-100=629,154,827 ns, whereas AP2 calculates that its
clock reading at the same time that AP4 transmitted the frame was
402,549,572-60=402,549,512 ns. This leads to the result that at the
instant that the clock of AP1 was reading 629,154,827 ns the clock
of AP2 was reading 402,549,512 ns and vice versa. The calculation
of a known propagation delay can however be subtracted at any time
in the calculation process because it remains a known constant in
case the position of the stations remain constant which is a
prerequisite. Thus a different order in the sequence of the
calculations is always possible. The fitting of a regression curve
such as a polynomial curve to an observed relative AP clock
behavior may be performed in the following manner. Preferably, a
relative clock behavior is monitored over a segment of time
containing n observations that span the instant when a relative
timing of a user observation needs to be established. For instance,
if the recorded n observation times, which eventually need to be
corrected for known propagation times of the same beacon
transmissions, are indicated by T.sub.AP1,i and T.sub.AP2,i, a
polynomial curve may be fitted to the observed data which is, for
example, a least squares fit to the observed data, e.g. in form
of
T.sub.AP2=a.sub.0+a.sub.1 T.sub.AP1+a.sub.2(T.sub.AP1).sup.2.
[0058] Then the coefficients a.sub.0, a.sub.1 and a.sub.2 can
readily be obtained from the observed data using a least squares
matrix technique by:
a=(X.sup.TX).sup.-1 X.sup.T Y
[0059] where a is a column vector consisting of a.sub.0, a.sub.1,
and a.sub.2, Y is a column vector of the n AP2 observations of
T.sub.AP2,1 to T.sub.AP2,n and X is an n.times.3 matrix formed from
the AP1 observations e.g.:
x = ( 1 , T AP 11 , ( T AP 11 ) 2 1 , T AP 1 n , ( T AP 1 n ) 2 )
##EQU00001##
[0060] Further details can be obtained from according to
http://mathworld.wolfram.com/LeastSquaresFittingPolynomial.html.
[0061] In this example a second order polynomial is used wherein:
a.sub.0 represents a fixed time offset; a.sub.1 represents a fixed
frequency offset and a.sub.2 represents a linear frequency drift
with time. Without restricting the invention, it is also possible
to use higher orders of polynomials although a second order
solution is likely to be good enough giving the shape of the curve
shown in FIG. 2 which is not a complicated curve. However, there is
a dependency between the order of the polynomial and a number of
measurement observations that are required for its determination
for a J.sup.th order polynomial at least J+1 measurement
observations are required. Preferably in practice more measurement
observations are needed if the impact of measurement noise needs to
be reduced.
[0062] FIG. 4 shows an example of a position determination process
in a flow-chart according to an embodiment of the present
invention.
[0063] At 405 the process is started. At 410 each station measures
its inaccurate signal arrival time against its own clock for every
signal it receives. At 415 in this embodiment the stations send the
observations meaning the arrival time of the individual signal and
the associated signal identification for instance to a location
server which may be employed to calculate a location. At 420 in
this case the server stores all measurements over a period of time,
for instance for the period of some 10 s of seconds. At 425, in
case of a location request (or alternatively periodically), each
station receiving a suitable signal e.g. a user frame, measures the
arrival time of each particular user frame, representative of a
user signal using its local clock and at 430 sends the observed
user signal measurement and the associated identification as well
as time to the location server. At 435 the server retrieves the
values stored for signal transmission between the wireless stations
corresponding to the wireless stations that have received the user
signal and at 440 corrects the observation regarding the retrieved
communication between the wireless station by subtracting the known
propagation delays based on propagation speed and positional
relationship, respectively known distances. Subsequently at 440 a
regression analysis such as fitting a polynomial is carried out on
the corrected value pairs for each pair of wireless stations which
received the common intercommunication between the wireless
stations. At 450 the server, for instance, uses the polynomials to
estimate the clock offsets of the wireless stations at times
corresponding to the time where the user signal was received. At
455 for each pair of wireless stations at 460 a time difference of
arrival for the user signal is calculated and corrected for the
factor determined on the basis of the stored values from the
corresponding pair of wireless stations and the calculated
polynomial estimate of the clock offset. At 465 the corrected time
difference of arrivals information are transmitted to the location
algorithm and at 470 the estimated user location is calculated by
intersecting the two hyperbolas as shown in FIG. 1 but marked by
reference sign 120 and 130 to determine the user location indicated
by arrow 140 which is retrieved at 495. Marked by 475 such a
process may be repeated to determine various user locations.
[0064] FIG. 5 shows an example of a computer program product
according to the present invention. Reference sign 500 indicates a
data carrier which comprises a program code 520 representative of
any of the method steps according to the present invention. Such a
computer program product constitutes a simple entity to transport
the method of the present invention and to implement it on the
transmitting and receiving stations AP1 to AP4 of the present
invention in case they are equipped with a network interface or a
corresponding data reader. The data carrier of the present
invention may be a suitable data carrier such as a magnetic- or
optical medium or a hardware storage medium such as a flash
storage. It may also be represented by a signal that is transmitted
on a network according to a certain network protocol implemented on
a wired network or on a wireless network for downloading the
program code from one computer to another computer.
[0065] FIG. 6 shows an example of a station that may be employed in
an arrangement according to the present invention. Reference sign
600 indicates a wireless station e.g. an access point or any other
wireless or wire bound device capable of transmitting and/or
receiving unique signals preferably identifiable by a unique
identifier such as e.g. a frame number. It has an input/output
interface 610 of any kind be it mechanical electrical or user
specific for data entry and display. Further the station contains a
receiver 615 and a transmitter 620 for instance capable of emitting
and receiving in the GSM, Bluetooth or WLAN range of frequencies,
or to handle standard packet communication over wire or optical
media. A controller or processor 625 is capable to control the
functions of the station 600 and has the power to perform the
required computations. Further a memory 630 is present which may be
any optical semiconductor or magnetic device for storing
communication and operation data. Reference sign 635 indicates a
power supply which may be a battery or a transformer connected to a
wall outlet. All the internal components are connected by a
suitable system bus 650 ensuring the proper operation of the
station 600. In a similar manner as the station 600 a computation
server 380 may be equipped with one or all of the components 610,
615, 620, 625, 630 and 650 dimensioned in a corresponding manner to
perform a location determination of a user device and the required
communication.
* * * * *
References