U.S. patent application number 16/266517 was filed with the patent office on 2019-06-06 for method to estimate and compensate for nlos bias in time difference of arrival estimate.
This patent application is currently assigned to Red Point Positioning Corporation. The applicant listed for this patent is Red Point Positioning Corporation. Invention is credited to Chunjie DUAN, Georgiy PEKHTERYEV, Zhenzhen YE, Yu Zhao.
Application Number | 20190174332 16/266517 |
Document ID | / |
Family ID | 58237289 |
Filed Date | 2019-06-06 |
![](/patent/app/20190174332/US20190174332A1-20190606-D00000.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00001.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00002.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00003.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00004.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00005.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00006.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00007.png)
![](/patent/app/20190174332/US20190174332A1-20190606-D00008.png)
United States Patent
Application |
20190174332 |
Kind Code |
A1 |
DUAN; Chunjie ; et
al. |
June 6, 2019 |
METHOD TO ESTIMATE AND COMPENSATE FOR NLOS BIAS IN TIME DIFFERENCE
OF ARRIVAL ESTIMATE
Abstract
Devices, systems, and method for compensating for
non-line-of-sight ("NLOS") bias in time difference of arrival
("TDOA") estimates between a first anchor and a second anchor in a
network having an obstacle in the line-of-sight therebetween are
provided. The systems and methods include transmitting a first
packet from a first anchor; indirectly receiving the first packet
by a second anchor, then transmitting a second packet by the second
anchor; indirectly receiving the second packet by the first anchor;
and receiving the first packet and the second packet by a mobile
node. The true fly-time of the first or second packets between the
first anchor and the second anchor and the bias in time of flight
of the first or second packets between the first anchor and the
second anchor are estimated. The time difference of arrival at the
mobile device between a direct path and an indirect path is further
estimated and the NLOS bias in the time difference of arrival
estimated at the mobile device is corrected.
Inventors: |
DUAN; Chunjie; (Brookline,
MA) ; PEKHTERYEV; Georgiy; (Brookline, MA) ;
YE; Zhenzhen; (Brookline, MA) ; Zhao; Yu;
(Brookline, MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Red Point Positioning Corporation |
Brookline |
MA |
US |
|
|
Assignee: |
Red Point Positioning
Corporation
Brookline
MA
|
Family ID: |
58237289 |
Appl. No.: |
16/266517 |
Filed: |
February 4, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15264924 |
Sep 14, 2016 |
10200886 |
|
|
16266517 |
|
|
|
|
62218070 |
Sep 14, 2015 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01S 5/06 20130101; G01S
5/0273 20130101; G01S 5/0226 20130101; H04W 24/02 20130101; G01S
1/20 20130101; G01S 5/0081 20130101; H04W 24/10 20130101; H04W
64/00 20130101 |
International
Class: |
H04W 24/02 20060101
H04W024/02; H04W 64/00 20060101 H04W064/00; G01S 5/02 20060101
G01S005/02; H04W 24/10 20060101 H04W024/10; G01S 1/20 20060101
G01S001/20; G01S 5/06 20060101 G01S005/06; G01S 5/00 20060101
G01S005/00 |
Claims
1. A method for compensating for non-line-of-sight ("NLOS") bias in
time difference of arrival ("TDOA") estimate between a first anchor
and a second anchor in a network having an obstacle in the
line-of-sight therebetween, the method comprising: transmitting a
first packet from a first anchor; indirectly receiving the first
packet by a second anchor, then transmitting a second packet by the
second anchor; indirectly receiving the second packet by the first
anchor; receiving the first packet and the second packet by a
mobile node; estimating the true fly-time of the first or second
packets between the first anchor and the second anchor; estimating
the bias in time of flight of the first or second packets between
the first anchor and the second anchor; estimating the time
difference of arrival at the mobile device between a direct path
and an indirect path; correcting the NLOS bias in the time
difference of arrival estimated at the mobile device.
2. The method of claim 1, wherein the estimating and correcting is
performed by the first and second anchors.
3. The method of claim 1, wherein the estimating of fly-time is
performed by the first anchor.
4. The method of claim 1, wherein correcting the NLOS bias in the
time difference of arrival is performed by the mobile device.
5. The method of claim 1, wherein the estimated bias is
filtered.
6. The method of claim 1, wherein the true locations of the first
anchor and the second anchor are known.
7. The method of claim 1, wherein the estimated bias is embedded in
a packet and transmitted.
8. The method of claim 1, wherein the bias between the first and
second anchors is compensated, according to:
.DELTA.R.sub.AB=t.sub.AB*c-R.sub.AB=.DELTA.t.sup.A/2*c-R.sub.AB,
where t.sub.AB is the time of travel of the first packet from the
first anchor to the second anchor, and wherein R.sub.AB is the
direct path between the first and second anchors.
9. The method of claim 1, wherein the estimated bias between
anchors is used directly in estimating a corrected distance
difference, according to:
.DELTA.R.sup.C.sub.AB=.DELTA.R.sup.M.sub.AB-.DELTA.R.sub.AB wherein
.DELTA.R.sup.M.sub.AB is the original distance difference measured
at the mobile device.
10. The method of claim 1, wherein the true fly-time between the
first and second anchor is embedded in a first packet and
transmitted.
11. The method of claim 1, wherein the true locations of one or
both of the first and second anchors are unknown.
12. The method of claim 1, wherein the NLOS bias between the first
and second anchors is estimated and updated a plurality of
times.
13. The method of claim 1, wherein the NLOS bias between the first
and second anchors is estimated during the initial setup of the
network.
14. The method of claim 1 further comprising, estimating the
position of the mobile device in an external computing device.
15. A method to compensate for non-line-of-sight ("NLOS") bias in a
time difference of arrival estimate, using an uplink time
difference of arrival ("UL-TDOA") scheme, the method comprising:
transmitting a first packet by a mobile device; receiving the first
packet by at least one first anchor and then transmitting a second
packet by the at least one first anchor; receiving the first packet
and the second packet by at least one second anchor in range of the
mobile device and at least one first anchor; estimating the time
differences of arrival at the at least one second anchor; and
correcting the time differences of arrival at the at least one
second anchor by subtracting the NLOS bias between the at least one
first anchor and at least one second anchor.
16. The method of claim 15 further comprising, estimating the
position of the mobile device using the corrected time differences
of arrival.
17. The method of claim 16, wherein estimating the position of the
mobile device is done in an external computing device.
18. A method to compensate for the non-line-of-sight ("NLOS") bias
in a beacon synchronized time difference of arrival ("BS-TDOA")
estimate, the method comprising: transmitting a first packet by a
first anchor; receiving the first packet by a mobile node and then
transmitting a second packet by the mobile node; receiving the
first packet and the second packet by at least one second anchor
that is within range of the first anchor and the mobile node; and
measuring the time differences of arrival between a direct packet
transmission path and an indirect packet transmission path by the
at least one second node; correcting the time differences of
arrival with an estimated bias of fly-time between the first anchor
and at least one second anchor.
19. The method of claim 18 further comprising, estimating the
position of the mobile device using the corrected time differences
of arrival.
20. The method of claim 19, wherein estimating the position of the
mobile device is done in an external computing device.
Description
RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Patent
Application No. 62/218,070, filed on Sep. 14, 2015, and entitled
"Method to Estimate and Compensate for NLOS Bias in Time Difference
of Arrival Estimate," which patent application is hereby
incorporated by reference in its entirety.
FIELD
[0002] The present disclosure relates generally to the localization
systems and more particularly to methods and systems to locating
objects wirelessly using time-of-flight information where there may
be no line of sight between objects.
BACKGROUND
[0003] Time-of-Flight (ToF), also known as Time-of-Arrival (ToA),
is often used to measure the distance between two wireless devices.
The distance R can be easily calculated as fly time t multiplied by
the traveling speed of the signal, i.e., R=c*t. For a radio
frequency signal, c is approximately 3.times.10.sup.8 m/s.
[0004] This direct conversion between time and distance is the
foundation of many ToF based location estimation technologies. If
the ToF is known between a device to be localized (DTBL), referred
as mobile device hereinafter, and multiple devices with known
locations (referred as anchors, or reference nodes hereinafter),
the distances between a mobile device and anchors can be computed
and subsequently the mobile location is estimated using
multi-lateration, or other techniques. The localization based on
ToA has been widely used in many wireless localization systems.
[0005] Given that a mobile node, or device, is generally not
time-synchronized to anchors in a given network, a technique called
Two-Way TOA (TW-TOA) is commonly used to estimate the location of
the mobile device. TW-TOA techniques may require signals to be
transmitted and received by both the anchor and the mobile device.
By doing so, the round trip fly time is measured and the distance
is calculated using the round trip delay as R=c*T/2, where T is the
round trip fly-time. Such an implementation using TW-TOA is
bandwidth and energy inefficient because of the large number of
transmissions needed for each localization operation. A system
using TW-TOA for localization operations often has a significant
capacity limit, i.e., the total number of nodes, or updates are
very limited.
[0006] A more efficient technique for localization than TW-TOA is
based on measuring the Time-Difference-of-Arrival (TDOA). TDOA
estimates the differences in the distance from the mobile device to
a plurality of different anchors. The differences in distance are
calculated by measuring the difference of time when signals arrive
at each receiver anchor, which subsequently determines the flight
time difference. There are a number of methods to realize
TDOA-based locationing, such as downlink TDOA (DL-TDOA), such as
the TDOA system described in U.S. Patent Application Publication
No. 2015/0156746, `Method and System for Estimating the Location of
a Receiving Device,` hereby incorporated by reference, uplink TDOA
(UL-TDOA), such as the UL-TDOA system described in US Patent
Application Publication No. 2015/0185309, `Method and System for
Estimating the Location of a Transmitting Device In a Wireless
Network,` hereby incorporated by reference, and Beacon synchronized
TDOA (BS-TDOA). The present disclosure describes techniques to
improve the performance of the TDOA systems and can be applicable
to all TDOA schemes.
[0007] A DL-TDOA system is illustrated in FIGS. 2(a) and 2(b). At
the start of a positioning ranging process, an anchor 2210 can
transmit a first request (REQ) packet 2208. One or more anchors
2202 can respond to the REQ packet 2208 by each transmitting
response (RSP) packets 2209. A RSP packet 2209 may only be
transmitted by an anchor 2202 after it receives a REQ packet 2208.
A mobile device 2103 can receive both the REQ and the RSP packets
2208, 2209, and can thereby determine time differences of arrival
and thereby estimate its own position.
[0008] A UL-TDOA system is illustrated in FIG. 7. The UL-TDOA
system can include at least one mobile device 3101 and a plurality
of anchors 3101. At the start of a localization process, a mobile
device, M 3103, can transmit a REQ packet 3208. One of the anchors
in range, for example anchor a0 3101 can responds to REQ packet
3208 by transmitting a RSP packet 3209. Other anchors in range, for
example anchors a1-a3 3101, can each receive both REQ 3208 and RSP
3209 packets to determine the time differences of arrival. The
location of the mobile device M 3103 can thus be estimated, as
described in greater detail in U.S. Patent Application Publication
No. 2015/0185309, `Method and System for Estimating the Location of
a Transmitting Device In a Wireless Network.`
[0009] Another scheme using a hybrid TW-ToA and TDOA, referred to
herein as Beacon Synchronized TDOA (BS-TDOA) scheme is illustrated
in FIG. 8. The BS-TDOA system can include at least one mobile
device M 4103 and a plurality of anchors 4101, 4104. At the start
of the process, one of the anchors, for example a0 4104, can
transmit a REQ packet 4208. The mobile device, for example M 4103
can, after receiving REQ packet 4208, transmit a response packet
RSP 4209. RSP packet 4209 can be received by all anchors 4101.
Anchors 4101 can determine the time differences between when each
RSP packet 209 is received. The location of the mobile device 4103
can then be estimated using the arrival time difference, as
described in greater detail in U.S. Pat. No. 8,259,699, `Method and
system for target positioning and tracking in cooperative relay
networks,` hereby incorporated by reference in its entirety.
[0010] In all three TDOA schemes described above, two types of
packets are transmitted, a request (REQ) packet, and one or more
response (RSP) packets. A RSP packet is only transmitted by a
device only after it receives a REQ packet. The difference lies in
the devices used to transmit these packets and to measure the time
difference of the arrivals. [0011] In DL-TDOA, an REQ is sent by an
anchor, and an RSP is sent by another anchor. The time differences
of arrival are measured by a mobile. [0012] In UL-TDOA, an REQ is
sent by a mobile, and one, or more RSP packets are sent by anchors.
The time differences of arrival are measured by anchors. [0013] In
BS-TDOA, an REQ is sent by an anchor, and an RSP is sent by a
mobile. The time differences of arrival are measured by
anchors.
[0014] For current ToA and TDOA localization schemes, the accuracy
of the location estimate will be affected by the presence of the
non-line of sight (NLOS) measurements. In aforementioned cases, the
NLOS measurements can introduce a time delay bias between anchor
and mobile that is a factor that needs to be mitigated.
Additionally, mobile nodes are often moving within an area covered
by a system. As mobile devices are moving through a system, it is
often not possible to avoid the occurrence of NLOS measurements,
though the bias is not always consistent as the mobile device moves
into a more favorable location, the bias will disappear. For TDOA
systems, the NLOS bias between anchor nodes can additionally have
the same effect on accuracy in location determinations. The bias
between anchors, however, is persistent as anchors are fixed and
thus do not change locations.
[0015] In all three cases described above, the time of flight
between anchors is used to estimate the position of mobile devices.
Accurate measurements of time of flight is necessary is for
obtaining accurate position estimates for the mobile devices.
[0016] The bias caused by NLOS packet transmissions can severely
degrade the accuracy of the position estimate of the mobile device.
The bias between anchors can be especially harmful as the bias is
always present for all the mobile devices. The bias can negatively
impact position estimates of all individual mobile devices using
the anchors not within the line of sight of each other.
[0017] To avoid the problems associated with non-line-of-sight
bias, it is common practice to carefully choose only the anchor
pairs that are within Line-of-Sight (LOS) to each other as TDOA
pairs. This, however, can sometimes be difficult to realize,
especially in complicated indoor environments. Even in systems
where it is possible to carefully chose anchor pairs within LOS, it
may require increasing the total number of anchors needed for the
coverage, or may significantly reduce the overall network
efficiency. LOS systems may also require time-consuming, manual
pairing, which can indirectly increase the installation complexity
and cost.
[0018] Accordingly, a need exists for systems and methods that
allow for a reduction in non-line-of-sight signal transmission bias
to enhance position estimates for mobile devices.
SUMMARY
[0019] Systems, devices, and methods are generally provided for
performing location estimates of mobile nodes. More specifically,
the systems, devices, and methods are designed to compensate for
systems where there is limited, or no-line-of-sight between anchor
nodes, or between anchor nodes and a mobile node. In one exemplary
method for compensating for non-line-of-sight ("NLOS") bias in time
difference of arrival ("TDOA") estimate between a first anchor and
a second anchor in a network having an obstacle in the
line-of-sight therebetween, the method includes transmitting a
first packet from a first anchor and indirectly receiving the first
packet by a second anchor. Then the second anchor transmits a
second packet and the first anchors indirectly receives the second
packet. A mobile node receives both the first packet and the second
packet. The true fly-time of the first or second packets between
the first anchor and the second anchor is estimated. Further, the
bias in time of flight of the first or second packets between the
first anchor and the second anchor is estimated. The time
difference of arrival at the mobile device between a direct path
and an indirect path is estimated and the NLOS bias in the time
difference of arrival is corrected at the mobile device.
[0020] In some embodiments the estimating and correcting can be
performed by the first and second anchors. In some other
embodiments, estimating of the fly-time can be performed by a first
anchor. Further, in some embodiments, the mobile device can perform
the correcting NLOS bias in the time difference of arrival. In some
embodiments, the estimated bias can be filtered. The true locations
of the first anchor and the second anchor can be known. The
estimated bias can be embedded in a packet and transmitted.
[0021] In some embodiments the bias between the first and second
anchors can be compensated, according to:
.DELTA.R.sub.AB=t.sub.AB*c-R.sub.AB=.DELTA.t.sup.A/2*c-R.sub.AB.
Where t.sub.AB can be the time of travel of the first packet from
the first anchor to the second anchor, and R.sub.AB can be the
direct path between the first and second anchors. The estimated
bias between anchors can be used directly in estimating a corrected
distance difference, according to:
.DELTA.R.sup.C.sub.AB=.DELTA.R.sup.M.sub.AB-.DELTA.R.sub.AB. Where
.DELTA.R.sup.M.sub.AB is the original distance difference measured
at the mobile device.
[0022] In other embodiments, the true fly-time between the first
and second anchor cam be embedded in a first packet and
transmitted. The true locations of one or both of the first and
second anchors can be unknown. The NLOS bias between the first and
second anchors can be estimated and updated a plurality of times.
The NLOS bias between the first and second anchors can be estimated
during the initial setup of the network. In some embodiments the
method can further include estimating the position of the mobile
device in an external computing device.
[0023] In another exemplary method for compensating for
non-line-of-sight ("NLOS") bias a time difference of arrival
estimate, using an uplink time difference of arrival ("UL-TDOA")
scheme, the method includes transmitting a first packet by a mobile
device; receiving the first packet by at least one first anchor and
then transmitting a second packet by the at least one first anchor.
The first packet and the second packet are received by at least one
second anchor in range of the mobile device and at least one first
anchor. Estimating the time differences of arrival at the at least
one second anchor and correcting the time differences of arrival at
the at least one second anchor by subtracting the NLOS bias between
the at least one first anchor and at least one second anchor.
[0024] In some embodiments the method can include estimating the
position of the mobile device using the corrected time differences
of arrival. Estimating the position of the mobile device can be
done in an external computing device.
[0025] In a further exemplary method to compensate for the
non-line-of-sight ("NLOS") bias, using a beacon synchronized time
difference of arrival ("BS-TDOA") estimate, the method includes
transmitting a first packet by a first anchor, receiving the first
packet by a mobile node and then transmitting a second packet by
the mobile node. The first packet and the second packet are
received by at least one second anchor that is within range of the
first anchor and the mobile node. Measuring the time differences of
arrival between a direct packet transmission path and an indirect
packet transmission path by the at least one second node and
correcting the time differences of arrival with an estimated bias
of fly-time between the first anchor and at least one second
anchor.
[0026] In some embodiments the method can include estimating the
position of the mobile device using the corrected time differences
of arrival. Estimating the position of the mobile device can be
done in an external computing device.
BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The following detailed description is provided with the
accompanying drawings, in which:
[0028] FIG. 1 is a prior art schematic of localization network
based on a downlink TDOA scheme;
[0029] FIG. 2(a) shows a prior art of a schematic of transmission
and reception of wireless signals;
[0030] FIG. 2(b) shows a prior art timing diagram of the
transmission and reception of the wireless signals;
[0031] FIG. 3(a) illustrates signal transmissions with NLOS path
between anchors in a DL-TDOA scheme;
[0032] FIG. 3(b) illustrates the timing with NLOS path between
anchors in a DL-TDOA transmission;
[0033] FIGS. 4(a)-(c) are the flow charts illustrating anchors and
mobile device for a TDOA location operation;
[0034] FIG. 5 is a schematic of a bias compensation for network
based on uplink TDOA scheme;
[0035] FIG. 6 is a schematic of a bias compensation for network
based on beacon synchronized TDOA scheme;
[0036] FIG. 7 is a schematic of a prior art UL-TDOA scheme;
[0037] FIG. 8 is a schematic of a prior art BS-TDOA system
DETAILED DESCRIPTION
[0038] This present disclosure describes systems and methods to
automatically estimate the fly-time bias for anchor pairs and
compensate for it during the position estimate process. As a
result, the system can deploy a TDOA network with anchors installed
in arbitrary locations and without having to worry about
introducing bias from non-line-of-sight packet transmissions.
According to the present disclosure and two anchors within range of
each other can be used as a TDOA pair. The present disclosure can
significantly simplify the network planning process as it allows
the network to be planned manually and at the same time reduces the
total number of anchor devices, increases the location coverage,
and offers more accurate position services.
[0039] FIG. 1 illustrates a prior art Time-of-Flight (TOF) based
locationing system 1100 which consists of a number of anchor
devices 1101 and mobile devices 1103. In a DL-TDOA configuration,
as described in U.S. Patent Application Publication No.
2015/0156746, which is hereby incorporated by reference, anchor
nodes are grouped in pairs (e.g., anchor pairs 1102). Each pair
1102 consists of two anchors 1101 that are within each other's
range. For example, anchors A and B form a pair 1102. Further,
anchors A, B, C, and D can additionally be paired into the
following pairs {B,D}, {C,D}, {C,A}, {B,C} and {A,D} if those nodes
are within each other's range. For TDOA location operation, all, or
some of the anchor pairs can transmit using RF signals 1112.
Alternatively, the present disclosure contemplates that additional
signal types can be used without departing from the spirit of the
present disclosure. The RF signals 1112 are received by the mobile
devices 1103. The reception time of the signals 1112 are estimated
and used to determine the distance difference from the mobile to
anchors respectively. With the distance differences, and the anchor
locations known, the position of the mobile device 1133 can be
estimated.
[0040] The present application assumes that signals transmitted by
radio devices are in the form of packets. Moreover, the anchors,
radio nodes, and mobile devices can be Ultra-Wideband (UWB) radio
devices. Alternatively, it is understood that other signal formats
can be used as long as timing information can be extracted. The
mobile devices, anchors, and other nodes can be formed together as
a single network.
[0041] During a downlink TDOA (DL-TDOA) operation, each anchor
2201, 2202 in an anchor pair {A, B} can transmit one packet. FIG.
2(a) illustrates two anchors 2201, 2202 and one mobile device 2103
in the DL-TDOA operation. The first anchor 2201 can transmit an REQ
packet 2208. The second anchor 2202 can receive the REQ packet 2208
and then can transmit an RSP packet 2209 immediately, or after a
brief delay. A mobile node, or mobile device, 2103 can receive both
the REQ packet 2208 and the RSP packet 2209. The mobile node can
then measure the time delay between the reception of REQ
(t.sup.M.sub.REQ) and RSP (t.sup.M.sub.RSP),
.DELTA.t=t.sup.M.sub.RSP-t.sup.M.sub.REQ.
[0042] The time-of-flight of the signals are illustrated in FIGS.
2(a) and 2(b). The REQ packet 2208 travels by distance R.sub.AM
2213 to reach a mobile device 2103. The REQ packet 2208
additionally travels by distance R.sub.AB 2212 to reach anchor B
2202. Upon receiving REQ packet 2208, anchor B 2202 transmits a
packet RSP 2209. The transmitted signal RSP 2209 travels by
distance R.sub.BM 2223 to reach the mobile device 2103. The path
from anchor A 2201 to mobile M 2103 is defined as the `direct
path`, and the path from anchor A 2201 to mobile M 2103 via a
second anchor 2202 (e.g., anchor B) as the `indirect path`. The
distance difference between the direct path and indirect path is
.DELTA.R=(R.sub.AB+R.sub.BM)-R.sub.AM. The distance difference from
the mobile device 2103 to different anchors, e.g.,
.DELTA.R.sup.M.sub.AB=R.sub.BM-R.sub.AM, can be used to estimate
the position of the mobile device. The distance difference is
estimated as follows:
.DELTA.R.sup.M.sub.AB=R.sub.BM-R.sub.AM=(R.sub.AB+R.sub.BM)-R.sub.AM-R.s-
ub.AB=.DELTA.t*c-R.sub.AB (1)
where .DELTA.t (time difference of arrival) is the time elapsed
from reception of REQ 2208 to the reception of RSP 2209 measured by
the mobile device node M 2103; R.sub.AB, and .DELTA.t can be
computed based on the known locations of anchor A and B. In case
the turnaround time at the second anchor device 2202 is non-zero,
it is subtracted from the measured time difference as well.
[0043] FIG. 2(b) shows the timing diagram of the packets. Where
t.sub.AM is the fly-time of signal from anchor A 2201 to mobile
device M 2103 directly (A.fwdarw.M), and where t.sub.AB, t.sub.BM
are the fly-times of the signals via the indirect path
(A.fwdarw.B.fwdarw.M). Again for simplicity, the turnaround time is
assumed to be zero.
[0044] Equation (1) can be used to accurately estimate the distance
difference .DELTA.R.sup.M.sub.AB. However, as shown in FIG. 3(a) a
bias can be present in the location estimate when the anchors 201,
202 are not in line-of-sight (LOS) from each other. In the case
where the LOS path R.sub.AB 112 between anchor A 201 and anchor B
202 is not available, the signal may reach anchor B 202 via a
non-line-of-sight (NLOS) path. As shown in FIG. 3, the NLOS path
consists of two segments R.sub.AC 311 and R.sub.BC 312 between
anchors and a reflector 333 due to the presence of a physical
obstacle 300. Obstacle 300 can be, for example, any of a wall, a
display, goods, etc. The total travel distance of the signal is
greater than the direct LOS path, i.e.,
R.sub.AC+R.sub.BC>R.sub.AB. As a result, the overall travel from
the indirect path has a positive bias
.DELTA.R.sub.AB=R.sub.AC+R.sub.BC-R.sub.AB. This is reflected as a
positive bias in the measured time difference of arrival
(TDOA).
[0045] In the above discussed NLOS situation, if equation (1) is
used directly, this bias .DELTA.R.sub.AB will be included in the
overall time difference measurement .DELTA.t, and as a result, the
position estimate can be significantly degraded. Estimating the
bias .DELTA.R.sub.AB and compensating for it can therefore improve
the position estimate accuracy.
[0046] FIG. 3(b) shows the timing diagram of the packet
transmission and reception at the anchors 201, 202 and the mobile
device 103. FIG. 3(b) additionally illustrates how to estimate the
bias .DELTA.R.sub.AB. For example, the sequence of packet
transmissions is unchanged, i.e., the first anchor A 201 transmits
a REQ packet 208, the second anchor 202 receives the first REQ
packet 208. The second anchor then transmits a RSP packet 209. The
mobile device 103 receives the REQ packet 208, from anchor A 201,
and the RSP packet 209, from anchor B 202. The mobile device 103
can then measure the difference between the direct path, and the
indirect path via the second anchor 202.
[0047] To estimate the bias, the first anchor 201 measures the time
elapsed between the transmission of the REQ packet to the reception
of the RSP packet. The first anchor 201 can estimate the round trip
fly-time of the signal via the NLOS path, as
.DELTA.t.sup.A=2*t.sub.AB. The fly-time from anchor A 201 to anchor
B 202 is t.sub.AB=.DELTA.t.sup.A/2. The distance bias is then
computed as
.DELTA.R.sub.AB=t.sub.AB*c-R.sub.AB=.DELTA.t.sup.A/2*c-R.sub.AB
(2)
[0048] Once the estimated bias .DELTA.R.sub.AB is calculated,
anchor A 201 can broadcast this information. The estimated bias can
be embedded into the following REQ packet sent by the anchor A 201.
Assuming that anchors 201, 202 are stationary and their locations
do not change over time, the bias between an anchor pair does not
change. The anchors 201, 202 can improve the accuracy of the bias
estimate by applying filtering to the estimated bias. The bias
between anchor pairs {A, B} can be measured continuously, or during
the initial network setup. Each anchor 201, 202 can store the bias
estimates of its neighboring anchors. The estimated bias, when
available, can be included in the REQ or RSP packets 208, 209.
[0049] The mobile node, after receiving the bias .DELTA.R.sub.AB,
corrects the original TDOA measurement with the bias, as
follows
.DELTA.R.sup.C.sub.AB=.DELTA.R.sup.M.sub.AB-.DELTA.R.sub.AB (3)
[0050] As we can see,
.DELTA.R.sup.C.sub.AB=.DELTA.R.sup.M.sub.AB-.DELTA.R.sub.AB=.DELTA.t*c-(R-
.sub.AB+.DELTA.R.sub.AB). Equation (3) can be rewritten as
.DELTA.R.sup.M.sub.AB=(.DELTA.t-.DELTA.t.sup.A/2)*c (4)
[0051] Equation (4) expresses the relationship that the distance
difference can be estimated using the measured time difference of
arrival at a mobile device, and the measured flight time between
anchors. It is not necessary to know the anchor locations to solve
for the distance difference.
[0052] FIGS. 4(a)-4(c) illustrate flow diagrams describing the
DL-TDOA scheme from the first anchor (initiating anchor), the
second anchor (responding anchor) and the mobile device. The flow
for each of the nodes, e.g. anchor A, anchor B, and the mobile
device, can be summarized as follows: [0053] Initiating anchor
(anchor A) [0054] transmits REQ packet 402, [0055] receives RSP
packet 404, [0056] estimates the round trip delay .DELTA.t.sup.A
and computes the distance bias .DELTA.R.sub.AB 406, [0057]
processing the bias estimate 408, and [0058] broadcasts the bias
(e.g., in the following REQ packet) 410. [0059] Responding anchor
(anchor B) [0060] receives REQ packet 412 and [0061] transmits RSP
packet 414. [0062] Mobile node (mobile device) [0063] receives REQ
and RSP packets 416, [0064] receives anchor bias .DELTA.R or
fly-time .DELTA.t.sup.A 418, [0065] measures fly-time difference
.DELTA.t between the direct path and indirect path 420, [0066]
correct the estimated fly-time difference with the anchor bias
.DELTA.R 422, and [0067] estimates position using corrected
distance differences for multiple anchor pairs 424.
[0068] The above described methods are applicable for other TDOA
schemes. Except as described below, or as will be readily
appreciated by one having ordinary skill in the art, the anchors
201', 202', the mobile device 103', and the physical obstacle 300'
are substantially similar to the anchors 201, 202, mobile device
103, and the physical obstacle 300 described above. A detailed
description of the structure and function thereof is thus omitted
here for the sake of brevity. For example, in the case of UL-TDOA,
a mobile device 103' can transmit a REQ packet 208' to all anchors
within range of the mobile device 103'. One or more of the anchors,
e.g. 201', 202', upon receiving the REQ packet 208', can transmit
RSP packets 209'. Anchors 201', 202' receive the RSP packets 209'.
FIG. 5 illustrates the signal path and packet transmission between
the two anchors A and B and a mobile device 103' during the UL-TDOA
operation. As shown, mobile device 103 initially transmits a REQ
packet 208'; then anchor A 201 receives the REQ packet 208'. Once
anchor A 208' receives the REQ packet 208', anchor A 208' can then
transmit a RSP packet 209'. Subsequently, anchor B 202 can receive
both the REQ and RSP packets 208, 209. As shown in FIG. 5, a
physical obstacle 300' between at least anchors A and B 201, 202,
that creates an NLOS bias between the anchors. The distance
difference between the mobile device 103' to the two anchors 201',
202' is calculated as
.DELTA.R.sup.M.sub.AB=R.sub.BM-R.sub.AM=R.sub.BM-(R.sub.AM+R.sub.AB)+R.s-
ub.AB=.DELTA.t*c-R.sub.AB (5)
where .DELTA.t is the measured time elapsed from the reception of
REQ packet 208' to the reception of RSP packet 208' at anchor B
202'.
[0069] The NLOS bias has similar impact on the overall estimate of
distance difference as described above. The NLOS bias can be
corrected similarly, provided the bias is measured. The anchors
201', 202' can perform the bias measurements prior to the UL-TDOA
operation, for example during the initialization of the system.
Alternatively, the anchors 201', 202' can perform the bias
measurements by transmitting an additional packet from anchor B
202' back to anchor A 201' at any time.
[0070] If the true fly-time between anchors is known, the NLOS bias
can be compensated for. To compensate for the NLOS bias, equation
(5) can be rewritten as
.DELTA.R.sup.C.sub.AB=R.sub.BM-R.sub.AM=.DELTA.t*c+(R.sub.AB+.DELTA.R.su-
b.AB) (6)
or
.DELTA.R.sup.C.sub.AB=R.sub.BM-R.sub.AM=(.DELTA.t+.DELTA.t.sup.A/2)*c
(7)
[0071] In a further alternative system, the above described bias
compensation can be applied to BS-TDOA scheme. Except as described
below, or as will be readily appreciated by one having ordinary
skill in the art, the anchors 201'', 202', the mobile device 103'',
and the physical obstacle 300'' are substantially similar to the
anchors 201, 202, mobile device 103, and the physical obstacle 300
described above. A detailed description of the structure and
function thereof is thus omitted here for the sake of brevity. In a
BS-TDOA scheme, as shown in FIG. 6, a first anchor, or beacon, A
201'' can transmit a REQ packet 208''. The mobile device 103'' can
receive the REQ packet 208'' and in response can transmit a RSP
packet 209''. Other anchors within range, e.g. anchor B 202'', can
receive both the REQ packet 208'' from anchor A 201'', and the RSP
packet 209'' from the mobile device 103''. FIG. 6 illustrates one
example that includes only one additional anchor B 202''. The
locations of both anchors 201'', 202'' is known, therefore anchor B
202'' can estimate the distance difference between the direct path
(A.fwdarw.B) and the indirect path (A.fwdarw.M.fwdarw.B) based on
the time difference of arrival measured at anchor B 202'' as
.DELTA.t,
.DELTA.t*c=R.sub.BM+R.sub.AM-R.sub.AB (8)
or
R.sub.BM+R.sub.AM=.DELTA.t*c+R.sub.AB (9)
[0072] If the path between the anchor A 201'' and the anchor B
202'' is non-line-of-sight, (NLOS), a bias .DELTA.R.sub.AB is
present in the total distance traveled by the signal. This bias is
present in the measurement of time of flight.
[0073] The correction of the NLOS bias can be applied as
R.sub.BM+R.sub.AM=.DELTA.t*c+(R.sub.AB+.DELTA.R.sub.AB) (10)
or
R.sub.BM+R.sub.AM=.DELTA.t+.DELTA.t.sup.A/2)*c (11)
[0074] Again, the distance, or time, bias between two anchors in
systems where no line of sight exists between nodes can be
estimated offline prior to the BS-TDOA transmissions, or during the
BS-TDOA operation by letting the second anchor B 202'' transmitting
an additional packet. Anchor A 201'' can then receive this
additional packet, that itself transmitted, and then estimates the
round trip delay bias.
[0075] The systems and methods described herein can effectively
compensate the bias in the time of flight estimation in NLOS
channels between anchors. With the calculated bias, the estimated
time of flight can be significantly reduced and subsequently, the
position estimate based on the time-of-flight or
time-difference-of-arrival is more accurate. Using this technology,
anchors in a real time location system can be used in buildings or
locations where non-line-of-sight conditions exist while
maintaining high accuracy of position estimates of mobile devices
based on TDOA schemes.
[0076] Although systems and methods have been described by way of
examples of preferred embodiments, it is to be understood that
various adaptations and modifications may be made within the spirit
of the scope of the concepts described herein.
* * * * *