WO2006098666A1

WO2006098666A1 - Method and device for assisted satellite-based positioning

Info

Publication number: WO2006098666A1
Application number: PCT/SE2005/000395
Authority: WO
Inventors: Ari Kangas; Janos TOTH-EGETÖ
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2005-03-17
Filing date: 2005-03-17
Publication date: 2006-09-21

Abstract

Methods and devices to solve the ambiguity problem of assisted satellite signal based positioning without suitable apriori location are presented. The method is based on frequency shift measurements (212) of the received satellite signals. The frequency or Doppler measurements are directly associated with location coordinates and location coordinate change rate determinations (220). By treating mobile terminal location coordinates as unknown parameters, mobile terminal location coordinates can be estimated from the Doppler measurements. Such estimated mobile terminal location coordinates can then be used as an initial position estimate in conventional assisted satellite signal positioning (230) based on truncated pseudorange measurements (210).

Description

METHOD AND DEVICE FOR ASSISTED SATELLITE-BASED

POSITIONING

TECHNICAL FIELD

The present invention relates in general to positioning of mobile equipment by use of satellites and in particular to such positioning assisted by land based communication nodes.

BACKGROUND

The possibility to determine a geographic position of an object, equipment or a person carrying the equipment has been more and more requested in recent years. Different types of applications, such as position determination for assistance in emergency situations, guiding assistance, position dependent services or pure surveillance can make use of such position determinations. One often used approach to achieve the positioning is to use well-defined signals emitted from satellites having well-known predetermined positions. Well-known examples of such systems are the Global Positioning System (GPS) and the coming GALILEO system. The position is typically given with respect to a specified coordinate system and is typically obtained by a triangulation/trilateration procedure based on a plurality of received satellite signals.

A stand-alone GPS receiver can obtain full locking to GPS satellite signals, without having any other information about the system except nominal carrier frequency and the rules by which data carried by the signals are modulated. The signals include a so-called Coarse /Acquisition (C/A) code that is unique for each satellite and repeats itself every 1 ms. Superimposed on the C/A code is a navigation data bit stream with a bit period of 20 ms.

Basically, the three-dimensional position as well as a receiver clock bias to the satellite time has to be determined in the position calculation step. However, a stand-alone GPS receiver that starts without any additional information normally needs to decode the complete navigation data stream before the receiver location can be calculated. This typically requires a long time to obtain a location estimate, relatively powerful computational resources as well as fairly large power supply, at least if the GPS receiver is powered by batteries adapted for portable equipment. Furthermore, a standalone GPS receiver typically requires certain minimum signal strength.

In order to solve some of those problems, Assisted GPS (AGPS) has been defined as an enhancement of GPS. Efforts have been taken for integration of GPS receivers into user equipment, i.e. mobile stations, of cellular communication systems. This invention is in general related to methods for locating user equipment (UE) by use of Assisted GPS. Assisted GPS in general aims at improving the performance of GPS receivers in many different respects, including detection sensitivity, time to obtain a location estimate, accuracy and saving battery power. This is done by moving some functionality from the GPS receiver in the mobile station to a network node in communicational contact and hence only performing a subset of the GPS tasks in the GPS receiver itself.

There are basically two types of AGPS, Mobile Station (or User Equipment) based and Mobile Station (or User Equipment) assisted. In Mobile Station based AGPS, the location of a mobile station is actually calculated in the mobile station. The calculation uses ranging signal measurement results determined by the mobile station itself together with assistance data provided by the network.

In Mobile Station assisted AGPS, also referred to as Network based AGPS, the mobile station only measures and reports timing of received ranging signals to a node, typically in a mobile communications network. The timing of the ranging signals reflects the pseudoranges to the Space Vehicles, i.e. satellites.

The position determination is then calculated in the nodes, using the reported timing as well as assistance data. For both types of AGPS, the receiver can determine the boundaries of the C/A code at a much lower signal strength than the one required to decode the entire navigation messages. The measured timing of the ranging signals is therefore normally truncated modulo 1 ms. This means that the μs timing information is completely known, but with an uncertainty of ±n ms, where n is an integer. 1 ms corresponds to a distance of 300km. When calculating the mobile station location, either in the mobile station itself or in a network location server, the complete pseudoranges need to be reconstructed. This is typically done using apriori information about the mobile station location together with the ranging signal measurement results determined by the mobile station. The complete pseudoranges are then used in order to compute the precise mobile station location. It can be shown that such a reconstruction can be unambiguously done when the apriori location uncertainty is less than 75km.

The present invention addresses a problem of range measurement reconstruction in cases where prior art initial location uncertainty is too large for prior art solutions to work properly. This is likely to happen e.g. in "user plane" solutions, where a cell location might not be known to a location server. In a typical case the only knowledge about the UE location may be the country it is currently in. This means that the uncertainty may be in the order of 1000km or more.

For UE based AGPS it is possible to circumvent this problem by determining the complete pseudoranges to all SVs directly by reconstructing the millisecond part of the SV transmit times. There are several known ways to do this in the art. However, for many UEs, the computational and power requirements are too high. Moreover, for UE assisted AGPS it is today only possible to report the millisecond part of the GPS system time at reception. Therefore, UE assisted AGPS will most likely not work in a roaming user plane scenario in cases where only apriori location known is in which country the mobile is. A second solution is to require that the UE measures not only the C/A code boundaries, but also the GPS data bit edges, which occur every 20ms. This way, the uncertainty interval can be stretched to 1500km which is sufficient for most countries. However it has the drawback that for UE assisted AGPS it requires standard changes and will thus not work on mobile stations that already exist in the market today.

SUMMARY

A general problem with prior art assisted satellite signal positioning is that an ambiguity appears when an apήori position is known with a too small accuracy.

A general object of the present invention is thus to provide improved methods and devices for assisted satellite signal positioning. A subsidiary object of the present invention is to provide a method and device enabling assisted satellite signal positioning essentially without any aprioή knowledge of the position.

The above object is solved by methods and devices according to the enclosed patent claims. In general words, the present invention outlines a method to solve the ambiguity problem based on frequency measurements of the received satellite signals. The frequency or Doppler measurements are directly associated with location coordinates and location coordinate change rate measurements. By treating mobile terminal location coordinates as unknown parameters in relations modelling Doppler shifts, mobile terminal location coordinates can be estimated from the Doppler measurements. Such estimated mobile terminal location coordinates can then be used as an initial position estimate in conventional assisted satellite signal positioning.

An advantage with the present invention is that the presented solution relies on existing measurements and is computationally simple. It provides an accuracy, which in the large majority of cases is sufficient as a starting point for more accurate position calculation based on conventional truncated pseudorange measurements.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention, together with further objects and advantages thereof, may best be understood by making reference to the following description taken together with the accompanying drawings, in which:

FIG. 1 is a schematic block diagram illustrating a communications system and satellite positioning system in which the present invention could be applied;

FIG. 2 is a schematic diagram of the time structure of GPS satellite signals;

FIG. 3 illustrates positioning using satellite signals when a full determination of propagation times is performed;

FIG. 4 illustrates the ambiguity problem that might occur when truncated satellite signals are measured;

FIG. 5 illustrates the maximum allowable inaccuracy of apήori locations usable in assisted satellite signal positioning; FIG. 6 illustrates range rate definitions in Doppler measurements;

FIG. 7 is a diagram of probability of location determination accuracy obtained by Doppler measurements;

FIG. 8 is a flow diagram of main steps of an embodiment of a method according to the present invention; FIG. 9 is a flow diagram of main steps of another embodiment of a method according to the present invention;

FIG. 10 is a flow diagram of presently preferred substeps of a step of the embodiments of Fig. 8 or 9;

FIG. 1 1 is a flow diagram of presently preferred substeps of another step of the embodiments of Fig. 8 or 9;

FIG. 12 is a flow diagram of main steps of yet another embodiment of a method according to the present invention; FIG. 13 is a flow diagram of main steps of an embodiment of a mobile terminal based method according to the present invention;

FIG. 14 is a block diagram of main parts of an embodiment of a system according to the present invention; FIG. 15 is a block diagram of main parts of another embodiment of a system according to the present invention; and

FIG. 16 is a block diagram of main parts of yet another embodiment of a system according to the present invention.

DETAILED DESCRIPTION

In the present disclosure, the following abbreviations are frequently used: GPS - Global Positioning System

AGPS - Assisted GPS SV - Space Vehicle

UE - User Equipment

In the detailed description below, embodiments implemented in a GPS system are illustrated. However, anyone skilled in the art realizes that the corresponding principles can be applied in any satellite signal based positioning system, such as the GLONASS or the coming European Galileo satellite navigation system.

Likewise, in the detailed description below, WCDMA systems will be used as model systems. However, the present invention is also applicable on other wireless communication systems. Non-exclusive examples of other systems on which the present invention is applicable are e.g. CDMA-2000 systems and GSM systems including GSM/GPRS systems. When applied to other wireless communication systems, the implementation of the different functionalities will be done in different terminals and nodes of such systems.

The term mobile terminal is in the present disclosure used to denote any kind of terminal that can be transported within a wireless communication system. Non-exclusive examples are cellular phones, personal digital assistants and portable computers.

Fig. 1 is a schematic view of an example communication system in which the present invention may be used. In this example scenario a basic wireless communication system 2 is used together with a GPS system 3 to provide mobile station assisted AGPS.

The example wireless communication system 2 illustrated in Fig. 1 is a WCDMA system, comprising a radio access network (RAN), e.g. UTRAN, and a core network 8. The RAN performs radio-related functions and is responsible for establishing connections 32 between user equipment/ mobile terminals 10 and the rest of the network. The RAN typically contains a large number of Base Transceiver Stations (BTS) 11 , also referred to as Node B, and Radio Network Controllers (RNC) 6. Each BTS serves the mobile terminals within its respective coverage area and several BTS are controlled by a RNC. Typical functions of the RNC are to assign frequencies, spreading or scrambling codes and channel power levels. The RNC 6 provides access to the core network 8, which e.g. comprises switching centres, support nodes and databases corresponding to those of a GSM/GPRS core network, and generally also includes multimedia processing equipment. The core network communicates with external networks, such as the Internet 4, and PSTNs, ISDNs and other PLMN 9.

In practice, most WCDMA networks, as well as most other wireless communication networks, present multiple network elements and nodes arranged in much more complex ways than in the basic example of Fig. 1.

In Fig. 1 , the GPS system 3 is represented by space vehicles (SV), i.e. typically satellites 20. Each SV transmits a respective ranging signal 30, and by using a number of such ranging signals 30, the position of the mobile terminal 10 or of the person carrying the mobile terminal can be determined. When a positioning request occurs, e.g. in the core network 8 of the wireless communication system 2, the positioning request is provided to a positioning node 7 associated with the RNC 6. The RNC 6 orders measurements of satellite ranging signals through control signals sent to the mobile terminal 10. The measurement order is accompanied by assistance data. The assistance data comprises satellite position data and satellite time reference data and can for example be retrieved by the RNC 6 via a reference receiver (not shown) connected to the communication system 2. Such a reference receiver can be provided as one unit or divided in parts, separating the determination of the satellite time reference and the satellite position data.

The assistance data is typically processed in the positioning node 7, which can e.g. determine which satellites that are in such positions that their ranging signals can be detected, before being sent to the mobile terminal 10.

The mobile terminal 10 is equipped with a receiver that is capable of detecting satellite ranging signals 30 and the terminal 10 uses the assistance data to facilitate the locking on and measuring of the satellite ranging signals 30. The measured ranging signals 30 are then used to calculate a position of the mobile terminal 10. If mobile station based AGPS is used, the processing of the ranging signals 30 is performed in the mobile terminal 10. If mobile station assisted AGPS is used, the ranging signals 30 or representations thereof are instead sent to a network node, typically the positioning node 7 in the illustrated scenario, where the positioning calculations are performed based on the reported measurement results from the mobile terminal 10 and on a priori information on where the mobile terminal 10 is located. In the example of Fig. 1 , the positioning functionality is provided in an external positioning node 7 connected to the RNC 6. The positioning node can for example be a stand-alone AGPS SMLC. Alternatively, the positioning functionality could be incorporated in the RNC 6 or in another network node.

In an alternative embodiment, the positioning node does not have to belong to the wireless communication system 2 itself. In Fig. 1 , an external positioning node 5 is depicted, which via e.g. the Internet 4 can be in communicational contact with the mobile terminal 10. Such an external positioning node 5 may also support the mobile terminal 10 by assistance data, typically over the user plane. A difficulty in such a system is that such an external positioning node 5 does normally not have access to any coarse positioning by the communications system 2.

Current GPS SVs transmit ranging signals with a spectrum centred at 1575.42 MHz. The signals include a C/A-code unique for each SV and which typically has a length of 1023 chips and a chip duration of l / 1.023xl0⁶ s.

The C/A code repeats itself every 1 ms. Superimposed on the C/A code is a navigation data bit stream with a bit period of 20 ms. An example of such a ranging signal 30 is illustrated in Fig. 2. Here the C/A code is denoted by 99 and the navigation data bit stream by 98. Among other things, the navigation data includes a set of ephemeris parameters that enables calculation of the precise position of the satellites at the time of signal transmission. The navigation data is divided into a number of subframes, each of length 6 seconds and each subframe is divided into 10 words. A time stamp, GPS Time Of Week (TOW), is transmitted in the second word (HOW) of every subframe. The indicated time is the time of transmission at the end of the subframe in question. The TOW is thus repeated every 6 seconds.

Each ranging signal basically defines a clock which is measured by the mobile station. The clock indicates the time of signal transmission. If the mobile station knows the GPS system time, then the clock reading can directly be used to determine the time delay , and hence the range from the SV transmitting the ranging signal to the mobile station. By measuring three ranges and utilizing the knowledge about Space Vehicle locations at time of transmission, the location of the mobile station can then be determined in three dimensions. However, normally the mobile station does not have knowledge about precise GPS system time, so one more measurement is needed to eliminate the mobile station clock bias. This is schematically illustrated in Fig. 3. A mobile terminal 10, residing at the earth surface 1, receives ranging signals 3OA, 3OB from satellites 2OA, 2OB. For simplicity, since the drawing is 2-dimensional, only two satellites are illustrated. By measuring the time of arrival of each ranging signal 3OA, 3OB, and knowing the emission times, a propagation time can be calculated. The propagation time in turn corresponds to a certain distance rtA, rtB, which spans a sphere 25A, 25B around each satellite 20A, 2OB. If no uncertainties would be present, the mobile terminal 10 position can be determined as the intersection 35 of the spheres 25A, 25B.

As mentioned above, in assisted GPS, the timing of the C/A code is measured. The measured timing of the ranging signals is thereby truncated modulo 1 ms. However, since the entire navigation messages are not detected, the mobile terminal 10 is unaware of which C/A it detects. This means that an ambiguity in determining the position is introduced. With reference to Fig. 2, the mobile terminal 10 can not distinguish whether it is measuring on C/A code 99, 99' or 99".

In Fig. 4, this ambiguity is illustrated for a simplified case where any clock bias considerations are neglected. As before, a mobile terminal 10 receives ranging signals 3OA, 3OB from two satellites 2OA, 2OB. The detected timing corresponds to a sphere around the corresponding satellite. However, since an offset of n times 1 ms may be present in the timing determination, such a truncated pseudorange C/A code measurement instead corresponds to a set of congruent spheres 25A: l-5, 25B: l-4. The mobile terminal 10 position could, when only such information is available, be determined in every intersection 35 between any of the spheres 25A: l-5, 25B: l-4. However, by using external knowledge about an approximate position, e.g. from the cellular communication network, such ambiguity can be resolved. If the mobile terminal 10 is apriori known to be present within a certain cell 40 of the communication network, there is only one or at least a limited number of possible intersections 35 that could be interpreted as the true position. However, if the apriori position knowledge is too inaccurate, e.g. as the illustrated area 45, there are too many intersection 35 alternatives to be handled.

In practical cases, clock bias is always present. In order to find possible solutions, differences between pairs of truncated pseudoranges can be calculated, which geometrically are represented by hyperboloids instead. This hyperboloid has to cut the original cell area to be allowable. However, the reasoning becomes analogue to the above described simplified model.

A reconstruction can be unambiguously done in AGPS when the initial location uncertainty is less than 75km. This can be explained as follows by using the illustration in Fig. 5. Let tsvA and tsvB denote the observed clocks from SV 2OA and SV 2OB, respectively, at a specific time t0 and at a specific location xθ.

The mobile terminal assisted AGPS receiver measures the SV 2OA clock tsvA completely but only the fractional (submillisecond) part of the SV 2OB clock tsvB. The measurement is done at (the unknown) time tθ. The question is now: Under what conditions is it possible to reconstruct unambiguously the integer millisecond part of the clock tsvB. tsvA and tsvB at the tentative mobile terminal locations 1OA and 1OB is now calculated: At mobile terminal 1OA:

tsvA = tO-(dA-Δ)/c (1) tsvB = tO-(dB+Δ)/c (2)

In (1) and (2), c is the speed of light. Subtracting (1) from (2) and rearranging, one gets:

tsvB = tsvA + (dA-dB-2Δ)/c (3)

Similarly at mobile terminal 1OB: tsvA = tO-(dA+Δ)/c (4) tsvB = tO-(dB-Δ)/c (5)

Subtracting (4) from (5) and rearranging, one gets:

tsvB = tsvA + (dA-dB+2Δ)/c (6)

Combining (3) and (6) we see that tsvB lies in the interval:

tsvB e (tsvA + (dA-dB-2Δ)/c, tsvA + (dA-dB+2Δ)/c) (7)

The size of this interval is 4Δ/c. In order to reconstruct the integer millisecond part of tsvB uniquely it is required that the interval is less than lms. Hence:

4Δ/c < 0.001 (8)

which leads to the requirement:

Δ < c*0.001/4 -75 km (9)

For so-called "control plane" solutions to AGPS, the aprioή location typically consists of the cell identity of the serving cell, c.f. area 40 in Fig. 4. The size of cells is typically less than 75km so this presents no problem in a majority of cases. Only in so-called extended range cells some ambiguities could remain. However, in "user plane" solutions to AGPS, the aprioή location can be unknown within far larger limits.

According to the present invention, an approximate location of the mobile terminal is estimated using Doppler measurements on the ranging satellite signals. This approximate solution is then used to limit the apriori location and serve as a starting point for the precise position calculation based on conventional pseudoranges.

Doppler measurements are performed inherently when analyzing the received pseudorange signals. For the so-called UE-assisted variants of

AGPS, Doppler shifts are also reported to a network location server as mandatory report according to present standards. These signals can also be further used by the positioning node. The original intent of such Doppler measurements was to enable the mobile terminal velocity to be determined. In such a procedure, it is then assumed that the location of the mobile terminal is first determined using code phase measurements. The mobile terminal velocity estimation is then reduced to a linear estimation problem.

In the present invention Doppler measurements are used for computing an approximate location of the mobile terminal, and, if needed, also the velocity of the mobile terminal. Use of Doppler measurements for solving the original ambiguity problem has to our knowledge not been presented in prior art. Determining an approximate location of the mobile terminal directly from the Doppler measurements has not been considered an important option previously, since the resulting accuracy is much poorer than the location estimate based on pseudorange measurements. For stand-alone GPS there is no need for any semi-accurate initial estimate, since the complete pseudoranges are determined. The position calculation function is in that case not very sensitive to initial location errors. The need for a semi-accurate initial estimate is most likely only relevant for the AGPS variants, since it is only there that truncated measurements are performed. Although not explicitly stated in the following detailed description, the invention is equally applicable for both mobile terminal based and mobile terminal assisted variants of AGPS.

Fig. 6 illustrates the relations between an ϊth satellite 2Oi and the mobile terminal 10. The mobile terminal 10 is moving by a velocity v and the ith satellite 2Oi is moving by a velocity v_SVi. The position of the mobile terminal 10 as referred to the earth centre 0 is r, and the position of the ith satellite 20i is r_Svi. The path that the satellite signal 30 follows then corresponds to r-rsvi. The ith satellite 2Oi emits the satellite signal 30 at a frequency f, and the mobile terminal receives a signal at the Doppler shifted frequency of f+Δf.

A relative range rate as observed by the mobile terminal 30 is the projection of the relative velocity vector on the unit length line of sight vector from the mobile terminal to the ith satellite, i.e. r_SVi-r. This can be expressed as:

P₁ = —^— (r_5vl - r). (v_JV, - v), i=l,..., n_sat (10) ^r,w - ^r

where all position and velocity vectors are expressed in a Cartesian coordinate system such that r = (x,y,z), r_sw = (x_Sw,ysw,z_SVi), v = (v_x,v_y,v_z), v_sυι = (vxsm, Vysυi, Vzsυi), and where • denotes scalar product. The measured range- rate is corrupted by noise and also contains a constant offset caused by the mobile terminal oscillator.

P_m = P, + b + e, (11)

The goal is to estimate the approximate mobile terminal location with an accuracy which is hopefully better than any previous apriori location uncertainty. A few things need to be considered.

The user location as well as the velocity are unknown and need to be estimated. If we assume a stationary mobile terminal but the mobile terminal is in fact moving, a huge error will occur. For a 3D-location and 3D velocity vector estimate, range rate measurements from 7 satellites are needed. This rather large number of measurements is normally not easy to obtain. However, it is possible to reduce the search by e.g. assuming an average altitude and not estimate the vertical velocity component. This way the number of parameters (and hence needed satellites) is reduced to 5. Since we aim at estimating only the horizontal coordinates and velocities it is advantageous to use a latitude longitude coordinate system.

However, there will still be cases where less than 5 measurements are available. One solution could then be to assume a stationary mobile terminal anyway. If then 4 measurements are available, the resulting accuracy can be estimated. Depending on the resulting accuracy one may in a subsequent step have to resort to multiple hypothesis position calculation. Such approaches will be further discussed further below.

If only 3 measurements are available, then there is no other alternative than to only estimate the initial horizontal position.

Estimation of horizontal location and velocity

An embodiment of a basic solution will now be outlined. Let the parameter vector be:

θ = [φ λ φ λ b] (12)

In (12), φ is the latitude, λ is the longitude, φ is the time-derivative of the latitude, i the time-derivative of the longitude and b is the range rate measurement bias. The relation (1 1) can now be rewritten as:

P_ml = f_i(0) + e_i (13)

with

fXΘ) = r-±— (r_s, - r(φ,λ)). (v_svi - y(φ, λ))+ b (14) l^r _Sv, - ^r The earth centred earth fixed coordinates r = (x,y,z) can be computed from the latitude (φ) and longitude (λ), and altitude (h) as is well known in the art. First define:

a = 6378137 (15a)

/= 1/298.257224 (15b) e = Pf-P)¹'² (15c)

C = l/( cos²(φ) + (1 -f)² sin²(φ) )^² ( 15d) s = C(I -β² (15e) a =(aC+h) (15f) c₂ = (aS+h) (15g)

Then one gets:

x = (ci)cos(φ)cos(λ) (15h) y = (ci)cos(φ)sin(λ) (15i) z = (ca)sin(φ) (15j)

The velocities v_x,v_y,v_z can now be found by differentiating (15h)-(15j) with respect to φ and λ. Furthermore, the influence of ci,C2 on φ (15d) is neglected, since it can be shown that this term is small. Straightforward differentiation gives:

V_x = -(a)sin(φ)cosβ)φ -(a)cos(φ)sin(λ)λ (16a) Vy = -(ci)sin(φ)sin(λ)φ +(ci)cos(φ)cos(λ)λ (16b)

By substituting (x,y,z,Vχ,v_y,v_z) in (10) with the expressions (15h)-( 15j), (16a)- ( 16c) the Doppler expression (13) can be expressed as a function of the parameter vector θ. The equation (13) can now be linearised around an initial estimate of θ = Qo, so that it can be written:

P_m = f. {0o) + G. <0oW - 0o) + e. , (17) where the row vector d(θ) contains the derivatives:

The exact expression for Gι(θ) and fi(θ) is listed below.

A = Xsvi-ci cos(φ)cosfλ) dA/ dφ=a sin(φ)cos(λ) dA/dλ=a cos(φ)sin(λ)

B = v_XSυι+cisin(φ)cos(λ)φ +acos(φ)sin(λ)λ

C = y_Sυι-cicos(φ)sin(λ)

dC/dλ=-a cos(φ)cos(λ) D = v_ysυι+cisin(φ)sin(λ)φ -cicos(φ)cosfλ) λ

F = Vzsυι-C2COS(φ)φ

dfjdφ= l/r (B dA/dφ+A{acos(φ)cos(λ)φ -asin(φ)sin(λ)λ})

+l/r- (D dC/ dφ+C{cicos(φ)sin(λ)φ +cisin(φ)cos(λ) λ }) +l/r- (F dE/dφ+Ec₂sin(φ)φ )

-l/r* (AB+CD+EF)( dA/dφ-A+ dC/dφ-C+ dE/dφ-E); dfjdλ = 1/r- (B dA/dλ +A{-asin(φ)sin(λ)φ +acos(φ)cos(λ) λ })

+l/r- (D dC/dλ +C{asin(φ)cos(λ)φ +acos(φ)sin(λ) A J) -l/r³-(AB+CD+EF)( dA/dλ-A+ dC/dλ-C); dfι/dφ = 1/r- (Acisin(φ)cos(λ)+Casin(φ)sin(λ)-Ec2cos(φ)) dfjdλ = 1/r- (Acicos(φ)sin(λ)-Ccicos(φ)cosβ)) df_l/db = l By stacking all measurements one gets a system of equations:

p_m = f(θ₀) + G(θ₀)(θ - θ₀) + e . (19)

Now, one can solve for θ in a least squares sense:

θ = θ_{o +} (G^τGYG^τ(p_m -ϊφ_o)). (20)

The equations (14), (18), (20) need to be iterated a few times until they converge. An estimate of the resulting accuracy can be obtained if one knows the range rate accuracy σ. In that case:

E{(θ - θ_ιr)(θ - θj }= σ^>(G^ι Gy (21)

If σ is not known it can be estimated based on the least squares residual:

The expected value of the residual is:

EV = σ²(n_sat-5) (23)

so that:

σ2= V/(n_sat-5) (24)

is an unbiased estimate of the variance. The latter expression can be used whenever n_sat >5.

By using (21), possibly combined with (24), an initial uncertainty area can be defined. Transformations of the error ellipsoid to a two-dimensional high- confidence shape (ellipse or circle) are well known in the art. The resulting shape is used as input for the position calculation based on truncated pseudorange measurements. If the resulting uncertainty area is still larger than 75km in at least one dimension, some type of multiple hypothesis position calculation will be needed.

Exemplary performance of the outlined algorithm is shown in Fig. 7 for test data generated from a GPS reference receiver. It can be seen that in the majority of cases the accuracy is much better than the required 75km. Only for a negligible number of cases, there might be problems.

If the number of available satellite signals is limited, one might have to limit the number of unknowns to estimate in the Doppler measurement treatment. One than may have to restrict the estimation procedure to incorporate only the location, assuming the velocity of the mobile terminal to be negligible.

Estimation of location only

Assume that the number of measurements are fewer than 5, or if it is decided that the initial positioning attempt is done assuming a stationary mobile terminal. In this case:

θ = [φ λ b] (25)

and G(θ) and ϊ(θ\ simplifies a bit. The principles leading to equations (14), (18), and (20) can still be used but with the changes:

/,(*) = (r_JV, - r(<M)). (v_sv,)+ ό ^r _Sv« - ^r

[dφ dλ db

A = Xsυι-ClCOS(φ)cθSβ)

B — Vxsυi

C = y_Sυι-cicos(φ)sin(λ) dC/dφ=cisin(φ)sin(λ) dC/dλ=-a cos(φ)cos(λ)

D ⁼ Vysυi

F = VzSUi r = (A²+C²+E²)¹/²

dfjdφ= 1/r (B dA/dφ+D dC/ dφ+F dE/ dφ)

-l/r³ (AB+CD+EF)( dA/dφ-A+ dC/dφ-C+ dE/dφ-E); dfi/dλ = 1/r- (B dA/dλ+D dC/dλ)

-l/r³-(AB+CD+EF)( dA/dλ-A+ dC/dλ-C); dfydb = I

Furthermore the expected value of the variance (24) is changed to:

σ²= V/(n_sat-3) (24)

where V is defined in (22). This expression can be used whenever n_sat >3.

Fig. 8 illustrates a flow diagram of main steps of an embodiment of a method according to the present invention. The procedure starts in step 200. In step 210, truncated pseudorange measurements are performed in a mobile terminal on signals from at least 3 satellites. In step 214, Doppler shifts in ranging signals are measured in the mobile terminal for at least 3 satellites. In step 220, an approximation of the mobile terminal position is estimated from the Doppler measurements according to the above outlined procedures. Step 214 and 220 thus together form a step 215 for achieving an approximate mobile terminal position. In step 230, an accurate mobile terminal position is determined based on the approximate mobile terminal position and the results of the truncated pseudorange measurements. The procedure ends in step 299.

In a mobile terminal assisted AGPS, a method according to the present invention typically involves a few additional steps. An embodiment of such a method is illustrated in Fig. 9. Steps that are similar to steps of previous embodiments are given the same reference numbers and are not discussed further. In step 21 1, truncated pseudorange measurement results are reported from the mobile terminal to a positioning node, in which the actual position determination is performed. Similarly, a step 218 is introduced between steps 214 and 220, where representations of measured Doppler shift are reported from the mobile terminal to the positioning node. A further step 212 is also introduced, where it is decided if an approximate mobile terminal position of acceptable accuracy is already known, e.g.. from a cellular communications system to which the mobile terminal is connected. In such a case, the entire step 215 may be omitted.

From the Doppler based estimations outlined above, one realises that the estimations from case to case have to include either a full set of parameters to be estimated or a reduced set. A full set, c.f. equation ( 12), can be required when an improved accuracy is needed. However, such computations are obviously more complex. On the other hand, if only a few satellite signals are available, one might have to rely on a reduced set of parameters, c.f. equation (25).

In Fig. 10, the step 220 from Figs. 8 and 9 are illustrated in more detail, in an embodiment that makes use of a combination of such alternatives. In step 221 , an estimation is performed using a reduced set of parameters. At the same time, an estimate of the variance is calculated, giving a hint of the accuracy of the position approximation. In step 222 it is determined if the acquired accuracy is good enough to serve as an apriori position in step 230. If that is the case, the flow continues to next main step. Also if there are too few available satellite signals to make an estimation for a full set of parameters, step 223, the flow continues to the next main step. If there are available satellite signals and the first estimation was not good enough, a second estimation using the full parameter set is performed in step 224.

Also in the final position determination step 230, there might be some advantageous part steps, illustrated in Fig. 1 1. In step 231, it is determined if the available position approximation is good enough to use for conventional truncated pseudorange measurement position estimations. If that is the case, the procedure continues to step 232, where such a standard positioning procedure is performed. This procedure can then be based on an approximate position obtained from any external source, from the Doppler estimation on the reduced set of parameters or from the Doppler estimation on the full set of parameters, depending on which of the procedures that will give an acceptable accuracy. In the case of AGPS, as in the examples above, the accuracy limit is 75km. However, depending on the actual design of the satellite signal format, such accuracy limits may be different in other systems. If it is decided that the accuracy is not good enough for removing ambiguities in the conventional assisted satellite signal based positioning procedures, the procedure continues to step 233. In step 233, a multihypothesis procedure is performed in order to try to resolve the ambiguities. When the uncertainty for an AGPS system is in the order of 100km it is possible to use such multihypothesis methods, since there are only a limited number of candidate pseudoranges that are feasible. However in the case of 1000km uncertainty there will be a combinatorial explosion which essentially prohibits also the use of that technique.

Fig. 12 illustrates another embodiment of a method according to the present invention. In this embodiment, the mobile terminal serves for the provision of an approximate position, while a positioning node performs the detailed position determination. Compared with Fig. 9, Fig. 12 has moved the reporting step 225 to occur after the step 220. The data that is reported then becomes the approximate position and preferably an accuracy estimation thereof.

In a pure mobile terminal based AGPS, an embodiment of a method according to the present invention may look like Fig. 13. Here, no reporting steps are involved until the actual position determination is finished.

Instead, a step 235 is introduced, to report the finally estimated position to any party that may have use of such information. This could be as simple as a display in the mobile terminal display for a user, or a communication to e.g. a service provider offering position dependent service.

Fig. 14 illustrates a block scheme of an embodiment of a satellite signal based positioning system according to the present invention. This embodiment is a typical mobile terminal assisted positioning system. A satellite 20 sends out a ranging signal 30, which is received by a mobile terminal 10. The mobile terminal 10 comprises means 12 for determining a Doppler shift in the received ranging signal 30. The mobile terminal 10 further comprises means 13 arranged for truncated pseudorange measurements on the received signal 30. A means 16 for reporting Doppler shift as well as truncated pseudorange measurement results is also present.

The report is signalled 50 to a base station 61 of the communication system. A receiving means 62 of a positioning node 60, connected to the communications system receives the report. The Doppler measurements are used in a means 64 for estimating an approximate position. The approximate position is used by a position determining means 65 to calculate an accurate position of the mobile terminal 10. The means 65 and 64 are typically implemented in processor 67.

Fig. 15 illustrates a block scheme of another embodiment of a satellite signal based positioning system according to the present invention. Here, some more functionalities are incorporated in the mobile terminal 10. A means 14 for estimating an approximate position is provided in the mobile terminal 10. The report means 16 therefore prepare a report 51 comprising truncated pseudorange measurement results and an approximate position of the mobile terminal. In the positioning node, the final position determination is performed.

Fig. 16 illustrates a block scheme of yet another embodiment of a satellite signal based positioning system according to the present invention. This embodiment is a typical mobile terminal based positioning system. In the mobile terminal 10, a processor 17 is provided, which comprises means 14 for estimating an approximate position as well as means 15 for performing an accurate position determination based on truncated pseudorange measurements and the approximate position. A report 52 may be provided to any node 69 having requested a position for the mobile terminal 10.

Prior art solutions to the presented ambiguity problem either requires changes of communication standards, or a computationally complex method, which cannot even be guaranteed to give a correct results in the case with very large uncertainty areas. The solution according to the present invention instead relies on existing measurements and communication standards and is moreover computationally simple. It provides an accuracy which in the large majority of cases is ₍ sufficient as a starting point for more accurate position calculation based on truncated pseudorange measurements.

The embodiments described above are to be understood as a few illustrative examples of the present invention. It will be understood by those skilled in the art that various modifications, combinations and changes may be made to the embodiments without departing from the scope of the present invention. In particular, different part solutions in the different embodiments can be combined in other configurations, where technically possible. The scope of the present invention is, however, defined by the appended claims.

Claims

1. Method for assisted satellite-based positioning, comprising the steps of: providing (215) an approximate location of a mobile terminal (10); performing (210) truncated pseudorange measurements of satellite signals (30) at said mobile terminal (10); determining (230) a position of said mobile terminal (10) based on said approximate location and said truncated pseudorange measurements, characterised in that said providing (215) step in turn comprises the steps of: measuring (212) Doppler shifts of said satellite signals (30) at said mobile terminal (10); estimating (220) said approximate location based on said Doppler shift measurements.

2. Method according to claim 1, characterised in that said step of estimating in turn comprises the step of obtaining an accuracy estimation of said approximate location.

3. Method according to claim 1 or 2, characterised in that said step of estimating is based on an assumption of a velocity of said mobile terminal (10).

4. Method according to claim 1 or 2, characterised in that said step of estimating is based on a set of unknown parameters comprising at least one position parameter and one velocity parameter of said mobile terminal (10).

5. Method according to claim 2, characterised in that said accuracy estimation is used for selecting an appropriate set of unknown parameters.

6. Mobile terminal (10), comprising: means for providing an approximate location of said mobile terminal

(10); means (13) for performing truncated pseudorange measurements of satellite signals (30); means (15) for determining a position of said mobile terminal (10) based on said approximate location and said truncated pseudorange measurements, characterized in that said means for providing an approximate location in turn comprises: means (12) for measuring Doppler shifts of said satellite signals (30); means (14) for estimating said approximate location based on said Doppler shift measurements.

7. Mobile terminal according to claim 6, characterised in that said means for estimating is further arranged for obtaining an accuracy estimation of said approximate location.

8. Mobile terminal according to claim 6 or 7, characterised in that said means for estimating is arranged to base said estimation on an assumption of a velocity of said mobile terminal (10).

9. Mobile terminal according to claim 6 or 7, characterised in that said means for estimating is arranged to base said estimation on a set of unknown parameters comprising at least one position parameter and one velocity parameter of said mobile terminal (10).

10. Mobile terminal according to claim 7, characterised in that said means for estimating is further arranged for selecting an appropriate set of unknown parameters using said accuracy estimation.

11. Node (5, 7, 60, 69) connected to a mobile communications system (2), comprising: means for providing an approximate location of a mobile terminal (10) connected to said mobile communications system (2); means (62) for receiving data (50, 51) representing truncated pseudorange measurements of satellite signals (30) performed at said mobile terminal (10); means (65) for determining a position of said mobile terminal (10) based on said approximate location and said truncated pseudorange measurements , characterized in that said means for providing an approximate location in turn comprises: means (62) for receiving data (50) representing Doppler shift data of said satellite signals (30) measured at said mobile terminal (10); means (64) for estimating said approximate location based on said Doppler shift measurements.

12. Node according to claim 11, characterised in that said means for estimating is further arranged for obtaining an accuracy estimation of said approximate location.

13. Node according to claim 11 or 12, characterised in that said means for estimating is arranged to base said estimation on an assumption of a velocity of said mobile terminal (10).

14. Node according to claim 11 or 12, characterised in that said means for estimating is arranged to base said estimation on a set of unknown parameters comprising at least one position parameter and one velocity parameter of said mobile terminal (10).

15. Node according to claim 12, characterised in that said means for estimating is further arranged for selecting an appropriate set of unknown parameters using said accuracy estimation.

16. Mobile communications system (2), comprising: a mobile terminal (10); a positioning node (5, 7, 60, 69); and means for providing an approximate location of said mobile terminal

(10); said mobile terminal (10) in turn comprising means (13) for performing truncated pseudorange measurements of satellite signals (30) and means (16) for sending data (50, 51) representing said truncated pseudorange measurements to said positioning node (5, 7, 60, 69); said positioning node (5, 7, 60, 69) in turn comprising means for receiving said data (50, 51) representing said truncated pseudorange measurements and means (65) for determining a position of said mobile terminal (10) based on said approximate location and said truncated pseudorange measurements, characterized in that said means for providing an approximate location in turn comprises: means (12, 62) for obtaining Doppler shifts of said satellite signals (30); means (14, 64) for estimating said approximate location based on said Doppler shift measurements.

17. Mobile communication system according to claim 16, characterised in that said mobile terminal (10) comprises means (12) for measuring Doppler shifts of said satellite signals (30) and means (16) for sending data (50) representing said Doppler shift measurements to said positioning node (5, 7, 60, 69); and said positioning node (5, 7, 60, 69) comprises said means for providing an approximate location, whereby said means (12, 62) for obtaining Doppler shifts comprises means (62) for receiving said data (50) representing Doppler shift data of said satellite signals (30) measured at said mobile terminal (10).

18. Mobile communication system according to claim 16, characterised in that said mobile terminal (10) comprises said means for providing an approximate location and means (16) for sending data (51) representing said approximate location to said positioning node (5, 7, 60, 69); and said positioning node (5, 7, 60, 69) comprises means (62) for receiving said data (51) representing said approximate location.

19. Mobile communication system according to any of the claims 16 to 18, characterised in that said means for estimating is further arranged for obtaining an accuracy estimation of said approximate location.

20. Mobile communication system according to any of the claims 16 to 19, characterised in that said means for estimating is arranged to base said estimation on an assumption of a velocity of said mobile terminal (10).

21. Mobile communication system according to any of the claims 16 to 19, characterised in that said means for estimating is arranged to base said estimation on a set of unknown parameters comprising at least one position parameter and one velocity parameter of said mobile terminal (10).

22. Mobile communication system according to claim 19, characterised in that said means for estimating is further arranged for selecting an appropriate set of unknown parameters using said accuracy estimation.