MOBILE POSITIONING IN MOBILE NETWORKS

The present invention relates to a method of geographically positioning a mobile terminal in a mobile network and a network to enable the method.

Background to the Invention

Third Generation (3G) UMTS network operators are known to provide geographical location services which allow the position of a mobile terminal to be determined. There are currently various ways to locate a mobile terminal. WO 01/95659 (Cellpoint Systems AB) and WO 02/35875 (Motorola Inc.) describe the use of storing time difference of arrival measurements (TDOA) from a plurality of base stations. WO 01/95659 defines measuring the real time difference (RTD) information between base stations during handovers. The measures RTD and the stored TDOA measurements are processed in a server to calculate the location of a mobile terminal. WO 02/35875 detects a measurable signal type and compares it with a stored signal operable to calculate the TDOA, to determine the mobile terminals position.

US 5394158 describes storing a set of ordered pairs corresponding to distance and signal characteristics associated with a particular location and comparing currently measured distance and signal characteristics with the stored information to determine the position of a mobile terminal.

US publication 2002/0193941 discloses a self locating devices capable of receiving signals, which defines its location by identifying the base stations from the GSM signals it receives. The base stations are identified from the common channel identifiers that are associated with the GSM signals sent from them.

At present it is known to use a global navigation satellite system (GNSS) such as global positioning system (GPS) in order to determine the position of a mobile

terminal. It is also possible to use cellular telephone signals to determine, to a less accurate degree, the position of a mobile telephone. The most commonly used techniques are Assisted Global Positioning System (A-GPS) and Cell_ID. The accuracy of A-GPS is around 20 meters. The accuracy of Cell__ID depends on the size of the cell which may vary from 300 m to 1.5 km in diameter. The details of these systems will not be described here as they are well known to those skilled in the art.

However, not all mobile terminals have A-GPS functionality as it requires complicated and expensive software and hardware. These mobile terminals therefore typically use CeIl ID when a location service is requested by the user. However, the accuracy of Cell_ID is not good enough for users to accurately pinpoint their position. There is therefore a need for a simple, inexpensive and more accurate method locating the position of a mobile terminal. System frame number (SFN) measurements are supported by all mobile terminals. Therefore, we propose that SFN measurements are combined with GNSS values to find the location of a mobile terminal , irrespective of whether the mobile terminal supports A-GPS or not. By using SFN and GPS measurements to deduce potential location information, the position of mobile terminals ( even ones that do not support A-GPS ) can also be determined with greater accuracy.

Summary of the invention

In a first aspect the present invention provides a method of generating a database of values indicative of a first characteristic of a plurality of pairs of base stations, comprising the steps of: establishing the position of a mobile terminal using global navigation satellite systems(GNSS) measuring a first timing difference, indicative of the position of said mobile terminal with respect to a pair of base stations;

calculating a first value indicative of a first characteristic for said pair of base stations from said established position of said mobile terminal and said first timing difference; and repeating the above steps for a plurality of mobile terminals and for a plurality of pairs of base stations and storing the results.

Thus, by using a mobile terminal which is enabled to establish its own position, a database of values can be built up which can then be used to determine the position of a mobile terminal in a second aspect of the present invention.

Preferably, the mobile terminal is a A-GPS enabled mobile which can establish its own position and the database of values is a database of random time differences between the system frames of pairs of cells (T _ran dom delay u)-

Preferably, the first timing difference is an SFN-SFN (System frame number) timing difference measurement. SFN is a common identifier allocated to each data frame. There are two types of SFN-SFN observed time difference measurements, type 1 and type 2. SFN-SFN type 1 measurements may be used for CELL_FACH

(forward access channel) and CELLJDCH (dedicated transport channel) connection in the intra-frequency mode. Type 1 measurements require the UE (user equipment) to read the Broadcast channels (BCH) of neighbouring cells. However this reading cannot be reported until the UE enters a CELL_FACH connection. Therefore the signals needed for type 1 are not always strong enough to ascertain time difference form multiple base stations.

SFN-SFN type 2 measurements can be used in both intra and inter-frequency for CELL_FACH and CELLJDCH. Preferably type 2 measurements are used in the invention since the signals required for SFN-SFN type 2 observed time difference measurements are strong enough to ascertain time difference from multiple base stations. However SFN-SFN type 1 measurements may be used in limited situations depending on signal strength.

Preferably, the present invention provides a method wherein said step of calculating a said first value comprises a step of calculating a theoretical value indicative of the position of said mobile terminal with respect to a said pair of base

stations on the basis of said established position of said mobile terminal and a step of calculating a said fist value on the basis of said theoretical value and said first timing difference.

Preferably, the theoretical value is the theoretical path delay time difference between two cells (Theoretical T _p d ^). The T _ran i _om de _l ay t-i _c is calculated according to the equations given in the detailed description of the invention.

In a second aspect the present invention provides a method of establishing the position of a mobile terminal in a mobile communications network using the database of first values indicative of a first characteristic of each of a plurality of pairs of base stations established in the first aspect of the invention, comprising the steps of: measuring a second timing difference, indicative of the position of said mobile terminal with respect to a pair of base stations; and determining the position of said mobile terminal from said second timing difference and a first value corresponding to said pair of base stations.

Thus, the position of a mobile terminal can be calculated using the database, which has been established, and values which have been already measured. The position is calculated without the use of A-GPS in the mobile terminals for which the position is being calculated.

As mentioned in the first aspect of the invention, the first values are random time differences between the system frames of pairs of cells (T _random de _la y t- _k ) which are calculated using A-GPS and System Frame Number (SFN) -SFN timing difference measurements, which are provided simultaneously by A-GPS enabled mobile terminals. Preferably, the second timing difference is an SFN-SFN timing difference measurement.

Preferably, the present invention provides a method wherein said step of calculating the position comprises a step of calculating a third timing difference indicative of the position of said mobile terminal with respect to said pair of base stations on the basis of said second timing difference and a first value corresponding

to said pair of base stations and a step of calculating the position of said mobile terminal on the basis of said third timing difference.

Preferably, the third timing difference is the path delay time difference between two cells (T _{pd Wc} ) which is calculated using the equations shown below in the detailed description.

Preferably, at least two T _{pd t} . _k values are used to determine the position of the mobile terminal by using for example, the known positions of the base stations and hyperbolic equations as shown below.

The accuracy of this method would be between Cell_ID and A-GPS.

In a third aspect the present invention provides a mobile communications system, comprising a plurality of base stations arranged to communicate with a plurality of mobile terminals, the system comprising: a database of first values indicative of a first characteristic of each of a plurality of pairs of base stations; first value means for calculating said first values on the basis of established positions of said mobile terminals and intermediate values indicative of positions of said mobile terminals with respect to pairs of base stations; and position means for calculating the position a said mobile terminal on the basis of the intermediate values indicative of the position of said mobile terminal with respect to said pair of base stations and a said first value corresponding to said pair of base stations.

Brief description of the drawings

So that the present invention be more readily understood, embodiments thereof will now be described by way of example only and with reference to the accompanying drawings in which:

Figure 1 shows a procedure diagram of an embodiment of the present invention;

Figure 2 show an example of hyperbolic positioning in an embodiment of the present invention; and

Figure 3 shows a database structure for T _{random de} ι _{ay t} . _k values between each pair of four cells in an embodiment of the present invention. Detailed description of the preferred embodiments

Figure 1 shows a procedure diagram of an embodiment of the present invention. The procedure will be described below.

In a typical 3 G mobile network there are a large number of mobile terminals with A-GPS capabilities. The Mobile switching centre (MSC) in 3 G core networks is responsible for maintaining databases for location registration, allocation resources, handover management, paging etc. The location of all mobile terminals that are present in the area served by an MSC is registered in databases maintained in the MSC. The latest location of a particular mobile terminal is stored in a Visitor location register (VLR). This contains the location of all the mobile terminals in the MSC area. Another register is maintained by the MSC called the Home location register (HLR). This register contains subscriber information of mobile users registered with a particular MSC. The data from the HLR of a mobile terminal is copied onto a VLR when a mobile enters (or switches ON) its home MSC or another MSC.

When a mobile terminal enters the area served by a new MSC, the VLR is updated and the location information in the VLR is communicated to the HLR of the mobile terminal. The location may be updated in the VLR based on the quality of signal strength each time the mobile terminal requests a service. A temporary mobile subscriber identifier is allocated to the mobile terminal by the VLR for tracking its changes. The location updates in the VLR and HLR enhances the paging procedure for a mobile terminal, since its latest location is known from the VLR, and the call may be easily routed.

If further location based services are required for additionally functionality a location application server (LAS) may be integrated into the structure of the MSC, for deploying a particular logic function. This server may be configured in a desired

fashion to perform and record location updates . Therefore if more precise database is required, the LAS may be configured to poll the different mobile units at regular intervals to provide their location information, which is registered at in a database. The LAS can also support GNSS services. The LAS may then be configured to perform predefined computation on the location data that it receives.

Therefore, for the operation of the present invention, mobile terminals report their A-GPS measurements to the Location Application Server (LAS) at regular intervals. Since the location of the mobile terminal is known from the A-GPS measurements, and the location of the base stations is known from a database in the LAS, the distance of the mobile terminal from each of two base stations i and k can be calculated. From these values it is possible to calculate the theoretical path delay from the mobile terminal to each of the two base stations i and k. The difference between these to path delays is the theoretical path delay time difference between two base stations i and k (theoretical T _pdi-k ).

The LAS receives the A-GPS measurements and calculates the theoretical T _pd i_ _k using the base station locations in the LAS's database and A-GPS measurements as follows:

Theoretical T _{pd i} . _k =

(VO* - ^χ _G ps Y + (y _t - y _G p _S Y - V<λ - ^X GPS Y + O* - yen Y ) ' ^c C ¹ )

where, (x;, yϊ) coordinates of base station i, known from LAS's database;

( ^χ _f o Y _k ) coordinates of base station k, known from LAS's database; (XAGPS, Y _A G _PS ) coordinates of the mobile terminal, known from A-GPS measurements; and c is the speed of light = 3x10 ⁸ m/sec.

Working the above in reverse, if the actual T _{pd i-k} is measured between a mobile terminal and two base stations, the coordinates of the mobile terminal could be

calculated, for example by using hyperbolic positioning. However, the equation has two solutions for the coordinates X _AGPS and y _AG p _S , and two values of T _{pd i-k} are therefore required. This can be achieved by using three base stations (a, b and c) instead of two. For any given mobile terminal position, there is a path delay time difference between each of the three pairs of base stations and the mobile terminal (Tpd i- _2> T _pd2-3 and T _{pd 1-3} ). Figure 2 shows an example of hyperbolic positioning and is for illustrative purposes only. Two T _{pd uk} could be used to solve the following hyperbolic equations to estimate the mobile terminal position (x _m , V _1n ):

V(*2 - ^χ .Ϋ + (yi -y _m ? - V(*i -*.) ^a + (λ - λ,,) ² = cτ _pdl2 (2)

V(*3 - ^χ _m Ϋ + (y ₃ - yJ - J(X ₁ - *.) ² + (κ - y, _n Ϋ = ^ ₁₃ (3)

where,

( ^χ _m5 Y _m ) coordinates of the mobile terminal, to be estimated; ( ^χ i _> Yi) coordinates of base station 1, known form LAS's database; (x ₂ , V ₂ ) coordinates of base station 2, known form LAS's database;

( ^χ _3> Yi) coordinates of base station 3, known form LAS's database; c is the speed of light; and T _pdn , T _pdl3 are the measured path delay time differences between cells 1 and 2 and cells 1 and 3 respectively.

Two equations (2 and 3) could be solved if two measurements are available from two measurements. If only one measurement is available mobile's position could not be estimated using this method.

However, for mobile terminals without A-GPS, T _{pd i-k} can not be measured directly. All mobile terminals do however measure the SFN-SFN timing difference between two base stations i and k. SFN-SFN timing difference measurements are performed in the mobile terminal to identify the frame time difference between two geographically separated sites. The present invention utilises these measurements as

follows to enhance positioning accuracy. The measured SFN-SFN time difference has three components as described below:

Measured SFN-SFN tuning _uk = T _{n t} . _k + Offset x 38400 (4)

Where, T _{1n t} . _k = T _pd u _k + T _{rmdom de} ι _ay ,. _/ .; (5)

T _pdi _ _k is the path delay time difference between two cells i and k; T r _an d _{om dela} y i- _k is random time difference between the System Frames of two cells i and k; and Offset is a number between 0 and 255 set by the operator.

Since the UMTS networks are not synchronized, there is always a time difference between two frames. This difference, T _{m ;-fo} has two components. The first one is a random delay due to asynchronous nature of network, T _{random de} ι _ay ^, and the second one is the path delay, T _{pd i} _ _k , which, as described above enables the location of a mobile terminal to be calculated.

From the above equation it can be seen that once the SFN-SFN timing difference has been measured, T _{m uk} can be calculated. In turn, as shown above, T _{pd uk} can be used as described above to calculate the mobiles position. However, to calculate T _pd ^ we need to know T _{random de} ι _{ay t} . _k . In order to work this value out, mobile terminals with A-GPS must be used.

The LAS regularly sends measurement requests to mobile terminals with A- GPS capabilities for the following:

• SFN-SFN time difference measurements • A-GPS measurements

The mobile terminal reports the SFN-SFN time difference and A-GPS measurements simultaneously. T _{random delay i-k} can be calculated by the LAS as follows:

Trandom delay i-k = Measured T _{m uk} - Theoretical T _{pd t} _ _k (6)

Theoretical T _{pd t} . _k is calculated from the A-GPS measurements as shown above in relation to equation one. Measured T _{m uk} is calculated from the SFN-SFN timing difference measurements as shown above in relation to equation four. Thus, the LAS can calculate a database for the T _{random de} ι _{ay t} - _k between each pair of cells i and k using above procedure. An example database is shown in Figure 3.

These measurements are repeated at regular intervals in the network to collect enough statistics and built up a reliable T _{random de]ay i-k} database in the LAS as shown in figure 3. The random delays between two system frames are expected to be stable. If any change is observed then the database would automatically update itself. The LAS then uses the T _{random de} i _ay ^ database when a location fix is requested by the user, in accordance with the following procedure:

1. The user requests a location service;

2. The network requests the last measured SFN-SFN measurement from that mobile terminal;

3. The mobile terminal sends the SFN-SFN timing difference measurement report to the LAS;

4. The LAS calculates the T _{m uk} using equation (4);

5. Using the T _{rcmdom de} ι _{ay uk} database, the LAS calculates T _{pd t} . _k using equation (5); and

6. Finally, the LAS can calculate the position of the mobile terminal, for example using hyperbolic positioning as shown above in relation to equations 2 and 3.

Thus, the present invention provides a mechanism which utilises measurements from mobile terminals with A-GPS capabilities, to enable the position of mobiles without A-GPS capabilities to be calculated.

Hyperbolic positioning has been described above as a method of calculating the mobile terminals position from the calculated T _{pd t} . _k . However, other techniques, including other triangulation techniques, may be used as will be appreciated by one

skilled in the art. It will also be appreciated that although the above technique has been described on the basis of obtaining two values for T _{pd ;-fo} the accuracy of the technique could be increased by obtaining three or more values of T _pd ^.

Furthermore, it is possible that the present invention could be used with other radio information in order to triangulate the position of the terminal. For example, if the mobile terminal can only receive signals form two base stations, then the present invention may only allow the position the terminal to be calculated along one hyperbola. However, other methods known to the person skilled in the art may can help calculate position along the hyperbola. For example, looking at the intersection of the hyperbola with known sections of each of the cells of the respective base stations or the intersection of the hyperbola with ranging estimates based on radio power. Other techniques will be apparent to the person skilled in the art for establishing the position of a mobile terminal along a single hyperbola.

It can therefore be appreciated that, although it is preferable to calculate the position of a mobile terminal along two hyperbolas using two values of T _pd ^, this is not essential to the present invention in view of the above described alternatives.

The second aspect of the present invention has been described above in relation to mobile terminals with A-GPS capabilities. While these mobile terminals may be mobile phones, it will be appreciated that alternative terminals may be used in accordance with the present invention. For example, in the event that the number of A-GPS mobile phones is limited, the network operator may wish to improve the database of T _{random delay t} . _k itself. This could be done by moving test equipment round the area covered by the network. The equipment must be capable of taking SFN-SFN timing difference measurements and measuring its location, by using GPS for example. The database of T _{random de} ι _{ay t} _ _k is then established in accordance with the method described above. This method is possible by virtue of the fact that the timing offset (J _rand o _{m d} e _l ay i- _k ) between two base stations is fairly constant over long periods of time. Once a base station is powered up, a test vehicle could drive round the cell building up a database of offset values for all neighbouring cells. These values would remain valid for a considerable period of time. This method could also be used if

there is any reason for not wanting to use user's mobile phones to take the relevant measurements.