WO2004077856A1

WO2004077856A1 - Improved handset location determination

Info

Publication number: WO2004077856A1
Application number: PCT/GB2003/005639
Authority: WO
Inventors: David Stanley Hill
Original assignee: Roke Manor Research Ltd
Current assignee: Roke Manor Research Ltd
Priority date: 2003-02-27
Filing date: 2003-12-24
Publication date: 2004-09-10
Anticipated expiration: 2005-08-27
Also published as: GB2398963B; AU2003290324A1; GB0304478D0; GB2398963A

Abstract

The present invention relates to location-aware systems in handsets, such as mobile telephones. In a deployment with a high degree of frequency reuse then there is a problem in differentiating between signals from different base stations which happen to share a common transmission frequency. The present invention addresses this problem, and aims to increase the number of base transceiver station (BTS) seen, by discriminating between two base transceiver station (BTS) operating on a same carrier frequency. In particular, the present invention provides the application of extended correlation, as implemented in uplink time difference of arrival (U-TDOA), to improve the location accuracy of enhanced observed time difference (E-OTD).

Description

IMPROVED HANDSET LOCATION DETERMINATION

The present invention relates to location-aware systems in handsets, such as mobile telephones. In a deployment with a high degree of frequency reuse then there 5 is a problem in differentiating between signals from different base stations which happen to share a common transmission frequency.

The present invention addresses this problem, and aims to increase the number of base transceiver station (BTS) seen, by discriminating between two base transceiver station (BTS) operating on a same carrier frequency. In particular, the

10 present invention provides the application of extended correlation, to improve the location accuracy of enhanced observed time difference (E-OTD).

The GSM frame structure will now be discussed, with reference to Fig. 6.

GSM utilises a combination of frequency division multiple access (FDMA) and time 15 division multiple access (TDMA). Here we are interested in the temporal domain. A mobile station (MS) will be assigned a time slot (TS) 61, one of eight in a time division multiple access (TDMA) frame 62. Each timeslot (TS) 61 lasts for 577 μs and carries 148 bits in 156.25 symbol durations. A frame 62 of 8 timeslot (TS) is repeated every 4.615 ms. Time division multiple access (TDMA) frames are arranged 20 in progressively larger multiframes 63, 64, superframes 65 and hyperframes 66, see

Figure 6. The periodicity of the hyperframe 66, is typically 3 hours 28 minutes 53.76 seconds, representing 2048 superframes and 2,715,648 TDMA frames 61.

Each time slot provides 148 bits per 4.615 ms TDMA frame. The multiframe may be a 26-frame multiframe 63 lasting 120ms, or a 51 -frame multiframe 64, lasting 25 235.38ms. The superframe 65 is made up of fifty-one 26-frame multiframes, or twenty-six 51-frame multiframes. The data is supplied at a rate of 32069 bits per second.

A typical, normal burst carried in timeslot 61 is made up of three symbols of tail bits 61a, fifty-eight symbols of encrypted bits 61b, a training sequence 61c of 30 twenty-six symbols, a further set of fifty-eight symbols of encrypted bits 61d, and a further three symbols of tail bits 61e. In GMSK modulation, each symbol is one bit long. In 8PSK modulation, each symbol is three bits long.

As illustrated in Fig. 7, the GSM air interface uses three main logical control channels in Base Transceiver Station (BTS) Downlink Signalling. These are the

Broadcast Channel (BCH), the Common Control Channel (CCCH) and the Dedicated

Control Channel DCCH).

The further logical control channels of interest will now be discussed. Broadcast Control CHannel (BCCH): in the downlink only, which is used to broadcast cell specific information; The broadcast control channel (BCCH) provides general information on a per-base transceiver station (BTS) basis, i.e. cell specific information. It is used for downlink point-to-multipoint communication and is unidirectional. The broadcast control channel (BCCH) includes information such as the base transceiver station (BTS) country code, network code, local area code public lands mobile network (PLMN) code, RF channels used by the cells of the base transceiver station (BTS) surrounding cells, hopping sequence number, mobile RF channel number for allocation, cell selection parameters and random access channel (RACH) description. The broadcast control channel (BCCH) also carries the important common control channel (CCCH)_CONF, which indicates the organisation of the common control channel (CCCH). The broadcast control channel (BCCH) is always transmitted on a designated carrier the 'broadcast control channel (BCCH) carrier' in time slot 0, denoted C0T0. The broadcast control channel (BCCH) contains some data that is highly likely to be identical between near spaced base transceiver station (BTS) e.g. base transceiver station (BTS) country code, network code, local area code public lands mobile network (PLMN) code, but there is also data which will be specific, and possibly unique to a base transceiver station (BTS).

Synchronisation CHannel (SCH), in the downlink only, which is used to broadcast synchronisation and base station system (BSS) identification information; With reference to Fig. 9, the synchronisation burst contains a 64 bit extended training sequence, and 78 (2 x 39) encrypted bits 92, 93. The extended training sequence was specifically designed for auto correlation, and provides a single correlation peak. The 78 encrypted bits 92, 93 contain the reduced time division multiple access (TDMA) frame number and base transceiver station identity code (BSIC). The base transceiver station identity code (BSIC) contains the public lands mobile network (PLMN) colour code which will be the same throughout any particular country and the base station colour code, see Figure 10, which will vary between base stations.

Paging CHannel (PCH): in the downlink only, which is used to send page requests to mobile station (MS)s; The paging channel (PCH) is used to alert the mobile station of an incoming call.

Random Access CHannel (RACH): in the uplink only, which is used to request a Dedicated Control CHannel; Access Grant CHannel (AGCH), in the downlink only, which is used to allocate a Standalone Dedicated Control CHannel; The Access Grant Channel

(AGCH) is used to allocate an standalone dedicated control channel (SDCCH) to a mobile for signalling, in order to obtain a dedicated channel, following a request on the random access channel (RACH). Standalone Dedicated Control CHannel (SDCCH): which is bi-directional;

The standalone dedicated control channel (SDCCH) is used for connection set-up and location update data.

Fast Associated Control CHannel (FACCH), which is bi-directional, and is associated with a Traffic CHannel; Slow Associated Control CHannel (SACCH), which is bi-directional, and is associated with a standalone dedicated control channel (SDCCH) or a Traffic Channel. The slow associated control channel (SACCH) is used for low rate, non- critical signalling. This includes System Information Type 5, 6 and optionally 5bis and 5ter messages¹¹. The System Information Type 5, 5bis and 5ter messages provide information on the broadcast control chamiel (BCCH) frequency of the neighbour cells. The System Information Type 6 message provides information on the location area of the current cell, possibly the status of the NCH, and an indication of whether paging channel restructuring has taken place.

With reference to Figs. 5 and 8, the Frequency Correction channel (FCCH), is carried in timeslot 0 (TSO) of every tenth TDMA frame 61₀. It is transmitted 5 times each 51 multiframe 62. The frequency correction burst 81 is used to synchronise the mobile station (MS) with the base transceiver station (BTS) and contains 148 bits which are all zero, effectively an unmodulated carrier signal, albeit with a frequency offset. The Cell Broadcast Channel (CBCH) is optional and is used for general, i.e. not point to point, short message information, such as weather forecasts and traffic information.

The present invention accordingly provides, in a mobile telephone network, a method for recognising a required base station (BTS k) which is transmitting a signal to a mobile terminal. The method comprises transmitting, from the base station, a

' signal comprising data frames having a predictable data content, said content comprising content unique to the required base station (BTS k); receiving, in the mobile terminal, the signal comprising data frames having a predictable content; independently providing information to the mobile terminal sufficient to allow it to calculate the data content of the data frames having a predictable data content; calculating, in the mobile terminal, the data content of the data frames having a predictable data content; correlating, in the mobile terminal, the calculated data content with the data content of the received signal; and, upon detection of a correlation above a certain threshold between the calculated data content and the content of the received signal, recognising the required base station as the source of the received signal.

The method may further comprise the steps of : independently providing to the mobile terminal, sufficient information to enable it to predict the time of arrival of the data frames having a predictable content; predicting, in the mobile terminal, the time of arrival of the data frames containing predictable data content; and correlating, in the mobile terminal, the calculated data content with the data content of a received signal only within a predetermined timeframe based on the predicted time of arrival. A plurality of base stations may transmit signals on the same frequency. In this case, the method may further comprise the step of: in response to the recognition of the required base station, ignoring signals from other base stations on the same frequency.

Preferably, a corresponding plurality of sets of information sufficient to allow to the mobile terminal to calculate the data content of data frames having a predictable data content from each of the plurality of base stations, is provided to the mobile terminal; the mobile terminal calculates the data content of the data frames having a predictable data content from each of the plurality of base stations; and the mobile terminal correlates the calculated data content of each of the sets of calculated data content with data content of a received signal. Upon detection of a correlation above a certain threshold between one set of the calculated data content and the data content of the received signal, the one of the required base stations is recognised as the source of the received signal. The method is repeated for at least one further base station. The method may further comprise the steps of: independently transmitting to the mobile terminal data sufficient to allow the mobile terminal to calculate a time of transmission of the data frames; detecting the time of arrival of the data frames at the mobile terminal; deriving a value of the transmission delay for the data frames; and using the derived value of transmission time in a calculation to determine the position of the mobile terminal with respect to at least one base station.

The above, and further characteristics, advantages and aims of the present invention will become more apparent with reference to the following description of certain embodiments, with reference to the appended drawings, wherein:

Fig. 1 illustrates the mapping of Logical channels;

Fig. 2 illustrates uplink data transmission;

Fig. 3 illustrates enhanced observed time difference (E-OTD) downlink data transmission;

Fig. 4 illustrates communication of a multiframe to a handset;

Fig. 5 illustrates a correlation using a sub-set of the multiframe data;

Fig. 6 illustrates a GSM frame structure;

Fig. 7 illustrates GSM logical channels; Fig. 8 illustrates a frequency correction burst;

Fig. 9 illustrates a synchronisation burst; and

Fig. 10 illustrates a base station identity code.

The present invention relates to the use of an extended correlation technique to aid the achievement of fifty-metre accuracy for enhanced observed time difference (E- OTD) location calculations, in particular, by enabling the measurement of base transceiver station (BTS) signals that are currently obscured by mutual interference.

The conventional timing estimation, correlation, process is used in both the location measurement unit (LMU) and the handset (mobile station (MS)). It is considered that an extended correlation scheme would only be applicable to use in the handset. The reasons are twofold; firstly, it is assumed that the optimum mode of operating an location measurement unit (LMU) is to operate in absolute time difference (ATD) mode with internals only and, second, that improving the mobile station (MS) measurements will yield the greater improvement in location performance.

A known system, enhanced observed time differeήee

illustrated in Fig. 3, uses measured differences in the time of arrival at a mobile station 10 of signals from various base transceiver stations (BTS). An alternative known system, uplink time difference of arrival (U-TDOA), illustrated in Fig. 2, uses measured differences in the time of arrival at various base transceiver stations (BTS) of a signal from a mobile station 10.

The frequency correction burst 81 (Fig. 8) is of no use in the correlation process and in fact would have a negative impact, significantly raising the spurious peak level. Hence, the correlation data length would either have to be bounded by the frequency correction bursts or a window function would have to be applied to the data to effectively remove the 142 zero bits.

In most applications of interest the existing enhanced observed time difference (E-OTD) measurement is sufficient for the measurement of the majority of the base transceiver station (BTS) signals. Only data transmission for the measurement of additional base transceiver station (BTS) from which signal reception is problematic needs to be addressed.

There are at least two ways in which handset location measurement performance could be improved: firstly, by increasing the accuracy of timing measurements; and secondly by increasing the number of base transceiver station (BTS) measured by the handset. Both of these could be enhanced by improving the signal to noise ratio of the correlation process. However, with tight frequency reuse, increasing the number of base transceiver station (BTS) seen can in some cases only be achieved by obtaining timing measurements from two or more base transceiver stations (BTS) that have the same broadcast control channel (BCCH) carrier frequency. The presnt invention employs an extended correlation process to enable the measurement of a weaker base transceiver station (BTS) signal in the presence of another base transceiver station (BTS) signal on the same broadcast control channel (BCCH) carrier frequency.

It is considered that in order to improve the timing accuracy over that achievable using just the synchronisation burst would require many multiframes of data to be correlated. Therefore, if the intention is to improve performance then this may be better achieved by increasing the number of base transceiver stations (BTS) used in the location estimate. In a deployment with high frequency reuse this might be achieved by improving the discrimination between base transceiver station (BTS) operating with the same broadcast control channel (BCCH) carrier frequency. The number of bits used in the auto-correlation process in the case of enhanced observed time difference (E-OTD) (Fig. 3) is 64 bits, being the length of the extended training sequence 91 in the synchronisation burst 61 (Fig. 9). Although the extended training sequence is only 64 bits long it does have the special property of only having a single correlation peak. The larger the number of bits the greater the potential processing gain, but for maximum effect the data should be truly random.

More data could be employed in the auto correlation process in order to increase the processing gain of the signal processing employed in the enhanced observed time difference (E-OTD) processing (Fig. 3). This may be beneficial in two respects. Firstly, the increased processing gain may increase the accuracy of measurements where the signal level is only marginally detectable, i.e. reduce the noise content of the measurement. Additionally, it may be possible to measure signals below the normal GSM signal processing threshold, and consequently add extra base transceiver station(s) (BTS) to a location estimate. The inclusion of an extra base transceiver station (BTS) may be particularly effective if the extra base transceiver station (BTS) provides a significant improvement in the geometry, i.e. geometrical dilution of precision (GDOP) is appreciably improved.

In an unsynchronised base station system (BSS), the GSM frames should be randomly aligned, and hence there should be a high probability that if you can determine where you are in a frame sequence then you would be able identify the base transceiver station (BTS) from which the frame sequence was transmitted. If you had a priori information indicating the approximate time of arrival of the defined frame sequence then the additional processing load on the handset may be reduced to an acceptable level.

A method according to the present invention takes advantage of the fact that a sub-set of the 51 multiframe data sequence can be readily predicted, minimising the requirement for data exchange, and that the detail in the sequence is unique to a particular base transceiver station (BTS) thus providing a basis upon which to discriminate between the two base transceiver station (BTS) signals. While the present invention particularly applies to additional processing in the mobile station (MS) measurement, as the location measurement units (LMU) may operate in absolute time difference (ATD) internals-only mode. However, the invention may also be applied to the location measurement unit (LMU) processing. The following discussion concentrates on the data available on the broadcast control channel (BCCH) carrier frequency, in time slot 0, as this frequency and time slot are already measured. However, the present invention may be applied to other frequencies and time slots.

In the contemporary implementation of enhanced observed time difference (E- OTD) (Fig. 3) the timing information is derived from the use of the Extended Training sequence¹ 91, of length 64 bits, within the synchronisation channel (Fig. 9). This 'α priori' information is stored in the location measurement unit (LMU). The Synchronisation channel is always transmitted on time slot 0 (TS 0) on the broadcast control channel (BCCH) carrier frequency. The synchronisation channel is only a part of the information transmitted in timeslot (TS) 0 on the broadcast control channel (BCCH) carrier frequency, the other information being the frequency correction channel, the broadcast common control channel and the dedicated control channel, see Figure 1.

Consider the data content of the channels that constitute the 51 multiframe. The present invention requires evaluation of two characteristics. Firstly, how different the data is between nearby base transceiver stations (BTS) and, second, how random is the data? The latter is important because it determines how good the correlation might be. Truly random data will have lower sidelobes resulting in better discrimination. The former is not uncomiected with the latter but it is of paramount importance in determining whether one base station can be detected in preference to another, i.e. is there sufficient difference between the two target base transceiver station (BTS) data streams to be able to reliably distinguish one base transceiver station (BTS) from the other?

An example multiframe is shown in the right-hand column of Figure 1. It consists of the following elements.

Frequency Correction Channel - FCCH in frames 0, 10, 20 etc.

Synchronisation Channel - SCH in frames 1, 11, 21 etc.

Broadcast Control Channel - BCCH in frames 2-5, 12-15, 22-25 etc. Access Grant Channel - AGCH and Paging Channel - PCH in frames 6-9, 12- 15 and 16-19.

Standalone Dedicated Control Channel - SDCCH in frames 26-29, 36-39. Cell Broadcast Channel - CBCH in frames 32-35. Slow Associated Control Channel - SACCH in frames 42-49.

The characteristics of these channels are summarised in Table 1.

Table 1 Summary of Control Channels

Table 1 attempts to characterise each of the channels, according to the two criteria of interest. The table shows a clear distinction between the FCCH, SCH and broadcast control channel (BCCH) and the other channels. The FCCH is of no use for correlation purposes as it only contains zeros, so only the synchronisation channel (SCH) and broadcast control channel (BCCH) are considered further.

The access grant channel (AGCH), paging channel (PCH), standalone dedicated control channel (SDCCH) and slow associated control channel (SACCH) channels are used for signalling and hence the data content, by definition, will vary from multiframe to multiframe. This means that this data would have to be distributed in real-time. It is not easy, partly due to the complexity and obscurity of ETSI specifications, to determine whether this data would exhibit random data characteristics. The most realistic option would be to examine a number of sets of real data.

The two channels BCCH and SCH have data that is largely static, that means that the majority of the information they contain does not change between multiframes. However, it may be that the data might not be different enough to be used as a discriminant between co-channel base transceiver stations (BTS).

The data in frames 1-5, being the SCH and the broadcast control channel

(BCCH), could quite easily be generated locally within the handset, as the majority of the data is available a priori from the SERVING MOBILE LOCATION CENTRE. The remainder, such as the frame number, could easily be updated from multiframe to multiframe. Hence, if the data in frame numbers 1 to 5 was sufficient to distinguish, via correlation, between base transceiver stations (BTS) then the extra performance provided by the ability to distinguish a further base transceiver station BTS may be gained without the necessity of continually passing a large amount of data between location measurement unit (LMU), Serving mobile location centre (SMLC) and handset.

In particular, the present invention provides advantages in many location estimation scenarios. A reasonably accurate position estimation requires a mobile station to identify signals from at least three different base stations. The invention may allow a mobile station to identify three base stations where previously only two, or one, could be identified. Even in a situation where three or more base stations could already be identified, the invention allows further base stations to be added, which improves the possible accuracy of the mobile station's position estimate. The mobile station 10 (Fig. 4) receives information telling it where it is. It can then identify some of the base stations that it should be able to receive.

Using the limited 'a priori' information it has, the mobile station can regenerate the leading five frames of the 51 multiframe for the particular base station it is looking for, then receive timing information to indicate approximately when the base station of interest should be transmitting those frames.

The mobile station then 'matches up', correlates, its internally- generated frame sequence, with received data streams, thereby identifying the base station transmitting the data stream. The identified base station may then be added to the repertoire of base stations received by the mobile station, in order to improve the mobile station's location estimate. Once a new base station has been identified, the method of the invention may begin again, with the objective of adding a further base station to the calculation.

The present invention addresses the requirement for improved location accuracy of the enhanced observed time difference (E-OTD) location estimation process. The enhanced observed time difference (E-OTD) location method requires a target handset 10 to make timing measurements of the transmission from a number of nearby base station transceivers. The conventional enhanced observed time difference (E-OTD) timing measurement uses 64 bits for correlation, being the length of the extended training sequence 91 in the synchronisation burst (Fig. 9), which is transmitted by the base station. The handset has a priori knowledge of this sequence and hence can undertake an auto correlation process to extract the relevant timing information by matching the a priori known extended training sequence to the received data stream. In a typical deployment there are only a small number of frequencies available and often the handset receives signals from two or more base stations on a same broadcast control channel (BCCH) frequency. The handset needs to be able to distinguish between these two or more base stations, but the same extended training sequence 91 is used by all base stations, and hence cannot be used to discriminate between base stations. The synchronisation burst (Fig. 9) is transmitted on the synchronisation channel that is a part of the information transmitted in timeslot (TS) 0 on the broadcast control channel (BCCH) carrier frequency. The synchronisation channel is a part of the information transmitted on the broadcast control channel (BCCH) carrier frequency, with the whole being described by the 51 multiframe 63.

In an embodiment of the present invention, the data used for the correlation is a demodulated, data stream corresponding to the GSM data rate, with 148 bits per burst. The sequential bursts would be concatenated together to form a single contiguous data stream. For example, as illustrated in Fig. 5, frames 1-5 of sequential multiframes are concatenated together.

The handset must therefore, as a prerequisite to making the timing measurement, identify a base station by decoding the relevant broadcast control channel (BCCH) data in the 51 multiframe. The minimum signal to noise ratio (SNR) that is required to decode the broadcast control channel (BCCH) data is higher than that required to obtain a sensible timing measurement. Accordingly, some base station signals may be too noisy to be decoded in order to provide base station identification in the usual manner. However, according to an aspect of the present invention, if the base station can be identified by an auto correlation process similar to that carried out for the timing measurement, then a timing measurement estimate could be made for a location at a lower SNR than that required to decode the broadcast control channel

(BCCH) signal, allowing the noisy base station signal to be identified and used by the handset 10 in calculating a position estimate with increased accuracy.

The present invention accordingly proposes using a sub-set of data from the 51 multiframe 65 to discriminate between two or more base stations transmitting on a same frequency. The data in frames 1-5 of the 51 multiframe are largely concerned with describing the transmitting base station and some of the elements are unique between base stations. According to an aspect of the present invention, the majority of the data in this sub-set of 5 frames, which is either static or can readily be calculated in the handset, be used in an auto correlation process to distinguish between base stations that use the same broadcast control channel (BCCH) frequency.

The handset 10 uses a priori known information to calculate an expected content of frames 1-5 of the multiframe from a required base station BTS. By correlating the a priori known subset of frames to the received signal, the base station

BTS transmitting a noisy signal may be identified, and timing measurements taken, thereby enabling a more accurate location estimate to be calculated.

The present invention particularly relates to handset processing. It is not thought feasible to mimic the uplink time difference of arrival (U-TDOA) process, where large amounts of data are stored, transferred and processed. This discussion will concentrate on the viability of using a sub-set of the data as mentioned in the preceding paragraph.

The purpose of the correlation is twofold, to synchronise the mobile station 10 to a base transceiver station (BTS) transmission, and to discriminate between different base transceiver stations (BTS) using a same broadcast control channel (BCCH) carrier frequency.

A GSM data bit has a duration of 3.6928 μs. Now in a GSM network, adjacent base transceiver stations (BTS) rarely have a radius greater than 35 km. Hence it is assumed that the target handset position can easily be estimated, by the serving mobile location centre (SMLC), to within ± 20 km. Now if we assume a propagation delay of 3 ns per metre, then ± 20 km corresponds to a propagation time of about 20 x

5 1000 x 3 ns, 60 μs, representing an error of approximately ± 16 data bits. It may be possible to significantly improve this estimate by the use of timing advance data ('« priori' known data).

In an unsynchronised network it is likely that the synchronisation bursts of two base transceiver stations (BTS) will be separated in time. In an embodiment of the 0 present invention, a timing estimate is obtained from both base transceiver stations (BTS) by correlating with a data sequence generated locally within the handset, for the sub-set defined by frame numbers 1 to 5, of the 51 multiframe.

One of the important performance criteria is the difference between the autocorrelation of the wanted data sequence, from one base transceiver station (BTS), and 5 the cross-correlation with the interference data sequence, from the other base transceiver station(s) using the same broadcast control channel frequency. The performance of various pseudo random sequences is well understood, and represents an optimally low cross-correlation response. The performance of the method of the present invention will not be as efficient as for a pseudo random sequence of the same 0 length, and it is not expected that this technique will be able to distinguish between two base transceiver station (BTS) on the same broadcast control channel (BCCH) carrier frequency if the 51 multiframes were significantly aligned.

The serving mobile location centre (SMLC) 41 knows the geometry of the base transceiver stations (BTS) and network measurements from the location 5 measurement unit (LMU), and also knows provide the frame number timing. Hence, the serving mobile location centre (SMLC) should be able to provide this data to the handset to tell the handset when to listen in order to receive a particular section of a multiframe from a particular base transceiver station (BTS). The serving mobile location centre (SMLC) should also be able to determine whether the multiframes 0 from two base transceiver station (BTS) are significantly time aligned, and which additional base transceiver station (BTS) would be most beneficial to the location estimate.

It may be considered impractical to ' transmit multiple multiframes from multiple base transceiver station (BTS j, k, m, n) via an Serving mobile location centre (SMLC) to a handset using location measurement unit location services protocol (LLP) and RRLP messaging. In addition, it may be considered that the handset would not be able to accommodate the extra data processing associated with the transfer of a significant amount of extra data. Hence, in order to realise a feasible solution the required data bandwidth needs to be reduced considerably.

Fig.4 illustrates an enhanced observed time difference (E-OTD) system, according to an embodiment of the present invention. The location measurement unit

(LMU) 40 receives a signal, measures and demodulates it to generate a demodulated data sequence for the base transceiver station of interest (e.g. BTS k). This data sequence is then communicated to the serving mobile location centre (SMLC) 41, via location measurement unit location services protocol (LLP) messaging 42, and then the serving mobile location centre (SMLC) would communicate this to the handset 10 via RRLP messaging 44.

The present invention provides methods and apparatus for improved location accuracy in location aware systems for portable communication device handsets, such as mobile telephones

Data, for example organised as the 51 multiframe 62 discussed with reference to Fig.l, is transmitted on selected timeslots (TS) of the broadcast control channel (BCCH) carrier. The 51 multiframe transmitted on timeslot TSO is not particularly suited to correlation, as its frequency correction bursts, contain 142 zeros, which appear as data discontinuities, significantly raising the spurious signal level.

It is assumed that the handset will demodulate the data, and be able to approximately align to the data bits. Also, it is assumed that the Serving mobile location centre (SMLC) will be able to predict the timing of the multiframe, to an accuracy greater than a single frame.

According to an aspect of the invention, the contents of frames 1 to 5 of the multiframe have been identified as containing data which is relatively static from multiframe to multiframe. According to an aspect of the invention the mobile station (MS) is adapted to internally calculate the contents of frames 1-5, thereby allowing these frames to be used for other purposes as discussed below. In this case the serving mobile location centre (SMLC) would only need to send the mobile station (MS) outline details of the base transceiver station (BTS) and an expected multiframe timing. The multiframe timing would most probably be related to the timing of the serving cell. The serving mobile location centre (SMLC) knows the timing of the respective base transceiver station (BTS) and should be able to calculate a reasonably accurate estimate of the relative timing at the mobile station (MS).

It may also be beneficial to concatenate a number of multiframe sub-sets, comprising frames 1-5 of several 51 multiframes, in order to improve the SNR. An example of such concatenation is shown in Fig.5. However, such concatenation could only be realistically achieved if the base transceiver station (BTS) multiframe sub-set of frames 1-5 can be readily identified, and located, in the data stream received at the handset.

The following sub-section provides a description of how the present invention may be employed to provide an improved location measurement function of a mobile station. (1) mobile station (MS) initiates an emergency 112 or 911 call.

(2) The serving mobile location centre (SMLC) receives location request, from the emergency service telephone exchange.

(3) The serving mobile location centre (SMLC) sends assistance data to mobile station (MS), including additional data for base transceiver station (BTS) and frame timing estimate.

(4) The serving mobile location centre (SMLC) sends a Measure Position Request.

(5) The mobile station (MS) scans and measures base transceiver station (BTS) broadcast control channel (BCCH) frequencies for all identifiable base transceiver stations.

(6) The mobile station (MS) determines which, if any, base transceiver station(s) (BTS) are missing from the identified base transceiver stations, and prioritises one base transceiver station (BTS) not yet measured, according to the assistance data, for identification according to an implementation of the present invention.

(7) The mobile station (MS) attempts to measure the additional base transceiver station (BTS), according to the present invention.

(8) The mobile station (MS) calculates and sends a Measure Position Response.

The present invention accordingly provides a method for improved calculation of extended correlation, to improve the location accuracy of enhanced observed time difference (E-OTD). There are problems with using the whole 51 multiframe (Fig.l) for autocorrelation, most notably the data discontinuities caused by the frequency correction burst transmitted in time slot 0 on the broadcast control channel (BCCH). There are also concerns that elements of the multiframe are not random, and hence the level of spurious correlation peaks will be relatively high. The present invention accordingly provides a system that minimises the extra data flow between the serving mobile location centre (SMLC) 41 and the handset 10. This takes advantage of a section of the 51 multiframe, within which the data is relatively static, thus minimising the data transmission requirement between the serving mobile location centre (SMLC) and handset. The system of the present invention is intended to be beneficial to the location estimate particularly through the provision of an extra base transceiver station (BTS k).

The following documents may be of assistance in the comprehension of the present invention, particularly where indicated in the description by their reference numeral. I. 3GPP timeslot (TS) 05.02, Radio Access Network; Multiplexing and multiple access on the radio path, V8.10.0 (2001-08)

Et. GSM 04.08, Mobile radio interface layer 3 specification, version 8.0.0 Release 1999.

The following document may also be of assistance: 3GPP timeslot (TS) 03.03, Numbering, addressing and identification, V7.6.0, June 2001.

Claims

CLAIMS:

1. In a mobile telephone network, a method for recognising a required base station (BTS k) which is transmitting a signal to a mobile terminal (10), the method comprising: transmitting, from the base station, a signal comprising data frames having a predictable data content, said content comprising content unique to the required base station (BTS k); receiving, in the mobile terminal, the signal comprising data frames having a predictable content; independently providing (44, 42) information to the mobile terminal sufficient to allow it to calculate the data content of the data frames having a predictable data content; calculating, in the mobile terminal, the data content of the data frames having a predictable data content; correlating, in the mobile terminal, the calculated data content with the data content of the received signal; upon detection of a correlation above a certain threshold between the calculated data content and the content of the received signal, recognising the required base station as the source of the received signal.

2. A method according to claim 2 further comprising the steps of: independently providing to the mobile terminal, sufficient information to enable it to predict the time of arrival of the data frames having a predictable content; predicting, in the mobile terminal, the time of arrival of the data frames containing predictable data content; and correlating, in the mobile terminal, the calculated data content with the data content of a received signal only within a predetermined timeframe based on the predicted time of arrival.

3. A method according to claim 2 wherein the step of independently providing sufficient information comprises the steps of receiving, in a location measurement unit, a signal transmitted by the required base station; demodulating the signal and generating a demodulated data sequence; corrununicating the demodulated data sequence to a serving mobile location centre (41), then cormiiunicating the data sequence to the mobile terminal.

4. A method according to any preceding claim wherein a plurality of base stations transmit signals on a same frequency; further comprising the step of: in response to the recognition of the required base station, ignoring signals from other base stations on the same frequency.

5. A method according to claim 4, wherein; a corresponding plurality of sets of information sufficient to allow to the mobile terminal to calculate the data content of data frames having a predictable data content from each of the plurality of base stations, is provided to the mobile terminal; the mobile terminal calculates the data content of the data frames having a predictable data content from each of the plurality of base stations; correlating, in the mobile terminal, the calculated data content of each of the sets of calculated data content with data content of the received signal; upon detection of a correlation above a certain threshold between one set of the calculated data content and the data content of the received signal, recognising the one of the required base stations as the source of the received signal; and repeating the method for at least one further base station.

6. A method according to any preceding claim further comprising the steps of: independently transmitting to the mobile terminal data sufficient to allow the mobile terminal to calculate a time of transmission of the data frames; detecting the time of arrival of the data frames at the mobile terminal; deriving a value of the transmission delay for the data frames; and using the derived value of transmission time in a calculation to determine the position of the mobile terminal with respect to at least one base station.

7. A method substantially as described, with reference to the accompanying drawings.