WO1999022545A1

WO1999022545A1 - Speed estimate method of a mobile part in a cellular telecommunication system

Info

Publication number: WO1999022545A1
Application number: PCT/EP1998/006976
Authority: WO
Inventors: Rossella De Benedittis; Giancarlo Rosina
Original assignee: Italtel Spa
Priority date: 1997-10-23
Filing date: 1998-10-19
Publication date: 1999-05-06
Also published as: ITMI972395A1; IT1295870B1

Abstract

Method to estimate the radial speed of a mobile unit (PP) inside the area covered by a digital cellular telecommunication system DECT. The estimate is made determining the variation of instants in which the receiving unit (mobile or fix) identifies the correlation word used to align in phase the slot synchronism of the receiving unit with the slot synchronism of the transmitting unit, inside an assigned time slot. The estimate is obtained comparing the variations in two consecutive measurement time slots, or in a sequence of instants averaged on more than two consecutive time slots.

Description

"SPEED ESTIMATE METHOD OF A MOBILE PART IN A CELLULAR TELECOMMUNICATION SYSTEM"

Field of the Invention

The present invention relates to a digital telecommunication system of the TDMA (Time Division Multiple Access) type with cellular structure, and more particularly it relates to a speed estimate method of a mobile unit inside the area covered by a digital telecommunication system , subdivided into a plurality of adjacent cells, each one equipped with a base station capable to manage the communications inside the cell, where the transmission of the useful signal takes place at predetermined time intervals, called "time slots" structured in frames

In cordless telecommunication systems such as the DECT (Digital Enhanced Cordless Telecommunications) employing small size cells, called microcells with radius in the range of 100-500 metres, a user travelling on a vehicle can easily reach a speed such to quickly cross different cells during a connection, with each crossing requiring a so-called "handover", that is a procedure controlling the passage of the connection under the control of the new cell The terminal or unit of a user who quickly moves through micro (or pico) cellular covering cells can therefore stress the network with an excessive number of handover requests Furthermore, a high number of handovers during a communication reduces the quality of the radio connection, particularly in systems like the DECT in which mtercell handovers (also defined "connection handovers") of the "seamless" type, that is without loss of information, are not assured The possibility to detect the movement speed of a mobile unit in due time, can therefore enable the optimization of the use of the resources (radio and non-radio) by the system operator Often, the area served by the DECT system is also covered by the cellular system GSM on which the "dual mode" portable part in quick movement could profitably be routed, thanks to the compatibility between the two systems Background art

Different proposals have been made to measure the speed of a mobile unit inside the area covered by a micro/pico cellular system, in order to deviate the mobile unit on a system having more extended cells (when these coexist and are able to mteroperate with the micro/pico cellular system), and therefore to control at best the system resources available In particular, WO 96/07279 describes an estimate method of the speed of the mobile unit based on measurements of the intensity of field received and on the relevant variations in time. The limit of these methods essentially consists in that the variations of the useful signal (due to multiple path or fading phenomena) can be caused also by the movement of object around the mobile unit and/or the base transceiver station, and can result low reliable. Objects of the Invention

The object of the present invention is to overcome the above mentioned drawbacks and limits. Summary of the Invention

The invention attains these objects through a method enabling to realize the estimate of the speed of the mobile unit measuring the variation in time of reception instants of the useful signal inside the time slot assigned to the same as described in claim 1. According to the invention, the above mentioned object is obtained through the estimate of the variation of instants in which, compared to a nominal reference, a radio (fixed or mobile) station correlates to the signal (i.e., it recognizes the same and identifies the starting) transmitted by the other radio part (mobile, fixed, respectively) inside a given time slot. In particular, the invention bases on the fact that the variation speed of correlation instants is tied to the radial movement speed of the mobile unit by a deterministic law. In the present description, the term "radial movement " means the motion of the mobile unit along the straight line connecting the two radio stations (fixed and mobile respectively). The correlation to the useful signal occurs recognizing a conventional word

(called correlation word) preceding the useful signal, that is the portion of signal destined to the receiving unit under examination, enabling it to align in phase its own slot synchronism with the slot synchronism of the transmitting unit.

The use of the correlation word of the DECT frame enables to obtain an accurate estimate of the radial speed, not altered by the noise of the surrounding environment, and without introducing dedicated circuits.

The solution proposed by the invention results less effective when the mobile unit moves along the cell periphery, however, in these conditions the stress/transition in adjacent cells by the mobile unit is reduced up to the point not to represent an important problem. The speed estimate can take place both in the base station and in the mobile one, and reference shall be made to this purpose hereafter to the receiving unit and to the transmitting unit, which can be, each one of them, in the base or in the mobile station. Additional features of the present invention are set forth in the appended claims. Brief description of figures.

The invention, together with further objects and advantages thereof, may be understood with reference to the following description referred to two preferred embodiments, taken in conjunction with the accompanying drawings, in which:

Figure 1 is a diagram showing the variation of the correlation instant according to the measurement time:

Figure 2 is a flow chart showing a first embodiment of the method according to the invention; Fig. 3 is a flow chart showing a second embodiment of the method according to the invention;

Making reference to the Figures, in both the method implementations according to the invention, to determine the speed of the mobile user, the instants are used in which the receiving station correlates, inside an assigned time slot, to the signal transmitted by the transmitting station, meaning by correlation the recognition and identification of the starting of the useful signal.

In the description of the method some parameters must be defined, namely:

• The minimum value of the speed of the mobile station, beyond which it is opportune that said mobile station moves towards the system having more extended cells;

• the stability, expressed in parts per million or ppm;

• the jitter of time references, and

• the size of the micro/pico cell.

For description sake, we shall assume hereafter that the speed estimate takes place in the fixed station DECT, that is the radio base station, called also RFP (Radio Fixed Part), and this shall not be construed as a limitation, and numeric values shall be assumed for the above mentioned parameters.

Making reference to the situation in which a mobile station DECT or PP (Portable Part) goes away from the fixed part DECT RFP inside the covering cell of this last, the deterministic law tying the variation, detected in the RFP, of the correlation instant ΔT_corr (in microseconds) in a second interval, at the movement speed V (expressed in m/sec) of the PP, is the following:

ΔTcorr = V . (2μs/300 m) The above mentioned law considers the fact that signals propagate in air at the light speed, and that the propagation delay affecting the signal transmitted by a PP and received by the RFP is tied to the double of the radio distance between PP and RFP.

According to a first embodiment of the method according to the invention, the lower limit of the movement speed of the PP is estimated comparing the variations of two correlation instants of the signal received by the RFP, averaged in two . consecutive measurement time slots.

The measurement interval (ΔTmis) for the estimate of the lower limit of the movement speed (V) beyond which the system operator desires that the control of the mobile unit is passed from a microcell to a macrocell, is selected according to the following factors:

1) Tmis must be sufficiently large to nullify the variations and/or oscillations of the correlation instant due to instability and inaccuracy of reference times of the mobile and base stations (jitter and p. p.m.); for DECT systems this condition translates into:

Tmis > 1 second;

2) Tmis must be sufficiently large in order that the variations of the correlation instant can be detected by the RFP; in particular, assuming that the minimum measurement solution is equal to the bit time duration Tbit, and therefore to 868 ns in the case of DECT system, the following bound shall be observed:

ΔTmis (s) > Tbit (μs) / ΔTcorr (μs/s). For DECT systems, and assuming that the minimum speed V of 50 km/h over which the transition towards a macrocell is requested, this condition translates into:

ΔTmis > 9.3 s. The measurement interval shall preferably have size equal to integer multiples of the minimum detection resolution of the station performing the measurement (i.e. 1 bit). In particular, said interval should be equal to:

N^*Tbit, with N integer > 1. In the above mentioned conditions, as small ΔTmis is, as handover requests are minimized (i.e. cell crossings) by the specific PP. In particular, at 50 km/h speed, the PP takes about 20 s to cross a cell having diameter equal to 300 mt, and therefore measurement intervals lower than or as much equal to 20 s can be considered satisfactory for an efficient application of the method according to the invention. Figure 1 is a diagram showing the variation of the correlation instant (in the y-axis) versus the measurement time (in the x-axis) for a movement speed of the PP equal to 50 km/h (assumed as wandering). K indicates the value of the correlation instant, versus a nominal reference, at the measurement starting (its value depends on the relative starting distance between PP and RFP). In the figure we assumed K equivalent to the bit time. Two lines are drawn in the diagram: the continuous line represents the real variation of the correlation instant; the hatched one represents the. variations of the correlation instant that the RFP is able to detect (that is by 1 bit steps).

For explicitness, the diagram of Figure 1 does not show the oscillations due to the inaccuracy and/or instability of reference times since said variations are substantially nullified performing the measurement in a sufficiently long time slot.

For speeds lower than the considered one, the minimum time to perform the measurement increases in a proportional way (e.g. for V = 25 km/h, the variation of a bit is recorded in about 18.6 s), but simultaneously increases also the staying time of the mobile station in the cell itself; therefore the method performance is maintained irrespective of the movement speed of the mobile station. Detailed description of a first embodiment of the Invention

Making reference to Fig. 2 we shall now describe more in detail the first embodiment of the method according to the invention. In particular:

• called "Ti" the instant in which RFP correlates to the signal transmitted by the PP in the i-th DECT frame;

• defined a suitable measurement interval ΔTmis on the basis of instabilities and/or jitter of time references and of the speed limit value V beyond which the transition of the PP towards the macrocellular system is desired,

• called "M" the number of frames contained in the pre-set measurement interval ΔTmis, are calculated two mean values (in microseconds) of the correlation instant (Tlmean and T2mean) in two consecutive ΔTmis intervals: i=M Tl mean = (1/M)^* ∑Ti i=1 and i=2M

T2mean = (1/M)^* ∑Ti i=M+1 The movement speed V of the mobile part is estimated starting from the two average values obtained above, as follows: V = [|(T2mean - T1 mean)|/ ΔTmis]*300/2

If the V value obtained is higher than or equal to the pre-set limit value, a change of cell shall be requested to the mobile part.

In relation to the above, the method described referring to said first preferred embodiment enables to precisely calculate if the movement speed of the mobile is higher than, or as much equal to the limit value set by the system operator; however it requires a comparatively long measurement time. Detailed description of a second embodiment of the Invention

According to an embodiment of the invention, the method applied enables to estimate the actual movement speed of the mobile unit, comparing a sequence of instants in which the RFP correlates to the signal transmitted by the PP averaged in consecutive measurement intervals.

The measurement interval (ΔTmis) for the estimate of the movement speed (V) is dimensioned in order to eliminate the jitter and the instabilities of reference times (e.g. it can be equal to 1 s for DECT systems). Making reference to Fig. 3 it shall be described more in detail the second embodiment of method according to the invention. In particular:

• called "Ti" the instant in which RFP correlates to the signal transmitted by the PP in the i-th DECT frame; • defined a suitable measurement interval ΔTmis on the basis of instabilities and/or jitter of time references;

• called "M" the number of frames contained in the measurement interval ΔTmis, are calculated as many mean values of the correlation instant (Timean) in consecutive measurement intervals as are required to detect variations that can be measured by the system between two adjacent mean correlation values. As soon as it occurs, a time counter Tmis is enabled, realized for instance counting the number of frames elapsing from its enabling to its stop, going on with the calculation and comparison of mean values of the correlation instants in subsequent measurement intervals up to the next difference that can be measured by the system between two consecutive mean values. As soon as it takes place, said difference is stored (ΔTmean), the time counter Tmis is "stopped/frozen", proceeding to the estimate of the movement speed V of the mobile unit using the following formula: V (mt/s) * [ΔTmean/ Tmis] ^* 300/2

If the V so calculated results higher than, or equal to the limit speed set in advance by the operator, the mobile shall be requested to pass to a macrocell.

This embodiment of the method according to the invention has the advantage to generally request a measurement time lower than that requested by the first embodiment. For instance, for a limit speed of 50 km/h, the measurement time is averagedly 14 s (min. 9,3 s and max. 18,6 s) versus the fixed time of 18,6 seconds requested in the first embodiment. However, the estimate of the movement speed of the mobile is affected by inaccuracy linearly tied to the ΔTmis value. For instance, a 50 km/h speed, in the case of ΔTmis = 1 s, is estimated equal to 0.868/10^*150 = 13.02 m/s = 47 km/h; while a 100km/h speed, is estimated equal to 94 km/h.

However, this imprecision is more than acceptable for the subject applications. Therefore, while two particular embodiments of the present invention have been shown and described, it should be understood that it is not limited thereto since other embodiments may be made by those skilled in the art without departing from the scope thereof. It is thus contemplated that the present invention encompasses any and all such embodiments covered by the following claims.

Claims

CLAIMS 1. Method to estimate the radial speed of a mobile unit (PP) inside the area covered by a digital cellular telecommunication system in which each cell is served by a radio fixed part (RFP) placed in central position compared to the cell, meaning by the term radial speed the speed component of the mobile equipment coinciding with the axis hypothetically connecting the mobile part to the fixed part, characterized in that said estimate is made determining the variation of the instants in which, versus a nominal reference, the mobile unit (the radio fixed part respectively) correlates to the signal transmitted by the base station (by the mobile unit respectively) inside an assigned time slot.

2. Method according to claim 1 , characterized in that said digital cellular telecommunication system is a DECT system.

3. Method according to claim 2, characterized in that said estimate is made in the base station and in that: ΓÇó said nominal reference consists of a reference signal inside the base station itself;

ΓÇó the correlation instant coincides with the recognition of the correlation word transmitted to enable the alignment in phase of the slot synchronism of the receiving unit with the slot synchronism of the transmitting unit.

4. Method according to claim 2, characterized in that said estimate is made in the mobile unit and in that:

ΓÇó said nominal reference consists of a first recognition of the correlation word transmitted to enable the alignment in phase of the slot synchronism of the receiving unit with the slot synchronism of the transmitting unit;

ΓÇó the above mentioned variation of instants in which the mobile unit correlates to the signal transmitted by the base station is determined measuring the time slot elapsing between said first recognition of the correlation word and at least a successive recognition of such word.

5. Method according to claims 2 or 3, characterized in that said estimate is obtained comparing the variations of two correlation instants of the signal received by the receiving station in two consecutive measurement time slots.

6. Method according to claim 5, in particular to transfer the control of the mobile unit from a cell to a cell having larger size or macrocell covering the same area when a pre-set radial speed is exceeded, characterized in that it foresees the following operational phases: a) - define a lower limit of the movement speed of the mobile unit (PP); b) - define a measurement interval ╬öTmis; c) - calculate two mean values (Tl mean and T2mean) of the correlation instant (Ti) in two consecutive intervals ╬öTmis: i=M

Tl mean = (1/M) * ΓêæTi i=1 and i=2M T2mean = (1/M) * ΓêæTi i=M+1 , where M is the number of frames contained in the per-set measurement interval

╬öTmis; d) - estimate the radial speed (V) of the mobile according to the relation: V = [|(T2mean - T1 mean)]/ ╬öTmis] * 300/2 e) - compare the speed value (V) so obtained with the lower limit of the movement speed.

7. Method according to claims 3 or 4, characterized in that said estimate is obtained comparing a sequence of instants in which the receiving unit correlates to the signal transmitted, averaged on more than two consecutive measurement intervals.

8. Method according to claim 7, in particular to transfer the control of the mobile unit from a cell to a cell having larger size or macrocell covering the same area when a given radial speed is exceeded, characterized in that it foresees the following operational phases: a) - define a lower limit of the moving speed of the mobile unit (PP); b) _ define a measurement interval ╬öTmis; c) - calculate a plurality of mean values of the correlation instant (Timean) according to the relation: i=M

Timean = (1/M) * ╬ú Ti i=1 at consecutive measurement intervals, where M is the number of frames contained in the measurement interval set ╬öTmis; d) - detect when, between two adjacent correlation mean values, a variation exists that can be measured by the system and enable a time counter (Tmis); e) - repeat steps c. and d. up to obtain a successive difference which can be measured by the system between two consecutive mean values; f) - store such difference (╬öTmean) and stop said time counter (Tmis); g) - estimate the radial speed (V) of the mobile according to the relation:

V = [╬öTmean/ Tmis] * 300/2 h) - compare the speed value (V) so obtained with the lower limit of the moving speed.