GB2358109A

GB2358109A - Power control during soft handoff

Info

Publication number: GB2358109A
Application number: GB0000437A
Authority: GB
Inventors: Behzad Mohebbi; Arnaud Mathwrin Georges Druet
Original assignee: Fujitsu Ltd
Current assignee: Fujitsu Ltd
Priority date: 2000-01-10
Filing date: 2000-01-10
Publication date: 2001-07-11
Anticipated expiration: 2020-01-10
Also published as: GB0000437D0; GB2358109B

Abstract

A mobile station in communication with two or more base stations during soft handoff measures the received signal strength from each base station. The difference in received signal strengths is compared against a threshold and if it is exceeded, a power imbalance message is transmitted to the network. Either the mobile station or the network may calculate the adjustment which is necessary to reduce the imbalance. In determining any necessary adjustment the mobile station may determine the path loss of the pilot signal received from each base station and also take into an intended transmission power offset set by the network.

Description

2358109 SOFT HAND-OFF IN CELLULAR MOBILE COMMUNICATIONS NETWORXS The

present invention relates to soft hand-off in cellular mobile communications networks. In particular, the present invention relates to controlling the downlink transmission powers of active cells involved in a soft hand-off operation.

Figure 1 of the accompanying drawings shows parts of a cellular mobile telecommunication network according to the Telecommunication Industries Association (TIA) /Electronic Industries Association (EIA) Standard TIA/EIA/IS-95 of October 1994 (hereinafter 11IS9511). Each of three base transceiver is stations (BTSs) 4 (BTS1, BTS2 and BTS3) is connected via a fixed network 5 to a base station controller (13SC) 6, which is in turn connected to a mobile switching centre (MSC) 7. The BSC 6 serves to manage the radio resources of its connected BTSs 4, for example by performing handoff and allocating radio channels. The MSC 7 serves to provide switching functions and coordinates location registration and call delivery.

Each BTS 4 serves a cell 8. When a mobile station (MS) 10 is in a so-called "soft hand-off I, (SHO) region 9 where two or more cells overlap, a mobile station can receive transmission signals (downlink signals) of comparable strength and quality from the respective BTSs of the overlapping cells. Transmission signals (uplink signals) produced by the mobile station (MS) can also be received at comparable strengths and qualities by these different BTSs when the mobile station is in the SHO region 9.

Figure 2 of the accompanying drawings shows a situation where the MS 10 is located within the SHO region 9, and is transmitting such uplink signals that are being received by plural BTSs 4. According to the IS95 standard, a BTS 4 that receives such an uplink signal from the MS 10 relays the signal to the BSC 6 via a dedicated connection line of the fixed network 5.

At the BSC 6, one of the relayed signals is selected based on a comparison of the quality of each of the received signals, and the selected signal is relayed to the MSC 7. This selection is referred to as Selection Diversity.

Similarly, Figure 3 of the accompanying drawings shows a situation where the MS 10 is located within the SHO region 9 and is receiving downlink signals from plural BTSs 4. According to the IS95 standard, downlink signals received by the BSC 6 from the MSC 7 is are relayed to all BTSs 4 involved in the soft hand-off via respective connection lines of the fixed network 5, and subsequently transmitted by all the BTSs 4 to the MS 10. At the MS 10 the multiple signals may be combined, for example, by using maximum ratio combination (MRC), or one of them may be selected based on the signal strength or quality, i.e. using Selection Diversity as for the uplink case.

In contrast to, for example, Global System for Mobile Communication (GSM) networks, in CDMA networks each BTS 4 transmits at the same frequency.

Consequently, careful control of transmission power must be maintained to minimise interference problems.

Signals are transmitted as a succession of frames according to the IS95 standard. As Figure 4 of the accompanying drawings shows, each frame is of duration ms, and comprises sixteen 1.25 ms time slots. In each time slot several bits of user data and/or control information can be transmitted.

Power control of transmissions from the MS 10 to the BTSs 4 (uplink power control) in IS95 is achieved as follows. When a BTS 4 receives a signal from the MS it determines whether a predetermined property of the received signal (for example absolute signal level, signal to noise ratio (SNR), signal -to- interference ratio (SIR), bit error rate (BER) or frame error rate (FER)) exceeds a preselected threshold level. Based on this determination, the BTS 4 instructs the MS 10 either to reduce or to increase its transmission power in the next time slot.

For this purpose, two bits in every time slot of a pilot channel (PCH) f rom. the BTS 4 to the MS 10 are allocated for uplink power control (see Figure 4).

Both bits have the same value, and accordingly will be referred to hereinafter as the "transmission power control bit" (or TPC bit) in the singular. The TPC bit is is assigned a value of zero by the BTS 4 if the MS 10 is required to increase transmission power by 1 dB, and a value of one if the MS 10 is required to decrease transmission power by 1 dE. The BTS 4 is not able to request directly that the MS 10 maintain the same transmission power; only by alternately transmitting ones and zeros in the TPC bit is the transmission power maintained at the same level.

When the MS 10 is in a SHO region 9, the MS 10 is required to make a decision on whether to increase or to decrease uplink transmission power based on a plurality of TPC bits received respectively from the BTSs 4 involved in the soft hand-off. Consequently, an OR function is performed on all the TPC bits. If the result of this OR function is zero then the MS 10 will increase power on uplink transmissions, and if the result is one then the MS 10 will decrease power on uplink transmissions. In this way, uplink transmission power is only increased if all BTSs 4 ask for an increase.

In the IS95 standard, power control of transmissions from the BTS 4 to the MS 10 (downlink power control) during soft handoff is not performed in the same way as for the uplink power control case. In particular, TPC bits are not sent from the MS 10 to the BTS 4.

However, in wideband CDMA (W-CDMA) networks such as the proposed European W-CDMA, network UTRA that are currently being designed, the use of TPC bits for downlink power control is being considered. UTRA stands for UMTS Terrestrial Radio Access, and UMTS stands for Universal Mobile Telecommunications System (a third generation mobile communications system). In this case, when the MS 10 receives a signal from each is BTS 4 it determines whether a predetermined property of the received signal (for example absolute signal level, SNR, SIR, BER or FER) exceeds a preselected threshold level. Based on this determination, the MS 10 instructs the BTS 4 either to reduce or to increase its transmission power in the next timeslot.

Referring now to Figure 5, a problem which can arise in controlling the downlink transmission powers of the active cells during a soft hand-off operation will be explained. In Figure 5, a MS 10 is in soft hand-off with two base transceiver stations BTS1 and BTS2. As previously indicated, for uplink power control the MS 10 receives TPC bits both from BTS1 and from BTS2. If either or both of BTS1 and BTS2 sends a "power down" TPC bit in a particular timeslot the MS 10 will reduce its uplink transmission power for the next timeslot. Thus, as illustrated in Figure 5, even when BTS2 issues a "power up" TPC bit at the same time as BTS1 issues a "power down" TPC bit, the MS 10 reduces its uplink transmission power. Because the MS 10 will always follow power-down TPC commands, the MS 10 is effectively power- controlled by the BTS which, at any given time, is enjoying the best-available uplink channel. The consequence of this is that any base station such as BTS2 which requests and needs a higher transmission power from the MS 10 may not be able to maintain its target received signal level (e.g. target SNR) for uplink signals, so that the BTS concerned will suffer a greater BER in the user data and control information received from the MS 10.

In the proposed UTRA network mentioned above, this control information sent from the MS 10 to each active BTS in soft hand-off includes TPC bits for use by the BTSs concerned to control their respective downlink transmission powers to the MS 10. The MS 10, however, may be transmitting its uplink signals (including the is control information) with a transmission power sufficient only to meet the SNR requirements at BTS1 and not sufficient to meet the SNR requirements at BTS2. This means that BTS2 will experience a high BER on the control information (TPC bits) it receives from the MS 10. This increased BER at BTS2, even if only momentary, can only be tolerated if the errors are symmetrical, i.e. there are as many "wrong" power-up commands (TPC bits which should have been power-down commands but were received as power-up commands) as there are "wrong" power-down commands (TPC bits which should have been power-up commands but were received as power-down commands).

In theory, when an infinite sample of TPC errors is considered, the number of wrong power-up commands would be equal to the number of wrong power-down commands. However, for any small sample of TPC errors, such as might be experienced over several timeslots, it is quite possible that a number of consecutive power down (is) or power-up (Os) errors can be experienced at BTS2.

If, for any reason, there is an imbalance in the number of power-up and power-down errors, a base station such as ETS2 in Figure 5 can end up transmitting too little or too much power in the downlink direction for the remainder of the soft hand off period. This will lead to a "permanent" power imbalance between the active BTSs involved in the soft hand-off operation, and this imbalance can be tens of dB. A power imbalance of this nature will lead to a non-optimum soft handoff operation, as well as asymmetric expansion of the soft handoff region and potentially severe loss of network capacity.

It is therefore desirable to provide a soft hand off method in which such a power imbalance situation is can be identified conveniently and reliably. It is also desirable to provide a convenient and simple mechanism for curing such a detected power imbalance.

According to a first aspect of the present invention there is provided a soft hand-off method, for use in a cellular mobile communications network, including the steps of: when a mobile station of the network is involved in soft hand-off with two or more base transceiver stations of the network, detecting an imbalance amongst the respective downlink transmission powers used by the two said or more involved base transceiver stations to transmit information signals to the mobile station; and following detection of such a power imbalance, adjusting the said downlink transmission power used by at least one of the said involved base transceiver stations to transmit its said information signals to the mobile station, so as to reduce the detected imbalance.

According to a second aspect of the present invention there is provided a mobile station, for use in a cellular mobile communications network, including:

power imbalance detection means operable, when the mobile station is involved in soft hand-off with two or more base transceiver stations of the network, to detect an imbalance amongst the respective downlink transmission powers used by the said two or more involved base transceiver stations to transmit information signals to the mobile station; and power imbalance notification means operable, following detection by the said power imbalance detection means of such a power imbalance, to transmit a power imbalance message to the network to indicate that adjustment of the downlink transmission power of at least one of the said involved base transceiver stations is required.

According to a third aspect of the present is invention there is provided a radio network controller, for use in a cellular mobile communications network, including: power imbalance notification receiving means operable to receive a power imbalance message transmitted by a mobile station of the network when the mobile station detects an imbalance amongst the respective downlink transmission powers used by two or more base transceiver stations, with which the mobile station is involved in soft hand-off, to transmit information signals to the mobile station; and power adjustment means, responsive to receipt by the power imbalance notification receiving means of such a power imbalance message, to calculate a power adjustment, if any, required in the said downlink transmission power of each involved base transceiver station and to cause each involved base transceiver station to adjust its said downlink transmission power in accordance with the calculated power adjustment.

According to a fourth aspect of the present invention there is provided a radio network controller, for use in a cellular mobile communications network, including: power imbalance detection means operable, when a mobile station of the network is involved in soft hand-off with two or more base transceiver stations of the network, to detect an imbalance amongst the respective downlink transmission powers used by the said two or more involved base transceiver stations to transmit information signals to the mobile station; and power adjustment means operable, following detection by the said power imbalance detection means of such a power imbalance, to calculate a power adjustment, if any, required in the said downlink transmission power of each involved base transceiver station and to cause each involved base transceiver station to adjust its said downlink transmission power in accordance with the calculated power adjustment.

is According to a fifth aspect of the present invention there is provided a cellular mobile communications network, including: a mobile station; a plurality of base transceiver stations; power imbalance detection means operable, when the said mobile station is involved in soft hand-off with two or more of the said base transceiver stations of the said plurality, to detect an imbalance amongst the respective downlink transmission powers used by the said two or more involved base transceiver stations to transmit information signals to the said mobile station; and power adjustment means operable, following detection of such a power imbalance by the said power imbalance detection means, to adjust the said downlink transmission power used by at least one of the said involved base transceiver stations to transmit its said information signals to the mobile station so as to reduce the detected imbalance.

Reference will now be made, by way of example, to the accompanying drawings, in which:

Figure 1, discussed hereinbefore, shows parts of a cellular mobile communications network according to IS95; Figure 2, also discussed hereinbefore, shows a schematic view for use in explaining processing of uplink signals in a soft hand-off operation performed by the Figure 1 network; Figure 3, also discussed hereinbefore, shows a schematic view for use in explaining processing of downlink signals in such a soft hand-off operation; Figure 4, also discussed hereinbefore, illustrates the format of a time frame in the Figure 1 network; Figure 5, also discussed hereinbefore, is a schematic view for illustrating a problem which can arise in a soft hand-off operation in a proposed UTRA network; and is Figure 6 shows a schematic view f or use in explaining a method of detecting power imbalance amongst active BTSs in a cellular communications network embodying the present invention; Figure 7 shows parts of a mobile station embodying the present invention; and Figure 8 shows in detail one of the parts of the Figure 7 mobile station.

Figure 6 shows a schematic view f or use in explaining how a power imbalance between active BTSs during soft hand-off can be detected in an embodiment of the present invention. The explanation given here relates to a UTRA network. However, embodiments of the present invention can be adapted to work in any cellular mobile communications network in which power imbalances between the active BTSs can arise in soft hand-off.

In Figure 6, a mobile station MS is in soft hand of f with two base stations BTS A and BTS B. BTS A broadcasts to all mobile stations in receiving range so-called pilot signals which are used by the mobile stations f or various purposes including BTS selection.

In one proposed version of the UTRA network these pilot signals are transmitted via a common pilot channel (CPICH). The pilot signals from BTS A are transmitted at a fixed transmission power T,,, and this power TA, is also made available to mobile stations in receiving range of BTS A by a BCH message broadcast by the BTS A via a broadcast channel BCHA thereof.

BTS A also has a downlink traffic channel allocated by the network for communication in the downlink direction from BTS A to the MS. In the proposed UTRA network, this downlink traffic channel (or forward traffic channel) may be referred to as a dedicated physical channel (DPCH) This DPCH is made up of a dedicated physical data channel (DPDCH) and a dedicated physical control channel (DPCCH). Thus, BTS A in Figure 6 uses DPCH, to send data and control information in the downlink direction to the MS.

The downlink transmission power TA, for DPCHA is adjusted in use of the network using TPC commands (TPC bits) transmitted from the MS in each timeslot.

BTS B also exchanges with the MS signals corresponding to those exchanged between the MS and BTS A. Thus, BTS B has a common pilot channel CPICHB which is transmitted at a fixed power T,,p. It also has a broadcast channel BCH, through which it sends a BCH message indicating its pilot signal transmission power T... Further, BTS B has a downlink traffic channel DPCH, allocated to the MS, and this downlink traffic channel is transmitted at an adjustable power level TI3T.

The power level T1IT is adjusted in dependence upon the TPC bits sent f rom the MS in each timeslot.

In the present embodiment an undesirable power imbalance between BTS A and BTS B is detected at the MS, as follows.

The MS receives the pilot signals broadcast by each BTS via its CPICH. For each CPICH, the MS measures the power level at which the CPICH signals are received (received power or received signal strength RSS). Thus, for CP1CH,, the MS produces a received power measure R... Similarly, for CPICH, the MS produces a received power measure R,,p, The MS also receives the BCH messages transmitted by STS A and BTS B via the respective broadcast channels BCHA and BCH, thereof. In each case, the received BCH message is decoded to extract therefrom the CPICH transmit power of the BTS concerned, i.e. TAp in the case of BTS A and T,,p in the case of BTS B. Using these CPICH transmission power levels, the MS can then calculate a path loss PLA or PL,, f or each CPICH as f ollows:

is PLA",,,, TAP- RAP PLB =TBp - RBp (2) The MS is also able to measure the receive power of each of the downlink traffic channels DPCHA and DPCH.. The MS produces a receive power measure RAT for DPCHA, and a receive power measure RBT for DPCH In the proposed UTRA network currently under investigation, it is not presently envisaged that explicit information relating to the instantaneous downlink transmission power of the DPCH allocated to a mobile station will be transmitted from the BTS concerned to the MS. However, it can validly be assumed that the path loss for the DPCH will be approximately the same as the path loss for the CPICH.

Accordingly, in the present embodiment, the transmission power TAT of DPCH, can be estimated at the MS as:

12- TAT= RAT+ PLA... (3) Similarly, the transmission power TBT of DPCH, can be estimated at the MS as:

TBT= RBT + PLB (4) Using equations 3 and 4 above, the MS can estimate TAT and T1IT and thus detect when there is an imbalance between them. For example, the MS can monitor the difference AT between TAT and TST:

AT =TAT-TBT... (5) This power imbalance measure AT, however, should not be compared directly with a fixed threshold value to detect a power imbalance because it will often be the case that BTS A and BTS B will deliberately be assigned different "reference" transmission powers.

These reference powers effectively are used to define the size of the cell served by the BTS concerned and may be varied from time to time to reflect different conditions in the cell. For example, the reference power of a BTS may be reduced by the network when the number of MSs in the cell reaches a desired maximum loading level for the cell, so as to keep interference within tolerable limits. The CPICH transmit power TAP or TBI is a (direct or indirect) measure of the assigned reference power of the BTS concerned, and the MS can therefore obtain a measure ATOs of the intended power imbalance between BTS A and BTS B using the BCH messages:

Tos = TAP- TBP... (6) Incidentally, it is also possible that the traf f ic channels DPCH of BTS A could be assigned an intended power offset from the traffic channels DPCH of BTS B, which power offset is different from the power offset of the assigned pilot-channel (CPICH) transmission powers (TAp and T.p). In this case, the BCH messages transmitted by the base stations include information specifying the intended DPCH power offset.

Then, to account for the intended power imbalance, the dif f erence between AT and AT0s can be compared with a suitable preselected threshold value THR:

is 1AT -ATosi>THR Thus, when 1 AT - AT,), 1 exceeds THR the MS determines that an unduly large unintended power imbalance has arisen.

In practice, the threshold value THR should not be set so low that power imbalances of small (and therefore manageable) proportions are detected. if such small power imbalances were detected and acted upon (as described in more detail below) significant network resources would be expended that would have adverse consequences for the network as a whole. For example, significant extra signalling traffic would be generated which would reduce the amount of data and other control information which could be transmitted at any given time. On the other hand, there will be a magnitude of power imbalance at which the amount of extra signalling and other network resources expended in detecting and dealing with the power imbalance can be justified as the consequences of leaving the power imbalance uncorrected, for example in terms of lost -14 capacity in the network, would far outweigh any capacity loss due to extra signalling. For example, the preselected threshold value THR may be 5dB.

once a power imbalance above the threshold value has been detected, the MS informs the network of the imbalance using a power imbalance message (PIM). This PIM indicates that adjustment of the power(s) of one or more of the BTSs involved in the soft hand-off operation is required. The PIM is, for example, a layer 3 message in the UTRA network, that is to say a network layer message transmitted from the MS to the BSC via one or more BTSs. Incidentally, in the UTRA network the BSC is sometimes referred to instead as a radio network controller RNC. This term RNC will be is used hereinafter.

There are a number of different approaches which can be used to bring about adjustments in the downlink transmission powers of the active BTSs following detection of a power imbalance.

A first approach is to use the RNC to determine the power adjustments. In this approach, in response to receipt of the PIM message the RNC interrogates the active BTSs to obtain therefrom their respective downlink transmission powers to the MS which produced the PIM. Alternatively, the RNC may already monitor and so have access to this information when the PIM is received.

A first possibility is then to adjust the power levels of all of the active BTSs to be the same as that of the BTS having the highest transmission power. It will be appreciated that the respective cells of the BTSs involved in the soft hand-off operation can have different respective sizes, due for example to different loading conditions or any other difference between the cells. This means that, as between any two BTSs, there will usually be a power offset (nominal offset) which is intended and specified by the network and is not caused by any TPC errors. This power offset corresponds to the difference in reference power levels set for the different BTSs, for example the difference between the CPICH transmission power TAP of BTS A and the CPICH transmission power T,,p of BTS B in Figure 5.

After any power adjustment has been performed, this power offset should of course be maintained. For example, say the intended power offset between BTS A and BTS B is ATOs (= T,,p - TBP). Also, assume that when the RNC compares the respective downlink transmission powers T, and T,, for the active BTSs, BTS A is found to be the BTS having the highest transmission power. In this case, the downlink transmission power for BTS B is is set to the downlink transmission power of BTS A less the power offset AT0s.

A second possibility is to adjust the transmission powers of all the active BTSs to be the same level as that of the BTS having the lowest transmission power.

A third possibility is to adjust the BTS downlink transmission powers to a preset level. Any intended power offsets ATOs between the active BTSs are of course applied to the preset level.

These three possibilities have the advantage of being simple and quick to implement commensurate with the short timeslot durations in the UTRA network.

A fourth possibility, requiring more processing power than each of the first and second possibilities, is to calculate a new power level f or each active BTS in dependence upon the existing power level of that one of the active BTSs which is experiencing the better uplink reception conditions. This possibility must be treated with caution, however, as the quality of the uplink signals is not necessarily correlated with that of the downlink signals. Indeed, it is partly for this reason that a power imbalance in the downlink transmission powers occurs.

However, in principle a BTS (e.g. BTS A) that is experiencing good-quality uplink signals is likely to have faithfully followed the TPC commands issued by the MS for its downlik transmission power. In this case, it is reasonable to assume that the downlink transmission power TAT of BTS A is likely to be more correct than the downlink transmission power T1IT of a BTS such as BTS B which is experiencing poor-quality uplink signals. Thus, upon receiving a PIM the RNC may leave the downlink transmission power TAT of the best uplink signal quality BTS A unchanged and adjust the downlink transmission power TBT Of BTS B to be the same as that of BTS A (after accounting for any intended is power of f set AT0s between BTS A and BTS B, i. e. TBT =TAT - AT0s).

The quality of the uplink signals received by each active BTS is known at the RNC as maximum ratio combining (MRC) or selection diversity of the uplink signals is carried out at the RNC. The quality measure used may be a history of the quality of the uplink signals over the period up to the power imbalance detection.

Instead of leaving the downlink transmission power TAT of the BTS A having the best uplink signal quality unchanged in response to the PIM, the RNC may also increase that power in a situation in which the power TST Of the BTS B will be decreased in a single step in the power adjustment operation carried out by the RNC in response to the PIM. Such an increase may be helpful because in a power imbalance situation in which BTS B is transmitting at too high a power, the MS will be sending power-down TPC commands but only BTS A (having the better uplink signal quality) is following the commands. Thus, by the time the power imbalance is rectified, the transmission power of BTS A may well have been reduced to a value below the appropriate level applicable when there is no power imbalance (i.e.

when BTS B is restored to the right level) Because of this, some form of weighting is preferably applied to increase the power TAT of BTS A at the same time as the power TBT of BTS B is reduced.

In all of the four possibilities mentioned above, instead of adjusting the BTS downlink transmission powers to the calculated new level immediately, the adjustment can be performed progressively over a number of timeslots. This can be useful in reducing the adverse effects of a possible "rogue" detection of a power imbalance between the active BTSs, i.e.

inadvertent detection of a power imbalance due to is random errors when no actual power imbalance exists.

For example, the steps could be such as to cause the BTS downlinktransmission powers to converge to the calculated level over 5 or 10 timeslots.

When progressive adjustment of the BTS downlink transmission powers is used, it is also possible for each BTS whose power is to be adjusted to employ a so-called adjustment loop of the kind described, for example, in a paper prepared by NEC entitled "Adjustment Loop in downlink power control during soft handoverll, TSG-RAN Working Group 1 meeting #7bis, Kyongju, Korea,October 4-5, 1999. In this paper, an adjustment loop is described for balancing downlink powers among active-set cells during soft hand-off. To implement the adjustment loop, a downlink reference power PRZF and downlink power convergence coefficient r (0<r<1) are set in the active-set cells during soft hand-of f so that P,, and r are used in common by all of the active-set cells. The adjustment loop works in addition to TPC commands (inner loop power control) Take, for example, a situation in which two BTSs, BTS A and BTS B are involved in soft hand-off. In each 18 timeslot i, each BTS calculates a new downlink transmission power for the next timeslot, as follows:

T,,(i + 1) = TIT(i) +(1 - r) (P,, - TIT (i)) +TPCAT(i) TBT(i+l)=TBT(i) +(1 - r) (P, - TBT(i)) +TPCBT(i) Assuming that TPC errors do not occur, i.e. TPCAT (i) = TPCBT (i) then the difference between the downlink transmission powers becomes: 10 TAT(i + 1) - TBT(i + I)= r (TAT W _TBTO) r i (TAT (1) _TBT (1)) It follows that the difference between the two powers TAT and TIT converges at 0 when r is smaller than 1.

Thus, in an embodiment of the present invention following detection of a power imbalance, the BTSs can be supplied with the parameters PREY and r by the RNC and convergence of the downlink transmission powers can be initiated.

In the first approach described above, the power adjustments were performed by the RNC after receiving a power imbalance message from the MS. However, in a second approach it is also possible for the MS to calculate the required power adjustments for the BTSs involved in the soft hand-off operation. Two such methods using the second approach will now be described in detail with reference to Figures 7 and 8.

Incidentally, it is also possible for the RNC to detect the power imbalance independently of the MS, although this is less reliable than using the MS to detect the imbalance. The way this is done depends on the information normally made available to the RNC by the BTSs it controls. In a network in which each BTS supplies to the RNC the TPC commands actually received from the MS the RNC can directly compare the TPC commands received at the different BTSs and so detect a TPC error at one of the BTSs when there is a mismatch amongst them, e.g. a power-up command at one BTS with power-down commands at all the remaining BTSs.

However, in other networks the TPC commands may be treated (like the pilot bits) as being layer-1 messages which are not forwarded from the receiving BTS to the RNC. In this case, the network would need to be is modified suitably to enable the RNC to detect a power imbalance. For example, the BTSs could regularly report their downlink transmission powers to the RNC which, knowing any intended power offset between the BTSs, could then detect a power imbalance as in equation 7 above. Alternatively, the TPC commands themselves received from the MS could be forwarded to the RNC by the BTSs. These approaches are generally less advantageous, however, as they increase the amount of signalling (backhaul) in the network from the BTSs to the RNC. For this reason, detection of power imbalance in the MS and reporting using a layer-3 message to the RNC is preferable.

Figure 7 shows parts of a mobile station for use in an embodiment of the present invention. The mobile station 20 has an antenna portion 22 connected (e.g.

via a duplexer - not shown) to a receiver portion 24 and a transmitter portion 26. The mobile station 20 also includes a soft hand-off control portion 28 which receives from the receiver portion 24 downlink signals DSi from the or each BTS with which the MS 20 is currently in communication. The soft hand-off control portion 28 also applies a power imbalance message PIM to the transmitter portion 26.

one example of the constitution of the soft hand off control portion 28 in the Figure 7 mobile station is shown in Figure 8.

In Figure 8 the soft hand-off control portion 28 comprises a power measuring section 282, a BCH message decoding section 284, a SIR measuring section 286, a BTS selection section 288, a power adjustment calculating section 290, a message section 292, a power imbalance detection section 294, subtractors 296, 298, 300 and 306, and adders 302 and 304.

The power measuring section 282 receives the downlink signals DSi received by the MS 20 from the is different BTSs involved in the soft hand-off operation.

For each such BTS the power measuring section produces receive power measures both for the pilot channel (CPICH) and for the downlink traffic channel (DPCH) allocated to it by the network. In the Figure 8 example it is assumed that two BTSs A and B are involved in the soft hand-off operation and accordingly the power measuring section produces respective receive power measures RAp and RAT for BTS A and respective receive power measures RBp and RBT for BTS B. The BCH message decoding section 284 also receives the downlink signals DSi received by the MS 20 and, when a BCH message is detected in among those signals, the decoding section 284 decodes the message concerned to extract therefrom a pilot transmission power T, or T,p for the BTS concerned.

Furthermore, the SIR measuring section 286 also receives the downlink signals DSi and produces for each traffic channel (DPCH) a measure of the signalto interference ratio (SIR) SIRAT or SIREIT for the BTS concerned.

The subtractor 296 subtracts the pilot transmission power T.. for BTS B from the pilot transmission power TAp for BTS A to produce a power offset value AT0s. This power offset value ATOs is applied to the BTS selection section 288, the power adjustment calculating section 290 and the power imbalance detection section 294. The subtractor 298 subtracts the pilot channel receive power measure RAp for BTS A from the pilot channel transmission power TAp for BTS A to produce a path loss measure PLA for BTS A.

The subtractor 300 performs the same operations for BTS B. The path loss measures PLA and PL, are applied to the adders 302 and 304 respectively and to the BTS selection section 288.

The adder 302 adds together the traffic channel is receive power measure R,, and the path loss measure PLA for BTS A to produce an estimate TAT of the traffic channel downlink transmission power of BTS A. The adder performs the same operations for BTS B to produce an estimate T, of the traffic channel downlink transmission power of that ETS.

The estimate TBT for BTS B is then subtracted from the estimate TAT for BTS A by the subtractor 306 to produce a power imbalance measure AT which is applied to the power imbalance detection section 294. The power imbalance detection section 294 also receives a preset threshold value THR, as well as the intended power offset AT0s. The power imbalance detection section 294 compares the power imbalance measure AT with the threshold value THR (taking into account the intended power offset AT0s) and produces a power imbalance detection signal DET when the absolute value of the difference between AT and ATOs exceeds the threshold value THR.

In the Figure 8 embodiment, the MS 28 employs a space selection diversity technique (SSDT) for the downlink traffic signals received from the active BTSs involved in the soft hand-off operation. This means that in any one timeslot, only the downlink signal from one of the involved BTSs is selected by the MS for receive processing to extract the information content therefrom. The selection of the BTS from which traffic signals are to be received may be carried out according to a number of different criteria but, in this embodiment, the MS selects the BTS whose downlink traffic (or pilot) signals have the highest receive power. Thus, when RAT > RST (or RAp > RBP) BTS A is selected, otherwise BTS B is selected. Alternatively the two path loss estimates PLA and PL, could be compared directly with one another to select the BTS having the lowest path loss, but this comparison will is not necessarily identify the BTS capable of providing the best-quality (e.g. strongest) downlink traffic signals to the MS 28. Incidentally, in the Figure 8 embodiment, instead of comparing the receive powers, the same result could be obtained by subtracting the intended power offset measure ATOs from the path loss estimate PLA for BTS A when carrying out the comparison, as follows:

when PLA ATOs < PLB then select BTS A when PLA ATOs 2t PL, then select BTS B. The BTS selection is carried out by the BTS selection section 288 which outputs a BTS selection signal IDBTS identifying the selected BTS. This selection signal ID1ITS is applied to the receiver portion 24 in Figure 7, where it is used to select the downlink traffic signal which is to be decoded in the next timeslot to extract the information content thereof. The selection signal IDSTS is also applied to the power adjustment calculating section 290.

The power adjustment calculating section 290 also receives the respective SIR measures SIRAT and SIRIT from the SIR measuring section 286. The power adjustment calculating section 290 also receives a target SIR value SIRTARGET which represents, for example, a SIR level at which it is expected that received downlink traffic signals can be decoded with a suitably-low BER when SSDT is used.

The power adjustment calculating section 290 also receives the power imbalance detection signal DET from the power imbalance detection section 294, as well as the estimates T., and T1IT Of the downlink transmission powers of BTS A and B. The power offset AT0s is also applied to the power adjustment calculating section 290.

In the power adjustment calculating section 290, power adjustments are calculated in response to receipt is of the detection signal DET as follows. Firstly, for the selected BTS identified by the selection signal IDITs a SIR error value ASIR is calculated to indicate the difference between the actual SIR of the downlink traffic signals from the selected BTS and the target SIR SIRTARGET. For example, when BTS A is selected:

SIR = SIRTARGET -SIRAT... (8) It is then assumed that, af ter any power adjustments have taken place, the interference and path loss for the selected BTS will remain the same as they were prior to the adjustment. Based on this assumption, an adjusted power TAT(ADJ) is calculated for the selected BTS (BTS A) using:

TAT(ADJ)= TAT +A SIR This adjusted transmission power TAT(Aj) should be sufficient to ensure that the target SIR SIRTMIET will be met.

As for the adjusted power TBT(ADJ) for BTS B, this can be calculated using:

TBT(ADJ):--TAT(ADJ) -A Tos (10) The adjusted powers TAT(ADj) and TBT(ADJ) are supplied by the power adjustment calculating section 290 to the message section 292. The message section 292 formulates a power imbalance message (layer 3 message) to be supplied to the network, which message includes the adjusted powers T.,T(ADj) and TBT(ADJ). The PIM is decoded by the RNC, and BTSs A and B are then informed by the RNC of their adjusted transmission powers.

The PIM can take many different forms. For example, instead of specifying the adjusted power levels TAT(ADJ) and TBT(ADJ) directly, the PIM could specify the change in downlink transmission power required for each involved BTS. This could be advantageous, for example, if the power adjustment is to be brought about progressively over a number of timeslots, rather than in one step. Furthermore, the adjusted powers could be specified as an offset from a nominal or reference power known both to the MS and to the network. This could reduce the amount of information that needs to be included in the PIM.

It would also be possible, in a network in which different TPC commands (TPC bits) can be supplied from the MS to the different BTSs involved in soft hand-off, for the power adjustments calculated by the power adjustment calculating section 290 to be brought about using individual TPC commands for the different involved BTSs.

In the Figure 8 embodiment, the MS 28 uses SSDT to receive downlink traffic signals from the involved BTSs during soft hand-off. However, it would also be possible for the MS to use a maximum ratio combining (MRC) technique to Combine the respective downlink signals of two or more BTSs to achieve satisfactory reception during soft hand-off. In this case, the BTS selection section 288 can be omitted or modified to select a subset of the BTSs involved in the soft hand off. The power adjustment calculating section operates as follows in this case. A combined SIR measure SIRcomB when MRC is used can be calculated using:

SIRCOMB = 10 10910 [10SIRAT110 + 10SIRBTIIO 1... (11) In order for the combined SIR SIRecmB to meet the is target SIR SIRTI1GET it therefore follows that:

10g10 llo(TAT(AD,9-TAT+SIRAT)110 + lo(TBT(AD,9-TBT+SIRBT)110 SIRTARGET .. (12) Prom this, the adjusted transmission power TAT (ADJ) for BTS A required to meet the SIR target can be calculated from:

TAT(ADJ) =SIR TARGET - 10 1091011 O(SIR AT- TAT)110 +io(SIRBT-TBTATos)110 (13) As in equation (10) above, the adjusted transmission power TBT (ADJ) for BTS B is:

TBT(ADJ)= TAT(ADJ) -A Tos... (14) 26- Although the present invention has been described above in relation to a UTRA network, it will be appreciated that it can also be applied to other networks in which it is desirable to avoid power imbalances between the BTSs involved in soft hand-off.

These networks could be, or could be adapted from, other CDMA networks such as a wideband CDMA (W-CDMA) network or an IS95 network. These networks could also be, or be adapted from, other mobile communication networks not using CDMA, for example networks using one or more of the following multipleaccess techniques:

time-division multiple-access techniques: time-division multiple access (TDMA), wave length- divi s ion multiple access (WDMA), f requencydivision multiple access is (FDMA) space-division multiple access (SDMA).

Also, although embodiments of the present invention have been described as having distinct "sections" such as the power adjustment calculating section, those skilled in the art will appreciated that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of the RNC, BTS and/or mobile station in embodiments of the present invention.

-27

Claims

CLAIMS:

1. A soft hand-off method, for use in a cellular mobile communications network, including the steps of:

when a mobile station of the network is involved in soft hand-off with two or more base transceiver stations of the network, detecting an imbalance amongst the respective downlink transmission powers used by the two said or more involved base transceiver stations to transmit information signals to the mobile station; and following detection of such a power imbalance, adjusting the said downlink transmission power used by at least one of the said involved base transceiver stations to transmit its said information signals to the mobile station, so as to reduce the detected is imbalance.

2. A method as claimed in claim 1, wherein the detecting step is carried out by the mobile station.'

3. A method as claimed in claim 2, wherein in the said detecting step the mobile station:

estimates the respective downlink transmission powers of the said involved BTSs; and compares the estimated downlink transmission powers and detects the said power imbalance.when a difference between the estimated downlink transmission powers exceeds a threshold value.

4. A method as claimed in claim 3, wherein, for each said involved base transceiver station, the mobile station:

measures an inf ormat ion- signal received power, at which the said information signals transmitted by the base transceiver station concerned are received by the mobile station; estimates an information-signal path loss experienced by the said information signals in transmission from the base transceiver station concerned to the mobile station; and calculates the said estimated downlink transmission power for the base transceiver station concerned based on the measured inf ormat ion- signal received power and the said estimated information signal path loss.

5. A method as claimed in claim 4, wherein, for each said involved base transceiver station, the mobile station:

measures a pilot-signal received power, at which pilot signals transmitted by the base transceiver station concerned are received by the mobile station; detects a message broadcast by the base transceiver station concerned, indicating a pilot signal transmission power at which the said pilot is signals are transmitted by the base transceiver station concerned; calculates a pilot-signal path loss experienced by the said pilot signals in transmission from the base transceiver station concerned to the mobile station based on a difference between the indicated pilot signal transmission power and the measured pilot-signal received power; and calculates the said estimated inf ormat ion -signal path loss based on the calculated pilot-signal path loss.

6. A method as claimed in claim 5, wherein the said estimated inf ormation- signal path loss is set equal to the calculated pilot-signal path loss.

7. A method as claimed in any one of claims 3 to 6, wherein, when comparing the differences between the estimated downlink transmission powers to detect the said power imbalance, the mobile station takes into account an intended transmission power offset, set by the network, between the respective downlink transmission powers of the involved base transceiver stations.

8. A method as claimed in any one of claims 2 to 7, wherein, upon detecting such a power imbalance, the mobile station transmits a power imbalance message to the network to indicate that such power adjustment is required.

9. A method as claimed in claim 8, wherein, upon receiving such a power imbalance message from the mobile station, the network calculates a power adjustment, if any, required in the downlink transmission power of each involved base transceiver station.

10. A method as claimed in any one of claims 1 to 8, wherein, following detection of such a power imbalance. the mobile station calculates a power is adjustment, if any, required in the said downlink transmission power of each involved base transceiver station, and transmits information relating to the calculated power adjustment (s) to the network.

11. A method as claimed in claim 1, wherein the detecting step is carried out by a radio network controller of the network which controls the said two or more involved base transceiver stations, and, following detection of such a power imbalance, the radio network controller calculates a power adjustment, if any, required in the downlink transmission power of each involved base transceiver station.

12. A method as claimed in claim 9, 10 or 11, wherein the said downlink transmission power is adjusted by the full amount of the calculated power adjustment in one step.

13. A method as claimed in claim 9, 10 or 11, wherein the said downlink transmission power is adjusted by the calculated power adjustment progressively using more than one adjustment step.

14. A method as claimed in any one of claims 9 to 13, wherein the calculated power adjustment for each involved base transceiver station is such as to make the said downlink transmission power of the stations after adjustment equal to the downlink transmission power of that one of the involved base transceiver stations having the highest downlink transmission power when the said power imbalance is detected.

15. A method as claimed in any one of claims 9 to 13, wherein the calculated power adjustment for each involved base transceiver station is such as to make the downlink transmission power of the station after adjustment equal to the downlink transmission power of that one of the involved base transceiver stations having the lowest downlink transmission power when the said power imbalance was detected.

is

16. A method as claimed in any one of claims 9 to 13, wherein the calculated power adjustment for each involved base transceiver station is such as to make the downlink transmission power of the station after adjustment equal to or dependent upon the downlink transmission power when the said power imbalance was detected of that one of the involved base transceiver stations which has the best quality reception for uplink signals from the mobile station.

17. A method as claimed in any preceding claim, wherein the said downlink transmission power for each said involved base transceiver station is set relative to a reference power determined by the network for the base transceiver station concerned.

18. A method as claimed in claim 11, wherein at any one time the mobile station processes the information signals transmitted from just a selected one of the said involved base transceiver stations to extract therefrom the information content thereof, and, when the said power imbalance is detected, the mobile station calculates a power adjustment for that selected base transceiver station so as to maintain a predetermined property of the said information signals received from the selected base transceiver station at a preselected target level.

19. A method as claimed in any claim 11, wherein the mobile station processes the respective information signals transmitted by at least two selected ones of the said involved base transceiver stations to derive therefrom a combined signal from which to extract the information content of the information signals, and, when the said power imbalance is detected, the mobile station calculates a power adjustment for each said selected one of the base transceiver stations so as to maintain a predetermined property of the said combined signal at a preselected target level.

is

20. A method as claimed in claim 18 or 19, wherein the said predetermined property is a signal-to interference ratio.

21. A mobile station, for use in a cellular mobile communications network, including:

22. A mobile station as claimed in claim 21, further including:

power adjustment calculating means operable, following said detection by the power imbalance detection means of such a power imbalance, to calculate a power adjustment, if any, required in the said downlink transmission power of each involved base transceiver station and to transmit information relating to the calculated power adjustment to the said network.

23. A radio network controller, for use in a cellular mobile communications network, including:

power imbalance notification receiving means operable to receive a power imbalance message transmitted by a mobile station of the network when the mobile station detects an imbalance amongst the respective downlink transmission powers used by two or more base transceiver stations, with which the mobile station is involved in soft hand-off, to transmit information signals to the mobile station; and power adjustment means, responsive to receipt by the power imbalance notification receiving means of such a power imbalance message, to calculate a power adjustment, if any, required in the said downlink transmission power of each involved base transceiver station and to cause each involved base transceiver station to adjust its said downlink transmission power in accordance with the calculated power adjustment.

24. A radio network controller, f or use in a cellular mobile communications network, including:

power imbalance detection means operable, when a mobile station of the network is involved in soft hand off with two or more base transceiver stations of the network, to detect an imbalance amongst the respective downlink transmission powers used by the said two or more involved base transceiver stations to transmit information signals to the mobile station; and power adjustment means operable, following detection by the said power imbalance detection means of such a power imbalance, to calculate a power adjustment, if any, required in the said downlink transmission power of each involved base transceiver station and to cause each involved base transceiver station to adjust its said downlink transmission power in accordance with the calculated power adjustment.

25. A cellular mobile communications network, including:

a mobile station; a plurality of base transceiver stations; power imbalance detection means operable, when the said mobile station is involved in soft hand-off with two or more of the said base transceiver stations of the said plurality, to detect an imbalance amongst the is respective downlink transmission powers used by the said two or more involved base transceiver stations to transmit information signals to the said mobile station; and power adjustment means operable, following detection of such a power imbalance by the said power imbalance detection means, to adjust the said downlink transmission power used by at least one of the said involved base transceiver stations to transmit its said information signals to the mobile station so as to reduce the detected imbalance.

26. A soft hand-off method substantially as hereinbefore described with reference to Figures 6 to 8 of the accompanying drawings.

27. A mobile station for use in a cellular mobile communications network substantially as hereinbefore described with reference to Figures 6 to 8 of the accompanying drawings.

28. A radio network controller for use in a cellular mobile communications network substantially as hereinbefore described with reference to Figures 6 to 8 of the accompanying drawings.

29. A cellular mobile communications network substantially as hereinbefore described with reference to Figures 6 to 8 of the accompanying drawings.