SE511167C2 - Downlink power control method during macrodiversity e.g. for spread spectrum multiple access systems - Google Patents

Abstract

The method involves transmitting power level control commands from a remote station to the base stations from which the remote station is receiving signals, to control the power levels of the signals transmitted by the base stations. A determination is made whether the remote station is simultaneously receiving a signal transmitted by a base station and a second signal transmitted by a second base station that include identical message information. A report of a power level of the first signal is transmitted to a controller from the base station. A second report of a power level of the second signal is transmitted to the controller from the second base station. A comparison is made in the controller the first report and the one second report. A command is transmitted from the controller to the first base station to adjust the power level of the first signal. The method transmits from the controller to the second base station a second command adjusting the power level of the second signal, the power level control commands are transmitted less frequently from the remote station to the base stations when the remote station receives signals from more than one base station than when the remote station receives signals from only one base station. The same power level control commands are transmitted several times from the remote station to the base stations when the remote station receives signals from more than one base station.

Description

511 167 2 Prentice Hall (1997). Usually, the same baseband signal is transmitted with appropriate spread from several base stations with overlapping coverage. Thus, the mobile phone can receive and utilize signals from several base stations simultaneously.

CDMA allows signals to overlap in both time and frequency. Thus, CDMA signals share the same frequency spectrum. In the frequency or time domain, the multiple access signals appear to overlap.

There are a number of advantages to CDMA communication techniques.

The capacity constraints of a CDMA-based cellular system are high. This is a consequence of the characteristics of a wideband CDMA system, such as improved interference diversity, voice control, and reuse of the same spectrum in interference diversity.

In principle, the information data stream to be transmitted in a CDMA system is superimposed on a much higher rate data stream as a signature sequence. Typically, the signature sequence data is binary, and constitutes a bit stream. One way to generate this signature sequence is with a PN (pseudo-noise) approach which appears random, but can be replicated by an authorized receiver. The information data stream and the signature sequence stream with the high bit rate are combined by multiplying the two bit streams, assuming that the binary values of the two bit streams are represented by +1 or -1. This combination of the higher rate bit signal with the lower bit rate data stream is called spreading the information data stream signal. Each information data stream or channel is assigned a unique spreading code. The ratio of the signature sequence bit rate to the information bit rate is called the spreading ratio.

Generally in the uplink, a CDMA base station attempts to maintain the same received power level in communications with each mobile phone it is currently handling.

To this end, a base station measures the received signal from each of the mobile phones to determine a parameter known as the Signal to Interference and Noise Ratio (SINR). For each mobile phone, the SINR is compared to a target SINR value. If the SINR measured for a particular mobile phone is less than the target SINR value, the base station commands the mobile phone to increase its power so that a stronger signal can be received at the base station. On the other hand, if the SINR determined for a particular terminal is greater than the target SINR value, the base station requests that the mobile phone reduce its power. In the downlink, on the other hand, the mobile phone listens to the base station and compares its perceived SINR to a target value and commands the base station to increase or decrease its output power depending on the perceived quality.

When it turns out that the target value for SINR also needs to be continuously updated for various reasons, such as increasing and decreasing the number of mobile phones handled by the base station. The update of the target value for SINR is usually part of another control loop.

An example of a control technique as described above is disclosed in U.S. Patent 623,484 to Muszynski. To transmit information (including a signal-to-interference and noise ratio SINR adjustment point) between the base station and the RNC node, Muszynski employs a common channel signaling technique known as Signaling System No. 7.

In conventional channel signaling, the signaling for a number of user channels/connections is carried in some number of signaling channels/connections. Furthermore, there is no static relationship between signaling channels/connections and user data.

In CDMA systems, the SINR target values must be updated frequently. When common channel signaling is used as in Muszynski to update a SINR target value, the load on the link or line connecting the RNC node to the base station increases greatly due to the updates. For example, sending a common channel signaling message of ten to fifteen octets at a rate of 10 to 511,167 4 times per second generates an additional signaling load of between one thousand and fifteen hundred bits per second per mobile station.

When mobile phone communications are established with a new base station or a new sector of an existing base station, the mobile phone has good communications with the new cell or sector before the old base station ceases to handle the call.

Such soft hand-off is in effect a do-before-break-unconnection procedure. The mobile telephone determines the best new base station, or sector, to which communications should be transferred from an old base station, or sector. Although it is preferred that the mobile telephone initiates the hand-off request and determines the new base station, hand-off decisions may be made as in conventional cellular telephone systems in which the base station determines when hand-off may be appropriate and, via an RNC, requests neighboring cells, or sectors, to search for the mobile telephone signal.

The base station receiving the strongest signal as determined by the RNC then accepts the hand-off.

In the CDMA cellular communication system, each base station typically transmits a control carrier signal in each of its sectors. This control signal is used by the mobile phones to achieve initial system synchronization and to provide stable time, frequency, and phase tracking of the base station's transmitted signals during a so-called air interface chip synchronization phase. The RNC maintains its synchronization with the Public Switched Telephone Network (PSTN).

In so-called macrodiversity, a mobile phone is handled by more than one base station. Macrodiversity must be utilized during a soft hand-off from a first base station to a second base station or to a new sector at the first base station.

Future CDMA systems will likely have fast power control in the downlink. Furthermore, most of these systems, like current systems, support soft hand-off. The closed-loop power control during soft hand-off can be handled in at least two different ways. Either the mobile phone sends identical power measurement reports to all base stations involved in the macrodiversity connection, or the mobile phone sends different power network reports depending on the specific power level value that has been measured from each individual base station. The first option can lead to the base stations transmitting at non-optimal power levels.

One way to prevent this is to only transmit from a single base station if the measured power levels from the other base stations are all below a specific threshold. Another solution is to use open-loop power control instead of closed-loop power control during the macrodiversity connection. Both of the latter options lead to a fairly low amount of control signaling between the RNC and the base stations. However, the SINR target values must be updated sufficiently frequently.

What is needed and an object of the present invention is therefore an efficient way to transmit control data from an RNC to its associated base stations during macrodiversity connections in a CDMA mobile telecommunications system.

SUMMARY OF THE INVENTION The present invention provides efficient closed-loop power control of the base stations involved in a macrodiversity handover to a particular mobile phone.

The proposed solution frees up network resources during the execution of a soft hand-off, i.e. during macrodiversity, by reducing the amount of control signaling between an RNC node and the base stations involved. The beneficial frequency diversity during soft hand-off reduces the detrimental effect on signal quality due to fast fading. Therefore, the proposed reduction of control signaling is possible.

The solution of having the RNC nodes or some other network entity adjusting the output power levels of the base stations is an effective method to maintain the output power levels of the traffic channels at a proper level under macrodiversity, even when errors in the reception of the downlink power control commands occur. However, the problem is that this requires a lot of extra signaling between the base stations and the RNC node. According to previous solutions, it was believed that the base stations need to send a power control message every 10 ms to keep the operation around 1-2 dB, in a system with a high proportion of uncoded or lightly coded power control commands, and a step size of 1 dB.

In accordance with the present invention, a first radio channel connection, e.g. of the code division multiple access type, is assumed to be active between a first base station and a mobile phone. A first control channel connection is also established between a radio network controller node and the first base station. Furthermore, power update commands are transmitted from the mobile phone to the first base station at a first rate, via the first radio channel connection. While the first radio channel connection is still active, a second radio channel connection is established between a second base station and the mobile phone. A second control channel connection is also established between the radio network controller node and the second base station. The mobile phone then sends the same second power control commands to the first and second base stations, or sends a third set of power control commands to the first base station and a fourth set of power control commands to the second base station. The bit rates of the different power control sets may be different.

The method according to the invention is characterized by the features of the characterizing part of claim 1.

According to a preferred embodiment of the invention, the second rate is achieved by sending at least one power control report, which is identical to a power control command previously sent. The redundancy thus introduced increases the probability that the power level reports are received correctly. As a result, power level reports will not be retransmitted as frequently and the resulting rate of power level reports is reduced.

According to an advantageous embodiment of the method according to the invention, the second rate of power level reports is achieved by subjecting the reports to a higher coding rate than when the mobile telephone communicates via non-macrodiversity connections, i.e. via a single base station.

According to another advantageous embodiment of the method according to the invention, the second rate of power level reports is achieved by transmitting optimized power control commands to each of the first and the second base stations.

The RNC node calculates which output power level is optimal for each base station with respect to the mobile telephone and sends the corresponding power control commands to the base stations involved in the macrodiversity transmission.

Further uses of the present invention will become apparent from the detailed description which follows. It should be understood, however, that the detailed description and specific examples, which at the same time indicate preferred embodiments of the invention, are given for illustrative purposes only, since various changes and modifications within the same spirit and scope of the invention will be apparent to those skilled in the art from this detailed description.

The proposed solution clearly reduces the required amount of signaling between an RNC node and the base stations involved during soft hand-off.

The invention also enables closed loop power control in the mobile telecommunications system during an entire connection session, without requiring excessive signaling between RNC nodes and base stations. 511 167 8 Finally, the invention provides a flexible telecommunications system, as it provides a possibility for an RNC node to send different commands to different base stations.

Finally, another possible method for reducing signaling is to reduce the step size of changes in the output power level.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows a per se known arrangement of nodes in a mobile telecommunications system in which the proposed method is carried out; Fig. 2 shows a flow chart of the first embodiment of the method according to the invention; Fig. 3 shows a flow chart of the second embodiment of the method according to the invention; Fig. 4 illustrates a method for controlling the downlink transmission power; Fig. 5 illustrates a soft hand-off involving a remote station and two base stations; Fig. 6 illustrates a method for overcoming a problem that a remote station's power control command is received with a signal level that is too low during soft hand-off mode; Fig. 7 illustrates another method for overcoming the problem that a remote station's power control commands are received with a signal level that is too low during soft hand-off mode; 9 511 167 Fig. 8A-8C illustrate the operation of a communication system according to the applicant's invention.

The invention will now be described in more detail with reference to preferred exemplary embodiments thereof and also with reference to the accompanying drawings.

DESCRIPTION OF PREFERRED EMBODIMENTS The overall problem that causes a power drift is that a mobile phone only sends one power control command which will be received by different base stations.

To reduce the required signaling between the base stations and the RNC node, the required power control command rate could be reduced during soft hand-off, the command rate could typically be reduced by a factor of 4.

The idea behind this setting is that the drift will be less if fewer messages are sent, because fewer messages will be interpreted differently. This is a very feasible solution, since the purpose of the fast closed loop power control is to follow the fast fading. In soft hand-off, the fast fading will not be as severe as it is under normal conditions due to good spatial diversity. Another possible way to exploit the good spatial diversity is to reduce the step size. The step size is the relative measure by which the output power is adjusted each time the output power is adjusted by the closed loop power control. For example, if the ordinary step size is 1 dB, it can be reduced to the typical value of 0.25 dB.

The idea of reducing the step size is to make the drift slower, since each time a power control message is received differently it will only result in a small drift. This means that the RNC node does not need to adjust the base stations' output power levels as often as it would have for larger step sizes, if an equal drift was allowed before the adjustment was required. The benefit of this approach is that it requires little or no signaling over the air interface. (The required power control command rate is the rate of power control commands needed to achieve the target SINR.) To further reduce the required signaling between base stations and the RNC node, one could choose to send the same power control command several times. Thus, the probability of receiving the power commands correctly increases, since the probability of receiving the majority of the received bits correctly increases as the number of bits considered increases. For example, if a power control command consists of 8 bits, it is less likely that 4 or more bits will be received incorrectly compared to a 2-bit message that has one or two incorrectly received bits. Another alternative is to use more coding of the required power control commands. These two approaches can be seen as using the bits that are not needed to change the transmitted power levels to increase the probability that the required power control commands will be received correctly under macrodiversity.

The lower power control rate required also makes it possible to send different power control messages to different base stations. The benefits of this approach are that we can reduce the unwanted power drift. The reduction in power drift is possible because the mobile phone can counteract the power drift by selecting power control commands to different base stations in such a way that the power drift is reduced.

To do this, however, the mobile phone needs information about the output power levels at different base stations. This is achieved by having the base stations send data regarding their output power levels. The mobile phone is unlikely to counteract power drift if it sends the same power command to all base stations to which it is currently connected, and may cause the output power levels from different base stations to be amplified due to the addition of no or little additional signaling.

By introducing additional signaling, other output schemes can also be supported.

The ideas of increasing the accuracy of the power control commands and the idea of having the mobile phone send different commands to different base stations can of course be combined. The process of having modified power control for one reason or another is possible in all systems. Having it modified due to macrodiversity is possible in all systems that use macrodiversity. Furthermore, it is particularly important to have this feature in systems that utilize downlink power control and where the base station output power levels are updated frequently.

In Fig. 1 a per se known arrangement of nodes in a mobile telecommunications system is depicted. An RNC node 100 is connected to one or more other RNC nodes or intermediate measurement lines to other networks. The RNC node 100 is also connected to two base stations 110; 120. The first base station is in connection with a mobile telephone 130, via a first radio channel connection which utilizes, for example, code division multiple access. During macrodiversity, the mobile telephone is further in connection with the second base station 120, via a second radio channel connection, which also utilizes code division multiple access. In addition to payload information, the mobile telephone 130 communicates control data with the base stations 110; 120. Measured power level reports PBSI; PBS2 about the received radio energy from the base stations 110; 120 is an example of such control data sent from the mobile phone 130 to the base stations 110; 120. The downlink power update commands from the mobile phone may be the same for both base stations 110 and 120 or the mobile phone may send different downlink power control commands to base station 110 compared to base station 120. If the power control commands are dedicated to both base stations, signaling is required between the RNC and the base stations to avoid downlink output power drift.

Fig. 2 shows a flow chart of the first embodiment of the method according to the invention. A first step 200 examines whether macrodiversity transmission should be used. If so, the procedure continues with a step 220. Otherwise, power control commands are sent from the mobile telephone to the active base station at a normal rate, after which the process returns to step 200. The power control commands in step 220 are sent from the mobile telephone to the base stations involved in the macrodiversity transmission at a reduced rate. This rate reduction is exemplified in an embodiment of the invention with further reference to Fig. 3. In a step 230, it is checked whether a time counting parameter t is equal to a rate determination value T. (If not, the procedure returns to step 200. Otherwise, the time counting parameter t indicates that it is time to update the downlink output power at the base stations involved in macrodiversity with a particular mobile phone. In a subsequent step 240, measured power level reports from the base stations are forwarded to the RNC node and SINR target values are sent from the RNC node to the base stations. In a subsequent step 250, the time counting parameter t is set to zero, after which the procedure returns to step 200.

Fig. 3 shows a flow chart of the second embodiment of the method according to the invention. A first step 300 examines whether macrodiversity transmission should be used. If so, the procedure continues with at least one of steps 321, 322 or 323, which can be performed in parallel. Otherwise, power control commands are sent to the active base station from the mobile telephone at a normal rate, after which the procedure returns to step 300.

In step 321, power control commands are sent from the mobile phone to the base stations involved in the macrodiversity transmission at a reduced rate by repeating at least some of the power control commands. A plurality of consecutive and identical power control commands increases the probability that the commands are received correctly. This will reduce the required signaling between the base stations and the RNC nodes.

In step 322, power control commands are sent from the mobile phone to the base stations involved in the macrodiversity transmission at a reduced rate by encoding more redundancy into the commands. The higher degree of encoding increases the probability that the power control commands are received correctly. This will reduce the required signaling between the base stations and the RNC nodes. 13,511,167 In step 323, power control commands are sent from the mobile phone to the base stations involved in the macrodiversity transmission at a reduced rate by transmitting separate power control commands to each base station involved in the macrodiversity transmission. Thus, the signaling between the base stations and the RNC nodes is reduced, as well as adding flexibility to the system.

A fourth alternative, not shown in Fig. 3, for reducing the power control signaling between the RNC node and the base stations is to reduce the step size for changes in the output power level.

In a step 330, it is checked whether a time counting parameter t is equal to a rate determining value T. (If not, the procedure returns to step 300. Otherwise, the time counting parameter t indicates that it is time to adjust the downlink output power level at the base stations involved in macrodiversity with a particular mobile phone. In a subsequent step 340, measured power level reports from the base stations are forwarded to the RNC node, and SINR target values are sent from the RNC node to the base stations. In a subsequent step 350, the time counting parameter t is set to zero, after which the procedure returns to step 300.

The following section will describe possible ways in which the signaling should be implemented in some cases. Of course, these are not the only embodiments. These examples only refer to the cases when the mobile phone enters macrodiversity with a new base station and when the mobile phone is initially not involved in a macrodiversity handover. Other scenarios can easily be imagined, e.g. cases when the mobile phone is connected to more than two base stations. Furthermore, cases are referred to when the mobile phone leaves macrodiversity with one or more base stations and simultaneously enters macrodiversity with one or more base stations. 511 167 14 Yet another scenario is when the base stations in question are connected to different RNC nodes.

The first scenario considers how the signaling would look if the decision on which power control scheme to use was made in the mobile phone. Initially, when the new macrodiversity base station is connected and the closed loop power control is up and running, the mobile phone will indicate that it requires a modification of the rate of power control commands. This is done by sending a message to the base stations. The message contains information about the mobile station as well as data on how the rate of power control commands should be changed. If any of the base stations, which are the intended recipients of the message, receives the message, that base station will respond with a positive acknowledgment signal (ACK).

To avoid that only one base station receives the message, some signaling is required in the fixed network, i.e. to verify that all involved base stations are aware of and prepared for the proposed modification as well as to make the RNC node aware and to allow it to prepare. As soon as the mobile phone receives at least an ACK message from the base stations, it will send a message indicating that it is starting to use the new power control scheme. It then prepares for the new transmission scheme and starts following the new scheme. When a base station receives the message indicating that the mobile phone is starting to use the new scheme, it tells the other involved base stations as well as the RNC node about this, and sends an ACK message to the mobile phone. At the same time, the base stations start following the new power control scheme. To utilize coded power control commands, including repetition, the base stations must store the received power control commands before adjusting their output powers. If the mobile phone sends different messages to different base stations, this can be achieved either by putting an address in each power control command message, or better yet by letting the base stations know which power control commands they should listen for. This could normally be achieved by utilizing a pattern for the power control commands, i.e. a base station knows that the power control commands in every other slot are dedicated to it. However, this is assuming that the base stations do not notice that the mobile phone changes the power control command rate or that the ACK message has disappeared. Then the mobile phone must retransmit the lost message. It is of course possible to change the power control scheme also during macrodiversity. The mobile station should return to the usual power control scheme just before the macrodiversity ends. Since both the network and the mobile phone know when this happens, little or no additional signaling is required.

If it is determined how the power control is to be changed in the network, the signaling could look like this. For example, the decision could be made in the RNC node. As soon as a mobile phone enters macrodiversity with a new base station and the closed loop is connected and in operation, the RNC node will inform the affected base stations about the needed power control scheme, and this message is also sent to the mobile phone. Both the base stations and the mobile phone prepare for the new schemes, and the mobile phone sends a message indicating that it is starting to use the new power control scheme. The base stations will respond with an ACK message. Since the base stations communicate with each other by using the fixed network, the mobile phone only needs to receive an ACK message. If no ACK message is received within a certain time, the mobile phone sends a new start message. Furthermore, if no base station has received a message from the mobile phone indicating that the mobile phone is starting to use the new schedule within a certain time, the system contacts the mobile phone again.

For the case where the step size is changed, the mobile phone does not need to be involved in any signaling, unless the mobile phone decides which step size to use, since the network knows when macrodiversity starts and the RNC node can simply order the involved base stations to change their step sizes. The normal power control should be maintained when the mobile phone leaves macrodiversity. This could also be achieved without adding much signaling over the air interface.

There will also be signaling between the base stations and RNC nodes when adjusting the output power of the base stations.

While this description is within the scope of cellular communication systems involving portable or mobile radiotelephones, it will be appreciated by those skilled in the art that the applicant's invention can be used in other communication applications. Furthermore, while the invention can be utilized in CDMA communication systems, it can also be used in other types of communication systems.

A fast uplink power control method involving the transmission of power control commands consisting of single uncoded bits, transmitted every 1.25 ms, is specified by the IS-95-A standard. This method can be understood by referring to Fig. 7, which is described in more detail below. A remote station RT would measure the signal-to-interference ratio of the downlink signal of a base station BS1 with which the station RT is communicating, and the remote station RT would send a report of the measured signal-to-interference ratio or an uncoded power control command to the base station. Based on such a report or command, the power level (and thereby the signal-to-interference ratio) of the downlink signal would be appropriately controlled by the base station or by another part of the network, such as a radio network controller RNC.

This method can be used for downlink transmit power control and the European publication EP 0 680 160 cited above describes the utility of this method for downlink control during soft hand-off. Although such a simple downlink power control method utilizing uncoded commands can be utilized when the communication system is not in a soft hand-off mode, this simple fast method does not perform well during soft hand-off, due to the possibility of reception errors and power level drift in the various base stations participating in the soft hand-off procedure.

Independent errors in downlink power control commands sent by a base station and received by different downlink transmitters can be expected to cause drift in the transmission power levels of the commanded downlink transmitters relative to each other. When the control power of downlink signals from two or more transmitters received at a remote terminal in soft hand-off mode is roughly equal, the degree of drift can be expected to be moderate. However, when the received downlink signal strengths are different, the transmission power levels of the downlink transmitters can rapidly drift apart, causing large losses in system capacity. The disadvantages of using the fast downlink power control method are overcome by the applicant's invention.

One way to solve the problem of downlink transmission power control according to the applicant's invention is a method that makes frequent adjustments to the downlink power levels transmitted from the base stations, which base stations are involved in soft hand-off. These base stations send the latest values of their downlink transmission power levels to a radio network controller, which compares these values and sends the respective adjustment commands PlDL(tn) and P2DL(tr1) back to the base stations.

This scenario is illustrated in Fig. 7, in which a base station BS1 sends a message PlDL(tn) indicating its downlink transmission power level Pl(tn) at time tn to a radio network controller RNC. In a cellular communication system, the controller RNC could be a base station controller or a mobile switching center. In addition, a base station BS2 sends a message P2DL(tn) indicating its downlink transmission power level P2(tn) at time tn to the controller RNC. In the soft hand-off mode shown in Fig. 7, a remote station RT receives signals from the base station BS1 and 511 167 1* receives signals with substantially the same message information from the base station BS2.

The controller RNC sends the respective adjustment commands PlDL(tn) and P2DL(m) to the base stations based on the messages PlDL(tn), P2DL(tn). In other embodiments, as described in more detail below, the controller may also determine the adjustment commands based on a report from the remote terminal RT, which identifies the base station from which the remote terminal receives the highest power level or the best signal interference ratio. The adjustment commands cause the base stations to adjust their downlink transmission power levels so that a desired relationship between P1(tn+1) and P2(tn+1) is maintained (e.g., both levels may be kept substantially equal or, as described in more detail below, one of the power levels may be controlled to be substantially zero). It will be appreciated that the transmission efficiency levels P1(tn), P2(tn) are derived from previous commands P1DL(tn 1) and P2DL(tn 1), as shown in Fig. 7.

The controller RNC may determine the adjustment commands in several different ways. For example, the power levels Pl(tn) and P2(tn) may be adjusted to their arithmetic mean values (Pl(tn) + P2(tr1))/2. Alternatively, the power levels may be adjusted to their geometric mean values. Alternatively, which has the advantage of fast response, the power levels may be adjusted so that the higher power level is lowered, e.g. to a level substantially equal to the lower power level. Alternatively, the power levels may be adjusted so that the lower power level is raised, e.g. to a level substantially equal to the higher power level. Alternatively, the power levels may be adjusted so that the lower power level is controlled to be substantially zero. The power level adjustments would normally be made once per frame, to keep the drift between the downlink transmit power levels of the downlink transmitters involved in a soft hand-off at a low level, e.g. less than 1 dB on average. 19 511 167 To minimize the delays between power level measurement and adjustment, in-band signaling could be utilized both for sending the messages P1DL(tn), P2DL(tn) from the base stations BS1, BS2 to the control unit RNC and for sending the adjustment commands PlDL(tn), P2DL(tn) from the control unit RNC to the base stations. By "in-band signaling" is meant that information is sent together with user data and not as separate messages on a separate control channel.

How the independent errors in a remote station's power control commands arise is illustrated in more detail in Fig. 5, which shows a remote terminal RT communicating with two base stations BS1, BS2, which are controlled by a radio network controller RNC. In the soft hand-off mode shown in Fig. 5, as well as in Fig. 7, the remote station RT receives signals from the base station BS1 at a power level P1 and receives signals with substantially the same message information from the base station BS2 at a power level P2". Power control commands PC' transmitted by the remote station RT are received by the base station BS1 as commands PC and are received by the base station BS2 as commands PC. Due to errors that can arise from many sources, a command PC need not be the same as either or both of commands PC°, PC", and the command PC" need not be the same as the command PC" in The order of magnitude of the difference between one downlink power level and another downlink power level affects the capacity of the system, since a base station transmission appears as interference to the other base station. In addition, the rate at which each downlink power level operates is important, since faster rates usually require that the control commands be transmitted more frequently. This increases the load of messages that must be carried by the links between the base stations BS1, BS2 and the controller RNC.

In a communication system operating in a manner as described in section 6.6.6.2.7.2 of the IS-95-A standard, the remote terminal's transmit power on the uplink to the base stations BS1, BS2 will be controlled substantially as commanded by the "nearer" base station BS1. As a result, the remote station's uplink transmit power level will be too low from the point of view of the "more distant" base station BS2. It will be appreciated that a "nearer" transmitter is not necessarily geographically closer than the receiver but is closer with respect to the received downlink power level, i.e. its downlink signal is received at a higher power level for a known transmit power level. Similarly, a "more distant" transmitter is further away with respect to the received downlink power level, i.e. its downlink signal is received at a lower power level for the known transmit power level.

Similarly, the remote station's power control commands PC" will be received with too low a signal level at the base station BS2, causing more power control command errors at the base station BS2. These additional errors are differences from the power control commands received at the base station BS1, i.e. PC i PC.

The difference between P1 and P2 can be large, for example as large as 8 dB in some communication systems, and such a large difference leads to a similarity that PC :ß PC around %.

Downlink transmit power control in accordance with another aspect of the applicant's invention avoids these problems by utilizing one of the quality-based downlink power control methods described above, such as the FER-based methods specified in the IS-95-A or J-STD-ÛOS standards or the BER-based method utilized in the CODIT system, when the remote terminal is in soft hand-off mode. As described in more detail below, the remote terminal measures the quality of the received downlink signal, either by determining the FER or BER, and then the remote terminal RT reports its quality measurement to the network by sending an appropriate quality message to the base stations involved in the hand-off.

Each base station forwards the remote terminal quality message (a Layer 3 message) to the controller RNC, which causes the base stations to adjust their downlink outputs appropriately. 21 511 167 It is important to understand that the remote terminal quality messages are more than just information about the forward link signal strength, as described in WO 95/12297, cited above. The applicant's quality messages are generated based on downlink signals that have been decoded, or demodulated, rather than on simple determinations of signal levels or signal interference ratios. The extra effort required to generate the quality messages is rewarded by the advantages of the applicant's invention.

The Applicant's method provides near-optimal CDMA communication system capacity at a very low cost in signaling between the network controller RNC and the base stations involved in the soft hand-off procedure (e.g., stations BS1, BS2 in Fig. 5). It will be noted that Layer 3 quality messages received by the network controller from a remote terminal through two or more base stations will generally be identical, and the controller will thereby not need to negotiate between different quality messages from the same mobile phone.

The gain achieved by utilizing the downlink power control method illustrated by Fig. 7 compared to utilizing one of the quality-based power control methods (Fig. 5) may be small during soft hand-off. Since the remote terminal in soft hand-off mode is similarly far from the base stations, it is likely that the remote terminal has good frequency diversity from each base station due to the multipath condition, and power fluctuations are likely to be slow compared to the non-soft hand-off mode. Therefore, the remote terminal can utilize signal energy from all base stations to which it is associated during soft hand-off, reducing the effects of fading, and the remote terminal can utilize "slow" signal quality-based power control methods, since "fast" methods are not needed. It will also be appreciated that the desirable quality reports can be achieved at a low cost in air interface and in transport capacity. 511 167 22 In another aspect of the applicant's invention, the problem of a remote station's power control commands PC being received at too low a signal level at a "more distant" base station BS2 during soft hand-off mode is overcome in several ways. A base station that is in "soft hand-off mode" is distinguished from a base station that is not in "soft hand-off mode" by the former's allocation of logical and physical resources, e.g., encoders/decoders, to different communication links.

For example, the remote station RT may either periodically send reports of its received downlink signal strength to the network or send a report whenever a signal from a new downlink transmitter is received with a strength higher than any other currently received downlink signal strength. It will be appreciated that the IS-95-A standard does not require the remote station to send reports of its received downlink signal strength. In such a communication system, the network controller RNC would respond to the base station signal strength reports by causing only the base station whose downlink signal is received with the highest strength at the base terminal RT to transmit.

This scenario is depicted in Fig. 6, which shows that when the remote terminal's received downlink signal level Pl from base station BS1 is greater than the received downlink power level P2 from base station BS2, the downlink signal level Pl transmitted by base station BS1 is greater than zero and the downlink power level P2 transmitted by base station BS2 is zero. (In practice, the power level P2 only needs to be significantly less than the power level P1). The weaker base station BS2 should operate differently, typically in soft hand-off mode. The network informs the remote terminal RT whenever the system chooses to change the transmitting base station to accommodate base station differences, such as different pseudo-noise signals, although different base stations generally look the same from the remote station's perspective in soft hand-off mode.

The problem of drift between downlink transmission power levels, i.e., that a remote station's power control commands PC are received with too low a signal level at a "more distant" base station BS2 during soft hand-off mode, can be overcome in another way in a communication system in which the uplink signal interference ratio is measured for each frame in the base stations involved in a soft hand-off and the signal interference ratio measurements are forwarded to the communication controller. The base stations involved in the soft hand-off procedure are processed and the communication controller causes the signal interference ratio values from only the base station whose downlink power level received at the remote terminal is the highest to be transmitted. It will be appreciated that signals received from different transmitters, e.g. different base stations, or received from a transmitter, e.g. a mobile telephone, at different receivers, e.g. different base stations, are advantageously processed by spreading each signal, combining the spread signals, e.g. by using a demultiplexer, flanking receiver or other equivalent equipment, and then decoding the combined signal.

This scenario is depicted in Fig. 7, which shows that the downlink power level Pl(tn+1) transmitted by the base station BSI at time tn+1 is greater than zero and the downlink power level P2(tn+1) transmitted by the base station BS2 at time tn+1 is zero. This relationship arises as a result of the respective link transmit power control command sets SetP1DL, SetP2DL transmitted by the radio network controller RNC to the base stations BS1, BS2. The network controller RNC generates the power set commands based on the base stations' respective determinations and reports of the uplink signal interference ratio values SIR1(tn), SIR2(tn) at the previous time In.

Thus, at times when the uplink power received at one base station, e.g. station BS1, is higher than the uplink power received at another base station, e.g. station BS2, the controller RNC sends messages to the base station BS1 causing the station BS1 to start transmitting to the remote terminal RT, to the base station BS2 causing the station BS2 to stop transmitting to the remote terminal RT, and to the remote terminal RT causing the terminal RT to only listen to the base station BS1 and not the base station BS2, starting from a particular frame number.

It is currently believed that the method depicted in Fig. 7 works best when the path losses in the uplink and downlink are highly correlated. This is likely to be the case for a communication system such as that described in the above-cited European Patent Publication No. 0 680 160 which has an uplink/downlink bandwidth of MHz. For such a wide channel bandwidth, multipath propagation to and from the remote station is likely, which reduces the problems with respect to Rayleigh fading.

Multipath propagation is even more likely in soft hand-off mode, in which the remote terminal is likely to be located on the boundary between cells or between cell sectors. Thus, the correlation between the uplink and downlink channel characteristics will generally be strong. For cases where this is not the case, the performance of the method depicted in Fig. 7 will depend to some extent on the movement of the remote terminal and the length of the filter in the network controller.

By using either of the methods depicted in Fig. 6 and Fig. 7, the remote station receives downlink power only from the base station that is received strongest. Thus, the problem of capacity losses due to downlink transmission power drift is eliminated, without significantly increasing the message load over the transport interface between the base station and the network controller. Another advantage of these methods is that the remote terminal remains synchronized with both base stations while the system is in soft hand-off mode; thus, it is unnecessary to increase the time and effort required to resynchronize the remote terminal, which would be necessary if the system learned the soft hand-off mode. A particular advantage of the method depicted in Fig. 7 is that it is independent of how messages in the air interface (the uplink and downlink between the remote terminal and the base stations) are specified; this is not the case with the method depicted in Fig. 6, which involves reports from the remote terminal of its received downlink power level. 511 167 Figs. 8A, 8B and 8C further illustrate the operation of a communication system in accordance with the applicant's invention. (The method illustrated by these figures is also illustrated in Fig. 7). In block 501, base station BS1 receives data and transmits power adjustment commands from a mobile phone RT. Base station BS1 identifies and decodes an adjustment command received at a first time, which may conveniently have the value either +1 or 1, and similarly adjusts its downlink transmit power level at the next time. For example, as indicated by block 503 in Fig. 8A, if the mobile phone sends adjustment commands PC = 1, the base station transmit power level P1(tn+1) = +1 dB with respect to its previous transmit power level Pl(tn). Similarly, if the mobile phone sends adjustment command PC = 1, then the base station transmit power level P1(tn+1) = 1 dB with respect to its previous transmit power level Pl(tn). In block 505, the message P1DL(tn) indicating the downlink transmit power level Pl(tn) at time tn is sent by the base station BS1 to the controller RNC. The base station preferably does not send such messages as often as the mobile phone sends power adjustment commands to reduce the signaling load on the base station-controller connection. For example, the base station may count the number of received adjustment commands n and send a message every Nth adjustment command, where N is a factor a times the number n. In fact a may be an integer in the range from about 16 to about 1600 for some communication systems. In block 507, the base station BS1 sets its downlink transmit power level to the commanded level.

As indicated by box 509 in Fig. 8B, the controller receives the messages P1DL(tn) and possibly messages from other base stations involved in a soft hand-off, e.g. P2DL(tn). The controller, which may include specialized circuitry or a programmable processor for the purpose, then determines new corresponding adjustment commands P1DL(tn+1) and P2DL(tn+1), which as indicated by box 513 in Fig. 8B may be based on differences (box 511) between the downlink transmit power levels P1(tn), P2(tn) of base stations BS1 and BS2. In the example illustrated, P1DL(tn) > P2DL(tn), and thus the new adjustment command 511 167 26 P1DL(tn+1) corresponds to the previous power level of the first base station, less than half the difference between the power levels, and the new adjustment command P2DL(tn+1) corresponds to the previous power level of the second base station plus half of the same difference. If P2DL(tn) > P1DL(tn), essentially the same operations would be performed but the identities of the base stations would be swapped. These commands then communicate in some appropriate manner with the corresponding base stations as indicated by box 515.

The operation of this type in communication systems is further illustrated by Fig. 5C, in which an adjustment command PIDL from the controller RNC is received by the base station BS1 as indicated by box 517. In response, the base station adjusts its transmit power level at a time tn + tm as indicated in box 519, where the interval tm is a suitable time interval, such as an integer number of communication lines. It will be appreciated that similar operations would be performed in the second base station (station BS2). Furthermore, if the power level of the weaker base station were to be commanded to be substantially zero by the operations indicated by boxes 511 - 519, then the weaker base station could advantageously be maintained in soft hand-off mode in all other respects as described above.

The invention having thus been described, it will be apparent that the same may be varied in various ways. Such variations are not to be regarded as departing from the spirit and scope of the invention, and all such modifications as would be obvious to those skilled in the art are intended to be included within the scope of the following claims.

Claims (1)

10 15 20 25 30 27 511 167 Patent claims A method for controlling power levels of signals transmitted by base stations in a communication system having base stations and remote stations, comprising the steps of: determining in a remote station whether the remote station simultaneously receives a first signal transmitted by a first base station and at least a second signal transmitted by at least a second base station which includes substantially identical message information; transmitting to a control unit from the first base station a first report of a power level for the first signal; transmitting to the control unit from the second base station at least a second report on a power level for the second signal; comparison in the control unit of the first report and the second report; transmitting from the control unit to the first base station a first command for controlling the power level of the first transmission from the control unit to the second base station at least one second signal; command for controlling the power level of the second signal; and where the power control commands are transmitted less frequently than when the tar station substantially utilizes message information only from the first base station. . A method for controlling power levels of signals transmitted by base stations in a communication system having base stations and remote stations, comprising the steps of: determining in a tar station whether the remote station simultaneously receives a first signal transmitted by a first base station and at least a second signal transmitted by at least one other base stations which include substantially identical message information; transmitting to a control unit from the first base station a first report of a power level for the first signal; Transmitting to the control unit from the second base station at least a second report on a power level of the second signal; accounting in the control unit of the first report and the second report; transmitting from the control unit to the first base station a first command for controlling the power level of the first signal; transmitting from the control unit to the second base station at least a second command for controlling the power level of the second signal; and where the step size of the power levels is lower than when the remote station substantially utilizes message information only from the first base station.