GB2402580A

GB2402580A - Scheduling one or more data packets in a cdma communication system based on computed power resource availability

Info

Publication number: GB2402580A
Application number: GB0313099A
Authority: GB
Inventors: Stephen Barrett; Eric Villier; Peter Legg
Original assignee: Motorola Inc
Current assignee: Motorola Solutions Inc
Priority date: 2003-06-06
Filing date: 2003-06-06
Publication date: 2004-12-08
Anticipated expiration: 2023-06-06
Also published as: GB2402580B; WO2004109989A2; EP1636947A2; GB0313099D0; WO2004109989A3

Abstract

A method (200) for scheduling one or more data packets in a code division multiple access (CDMA) communication system (100) comprises the steps of measuring power by the one or more base sites (122-132) and computing an amount of downlink power resource needed to transmit one or more data packets using measurements made in the one or more of the base sites (122-132); and/or an amount of power resource availability into which one or more data packets can be scheduled. The method also comprises scheduling the one or more data packets based on the computed amount of power. In this manner, a CDMA communication system is able to determine whether sufficient power resource exists in order to schedule one or more data packets, thereby improving scheduling efficiency in scenarios where transmit power in the communication system is highly variable.

Description

COMMUNICATION SYSTEM, COMMUNICATION UNIT AND METHODS OF

SCHEDULING TRANSMISSIONS THEREIN

Field of the Invention

This invention relates to scheduling of transmissions by a communication unit such as a radio network controller operating in a wireless communication system. The invention is applicable to, but not limited to, efficient 1 0 scheduling of transmissions based on power computations.

Background of the Invention

Wireless communication systems, for example cellular 1 5 telephony or private mobile radio communication systems, typically provide for radio telecommunication links to be arranged between a plurality of base transceiver stations (BTSs) and a plurality of subscriber units, often termed mobile stations (MSs). 2 0

Wireless communication systems are distinguished over fixed communication systems, such as the public switched telephone network (PSTN), principally in that mobile stations move between BTS (and/or different service 2 5 providers) and in doing so encounter varying radio propagation environments.

In a wireless communication system, each BTS has associated with it a particular geographical coverage 3 0 area (or cell). A particular range defines the coverage area where the BTS can maintain acceptable communications with MSs operating within its serving cell. Often these cells combine to produce an extensive coverage area.

Present day communications systems, both wireless and wire-line, typically have a requirement to transfer data between communications units. Data, in this context, includes speech communication. Such data transfer needs to be effectively and efficiently provided for, in order to optimise use of limited communication resources.

One such wireless communication system is a second 1 0 generation mobile telecommunication system, known as the Global System for Mobile communications (GSM). GSM is an established wireless communication system. The communication technology under-pinning the GSM standard has recently been supplanted by technologies offering 1 5 additional features. One of these technologies is the General Packet Radio System (GPRS), which complements the GSM system by carrying packet data on a sub-set of GSM time-slots. GPRS is a time division multiple access (TDMA) communication system, which makes use of the known 2 0 cell frequency re-use principle, wherein a frequency is not re-used in another (adjacent) cell.

In such wireless communication systems, it is imperative to efficiently use the limited radio (bandwidth) 2 5 resource. Thus, scheduling mechanisms are often employed to ensure that data packets are transmitted efficiently throughout the system. In TDMA systems, such as a GPRS system, the scheduler always schedules timeslots and transmits data in a timeslot at approximately the same 3 0 power level. The scheduler may optionally make use of carrier-to- interference (C/I) measurements from the subscriber unit in order to modify the coding and/or modulation scheme that is/are used to maximise data - 3 - throughput. Any effects of wireless communication unit / BTS power levels and/or interference in a GPRS system are typically managed off-line, for example once a week via a frequency re-planning process.

High Speed Downlink Packet Access (HSDPA) is an enhancement of R99 UMTS. In many ways, HSDPA performs like a GPRS/EDGE scheduler in that the transmit power per subscriber unit (termed user equipment (UK) in UMTS 1 0 parlance) is fixed and the data rate is modified dependent upon the C/I measurement received from the UK.

A HSDPA data packet scheduler performs scheduling based on C/I information reported solely by the UK. Notably, data packets cannot be scheduled in soft hand-off. 1 5

The GSM system has also been the forerunner for third generation mobile communications technologies, such as the Universal mobile telecommunication system (UMTS).

UMTS employs code division multiple access (CDMA) 2 0 technology. In a CDMA system, both the frequency resource and the time resource are shared amongst users using individual 'codes' to distinguish between different communications sent on the same frequency at the same time. Thus, multiple users are able to transmit on the 2 5 same frequencies at the same time.

In a CDMA system, the frequency re-use factor is '1', i.e. frequencies are re-used in adjacent cells. In this regard, it is the responsibility of the system to 3 0 dynamically manage interference (transmit power) levels.

Furthermore, there is much more variability in the required power per channel in a CDMA system than in a 4 - comparable TDMA system. It is also noteworthy that users in different locations may require vastly different transmit powers in order to receive the same data rate at a given quality level.

Thus, there exists a need in the field of CDMA

communication systems, to provide an improved scheduling mechanism, wherein the abovementioned disadvantages may be alleviated. 1 0

Statement of Invention

In summary, the inventive concepts of the present

invention provide a scheduling mechanism to allow 1 5 transmitter power levels to float in a CDMA communication system, such that power is managed dynamically. Such an approach effectively results in different scheduler designs to those currently used.

2 0 The inventive concepts herein described also provide a way of determining whether enough downlink power resource exists to serve a quantity of data packets. In particular, the inventive concepts provide an improved mechanism for scheduling packet data onto a power 2 5 controlled CDMA channel, for example to support a UMTS downlink shared channel (DSCH). Advantageously, an enhanced embodiment of the present invention may also be used to schedule access to dedicated channels (DCHs).

3 0 Brief Description of the Drawings

Exemplary embodiments of the present invention will now be described, with reference to the accompanying - 5 - drawings, in which: FIG. 1 illustrates a block diagram of a UMTS CDMA cellular radio communications system adapted to support the various inventive concepts of embodiments of the present invention; FIG. 2 illustrates a flowchart of a scheduler scheduling data packets for transmission dependent upon one or more 1 0 power computation(s) in accordance with an embodiment of the present invention; FIG. 3 illustrates a number of timing diagrams associated with scheduling data packets for transmission dependent 1 5 upon one or more power computation(s) in accordance with an embodiment of the present invention; FIG. 4 illustrates a flowchart of a data packet scheduling operation for a communication system 2 0 supporting a variety of transmission time intervals in accordance with an embodiment of the present invention; FIG. 5 illustrates a number of timing diagrams associated with scheduling data packets for transmission in an 2 5 asynchronous communication system in accordance with an embodiment of the present invention; FIG. 6 illustrates a flowchart of a data packet scheduling operation for a communication system 3 0 supporting both asynchronous and synchronous transmissions in accordance with an embodiment of the present invention; and FIG. 7 illustrates a conceptual data flow diagram for a communication system supporting a super-scheduler operation in accordance with an embodiment of the present invention.

Description of Preferred Embodiments

Referring now to FIG. 1, a cellular-based telephone communication system 110 supporting a Universal Mobile 1 0 Telecommunications Standard (UMTS) air-interface is illustrated, in outline, in accordance with a preferred embodiment of the invention. Pluralities of subscriber units (UEs) 112-116 communicate over the UMTS air- interface 118-120 with a plurality of Node Bs 122-132. A 1 5 limited number of UEs 112-116 and Node Bs 122-132 are shown for clarity purposes only. Each Node B 122-132 contains one or more transceiver units and communicates with the rest of the cellular system infrastructure via an Tub interface. The Node Bs 122-132 may be connected 2 0 to external networks, for example, the public-switched telephone network (PSTN) or the Internet 134 through Radio Network Controller stations (RNC) 136, 138, 140 and any number of mobile switching centres (MSCs) 142 and Serving GPRS Support Nodes (SGSN) 144. 2 5

Each RNC 136, 138, 140 may control one or more Node Bs 122-132. Each MSC 142 (only one shown for clarity purposes) provides a gateway to an external network 134, whilst the SGSN 144 links to external packet data 3 0 networks.

The Operations and Management Centre (OMC) 146 is operably connected to RNCs 136, 138, 140 and Node Bs 122 - 7 - 132, and administers and manages functions within the cellular telephone communication system 110, as will be understood by those skilled in the art.

In summary, the inventive concepts of the present

invention provide mechanisms for determining whether enough downlink power resource exists to serve a given quantity of data packets. In the preferred embodiment of the present invention, the algorithm runs on per node B 1 0 or per cell basis, i.e. there is one algorithm per cell or per Node B. Preferably, the scheduling algorithm runs once every 10 msec., which is the smallest transmission time interval (TTI) employed in a UMTS communication system. 1 5

In accordance with a preferred embodiment of the present invention, one or more RNCs, for example RNC 136 comprise a scheduler function 137. The scheduler function 137 includes one or more processors to perform one or more of 2 0 the improved scheduling algorithm(s) described hereafter.

The scheduler function 137 has preferably been adapted, inter alla, to compute power requirements per data packet. The scheduler function 137 has also been adapted 2 5 to compute a power budget for scheduling data packets, as described below with reference to FIG. 2.

Advantageously, when applied to a scenario where one or more dedicated channel (DCH) and/or one or more downlink shared channel(s) exists, the scheduler is configured to 3 0 build a single queue into which packets (destined for either the DCH or DSCH) are arranged and served in order of a global priority, as described below with reference to FIG. 6. In this scenario, it is noted that base sites supporting DCH transmissions are configured asynchronously. As such, the scheduler has been adapted to provide for, and utilise, multiple power bins in the timing structure to handle multiple, variable TTIs.

More generally, one or more RNCs may be adapted to incorporate the improved scheduler 137 in any suitable manner, according to the preferred embodiments of the present invention. For example, new apparatus may be 1 0 added to a conventional communication unit (for example RNC 136). Alternatively existing parts of a conventional RNC unit may be adapted, for example, by re-programming one or more processors therein. As such the required adaptation (say, to introduce power computation 1 5 algorithms and/or scheduling of different channel-types) may be implemented in the form of processor-implementable instructions stored on a storage medium, such as a floppy disk, hard disk, programmable read only memory (PROM), random access memory (RAM) or any combination of these or 2 0 other storage media.

It is also within the contemplation of the invention that such aforementioned functions/algorithms may reside in other network elements, or alternatively be distributed 2 5 amongst two or more of such network elements in other wireless communication systems. Furthermore, alternative radio communication architectures could benefit from the inventive concepts described herein, and the inventive concepts are not considered as being limited to the 3 0 specific UMTS configuration illustrated in FIG. 1.

Referring now to FIG. 2, a flowchart 200 illustrates the preferred embodiment of the present invention. The 9 - flowchart 200 starts in step 205, followed by a determination of whether a new carrier power measurement has arrived at the RNC, as shown in step 210. If a new carrier power measurement has arrived at the RNC, in step 210, then the process then moves to a first stage where that new carrier power measurement is used to compute a new value for the power resource, which is predicted to be available for the transmission of scheduled traffic in frames for which no traffic has yet been scheduled, as 1 0 shown in step 218.

If a new carrier power measurement has not arrived at the RNC, in step 210, then the process moves to an alternative first stage, in step 215, where a power 1 5 budget consumed by one or more channels is computed based on the latest computation. In a preferred enhancement of the inventive concepts of the present invention proposes to compute the power budget required to transmit one or more data packets. In accordance with an enhanced 2 0 embodiment of the present invention, power is preferably categorized in two ways, i.e. the power budget that is under the control of the scheduler, Pscheduled, and the power that is not under the control of the scheduler, Punscheduled. Indeed, the above two power computations are 2 5 inextricably linked in the algorithm discussed herein, since the estimate for PUnscheduled is essentially the total power measured over some previous interval minus the power budget PScheduled measured in that previous interval. 3 0

In the context of the present invention, the term "power budget" is used to refer at least to the fact that the - 10 Node B power amplifier has some maximum transmit power limitation. As an example, let us assume that a Node B cannot output a power level higher than 20W. This output power level is then defined as the "power budget", i.e. the total amount of transmit power available.

In the enhanced embodiment of the present invention, the first stage in step 215 determines, Punscheduled' idea how much power channels that are outside the control of the 1 0 scheduler consume. Preferably, Punscheduled is further sub-divided into: (i) The power that is attributable to power controlled sources: hereinafter termed Punscheduled-pc; and (ii) The power that is attributable to non-power 1 5 controlled sources (common channels): hereinafter termed PunscheduIed _cch.

Preferably, in this first stage, the scheduler estimates the power consumed by unscheduled services every T 2 0 seconds, for example, every lOOmsec's. In this regard, the scheduler computes: Punscheduled (nT) = Pcarrier(nT)Pscheduled (nT) [1] 2 5 Where: Pcarrier (nT) is the averaged total transmit power per carrier and per cell measured by the Node B over the period n-l(T) to nT; and Pscheduled (nT) is the averaged amount of scheduled 3 0 power over the period (n-l)T to nT, computed by the scheduler in the RNC and based on measurements that are made at the Node B. Scheduled (nT) is preferably computed using the following equation: f =r, UK= NUE PdpCCh (UK, f, nT)K(UE, f, nT) Pschelu/e (rl T) = f UK=! FT [ 2] Where: 1 0 K(UE, f,nT) is a scalar value that is applied to the dedicated physical control channel (DPCCH) power, in order to provide the dedicated physical data channel (DPDCH) power for packets scheduled to, say a UE in frame f of measurement interval (n-l) T to nT.

1 5 NUE - is the number of scheduled UEs.

PdpCCh (UK, f, nT) - is the measurement of DPCCH transmit power which was made at the Node B for the UE at the time which is closest to and preceding frame f of period (n 1)T to nT. . 2 O FT - is the number of frames in the interval T (i.e. x lOmsec. frames in lOOmsec.).

It is noteworthy that the DPCCH carries a limited amount of information that is used by the UE physical layer 2 5 (e.g. pilot, power control commands, eta). In contrast, the DPDCH carries the user plane data (i.e. the application traffic). The amount of data carried on the DPDCH can vary greatly (depending on how much data is scheduled), whilst the required DPCCH power remains 3 0 relatively stable from one service type to another. For this reason the ratio of DPDCH to DPCCH power will vary - 12 from one frame to another (or from one service to another) depending on how much data is carried.

It is also noteworthy that K(UE, f,nT) is a function of the amount of data scheduled.

In order to compute how much power that unscheduled channels consume in the interval being scheduled, two approaches are proposed: 1 0 Approach 1: In this first approach it is assumed that unscheduled power remains the same in the interval being scheduled, as it was in the previous interval ((n-l)T to nT), for 1 5 which the last PCarrier measurement result is available.

In this regard, the power available for transmission of data packets, Pbudget is given by: 2 0 Pbudget=PmaxPunscheduled(nT) [3] Where: PmaX is the total power available.

It is noteworthy that Pbudget also varies as a function of 2 5 time, in that it is re-computed every time a new Pcarrier measurement becomes available (e.g. every lOOmsec).

Approach 2: In this second approach, it is assumed that the power of 3 0 the unscheduled power controlled channels increases in proportion to Pmax/Pcarrier, where PmaX is again the total power available (e.g. where a cell is served by a power amplifier that has a rated power output of 20 W. then PmaX would be 20W). In the preferred embodiment of the present invention, the underlying assumption is that the amount of interference is scaled in direct proportion to the total power transmitted. The rationale for this assumption is generally valid, not least for a single cell system where only intra-cell interference exists.

1 0 For the purposes of the preferred scheduling algorithm, the total power transmitted is assumed to be PmaX. In other words in determining how much power each packet will require, the scheduler assumes that at the end of the schedule all available power will be used. 1 5

Pbudge t = Pmax - ( Pcch+ Puns cheduled pa ( nT).Pmax/Pcarrier(nT)) [4] Where: 2 0 Punscheduled_pc (nT) = Punscheduled (nT) - Pcch [5] Here, PCch may be assumed to be fixed. Alternatively, in a preferred embodiment of the present invention, PCch is more accurately determined using knowledge about how much 2 5 data was transmitted, for example on the forward access channel (FACH) and the paging channel (PCH), during the interval (nT).

As mentioned, this process for calculating Pbudget is 3 0 performed once every T msec, say typically every lOOmsec, when a new PCarrier measurement is received from the Node B. This value of PUnscheduled is used for every frame that is scheduled, up until the time that a new value of Punscheduled is calculated.

The flowchart of FIG. 2 then moves on to stage 2, in step 220, where one or more data packet(s) is/are taken from the queue of data packets that is to be transmitted.

Notably, the scheduler computes the amount of power that would be required to transmit the data packet in any of a 1 0 number of ways. Note that in current TDMA scheduling techniques, such as those used in GPRS or HSDPA-based systems, the power per packet is assumed to be fixed. In such systems the data rate is modified, by changing the modulation and channel coding used in order to make best 1 5 use of the 'fixed' power allocation.

However, in a preferred embodiment of the present invention, for each data packet an estimate is made of how much power it will consume. Three novel techniques 2 0 have been identified, and described below, to estimate/compute the amount of power required by each data packet. An underlying commonality of all three techniques is that they all preferably make use of the Node B Application Part (NBAP) "Transmitted Code Power" 2 5 measurement, which is performed in the Node B. The NBAP is the UMTS RNC <-a Node B signalling protocol. Notably, the PCarrler measurement is actually the NBAP measurement of "Transmitted carrier power".

3 0 Power estimate/computation techniques: (i) For simplicity, a first technique will be referred to as the "AS_IS" technique. In this first technique, an assumption is made that the DPCCH power measured during some preceding interval will be substantially equivalent to the power that would be seen during a future scheduling interval. Following this assumption, the power required to transmit a data packet on the DPDCH is therefore provided by the last available PdpCch measurement multiplied by the appropriate 1 0 DPDCH/DPCCH scaling factor 'K', which is dependent upon the number of bits to be transmitted, as shown below in equation [6].

PDPDC.H PdPCCh K [ 6] 1 5 (ii) A second technique will be referred to as a "MAX" technique. Equation [6] above is employed, but in this technique the PdpCch term that is used is the largest power level measured over some preceding 2 0 interval, Tprecede.

(iii) A third technique will be referred to as a "FRACTION" technique. In this third technique, it is assumed that the proportion of the total transmit power 2 5 that is consumed by the user's DPCCH (as measured in the previous measurement period) will remain the same during the future interval for which the schedule is being built.

3 0 Notably, the rationale is similar to that discussed previously when calculating Pbudget for "Approach 2".

The objective here is to factor in the situation where a - 16 scheduler may pack power up to PmaX (when previously it had not). Here, the amount of intra-cell interference will increase (along with inter-cell interference to some extent), such that the absolute power required to transmit the packet will be greater. In this regard, the power estimate/computation equation is: P. -p Pmax K packet dpcch p [ 7] carrier 1 0 A skilled artisan will appreciate that in an enhanced embodiment of the present invention, the "FRACTION" technique is preferably used in conjunction with the "Approach 2" method for computing the Pbudget. In this 1 5 manner, both techniques attempt to take into account the same issue, i.e. that the power consumed to transmit a certain amount of data will scale in proportion to the total carrier transmit power.

2 0 A skilled artisan will also appreciate that the timing of such measurements may vary from system to system, and scheduler to scheduler. Thus, the time subscripts for the equation variables given above may well vary. For example, PdpCch measurements and PCarrier measurements may 2 5 arrive asynchronously. Even though these measurements may arrive with the same period of 100 msec., the scheduler is likely to be building a schedule for a different 10 msec period. In scenarios such as this, a skilled artisan will appreciate that the time subscripts 3 0 for the equation variables should be adapted accordingly. - 17

Returning now to the flowchart 200 of FIG. 2, once the power required per data packet has been computed/estimated in accordance with any one (or more) of the above techniques, the process moves on to stage 3 in step 225.

A determination is made as to whether sufficient power resource is available to serve the data packet that is next in the queue to be transmitted, in step 225. In 1 0 this regard, as the scheduler is building a schedule of data packet transmissions, the scheduler preferably maintains a variable, PUsed' which is the total power allocated up to that point in the schedule. At the beginning of the schedule operation, the preferred 1 5 approach is to set: Pused= Pmax- Pbudget (calculated according to approaches described in Stage 1) Thereafter, as each data packet is considered for 2 0 scheduling, the following check algorithm is preferably performed: If: Fused + Ppacket < Pmax (Stage 3, step 225) 2 5 AND code resource exists (Stage 4, step 230)

AND

Transport Format Combination (TFC) would be valid (Stage 4, step 230) 3 0 then: Schedule the data packet (Stage 6, step 235) Fused = Pused + Ppacket - 18 Update record of code power usage for that frame and that cell (if packet transmitted on DSCH) Update record of TFC usage in that frame for that UE Else: skip packet and take another data packet from the queue (Stage 5, step 240) End.

1 0 A Transport Format Combination (TFC) is defined as the combination of currently valid transport formats on all transport channels of a UK, i.e. containing one transport format from each transport channel. Although the preferred embodiment of the present invention comprises 1 5 determining whether both a code resource exists and that the TFC would be valid, it is envisaged that a less- efficient, but possibly acceptable, scheduling algorithm may use one or indeed neither of these optional steps.

For example, it is envisaged that a code check may not be 2 0 required for scheduling DCH traffic or a TFC check may not need to be checked if it is checked elsewhere in the system, such as by the medium access control (MAC) layer.

Indeed, in alternative embodiments, it is envisaged that the code and TFC checks may be more critical in many 2 5 respects than a power level check.

In the preferred embodiment of the present invention, Stage 6 (step 235) is preferably enhanced, such that when a data packet is scheduled the scheduler updates its 3 0 records of available power and code resource used for that particular scheduling interval. The key variable relating to power usage that is preferably updated following the scheduling of a packet is the variable Pused. Pused is a record of the predicted amount of power consumed by both unscheduled traffic and data packets that have been previously scheduled. There will be one value of Bused for each power bin and for each cell.

Associated with each cell and each frame for which there is a power bin, there is also preferably a record of DSCH code usage. This record could take the form of a list of all the possible DSCH channelisation codes. For each 1 0 DSCH channelisation code there is preferably a binary indicator, indicating whether the code has been previously assigned or not. When a data packet is scheduled on the DSCH, then the binary indicator corresponding to the channelisation code selected to 1 5 carry the data packet is preferably toggled.

Associated with each scheduled user and each frame for which there is a power bin, there is preferably an indication of the transport formats of data packets that 2 0 have already been scheduled for transmission. When a new data packet is scheduled for that user, then this record will be updated. The preferred algorithm uses the record to check that if a new data packet is scheduled, then the resulting transport format combination will be valid. 2 5

In an enhanced embodiment of the present invention, the aforementioned process(es) is/are preferably applied by the scheduler in order to handle a number of different Transmission Time Intervals (TTIs). Transmission Time 3 0 Interval is defined as the period over which the bits of the scheduled transport block are coded, interleaved and then transmitted. 20

Referring now to FIG. 3, a timing diagram 300 illustrates various durations over which data packets may be transmitted. The lOmsec. framing structure for the cell is illustrated in time line 315. An offset 305 is shown between the CCH framing and the framing of the DCH (illustrated in time line 330. The power transmission of data packets for a lOmsec period is illustrated in time line 325. The power transmission may be used in a variety of channel types, for example, a forward access 1 0 channel (FACH), a broadcast channel (BCH) or a paging channel (PCH) 320, or the higher power transmission of a physical downlink shared channel (PDSCH) 325. A first example transmission time line is shown for UE#1 330, which transmits data packets over a 40-msec. period on a 1 5 DCH (i.e. TTI=40msec). A second example transmission time line is shown for UE#2 335, which has a lOmsec TTI to transmit data on a DCH.

In order to handle various TTIs, a scheduling algorithm 2 0 employed by the scheduler maintains "power bins" that span from the current frame to at least eight frames into the future, since there is an 80msec. TTI frame.

In this context, the "power bins" refer to data 2 5 structures within the RNC/scheduler. Each power bin is preferably associated with one of the CCH frames. Each of the power bins/data structures map to successive CCHframes. The number of frames covered has to be as large as the longest duration TTI that can be scheduled and 3 0 would be no larger than eight, since the largest TTI is 80msec. and each frame lasts lOmsec. - 21

In an enhanced embodiment of this aspect of the present invention, nine frames/power bins are maintained in order to account for the fact that DCH traffic is asynchronous and, as described previously, the TTI for any one DCH will typically spill over CCH frame boundaries.

To provide further clarification, let us define a point of time at which the schedule is running as a "Frame_ now" point (i.e. the current CCH frame). In this case, the 1 0 CCH frame corresponding to the first power bin will be "Frame_ now+p", (where p might be set at 3). A delay is inserted to provide sufficient time for the data packets to be transferred from the RNC to the Node B. The delay also supports DSCH, for which Transport Format 1 5 Combination Indicator (TFCI) signalling information has to be sent in advance of the packet arriving on the DSCH.

TFCI provides, in effect, a means to inform a UE whether a data packet has been scheduled for it on the DSCH and what physical layer processing will have been applied. 2 0

The data structure corresponding to each power bin will preferably include the following types of information: (i) A record of the total power resource allocated to date in the corresponding CCH frame (this includes 2 5 unscheduled traffic on DCH, CCH channels, scheduled traffic, etc.); (ii) A record of which blocks have been scheduled to each user (this may be required when performing a check to see whether any additional blocks that are scheduled 3 0 would still result in a valid TFC in step 230); and (iii) A record of which code resources have been allocated to DSCH users. - 22

When a new power bin is opened, say at the beginning of a new frame/scheduling interval, it is initialised by setting the amount of power consumed to that predicted to be consumed by common channels and unscheduled traffic.

Likewise, the record of code usage will be initialized, by recording as allocated all codes known to be assigned to CCH and DCH channels (i.e. all channels excluding DSCH).

1 0 Hence, in accordance with the enhanced embodiment of the present invention, if a data packet is scheduled whilst the scheduler is running (in its lOmsec. periods) with a TTI greater than lOmsec., the scheduler will start setting aside power resource in the corresponding future 1 5 frames (i.e. in future power bins), as illustrated in timing diagram 330.

Referring now to flowchart 400 of FIG. 4, a preferred scheduler algorithm is described for handling different 2 0 TTIs. In this regard, stage 3 in step 220 of FIG. 2 is extended by the use of multiple power bins to handle scheduling of different TTIs.

In particular, the process of taking the next data packet 2 5 from the queue in step 220 is adapted by then determining a TTI associated with the data packet, as shown in step 410. If it is determined by the scheduler that this data packet has a lOmsec. (or less) TTI, in step 410, then the scheduler determines whether the necessary code and power 3 0 resources exist in each of the power bin(s) corresponding to the TTI, in step 420. In addition, it is envisaged that the scheduler may make other determinations, such as whether the TFC is valid. This is in a similar manner to - 23 step 220 and step 225 of FIG. 2. If adequate resource exists, in the determination in step 420, then the data packet is scheduled in the normal manner, in step 240.

However, if the scheduler determines that the data packet has a TTI of greater than lOmsec., in step 410, the scheduler determines whether a 'Currentframe' value, which is being scheduled, maps to the first frame in the TTI super-framing. In this regard, TTI super-framing 1 0 refers to the notion that a packet/transport block cannot be transmitted in just any frame. For example the first frame of a transport block that is transmitted with an 80 msec TTI, could be transmitted in frame n, n+8, n+16, but not in frame nil, n+2. For a transport block with a TTI 1 5 of 40 ms, the first part of the transport block could be transmitted in frame n, n+4, and n+8. In this context Currentframe' is the lowest CCH frame number associated with any of the power bins.

2 0 If the scheduler determines that the 'CurrentErame' being scheduled does map to the first frame in the TTI super- framing structure for the UE corresponding to the packet being scheduled, in step 415, the process moves to step 420 with the use of multiple power bins enabling the 2 5 provision of TTIs of greater than lOmsec. If the scheduler determines that the 'Currentframe' being scheduled does not map to the first frame in the TTI super-framing, the scheduler skips this data packet in step 430 and moves to the next data packet in step 220. 3 0

In this manner, by use of multiple power bins that span the current frame as well as a number of subsequent - 24 frames, the scheduler is able to handle various TTIs in the data packets to be transmitted.

The UMTS standard was written such that Node Bs do not have to be synchronized. As the Node Bs are not synchronized, the common channel framing is offset between each Node B. In addition, the DCH framing of one UP is offset from that of another. Indeed, different offsets for DCH channels may be selected deliberately by 1 0 the respective Node Bs in order to smooth the traffic load, to ensure that multiple frames do not arrive for processing at substantially the same time. However, it is noteworthy that in the UMTS standard, since the DSCH is transmitted in one-way hand-off (i.e. hard-hand-off), 1 5 DSCH frames are always synchronous with the framing on the common channels.

Notably, there are no known current systems that employ any scheduling of either DCH or DSCH channels. By 2 0 implication, the aforementioned scheduling operations of the preferred embodiments of the present invention can readily be applied to a synchronous DSCH case.

However, in a further enhanced embodiment of the present 2 5 invention, the aforementioned processes are preferably applied in a scenario where a scheduler needs to accommodate asynchronous framing in different Node Bs, for scheduling access to asynchronous channels such as DCHs. 3 0

The preferred scheduling algorithm addresses the asynchronicity of DCH scheduled power transmissions.

-

In this further enhanced embodiment, it is assumed that the "power bins" (unlike the asynchronous timing structure) are aligned with the common channel (CCH) framing. When power bins are allocated in this manner, no special procedures are needed to deal with asynchronicity when scheduling the DSCH. The preferred approach to asynchronicity of a DCH is to apportion power consumption in proportion to an amount of overlap of the asynchronous transmission in each frame. 1 0

This principle is illustrated in the timing diagrams 500 of FIG. 5. The common channel timing structure is shown in a first timing diagram 510, with cell framing of lOmsec periods. The timing of a first UE (UE#1) is shown 1 5 in timing diagram 520, which illustrates the asynchronicity of the DCH channel with the common channels, i.e. a DCH offset 530 exists. A data packet 535 is taken by the scheduler and scheduled for transmission in frame 'm' 515 and frame m+1. The amount 2 0 of power 540 required to transmit the data packet is given as PO x power_scale. The scheduler allocates resource according to a general principle, as follows: (i) power_scale x Po; is allocated in the first 2 5 overlapping frame; (ii) (1-power_scale) x PO; is allocated in the last overlapping frame; and (iii) PO; is allocated in all frames in between.

3 0 In the particular example of FIG. 5, the scheduler allocates Po x power_scale in frame m; and PO x (1 - 26 power_scale) in frame 'm+1', where the power_scale is defined as: Power_scale = (10-DCH offset)/10 [8] Where '10' relates to the lOmsec frame duration and DCH_offset is measured in msec.

In a yet further enhanced embodiment of the present 1 0 invention, a scheduler in a scenario may apply the aforementioned process where both DSCH and DCH resources exist. In this regard, stage 3 (step 220) of FIG. 2 is preferably adapted according to the flowchart 600 of FIG. 6. Thus, following a power computation of the required 1 5 power to transmit a data packet in step 605, the scheduler ascertains whether the data packet is to be transmitted on a DCH or a DSCH, as shown in step 610. If the data packet is to be transmitted on a DSCH in step 610, the process follows that shown in the flowchart 200 2 0 of FIG. 2. However, if the data packet is to be transmitted on a DCH in step 610, the scheduler distributes power according to the previous steps (i), (ii) and (iii).

2 5 Advantageously, the aforementioned scheduling operation applied to both DCH and DSCH channels can be extended to other channels, for example, a Forward Access Channel (FACH) could be optionally included in the queue in the same manner as above. Notably, the FACH channel is not 3 0 power controlled. Thus, transmit power settings on the FACH are designed so that coverage over the entire cell area is achieved. Hence, the difficulty associated with determining how much power will be consumed when - 27 scheduling a data packet on a FACH is straightforward, as illustrated in equation [9] below: Power consumed in each frame of the TTI = (Eb x Number of bits in frame)/lOmsec. [9] Where: Eb is the energy required to transmit one bit of information on the FACH. In addition there are no asynchronicity problems to deal with for the FACH. 1 0

Data packets can therefore be ordered in the queue according to a priority - irrespective of whether the supported channels are DCH and/or DSCH and/or FACH and/or any other suitable channel. 1 5

In this yet further enhanced embodiment of the present invention, the primary difference is that for the DCH scenario there is no code check, as code resource for DCH users is typically allocated and fixed at call set-up.

2 O In addition, where packets are scheduled on DCH, the asynchronicity with respect to the CCH framing has to be dealt with as described earlier. Furthermore, super- scheduler approaches can be applied as described in the next section. 2 5

Super-scheduler option: An optional enhancement to each of the aforementioned scheduling algorithm embodiments is the use of super 3 O scheduling within the cells of a Node B. The use of a super-scheduler is illustrated in the data flow diagram of FIG. 7. In this scenario, it is assumed that power bins are allocated for every cell, where a 'super scheduler' is configured to optimise resource across a number of Cells, e. g. Cell #A 720, Cell #B 730 and Cell #C 740 as shown. A queue of traffic channel data packets 705 is waiting scheduling in a Node B. It is shown that data packet '4', which is carried on DCH in SHO, consumes scheduler power in the power bins for Cell #A 720 and Cell #B 730 of the Node B. Data packet '1', which is a DSCH packet, is solely carried by Cell #A 720.

1 0 "Unscheduled power" 725, 735, 745 for each Cell is taken into account by the super-scheduler. The super-scheduler deducts the unscheduled power, preferably computed as previously described, to determine the power budget in each cell. In this manner, the super-scheduler is able 1 5 to determine how much of the power budget remains for the super- scheduler to schedule non-real-time (NRT) traffic.

The advantage of this super-scheduler configuration is that it manages power more tightly, provides greater throughput and improved quality of service (QoS) over the 2 0 alternative technique of having one scheduler per cell.

In such alternative schemes, where there is just one scheduler per cell, NRT traffic that is scheduled in one cell will appear as "unscheduled power" from the perspective of the schedulers associated with the other 2 5 cells.

It is within the contemplation of the present invention that a downlink super-scheduler would preferably include an ability to schedule data packets across multiple Node 3 0 Bs, and therefore the inventive concepts are not limited to scheduling data packets solely within a single node B as discussed above. In this regard, the aforementioned inventive concepts would be substantially the same where - 29 the scheduler is able to deal with multiple Node Bs.

However, the scheduling entity would be extended to maintain a single queue containing data packets destined for many different Node Bs.

Scheduler cell-selection option: It is envisaged that a further embodiment of the present invention can be applied where the DCH of a UE is carried 1 0 in SHO. Here, since the scheduler runs in a distributed manner, it is necessary to decide into which Node B queue (or cell queue) the data packets should be sent. In this regard, let us assume that there is an individual scheduler associated with each cell (or Node B or 1 5 collection of Node Bs if the super-scheduler concept is used). The scheduling algorithms preferably run autonomously. Hence, if NRT data is scheduled on a DCH that has soft handover legs spanning for example three cells, then potentially there could be different 2 0 schedulers associated with each cell, and a selection has to be made of which scheduler will be in control for that user.

This algorithm is preferably implemented as follows. A 2 5 simple scheduler cell-selection technique is to make the scheduler cell that which is identified as the best cell by, say, a SHO event type ID, as supported in the UMTS standard. This event type indicates that a change in the best cell has occurred based on measurements made by the 3 0 Node B. Alternatively, if there is a scheduler per Node B (or some larger grouping of Node Bs), then the "controlling"

-

scheduler for scheduling traffic on a certain DCH could be selected as the node B (or larger grouping of Node Bs) that includes the best cell.

It is also within the contemplation of the present invention that any combination of one or more of the above inventive concepts could be employed in the improved scheduler function in the RNC.

1 0 Although the preferred embodiment of the present invention has been described with reference to a UMTS communication system, it is envisaged that the inventive concepts are equally applicable to other telecommunication systems, either wireless or wireline, 1 5 including for example core networks or backbone networks.

Furthermore, it is envisaged that the inventive concepts described herein are, inter alla, equally applicable to IS-95 systems and their derivatives, as well as wireless Local Area Networks and even satellitebased systems. 2 0

Therefore, it will be understood that the improved scheduling mechanisms, as described above, provide at least one or more of the following advantages: (i) The preferred scheduling mechanisms improve 2 5 the scheduling efficiency in scenarios where transmit power in the communication system is highly variable, for example in CDMA-based systems.

(ii) The preferred scheduler provides a mechanism for determining whether sufficient downlink 3 0 power resource exists to serve a particular number of data packets, to optimise the use of the limited channel resources.

- - - - - 31

(iii) The preferred scheduler provides a mechanism for determining an amount of power per data packet, for example in the context of ascertaining whether sufficient downlink power resource exists to serve a particular number of data packets.

(iv) The preferred scheduler provides a mechanism for scheduling in DSCH and/or asynchronous DCH and/or FACH.

(v) The preferred scheduler provides a 1 0 mechanism for handling multiple TTI.

(vi) The preferred scheduler may be a Super- scheduler, able to schedule data packets across multiple cells.

1 5 Whilst the specific and preferred implementations of the embodiments of the present invention are described above, it is clear that one skilled in the art could readily apply variations and modifications to the preferred embodiments that fall within the inventive concepts. 2 0

Thus, a communication system, a radio network controller, a scheduler and a number of scheduling methods have been provided wherein the aforementioned disadvantages of the

prior art have been substantially alleviated. 2 5 ( - 32

Claims

Claims 1. A method (200) for scheduling one or more data packets in a code

division multiple access (CDMA) communication system (100) comprising one or more base sites (136-140), the method characterized by the steps of: measuring power by the one or more base sites (122-132); 1 0 computing: an amount of downlink power resource needed to transmit one or more data packets using measurements made in the one or more of the base sites (122-132); and/or 1 5 an amount of power resource availability into which one or more data packets can be scheduled; and scheduling the one or more data packets based on the computed amount of power.
2 0 2. A method (200) for scheduling one or more data packets according to Claim 1, wherein the one or more base sites measurements comprise one or more of the following measurements: (i) a downlink transmit power measurement, 2 5 preferably a total power for a cell and carrier; and (ii) a transmit power per user channel measurement, for example a power measurement made on a dedicated physical control channel.
3 0 3. A method (200) for scheduling one or more data packets according to Claim 1 or Claim 2, wherein the step of computing is further characterized by one or more of the following steps: l - 33 computing, by a scheduler, an amount of power that is under control of the scheduler; and/or computing, by a scheduler, an amount of power that is not under control of the scheduler.
4. A method (200) for scheduling one or more data packets according to Claim 3, wherein the step of computing applies to power that is under control of the scheduler (Pscheduled) and power that is not under control 1 0 of the scheduler (PUnscheduled), the method further characterized by the step of: estimating Punscheduled by measuring a total power level used over a previous interval minus the power Pscheduled obtained for that previous interval. 1 5
5. A method (200) for scheduling one or more data packets according to Claim 3 or Claim 4, wherein the step of computing an amount of power that is not under control of the scheduler in an interval being scheduled is 2 0 further characterized by the steps of: assuming that an amount of unscheduled power remains substantially the same in the interval being scheduled as it was in a previous interval; or assuming that the amount of unscheduled power 2 5 consumed by power-controlled channels increases in proportion to a relationship of Pmax/Pcarrier
6. A method (200) for scheduling one or more data packets according to any preceding Claim, wherein the 3 0 step of scheduling is further characterized by the step of: - 34 identifying whether code resource is available; and/or identifying whether a transport format combination is valid; in determining whether to schedule one or more data packets for transmission.
7. A method (200) for scheduling one or more data packets according to any preceding Claim, wherein the 1 0 method is further characterized by the steps of: determining whether one or more data packets exceeds a transmission time threshold, for example a transmission time interval of greater than 10 msec, and scheduling transmission of the one or more data 1 5 packets in a plurality of transmission periods when the one or more data packets exceeds the transmission time threshold.
8. A method (200) for scheduling one or more data 2 0 packets according to any preceding Claim, wherein the step of computing is further characterized by the steps of: measuring a power level used on a dedicated control channel, for example a dedicated physical control 2 5 channel, in a first time interval; and estimating a power level required to transmit a data packet on a dedicated data channel, for example a dedicated physical data channel, in a second time interval by multiplying the measured dedicated control 3 0 channel power by a scaling factor, for example a scaling factor (K) dependent upon a number of bits to be transmitted.
9. A method (200) for scheduling one or more data packets according to Claim 8, wherein the step of estimating is further characterized by the step of identifying a largest value of measured dedicated control channel power over a preceding interval and multiplying a largest identified value of measured dedicated control channel power by the scaling factor to estimate the power required to transmit a data packet on the dedicated data channel in the second time interval. 1 0
10. A method (200) for scheduling one or more data packets according to Claim 8, wherein the step of estimating is further characterized by the steps of: determining a proportion of total power consumed 1 5 by the dedicated control channel in the first time interval; and assuming that the proportion will remain the same during the second time interval, for which the schedule is being built. 2 0
11. A method (200) for scheduling one or more data packets according to any of preceding Claims 8 to 10, wherein the CDMA communication system is a UMTS communication system and the step of measuring a power 2 5 level uses a Node B Application Part transmit code power measurement and/or a Node B Application Part transmitted carrier power measurement.
12. A method (200) for scheduling one or more data 3 0 packets according to any preceding Claim, wherein the CDMA communication system (100) supports an asynchronous timing structure used by a dedicated channel, the method further characterized by the steps of: - 36 maintaining multiple power bins corresponding to a number of consecutive frames, during which packets with different transmission time intervals are transmitted; and scheduling a use of the dedicated channel between asynchronous base sites of the CDMA wireless communication system (100) by allocating power resource to the maintained multiple power bins.

1 0
13. A method (200) for scheduling one or more data packets according to Claim 12, wherein the multiple power bins are aligned with control channel frames and the step of allocating power resource to the maintained multiple power bins comprises allocating power resource across 1 5 multiple subsequent frames from a frame being scheduled, such that scheduling of one or more data packets is based on a transmission time interval.
14. A method (200) for scheduling one or more data 2 0 packets according to Claim 12 or Claim 13, wherein the method is further characterized by the step of: maintaining one or more records of consumed power resource of the multiple power bins.

2 5
15. A method (200) for scheduling one or more data packets according to Claim 14, wherein the method is further characterized by the steps of: updating the one or more records of consumed power resource of the multiple power bins when scheduling 3 0 a data packet that exceeds a frame, for example scheduling a data packet with a transmission time interval of greater than lOmsec. - 37
16. A method (200) for scheduling one or more data packets according to any of preceding Claims 12 to 15, where an overlap exists between a downlink channel framing relative to a control channel framing, wherein the step of scheduling a use of the dedicated channel is further characterized by the step of: apportioning power resource substantially in proportion to an amount of overlap of an asynchronous transmission in one or more frames. 1 0
17. A method (200) for scheduling one or more data packets according to any preceding Claim, wherein the CDMA communication system (100) supports both synchronous and asynchronous channels, the method further 1 5 characterized by the steps of: ascertaining whether one or more data packet(s) is/are to be transmitted on the synchronous and/or asynchronous channels (610); and building a single queue into which the one or 2 0 more data packet(s) are arranged and served in order of priority.
18. A method (200) for scheduling one or more data packets according to Claim 17, wherein the CDMA 2 5 communication system is a UMTS communication system and the synchronous and asynchronous channels comprise one or more of the following: a dedicated channel, a Forward Access Channel (FACH), a downlink shared channel (DSCH).

3 0
19. A method (200) for scheduling one or more data packets according to any preceding Claim, wherein a super-scheduler performs the step of computing and scheduling in the following manner: - 38 computing an amount of downlink power resource needed to transmit one or more data packets across multiple cells or base sites; and/or computing how much power resource is available into which one or more data packets can be scheduled; and scheduling the one or more data packets across multiple cells or base sites, in response to the computed downlink power resource and/or computed available power resource. 1 0
20. A method (200) for scheduling one or more data packets according to Claim 19, wherein the super scheduler schedules the one or more data packets on downlink channels that are operating in soft hangover. 1 5
21. A method (200) for scheduling one or more data packets according to Claim 19 or Claim 20, the method further characterized by the step of: performing checks across multiple cells to see 2 0 whether the one or more data packets can be scheduled for transmission within respective cells.
22. A method (200) for scheduling one or more data packets according to any of preceding Claims 19 to 21, 2 5 the method further characterized by the steps of: allocating a scheduler and/or individual queue associated with each cell or each base site or each cluster of cells or each cluster of base sites, to facilitate the step of scheduling the one or more data 3 0 packets across multiple cells and/or base sites. - 39
23. A method (200) for scheduling one or more data packets according to any of preceding Claims 19 to 22, the method further characterized by the steps of: selecting a scheduler of a cell for transmission of one or more data packets by, say, a soft handover event; or selecting a scheduler of a Node B or grouping of Node Bs as the one that includes the best cell.

1 0
24. A radio network controller (136), adapted to support the method steps of any of the preceding Claims.
25. A communication system (100), for example a UMTS wireless communication system, adapted such that a 1 5 processor operational within the communication system performs the method steps of any of Claims 1 to 22. - 40
26. A radio network controller (136) managing base site resources in a CDMA communication system, the radio network controller (136) comprising: a receiving function, for receiving power measurements from one or more base sites (122-132); and a scheduler (137) comprising one or more processors configured to perform scheduling of one or more data packets within the CDMA communication system, wherein the radio network controller (136) is 1 0 characterized by the one or more processors arranged to compute an amount of downlink power resource needed to transmit one or more data packets using measurements made in one or more of the base sites (122-132) and/or compute how much power resource is available into which one or 1 5 more data packets can be scheduled and schedule the one or more data packets based on the computed amount of power.
27. A radio network controller (136) according to 2 0 Claim 26, wherein the one or more base sites measurements comprise one or more of the following measurements: (i) a downlink transmit power measurement, preferably a total power for a cell and carrier; and (ii) a transmit power per user channel 2 5 measurement, for example a power measurement made on a dedicated physical control channel.
28. A radio network controller (136) according to Claim 26 or Claim 27, wherein the scheduler (137) 3 0 computes an amount of power that is under control of the scheduler; and/or computes an amount of power that is not under control of the scheduler. - 41
29. A radio network controller (136) according to Claim 28, wherein the scheduler estimates power that is under control of the scheduler (Punscheue) by measuring a total power level used over a previous interval minus the power that is not under control of the scheduler (Pscheue0) obtained for that previous interval.
30. A radio network controller (136) according to any of preceding Claims 26 to 29, wherein the scheduler 1 0 identifies whether code resource is available; and/or identifies whether a transport format combination is valid; in determining whether to schedule one or more data packets for transmission.

1 5
31. A radio network controller (136) according to any of preceding Claims 26 to 30, wherein the scheduler is arranged to determine whether one or more data packets exceeds a transmission time threshold, for example a transmission time interval of greater than 10 msec, and 2 0 schedules transmission of the one or more data packets in a plurality of transmission periods when the one or more data packets exceeds the transmission time threshold.

32. A radio network controller (136) according to any 2 5 of preceding Claims 26 to 31, wherein the receiver receives a power level measurement of a dedicated control channel, for example a dedicated physical control channel, in a first time interval; and the scheduler estimates a power level required to transmit a data 3 0 packet on a dedicated data channel, for example a dedicated physical data channel, in a second time interval by multiplying the measured dedicated control channel power by a scaling factor, for example a scaling factor (K) dependent upon a number of bits to be transmitted.
32. A radio network controller (136) according to Claim 31, wherein the scheduler identifies a largest value of measured dedicated control channel power over a preceding interval and multiplies value by the scaling factor to estimate the power required to transmit a data packet on the dedicated data channel in the second time 1 0 interval.
33. A radio network controller (136) according to Claim 31, wherein the scheduler determines a proportion of total power consumed by the dedicated control channel 1 5 in the first time interval; and assumes that the proportion will remain the same during the second time interval, for which the schedule is being built.
34. A radio network controller (136) according to any 2 0 of preceding Claims 26 to 33, wherein the CDMA communication system is a UMTS communication system and the scheduler uses a Node B Application Part transmit code power measurement and/or a Node B Application Part Transmitted carrier power measurement in determining a 2 5 scheduling process.
35. A radio network controller (136) according to any of preceding Claims 26 to 34, wherein the CDMA communication system (100) supports an asynchronous 3 0 timing structure used by a dedicated channel and the scheduler (137) maintains multiple power bins corresponding to a number of consecutive frames, during which packets with different transmission time intervals - 43 are transmitted; and schedules a use of the dedicated channel between asynchronous base sites of the CDMA wireless communication system (100) by allocating power resource to the maintained multiple power bins.
36. A radio network controller (136) according to Claim 35, wherein the multiple power bins are aligned with control channel frames and the scheduler allocates power resource across multiple subsequent frames from a 1 0 frame being scheduled, such that scheduling of one or more data packets is based on a transmission time interval.
37. A radio network controller (136) according to any 1 5 of preceding Claims 26 to 36, wherein the scheduler is operably coupled to a memory element maintaining one or more records of consumed power resource.
38. A radio network controller (136) according to 2 0 Claim 37, wherein the scheduler updates the one or more records of consumed power resource in the memory element when scheduling a data packet that exceeds a frame, for example scheduling a data packet with a transmission time interval of greater than lOmsec. 2 5
39. A radio network controller (136) according to any of preceding Claims 35 to 38, where an overlap exists in a framing structure for data packets being scheduled, for example an overlap between a downlink channel framing 3 0 relative to a control channel framing, wherein the scheduler apportions power resource substantially in proportion to an amount of overlap of an asynchronous transmission in one or more frames.
40. A radio network controller (136) according to any of preceding Claims 26 to 39, wherein the CDMA communication system (100) supports both synchronous and asynchronous channels, and the scheduler ascertains whether one or more data packet(s) is/are to be transmitted on the synchronous and/or asynchronous channels (610); and builds a single queue into which the one or more data packet(s) are arranged and served in 1 0 priority order.
41. A radio network controller (136) according to Claims 40, wherein the CDMA communication system is a UMTS communication system and the synchronous and 1 5 asynchronous channels comprise one or more of a dedicated channel, a Forward Access Channel (FACH), a downlink shared channel (DSCH).
42. A super-scheduler of a CDMA communication system, 2 0 comprising multiple cells and multiple base sites, arranged to compute an amount of downlink power resource needed to transmit one or more data packets across multiple cells or base sites; and/or compute how much power resource is available into which one or more data 2 5 packets can be scheduled; and schedule the one or more data packets across multiple cells or base sites, in response to the computed downlink power resource and/or computed available power resource.

3 0
43. A super-scheduler according to Claim 42, wherein the superscheduler schedules the one or more data packets on downlink channels that are operating in soft handover. -
44. A super-scheduler according to Claim 42 or Claim 43, the super-scheduler further characterized in that it is arranged to perform checks across multiple cells to see whether the one or more data packets can be scheduled for transmission within the cells.
45. A super-scheduler according to any of preceding Claims 42 to 44, the super-scheduler further 1 0 characterized in that it is arranged to allocating a individual queues associated with each cell or each base site or each cluster of cells or each cluster of base sites to facilitate scheduling the one or more data packets across multiple cells and/or base sites. 1 5
46. A radio network controller (136), comprising the super-scheduler of any of preceding Claims 42 to 45.
47. A communication system (100), for example a UMTS 2 0 wireless communication system, comprising the radio network controller of any of Claims 26 to 41 or the super-scheduler of any of Claims 42 to 45. - 46
48. A method (200) for computing a power level per data packet in a data packet scheduling process in a code division multiple access (CDMA) communication system (100), the method characterized by the steps of: measuring a power used on a dedicated control channel, for example a dedicated physical control channel, in a first time interval; and estimating a power required to transmit a data packet on a dedicated data channel, for example a 1 0 dedicated physical data channel, in a second time interval by multiplying the measured dedicated control channel power by a scaling factor, for example a scaling factor (K) dependent upon a number of bits to be transmitted. 1 5
49. A method (200) for computing a power level per data packet according to Claim 48, wherein the step of estimating is further characterized by the step of: identifying a largest value of measured power 2 0 over a preceding interval and multiplying the largest value of measured power by the scaling factor to estimate the power required to transmit a data packet on the dedicated data channel in the second time interval.

2 5
50. A method (200) for computing a power level per data packet according to Claim 48, wherein the step of estimating is further characterized by the steps of: determining a proportion of total power consumed by the dedicated control channel in the first time 3 0 interval; and assuming that the proportion will remain the same during the second time interval, for which the schedule is being built. c - 47
51. A method (200) for computing a power level per data packet according to any of preceding Claims 48 to 50, wherein the CDMA communication system is a UMTS communication system and the step of measuring uses a Node B Application Part "transmit code power" measurement.
52. A radio network controller (136), adapted to support the method steps of any of preceding Claims 48 to 1 0 51.
53. A storage medium storing processor-implementable instructions for controlling a processor to carry out the method steps of any of Claims 48 to 51. 1 5
54. A communication system (100), for example a UMTS wireless communication system, adapted to facilitate the method steps of any of preceding Claims 48 to 51. 2 0 - 48
55. A method (600) for scheduling of data packets in a code division multiple access (CDMA) communication system (100) supporting an asynchronous timing structure used by a dedicated channel, the method characterized by the steps of: maintaining multiple power bins across a number of transmission time intervals for one or more data packets to be transmitted in the dedicated channel; and scheduling a use of the dedicated channel between 1 0 asynchronous base sites of the CDMA wireless communication system (100) by allocating power resource from the maintained multiple power bins.
56. A method (600) for scheduling of data packets 1 5 according to Claim 55, wherein the step of allocating power resource from the maintained multiple power bins comprises allocating power resource across multiple subsequent frames from a frame being scheduled.

2 0
57. A method (600) for scheduling of data packets according to Claim 55 or Claim 56, wherein the step of scheduling a use of the dedicated channel is further characterized by the step of: apportioning power resource in proportion to an 2 5 amount of overlap of an asynchronous transmission in one or more frames.
58. A radio network controller (136), adapted to support the method steps of any of preceding Claims 55 to 3 0 57.
59. A storage medium storing processor-implementable instructions for controlling a processor to carry out the method steps of any of Claims 55 to 57.
60. A communication system (100), for example a UMTS wireless communication system, adapted to facilitate the method steps of any of preceding Claims 55 to 57. 1 0
61. A method (200) for scheduling of data packets in a code division multiple access (CDMA) communication system (100) supporting both synchronous and asynchronous channels, the method characterized by the steps of: ascertaining whether one or more data packet(s) is/are to be transmitted on the synchronous and/or asynchronous channels (610); and building a single queue into which the one or more data packet(s) are arranged and served in priority 1 0 order.
62. A method (200) for scheduling of data packets according to Claim 61, wherein the CDMA communication system is a UMTS communication system and the synchronous 1 5 and asynchronous channels comprise one or more of a dedicated channel, a Forward Access Channel (FACH), a downlink shared channel (DSCH).
63. A radio network controller (136), adapted to 2 0 support the method steps of Claim 61 or Claim 62.
64. A storage medium storing processor-implementable instructions for controlling a processor to carry out the method steps of Claim 61 or Claim 62. 2 5
65. A communication system (100), for example a UMTS wireless communication system, adapted to facilitate the method steps of any of preceding Claim 61 or Claim 62.

3 0
66. A UMTS communication system substantially as hereinbefore described with reference to, and/or as illustrated by, FIG. 1 of the accompanying drawings.
67. A method of scheduling data packets for transmission substantially as hereinbefore described with reference to, and/or as illustrated by, FIG. 2 or FIG. 4 or FIG. 6 of the accompanying drawings.