WO2004049648A2

WO2004049648A2 - Channel configuration and timing for enhanced uplink dch

Info

Publication number: WO2004049648A2
Application number: PCT/EP2003/013260
Authority: WO
Inventors: Martin DÖTTLING; Bernhard Raaf
Original assignee: Siemens AG; Siemens Corp
Current assignee: Siemens AG; Siemens Corp
Priority date: 2002-11-25
Filing date: 2003-11-25
Publication date: 2004-06-10
Anticipated expiration: 2005-05-25
Also published as: EP1566027A2; AU2003302343A1; AU2003302343A8; WO2004049648A3

Abstract

A method for transmitting information between a UE and a Node B comprising the steps of establishing a HS-DPCCH channel between the UE and the Node B, establishing a EU-DCH channel between the UE and the Node B, aligning the HS-DPCCH channel and the EU-DCH in time in the uplink per UE using a timing such that EU-DCH starts at the beginning of a CQI field or a ACK/NACK field of the HS-DPCCH channel.

Description

Channel Configuration and Timing for Enhanced Uplink DCH

The invention relates to channel configuration and timing for Enhanced Uplink DCH (EU-DCH) , in particular to a method for transmitting information between a UE (mobile station) and a Node B (base station or base station system) and a radio set (mobile station or base station) .

In 3GPP TSG RAN G1 a feasibility study on enhanced uplink for UTRA FDD has started. Since the use of IP based services becomes more important there is an increasing demand to improve the coverage and throughput as well as to reduce the delay of the uplink. Applications that could benefit from an enhanced uplink may include services like video-clips, multimedia, e-mail, telematics, gaming, video-streaming etc. The general idea of this study is to investigate performance enhancement techniques, like adaptive modulation and coding, HARQ and fast scheduling in combinations with shorter frame sizes. For simplicity the enhanced uplink dedicated data channel will be denoted as EU-DCH (Enhanced Uplink Dedicated Channel) . Several problems arise if a new high-rate uplink channel suited for packet-oriented services is introduced in UMTS:

• To ensure efficient scheduling, latency in the system must be minimised. Especially, the delay between downlink signalling and corresponding uplink data transmission needs to be minimised. Similarly a short round-trip delay must be ensured, i.e., also the uplink control information required for Node-B based scheduling needs to be sent with minimum delay • The uplink tends to be interference-limited and therefore precautions need to be taken that would allow the Node B to control and manage the interference created by a particular UE . This applies especially to channels which are likely to cause bursty interference. In a further step, interference management among different UEs needs to be considered.

• It must be borne in mind that total transmission power at UE is a limited and precious resource.

• It can be envisaged that EU-DCH will be used together with HSDPA and UE complexity as well as required processing power needs to be kept as low as possible.

• The EU-DCH should be able to accommodate different types of data traffic (long versus short packet calls) and different degrees of mobility (i.e. varying channel coherence times) .

• The EU-DCH data channel should be implemented in a backward-compatible evolutionary way and should reuse existing UMTS features wherever possible.

Since the study is only in its initial phase, no complete solution of the aforementioned problems within the UMTS context has yet been presented. The following single elements of a possible EU-DCH have been discussed [Mot 1, Eril, Nokl] : • usage of shorter frame sizes, in particular 2 ms as in HSDPA, to reduce delay, • Node-B based scheduling of the EU-DCH, • re-use of HSDPA control channels for EU-DCH using additional frame formats.

Motorola [Mot2] proposes to use uplink and downlink control channels with 10 ms frame size, i.e. scheduling is basically done by sending a map of resource grants for 5 consecutive 3- slot TTIs of data within one control message. A 3-slot (2 ms) format is used for the downlink ACK/NACK control channel. The timing of EU-DCH (and their relative timing to HSDPA channels) is not addressed in this paper.

Nokia [Nok2] discusses a method, which allows to operate EU- DCH with very low physical layer signalling. The UE issues a rate request (RR) message and the corresponding answer of the Node B is a rate grant (RG) message. A figure in this contribution implicitly shows that the EU-DCH data channels of different UEs are time-aligned. The critical issue of the time-alignment of channels per UE and the timing of all channels of different UEs involved, however, is not investigated.

In summary, there has been no discussion of a complete configuration and timing of the different channels required for EU-DCH that addresses all issues mentioned previously.

It is an object of the invention to provide a solution to the EU-DCH physical channel configuration that addresses one or more of the aforementioned problems and constraints.

The object of the invention will be achieved with a method for transmitting information between a UE and a node B, which is defined by what is disclosed in the appended independent claims. Advantageous embodiments of the present invention will be presented in the dependent claims. Radio sets and further developments of the radio sets corresponding to the method claims also lie within the scope of the invention

Thus the invention is based on the following thoughts, which are in no way limiting for the scope of the invention:

Time alignment of channels: In order to enable efficient interference management at Node B, the timing of all uplink channels which create bursty interference should be aligned. To allow this interference management and additionally to support power management at UE the uplink control channel of HSDPA (HS-DPCCH) and the data channel of the enhanced uplink (EU-DCH) shall be time aligned. These are the most important channels with bursty traffic, i.e., channels that cause bursty interference. Node B can manage the expected interference in the forthcoming TTIs and slots since it has knowledge on the time when it expects CQI, ACK/NACK messages, and data on EU-DCH. Based on this expected noise rise, Node B can control the received interference by appropriate scheduling and the time alignment of EU-DCH and HS-DPCCH ensures maximum efficiency and minimum complexity of this process. Fig. 1 gives an overview over the discussed data and control channels for EU-DCH and HSDPA.

Alternatively, an alignment with Rel . 99 UL DPCH would have been possible. However, this is a channel which causes less bursty interference, since it is power controlled and does not as frequently use DTX (it basically uses varying SF on a slower per-fra e basis) . Therefore it is preferred to use time alignment with HS-DPCCH. The aforementioned alignment of EU-DCH and HS-DPCCH channels allows power and interference management per UE . From a network perspective it could be envisaged that also such a management is possible between all active UEs, i.e. the network has the capability of shifting the EU-DCH and HS- DPCCH of each UE in a way that the reception of this data at Node B occur on a 3-slot grid for all UEs. In principle, this could be achieved by the timing parameters T_n and/or ιcι as defined in [25.211], which adjust the start of the DCH and HS-DPCCH with respect to the frame border. However, this would lead also to an alignment of the DCH of all users, and in particular, to an alignment of the TFC, TPC and pilot bits, which are transmitted with high power, which may well not be desirable. An alternative solution would be to introduce a new timing adjustment parameter (denoted as T_e) with dynamic range of 3 slots. T_e could either adjust the timing of both, HS-DPCCH and EU-DCH or only of EU-DCH. The second solution, however, would trade the alignment of HS- DPCCH and EU-DCH per UE with the alignment of all EU-DCHs to a 3-slot grid. Note, that a general drawback of this additional timing offset T_e is that it increases total round trip delay. Therefore another ^'solution would be not to use active per-TTI interference management but semi-static interference averaging by an assignment strategy for T_n that jointly optimises power and interference for DCH and EU-DCH.

For implementation of an EU-DCH, control information, like request for uplink resources, UE buffer status, transport format of the data, or information on available transmission power resources at UE must be sent from UE to the network. This uplink control information for EU-DCH (denoted as EU- DPCCH in this text) can either be a separate DPCCH or be multiplexed into HS-DPCCH by reducing the spreading factor and using blind format detection at Node B. Note that the separate DPCCH results in higher peak-to-average-power ratio due to multi-code transmission, while multiplexing with HS- DPCCH might reduce CQI decoding performance at Node B.

In any case, the uplink control information for EU-DCH (denoted as EU-DPCCH in this text) shall be time aligned with the HS-DPCCH field that on average required the smallest transmit power, e.g. the CQI field of HS-DPCCH. For UE power management it is favourable not to send the EU-DPCCH in parallel to the ACK/NACK (A/N) of HS-DPCCH, since the latter will use higher transmission power than the CQI message. Additionally, time alignment with the CQI slots allows a favourable split between processing time available at UE and at Node B, as will be detailed in the following. As a result, a 2-slot format for EU-DPCCH should be adopted.

Timing of Enhanced Uplink Channels and Available Processing Time

The processing time for EU-DCH available at the UE TUEP,_EU-_{DCH /} i.e., the time interval from end of reception of the downlink control channel, denoted as EU-SCCH in the sequel, until beginning of EU-DCH transmission, is required for EU-SCCH decoding and EU-DCH encoding. Without the constraint that HS-

SCCH timing shall be reused for EU-SCCH, this time is a free parameter and can be adjusted such that it is equal to the actually required processing time in the UE tuεp,re :

TuEP,EU-DCH - tuEP.req • ( 9 ) This constant time delay between downlink control channel and uplink data channel is denoted as fixed timing in the following. Furthermore, we require that EU-DCH and HS-DPCCH are time aligned in a way that the beginning of e EU-DCH TTI starts at the beginning of a ACK/NACK field or a CQI field of

HS-DPCCH. By this, the start of a EU-SCCH transmission is implicitly defined, simply as TUEP,EU-DCH before the begin of the corresponding EU-DCH. These timing will result in an available processing time at

Node B T_NP,Eu-DCH (i.e. time between reception of EU-DCH and start of corresponding EU-SCCH transmission)

^TNP, EU-DCH ~ TRT-V EU-SCH + ^TUEP,EU-DCH ^+t EU-DCH + ² " * prop ) r ( 1 0 ) where T_RT is the total round-trip delay (a system design variable) , t_Eu-sc_H is the duration of the EU-SCCH sub-frame, tEu-DCH is the duration of the EU-DCH TTI, and t_prθp is the propagation time between Node B and UE. This time is available for EU-DCH decoding, scheduling decision and EU-SCCH. In practical systems the following parameter choices are preferred: t_EU-scH ≤ t_EU-DCH and T_RT = p ^■ t_Eu-DCH . i.e. the sub-frame length of the downlink control channel corresponds to the TTI length and the total round-trip delay is a multiple thereof, i.e. p is an integer.

For power and interference management reason EU-DPCCH is time aligned with a CQI field of HS-DPCCH. Therefore the time interval between the end of reception of EU-SCCH and the start of EU-DPCCH transmission TUEP,EU-DPCCH can be set to:

Tu_EP, _EU-_DPCC_H ⁼ AEP, _EU-_DCH + 2⁽3 ^• t_s!ot + D - t_EU__DCH , ( 1 1 ) where b is an integer and a = 0 if EU-DCH is time-aligned with the CQI field of HS-DPCCH and a = 1 if EU-DCH is time- aligned with the ACK/NACK field of HS-DPCCH . t_βiot s the duration of one slot . The processing time for decoding EU- DPCCH and using the information in the scheduling decision TNP, Eu-DPccH becomes : ^WP, EU-DP_CCH ⁼ TNP_<EU-DPCCH ^{+ t} _si_ot ~ 2a - t_slot -b - t_EU__DCH . ( 12 ) The above mentioned parameters should be chosen in a way that provides sufficient processing time at the network and UE and at the same time minimises round trip delay and latency.

Different lengths of EU-SCCH sub-frame and EU-DCH TTI:

If the downlink signalling payload is low t_Eu-scH < t_Eu-DCH m y be used and control information for more UEs can be accommodated in the shared channel. For example if t_Eu- _DCH - 3 slots and t_Eu-scH = 1 slot, three different UEs can receive control information in time multiplex. This would be beneficial to save downlink code resources. However, to keep EU-DCH and HS-DPCCH time aligned different values of T_UEP,_EU-_DCH need to be used for UEs using the first, the second and the third slot of a 3-slot sub-frame (denoted as variable timing in the following) . Note, that the different values of T_UEP._EU- _DCH need not to be signalled to UE explicitly but can be determined using the slot number of the start of the EU-SCCH sub-frame. For example the following equation can be used:

TU_EP,EU-_DC_H ⁼ Tu_EP,_EU-_DC_H,0 + TTI ~ ψs_lot ~~θdN_m ))-t_slot , ( 13 ) where TUEP,EU-DCH,O is the basic UE processing time, _sι₀t e {1, 2, ... , 15} the slot number of the end of the EU-SCCH sub- frame, and N_TTι is the number of slots per TTI used for EU-DCH transmission. At the Node B, the processing time becomes

TNP,EU-D_CH ⁼ AP,EU-D_CH,_O + (ψs t ^~ l)mod N_TTI )-t_sht , ( 14 ) where n_slot denotes the slot number of the begin of the EU-SCCH sub-frame .

Shorter control information sub-frames will increase the average available processing time for the same round-trip delay or equivalently allows shorter round-trip delay for identical processing time available. However, the worst-case processing time at the UE and Node B remains unchanged. In summary, for small downlink signalling payload shorter con- trol messages allow to save downlink code consumption. The reduced downlink code consumption in turn alleviates the need for fast switching between connection states (e.g. between

CELL_FACH and CELL_DCH) . Therefore this is considered an interesting option. Note, that in principle, the same procedure (determination of T_UEP,_EU-DCH based on the relative slot number of EU-SCCH) can be applied to cases where t_Eu-scH

The basic UE processing time TUEP,EU-DCH,O can be easily derived based on the considerations and equations in the previous sections. The use of a separate downlink control channel can result in low overall decoding processes required in the UE.

Enhanced EU-DCH formats

In general the adoption of a 3-slot TTI for EU-DCH is advantageous since it provides low delay, reasonable packet sizes and allows efficient reuse of HSDPA resources. Nevertheless, in the uplink, code resources are not as scarce as in the downlink and the EU-DCH is not a shared channel. Furthermore, services that inherently produce longer packet calls should also be accommodated efficiently.

Therefore a multi-TTI transmission can be used which comprises L TTIs, where L is a parameter that the Node B assigns depending on various criteria, like data backlog, HARQ process states, QoS parameters, channel coherence time and interference management considerations. A multi-TTI transmission consists of L individual packets. The major advantage of the multi-TTI transmission is that only one EU- SCCH control information is required for L packets and scarce downlink code resources are saved. Furthermore the UE needs not to monitor EU-SCCH during the duration of the multi-TTI transmission. Multi-TTI transmission allow a simple adaptation to the traffic type and to channel coherence time while maintaining the simplicity of a N-channel Stop-And-Wait HARQ protocol. The parameter L must be signalled in the resource grant message from the Node B.

Different realisations of multi-TTI transmissions are possible. Either each packet can be sent with its individual format and no resources on the uplink control channel EU- DPCCH can be saved. Alternatively, a multi-TTI uplink control format can be introduced, which is applied if all packets use the same transport format. This is especially useful for UEs that have slow or small variations in the channel conditions. For such a multi-TTI uplink control format various alternatives are possible:

• new data indicator and the HARQ process ID are sent for each packet individually: highest flexibility at the expense of uplink control signalling load, • common new data indicator: applies to a multi-TTI transmission consisting entirely of initial transmissions (or re-transmissions) , saves signalling but has restrictions in applicability,

• sequential HARQ processes and signalling only of start process ID: useful as long as no data in the HARQ processes has considerably higher waiting time.

Again it should be noted that these thoughts described above are in no way limiting for the scope of the invention. In fact the invention encloses the following elements and any combinations of these elements:

• time alignment of HS-DPCCH channel and EU-DCH in the uplink per UE using a timing such that EU-DCH starts at the beginning of a CQI field or a ACK/NACK field of the

HS-DPCCH channel,

• for a sub-frame length of EU-SCCH equal to the TTI length of EU-DCH using a fixed timing and adjusting the relative timing of EU-SCCH accordingly,

• for a sub-frame length of EU-SCCH not equal to the TTI of EU-DCH using a variable timing that the UE can calculate based on the slot number of the EU-SCCH without explicit signalling, • using identical number of slots and time alignment of the uplink control information of EU-DCH with the HS- DPCCH field that requires the smallest average transmit power for efficient power management at UE,

• the adoption of a 2-slot format for the uplink control information of EU-DCH and time alignment of the uplink control information of EU-DCH (EU-DPCCH) with the CQI field of HS-DPCCH,

• using a timing such that the EU-DPCCH starts at

(2a +3b)- t_sio_t after the beginning of the corresponding EU-DCH transmission, where b is an integer and a = 0 if EU-DCH is time-aligned with the CQI field of HS-DPCCH and a = 1 if EU-DCH is time-aligned with the ACK/NACK field of HS- DPCCH, • time alignment of HS-DPCCH control channel and EU-DCH between different UEs for effective uplink interference management at Node B, • the definition of a multi-TTI transmission of L TTIs, which are granted to one UE using one single downlink resource grant message,

• the signalling of the number of blocks L in a multi- TTI transmission in the downlink control message, • the definition of a multi-TTI uplink control information format, which allows to reduce uplink signalling by communicating control parameters identical to all packets of a multi-TTI transmission only once for L TTIs.

Other objects and features of the present invention will become apparent from the following detailed descriptions considered in conjunction with the accompanying drawings. It is to be understood, however, that the drawings are designed solely for the purposes of illustration and not as a definition of the limits of the invention, for which reference should be made to the appended claims .

Fig 1 Overview over data and control channels of EU-DCH and HSPDA,

Fig 2 Exemplary channel configuration and timing diagram,

Fig 3 Exemplary channel configuration and timing diagram for

In Fig. 2 the channel configuration (time alignment of EU-DCH and HS-DPCCH) and timing is exemplified for one particular parameter setting, that provides reasonable processing time at both, Node B and UE, and at the same time results in equal round-trip delay T_RT for HSDPA and EU-DCH transmission. A separate downlink control channel (EU-SCCH) is used. In this case, the relative timing of EU-SCCH is simply determined by TU_EP,Eu-DCH _r which in turn is chosen simply as the required processing time at the UE . In this example, the EU-DPCCH is sent in parallel to the first two slots of EU-DCH (and of course time aligned with a CQI field of HS-DPCCH) . Such a timing can be useful if the transport format of EU-DCH is (at least partly) determined by the UE . Note, that also for

Rel. 99 DCH control information and data are sent in parallel .

Fig. 3 shows the same implementation as depicted in Fig. 2. The only difference is that t_Eu-scH < t_Eu-DCH . i.e., a 1-slot EU-SCCH format is used and three different UEs can receive EU-SCCH within 3 slots. One additional slot processing time is now available at both, Node B and UE, if the "middle" EU- SCCH slot is used (orange coloured slot of EU-SCCH) . The grey slots of EU-SCCH and the dotted lines correspond to the timing of Fig. 2. For the first slot, only the UE would benefit from additional processing time (2 slots) , while for the last slot only Node B would have two additional slots of processing time. As already stated above, the worst-case processing time remains unchanged. However, more UEs can share one EU-SCCH channel in time multiplex. Abbreviations :

3GPP Third Generation Partnership Project

CDMA Code Division Multiple Access

DL Downlink

FDD Frequency Division Duplex

IP Internet Protocol

QoS Quality of Service

UL Uplink

UMTS Universal Mobile Telecommunications

System

UTRAN UMTS Terrestrial Radio Access Network

HSDPA High Speed Downlink Packet Acces s HS-DSCH High Speed Downlink Shared Channel (HSDPA data channel) HS-SCCH High Speed Shared Control Channel (HSDPA downlink control channel) HS-DPCCH High Speed Dedicated Physical Control Channel

(HSDPA uplink control channel) EU-DCH Enhanced Uplink Dedicated Channel (enhanced uplink data channel) EU-SCCH Enhanced Uplink Shared Control Channel (EU-DCH downlink control channel)

EU-DPCCH Enhanced Uplink Dedicated Physical Control Channel

(EU-DCH uplink control channel UE User Equipment TTI Transmission Time Interval

References

[Motl] 3GPP TSG RAN WG 1 Tdoc Rl-02-1250, Motorola,

"Uplink enhancements for dedicated transport channels," Espoo, Finland, October 2002,

[Mot2] 3GPP TSG RAN WG 1 Tdoc Rl-02-1350, Motorola,

"Design Considerations for Enhanced Uplink Dedicated

Channel," Shanghei, China, November 2002, [Eril] 3GPP TSG RAN WG 1 Tdoc Rl-02-1225, Ericsson, "Techniques for Uplink Enhancements for Dedicated

Transport Channels," Espoo, Finland, October 2002, [Nokl] 3GPP TSG RAN WG 1 Tdoc Rl-02-1219, Nokia, "Issues to be studied for Enhanced Uplink DCH, " Espoo, Finland,

October 2002, [Nok2] 3GPP TSG RAN WG 1 Tdoc Rl-02-1219, Nokia, "Two

Threshold Node B Packet Scheduling, " Shanghai, China,

November 2002, [Nok3] 3GPP TSG RAN WG 1 Tdoc Rl-02-0018, Nokia, "Compact signalling of multi-code allocation for HSDPA, version

2," Espoo, Finland, January 2002,

[25.211] 3GPP Technical Specification TS 25.211, Physical channels and mapping of transport channels onto physical channels (FDD) (Release 5) ,

Claims

1. A method for transmitting information between a UE and a Node B comprising the steps of: • establishing a HS-DPCCH channel between the UE and the Node B

• establishing an enhanced uplink data channel between the UE and the Node B

• aligning the HS-DPCCH channel and the enhanced uplink data channel in time in the uplink using a timing such that the enhanced uplink data channel starts at the beginning of a CQI field or a ACK/NACK field of the HS- DPCCH channel.

2. The method according to claim 1

• wherein the sub-frame length of a downlink control channel for enhanced uplink is equal to the TTI length of a enhanced uplink data channel

• wherein a fixed timing between downlink control channel and uplink data channel is used and

• wherein the relative timing of the downlink control channel for enhanced uplink is adjusted accordingly .

• 3. The method according to claim 1 • wherein the sub-frame length of a downlink control channel for enhanced uplink is not equal to the TTI of a enhanced uplink data channel and

• wherein a variable timing between downlink control channel and uplink data channel is used that is calculated by the UE based on the slot number of the downlink control channel for enhanced uplink without explicit signalling.

4. A method for transmitting information between a UE and a Node B comprising the steps of:

• establishing a HS-DPCCH channel between the UE and the Node B

• establishing an enhanced uplink data channel between the UE and the Node B

• using identical number of slots and

• aligning the uplink control information of EU-DCH with the HS-DPCCH field that requires the smallest average transmit power for efficient power management at UE in time.

5. The method according to any of the preceeding claims, wherein

• a 2-slot format is adopted for the uplink control information of enhanced uplink and the uplink control information of enhanced uplink is time aligned with the CQI field of HS-DPCCH.

6. The method according to the preceeding claim, which comprises the step of

• using a timing such that the uplink control information for enhanced uplink starts at (2a +3b)- t_siot after the beginning of the corresponding enhanced uplink data transmission, where b is an integer and a = 0 if enhanced uplink data channel is time-aligned with the CQI field of HS-DPCCH and a = 1 if enhanced uplink data channel is time-aligned with the ACK/NACK field of HS-DPCCH.

7. A method for transmitting information between a UE and a

Node B comprising the step of:

• aligning HS-DPCCH control channel with EU-DCH in time between different UEs for effective uplink interference management at Node B .

8. A method for transmitting information between a UE and a Node B comprising the step of:

• defining a multi-TTI transmission of L TTIs, which are granted to one UE using one single downlink resource grant message.

9. The method according to the preceeding claim, which comprises • the signalling of the number of blocks L in a multi- TTI transmission in the downlink control message.

10. The method according claim 8 or 9, which comprises

• the definition of a multi-TTI uplink control information format, which allows to reduce uplink signalling by communicating control parameters identical to all packets of a multi-TTI transmission only once for L TTIs.

11. A radio set comprising processing means, that are arranged to perform the steps of any of the preceeding claims.