WO2006110072A1

WO2006110072A1 - Mac header for enhanched uplink multiplexing

Info

Publication number: WO2006110072A1
Application number: PCT/SE2005/000551
Authority: WO
Inventors: Janne Johannes Peisa; Johan Torsner
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2005-04-15
Filing date: 2005-04-15
Publication date: 2006-10-19

Abstract

The present invention relates to a User Equipment and a Node B connectable to a UMTS terrestrial radio access network, UTRAN, comprising a medium access control, MAC, entity for multiplexing and demultiplexing logical uplink channels onto transport uplink channels. The MAC entity of the UE and the Node B comprise means for handling an enhanced MAC header, MAC-e header, comprising information of an identity of a logical channel, an identity of a MAC-d flow onto which the logical channel is mapped and an associated MAC Protocol Data Unit, PDU, size or RLC PDU size in order to be able to multiplex/demultiplex at least two logical channels onto a single transport channel.

Description

MAC HEADER FOR ENHANCHED UPLINK MULTIPLEXING

FIELD OF THE INVENTION

The present invention relates to a cellular telecommunication system, and in particular to an enhanced uplink structure in the radio network in a Universal Mobile Telecommunication System (UMTS) .

BACKGROUND OF THE INVENTION

The present invention relates to methods and arrangements for improving the uplink in a UMTS terrestrial radio access network (UTRAN) . The UTRAN is illustrated in figure 1 and comprises at least one Radio Network System 100 connected to a Core Network (CN) 200. The Core Network is connectable to other networks such as the Internet or other mobile networks, e.g. GSM systems and fixed telephony networks. The RNS 100 comprises at least one Radio Network Controller 110. Furthermore, the respective RNC 110 controls a plurality of Node-Bs 120,130 that are connected to the RNC by means of the Iub interface 140. Each Node B covers one or more cells and is arranged to serve User Equipments (UE) 300 within said cell. Finally, the User Equipments 300, also referred to as mobile terminals, are connected to one or more Node Bs over the radio interface 150.

Figure 2 shows the radio interface protocol architecture for the UTRAN. The physical layer interfaces the Medium Access Control (MAC) sub-layer of Layer 2, which interfaces the Radio Link Control (RLC) , a further sublayer of layer 2 (not shown) . RLC interfaces the Radio Resource Control (RRC) of Layer 3. The RRC controls the physical layer. The physical layer offers different transport channels to MAC where the transport channel is characterized by how the information is transferred over the radio interface, e.g. a certain transmission time interval or a certain type of forward error correcting coding (FEC) . In 3GPP release 5 a new transport channel has been introduced, the high-speed downlink shared channel (HS-DSCH) . For HS-DSCH there is only one transport channel per User Equipment. Different quality of service is instead provided by different MAC-d flows where each MAC-d flow can have different block error rate and different priority over the transport network. MAC offers different logical channels to RLC, wherein the logical channels are mapped on the MAC-d flows. A logical channel has its own RLC entity and is characterized by the type of information that is transferred on the logical channel .

Currently, there are discussions, e.g. in the document 3GPP TR 25.896 issued by the 3^rd Generation Partnership Project (3GPP) , to enhance the WCDMA (Wideband Code Division Multiplex Access) uplink with features to be applied on an enhanced dedicated transport channel (E-DCH) . The uplink features are for example shorter Transmission Time interval, (TTI), Hybrid Automatic Repeat Request (HARQ) and fast rate control.

The Transmission Time interval, (TTI) is the frame length used for transmission over the radio interface, i.e. even if a very small amount of data is to be transmitted it is anyway required to be interleaved over one TTI. By using a shorter TTI the delay in the system will therefore be reduced.

HARQ is a more advanced form of an ARQ retransmission scheme. In conventional ARQ schemes the receiver checks if a packet is received correctly. If it is not received correctly, the erroneous packet is discarded and a retransmission is requested. With HARQ the erroneous packet is not discarded. Instead the packet is kept and combined with a result of the retransmission. That implies that even if both the first transmission and the retransmission are erroneous, they may be combined to a correct packet. This means that fewer retransmissions are required.

Fast rate control (or fast scheduling) means that the Node B can indicate to each UE the rate the UE is allowed to transmit with. This can be done every TTI, i.e. in a fast way. Thus, the network is able to control the interference in the system very well.

When introducing a shorter TTI with HARQ, the WCDMA radio- interface needs to be partly re-designed. Also solutions for transport channel handling and multiplexing needs to be specified for E-DCH. In the existing release 5 of the WCDMA-' specification issued by 3GPP, a User Equipment can be configured with several transport channels where several logical channels can be multiplexed on each transport channel. When considering, as an example, the use case when two packet-switched (PS) radio access bearers (RAB) are configured, each PS-RAB is using a separate logical channel and typically the two logical channels are multiplexed on the same transport channel (they could also be mapped on two transport channels, e.g. if there is a need to support different Quality of Service (QoS) for the two RABs) . In the control plane, the signalling radio bearers 1-4 are typically multiplexed on the same transport channel, i.e. in total there are typically two or three transport channels for a UE (not including broadcast and paging Channels) .

It is possible to configure a separate priority for each logical channel. In the uplink, a Transport Format Combination (TFC) selection algorithm is specified by 3GPP using absolute priorities between the logical channels. This means that when a TFC is selected, the amount of data carried for the highest priority is maximised. Then, if the configured TFCs allow data with lower priority to be transmitted simultaneously, the algorithm maximises the amount of the second highest priority etc., i.e. after the data with the highest priority is maximised. Thus, it is only possible to transmit the data with the lower priority if there are available resources when the high priority data is transmitted. This rule implies that, if there is no possibility to transmit data with different priorities at the same time, i.e. in the same transmission time interval

(TTI), the lower priority data is blocked - or starved - when there is high priority data to transmit. The starving problem of low priority data depends inter alia on the burstiness of the traffic, i.e. the, burstier the transmission is the smaller is the problem.

Figure 3a describes the multiplexing according to 3GPP Release-5, which illustrates that it is possible to multiplex a number of logical channels on the same transport Channel and that it is possible to configure several transport channels for a particular User Equipment. The logical channels can be assigned different priorities and they can have different RLC PDU sizes. The use of several transport channels is beneficial in 3GPP Release-5 if different quality of service (QoS), e.g. block error rates, needs to be supported or if the logical channels have different RLC PDU sizes. However, a disadvantage with the approach of release 5 is that even if different RLC PDU sizes can be multiplexed on the same TrCH, they cannot be transmitted in the same TTI as discussed in the section above .

A transport channel structure and multiplexing functionality need hence to be specified for E-DCH. In the design of the channel structure, there are several issues to consider:

For the enhanced uplink dedicated channel (E-DCH) the use of several transport channels would be complex since the HARQ protocol would need to operate on a transport Channel level. Further, the need to support several transport channels for E-DCH is smaller than for earlier 3GPP releases of WCDMA.

The solution to the starving of low priority data mentioned in the previous section is handled in an earlier release (Release-99) of the WCDMA-specification by configuring the TFCs in such a way that data always can be transmitted simultaneously for all transport channels. This means that data from a transport channel with high priority does not block data from a transport Channel with low priority. This is however highly inefficient since it also means that only parts of the physical channel data rate can be used for a particular transport channel, even if there is no data for other transport channels. This is illustrated by the following example: Supposing there are two transport channels mapped on one physical channel, and the physical channel can support 100 kbps and further supposing that the TFCs are configured as in the release 99 reference bearers specified by 3GPP. Then it is required to fix the rate available for each of the transport channels, e.g. 50 kbps for TrCHl (transport channel 1) and 50 kbps for TrCh2

(transport channel 2), or any other split. If there is no data to transmit for TrCH2 it is not possible to transmit more than 50 kbps for TrCHl. With a more flexible method, TrCHl could transmit 100 kbps in this case. Thus it is desirable to avoid the fix rate of the transport channel for the E-DCH. SUMMARY OF THE INVENTION .

The objective problem of the present invention is to achieve an uplink channel structure that provides an efficient MAC- multiplexing.

The objective problem is solved by a User Equipment comprising a medium access control entity for multiplexing logical uplink channels, potentially mapped on different MAC-d flows, onto transport uplink channels wherein the MAC entity of the UE comprises means for transmitting an enhanced MAC header (MAC-e header) . The MAC-e header comprising information of an identity of a logical channel, an identity of a MAC-d flow onto which the logical channel is mapped and an associated Protocol Data Unit, PDU, size for the logical channel in order to be able to multiplex at least two logical channels potentially mapped on two different MAC-d flows onto a single transport channel makes it possible to achieve an uplink channel structure that provides an efficient MAC-multiplexing.

The objective problem is also solved by a Node B comprising a medium access control entity for demultiplexing logical uplink channels and MAC-d flows from transport uplink channels, wherein the MAC entity of the Node-B comprises means for receiving an enhanced MAC header (MAC-e header) . The MAC-e header comprising information of an identity of a logical channel, an identity of a MAC-d flow onto the logical channel is mapped and an associated Protocol Data Unit size for the logical channel in order to be able to demultiplex at least two logical channels and MAC-d flows from a single transport channel makes it possible to achieve an uplink channel structure that provides an efficient MAC- multiplexing. An advantage with the present invention is that starving/blocking of low priority data may be avoided. According to the present invention, it is in principle possible for data from a given logical channel to use the whole available data rate when there is no data from other logical channels available for transmission. When data from several logical channels are available for transmission simultaneously, the amount of data from each logical channel can be restricted, e.g. by a UTRAN configuration, so that the high priority data is only allowed to use a part of the available data rate. In this way, it can be assured that data with lower priority always can be transmitted. Alternatively, this restriction can be made on MAC-d flow/transport channel level.

A further advantage with present invention is that it also provides means for multiplexing data_^ with different RLC PDU sizes and from different logical channels and MAC-d flows in the same TTI, which is beneficial from an efficiency perspective .

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows a UMTS terrestrial radio network (UTRAN) wherein the present invention may be implemented.

Figure 2 shows the radio interface protocol architecture for the UTRAN.

Figure 3a shows the MAC-d multiplexing according to the prior art while figure 3b shows an overview of the corresponding MAC-e multiplexing according to the present invention. Figure 4 shows a MAC-e header format according to the present invention.

Figure 5 shows the Transport channel structure for E-DCH according to the present invention.

DESCRIPTION OF THE INVENTION

According to the present invention, it is possible to use a single E-DCH transport channel by multiplexing several logical channels belonging to the same UE onto the single E- DCH transport channel via one or more MAC-d flows on the MAC layer. This simplifies the channel structure since the transport channel dependent rate matching on E-DCH is avoided and the HARQ-protocol is only required to consider the single transport '/'channel as shown in figure 5. An enhanced MAC header, i .le. ^a-new ^■;MAC-e header, is introduced which makes the MAC multiplexing, onto the single E-DCH possible. The new MAC-e header comprises information of an identity of a logical channel, an identity of a MAC-d flow onto which the logical channel is mapped and an associated MAC Protocol Data Unit (PDU) size or RLC PDU size for the logical channel .

In a preferred embodiment according to the present invention, the information comprises a header value indicative of the logical channel identity, MAC-d flow identity and the associated MAC-d PDU size or RLC PDU size.

In the example shown in figure 3b all the signalling radio bearers (SRB) are mapped on the E-DCH. In practice, all the five signalling radio bearers may also be mapped on the normal DCH but that does not change the issues discussed in the following. The difference between the proposed MAC-e multiplexing and the known MAC-d multiplexing is that all logical channels for a UE are multiplexed together on MAC-e, i.e. the MAC-e header is unique per UE, in contrast to the MAC-d header which is unique per transport channel.

According to the preferred embodiment of the present invention, for each used MAC-e header value a logical channel identification (ID), a MAC-d flow identity, and the associated MAC-d PDU or RLC size are mapped. This means the receiver of the Node B is able to identify the logical channel, MAC-d flow and the MAC-d PDU or RLC size from the received MAC-e header value. This implies that the Transport Format Combination Identifier (TFCI) only needs to indicate the transport block size. In Rel-99 the TFCI indicates the transport block size and the number of transport blocks in a TTI. That is required in order to make the receiver able to decode the data since . the data on different transport channels can be .coded differently. For the E-DCH, there is only one transport channel, and the TFCI does not need to be used for indicating the 'amount of data for each transport channel .

Logically, several MAC-d flows, on which the logical channels are mapped, still can be supported, wherein all

MAC-d flows are mapped onto the same transport channel.

Since both the logical channel identity and the MAC-d flow identity are extractable from the MAC-e header value it is possible to logically maintain the same mapping of logical channels to MAC-d flows that is used for a normal DCH.

The MAC-e PDU format is shown in figure 4. The MAC-e header value Hl is mapped to logical channel 1 and MAC-d flow 1 while the header value H2 is mapped to logical channel 2 and MAC-d flow 2. The logical channels 1 and 2 have different PDU sizes. Header value H3 is also mapped to logical Channel 2 and MAC-d flow 2 but with a different PDU size. The header value HO is a special reserved header value that indicates that the rest of the MAC-e PDU is padding according to one embodiment of the invention. As shown in the example it is possible to have several PDU sizes for the same logical channel by allocating a separate MAC-e header value to each PDU size for the logical channel. The PDU size is hence part of the information that can be extracted from the header value. In the prior art, the C/T field only indicates the logical channel, i.e. the there is no size information. The MAC-d PDU size in one TTI is according to prior art instead indicated by the TFCI, which implies that it is difficult to have different MAC-d PDU sizes in one TTI.

Thus, the proposed solution makes it possible to multiplex a plurality of logical channels potentially mapped on different MAC-d flows, wherein the logical channels may have different MAC-d PDU size, in the same TTI, which is beneficial for .performance reasons.

The MAC-e header, may be similar to the C/T field in the. MAC- d header in Releas_eτ5_{. ,} of. the WCDMA-specification, except that all logical channels are multiplexed together on the same E-DCH transport Channel, even if they logically belong to separate MAC-d flows. The MAC-e header format may be the same as the MAC-d C/T field in said Relase-5, i.e. 4 bits allowing for a total of 16 logical Channels on E-DCH. However, the MAC-e header format may also be different, e.g. by extending the number of bits to allow for more logical channels. Preferably, the MAC-e header comprises a value of 6 bits. The value of 6 bits is adapted to be mapped to the logical channel identity, MAC-d flow identity and the PDU size of the identified logical channel. Thus, a number of consecutive PDUs may have the same value. The MAC-e header is only needed when multiplexing is performed, i.e. when there is more than one logical channel to be mapped onto the E-DCH. In other cases the MAC-e header is not required.

With the present invention, it is not necessary to have a transport block (TB) size defined for all possible payload sizes. If the payload does not fit exactly into the TB, padding is present in the end of the MAC-e PDU. There are two ways that may be implemented in the arrangements according to the embodiments of the present invention to indicate that the padding starts:

1) A special reserved value of the MAC-e header is used, which indicates that the remaining of the MAC-e PDU consists of padding. This alternative is discussed above in conjunction with figure 4.

2) A rule is defined which states that the amount of padding is always smaller than the smallest configured MAC-d PDU size. Thus, if the remaining number of bits in the MAC-e PDU is less than the smallest MAC-d PDU size, the bits contain padding. If there is room for at least one MAC-d PDU_. of the smallest configured MAC-d PDU size, the bits contains data.

The enhanced MAC header is implemented in a UE according to , the present invention.^' Thus, the UE comprises a MAC entity for multiplexing logical channels onto transport channels. The MAC entity comprises means for transmitting an enhanced MAC header comprising information of an identity of a logical channel, an identity of a MAC-d flow onto which the logical channel is mapped and an associated PDU size for the logical channel in order to be able to multiplex at least two logical channels onto a single transport channel.

The enhanced MAC header is implemented in a Node B according to the present invention. Thus, the Node B comprises a MAC entity for de-multiplexing logical channels from transport channels. The MAC entity comprises means for receiving an enhanced MAC header comprising information of an identity of a logical channel, an identity of a MAC-d flow onto which the logical channel is mapped and an associated PDU size for the logical channel in order to be able to de-multiplex at least two logical channels from a single transport channel. In the_. drawings and specification, there have been disclosed typical preferred embodiments of the invention and, although specific terms are employed, they are used in a generic and descriptive sense only and not for purposes of limitation, the scope of the invention being set forth in the following claims .

Claims

1. A User Equipment, UE, (300) connectable to a UMTS terrestrial radio access network, UTRAN, comprising a medium access control, MAC, entity for multiplexing logical uplink channels onto transport uplink channels, charcaterised in that the MAC entity of the User Equipment comprises means for transmitting an enhanced MAC header, MAC-e header, said header comprising information of an identity of a logical channel, an identity of a MAC-d flow onto which the logical channel is mapped and an associated MAC-d or RLC Protocol Data Unit, PDU, size for the logical channel in order to be able to multiplex at least two logical channels onto a single transport channel .

2. The User Equipment (300) according to claim 1, characterised in that the MAC-e header comprises < a..-'-header value indicative of the logical channel identity, MAC-d flow identity, and the associated MAC-d or RLC PDU size.

3. The User Equipment (300) according to claim 1 or 2, characterised in that the enhanced MAC-e header is adapted to comprise a reserved value indicative of that the remaining of the MAC PDU consists of padding.

4. The User Equipment (300) according to claim 1 or 2, characterised in that the amount of padding in one MAC PDU is always smaller than the smallest configured MAC-d PDU size in order to identify the padding.

5. A Node B (120,130) connectable to a UMTS terrestrial radio access network, UTRAN, comprising a medium access control, MAC, entity for demultiplexing logical uplink channels from transport uplink channels, charcaterised in that the MAC entity of the Node-B comprises means for receiving an enhanced MAC header, MAC-e header, comprising information of an identity of a logical channel, an identity of a MAC-d flow onto the logical channel is mapped and an associated MAC-d or RLC Protocol Data Unit, PDU, size in order to be able to demultiplex at least two logical channels from a single transport channel.

6. The Node B (120,130) according to claim 5, characterised in that the mac-e header comprises a header value indicative of the logical channel identity, MAC-d flow identity and the associated MAC-d or RLC PDU size.

7. The Node B (120,130) according to claim 5 or 6, characterised in that the enhanced MAC header is adapted to comprise a reserved value indicative of that the remaining of the MAC PDU consists of padding.

8. The Node B (120,130) according to claim 5 or 6, characterised in that the amount of padding in one MAC PDU is always smaller than the smallest configured MAC-d PDU size in order to identify the padding.