A METHOD OF IMPROVED SCHEDULING FOR ENHANCED UPLINK OF UTRA FDD AT A BASE STATION OF A COMMUNICATION SYSTEM
This invention relates to a method of scheduling at a base station of a communication system, in particular for mobile communications. The present invention addresses the need to take account of urgency information in scheduling. Various methods for taking into account urgency information in scheduling are possible, but these all have certain disadvantages. If the node B predicts urgency by examining and comparing, for example, data rate requests against granted data rate, this leads to inaccuracies since, for example latency requirements on the information might not be known a priori. Another option is to incorporate priority information into the data rate request information. However this restricts scheduling in the case that the node B has additional capacity to allocate. Simply not incorporating any urgency information decreases the efficiency of the scheduling. In accordance with a first aspect of the present invention, a method of scheduling at a base station of a communication system comprises determining in a terminal a short term urgency indication from buffer occupancy, data type and quality of service; sending the short term urgency indication from the terminal to the base station, which base station is adapted to dynamically allocate resources between terminals; wherein the base station varies precedence for resource allocation from the terminal according to the short term urgency indication received. In the present invention, the terminal calculates the .urgency and sends an indication to the base station, taking account of buffer occupancy, data type and quality of service (QoS). A scheduler in the base station can maintain a quality of service (QoS) target with this additional signalling, indicating ho ^ urgently the terminal wishes to transmit its data, so that the scheduler can increase the precedence of a terminal for a specific transmission period without allocating a higher precedence channel on a permanent basis. This makes more efficient use of the available resources. Preferably, the terminal provides the urgency indication by dynamic physical layer signalling on a common, shared or dedicated physical channel.
Alternatively, the terminal provides the urgency information within a medium access control (MAC) header block that is transmitted to the base station Preferably, the resources allocated to a channel are modified for individual transmit time intervals according to the urgency indications and data rate requirements of all communicating terminals in the system. Preferably, the short term urgency indication is only signalled by the terminal if a change in urgency occurs. Alternatively, the short term urgency indication is signalled at regular intervals. This improves the efficiency of scheduling. Preferably, an average quality of service is negotiated between the terminal and the base station at the beginning of a transmission. Preferably, the urgency indication is transmitted at the same time as a predicted transport format combination indicator. Typically, the base station is one of a Node B or radio network controller (RNC). In accordance with a second aspect of the present invention, a terminal comprises one of a mobile phone, laptop, PC card or personal digital assistant, the terminal being adapted to carry out the method of the first aspect. An example of a method of scheduling at a base station of a communication system in accordance with the present invention will now be described with reference to the accompanying drawings in which:- Figure 1 illustrates a possible time line for the method of the present invention.
There is interest in producing an "enhanced uplink" channel for the FDD mode of 3GPP specifications. The purpose of the enhanced uplink (UL) is to increase the throughput and capacity and, where possible, also reduce packet call delay in the uplink. One of the prime features of the enhanced uplink is fast allocation of uplink "noise rise" at the base station or node B, otherwise known as node B scheduling. The mechanism of node B scheduling is that the node B signals to terminals or user equipment (UE), using fast physical layer signalling, restrictions on the maximum data rate at which the terminals may transmit in the uplink in order to restrict the amount of interference created by the terminals. Since the function is located at the node B and the signalling is carried out at the physical layer, the node B is able to track
rapid variations in the UE's transmit capability and buffer status, in order that it can restrict terminal data rates according to their instantaneous needs and capabilities and thus more effectively utilise the noise rise at it's receiver. To facilitate efficient scheduling, some additional information may be transmitted in the uplink including UE buffer status, UE transmitter (TX) power for the current transport format combination (TFC) relative to its maximum TX power and the status of other services the UE is supporting. Such information informs the scheduler of the data rate that the UE requires and is able to support. Commonly, the UE and network negotiate a channel precedence via high level signalling and having agreed a precedence, that precedence is fixed for the particular channel. Instead of this, the UE and network negotiate an average QoS and the method of the present invention aids the scheduler in maintaining this QoS target by additional signalling, indicating how urgently the UE wishes to transmit a particular block of data. Reasons for urgency may be that the UE transmit buffer is becoming full, .or that the data is low latency e.g. gaming. The urgency information is taken into account by the scheduler when allocating UL resources, so if the buffer is not full the UE does not to send an urgency indicator, but at a later transmit time interval, when further data has arrived in the buffer, it may become necessary to transmit the same data which previously was not urgent. A physical layer urgency indicator, calculated in the UE, preferably transmitted together with an associated predicted TFC indicator (TFCI), is signalled from the UE, so that the scheduler can take this into account in order to maintain a QoS target. This differs from the known technique of higher layer priority labelling of logical channels in several ways. In the present invention, the urgency indication is dynamic, i.e. is able to change between transmission intervals on a fast basis. It is signalled by physical layer signalling or within a medium access control
(MAC) header block. The scheduler is responsible for maintaining overall QoS targets and the scheduler has autonomy to vary precedence between channels and terminals according to the urgency indicators. There are a number of possible implementations of the present invention, for example, the urgency indicator can be provided as one or several bits on an "Enhanced UL" UL signalling channel, as one or several bits on another (non EUL) signalling channel, or as one or several bits embedded in a previous enhanced UL transmission. One example of a time line for the method of the present invention is given in
Fig.l. A UE 10 indicates 1 a supportable TFC or data rate, and current traffic priority, precedence or urgency indicator for the immediate future to a network traffic scheduler in a base station 11. The network scheduler 2 uses desires of competing UE's to decide resources and signals 3 the allocation to use to the UE 10. The UE prepares 4 current data into the block size signalled by the base station 11 , and calculates the desired capacity and precedence for the next time. The UE sends 5 uplink data and indicates 6 the desired capacity and current traffic precedence for the immediate future. This cycle of desire 6 and allocation 3 is repeated. The cycles may be back-to-back, or extended to interleave with traffic from other UE's depending on network management requirements. It is possible to streamline the system by having the user equipment only indicate changes in traffic precedence, for example raising the precedence for an overflow condition, where data to be sent was accumulating faster than it could, be transmitted, or reducing precedence once such a condition had passed. However, in rapidly fluctuating environments it may be easier in operation to send the same priority/ precedence/urgency information again, when change is more common than continuity, or if the data flow is burst-like. The examples are described with reference to the frequency division duplex (FDD) mode of 3rd generation project partnership (3GPP) specifications, but the invention is not limited to this, nor even to mobile phone applications. The same techniques although with different data formats, could be deployed as part of the traffic control layers of any data transfer system limited by capacity at a central node. In addition to 3GPP TDD systems or CDMA 2000 systems, or any enhancements thereof, are computers making repeated requests to a central file server in a network (either wired or wireless), or a satellite modem system. At the moment systems using the normal TCP/IP simply drop packets for which there is no capacity, and it is up to the sender to send them again if no acknowledgement is received, a slow and wasteful process.