GB2516941A

GB2516941A - Scheduling Resources at a Relay Station in a Mobile Communications Network

Info

Publication number: GB2516941A
Application number: GB1314174.2A
Authority: GB
Inventors: Mahmoud Hadef; Jim O'reilly; Kwang Taik Kim; Kyunghun Jang
Original assignee: Samsung Electronics Co Ltd
Current assignee: Samsung Electronics Co Ltd
Priority date: 2013-08-07
Filing date: 2013-08-07
Publication date: 2015-02-11
Anticipated expiration: 2033-08-07
Also published as: GB201314174D0; KR20150017652A; GB2516941B

Abstract

A relay station 121-124 and a method of performing scheduling at a relay station, in a mobile communications network 100, are disclosed. Scheduling information is received, defining scheduling parameters for downlink transmission of data stored in a buffer, and channel state information is received from a device associated with the relay station. Updated scheduling parameters are obtained based on the received scheduling information, the channel state information, and the amount of data currently stored in the buffer. Data stored in the buffer can then be transmitted to the associated device, according lo the updated scheduling parameters. The channel stale information can be transmitted to a base station or intermediate relay station to provide feedback about the channel state. The channel state information can be quantized before being transmitted, to reduce the volume of data being transmitted over the network. The relay station may be part of a device to device D2D or Wimax communication system.

Description

Scheduling Resources at a Relay Station in a Mobile Communications Network

Technical Field

The present invention relates to scheduling resources in a mobile communications network. In particular, the present invention relates to scheduling resources at a relay station in a relay-based network.

Background of the Invention

Mobile communications networks enable data to be transmitted wirelessly between Base Stations (BS) and mobile devices, for example mobile telephones, laptops, and tab'et computers. In the thst decade relay-based networks have been developed which can enhance the coverage of a network as well as improving overall system throughput and enert efficiency. In a relay-based network, a Relay Station (RS) relays contr& i signalling and/or data between a BS and a Mobile Station (MS). Multi-relay-assisted communication has been standardized, for example in both 4G LTE-Advanced and WiMAX IEEE 8o2.16m mobile broadband communication systems.

The scheduling of resources between MSs in the network is an important task which can significantly impact the overall system performance. For example, in an Orthogonal Frequency Division Multiple Access (OFDMA) network, such as a 4G LTE network or a WiMAX network, scheduling can be performed to allocate available orthogonal subcarriers (frequency bandwidth) to MSs, and to identify the most appropriate modulation (MOD) and channel coding (COD) schemes for each MS.

The scheduling process takes into account a number of parameters, including feedback information provided by involved nodes (MSs and RSs) which quantifies the status of the wireless channel (link) between two nodes within the network. In a conventional relay-based network, scheduling is performed centrally at each BS. However, as the number of RSs in the network increases, and particular'y when multi-hop chains including a phirality of RSs are considered, the overafi system performance can be significantly decreased as the volume of channel state feedback information being sent over the network increases.

3s The invention is made in this context.

Srnnmary of the Invenlion According to the present invention, there is provided a method of performing scheduling at a r&ay station in a mobile communications network, the method comprising: receiving scheduling information defining scheduling parameters for downlink transmission of data stored in a buffer, and channel state information from a device associated with the relity station; and obtaining updated scheduling parameters based on the received scheduling information, the channel state information, and the amount of data currently stored in the buffer.

The updated schedufing parameters can be obtained by: identifying a plurality of avai1abe scheduling parameters which require tess than or equal to the amount of data currently stored in the buffer; and selecting the updated scheduling parameters from the plurality of available scheduling parameters according to one or more predetermined conditions.

S&ecting the updated schedufing parameters according to one or more predetermined conditions can comprise: selecting the scheduling parameters which provide the maximum throughput out of the available scheduling parameters; and/or selecting the scheduling parameters which provide a desired Quality of Service QoS. For example, the desired QoS can be a predefined QoS, or can be a QoS requested by the device associated with the relay station.

In response to none of the available scheduling parameters satis'ing the one or more predetermined conditions, the method can further comprise: postponing the transmission of the data stored in the buffer.

The received channel state information can comprise a channel quality metric, and the mcthod can further comprise: calcu1atiig a differcime bctwccn the current value of the channel quality metric in the received chann& state information, and a previously-received value of the channel quaBty metric; and transmitting the cakulated difference to a base station or intermediate r&ay station as channel state information.

The method can further comprise: quantizing the channel state information by adjusting the channel state information to one of a plurality of predefined levels; and transmitting the quantized channel state information to a base station or intermediate relay station. The received channel state information can be quantized and transmitted, or the calculated difference can be quantized and transmitted.

The method can further comprise selecting the number of predefined eves for use when quantizing the channel state information by: selecting a lower number of predefined levels for a higher number of hops on a multi-hop chain between a mobile station and a base station; and/or selecting a lower number of predefined levels for higher channel fading conditions.

io The received channel state information can be information about a state of a communications channd between the relay station and the device associated with the relay station in a previous time period, and the method can further comprise: obtaining predicted channel state information about a state of the communications channel in a current time period, based on the received channel state information, wherein the updated schedufing parameters can be obtained based on the predicted channel state information.

The relay station can be operated according to a decode-and-forward DF relaying method, and obtaining the updated scheduling parameters can comprise obtaining a new modulation and coding scheme.

The received scheduling information can define a bandwidth allocated to the device associated with the relay station, and the defined bandwidth can be retained when obtaining the updated scheduling parameters.

The received scheduling information can define a bandwidth allocated to the device associated with the relay station, and obtaining the updated scheduling parameters comprises obtaining a new bandwidth for the device associated with the relay station by re-allocating bandwidths allocated to a plurality of devices.

The bandwidth can be re-aflocated by reducing a bandwidth allocated to one of the devices wIth a poor channàl state and increasing a bandwidth allocated to one of the devices with a better channel state.

The relay station can be operated according to an amplify-and-forward AF relaying method, and obtaining the updated scheduling parameters can comprise obtaining a new bandwidth and power allocation. Mternatively, the relay station can be operated according to a compress-and-forward CF relaying method, and obtaining the updated scheduhng parameters can comprise obtaining a new bandwidth and modulation scheme.

Obtaining the updated scheduling parameters can comprise obtaining the updated scheduling parameters based on a round robin algorithm, a proportional fairness algorithm, or an adaptive proportional fairness algorithm.

io The scheduling information can define scheduling parameters allocated by a base station and can be received directly from the base station or indirectly via one or more intermediate relay stations. Alternatively, the scheduling information can define scheduling parameters allocated by an intermediate relay station and can be received from the intermediate relay station.

The method can further comprise: receiving downhnk data; and adding the received downlink data to the buffer, before obtaining the updated scheduling parameters.

The channel state information can include one or more of a Channel Quality Indicator CQI, a Signal to Noise Ratio SNR, a Signal to Interference and Noise Ratio SINR, a Physical Carrier to Interference and Noise Ratio (CINR), effective CINR, Multiple In Multiple Out (MIMO) mode selection and frequency selective sub-channel selection.

The mobile communications network can be an LTE or WiMAX network.

The method can be performed by a mobile station configured to operate as a relay station according to a Device to Device D2D communication method.

According to the present invention, there is also provided a computer-readable storage medium arranged to store a computer program which, when executed, performs the method.

According to the present invention, there is also provide a relay station for use in a mobile communications network, the relay station comprising: a buffer arranged to store data; a receiving module arranged to receive scheduling information defining scheduling parameters for downlink transmission of the data stored in the buffer, and channel state infonnation from a device associated with the relay station; and a scheduling module arranged to obtain updated scheduling parameters based on the received scheduhng information, the channd state information, and the amount of data currently stored in the buffer.

The scheduling module can be arranged to identify a plurality of available scheduling parameters which require less than or equal to the amount of data currently stored in the buffer, and select the updated scheduling parameters from the plurality of available scheduling parameters according to one or more predetermined conditions. I0

The scheduling module can be arranged to select the scheduhng parameters which provide the maximum throughput out of the available scheduling parameters, and/or to select the scheduling parameters which provide a desired Quality of Service QoS.

In response to none of the available scheduling parameters satisfying the one or more predetermined conditions, the relay station can be arranged to postpone the transmission of the data stored in the buffer.

The received channel state information can comprise a channel quality metric, and the relay station can further comprise: a difference calculating module arranged to calculate a difference between the current value of the channel quality metric in the received channel state information, and a previously-received value of the channel quality metric; and a transmission module arranged to transmit the calculated difference to a base station or intermediate relay station as channel state information.

The relay station can further comprise: a quantization module arranged to quantize the channel state information by adjusting the channel state information to one of a plurality of predefined levels; and a transmission module arranged to transmit the quantized channel state information to a base station or intermediate relay station.

The quantization modifie can be arranged to select the number of predefined evels for use when quantizing the channel state information by selecting a lower number of predefined levels for a higher number of hops on a multi-hop chain between a mobile station and a base station, and/or by selecting a lower number of predefined levels for higher channel fading conditions.

The received channel state information can be information about a state of a communications channel between the relay station and the device associated with the relay station in a previous time period, and the relay station can further comprise: a prediction module arranged to predict channel state information about a state of the communications channel in a current time period, based on the received channel state information, wherein the scheduling module can be arranged to obtain the updated scheduling parameters based on the predicted channel state information.

The relay station can be arranged to be operated according to a decode-and-forward DF io relaying method, and the scheduling module can be arranged to obtain a new modulation and coding scheme as the updated scheduling parameters.

The received scheduling information can define a bandwidth allocated to the device associated with the relay station, and the scheduling module can be arranged to retain is the defined bandwidth when obtaining the updated scheduling parameters.

The received scheduling information can define a bandwidth allocated to the device associated with the relay station, and the scheduling module can be arranged to obtain a new bandwidth for the device associated with the relay station by re-allocating bandwidths allocated to a plurality of devices associated with the relay station, when obtaining the updated scheduling parameters.

The scheduling module can be arranged to re-allocate the bandwidth by reducing a bandwidth allocated to one of the devices associated with the relay station with a poor channel state and increasing a bandwidth allocated to one of the devices associated with the relay station with a better channel state.

The relay station can be arranged to be operated according to an amplify-and-forward AF relaying method, and the scheduling moduk can be arranged to obtain a new bandwidth and power allocation when obtaining the updated scheduling parameters, or the relay station can be arranged to be operated according to a compress-and-forward CF relaying method, and the scheduling module can be arranged to obtain a new bandwidth and modulation scheme when obtaining the updated scheduling parameters.

The scheduling module can be arranged to obtain the updated scheduling parameters based on a round robin algorithm, a proportional fairness algorithm, an adaptive proportional fairness algorithm, or a greedy scheduling algorithm.

The scheduling information can define scheduling parameters allocated by a base station and the receiving module can be arranged to receive the scheduling information directly from the base station or indirectly via one or more intermediate relay stations, or the scheduling information can define scheduling parameters allocated by an intermediate relay station and the receiving module is arranged to receive the io schedtiling information from the intermediate rehiy station.

The receiving module can be arranged to receive downlink data, and the relay station can be arranged to add the received downlink data to the buffer before the scheduling module obtains the updated scheduling parameters.

The relay station can be configured for use in an LTE or WiMAX network. The relay station can be a mobile station configured to operate as a relay station according to a Device to Device (D2D) communication method.

Brief Description of the Drawings

Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 illustrates a mobile communications network, according to an embodiment of the present invention; Figure 2 illustrates a method of performing scheduling at a rethy station, according to an embodiment of the present invention; Figure 3 illustrates a method of providing quantized channel state feedback from a relay station, according to an embodiment of the present invention; Figure 4 illustrates a method of providing channel state feedback from a relay station using a difference between current and previous values of channel state information, according to an embodiment of the present invention; Figure 5 illustrates a method of performing scheduling at a relay station, according to an embodiment of the present invention; Figure 6 schematically illustrates a r&ay station arranged to perform scheduling for a decode-and-forward (DF) relaying scheme, according to an embodiment of the present invention; Figure 7 schematically illustrates a relay station arranged to perform scheduling for an amplify-and-forward (AF) relaying scheme, according to an embodiment of the present invention; Figure 8 illustrates a mobile communications network in which bandwidth is re-io allocated by relay stations, according to an embodiment of the present invention; and Figure 9 illustrates a timellne showing the two-stage scheduling process, according to an embodiment of the present invention.

Detailed Description

Referring now to Fig. 1, a mobile communications network is illustrated, according to an embodiment of the present invention. The mobile communications network 100 is a relay-based network comprising a Base Station (BS) 110, a phirallty of Relay Stations (RS) 121, 122, 123,124, md a plurality of Mobile Stations (MS) 131, 132, 133, 134. Each MS can be, for example, a mobile telephone, laptop, tablet computer, or any other suitable device able to connect to the network 100. The mobile communications network 100 can also be referred to as a cellular mobile network, since each BS 110 and RB 121, 122, 123, 124 defines a cell in which a MS or another RB can be served by the particular BS or RS. The RS to which a device (MS or other RS) is associated can be referred to as the "serving RS" for that device.

In the present embodiment, the network includes first, second, third and fourth MSs 131, 132, 133, 134. Each MS can be served by the BS 110, by a RB connected to the BS 110, or by a RB connected to the BS 110 through a multi-hop chain comprising one or more other RSs. In the configuration shown in Fig. 1, the third MS 133 is served direcfly by the BS 110. The first and second MSs 131, 132 are served by first and second RSs 121, 122 connected to the BS 110. The fourth MS 134 is connected to the BS 110 through a multi-hop chain comprising third and fourth RSs 123, 124. The fourth RS 124 can be referred to as the serving RS for the fourth MS 134, and the third RS 123 can be referred to as an intermediate RS.

It will be understood that the above-described configuration is not fixed, and the links between the MSs 131, 132, 133, 134, RSs 121, 122, 123, 124 and the BS 110 may change over time as the MSs 131, 132, 133, 134 move wfthm the network coverage area or join and leave the network 100.

In the present embodiment, radio resources in the network 100 are allocated through a two-stage scheduling procedure. In the first stage, resources are initially allocated to MSs by the BS in a centralized fashion, by obtaining scheduling parameters for each MS in the network sector controlled by that BS. That is, the BS allocates resources not only io to MSs direcfly associated with the BS itself, but also to any RSs operating within the sector controfled by the BS. Then, in the second stage, post-processing is carried out at the RS to obtain updated scheduling parameters. In contrast to a conventional centralized scheduling method in which only the BS performs scheduling, methods performed by embodiments of the present invention can be referred to as semi-is centralized scheduling methods.

When scheduling parameters are updated at the RS, more up-to-date feedback information on the channel state can be used. Specifically, in terms of a BS-RS-MS relay chain, the RS is by definition closer than the BS to the MS. The time delay in receiving feedback information from the MS at the RS is less than the delay experienced by the BS, and so the feedback information available at the RS will be more up-to-date than the feedback information available at the BS. The impact of delayed feedback is worsened in the case of multi-hop relaying transmission, where several relays are involved into the transmission process in sequential fashion, and also in the case of fast fading channels (e.g. when an MS is travelling rapidly). In embodiments of the present invention, semi-centralized scheduling methods can be used to obtain updated scheduling parameters which are better suited to the current channel conditions.

In addition to pcrforming second-stage scheduling, an RS in the prcscnt cmbodimcnt can also relay feedback information about the channel state back to the BS, or to another RS higher up the chain. Because scheduhng parameters wifi be updated at the RS, in comparison to a conventional centralized scheduling method it is tess critical in the present embodiment for the BS to be provided with accurate feedback information.

Therefore in the present embodiment the channel state information is quantized before being fed back along the chain, reducing the number of bits to be transmitted as feedback information and thereby reducing system overheads. -10-

In the present embodiment the network 100 is an LTE network. However, embodiments of the present invention are not limited to LTE networks and can be applied to any type of relay-based network. Examples of types of network to which embodiments of the present invention can be applied include, but are not limited to, a WiMAX network configured according to the IEEE 8o2.16m standard, a mesh network configured according to the IEEE 802.15.5 WPAN standard, and a Body Area Network (BAN) configured according to the IEEE 802.15.6 standard.

io Referring now to Fig. 2, a method of performing schedtiling at a relay station is iflustrated, according to an embodiment of the present invention. The method can be used by any of the RSs shown in Fig. 1, to perform second-stage scheduling after the BS has performed first-stage centralized scheduling.

To perform first-stage scheduhng, the BS receives delayed channel state information from a MS through one or more RSs. In the present embodiment the channel state information is provided in the form of a Channel Quality Indicator (CQI). The channel state information can include the numerical value of the CQI, or any suitable information to allow the BS to identifr a numerical value, for example an index corresponding to one of a plurality of predefined quantized levels. In other embodiments, instead of or as well as the CQI, the channel state information can include one or more other metrics of channel quality. Suitable metrics of channel quality include, but are not limited to, a Signal to Noise Ratio (SNR),r a Signal to Interference and Noise Ratio (SINR), a Physical Carrier to Interference and Noise Ratio (CINR), effective CINR, Multiple In Multiple Out (MIMO) mode selection and frequency selective sub-channel selection. The BS similarly obtains channel state information for all other users (MS, RS) within the BS sector.

Thc BS then performs centralized scheduling to achieve a fair allocation of resources.

In the present embodiment the network is operated according to a decode-and-forward (DF) relaying method, and the centralized scheduling pmcedure includes the allocation of frequency bandwidth (BW) and identification of the most efficient modulation (MOD) and channel coding rate (COD) schemes for each channel to ensure maximum throughput at the user end. In this way, scheduling parameters (BW, MOD, COD) are obtained for each user. Scheduling information defining the obtained scheduling -11 -parameters are forwarded to devices in the network (MSs and RSs), through one or more intermediate RSs where applicable.

Once the BS has performed first-stage schedufing, a RS in the mobfle communications network performs second-stage scheduling using the method shown in Fig. 2.

First, in step 5201 the RS receives scheduling information and downlink data (D_1). In the present embodiment the scheduling information is received from a BS and defines the scheduling parameters allocated by the BS during first stage scheduling. In another Jo embodiment the scheduling information can be received from another RS, i.e. an intermediate RS, and can define updated scheduling parameters that have been obtained by the intermediate RS. The scheduling parameters are used to control the downlink transmission of data over a communications link between the RS and a device associated with the RS, which in the present embodiment is a MS. The RS stores the is downlink data in a buffer prior to transmission to the MS.

In step 5202, the RS receives channd state information as feedback from the MS. The channel state information comprises information about the state of a communications channel between the RS and the MS. Although in Fig. 2 the channel state information is shown as being received after the scheduling information, in general steps S2o1 and S2o2 can be performed in the reverse order, or at the same time. Transmission and processing delays mean that the received channel state information conveys information about a state of a communications channel between the relay station and a mobile station in a previous time period.

Then, in step 5203 the buffer is updated by adding the received data D11_1 to the data already queued in the buffer, Q_1. When the received data D0_1 is coded, for example when the RS is operated according to a Decode and Forward DF relay method, the received data D-1 can first bc dccodcd using the MOD, COD schemes indicated in the o received scheduling information, before updating the buffer with the decoded data.

The buffer can be updated at any point between receiving the downlink data (step S2o1) and updating the scheduling parameters (step S2o5).

Next, in step S2o4 the RS obtains predicted channel state information about a state of the communications channel in a current time period, based on the received channel state information. The channel state information can be predicted based on previously -12 -stored historical feedback, and the prediction process can try to identify channel behaviour from predeflned channel models.

Then, in step S2o5 the RS obtains updated scheduhng parameters based on the predicted channel state information. In other embodiments step S2o4 can be omitted and the scheduling parameters can be updated using the received channel state information, for example in conditions where the channel state is only slowly changing (e.g. stationary MS) it may be assumed that the received channel state information is a good approximation to the current channel state. As another example, channel io feedback fluctuations may show random behaviour, in which case the RS can be arranged to not perform prediction but to use the received channel state information instead when obtaining updated scheduling parameters.

In step S2o5, the scheduling parameters can be updated based on, for example, a round robin algorithm, a proportiona' fairness algorithm, or an adaptive proportional fairness algorithm.

In addition, when obtaining the updated scheduling parameters in step S205, the RS also takes into account the received scheduling information and the amount of data for downlink transmission currently stored in the buffer. In the present embodiment, when updating the scheduling parameters the RS retains the BW allocated by the BS, but obtains new MOD, COD schemes if these will achieve a higher throughput under the current channel conditions, provided that the new MOD, COD schemes require less than or equal to the amount of data currently stored in the buffer.

By taking into account more tip-to-date feedback information about the channel state than was available to the BS during first stage scheduling, and also the current state of the local buffer, the RS can maximise throughput by obtaining updated scheduling parameters which are better suitcd to the currcnt conditions than those originally o allocated by the BS, during first-stage scheduling.

Then, in step S2o6, at least part of the downllnk data stored in the buffer is transmitted to the MS, in accordance with the updated séheduhng parameters. The downlink data can be transmitted by packaging the data based on the updated scheduling parameters, and transmitting the packaged data.

-13 -Referring now to Fig. 3, a method of providing quantized channel state feedback from a relay station is illustrated, according to an embodiment of the present invention.

First, in step 5301 the channel state information is received from a device associated with the relay station. In the present embodiment the channel state information is a CQI value, and the associated device is a MS. Step S3o1 corresponds to step S2o2 in the method of Fig. 2. Then, in step 5302 a number of predefined levels are selected within a range of values. The levels can be evenly spaced within the range of values.

io When more quantization levds are selected, the spacing between quantization levds is reduced and the quantized CQI will be doser to the original value. However, when more quantization levels are selected, the number of bits required to transmit the quantized CQI may increase. To reduce the volume of data being transmitted over the network as channel feedback information, in step 5302 a lower number of predefined levds can be sdected when there is a higher number of hops on a multi-hop chain between a mobile station and a base station, and/or a lower number of predefined levds can be sdected when there are higher channd fading conditions. Under each of these conditions it is less important to provide less accurate feedback to the BS or intermediate RSs further along the chain, since the channel state information is more likely to be out of date by the time it is received.

In some embodiments, step S3o2 can be omitted and a predetermined fixed number of quantization levels can be used.

Next, in step 5303 the CQI value is quantized by adjusting the received channel state information to one of a plurality of predefined levels. Then, in step 5304 the quantized CQI is transmitted to a base station or intermediate relay station.

Rcfcrring now to Fig. 4, a mcthod of providing chann& statc fccdback from a rclay station using a difference between current and previous values of channel state information is iflustrated, according to an embodiment of the present invention.

Steps S4o1, S4o3, S4o4 and S4o5 are similar to steps 5301, S3o2, S3o3 and S3o4 of Fig. 3, respectively, and a detailed description will not be repeated here. However, unlike the method of Fig. 3, in the present embodiment a difference is calculated between the current CQI value CQI, and the CQI value that was previously sent to the BS. In the present embodiment the channel state information is a CQI value, but this approach can also be used for any channel quality metric, that is, any quantified numerical vahle representing a measure of the channel quality. The difference, ACQI, is then quantized and transmitted. The vahle of ACQI may be substantially smafler than the absolute value of CQI, meaning that fewer bits may be required to transmit the difference value ACQI than to transmit the actual CQI value.

In some embodiments the difference value ACQI can be transmitted without quantization, in which case steps 5403 and 5404 can be omitted. I0

The methods of Figs. 3 and 4 enable the number of bits required when transmitting channel state information as feedback to devices further along the chain (e.g. BS or intermediate RSs). A RS can use the method of Fig. 3 or 4 to provide feedback at any time after receiving channel state information (step S2o2 of Fig. 2). That is, the steps shown in Figs. 3 or 4 can be performed before, after, or at substantiay the same time as, steps 5204 to 5206 of Fig. 2.

Referring now to Fig. 5, a method of performing scheduling at a relay station is illustrated, according to an embodiment of the present invention.

In step 5501, scheduling information is received from a BS or intermediate RS, similar to step S2o1 of Fig. 2. In addition, downlink data (D1) to be transmitted to the associated device, such as a MS or intermediate RS, is also received.

In step S5o2, channel state information is received from a device associated with the RS, for example a MS or intermediate RS, in a similar manner to steps 5202, 5301 and 5401 of Figs. 2, 3 and 4 respectively. In addition, information about a desired Quality of Service (QoS) is also received from the associated device.

o Then, in step S5o3 the buffer is updated by adding the received data D_1 to the data already queued in the buffer, Q. Step 5503 corresponds to step 8203 in Fig. 2, and the method of Fig. 5 can also indude the step of decoding the received data if required, for example when the RS is operated according to a Decode and Forward DF rday method.

-15 -In step S5o4, available scheduling parameters are identified by identifying the scheduling parameters which require less than or equal to the amount of data currently stored in the buffer (D «= Q), and which provide the desired QoS requested by the associated device. In some embodiments, instead of allowing a device to request a desired QoS, the RB can select updated scheduling parameters which provide a predefined QoS. For example, a predefined QoS level can be set using common block error rate (BLER) values, to ensure a certain level of successful decoding at the receiver (MS or other RS).

io In the present embodiment the scheduling parameters include a BW allocation and MOD, COD schemes, and the schedufing parameters are updated by obtaining new MOD, COD schemes whilst retaining the originally-allocated BW. In another embodiment however, the BW can be re-allocated when updating the scheduling parameters. For example, when there are a plurality of devices (MS, RS) associated is with the RS, the bandwidth can be re-aflocated amongst the devices by reducing a bandwidth allocated to one of the devices associated with the RS with a poor channel state and increasing a bandwidth allocated to one of the devices with a better channel state. This can result in more efficient scheduling, since the bandwidth allocated to users with poor channel conditions can be re-allocated to other users with better channel conditions. When re-allocating BW amongst a plurality of users, the RS may attempt to maximise the overall system capacity instead of maximising individual user throughputs.

Next, in step S505 scheduling parameters which provide the maximum throughput are selected from the set of available parameters identified in step S5o4. Although in the present embodiment the scheduling parameters are selected based on throughput, in other embodiments a different predetermined condition can be used to select the scheduling parameters in step S5o5, for example the parameters which provide the highcst QoS can be sciected.

In step So6, as long as a set of scheduling parameters have been found in steps S5o4 and then in step S5o7 at least some of the data is transmitted according to the updated schedufing parameters. If not all of the data is transmitted, unsent data is retained in the buffer and can be sent at a later time, when a better channel state occurs which allows higher-throughput MOD, COD schemes to be used.

In the event that no suitable scheduling parameters were found in steps 5504 and S5o5, for example if no scheduling parameters were available which could provide the desired QoS and which require D «= Q, then in step 5508 the RS postpones transmission until the next time s'ot. In some embodiments, instead of postponing transmission the RS could transmit the data using other scheduling parameters, for example the parameters defined in the received scheduling information, or by selecting updated parameters which provide a lower QoS than that requested by the associated device.

The updated scheduling parameters can also be selected in consideration of the io bandwidth (BW) allocated to the MS by the BS, by ensuring that on'y schedullng parameters are selected which are able to provide the BW allocation.

Although in the present embodiment the buffer is updated before performing second-stage scheduling in steps S5o4 and S5o5, in other embodiments the updated scheduling parameters may be obtained without first updating the buffer. For example, in the case of buffer overflow, it may not be possible to add the new data to the buffer until the data already present has been transmitted. Therefore in some use-cases step 5503 may be omitted.

Referring now to Fig. 6, a relay station arranged to perform scheduling is schematically illustrated, according to an embodiment of the present invention. Certain modules shown in Fig. 6 may be embodied as software instructions within a computer program executed by a processor, or may be embodied as physically separate hardware components, depending on the embodiment. The RS 620 of the present embodiment is configured to operate according to a DF relaying scheme.

The RS 620 comprises a BS receive/transmit module 621 arranged to communicate with a BS 610, a scheduling module 622 arranged to obtain updated scheduling parameters, and a MS receive/transmit module 623 arranged to communicate with a MS 630. Each receive/transmit module 621, 623 can include a separate transmitter and receiver, or a combined transmitter/receiver. Although separate receive/transmit modules 621, 623 are shown in Fig. 6, the same transmitter and/or receiver maybe used when communicating with the BS and MS. -17-

Also, although in the present embodiment the ItS 620 is shown communicating with a BS 6io and MS 630, in other embodiments one or both of the BS 610 and MS 630 may be repthced with another RS.

The scheduling module 622 can perform second-stage scheduling at the RS 620 according to any of the methods described herein, by updating the received scheduling parameters. In Fig. 6, the subscript "U" denotes the updated scheduling parameters, which in the present embodiment are updated MODLI, COD and BWLF parameters.

io The P.S 620 also comprises a data decoding module 624 arranged to demodulate and decode data received from the BS 610, according to the MOD, COD scheme indicated in the received scheduling parameters. In some embodiments the data decoding module 624 can be omifted, for example when the RS 620 operates according to an amplify-and-forward AF or compress-and-forward CF relaying scheme. The RS 620 further comprises a buffer 625 for the associated MS 630, which is arranged to store downUnk data Q for transmission to the MS 630, and a data encoding module 626 which encodes and modulates data (D) output from the buffer 626 before transmission to the MS 63o. As shown in Fig. 6, the scheduling module 622 also controls the buffer 625 to output the data D for transmission, by sending information to the buffer 625 about the amount of data (ND) to be transmitted in this time slot. The buffer outputs the data D indicated by the scheduling module 622, which will be less than or equal to the total amount of data queued in the buffer 625 (Q). The scheduling module 622 also informs the data encoding module 626 of the updated MODE, CODu schemes to use when encoding the data D. In addition, the RS 620 further comprises a prediction module 627, a difference module 628, and a quantization module 629, which can be arranged to provide channel state information feedback to the BS 610 (or an intermediate P.5) using a method such as the onc shown in Fig. 3 or Fig. 4. In the prcscnt cmbodimcnt the chann& state information is a CQI value, bitt in other embodiments different types of channel state information can be used.

As described above with reference to Figs. 2 to 4, in some embodiments of the present invention certain method steps can be omitted, for example CQI prediction, difference calculation and quantization. Accordingly, in some embodiments of the RS, one or more of the CQI prediction module 627, CQI difference module 628, and CQI -18-quantization module 629 can be omitted. Mso, although data decoding and encoding modules 624, 626 are included in the present embodiment for use in a DF relaying scheme, these can be omitted in embodiments which operate according to different relaying schemes.

Referring now to Fig. 7, a relay station arranged to perform scheduling is schematically illustrated, according to an embodiment of the present invention. As with Fig. 6, certain modules shown in Fig. 7 may be embodied as software instructions within a computer program executed by a processor, or may be embodied as physically separate io hardware components, depending on the embodiment.

The RS 720 of the present embodiment is configured to operate according to an AF relaying scheme. Like the P.S 620 of Fig. 6, the RS 720 of the present embodiment includes a BS receive/transmit module 721 arranged to communicate with a BS 710, a schedtiBng module 722, a MS receive/transmit module 723 arranged to communicate with a MS 730, a buffer 725, a CQI prediction module 727, a CQI difference cakulating module 728, and a CQI quantization module 729.

Because the RS 720 of the present embodiment operates according to an AF relaying scheme instead of a DF relaying scheme, the RS 720 includes an analogue-to-digital converter ADC 724 instead of a data decoding module, and a digital-to-analogue converter DAC 726 instead of a data encoding module. The ADC 724 digitises received analogue data for storing in the buffer 725, and the DAC 726 converts digital data D output by the buffer into analogue data for transmission. The MS receive/transmit module 723 amplifies the analogue data according to the updated transmission power (PWRu) set by the scheduling module, and the BW allocated to the MS 730 by the BS 710.

In both Fig. 6 and Fig. 7, a RS retains the BW allocated to an MS by a BS, when updating the received scheduling parameters for that MS. However, as described above, in some embodiments a P.S can re-a&cate BW amongst a plurality of devices associated to that P.S.

Referring now to Fig. 8, a mobile communications network in which bandwidth is re-allocated by relay stations is illustrated, according to an embodiment of the present invention. For clarity, only bandwidth allocations amongst the scheduling parameters will be described here, but depending on the embodiment the scheduling parameters may include other parameters, such as MOD, COD or PWR parameters.

The network includes a BS 810, first and second RSs 821, 822, and first, second, third, fourth and fifth MSs 834, 835. The first, second and third MSs 831, 832, 833 are associated with the first RS 821, and the fourth md fifth MSs 834, 835 are associated with the second RS 822. The BS 8io allocates a first bandwidth BW to the first MS 831, a second bandwidth BW2 to the second MS 832, a third bandwidth BW3 to the third MS 833, a fourth bandwidth BW4 to the fourth MS 834, and a fifth bandwidth io BW5 to the fifth MS 835.

The BS 8io signals the allocated bandwidths to the first RS 821 as scheduling parameters. When obtaining updated scheduling parameters during second-stage scheduling, the first RS 821 re-allocates the first, second and third bandwidths BW1, is BW2, BW3 amongst the first, second and third MSs 831, 832, 833, thereby obtaining updated bandwidth allocations BWIU, BW2U, BW3u. The first RS 821 does not re-allocate bandwidth for devices not associated with itself, namely the fourth and fifth MSs 834, 835. For these devices, the first RS 821 retains the originally-allocated bandwidths BW4, BW5 and signals these to the second RS 822. This avoids possible interference between devices in neighbouring cells, since each RS only alters bandwidth usage within its own cell.

Then, the second RS 822 re-allocates bandwidth to the fourth and fifth MSs 834, 835, obtaining updated bandwidths BW4u, BW5u. Although a two-hop relay chain is illustrated in the present embodiment, it will be understood that the principle can be extended to relay chains of any length.

Referring now to Fig. 9, a timeline is illustrated showing the two-stage scheduling process, according to an embodiment of the present invention. The timeline illustrates steps performed in a network by a BS 910, a first RS 921, a klh RS 922, and a MS 931. In the present embodiment the network is operated according to a DF relaying scheme, channel state feedback is provided in the form of quantized ACQI values, and RSs do not re-allocate BW, but it will be understood from the above description that embodiments of the invention are not limited to these particular examples.

The BS 910 receives delayed channel state information for each MS associated with the BS 910, and performs a first-stage centralized scheduling for all of the associated MSs.

Here, the BS 910 obtains scheduling parameters BW1, MODI, CODI for the MS 931. The BS 910 transmits the scheduling parameters BW1, MOD1, COD1 to the first RS 921 along with downlink data for transmission to the MS 931.

The first RS 921 forwards the scheduling parameters BW1, MOD1, COD1 and downlink data along a relay chain to the kthl RS 922, which is the serving RS for the MS 931. In the present embodiment, scheduling parameters for an MS are only updated by the io serving RS for that MS, but in other embodiments the scheduHng parameters could be updated in a progressive fashion by each relay along the BS-RS1-RSk-MS relay chain.

The kth RS 922 receives the scheduling parameters BW1, MOD1, COD1 and downlink data, and also receives new CQI feedback from the MS 931. The ktll RS 922 then is updates its oca buffer for the MS 931 with the new downlink data, and performs second-stage scheduling to obtain updated MOD, COD parameters MOD1, CODIL for the MS 931. It can be seen from Fig. 9 that the CQI feedback available to the kthl RS 922 during second-stage scheduling is more up-to-date than the CQI feedback that was available to the BS 910 during first-stage scheduling.

Finally, the ktll RS 922 transmits downlink data from the buffer to the MS 931 in accordance with the originally-allocated BW1 and the updated MOD111, COD111 scheduling parameters.

In some network standards, including LTE-Advanced, WiMAX IEEE 8o2.16m, from a physical layer perspective a RS can operate in one of two modes: non transparent (NT-RS) or transparent (T-RS). A non-transparent RS can also be referred to as a Type-I RS, and a transparent RS can also be referred to as a Type-Il RS. A NT-RS can be used to cxtcnd signal and service covcragc, and transmits a preamble and other broadcast messages as well as relaying data traffic. AT-RS can be used to improve service quafity and link capacity through multipath diversity, for a MS which also has a direct link to a BS. However, the T-RS on'y relays data traffic, and the MS receives control signals direcfly from the BS. Because a T-RS is not able to generate or amend control messages produced by a BS, the T-RS cannot participate in scheduling. Therefore in networks which feature both Type I (NT-RS) and Type II (T-RS) relay stations, embodiments of the invention can be used to perform scheduling at a NT-RS but not at a T-RS.

In simulations of an Above Rooftop (ART) relaying scenario in a WiMAX system with 19 cells, in which 20 mobile stations (MS) are associated with the serving node in each sector, semi-centralized scheduling methods such as the ones described above achieve a gain of around 9% in terms of average throughput per cell, where only data delivered by relays are considered is observed. The simulations were performed for a simple three- hop scenario with stow fading channels, and a larger gain would be expected in real-world situations with more hops between BS and MS and/or rapidly fading channe's.

Jo Embodiments of the present invention have been described in which a RS is operated according to a decode-and-forward DF relaying scheme, and new modifiation and coding schemes are obtained when updating the scheduling parameters. However, in other embodiments a second-stage scheduling procedure can be performed by a RS operated according to a different relaying scheme.

For examp'e, in an ampEfy-and -forward AF relaying scheme the RS does not perform demodifiation and decoding, and so it is not possib'e to after the MOD, COD scheme.

When the RS is operated according to the AF relaying scheme, a new bandwidth and power allocation can be obtained when updating the scheduling parameters.

Alternatively, svhen the RS is operated according to a compress-and-forward CF relaying scheme, a new bandwidth and modulation scheme can be obtained when updating the scheduling parameters.

Embodiments of the present invention have been described in which a RS performs a second-stage scheduling procedure. In multi-hop chains including a plurality of intermediate RSs, the second-stage scheduling for a MS may only be performed by the serving RS (the RS to which the MS is associated), or may alternatively be performed in a progressive manner by performing second-stage scheduling at each intermediate RS &ong the chain, in turn.

By updating scheduling parameters at a RS in a mobile communications network, a number of advantages can achieved in comparison to a centr&ized scheduling method.

System throughput can be improved because more up-to-date feedback is available at the RS level, and available buffers at relays can be exploited to avoid transmitting data over corrupted links, without having to exchange buffer state information with a BS. -22-

Energy efficiency can be improved as a result, because not sending data over a corrupted channel will reduce the reoccurrence of HARQ (request of resending failed frames). Also, when rescheduling is performed at a RS, it is not necessary to provide accurate chann& state feedback to a BS since a refined second stage rescheduflng stage will take place at the relay level. The RS may only be required to send quantized and differential estimations about channel states, instead of accurate values, which mean fewer bits are needed to convey the channel status. This will reduce the amount of data overhead on the system.

Whilst certain embodiments of the invention have been described herein with reference to the drawings, it will be understood that many variations and modifications will be possible without departing from the scope of the invention as defined in the accompanying claims.

Claims

-23 -Claims A method of performing scheduling at a relay station in a mobile communications network, the method comprising: receiving scheduling information defining scheduling parameters for downlink transmission of data stored in a buffer, and channel state information from a device associated with the relay station; and obtaining updated scheduling parameters based on the received scheduling information, the channel state information, and the amount of data currently stored in io the buffer.
2. The method of claim 1, wherein the updated scheduling parameters are obtained by: identifying a plurality of available scheduling parameters which require less than or equal to the amount of data currenfly stored in the buffer; and selecting the updated scheduhng parameters from the plurality of available scheduBng parameters according to one or more predetermined conditions.
3. The method of claim 2, wherein selecting the updated scheduling parameters according to one or more predetermined conditions comprises: selecting the scheduling parameters which provide the maximum throughput out of the available scheduling parameters; and/or selecting the scheduling parameters which provide a desired Quality of Service QoS.
4. The method of claim 2 or 3, wherein in response to none of the available scheduling parameters satisfying the one or more predetermined conditions, the method further comprises: postponing the transmission of the data stored in the buffer.
5. The method of any one of the preceding claims, wherein the received channàl state information comprises a channel quality metric, the method further compnsing: calculating a difference between the culTent value of the channel quafity metric in the received channel state information, and a previously-received value of the channel quality metric; and -24 -transmitting the calculated difference to a base station or intermediate relay station as channel state information.
6. The method of any one of the preceding claims, further comprising: quantizing the channel state information by adjusting the channel state information to one of a plurality of predefined levels; and transmitting the quantized channel state information to a base station or intermediate relay station.

Jo
7. The method of claim 6, further comprising selecting the number of predefined levds for use when quantizing the channel state information by: selecting a lower number of predefined levels for a higher number of hops on a multi-hop chain between a mobile station and a base station; and/or selecting a lower number of predefined levels for higher channel fading conditions.
8. The method of any one of the preceding claims, wherein the received channàl state information is information about a state of a communications channel between the relay station and the device associated with the relay station in a previous time period, and the method further comprises: obtaining predicted channel state information about a state of the communications channel in a current time period, based on the received channel state information, wherein the updated scheduling parameters are obtained based on the predicted channel state information.g. The method of any one of the preceding claims, wherein the relay station is operated according to a decode-and-forward DF relaying method, and obtaining the updatcd scheduling parameters comprises obtaining a new modulation and coding scheme.10. The method of claim 9, wherein the received scheduBng information defines a bandwidth aflocated to the device associated with the r&ay station, and the defined bandwidth is retained when obtaining the updated scheduling parameters.-25 - 11. The method of any one of claims 1 to 9, wherein the received scheduling information defines a bandwidth allocated to the device associated with the relay station, and obtaining the updated scheduhng parameters comprises obtaining a new bandwidth for the device associated with the relay station by re-aflocating bandwidths allocated to a plurality of devices.12. The method of claim ii, wherein the bandwidth is re-allocated by reducing a bandwidth allocated to one of the devices with a poor channel state and increasing a bandwidth allocated to one of the devices with a better channel state. I013. The method of any one of daims ito 8, wherein the relay station is operated according to an amplify-and-forward AF relaying method, and obtaining the updated scheduling parameters comprises obtaining a new bandwidth and power allocation, or wherein the relay station is operated according to a compress-and-forward CF relaying method, and obtaining the updated scheduling parameters comprises obtaining a new bandwidth and modulation scheme.14. The method of any one of the preceding claims, wherein obtaining the updated scheduling parameters comprises obtaining the updated scheduling parameters based on a round robin algorithm, a proportional fairness algorithm, or an adaptive proportional fairness algorithm.15. The method of any one of the preceding claims, wherein the scheduling information defines scheduling parameters allocated by a base station and is received directly from the base station or indirectly via one or more intermediate relay stations, or wherein the scheduling information defines scheduling parameters allocated by an intermediate relay station and is received from the intermediate relay station.16. The method of any one of the preceding claims, further comprising: receiving downhnk data; and adding the received downhnk data to the buffer, before obtaining the updated scheduling parameters.17. The method of any one of the preceding claims, wherein the channel state information includes one or more of a Channel Quality Indicator CQI, a Signal to Noise -26 -Ratio SNR, a Signal to Interference and Noise Ratio SINR, a Physical Carrier to Interference and Noise Ratio (CINR), effective CINR, Multiple In Multiple Out (MIMO) mode selection and frequency selective sub-channel selection.i8. The method of any one of the preceding claims, wherein the mobile communications network is an LTE or WiMAX network.19. The method of any one of the preceding claims, wherein the method is performed by a mobile station configured to operate as a relay station according to a io Device to Device D2D communication method.20. A computer-readable storage medium arranged to store a computer program which, when executed, performs the method according to any one of the preceding claims.21. A relay station for use in a mobile communications network, the relay station comprising: a buffer arranged to store data; a receiving module arranged to receive scheduling information defining scheduling parameters for downlink transmission of the data stored in the buffer, and channel state information from a device associated with the relay station; and a scheduling module arranged to obtain updated scheduling parameters based on the received scheduling information, the channel state information, and the amount of data currently stored in the buffer.22. The relay station of claim 21, wherein the scheduling module is arranged to identify a plurality of available scheduling parameters which require less than or equal to the amount of data currently stored in the buffer, and select the updated scheduling parameters from the phirality of available scheduling parameters according to one or more predetermined conditions.23. The relay station of claim 22, wherein the scheduling modifle is arranged to select the scheduling parameters which provide the maximum throughput out of the available scheduling parameters, and/or to select the scheduling parameters which provide a desired Quality of Service QoS. -27-24. The relay station of claim 22 or 23, wherein in response to none of the available scheduling parameters satisfying the one or more predetermined conditions, the relay station is arranged to postpone the transmission of the data stored in the buffer.25. The relay station of any one of claims 21 to 24, wherein the received channel state information comprises a channel quality metric, and the relay station further comprises: a difference calculating module arranged to calculate a difference between the current value of the channel quality metric in the received channel state information, io and a previously-received value of the channd quality metric; and a transmission modifle arranged to transmit the calculated difference to a base station or intermediate relay station as channel state information.26. The relay station of any one of claims 21 to 25, further comprising: a quantization module arranged to quantize the channel state information by adjusting the channd state information to one of a plurality of predefined levds; and a transmission modifle arranged to transmit the quantized channd state information to a base station or intermediate relay station.27. The relay station of claim 26, wherein the quantization module is arranged to select the number of predefined levels for use when quantizing the channel state information by selecting a lower number of predefined levels for a higher number of hops on a multi-hop chain between a mobile station and a base station, and/or by selecting a lower number of predefined levels for higher channel fading conditions.28. The relay station of any one of claims 21 to 27, wherein the received channel state information is information about a state of a communications channel between the relay station and the device associated with the relay station in a previous time period, and the relay station further comprises: a prediction module arranged to predict channel state information about a state of the communications channel in a current time period, based on the received channel state information, wherein the scheduling modifie is arranged to obtain the updated scheduhng parameters based on the predicted channel state information.29. The relay station of any one of claims 21 to 28, wherein the relay station is arranged to be operated according to a decode-and-forward DF relaying method, and the scheduhng modifie is arranged to obtain a new modifiation and coding scheme as the updated scheduling parameters.30. The relay station of claim 29, wherein the received scheduling information defines a bandwidth allocated to the device associated with the relay station, and the scheduling module is arranged to retain the defined bandwidth when obtaining the updated scheduling parameters. I031. The relay station of any one of daims 21 to 29, *herein the received scheduhng information defines a bandwidth allocated to the device associated with the relay station, and the scheduling module is arranged to obtain a new bandwidth for the device associated with the relay station by re-allocating bandwidths allocated to a is phirabty of devices, when obtaining the updated scheduling parameters.32. The relay station of claim 31, wherein the scheduling modifle is arranged to re-allocate the bandwidth by reducing a bandwidth allocated to one of the devices with a poor channel state and increasing a bandwidth allocated to one of the devices with a better channel state.33. The relay station of any one of claims 21 to 28, wherein the relay station is arranged to be operated according to an amplif'-and-forward AF relaying method, and the scheduling module is arranged to obtain a new bandwidth and power allocation when obtaining the updated scheduling parameters, or wherein the relay station is arranged to be operated according to a compress-and-forward CF relaying method, and the scheduling module is arranged to obtain a new bandwidth and modulation scheme when obtaining the updated scheduling parameters.34. The relay station of any one of daims 21 to 33, wherein the schedding module is arranged to obtain the updated scheduling parameters based on a round robin algorithm, a proportional fairness algorithm, or an adaptive proportional fairness algorithm.-29 - 35. The relay station of any one of claims 21 to 34, wherein the scheduling information defines scheduling parameters allocated by a base station and the receiving module is arranged to receive the scheduling information direcfly from the base station or indirecfly via one or more intermediate r&ay stations, or wherein the scheduling information defines scheduling parameters allocated by an intermediate relay station and the receiving module is arranged to receive the scheduling information from the intermediate relay station.36. The relay station of any one of claims 21 to 35, wherein the receiving module is io arranged to receive downhnk data, and the relay station is arranged to add the received downlink data to the buffer before the scheduling module obtains the updated scheduling parameters.37. The relay station of any one of claims 21 to 36, wherein the channel state information includes one or more of a Channel Quality Indicator CQI, a Signal to Noise Ratio SNR, a Signal to Interference and Noise Ratio SINR, a Physic& Carrier to Interference and Noise Ratio (CINR), effective CINR, Multiple In Muhipe Out (MIMO) mode selection and frequency selective sub-channel selection.38. The relay station of any one of claims 21 to 37, configured for use in an LTE or 39. The relay station of any one of claims 21 to 38, wherein the relay station is a mobile station configured to operate as a relay station according to a Device to Device D2D communication method.