WO2009067060A1

WO2009067060A1 - Method and arrangement in a telecommunications system

Info

Publication number: WO2009067060A1
Application number: PCT/SE2007/050884
Authority: WO
Inventors: David Astely; Hannes Ekström; Tobias Tynderfeldt; Erik Westerberg
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2007-11-21
Filing date: 2007-11-21
Publication date: 2009-05-28

Abstract

A method, scheduler and a network node for scheduling retransmissions in a telecommunications system is described. The method comprises determining a scheduling point of time for a retransmission in a periodic interval cycle for transmissions in the network node, which point of time in the cycle is different as compared to the point of time in the cycle when the previous transmission was performed. The retransmission is then scheduled to be performed at that different point of time in the cycle, hence providing the placing of the data in a different position in the frame as compared to the data position of a previously interfered transmission of that data. This provides the advantage of providing an increased probability of avoiding that retransmitted data is again interfered in transmissions suffering from periodic interference problems.

Description

Method and Arrangement in a Telecommunications System TECHNICAL FIELD

The present invention relates to a method and arrangement in a telecommunications system, in particular it relates to a method and arrangement for scheduling retransmissions in a telecommunications system.

BACKGROUND

Interference disturbances between uplink (UL) and downlink (DL) transmissions in a cellular radio system, and especially in one using Time-division duplex (TDD) , may disrupt or corrupt data being transmitted therein. Such disturbances risk impairing the quality of transmission contents and thus negatively affecting the experience for the receivers of transmitted data, i.e. end users. In the case of real-time-services involving streaming audio and/or video such disturbances may cause data quality corruption to a degree perceived as unacceptable by receivers .

In a Time-division duplex (TDD) system the uplink and downlink transmission alternate periodically in time. Due to long distances in a macro network and also possible synchronization inaccuracies in the network, some radio frames may be significantly more interfered than other because of interference between radio-base-stations (RBS) . Also, due to synchronization inaccuracies, there may also be interference between mobile terminals or the like user equipment (UE's) in the same way. Moreover, such interference peaks typically occur periodically.

The described problem concerns interference both within the same carrier as well as interference between uncoordinated TDD systems operating on adjacent frequencies. In TDD systems it is important that UL and DL transmissions are kept separated in time. In particular it is important that a UE receiving weak radio signals from a far away positioned RBS is not interfered by a second UE that is positioned nearby and transmitting comparatively strong radio signals to another RBS. An example of such a situation is illustrated in Fig. 1 exemplifying UE to UE interference problems in TDD. When UE B transmits radio signals (unbroken line arrows) while at the same time UE A is trying to receive a much weaker radio signal from the RBS (broken line arrow) this may result in that UE A cannot decode the signal from the RBS due to the interference from UE B .

A similar problem is RBS to RBS interference. Such interference may occur when one RBS is receiving weak signals from a UE while at the same time being interfered, typically by a much stronger RBS. An example of such a situation is depicted in Fig. 2 where the Time-division duplex (TDD) RBS to RBS interference problem occurs when the relatively strong signals (unbroken line arrow) from RBS B hits RBS A while RBS A is attempting to receive the comparatively much weaker signal (broken line arrow) from the UE. This results in that the radio channel from the UE to RBS A is severely interfered by transmission from RBS B. A known method to reduce the effects of both UE to UE and RBS to RBS interference is to synchronize the TDD system so that all RBS 's transmit one or several DL frames at the same time period while the UE's transmit one or several UL frames during another time period. To allow for the finite time it takes for a radio signal to travel between two RBS 's, guard times, also called guard periods or guard intervals, are used between the DL time periods and the UL time periods. Guard times are silent periods, i.e. periods with no transmission content, arranged between the radio frames .

In Fig. 3 is depicted an example of guard periods between DL and UL frames in a TDD system to account for propagation delays and hardware switching times . The guard periods are typically given by the standard and usually the same guard periods are used in all cells.

One problem with the existing solution occurs when the guard time is not sufficiently long to protect from RBS to RBS interference between two RBS 's that are positioned far apart but still within interference distance, given the wave propagation environment. In such case there will be a systematic interference in every first DL frame.

Another problem is when an RBS drifts out of synch, or plainly looses synchronization with the surrounding system. If this situation occurs, the out-of-synch RBS will cause both RBS to RBS interference problems to surrounding RBS 's as well as UE to UE interference problems to UE 's connected to neighbouring RBS "s. The faulty RBS and its UE's, i.e. the UE's that it serves, will themselves suffer from the same effects induced by the rest of the system.

Fig. 4a depicts an example of a communication system where the guard time is not sufficiently large to protect RBS B from interference from RBS A. As can be seen the first part of each UL frame will be interfered by the signals from RBS A. Note that this interference is periodic, e.g. it occurs in the same part of the UL frame at every RBS B UL frame.

Fig. 4b illustrates an example of a situation where RBS A is out of synchronization. A wave propagation delay from

RBS A to RBS B is not included since the inclusion of such a delay would make the figure unnecessarily complex. The basic problem would however still be the same. The UL frames overlap partly with the DL frames of RBS B. This leads to UE to UE interference problems. Similarly the UL frames of RBS B are partly interfered by the DL frames of RBS A. Note that just as in Fig. 4a the interference is periodic in time.

The document "Duplexing, Resource Allocation and Inter- cell Coordination: Design Recommendations for Next Generation Systems" by Angeliki Alexiou, Lucent Technologies, 2005-01-01, discloses discussions on the periodic interference dilemma and in this context mentions the use of asymmetric allocation and block interference averaging as a possible solution.

Furthermore, WO2007/024936 discloses compensation techniques for periodic interference in TDD based transmissions, such as retransmitting disturbed or not successfully decoded data in a maximum number of retransmissions, each occurring at a lower rate.

Cited prior art however do not address how to systematically reduce the risk that a re- transmission of previously interfered data is again scheduled in an interfered frame .

SUMMARY

The present invention aims at providing a solution that alleviates at least one of the problems indicated above.

It is therefore an object of the present invention to provide methods and arrangements for scheduling retransmissions in a telecommunications system, which provide means for at least reducing the negative performance impact from periodic interference patterns. In particular it is an objective of the invention to reduce the user-data delay and the delay jitter imposed by periodic interference and thereby to improve the performance of delay sensitive applications, e.g. voice, video telephony and interactive gaming.

At least one of the above objects is achieved with a method, scheduler or a radio network node according to the appended independent claims.

Further objects and advantages are evident from the dependent claims.

A first aspect of the present invention relates to a method in a radio network node of a telecommunications system for scheduling a retransmission of radio block data that was interfered in a previous transmission, which could be a first retransmission or a retransmission of yet a previously interfered transmission or retransmission. A radio frame position for the data to be retransmitted is selected or determined to be different when compared to the position of that data in the frame of the previous transmission. This selecting or determining of a different position in the radio frame for the data is provided by determining a scheduling point of time for the retransmission in a periodic interval cycle for transmissions in the communication system, which point of time in the cycle is different as compared to the point of time in the cycle when the previous transmission was performed. The retransmission is then scheduled to be performed at that different point in time of the periodic cycle, hence providing the placing of the data in a different position in the frame.

This provides the advantage of providing an increased likelihood of avoiding that retransmitted data is again interfered in transmissions suffering from periodic interference problems, i.e. when the same frame parts are repeatedly interfered in subsequent transmissions of the same data. According to one embodiment of this aspect of the invention, the length of the periodic interval cycle for transmissions is determined depending on the radio interface configuration of the network node, such as Time Division Duplexing, Frequency Division Duplexing, or other configuration.

This provides the advantage of providing a retransmission cycle that provides scheduling diversity relative interference with the same periodicity as the radio interface frame structure. This includes the type of interference that is caused by non-perfect RBS time synchronization in a TDD system and the interferences induced by a too short guard time in a TDD system. Another advantage of this periodic interval cycle is that it provides scheduling diversity relative possible interference from broadcast information that is periodically transmitted by other cells in the same geographic area.

According to another embodiment of this aspect of the invention, a HARQ process, such as a synchronous or asynchronous HARQ process, is used for the retransmission scheduling.

This provides the advantage of avoiding explicitly signaling the retransmission radio slots to the mobile station. Rather, with synchronous HARQ the retransmission radio slots are derived by the mobile station from the description of the radio slot of the first transmission. Thus synchronous HARQ has the advantage of reducing the radio grant signaling in a network. In a set of radio systems, including LTE and WCDMA, while operating in synchronous HARQ mode, the time difference between a first transmission and the corresponding retransmission is determined by the number of HARQ processes agreed upon between the network and the mobile station.

According to a further embodiment of this aspect of the invention, in a system using synchronous HARQ, the scheduling at the different point of time in the periodic cycle is provided by setting a number of HARQ processes to a number that is not a multiple of a number of Transmission Time Intervals TTI 's used in the TDD mode of operation for the UL transmission and DL transmission period. Alternatively, the different point of time in the periodic cycle is provided by setting an uneven number of HARQ processes, such as 3 HARQ processes, for a number of transmission time intervals used per transmission period which is even, such as 2, or vice versa.

This again gives the advantage of providing scheduling diversity relative interference of the same periodicity as the transmission cycles, in this case two transmission intervals . According to yet a further embodiment of this aspect of the invention, a retransmission interval is determined as a required time length to the retransmission from the previous transmission for the retransmission to be performed at a different point of time in the periodic cycle, and wherein the retransmission is scheduled with the thus determined retransmission interval. According to a further embodiment of this aspect of the invention, the number of HARQ processes to be used for a transmission period is set in the network node based on the thus determined retransmission interval. According to a further embodiment of this aspect of the invention, the number of HARQ processes is signaled to a mobile terminal, either as part of a broadcast system information message or in dedicated control signaling, and wherein the scheduling of a retransmission uplink from the terminal is based on that number of HARQ processes.

According to a further embodiment of this aspect of the invention, scheduling information comprising at least one of the number of HARQ processes and the retransmission interval is signaled to a mobile terminal for controlling when the terminal shall perform a first uplink transmission of data and when the terminal shall perform a retransmission of that data.

According to a further embodiment of this aspect of the invention, the method is performed in a radio base station, such as a NodeB or eNodeB base station.

A second aspect of the present invention relates to a data scheduler of a radio network node adapted for scheduling a retransmission of radio block data interfered in a previous transmission in a telecommunications system. The scheduler comprises an information requesting unit arranged for requesting information on a radio interface configuration of the network node. The requested information includes a time length of a periodic transmission interval cycle of the specific node configuration, such as a TDMA interval if the node is configured for TDMA transmissions. The scheduler further comprises a determining unit arranged for determining a retransmission interval such that the transmission and retransmission, or e.g. first retransmission and second retransmission, will be performed at different points of time in said periodic interval cycle. This will provide the positioning, i.e. placing, of the data in different places in the radio frame. The scheduler furthermore comprises a data scheduling unit arranged for scheduling the retransmission according to the thus determined retransmission interval.

A third aspect of the present invention relates to a radio network node capable of scheduling a retransmission of radio block data interfered in a previous transmission in a telecommunications system, and comprising an antenna system connected to a transmitter and a receiver for communicating with a mobile station. The network node comprises a data scheduler according to that of the second aspect of the invention.

According to one embodiment of this aspect of the invention, the node is arranged configured with a set of HARQ buffers adapted for a number of HARQ processes of a transmission period that is not a multiple of a number of transmission time intervals used for the transmission period. Alternatively, the node is arranged configured with a set of HARQ buffers adapted for a number of HARQ processes of a transmission period that is uneven, such as three HARQ processes, for a number of transmission time intervals per transmission period which is even, such as two, or vice versa.

According to another embodiment of this aspect of the invention the node is a radio-base-station, such as a NodeB or eNodeB base station. Alternatively, the node may be a radio network controller, a wireless access point or any other network node in charge of scheduling transmissions and retransmissions across the radio interface .

The features described above in relation to the method according to the invention may, where applicable, also be implemented in an arrangement, such as a data scheduler or network node, according to the invention with the same advantages as described in relation to the method.

It goes without saying that the above aspects of the invention may be combined in the same embodiment. In the following, preferred embodiments of the invention will be described with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

Fig. 1 schematically illustrates UE to UE interference problems in a TDD system.

Fig. 2 in turn schematically illustrates RBS to RBS interference problems in a TDD system. Fig. 3 exemplifies guard times between DL and UL frames in a TDD system.

Fig. 4a illustrates an example of a communication system, such as a TDD system, where the guard time is not sufficiently large to protect RBS B from interference from RBS A.

Fig. 4b shows an example of an out of synch Radio-Base- Station (RBS) A.

Fig. 5a shortly illustrates a procedure for scheduling retransmissions according to one embodiment of the present invention. Fig. 5b depicts an example of a radio-network-node configuration for scheduling retransmissions according to one embodiment of the invention.

Fig. 6a shows a prior art procedure for scheduling retransmissions of radio blocks.

Fig. 6b schematically illustrates a DL scheduling for a non-synchronized RBS according to one embodiment of the present invention.

Fig. 7a exemplifies an uplink scheduling of data transmissions according to previously known techniques.

Fig. 7b illustrates an example embodiment of the present invention for UL scheduling of retransmissions.

Fig. 8 schematically illustrates a network node, here exemplified in the preferred form of a radio-base-station, adapted to perform method steps of the present invention for scheduling retransmissions.

Fig. 9 in a block diagram illustrates a procedure for scheduling retransmissions according to one embodiment of the present invention. DETAILED DESCRIPTION

In the following, various embodiments of the invention will be described.

The basic idea of the present invention is to battle the systematic, i.e. periodic, interference by time diversity in the scheduling. According to the present invention, one way of accomplishing this may be to consistently schedule retransmissions in a different part of the frame as compared to the previous transmissions of the same radio block. That is, to address the periodic interference problem by applying a systematic approach for minimizing the risk that a re-transmission of previously interfered data is again scheduled in an interfered frame, or more precisely scheduled to be placed in an interfered part of a frame. By doing this the risk that both first-, second- and third transmissions are all interfered is reduced. This is of particular importance for real-time services like Voice over IP (VoIP) and furthermore provides that the performance of best effort services increases.

Fig. 5a in a flowchart exemplifies method steps according to one embodiment of the present invention which may be applied in a network node, such as radio base station, a radio network controller, a wireless access point or any other network node in charge of scheduling transmissions and retransmissions across the radio interface, of a telecommunications system for scheduling a retransmission of radio block data that was interfered in a previous transmission, which could be a first transmission or a retransmission of yet a previously interfered transmission or retransmission. Here, a first method step 510 involves determining or selecting a radio frame position for the data to be retransmitted to be different when compared to the position of that data in the frame of the previous transmission, which was interfered. The determining of a different position in the radio frame for the data may here be provided by determining a different point of time in a periodic interval cycle for transmissions in the communication system for scheduling the retransmission as compared to the point of time in the cycle when the previous transmission was performed. The periodic interval cycle for transmissions may for example be a TDMA interval, or any other periodic interval cycle for transmission depending on the utilized transmission method, FDD or TDD or other transmission method. In step 520 the retransmission is thereafter scheduled to be performed at that different point of time in the periodic interval cycle as determined, hence providing the placing of the data in a different position in the frame reducing the risk of the same data being repeatedly interfered.

A simplified block diagram of a network node here exemplified by a base station, such as an eNodeB, adapted to operate in accordance with at least the embodiments described above, will now be described with reference to Fig. 5b. It is to be understood that, for simplicity reasons, units which are not necessary for the understanding of the claimed invention have been omitted. It is also to be understood that all units mentioned in this document are to be interpreted as exemplified logical units, which may be implemented as single units or in combination with other units in any of various possible ways .

The eNodeB 600 comprises a scheduler 601, adapted to administrate scheduling between the eNodeB and at least one mobile terminal i.e. UE, represented here by UE 700. The scheduler 601, which typically includes separate uplink and downlink scheduling functions (not shown) , comprises an information requesting unit 602, adapted to request or interrogate on the radio interface configuration of the eNodeB 600, especially on the periodic interval cycle of the radio interface configuration thereof, and the scheduler also comprises a determining unit 603, adapted to determine a retransmission interval such that the transmission and its retransmission (s) may be performed at different times in the periodic interval cycle, to avoid periodic interferences. The scheduler 601 further comprises a scheduling unit 604 adapted to schedule the retransmission according to the determined retransmission interval for avoiding or at least minimizing the risk that data for retransmission is again interfered. A transceiver 605 of the eNodeB comprising a transmitting unit 606 and a receiving unit 607 provides for the scheduling information, such as the uplink retransmission interval, to be transmitted to the UE in question and for receiving ACK/NACK messages in return from the UE. A scheduler, such as the one described above in connection with Fig. 5b, adapted for implementing the invention as claimed may be provided knowledge of the periodic cycle of the radio interface of a network node either by such configuration data being downloaded from an O&M (operation and maintenance) system or through a command file compiled by an operator. Alternatively, the network node may be preconfigured with the periodic cycle structure.

Downlink Scheduling

Figure 4b exemplifies an interference pattern for the downlink (DL) in respect of RBS B. An even more detailed illustration of this interference pattern is shown in Figs. 6a and 6b respectively, where the DL frames are periodically interfered, as illustrated with grid parts, by the UE 's belonging to the RBS out of synch, i.e. RBS A of Figure 4b.

Each labeled block of Figures 6a and 6b correspond to one HARQ (Hybrid Automatic Repeat Request) process, e.g. radio block, with radio block identity, e.g. letter, and transmission sequence, e.g. number. In this example the letters A-F are used for defining radio block identity and numbers 1-3 for defining transmission sequence of the respective radio blocks. The HARQ processes are scheduled consecutively in time, i.e., in consecutive TTI 's (Transmission Time Intervals) . Figures 6a and 6b both show the second transmission i.e. the first retransmission C2 of transport block C. ARQ is a known error control method for data transmissions which uses acknowledgments, negative acknowledgements, i.e. NACK, timeouts and retransmissions to achieve reliable data transmission. An acknowledgment, i.e. ACK, is a message sent by the receiver to the transmitter to indicate that it has correctly received a data frame. Similarly a non-acknowledgement is a message sent by the receiver to the transmitter to indicate that an expected data frame has not been correctly received. At reception of an acknowledgement message the transmitter considers the corresponding data frame to be correctly received by the receiver and erases that data frame from the transmitter send buffer. In contrast, at the reception of a non-acknowledgement message the transmitter concludes that the corresponding data frame has not been correctly received by the receiver. The reception of a non- acknowledgement message typically triggers the transmitting side to re-send the corresponding data frame. A timeout is a point in time after the sender sends the data frame; if the sender does not receive an acknowledgment before the timeout the corresponding data frame is typically re-transmitted. By using acknowledge messages in combination with non-acknowledgement messages and/or time-out procedures each data frame is transmitted and retransmitted until the sender receives an acknowledgment or exceeds a predefined number of retransmissions .

Hybrid ARQ (HARQ) is a variation of ARQ (Automatic Repeat reQuest) that has better performance, particularly over wireless channels, at the cost of increased implementation complexity. In addition to the ARQ procedures described above the HARQ technology combines the information in each transmission and retransmission of a particular data frame. This means that the receiver stores whatever information content that was received in each reception of the data frame (even though that information was not sufficient to decode the data frame correctly) . At the reception of each re-transmission of the data frame the receiver then combines the re-transmission information with the information stored from each previous transmission of that particular data frame. This means that for each retransmission more information is added to the already received information thus increasing the chances of correctly decoding the data frame for each re-transmission.

The present invention can be applied both to systems using ARQ and for systems using HARQ. Though herein described primarily for a system using the HARQ technique it should be obvious for anyone skilled in the art how to generalize the inventive method to include a system using ARQ.

A useful and known concept when describing transmissions and retransmissions in a communication system is a so- called HARQ process. A HARQ process is the set of transmissions and retransmissions relating to the transfer of a single data frame across a communication link. When a

HARQ process has succeeded in transmitting a data frame

(typically finished by the reception of an acknowledgement message) the HARQ process can be used to transmit a second data frame. In state of the art technologies using HARQ processes more than one HARQ process operates in parallel. This has the benefit that while one HARQ buffer does not use the communication link, e.g. while waiting for an acknowledgement message or time out, a second HARQ process can transmit its data frame. Thus the usage of a multitude of HARQ processes increases the ability to fully use the link capacity.

A HARQ process can work in either synchronous mode or in asynchronous mode. In synchronous mode the timing of a possible re- transmission is uniquely determined by the timing of the first transmission in that HARQ process. In asynchronous mode the retransmission can occur at any time quite independent of the timing of the first transmission. The advantage of the asynchronous HARQ mode is the flexibility to schedule transmissions and retransmissions independently which increases the system's capability to control latency and quality of service requirements . The advantage of the synchronous mode is that it requires less signaling bandwidth across the link since the exact timing of the retransmission needs not to be signaled to the receiver .

The methodology of the present invention is applicable with any type of HARQ process used for retransmission of erroneously received data. Also, as already mentioned, an ARQ process could be used with the method of the present invention for retransmission of such data depending on compatibility with the hardware configuration of and transmission method used in the communication system in which the concept of the present invention is applied.

In Figure 6a a periodic scheduling according to prior art methodology is applied. Here is illustrated that the radio block C is consistently transmitted during the interfered period. This consistent scheduling may be due to the scheduling implementation. After the third attempt radio block C has still not been decoded and the HARQ process is terminated. The maximum number of HARQ transmissions, i.e., the number of transmissions of the same transport block that are attempted before the HARQ processes terminates is configurable. In this example, it has been assumed to be set to 3. If RLC (Radio Link Control) has been configured in so-called acknowledged mode, an RLC retransmission will then be triggered, so the data would in such case not be lost. Still it would have suffered a substantial time delay with negative impact on e.g. a VoIP service, or the like service heavily depending on the data stream being more or less uninterrupted for the QoS (Quality of Service) to be acceptable. In case RLC instead was configured in unacknowledged mode, data would be lost and could potentially only be recovered by higher- layer protocols such as TCP (Transmission Control Protocol) . It is, however, not likely that a VoIP service uses such reliable transport protocols, in which case the loss of data would impact the voice quality.

An example where the drawbacks of the previously known solutions for scheduling retransmissions could cause real trouble would be if so-called persistent scheduling was used in which case both transmissions and retransmissions risk be systematically scheduled in one part of the frame. In the prior art example of above this would mean that only the C frames belong to the VoIP stream while the other frames belong to a best effort stream, possibly even to other users . The result would be that stream C would be interfered all the time. In the worst case, this would lead to a dropped voice call.

Figure 6b provides an alternative illustration of the example embodiment of Fig. 5a describing a method for avoiding or at least minimizing the risk of the same transmission data being consistently interfered. The solution according to this embodiment of the invention assures that retransmissions are systematically placed in different parts of the frame than were the previous transmissions. Accordingly, as described above, a radio frame position for the data to be retransmitted is selected or determined to be different when compared to the data position in the frame of the previous transmission. This selecting or determining of a different position in the radio frame for the data is accomplished by determining a point of time for said retransmission in a periodic interval cycle for transmissions in the communication system, which point of time in the cycle is different as compared to the point of time in the cycle when the previous transmission was performed. The retransmission is then scheduled to be performed at that different point in time of the periodic cycle to hence provide the placing of the data in a different position in the frame. The elapsed time from the previous transmission and the determined time for the retransmission constitutes a retransmission interval Through the method of the invention, for example according to the Fig. 5a and 6b embodiment in the downlink, the above described problems relating to prior art solutions may be avoided completely or at least reduced significantly. Assumedly the first scheduling of C, i.e. block Cl, is scheduled during the interfered period, i.e. grid part, but due to the scheduling strategy of putting retransmission elsewhere in the frame according to the inventive solution already the first retransmission, i.e. C2 , will be placed in a non- interfered part of the frame leading to a successful decoding. In this way all the data streams share the good and bad parts of the frames, no stream is hurt consistently and every data stream gets its share of the average radio link quality of the DL frames. Uplink Scheduling

It should be noted that a corresponding methodology of distributing retransmissions to less interfered Transmission Time Intervals (TTI's) as described above for the DL scheduling procedure according to the invention may be applied also to the UL scheduling with the corresponding effect and advantages.

One embodiment of the present invention for uplink scheduling, which could be applied also to the downlink scheduling with the corresponding effect and advantages, comprises utilizing the synchronized time difference between a first- and a second transmission in the uplink, so-called "Synchronous HARQ", which can be used in an UMTS communication system such as LTE as well as in systems utilizing other standards for data- and telecommunication.

According to one embodiment of the present invention for scheduling of retransmissions, the periodic interference problem can be solved, or at least the risk for such interference greatly reduced, by setting the synchronized HARQ so that consecutive transmissions of the same data get a different placement, i.e. position, in the radio frame. Furthermore, due to the synchronous HARQ, this may be achieved by controlling the number of HARQ processes for the uplink with which the system is configured as illustrated in Fig. 7b. Figs. 7a and 7b respectively show a UE communicating with an RBS in a TDD cell that has been configured with a 3+2 asymmetry. That is each downlink period consists of 3 downlink TTI's 10 and is followed by an uplink period consisting of 2 uplink TTI's 20. The first TTI in an uplink period is especially interfered, so-called periodic interference, in Figs 7a and 7b. In Fig.7a, exemplifying a prior art solution for the UL, the number of HARQ processes is set to a multiple of the number of uplink TTI 's per UL period providing that if data of a first transmission is interfered also its retransmission is interfered. According to this example the number of HARQ processes 1-4 has been configured to 4. The UE is here assumed to have enough data to fill 2 HARQ processes. The first transmission Tl is sent in the first HARQ process 1. Being the first TTI in the uplink period, this TTI is interfered, which is illustrated with the radio wave or flash symbol. As is evident from the figure, the retransmission Rl of the previous transmission Tl will also be interfered. On the other hand, the data placed in HARQ process 2 here exemplified with T2 and R2 will not be interfered. The reason for this is that the number of synchronous HARQ processes has been chosen as a multiple of the number of uplink TTIs that are contained in an UL period.

Figure 7b, as an example, illustrates one embodiment of the invention, in which the number of HARQ processes is set or configured not a multiple of the number of uplink TTIs per UL period. Thereby is achieved that data to be retransmitted, i.e. not correctly received or decoded data from a previous transmission, is scheduled in another part of the frame than it was in the frame of the previous transmission, i.e. scheduled at a different time in the periodic transmission cycle than the previous transmission of the same data. The solution according to this example is to choose i.e. set or configure the number of synchronous HARQ processes 1-3 to 3 or to a number that is not a multiple of 2, which is the number of TTIs per UL period in this example. Thereby, although the first transmission Tl is in an interfered part of the frame i.e. in an interfered TTI, its retransmission Rl will consequently be placed in a TTI, i.e. frame part, that is most likely not interfered.

The following two situations may be contemplated: (1) : The periodic interference is known by the base station scheduler, e.g. RBS B scheduler, or

(2) : The periodic interference is not known by the scheduler.

In situation (1) more clever things could in principle be done as an alternative to putting transmissions and retransmission in different positions in the frame. So could e.g. the RBS schedule selected UE " s away from the interference while scheduling others that are not so impacted on the interfered part of the frame. Hence it is in particular in situation (2) the solution according to the present invention would be of most use since there is no harm done applying the inventive scheduling also in the absence of the interferer. It would be advantageous to apply the inventive scheduling e.g. in a TDD deployment of two different, non-synchronized TDD systems in neighboring spectra. The inventive scheduling could also be adopted as a general scheduling principle in any communication system to minimize in particular UE to UE interference impact, should any part of the system fall out of synchronization.

Furthermore, even if believed to be of most use in a TDD system, as periodic interference patterns are expected to be more common in TDD systems, the principles of the present invention are also applicable to a Frequency- division duplex (FDD) system as well as other systems for wireless communication. As described above, for both uplink and downlink transmission, the inventive method as claimed may be used both if dynamic synchronous retransmission scheduling is used and if asynchronous scheduling is applied. With asynchronous retransmission, the base station dynamically determines when and where a retransmission shall occur. Here the base station uses the method of the invention, preferably but not necessarily implemented through an algorithm or software forming part of base station control logic, to decide where and when to perform a retransmission. If synchronous retransmission is used, either in the uplink, in the downlink or in both, the inventive solution is applied in the definition of the relationship between a first transmission and a subsequent retransmission of a radio block. The solution according to the invention as claimed can be used both if the relationship between transmission and retransmissions is set explicitly by signaling and in systems where the relationship is set indirectly, e.g. by the number of HARQ processes used. It should also be understood that the inventive solution can be applied to systems using both dynamic and persistent scheduling of radio blocks .

Fig. 8 shows, as an example, a network node here in the form of a radio-base-station (RBS) 100, adapted for performing a systematic scheduling of retransmissions in non-disturbed frames according to method steps of the present invention. The node, which may alternatively be a radio network controller, a wireless access point or any other network node in charge of scheduling transmissions and retransmissions across the radio interface, to this end comprises a Digital Unit 110 and a Radio Unit 120. The node is connected to an antenna system 130, such as a MIMO

(Multiple Input Multiple Output) , antenna system or SISO- (Single Input Single Output) , SIMO- (Single Input Multiple Output) or MISO- (Multiple Input Single Output) antenna system. The Radio Unit includes a transmitter 150 that is connected to the antenna system 130 via the transmission filter 160. The Radio Unit further has a receiver 180 that connects to the antenna system 130 through the receive filter 170. The Digital Unit 110 comprises a receive MAC buffer 190 which in turn hosts one or more HARQ receive data buffers 200. Further residing in the Digital Unit 110 is the uplink RLC buffer 210 and the radio data scheduler 140, which comprises both an uplink- and downlink scheduler. Also in the digital unit 110 is the radio interface configuration controller 220.

A possible communication scheme involving the inventive method for scheduling retransmissions of radio data in the uplink is illustrated as an example in Fig. 9. Although this example describes the method in a radio base station, the method is applicable also to other radio network nodes, such as a radio network controller, a wireless access point or any other network node in charge of scheduling transmissions and retransmissions across the radio interface. The methodology of the Fig. 9 example described below may be applied also to the downlink scheduling with the corresponding effect and advantages. The method steps of Fig. 9 may be performed through a network node, such as an eNodeB, as that of Fig. 5b or 8 communicating with a mobile station/terminal/UE, or any other NW node or base station adapted for performing the method of the invention.

In step 901 the base station data scheduler 140 interrogates the radio interface configuration controller

220 for information about the prevailing configuration of the radio interface, for example, the TDD configuration. In step 902 the radio interface configuration controller 220 responds by sending information to the data scheduler 140 about which time frames are configured for uplink transmission and which are configured for downlink transmission and for example also about the duration in time of the periodic transmission interval cycle of the particular configuration.

In step 903 the data scheduler 140 uses the information about the radio interface configuration to determine a set of retransmission intervals, i.e. time from first data transmission to time of data retransmission, such that transmission and retransmissions will occur at different times in the periodic TDD interval cycle and the transmission data thereby be placed in a different position in the radio frame for every retransmission of the same data.

Based on the retransmission intervals derived in step 903 above, the scheduler then in step 904 sends the corresponding scheduling information via the Radio Unit 120 and antenna system 130 across the radio interface to the mobile station. The scheduling information either explicitly contains information about when the mobile station shall perform a first transmission of data and when the mobile station shall perform retransmission of erroneous data units. Alternatively, in the case of synchronous uplink HARQ retransmissions, it is sufficient to send to the mobile station information about the number of HARQ buffers 200 the radio base station uses plus continuous information about when to send data and about which data has been correctly received. With synchronous HARQ retransmission the number of HARQ buffers in the RBS implicitly determines when the mobile station shall do a retransmission of an erroneous data unit. In this case the data scheduler uses the information about the appropriate retransmission interval to set the number of HARQ buffers accordingly.

The mobile station thereafter, in step 905, uses the scheduling information to transmit data across the radio interface via the antenna system 130 and radio unit 120 to the radio base station MAC buffer 190 and HARQ buffers 200.

In step 906 the mobile station then retransmits data units that are not acknowledged by the radio base station, further using the scheduling information from the radio base station in combination with knowledge of the number of HARQ buffers 200.

As data units are correctly received in the base station HARQ buffer 200, possibly after one or more retransmissions, the data units from the mobile station are in step 907 forwarded to the base station RLC buffer 210.

In step 908 the data in the RLC buffer is processed and possibly deciphered after which the corresponding user data is transmitted from the radio base station to appropriate other node in the system.

The solution according to the present invention has the advantage of reducing the negative impact of radio-base- stations RBS 's out of synch and of RBS to RBS interference in situations where the guard period is not sufficiently large. The inventive solution may, for example, be very useful in deployments of non- synchronized TDD networks in the same geographic area in neighboring spectra.

The method steps of the invention described herein may be implemented by software executed by a processor in a network node, such as a radio base station also called NodeB or eNodeB, and/or user terminal, sometimes called user equipment, mobile terminal or mobile station.

Any examples and terminology relating to 3GPP LTE standard being used herein should not be seen as limiting the scope of the invention, the methodology of which in principle can be applied to any communication system.

The described subject matter is of course not limited to the above described and in the drawings shown embodiments, but can be modified within the scope of the enclosed claims.

Claims

1. A method in a radio network node (100, 600) of a telecommunications system for scheduling a retransmission of radio block data interfered in a previous transmission c h a r a c t e r i z e d by determining (510) a radio frame position for the data of said retransmission which is different as compared to the frame position of said previous transmission by determining a point of time in a periodic transmission interval cycle of said communication system for scheduling said retransmission, which point of time in said cycle is different as compared to the point of time in said cycle of said previous transmission, and - scheduling (520) said retransmission to be performed at said different point of time.

2. A method according to claim 1, wherein the length of said periodic interval cycle for transmissions is determined depending on the radio interface configuration of said network node, such as Time Division Duplexing, Frequency Division Duplexing, or other configuration.

3. A method according to claim 1 or 2 wherein a HARQ process, such as a synchronous or asynchronous HARQ process, is used for said retransmission scheduling.

4. A method according to any one of the previous claims wherein said scheduling at said different point of time in said periodic cycle is provided by setting a number of HARQ processes to a number that is not a multiple of a number of Transmission Time Intervals TTI's used per transmission period.

5. A method according to any one of claims 1-3 wherein said scheduling at said different point in time in said periodic cycle is provided by setting an uneven number of HARQ processes, such as 3 HARQ processes, for a number of transmission time ntervals used per transmission period which is even, such as 2, or vice versa.

6. A method according to claim 1, wherein a retransmission interval is determined as a required time length to said retransmission from said previous transmission for said retransmission to be performed at a different point of time in said periodic cycle, and wherein said retransmission is scheduled with said retransmission interval.

7. A method according to claim 6, wherein a number of HARQ processes to be used for a transmission period is set in the network node based on the determined retransmission interval.

8. A method according to claim 7 wherein said number of HARQ processes is signaled to a mobile terminal, and wherein the scheduling of a retransmission uplink from said terminal is based on said number of HARQ processes.

9. A method according to any one of the previous claims, wherein scheduling information comprising at least one of the number of HARQ processes and the retransmission interval is signaled to a mobile terminal for controlling when the terminal shall perform a first uplink transmission of data and when said terminal shall perform a retransmission of said data .

10. A method according to any one of the previous claims wherein said method is performed in a radio base station, such as a NodeB or eNodeB base station.

11. A data scheduler (140, 601) of a radio network node (100, 600) adapted for scheduling a retransmission of radio block data interfered in a previous transmission in a telecommunications system, c h a r a c t e r i z e d by:

- an information requesting unit (602) arranged for requesting information on a radio interface configuration of said base station comprising a time length of a periodic transmission interval cycle of said configuration,

- a determining unit (603) arranged for determining a retransmission interval such that said transmission and retransmission will be performed at different points of time in said periodic interval cycle, and - a data scheduling unit (604) arranged for scheduling the retransmission according to said retransmission interval .

12. A radio network node (100, 600) capable of scheduling a retransmission of radio block data interfered in a previous transmission in a telecommunications system, said network node comprising an antenna system (130) connected to a transmitter (150, 606) and a receiver (180, 607) for communicating with a mobile station (700), c h a r a c t e r i z e d in that said node comprises a data scheduler (140, 601) according to claim 11.

13. A network node according to claim 12, wherein said node is arranged configured with a set of HARQ buffers (200) adapted for a number of HARQ processes of a transmission period that is not a multiple of a number of transmission time intervals used per transmission period.

14. A network node according to claim 12, wherein said node is arranged configured with a set of HARQ buffers

(200) adapted for a number of HARQ processes of a transmission period that is uneven, such as 3 HARQ processes, for a number of transmission time intervals per transmission period which is even, such as 2, or vice versa.

15. A network node according to any one of claims 12- 14, wherein said node is a radio-base-station, such as a NodeB or eNodeB base station, a radio network controller, a wireless access point or any other network node in charge of scheduling transmissions and retransmissions across the radio interface.