WO2010031438A1

WO2010031438A1 - Network element and method of operating a network element

Info

Publication number: WO2010031438A1
Application number: PCT/EP2008/062504
Authority: WO
Inventors: Bernhard Raaf; Wolfgang Zirwas
Original assignee: Nokia Siemens Networks Oy
Priority date: 2008-09-19
Filing date: 2008-09-19
Publication date: 2010-03-25

Abstract

A network element is provided wherein the network element is adapted to switch between a first state and a second state different to the first state during a subframe of a communication in the communication network.

Description

DESCRIPTION

Network element and method of operating a network element

Field of invention

The present invention relates to the field of network elements, in particular to network elements of a cellular commu- nication network. Furthermore, the invention relates to a method of operating a network element.

Art Background

For cellular systems like LTE multihop relay nodes are considered for range extension and capability enhancement, e.g. by signal to noise Ratio (SNR) improvement. According to some proposals relaying networks seem to be an option for broad- band wireless data networks like LTE, e.g. for LTE R9 or LTE- A.

For frequency division duplex (FDD) systems an additional time division multiplexing (TDM) structure, for example on subframe basis, may be a suitable relying protocol on layer 2, as it allows for orthogonal transmissions on nodeB - relaying node (NB-RN) and RN - user equipment (RN-UE) links. In the resulting duplex scheme the RN is either connected to the NB or to its attached UEs. In the first case the RN provides UE functionality to the NB and in the second case NB functionality to the UEs.

The basic scheme of such a band structure of a relaying network is schematically depicted in Fig. 1. In the downlink band 101 a relay node (RN) transmits signals to a user equipment (UE) in a first subframe 102 and a nodeB (NB) will transmit signals to the UE or the RN in a second subframe 103. In the uplink band 104 the RN receives signals from the UE in one subframe 105 and the NB will receive signals from the UE or the RN in the next subframe 106. Between the downlink band 101 and the uplink band 104 a duplex gap 107 is present which separates the both in frequencies, which separation in frequencies is schematically depicted by coordinate system 108.

However, there may be a need for a network element and a method of operating a network element which allows an improved performance, in particular which may enable a relaying solution while using LTE R8 conform user equipments.

Summary of the Invention

This need may be met by a network element, a communication network, a method of operating a network element, a program element, and a computer-readable medium according to the independent claims. Further embodiments of the present invention are described by the dependent claims.

According to an exemplary aspect of the invention a network element is provided wherein the network element is adapted to switch between a first state and a second state different to the first state during a subframe of a communication in the communication network.

The term "subframe" may particularly denote a part of a frame defining a specific time period of a communication. The lengths and layout of such frames are typically defined in the context of different networks, e.g. in the context of LTE networks. A subframe may comprise several symbols while several subframes are included in a frame. In specific cases a subframe may correspond to a resource block.

According to an exemplary aspect a communication network is provided which comprises at least two network elements according to an exemplary aspect of the invention wherein one of the at least two network elements is in a transmitting state while the other one of the at least two network elements is in a silent state.

According to an exemplary aspect a method of operating a network element is provided, wherein the method comprises re- ceiving signals during the second state, and transmitting signals during the first state. In particular, the received signals may be data signals which may be received during a predetermined sequence of symbols while the transmitted signals may be reference signals which are transmitted during the rest of the symbols of the subframe.

According to an exemplary aspect a program element is provided, which, when being executed by a processor, is adapted to control or carry out a method according to an exemplary aspect of the invention.

According to an exemplary embodiment a computer-readable medium, in which a computer program is stored which, when being executed by a processor, is adapted to control or carry out a method according to an exemplary aspect of the invention.

By providing a network element which is able to switch between two states during a single subframe it may be possible to avoid problems arising due to the fact that a network ele- ment may not be in a transmitting state and a receiving state at the same time with respect to one specific communication, e.g. in a specific frequency band i.e. uplink or downlink. That is, in general a relay node, for example, may at one point in time only be in receiving mode or in transmitting mode but not in both. Otherwise the relay node may not be able to receive weak signals correctly, given that it transmits on the same frequency and this transmission is much stronger at the location of the relay node than the received signal, e.g. several Watts compared to 10^~33 Watts or e.g. 30 dBm transmit power and -100 dBm reception power. The strong transmit signal may then make it virtually impossible to receive the weak reception signal. However, a switching of a relay node, for example, may enable that user equipments are used which demands the presence of reference signals in each subframe in order to perform a channel estimation. Thus, it may be possible to use simple known user equipments even in a relaying communication network which averages reference signals of subsequent subframes in order to perform the cannel estimation. Such an averaging (or at higher speeds interpolation) may enable the user equipment to make a better estimate of the channels impulse response and thus also may allow a more accurate detection of the transmitted data. However, if some reference signals have not been sent then a deteriora- tion of the reception may be caused.

A gist of an exemplary aspect of the invention may be seen in the providing of network elements which are able to switch between two different states during a single subframe which, according to communication protocols, are used either for transmitting or receiving. For example, during a subframe which is, according to a used protocol or communication scheme, used for transmitting some symbols of the frame may be used to perform a receiving action. Next, further exemplary embodiments of the network element are described. However, these embodiments also apply to the communication network, the method of operating a network ele- ment, the program element, and the computer-readable medium.

According to another exemplary embodiment of the network element the subframe comprises a number of symbols which is in accordance with a communication protocol. In particular, the communication protocol may be an LTE protocol, i.e. a protocol which ensures an LTE conform transmission. For example, each subframe may comprise a predetermined number of symbols, e.g. 14 in case of a LTE R8 procedure for example, and predetermined symbols or time positions during the subframe may be used for transmitting or broadcasting reference signals, e.g. symbol 2, 6, 9 and 13. However, all other time positions may also be used for transmitting the reference signals.

According to another exemplary embodiment of the network ele- ment the network element is a relay node comprising a transmitting unit and a receiving unit, wherein the first state is a transmitting state in which the relay node transmits signals by using the transmitting unit, and wherein the second state is a receiving state in which the relay node receives signals by using the receiving unit.

That is, the relay node may be adapted to switch between receiving and transmitting mode in one single subframe while using a frequency division duplex procedure. For example, during the time periods associated to some symbols of a single subframe the relay node may be in receiving mode while during the time periods associated with the other symbols of the single subframe the relay node may be in transmitting mode. In case of an LTE conform transmission each subframe may comprise 14 symbols and during the time periods of symbol 2, 6, 9 and 13 the relay node may be in transmitting mode while during the time periods associated to the other symbols the relay node may be in receiving mode, for example. This example may be in particular useful for the downlink band of a communication while on the uplink band of the communication the conditions may be vice versa.

According to another exemplary embodiment of the network ele- ment during the transmitting state reference signals are transmitted and/or wherein during the receiving state data signals are received.

The term "reference signals" may particularly denote signals which are transmitted in order to enable a channel estimation based on the received quality of the reference signals.

According to another exemplary embodiment of the network element signals received in the receiving state comprises a guard time in the beginning and/or the end of the signal. In particular, the guard time may be chosen in such a way that the effects of a switching between different states on the timing are compensated for. For example, the guard time may be increased to compensate for these effects, e.g. to take into account the switching times needed to switch between the states and thus may influence the timing and/or the amount of data which is transmittable .

That is, an approach similar to SC-FDMA (subcarrier frequency division multiplexing access) on a link to the relay node may be possible, in particular for cases where switching time may be of relevance. In general SC-FDMA may generate a signal similar to a time domain multiple access (TDMA) signal but may embed it into an OFDMA like signal. For such an SC-FDMA signal it may be possible to provide some guard time in the beginning and the end to take into account the switching times. In this context the term "guard time" may particularly denote a time period in which no information is transmitted and which may be used in order to decrease the impact of the switching time. The guard time may correspond to a cyclic prefix which may be elongated compared to a common cyclic prefix and which may be associated with a single carrier data block. The rest of the time of the subframe or the rest of the symbol of the subframe may be used to carry information. In this way it may be possible to take into account the limited available reception time at the RN without compromising in any way the reception of concurrently receiving UEs.

According to another exemplary embodiment of the network element the received signal is an SC-FDMA signal transformed into frequency domain.

In case of the use of a guard time there may still be some inter subcarrier interference from signals transmitted to concurrently receiving UEs to signals transmitted to the RN. This inter subcarrier interference may be partially taken care of by such a back transforming of the SC-FDMA signals into the frequency domain. In this case carriers close to the borders of the RN transmission may be affected which may be taken into account by, e.g. loading less information to these carriers, e.g. use 4 QAM and 16 QAM instead of 16 QAM and 64 QAM, respectively.

According to another exemplary embodiment of the network element the network element is a base station comprising a transmitting unit, wherein the first state is a transmitting state in which the base station transmits signals on a subcarrier by using the transmitting unit and wherein the second state is a silent state in which the base station does not transmit signals on the subcarrier. That is, the base station or nodeB may be adapted to switch between transmitting and silent mode in one single subframe while using a frequency division duplex procedure. For example, during the time periods associated to some symbols of a single subframe the base station may be in transmitting mode while during the time periods associated with the other symbols of the single subframe the base station may be in silent mode. In case of an LTE conform transmission each subframe may comprise 14 symbols and during the time periods of symbol 2, 6, 9 and 13 the base station may be in silent mode while during the time periods associated to the other symbols the base station may be in transmitting mode, for example. This example may be in particular useful for the downlink band of a communication. In particular, the silent state of the base station is a state in which it does not transmit data and may further not receive data. In particular, in the silent state the base station may be silent only on one or some specific subcarri- ers or may be silent on all subcarriers .

According to another exemplary embodiment of the network element the base station is adapted to communicate with at least two network elements, wherein the base station is further adapted to be in the silent state for the communication with a first one of the at least two network elements, and to be in the transmitting state for the communication with a second one of the at least two network elements.

According to another exemplary embodiment of the network element the base station is adapted to use power which is not used by the base station during the silent state in the communication with the first one of the at least two network elements to increase the power level of the transmission of signals to the second one of the at least two network elements .

That is, a shifting of the available power resources of the base station or another network element, e.g. a relay node, may be performed in such a way that power which is saved since the base station is in the silent mode with respect to a first network element, e.g. a first relay node, is used in order to increase the power level of a transmission to a sec- ond network element which is performed at the same time.

According to another exemplary embodiment of the network element in the first state a coding of the signals is increased compared to a subframe in which no switching is performed be- tween the first state and the second state. An increased coding or a strong coding may particularly denote the fact that the redundancy of the coding is increased, i.e. that a lower coding rate is used. In particular, the first state may be a state of the base station or nodeB in which it transmits sig- nals, e.g. data or reference signals, to a relay node of the communication network.

Such a stronger coding may be a suitable measure to compensate for some degradation which may arise due to the switch- ing during a subframe. Such degradation may be introduced in case the switching cannot be done fast enough or in case of uplink connection. In such uplink connection the transmission may have to start earlier due to time advance possibly leading to a decrease of time for transmission from NB to RN which may be compensated by redefining the transmission scheme in the symbols that are used for the communication to the RN as described above, i.e. by performing a stronger coding. If the channel to the RN is flat, then the base station or NB may know the interference that is generated from par- tially received other subcarriers and may compensate this interference already in advance (similar to the so-called dirty paper precoding where some subcarriers may need to be reserved for cancellation purpose and may not be available to carry data however, but the remaining carriers may not suffer from inter carrier interference) .

Alternatively, the impact of the switching in a single sub- frame may be cared for by reducing a symbol duration in the subframe, i.e. the subcarrier spacing is enlarged, or by combining two symbols into a single one while a carrier spacing may be used accordingly. In case of the combining of two symbols into a single one it may not be possible to transmit data to other UEs at the same time however, because it may be advantageous to use a single FFT to generate the signals, in this case the relevant subframe may only be used for base station - RN communication, but it may still be possible to communicate to several RNs. The use of a single FFT however may induce significant interference compared to the case when two FFTs with different symbol lengths are used for transmission .

It can be noted, that if the switching takes more time than the time provided by the cyclic prefix, then the RN may not be able to receive an entire OFDM symbol and consequently some information contained in that OFDM symbol may be lost. This is in particular relevant for the information that is supposed to be transmitted from the Base station to the RN. In order to allow the RN still to receive the desired infor- mation, according to a further embodiment, the information for the RN may be concentrated within the OFDM symbol to that part in time of the OFDM symbol, where the RN is able to receive. One simple way to achieve this may be to reduce the time duration of the OFDM symbol accordingly, however this may be inappropriate if at the same time also information is to be transmitted within the same OFDM symbol to other nodes, in particular to legacy UEs, because the latter are not prepared to handle a shorter OFDM symbol and if two OFDM symbols are transmitted with incompatible duration at the same time there may be cross interference between the two.

It may be an objective of a further embodiment to achieve such a concentration in time without impacting the reception of legacy R8 terminal scheduled in the same OFDM symbol. According to one exemplary embodiment of the invention, therefore, the concentration in time may be obtained by applying a technique similar to the modulation used in LTE for uplink transmission and called SC-FDMA (Single Carrier Frequency Do- main Multiple Access) and well known in the art. Shortly summarizing this modulation scheme, the data to be transmitted are first processed by a DFT (Discrete Fourier Transformation) before they are input on the desired subcarriers to the IFFT that forms the OFDM modulation. Basically this will cre- ate a single carrier time domain signal, but due to the DFT - IFFT combination, the signal can be placed on any block of subcarriers and the bandwidth of the signal and the frequency area it occupies can be selected/determined in this way. Further more, such a signal can coexist with other SC-FDMA sig- nals on other subcarriers and also with a standard OFDM signal on other subcarriers. Furthermore, because the IFFT is preceded by a DFT, the data are basically transmitted in a time domain fashion. Therefore, if some of the data put into the DFT are replaced by 0 (i.e. no data, or more generally speaking some of the data input to the DFT are not data intended to be used by the RN, but possibly for another one as explained in another embodiment) then the remaining data which are intended for the RN are concentrated in time. This may be exactly what is intended in order to make sure that the information content intended for the RN is concentrated in that time area, where the RN is able to receive. It should be noted that also other variants of signal generation schemes may be used to obtain a similar effect e.g. by using other transform functions than Fourier Transform.

Next, further exemplary embodiments of the communication network are described. However, these embodiments also apply to the network element, method of operating a network element, the program element, and the computer-readable medium.

According to another exemplary embodiment the communication network further comprises a user equipment, wherein the relay node transmits reference signals to the user equipment in the transmitting state.

It should be noted that while the above description mainly relates to distinct states of a first network element, e.g. a base station, these distinct states relates to a communica- tion link to a single other network element, while for a communication to another network element the first network element may be in another state. For example, a base station may be in the transmitting state with respect to a specific UE or RN while with respect to another UE or RN it may be in a si- lent or receiving state. Of course the communication network may comprise a plurality of base stations, relay nodes and/or user equipments, wherein each or at least some of these elements are network elements according to an exemplary aspect of the invention.

Summarizing an exemplary aspect of the invention may be seen in providing a relay node which is adapted to enable the use of common UEs demanding the presence of reference signals in each subframe. For enabling the use of such UEs the relay node may be adapted to perform a switching between transmitting mode and receiving mode during a single subframe. Thus, it may be possible to omit the necessity to provide a relay node which can simultaneously receive and transmit signals which will lead to a feedback loop over the air. Furthermore, in case of frequency division duplexing (FDD) systems the frequency duplex bands for up- and downlink (UL/DL) have to change when the RN is switching between UE and NB functionality. Thus, in all subframes where the RN is listening to the NB messages it will not be able to broadcast the scattered reference signals for channel estimation as defined in LTE R8. Thus, the simple switching on a pure subframe basis may be simple but not appropriate while a switching performed during a single subframe may omit at least some of the prob- lems which would arise when switching is performed only on subframes borders, i.e. at the beginning and end of a subframe. In particular, UEs using techniques averaging reference signals (RSs) from adjacent subframes in order to estimate channel performances may still be suitable to operate together with a relay node according to an exemplary aspect of the invention in a relaying network. Also for neighbouring cell measurements an UE may make use of all potentially available RS, in particular for measurements on other frequencies, for the latter case the UE may not have much time available, since it has to retune its receiver for a short while, and therefore may use all available RS as well, which may be difficult when using a pure subframe border switching while may be possible when using a switching during a subframe .

Thus, a network element, e.g. a relay node, according to an exemplary aspect of the invention may enable a FDD relaying solution which can serve LTE R8 conform UEs. This may in particular be beneficial to allow LTE R8 conform user equip- merits, that interpolate or average over several reference signals or in general derive channel information from reference signals in multiple subframes.

It has to be noted that embodiments of the invention have been described with reference to different subject matters. In particular, some embodiments have been described with reference to apparatus type claims whereas other embodiments have been described with reference to method type claims. However, a person skilled in the art will gather from the above and the following description that, unless otherwise notified, in addition to any combination of features belonging to one type of subject matter also any combination between features relating to different subject matters, in par- ticular between features of the apparatus type claims and features of the method type claims is considered to be disclosed within this application.

The aspects and exemplary embodiments defined above and further aspects of the invention are apparent from the example of embodiment to be described hereinafter and are explained with reference to these examples of embodiment.

Brief Description of the Drawings

Fig. 1 schematically shows a band structure of a relaying network .

Fig. 2 schematically shows a basic relaying concept.

Fig. 3 schematically shows not R8 conform subframe allocation at relay nodes. Fig. 4 schematically shows a R8 conform relaying solution.

Fig. 5 schematically shows a relay multiplexing scheme adapted to achieve a concentration in time.

Fig. 6 schematically shows an alternative relay multiplexing scheme adapted to achieve a concentration in time.

Fig. 7 schematically shows a relay reception scheme according to an exemplary embodiment.

Fig. 8 schematically shows a pre-cancellation of inter symbol interference .

Fig. 9 schematically shows a pre-cancellation of inter symbol interference using channel knowledge.

Detailed Description

The illustration in the drawings is schematic. It is noted that in different figures, similar or identical elements are provided with the similar or identical reference signs.

With reference to Figs. 2 and 3 some basics of a relaying network will be described which may be useful for understanding exemplary embodiments of the invention.

Fig. 2 depicts the basic relaying concept in a communication network 200 with a nodeB (NB) 201 supporting own user equipments (UEs) 202 in parallel to the relay node (RN) 203, which in turn serves its own UEs 204. As we assume a LTE system, OFDMA allows a NB to support UEs and RNs in parallel on dif- ferent resource blocks (RBs), i.e. on some RBs the NB transmits data to UEs and on other RNs it transmits data to RNs.

In FDD systems the RN has intermittently to listen and to transmit on the same duplex frequency band, i.e. has to switch between receive and transmit mode. It should be noted that simultaneous transmission and reception at the RNs (direct repeater) may not be suitable due to its restricted applicability and difficult feedback instabilities.

An example for a typical relaying frame structure is shown in Fig. 3, where the DL frequency band 301 is shown for a sequence of subframes 302, 303 and 304. In subframe 302 the RN transmits data to its UEs together with a reference pilot grid as defined in LTE (poles 305 in Fig. 3) . In subframe 303 as well as each other even subframe it has to listen to the NB to receive new data. For this reason the RN is not able to broadcast the required reference signal (RS) grid to its UEs during even subframes. UEs which would try to average their channel estimation over several subframes to maximize interpolation gain may fail due to the completely missing RSs in each second subframe. In Releaseδ (R8) no control message has been defined, which could inform the UEs about subframes without RSs. For that reason this simple approach for relay- ing is not possible for R8 UEs.

Fig. 4 schematically shows a R8 conform relaying solution. A basic idea of this exemplary embodiment of the invention may be clear from Fig. 4, where the RN is switched within a sub- frame several times between Rx- and Tx-mode. In particular, the Fig. 4A shows schematically the switching between transmitting states 401 and receiving states 402 for a relay node, wherein in a first subframe 403 the relay node is in transmitting state, while in a second subframe 404 the switching is performed. In the bottom part of Fig. 4A the downlink band is shown in a schematically two dimensional form (similar to Fig. 3) in order to show that in each subframe different frequencies may be used for transmitting and receiving, i.e. a plurality of frequencies may be used for the communication. As in Fig. 3 poles 405 are indicating reference signals transmitted from the RN to UEs. The hatched bands 406 indicate the transmission of data (RBs) from the RN to its UEs) .

In Fig. 4B the communication scheme between a nodeB and its UEs and RN is schematically depicted which scheme may be implemented. In this implementation some specific RB formats are used, where no data are transmitted on the OFDM symbols carrying RSs, i.e. symbol 2, 6, 9 and 13. This may allow for data transmission from NB to its attached UEs on not used resources, but this feature is optional, as the NB might send full RBs as well to its UEs. If RSs from NB and RN are orthogonal - and for a proper relaying solution this would be advantageous - there may be no need for the NB to stop data transmission to its UEs on RBs not intended for data transfer to the RN. The data transmission (RBs) from NB to its UEs is schematically depicted in Fig. 4B by the continuous hatched bands 411, 412, and 413, while the transmission of data (RBs) from the NB to the RN is indicated by the columns 414 in Fig. 4B.

The switching allows the RN to transmit RSs signals in those four symbols of the subframe which contain RSs according to LTE R8, namely symbols 2, 6, 9, and 13.

For those symbols, where the RN is in transmit mode the NB may stop transmission of data to the RN. So instead of RBs of length 14 symbols the NB will send several data packets of shorter length to the RN. Therefore the NB-RN may not longer support LTE conform transmission according to the definition of RBs, but will be proprietary, which may be of small importance compared to a proprietary radio interface to the UEs. The latter interface should be maintained, otherwise legacy UEs may not be used in a relay enhanced network. However, base stations may typically be equipped with a software update, so that it is feasible to implement changes on the link from the BS to RN.

For symbols containing RSs the NB will have to stop data transmission to the RN. The unused subcarriers (SCs) for these symbols may not be used for data transmission to UEs as UEs always expect RBs of length 14 symbols.

So, one option would be to use the unused power of the data SCs in these symbols for boosting of data sent to other UEs or for boosting the RSs. Boosting of RSs may be advantageous in LTE, and the use of unused power may enable to omit the necessity to take away the power from data that are transmit- ted in these symbols. Such a removing of power may impact boosting of RSs and may even make this boosting concept pointless. However, when using the unused power this may improve the possibilities of boosting.

It may be possible for new UEs i.e. UEs for a later release than R8 to define ^λshort RBs', where such UEs with small data rate requirements might be supported by the data SCs of the RS symbols .

For implementations, where the power amplifier (PA) itself is not switched off during receive mode, but where instead e.g. some circuitry is used to disconnect the PA from the antenna, fast switching in the order of ns to μs should be possible so that the switching loss will be small. All symbols may con- tain a CP (cyclic prefix) in order to ease processing also in the case of multipath propagation. However, if RN are placed so that there is a strong LOS (Line of Sight) component, the effective delay spread may be short, so a major part of the CP may not be needed on the link between BS and RN Therefore at least some part of the CP may be used to switch off the transmitter (and possibly switch on the receiver instead) .

The resulting number of resource elements (RE) per subframe, which cannot be used for data transmission by the RN is <(4- 4/3) /14=19%, as there are 4 RS symbols in each subframe of length 14 symbols (here only frames with short GI is assumed) and each OFDM symbol with RSs has a RS at each third subcar- rier. For the calculation a system with 2 Tx antennas at the NodeB is assumed. It should be noted that in case of 4 Tx antennas there would be even less usable data SCs in each RS symbol .

The "lesser than" sign is used in the above equation as only those RBs, where a RN transmits data to one of its UEs might be affected, while the other RBs can be used by the NB for data transmission to some of its UEs. This may work well in case of semi-static allocation of RBs to RNs and NBs.

In case of power boosting of the RSs even less power- resources may be affected. Additionally in many cases it may take more than one subframe to send the data that the RN has been receiving from the NB in one of the subframes before to its UEs, as UEs might be in bad radio channel conditions while the NB-RN link is assumed to have optimum radio channel conditions. In that case less than every second subframe may have to perform the proposed switching procedure and therefore overall fewer resources are affected by the proposed scheme . Optimum channel conditions between NB and RN - compared to channel conditions from RNs to UEs - are assumed due to the application of directional antennas and line of sight (LOS) connections for the NB-RN links. Therefore the short intermittent data packets between the 4 data symbols for transmission of RN RSs may be sufficient for the data transfer from NodeBs to RNs.

So taking all issues together, the number of 19% of REs, which are not available for data transmission, is an upper bound, which typically will never be reached.

Looking at it from the RN point of view, the RN can receive data in 10 out of 14 symbols, because the latter are used to transmit RS from the RN. However, out of these 4 symbols 1/3 of the subcarriers carry RS anyhow and only the rest carries data, so the fraction of data bits that can be received can

be — = 79% i.e. a loss of 21%. Actually, some of those

10 +— -4

3 bits may have to be used for transmitting reference signals to the RN because the RN cannot receive the RS transmitted by the eNB, because it is busy transmitting its on RS during that time, but because the RN is expected not to move, the channel will be quite static. If the RN experiences LOS or close to LOS conditions the delay spread may be short and so the channel may be frequency flat. For both reasons the channel may not vary quickly in time or frequency and therefore few RS will be sufficient, so only a small fraction of the data symbols will need to be use for RS.

Summarizing, according to an exemplary aspect of the invention a method may be provided which may enable the use of common R8 UEs in a relaying communication network by providing relay nodes which are able to switch between a transmission mode and a receiving mode during a single subframe. A main advantage of such a method, i.e. the fast insertion of RSs during subframes were the RN is normally in listening mode, is that LTE Release 8 conform UEs may be supported by new relaying or multihop systems. The UEs may see on air a full RS grid as expected from the standard, while data transfer between NB and RNs can be organized simultaneously.

Fig. 5 schematically shows a relay multiplexing scheme adapted to achieve a concentration in time. In this example it is assumed that the RN cannot receive the very beginning of the OFDM symbol, i.e. it does miss more than the cyclic prefix (CP) . The basic structure of the OFDM symbol is depicted on the right hand part 501 of Fig. 5. It shows where the different information is concentrated in the time- frequency domain. It should be noted that this is just to be understood as a schematic illustration of the basic principle as it is not fully possible to completely confine a signal sharply in both time and frequency domain. The RN cannot receive the initial part of the OFDM symbol that is indicated by the dotted area 502. The left hand part 503 shows the generation of the signal. In the lower part a signal is gener- ated according to the R8 LTE standard: The data are converted from serial to parallel (S->P) 504, then input to an IFFT 505 and then the cyclic prefix is added (CP) 506. The data for the RN are first padded 507 with 0, then input to a DFT 508 and then combined with the R8 data in the IFFT 505. Cyclic prefix is appended to the combined data stream. As a result, the data for the RN are concentrated in time in the time span where the RN can receive data. Actually also some RN data are within the cyclic prefix, that the RN was assumed not to be able to receive, but this is not to be understood to be in contradiction to this embodiment, as the cyclic prefix is redundant, at least if there is no delay spread in the channel, which is quite likely for well placed RNs. Furthermore, because after the application of the DTF the RN data are proc- essed in the same way as the R8 data, the R8 data are not disturbed in any way by the RN signal. In particular there may be no interference from the RN signal to the R8 signal, so the performance of the R8 terminals may not compromised.

Fig. 6 schematically shows an alternative relay multiplexing scheme adapted to achieve a concentration in time. The scheme of Fig. 6 is similar to the one shown in Fig. 5 so that it is not described in detail. However, contrary to the scheme of Fig. 5 an IDFT 610 is performed in the beginning and a guard band 611 is schematically shown in the right part.

Fig. 7 schematically shows a relay reception scheme according to an exemplary embodiment that can be used to receive the data in the RN. The upper part shows the generation of the signal as in the Fig. 5. The lower part shows an example of a possible reception processing. First, the samples that could not be received due to the switching are padded with 0. Then basically the signal processing in the transmitter is mirrored by appropriate inverse processing steps: The signal is fed into a FFT 720 to be converted from time to frequency domain. Then the data for the R8 terminal are dropped 721 and the data for the RN are put into an IDFT (Inverse DFT) 722. Finally the data for the RN are extracted by deleting the 0- es 723 that had been padded in on the transmitter side to concentrate the RN data in the time span where the RN can receive. It will be apparent to those versed in the art that in a real receiver additional signal processing steps may be required e.g. to compensate distortions that are introduced to the signal due to the wireless channel and/or due to imper- fections in analogue parts. For the sake of clarity of displaying the essential aspects of the embodiments these steps are not shown in the figure. It will also be apparent that signal processing steps can be combined or further subdivided or modified to obtain the desired result.

The exemplary embodiments are shown for the example where the RN cannot receive samples at the beginning of the OFDM symbol. The principle can however easily be extended to the case that the RN does miss some samples at the end or both at the beginning and the end of the OFDM symbol.

Applying this signal processing scheme, the RN data can be exactly reconstructed if no R8 data are transmitted in the same OFDM symbol, there is however some interference if a R8 signal is present to the RN signal: The samples that are lost at the beginning of the OFDM symbol, will also generate some output to the frequency domain representation after the receivers FFT in the subcarriers that are assigned to the RN and will therefore cause some interference there. This is visualized by the dashed arrow indicating the source and victim of this intra symbol interference (ISI) . It may be an objective of the following embodiment to reduce or eliminate this interference.

In a particular exemplary embodiment, the subcarriers that are close to a R8 signal, or the subcarriers that are at the edges of the RN signals may not used but are padded with 0- es . These subcarriers would otherwise be most seriously af- fected by the above mentioned interference, because they are so close to the interfering signal in the frequency domain. By not using those badly interfered subcarriers the average signal quality and subsequently the achievable data rate may be increased, despite sacrificing the subcarriers at the edge (s) .

In another exemplary embodiment, the interference that is generated from the R8 signals to the RN signal may already be taken into account and compensated on the transmitter side. This may be possible, because the transmitter can anticipate the processing steps that will be done at the receiver side and therefore can also anticipate the resulting interference. The transmitter can then input the desired data minus the interference instead of the original RN data, thus pre- distorting the RN data. Then, when the interference adds to the transmitted data in the receiver due to the combination of transmitter and receiver processing, the actual interfer- ence cancels with the intentional predistortion .

An example for such a processing is shown in Fig. 8. In particular, Fig. 8 schematically shows a pre-cancellation of inter symbol interference. On the top left side there is the generation of the R8 signal, similarly as in the previous figures. Then in the lower left box 830 the interference from the R8 signal is predicted. For this purpose, the reception steps in the RN are performed. In particular, a FFT is performed 831, data for the R8 terminal are dropped 832 and the data on the subcarriers used for the RN are put into an IDFT 833. Finally the 0-es are deleted 834 and the inter symbol interference is estimated and forms the output of box 830. This interference estimate is then combined in the lower right hand box with the intended RN data and processed ac- cording to the already presented RN TX data processing. Then, in the upper right hand part the predistorted RN signal 835 and the R8 signal 836 are combined 837 and finally the cyclic prefix is added 838. The processing has been presented here in the most simple way to show the essence of the embodiment, it will be apparent to those versed in the art that various modifications and optimizations of the processing flow are possible e.g. combining Fourier and inverse Fourier transforms but the basic principle to predistort the RN signal to take into account predictable interference from R8 signals remains .

Again the previous embodiment did not take into account the influence of the wireless channel. It should be noted how- ever, that if the channel is not frequency selective or in other words flat in the frequency domain, i.e. all subcarri- ers are subject to the same distortion, then this distortion may be a linear operation on all subcarriers and therefore will not influence the distortion or the required predistor- tion. In this case it is therefore possible to apply a proper predistortion without channel knowledge and only apply the channel compensation at the receiver in the standard way. However, if the channel is frequency selective, it does have an influence on the distortion. Then it is still possible to generate a predistorted signal by taking the channel variations into account.

This is shown in Fig. 9 which contains the same basic building blocks as Fig. 8 i.e. generation of R8 signal (top left block) 940, prediction of the interference (top right block) 941 and predistortion of the RN signal with the predicted interference (lower block) 942. However, an estimation of the channel variations is now taken into account when calculating the predicted interference. It is taken into account both for the signal generation, here the effect of the wireless channel on the R8 signal is taken into account, and in the calculation of the interference, here the channel compensation that will be performed in the receivers processing chain are taken into account. Because the generation of the R8 signal used for the interference prediction now contains the estimated channel influence, another version without this influence has to be generated for the actual R8 transmit signal generation. Again various modifications and optimizations of the general signal processing schemes are possible.

In order to perform this channel compensation a channel estimate may need to be known at the transmitter side. It may be provided from the RN via feed back signalling or can be esti- mated by the base station itself from own measurements, in particular for TDD systems. As the RN will be typically placed in positions with favourable channel conditions, and is typically not moving but installed on fixed locations, the channel variations can be expected to be only slowly time varying (due to the stationary of the RN) and also only slightly variable in the frequency domain, because the channel will not exhibit a strong delay spread but have a strong Rician component (i.e. line of sight component or rather direct component) . Therefore the amount of channel feedback that is necessary may not be very huge and thus not absorb much link capacity on the feedback link.

Fig. 9 also shows an additional possible modification of the signal processing scheme, similarly to the one presented in Fig 6 (610) : Optionally, the RN data can be processed in a IDFT step 943 prior to be 0-padded 944 and fed into the DFT 945. This may cause the RN signal to become similar to an OFDM signal and this allows to perform frequency dependent optimizations on the different RN samples that now correspond to OFDM subcarriers, e.g. frequency adaptive modulation or power loading or scheduling can be applied. This modification may also be applied to the other embodiments presented before and may of course also require appropriate signal processing in the receiver. Finally, it should be noted that the above-mentioned embodiments illustrate rather then limit the invention, and that those skilled in the art will be capable of designing many alternative embodiments without departing from the scope of the invention as defined by the appended claims. In the claims, any reference signs shall not be construed as limiting the claims. The word "comprising" and "comprises", and the like, does not exclude the presence of elements or steps other than those listed in any claim or the specification as a whole. The singular reference of an element does not exclude the plural reference of such elements and vice-versa. In a device claim enumerating several means, several of these means may be embodied by one and the same item of software or hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

List of reference signs:

101 Downlink band

102 First subframe 103 Second subframe

104 Uplink band

105 One subframe 106 Next subframe 107 Duplex gap 108 Coordinate system

200 Communication network

201 NodeB

202 User equipment

203 Relay node 204 User equipment

301 Downlink band

302 Subframe

303 Subframe

304 Subframe 305 Reference signals

401 Transmitting state

402 Receiving state

403 First subframe

404 Second subframe 405 Reference signals

406 Data transmission

411 Data transmission

412 Data transmission

413 Data transmission 414 Data transmission

501 Basic structure of OFDM symbol

502 Initial part of OFDM symbol

503 Generation of signals

504 Serial to parallel conversion 505 I FFT

506 Adding cyclic prefix

507 Padding with 0

508 DFT 610 IDFT

611 Guard time

720 FFT

721 Dropping of unused data

722 IDFT 723 Deleting 0

830 Prediction of interference

831 FFT

832 Dropping of unused data

833 IDFT 834 Deleting 0

835 Predistorted RN signal

836 R8 signal

837 Combining predistorted and R8 signal

838 Adding cyclic prefix 940 Generation of R8 signal

941 Prediction of interference

942 Predistortion

943 IDFT

944 0-padding 945 DFT

Claims

CLAIMS :

1. A network element (201, 203) for a communication network (200), wherein the network element (201, 203) is adapted to switch between a first state and a second state different to the first state during a subframe (403) of a communication in the communication network (200).

2. The network element (201, 203) according to claim 1, wherein the subframe (403) comprises a number of symbols which is in accordance with a communication protocol.

3. The network element (201, 203) according to claim 1, wherein the network element (201, 203) is a relay node (203) comprising: a transmitting unit; and a receiving unit, wherein the first state is a transmitting state in which the relay node (203) transmits signals by using the transmitting unit; and wherein the second state is a receiving state in which the relay node (203) receives signals by using the receiving unit.

4. The network element (201, 203) according to claim 3, wherein during the transmitting state reference signals are transmitted; and/or wherein during the receiving state data signals are received.

5. The network element (201, 203) according to claim 3, wherein signals received in the receiving state comprises a guard time in the beginning and/or the end of the signal .

6. The network element (201, 203) according to claim 5, wherein the received signal is an SC-FDMA signal transformed into frequency domain.

7. The network element (201, 203) according to claim 1, wherein the network element is a base station (201) comprising: a transmitting unit, wherein the first state is a transmitting state in which the base station (201) transmits signals on a subcarrier by using the transmitting unit; and wherein the second state is a silent state in which the base station (201) does not transmit on the subcarrier.

8. The network element (201, 203) according to claim 7, wherein the base station is adapted to communicate with at least two network elements (203) , wherein the base station (201) is further adapted to be in the silent state for the communication with a first one of the at least two network elements, and to be in the transmit- ting state for the communication with a second one of the at least two network elements.

9. The network element (201, 203) according to claim 8, wherein the base station (201) is adapted to use power which is not used by the base station during the silent state in the communication with the first one of the at least two network elements to increase the power level of the transmission of signals to the second one of the at least two network elements .

10. The network element (201, 203) according to claim 7, wherein in the first state a coding of the signals is increased compared to a subframe in which no switching is performed between the first state and the second state.

11. A communication network (200) comprising: a network element (203) according to claim 3, and a network element (201) according to claim 7, wherein the network element (201) according to claim 3 is in the transmitting state while the network element according to claim 7 is in the silent state.

12. The communication network (201) according to claim 11, further comprising: a user equipment (204), wherein the network element (203) according to claim 3 transmits reference signals to the user equipment (204) in the transmitting state.

13. A method of operating a network element according to claim 1, the method comprising: receiving signals during the second state, and transmitting signals during the first state.

14. A program element, which, when being executed by a processor, is adapted to control or carry out a method according claim 13.

15. A computer-readable medium, in which a computer program is stored which, when being executed by a processor, is adapted to control or carry out a method according claim 13.