Method, device, and system for coherent joint transmission

Abstract

The invention relates to a method, a network node, a system for supporting coherent joint transmission, CJT, in a distributed-multiple input multiple output, D-MIMO, system, by obtaining (601) information on frequency update of two or more transmit-and-receive points, TRPs, configured for CJT of a downlink, DL, communications channel to a wireless device, WD; predicting (603) one or more phase offsets between the TRPs, based on historical information on the DL communications channels between the TRPs and the WD, and the corresponding frequency updates; and performing (605) a pre-compensation of the DL communications channel transmitted from the TRPs, based on the one or more predicted phase offset.

Description

[0001]METHOD, DEVICE, AND SYSTEM FOR COHERENT JOINT TRANSMISSION TECHNICAL FIELD The invention relates to a method for supporting coherent joint transmission (CJT), a network node for supporting CJT, a system for supporting CJT, and corresponding computer program, computer program product, and data carrier signal. BACKGROUND In 4G, 5G, and 6G mobile communications systems, beamforming features are becoming more and more critical for improved system performance. Beamforming parameters (such as, beam direction and nulling directions) are determined based on estimations of the wireless communications channel between a transmit-and-receive point (TRP) and a user equipment (UE). The channel estimation is based on measured channel parameters which require specific reference symbols (e.g., 3rd Generation Partnership Project (3GPP) defined downlink (DL) Channel State Information Reference Signal (CSI-RS) symbols or uplink (UL) Sounding Reference Signal (SRS) symbols) that are transmitted at regular time intervals. On one hand, channel estimations may add overhead if performed too often, one the other hand, channel estimations should be done often enough to avoid channel estimation aging. Channel estimation aging is due to the UE mobility which causes the communications channel to continuously evolve. Distributed Multiple Input Multiple Output (D-MIMO) systems aim at increasing coverage and at improving spectral efficiency. In a D-MIMO system with coherent joint transmission (CJT) multiple TRPs simultaneously serve a same device, e.g., a UE, over the same time / frequency resources based on directly measured channel characteristics. CJT for DL communications channel may improve interference, allowing to serve multiple UEs simultaneously. To obtain good performance, CJT requires good frequency synchronization between TRPs and needs to take into account channel aging. Further information on CJT may be found in WO 2023 / 175513 A1. SUMMARY   An object of the invention is to improve performance of coherent joint transmission (CJT). This and other objects of the invention are achieved by means of different aspects of the invention, as defined by the independent claims. Embodiments of the invention are characterized by the dependent claims. According to a first aspect of the invention, a method for supporting CJT in a D-MIMO system is provided. The D-MIMO system comprises two or more transmit-and-receive points (TRPs) configured for CJT of a downlink (DL) communications channel to a wireless device (WD). The method is performed by a network node. The method comprises obtaining information on frequency update for each of the two or more TRPs. The method comprises predicting one or more phase offsets between TRPs. The prediction of the one or more phase offsets between TRPs is based on historical information on the DL communications channels between the TRPs and the WD and the corresponding frequency updates. The method comprises performing a pre-compensation of the DL communications channel transmitted from the TRPs. The pre-compensation is based on the one or more predicted phase offsets. According to a second aspect of the invention, a network node for supporting CJT in a D- MIMO system is provided. The D-MIMO system comprises two or more TRPs configured for CJT of a DL communications channel to a WD. The network node comprises a processor and a memory. The memory has stored thereon instructions executable by the processor. The instructions, when executed by the processor, cause the network node to obtain information on frequency update of each of the two or more TRPs. The instructions, when executed by the processor, cause the network node to predict one or more phase offsets of between TRPs, based on historical information on the DL communications channels between the TRPs and the WD and the corresponding frequency updates. The instructions, when executed by the processor, cause the network node to perform a pre-compensation of the DL communications channel transmitted from the TRPs, based on the one or more predicted phase offset. According to a third aspect of the invention, a system node for supporting CJT in a D-MIMO is provided. The system comprises a network node for supporting CJT and two or more TRPs. The system comprises the network node configured to obtain information on frequency update from each of the two or more TRPs. The system comprises the network node configured to predict one or more phase offsets between TRPs, based on historical information on the DL communications channels transmitted from the TRPs to the WD and the corresponding   frequency updates. The system comprises the network node configured to transmit to the two or more TRP the corresponding predicted phase offset. The system comprises the two or more TRPs configured to perform a pre-compensation of the corresponding DL communications channel, based on the received phase offsets. According to a fourth aspect of the invention, there is provided a computer program. The computer program comprises instructions which, when run in a processing unit on a network node, cause the network node to perform the method according to the first aspect. According to a fifth aspect of the invention, there is provided a computer program product. The computer program product comprises a computer readable storage medium on which the computer program according to the fourth aspect is stored. According to a sixth aspect of the invention, there is provided a data carrier signal. The data carrier signal carries the computer program according to the fourth aspect. Certain embodiments may provide one or more of the following technical advantages: - improvement of the accuracy of the precoding of the signals transmitted from the TRPs to the WD; - compensation of the phase variations vs time due to (periodical) updates of the reference clock frequency; - reduction of requirements on absolute frequency synchronization between the TRPs; - seamless integration with already existing communication channel prediction functionality. BRIEF DESCRIPTION OF THE DRAWINGS For better understanding of the present disclosure, and to show more readily how the invention may be carried into effect, reference will now be made, by way of example, to the following drawings, in which: Figure 1 shows an example of distributed-multiple input multiple output (D-MIMO) system according to embodiments;   Figure 2 shows a simplified view of a cumulative phase drift in case of wireless device (WD) mobility and constant transmit-and-receive point (TRP) frequency offset; Figure 3 shows a simplified view of a cumulative phase drift in case of WD mobility and frequency update; Figure 4a shows a first example of a system according to embodiments; Figure 4b shows a second example of a system according to embodiments; Figure 5 shows a simplified view of a cumulative phase drift in case of WD mobility and frequency update when the invention is implemented; Figure 6 shows a flow chart illustrating a method according to embodiments; Figure 7 shows a timeline for two TRPs, TRP1 and TRP2, of a D-MIMO system: illustrating how frequency offset compensation may be affected by an error because of a delay between the point in time a synchronization algorithm applies the frequency update at a TRP and the point in time wherein a network node applies the compensation; Figure 8 shows a timeline illustrating how to post-compensate for control loop delays between application of a frequency update and related predictor corrections; Figure 9 shows a block diagram depicting a network node according to embodiments of the invention; and Figure 10 shows a simplified view of a cumulative phase drift in case of frequency update if the second example of a system according to embodiments is implemented. DETAILED DESCRIPTION Embodiments will be illustrated herein with reference to the accompanying drawings. These embodiments are provided by way of example so that this disclosure will be thorough and complete, and will fully convey the scope of the inventive concept to those skilled in the art.   "First", "second", "third", etc. are used as a manner of distinguishing between different instances of a term, they are not intended to confer a cumulative or chronological meaning to the terms. Performance of coherent joint transmission (CJT) in distributed-multiple input multiple output (D-MIMO) systems comprising multiple transmit-and-receive points (TRPs) depends on channel aging and frequency synchronization accuracy between the TRPs. Channel aging is due to the relative movement between wireless device (WD), e.g., a user equipment (UE), and TRPs serving the WD. This relative movement causes a mismatch between estimated channel and current channel. As the relative movement between WD and TRPs increases, channel aging impact on performance increases. The frequency synchronization accuracy between TRPs depends on factors, such as synchronization source(s), characteristics of fronthaul synchronization distribution, TRPs reference oscillator designs, temperature variations, etc. Any frequency offset between TRPs will cause phase drift over time. Specifically, a control algorithm generates a reference clock for the TRPs, but relative phase and frequency offset between the TRPs may vary because of periodical updates of the reference clock in each TRP and / or accuracy of the reference clock. In case of constant frequency offset between TRPs, the phase variation / drift would be a linear function, and therefore a communications channel predictor could easy predict it. However, in real case scenarios, the frequency offset is not constant and varies over time, therefore (regular) frequency updates will result in abrupt phase changes. The invention disclosed herein makes it possible to improve performance of CJT by a method comprising obtaining information on frequency update from two or more TRPs configured for CJT of a DL communications channel to a WD. One or more phase offsets between TRPs are predicted based on historical information on the DL communications channels between the TRPs and the WD, and the corresponding frequency updates. The method further comprises performing a pre-compensation of the DL communications signals when transmitted from the TRPs, based on the predicted phase offsets. The invention disclosed herein enables - to improve the accuracy of the precoding of the signals transmitted from the TRPs to the WD; - to compensate the phase variations vs time due to (periodical) updates of the reference clock frequency;   - to reduce requirements on absolute frequency synchronization between the TRPs; - a seamless integration with already existing communications channel prediction functionality. Figure 1 shows a D-MIMO system wherein the invention disclosed herein may be implemented. The D-MIMO system comprises two TRPs, TRP1103a and TRP2103b, but the invention also applies to more than two TRPs. The two TRPs, 103a-b, are configured for CJT of a DL communications channel to a WD 105, e.g., a UE. The WD 105 may move at a speed v. A TRP, 103a-b, is an antenna array with one or more antenna elements and located at a specific geographical location. A TRP comprises a radio transceiver unit (RU) and frequency and time synchronization functionality. Each TRP in the D-MIMO system is separated in space, therefore antennas in one TRP experience the same frequency offset and phase drift, whereas antennas from different TRPs experience different frequency offset and phase drift. It should be noted that in some cases, multiple TRPs may be served by the same RU, where radio frequency (RF) signal distribution is performed (via cables), and may then experience the same frequency offset and phase drift even though they are not co-located. The transmission to / from each TRP is controlled by a network node 101. The RU may host the frequency and time generator generating a synchronization reference for the TRPs, the RU may in turn receive a synchronization reference from the network node 101, but other synchronization alternatives are possible. The network node 101 may be connected to the two or more TRPs, 103a-b, via a (point-to-point or packet based switched network) digital fronthaul solution. The network node 101 may comprise a CJT precoder entity and a communications channel predictor entity. With reference to Figure 1, H1, 107a, represents the wireless communications channel between TRP1, 103a, and the WD, 105, and H2, 107b, represents the wireless communications channel between TRP2, 103b, and the WD, 105. t1 and t2 represent transmitter impairments of TRP1, 103a, and TRP2, 103b, respectively, and rWDrepresents the receiver impairments for the WD, 105. The transmitter impairments, such as frequency errors and frequency drift, cause RF phase instability and drift over time. The receiver impairments, assumed to be common for the two paths (i.e., TRP1-WD and TRP2-WD), may be frequency error and drift from frequency reference. The parameters H1, H2, t1, t2, and rWDare complex-valued narrowband representation of the communications channel (complex baseband notation). The signal received at the WD 105 may be represented with the formula: ^^ ൌ ^^^ ^ ^^ଶ ൌ ^^^^^^^^^^^^ ^ ^^^^^^ଶ^^ଶ^^ ൌ ^^^^^^^^^^^ ^ ^^ଶ^^ଶ^ ^^   wherein, ^^^is the signal received at the WD 105 from TRP1103a, ^^ is the signal transmitted by TRP1103a and TRP2103b to the WD 105, and ^^ଶis the signal received at the WD 105 from TRP2103b. The task of the network node 101, specifically the DL precoder entity, is to add precoding weights to the signal ^^ sent from each TRP, TRP1 and TRP2, so that the signals received by the WD 105 are added in phase (and amplitude) thereby maximizing the received signal level at the WD 105. A relative communications channel d^^ from each TRP, 103a-b, to the WD 105 may be defined as: d^^ ൌ^^^^^ ൌ^^^^and a phase relationship Δφ may be ^^ ൌ∠d^^ ൌ ∠൬ ^ ^ ^ ൌ ∠൬ ^^^ Δ^^ ^^^^ ^ ∠൬^^^ ൌ Δ^^ு ^ Δ^^௧Note that the phase is additive. phase relationship Δ^^ varies over time: the variation of the first term (Δ^^ு) over time is due to the WD 105 movements which change the relative phase between the wireless communications channels H1 and H2, and the variation of the second term (Δ^^௧) over time is a function of the frequency offset between the TRPs, 107a-b. Figures 2-5 will now be described to further explain the problem to solve and how the problem is solved by the invention. Figure 2 shows a simplified view of a cumulative phase drift Δφ, for example, in a case where a WD has a good line-of-sight (LoS) channel to each TRP, and the WD is initially moving fast, and then slowing down after a certain time interval until it stops and stands still. The dashed line 203 represents the phase drift Δ^^ுdue to the WD mobility. The dot-dashed line 205 represents the phase drift Δ^^௧due to the frequency offset. The solid line 201 represents the total phase drift (cumulative) Δφ that is the sum of the twoeffects (Δ^^ு ^ Δ^^௧). At each channel estimation instance (i.e., at the points in time 207a-e onthe solid line in Figure 2), a new phase drift, 209a-e, is estimated and modeled as an estimated phase derivative around the channel estimation instance. The predicted phase drifts are represented by the dotted lines 209a-e. Figure 2 shows that a nonlinear phase drift introduces prediction errors. Channel prediction is performed by using historical channel estimations to predict future channel outcome. In case of channel prediction for multiple TRPs, part of the channel   prediction problem is to estimate and predict the relative channel changes due to the WD movement (Δ^^ு) and to the TRP frequency offset (Δ^^௧), so that the precoder for CJT may pre- code the signals sent by the TRPs to the WD with the correct phase offset to create coherent summation at the WD receiving antenna. In real case scenarios, the frequency offset between TRPs is not constant but changes over time. In fact, the frequency of each TRP is regulated with periodical frequency updates that cause an abrupt change in the frequency offset and therefore in the phase slope vs time. This makes the channel prediction difficult since historical channel information does not give any indication on the frequency updates, and the frequency updates are a function of a noisy estimate where the noise could e.g. be due to noise introduced in a packet based fronthaul when distributing synchronization to the RU. Note that e.g., oscillator frequency changes caused by temperature or aging are considered slow effects and would not cause the abrupt change referred above. Figure 3 shows a simplified view of the cumulative phase drift Δφ, in a case where, as in Figure 2, the WD is initially moving fast, and then slows down after a certain time interval until it stops and stands still. The phase drift Δ^^ுrepresented by the dashed line 303 in Figure 3 is the same 203 of Figure 2, Differently from Figure 2, in Figure 3, the phase drift Δ^^௧due to the frequency offset (represented by the dot- dashed line 305) has an abrupt change due to a frequency update 311. Figure 3 shows that the predicted phase drift 309a-e (dotted line) at the points in time 307a-e is too slow to react to the phase slope change due to the frequency update, and therefore the predicted phase drift 309c is inaccurate when the frequency changes 311. The invention disclosed herein solves the problem shown in Figure 3 by predicting one or more phase offsets between TRPs, based on historical information on the DL communications channels between the TRPs and the WD, and the corresponding frequency updates. Figure 4a and 4b show a system wherein the invention as disclosed herein may be implemented. The system comprises: - two TRPs 103a-b configured for CJT of a DL communications channel to a WD 105 moving at a speed v; the two TRPs 103a-b may comprise a frequency and time generator 109a-b that generates a frequency update for the TRPs based on a frequency and timing control algorithm 407a-b; the TRPs 103a-b send the information on a frequency update 409 to a network node 101; - the network node 101 comprising:   - a channel predictor entity 405 estimating the communications channels between the TRPs 103a-b and the WD 105; the channel predictor entity 405 takes as input the frequency update generated by the frequency and timing control algorithm 407a-b; and - a CJT precoder entity 402 taking as input the output of the channel predictor entity 405. As shown in Figure 4a, the CJT precoder entity 402 may comprise a CJT precoder decision entity 403, that determines the precoding weights for the signal to be transmitted from the TRPs 103a-b, and a DL precoder entity 401, that performs the pre-coding. Alternatively, as shown in Figure 4b, the DL precoder entity 401a-b may be comprised in each TRP. Figure 5 shows the same scenario of Figures 3, but applying the invention as disclosed herein. Figure 5 shows that the predicted phase drift (dotted line 309a-b, 509c, 309d-e) is accurate (i.e., the dotted line 509c follows the solid line 301 representing the cumulative phase drift Δφ) because the network node takes into account also the frequency update of the TRPs. Figure 6 shows a method 600 for supporting CJT in a D-MIMO system. The D-MIMO system comprises two or more TRPs configured for CJT of a DL communications channel to a WD. The method 600 is performed by a network node. The method 600 comprises obtaining 601 information on frequency update for each of the two or more TRPs. The information on frequency update may be obtained from the corresponding TRP. The frequency update of a TRP may comprise one or more of - a value of the frequency update of the TRP, - a timestamp associated with the frequency update indicating the point in time the frequency update is applied to the TRP, thus allowing the network node (e.g., a channel predictor entity comprised in the network node) to know the validity of the frequency update; - a value indicating time and occurrence of further frequency updates, e.g., a period of time. The method 600 comprises predicting 603 one or more phase offsets between TRPs. The prediction of the one or more phase offsets is based on historical information on the DL communications channels between the TRPs and the WD, and the corresponding frequency updates. If the WD is stationary, channel aging may not impact the performance, therefore, the prediction of the one or more phase offsets does not take into account the WD movement. With reference to the formulas used above, in this case Δ^^ = Δ^^௧. If the WD is moving, the prediction of the one or more phase offsets may be obtained by DL communications channel prediction, i.e., considering more than one historical information on the DL communications   channels between the TRPs and the WD, so that the WD movement is taken into account. With reference to the formulas used above, in this case Δ^^ = Δ^^ு ^ Δ^^௧.The prediction of the one or more phase offsets may be performed by - selecting 607, as a reference TRP, one of the two or more TRPs; - for each of the remaining TRPs (i.e., all the TRPs excluded the TRP selected as reference TRP), determining 609 a relative communications channel associated with the reference TRP and each of the remaining TRPs. A relative communications channel is based on the DL communications channel of the reference TRP and the DL communications channel of the corresponding remaining TRP; - predicting 611 the phase offset for each of the relative communications channels. For example, in case of n TRPs (e.g., TRP0, TPR1, …, TPRn), n-1 relative communications channels (e.g., ^^^^^, …, ^^^^^ି^) may be determined. Selecting, for example, TRP0 as thereference TRP, the n-1 relative communications channels may be: ^^^^ ^భ^^షభ^ൌ ^బ, …, ^^^^^ି^ൌ ^బ, signal received by the WD from TRPn. may be predicted, wherein Δ^^^ି^ ൌ ൌ ∠d^^^ି^ and ∠^∙^ isthe phase (angle) After predicting 611 one or more phase offsets between the TRPs, the method 600 comprises performing 605 a pre-compensation of the DL communications channels transmitted from the TRPs, based on the one or more predicted phase offsets. In case of two TRPs, wherein one of the two TRPs has been selected as the reference TRP, the pre-compensation of the DL communications channel transmitted to the WD may comprise performing 613 the pre- compensation only of the DL communications channel transmitted from the reference TRP. In case of more than two TRPs, wherein one of the more than two TRPs has been selected as the reference TRP, the pre-compensation of the DL communications channel transmitted to the WD may comprise performing 615 the pre-compensation of the DL communications channel transmitted from all the TRPs excluding the DL communications channel transmitted from the reference TRP. The frequency offset compensation may be affected by an error because of a delay between the point in time a synchronization algorithm applies the frequency update at a TRP and the point   in time wherein the network node applies the compensation. For example, Figure 7 shows a timeline for two TRPs, TRP1 and TRP2, of a D-MIMO system: - t0 and t3 are the points in time wherein a frequency update in TRP1 and TRP2 respectively, is determined and sent to the network node; - t1 and t4 are the points in time wherein the frequency update is applied at TRP1 and TRP2, respectively, and - t2 and t4 are the points in time wherein the network node or the TRP(s) applies the compensation (i.e., performs 605 the pre-compensation) for TRP1 and TRP2, respectively. The delays between the point in time a synchronization algorithm applies the frequency update at a TRP and the point in time wherein the network node applies the compensation, i.e., Δt2_t1705a for TRP1 and Δt5_t4705b for TRP2, may introduce an uncompensated phase error. The delays Δt2_t1705a and Δt5_t4705b should be as small as possible and preferably zero to reduce / avoid the uncompensated phase error. The delays Δt2_t1705a and Δt5_t4705b are caused by a predictor control loop delay (i.e., the time interval between the point in time wherein the frequency update is determined and the point in time wherein the frequency update is compensated, i.e., Δt2_t0701a and Δt5_t3701b for TRP1 and TRP2 respectively) being larger than the delay for the synchronization algorithms to apply the frequency update, i.e., Δt1_t0703a and Δt4_t3703b for TRP1 and TRP2 respectively. With reference to Figure 7, this may be expressed as: Δt2_t0 > Δt1_t0 and Δt5_t3> Δt4_t3. This problem may be reduced by dimensioning, at the corresponding TRP, the delay Δt1_t0703a for TRP1 and Δt4_t3703b for TRP2, so that it is equal or higher than the predictor control loop delay (i.e., Δt2_t0, Δt5_t3). Therefore, the method 600 may further comprise, for each TRP, obtaining 617 (by the network node) a first time interval value 701a 701b (e.g., Δt2_t0, Δt5_t3), wherein the first time interval is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time the frequency update is compensated in the network node and sending the determined first time interval value to the corresponding TRP. The network node may obtain the first time interval (e.g., Δt2_t0, Δt5_t3) based on the point in time the network node performs the pre-compensation (e.g., t2, t4) and the information on frequency update received from e.g. the TRP, comprising information on the point in time wherein the frequency update is applied (e.g., t1, t4) and on the point in time wherein the frequency update is determined (e.g., t0, t3) or on the interval between the two points (e.g., Δt1_t0, Δt4_t3). In case of coordinated approach of the TRPs, i.e., when the TRPs applies the frequency update at the same time, the method 600 may comprise, for each TRP, obtaining 617 (by the network node) the first time interval 701a 701b (e.g., Δt2_t0, Δt5_t3). Between all the obtained first   time intervals, the method 600 further comprises selecting 619 the first time interval with the highest value and transmitting 621 it to the two or more TRPs. In both cases of non-coordinated and coordinated approach, each TRP may receive the first time interval value and determine a fifth time interval 703a-b (e.g., Δt1_t0, Δt4_t3). The fifth time interval value is equal or higher than the first interval (e.g., Δt2_t0, Δt5_t3) and is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time wherein the frequency update is applied. This may be expressed with the following formulas: Δt1_t0 ≥ Δt2_t0 for TRP1 Δt4_t3 ≥ Δt5_t3 for TRP2 With reference to Figure 8 showing a timeline for a TRP (i.e., TRP1) of a D-MIMO system, if the delay (or third time interval 805) (i.e., Δt2_t1 representing the time interval between the point in time wherein the frequency update is applied in the TRP and the point in time wherein the pre-compensation is performed) may not be reduced or eliminated, and if during the delay (or third time interval 805) (Δt2_t1) the phase error is higher than a threshold, a post- compensation may be performed by scheduling restrictions for CJT. The method 600 may further comprise determining 625 the third time interval value 805 (e.g., Δt2_t1). The third time interval value is indicative of the time interval between the point in time wherein the frequency update is applied in the TRP (e.g., t1) and the point in time wherein the pre-compensation is performed (e.g., t2). The third interval may be determined based on - the obtained information on frequency update (comprising information on the point in time wherein the frequency update is applied, e.g., t1) and - the point in time (e.g., t2) wherein the pre-compensation is performed (t2 is known by the network node since the network node applies the pre-compensation). The method 600 may further comprise estimating 627 a phase offset of each TRP during the third time interval (e.g., Δt2_t1), and applying 629 a scheduling restriction to the WD from the two or more TRPs (i.e., the TRPs do not transmit to the WD) during a fourth time interval 807 (Δt2_t6), if the estimated phase offset in the third time interval (e.g., Δt2_t1) is higher than a threshold. The fourth time interval (Δt2_t6) may be equal to the third time interval (e.g., Δt2_t1) or shorter. It will be appreciated that the method 600 may comprise additional, alternative, or modified, steps in accordance with what is described throughout this disclosure.   Figure 9 shows a block diagram illustrating an embodiment of a network node 101, 900 according to embodiments of the invention, comprising processing circuitry 901, a computer- readable data carrier, such as the memory 902, and the network interface circuitry 903. The processing circuitry 901 may comprise one or more processors, such as CPUs, microprocessors, application processors, application-specific processors, Graphics Processing Units (GPUs), and Digital Signal Processors (DSPs) including image processors, or a combination thereof, and the memory 902 comprises the computer program comprising instructions. When executed by the processor(s), the instructions cause the network node 900 to become operative in accordance with embodiments of the invention described herein, in particular with reference to Figure 6. More specifically, the network node 900 becomes operative to obtain information on frequency update of each of the two or more TRPs. The frequency update may be obtained from the corresponding TRP. The frequency update of a TRP may comprise one or more of a value of the frequency update of the TRP, a timestamp associated with the frequency update indicating the point in time the frequency update is applied to the TRP, a value indicating time and occurrence of further frequency updates. The network node 900 becomes operative to predict one or more phase offsets of between TRPs, based on historical information on the DL communications channels between the TRPs and the WD and the corresponding frequency updates. The network node 900 becomes operative to perform a pre-compensation of the DL communications channel transmitted from the TRPs, based on the one or more predicted phase offset. The network node 900 may become operative to predict one or more phase offsets by predicting a phase offset by DL communications channel prediction. The network node may become operative to predict a phase offset by selecting, as a reference TRP, one of the two or more TRPs; for each of the remaining TRPs, determining a relative communications channel associated with the reference TRP and each of the remaining TRPs, wherein a relative communications channel is based on the DL communications channel of the reference TRP and the DL communications channel of the corresponding remaining TRP; and predicting the phase offset for each of the relative communications channels. The network node 900 may become operative to perform a pre-compensation of the DL channel transmitted to the WD based on the phase offset by performing the pre-compensation of the DL communications channel transmitted from the reference TRP. Alternatively, the network node may become operative to perform a pre-compensation of the DL channel transmitted to   the WD based on the phase offset by performing the pre-compensation of the DL communications channel transmitted from the remaining TRPs. The network node may become operative to obtain, for each TRP, a first time interval, wherein the first time interval is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time the frequency update is compensated in the network node. The network node may become operative to obtain as first time interval value for all the two or more TRPs the highest between the first time intervals of the two or more TRPs. The network node may become operative to send the determined first time interval value to the corresponding TRP. Alternatively, the network node may become operative to determine a third time interval value, wherein the third time interval value is indicative of the time interval between the point in time wherein the frequency update is applied in the TRP and the point in time wherein the pre- compensation is performed; estimate a phase offset of each TRP during the third time interval; and apply a scheduling restriction to the WD from the two or more TRPs during a fourth time interval, if the estimated phase offset in the third time interval is higher than a threshold. The computer program 904 may be stored in a computer-readable data carrier, such as a memory 902. Alternatively, the computer program 904 may be carried by a data carrier signal, e.g., downloaded to the memory 902 via a network interface circuitry 903. The memory 902 may, e.g., be a RAM, a ROM, a Flash memory, or the like. The computer program 904 may be downloaded to the memory 902 by means of the network interface circuitry 903, as a data carrier signal carrying the computer program 904. The network interface circuitry 903 may comprise one or more of a cellular modem (e.g., GSM, UMTS, LTE, 5G, or higher generation), a WLAN / Wi-Fi modem, a Bluetooth modem, an Ethernet interface, an optical interface, or the like, for exchanging data between the device manager and the network node 900 and other computing devices, communications devices, a radio-access network, and / or the Internet. The processing circuitry 901 may alternatively or additionally comprise one or more ASICs, FPGAs, or the like, which are operative to cause the network node 900 to become operative in accordance with embodiments of the invention described herein. It is to be understood that the structures as illustrated in Figure 9 are merely schematic and that the network node 900 may include further components which, for the sake of clarity, have not been illustrated, e.g., further interfaces or processors. Also, it is to be understood that the memory, 902, may include further program code for implementing other and / or known functionalities. It is also to be understood that the network node 900 may be provided as a   virtual apparatus. In one embodiment, the network node 900 may be provided in distributed resources, such as in cloud resources. When provided as virtual apparatus, it will be appreciated that the memory, 902, processing circuitry, 901, and communications circuitry, 903, may be provided as functional elements. The functional elements may be distributed in a logical network and not necessarily be directly physically connected. It is also to be understood that the network node 900 may be provided as a single-node device, or as a multi-node system. The invention as disclosed herein may be implemented in a system comprising a network node for supporting CJT and two or more TRPs 103a-b. The network node may comprise a CJT precoder decision entity 403 and a channel predictor entity 405. The two or more TRPs may comprise a DL precoder entity 401a-b. The system may be the one illustrated in Figure 4b. The system comprises the network node configured to obtain information on frequency update from each of the two or more TRPs. The frequency update of a TRP may comprise one or more of - a value of the frequency update of the TRP, - a timestamp associated with the frequency update indicating the point in time the frequency update is applied to the TRP, thus allowing the network node (e.g., a channel predictor entity comprised in the network node) to know the validity of the frequency update; - a value indicating time and occurrence of further frequency updates, e.g., a period of time. The network node is further configured to predict one or more phase offsets between TRPs, based on historical information on the DL communications channels transmitted from the TRPs to the WD, and the corresponding frequency updates; and to transmit, to the two or more TRPs, the corresponding predicted phase offset. The two or more TRPs are configured to perform a pre-compensation of the corresponding DL communications channel, based on the received phase offset. The network node may be configured to predict a phase offset by DL communications channel prediction. Between frequency updates, a TRP (e.g., the DL precoder entity) may execute only the decisions of performing the pre-conding, received from the network node, without taking into account the frequency update. When a frequency update is detected, the DL precoder entity may directly apply the frequency update to the predicted phase drift. For example, Figure 10 shows a simplified view of a cumulative phase drift Δφ 301 (solid line) that is the sum of a phase drift Δ^^ுdue to the UE mobility 303 (dashed line) and a phase drift Δ^^௧due to the frequency offset 305 (dot-dashed line), in a case where a WD has a good LoS channel to each   TRP. 1003a-e represent the points in time wherein the CJT precoder takes a decision of performing the pre-conding. When the frequency update (e.g., 311 in Figure 10) occurs between the CJT precoder decisions (e.g., 1003c and 1003d in Figure 10), the DL precoder entity in the TRP applies a phase compensation 1001 based on the frequency update to the predicted phase drift (dotted line in a circle 1001) until the DL precoder entity receives the next precoding instructions from the CJT precoder decision entity 1003d. When the DL precoder entity receives the next precoding instructions from the CJT precoder decision entity, 1003d, the channel predictor entity has received new estimates from reference symbols that has not been compensated by the DL precoder entity. From then on, the DL precoder entity again follows only the CJT precoder decisions (1003d, 1003e), that now includes the effect of the new frequency value, until the next frequency update. Therefore, the system may further comprise, the two or more TRPs configured to obtain information on frequency update; and to perform a pre-compensation of the corresponding DL communications channel, based on the information on the frequency update. The system may further comprise, the network node configured to predict a phase offset by - selecting, as a reference TRP, one of the two or more TRPs; - for each of the remaining TRPs, determining a relative communications channel associated with the reference TRP and each of the remaining TRPs, wherein a relative communication channel is based on the DL communications channel of the reference TRP and the DL communications channel of the corresponding remaining TRP; - predicting the phase offset for each of the relative communications channels. The system may further comprise, the network node configured to obtain, for each TRP, a first time interval, wherein the first time interval is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time the frequency update is compensated in the network node; and to transmit the determined first time interval value to the corresponding TRP. In case of coordinated approach of the TRPs, i.e. when the TRPs applies the frequency update at the same time, the method may comprise for each TRP, obtaining a first time interval, wherein the first time interval is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time the frequency update is compensated in the network node. Between all the obtained first time intervals, the one with the highest value is selected as first time interval value and transmitted to the two or more TRPs. In both cases of non- coordinated and coordinated approach, each   TRP may be configured to determine a fifth time interval, wherein the fifth time interval value is equal or higher than the first interval, wherein the fifth time interval value is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time wherein the frequency update is applied. The system may further comprise, the network node configured to determine a third time interval value. The third time interval value is indicative of the time interval between the point in time wherein the frequency update is applied in the TRP and the point in time wherein the pre-compensation is performed. The network node may be further configured to estimate a phase offset of each TRP during the third time interval, and applying a scheduling restriction to the WD from the two or more TRPs during a fourth time interval, if the estimated phase offset in the third time interval is higher than a threshold.

Claims

CLAIMS 1. A method (600) for supporting coherent joint transmission, CJT, in a distributed-multiple input multiple output, D-MIMO, system, wherein the D-MIMO system comprises two or more transmit-and-receive points, TRPs, configured for CJT of a downlink, DL, communications channel to a wireless device, WD, the method performed by a network node, and comprising: - obtaining (601) information on frequency update for each of the two or more TRPs; - predicting (603) one or more phase offsets between TRPs, based on historical information on the DL communications channels between the TRPs and the WD, and the corresponding frequency updates; and - performing (605) a pre-compensation of the DL communications channel transmitted from the TRPs, based on the one or more predicted phase offsets.

2. The method (600) according to claim 1, wherein predicting (603) one or more phase offsets comprises predicting a phase offset by DL communications channel prediction.

3. The method (600) according to any of claims 1 or 2, wherein predicting (603) one or more phase offsets comprises - selecting (607), as a reference TRP, one of the two or more TRPs; - for each of the remaining TRPs, determining (609) a relative communications channel associated with the reference TRP and each of the remaining TRPs, wherein a relative communications channel is based on the DL communications channel of the reference TRP and the DL communications channel of the corresponding remaining TRP; - predicting (611) the phase offset for each of the relative communications channels.

4. The method (600) according to claim 3, wherein performing (605) a pre-compensation of the DL channel transmitted to the WD based on the phase offset comprises - performing (613) the pre-compensation of the DL communications channel transmitted from the reference TRP.

5. The method (600) according to claim 3, wherein performing (605) a pre-compensation of the DL channel transmitted to the WD based on the phase offset comprises  - performing (613) the pre-compensation of the DL communications channel transmitted from the remaining TRPs.

6. The method (600) according to any of claims 1-5, wherein obtaining (601) information on frequency update from each of the two or more TRPs comprises receiving the frequency update of a TRP from the corresponding TRP.

7. The method (600) according to any of claims 1-6, further comprising: - for each TRP, obtaining (617) a first time interval, wherein the first time interval is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time the frequency update is compensated in the network node; - sending (621) the determined first time interval value to the corresponding TRP.

8. The method (600) according to claim 7, comprises - obtaining (619) as first time interval value for all the two or more TRPs the highest between the first time intervals of the two or more TRPs.

9. The method (600) according to any of claims 1-6 further comprising: - determining (625) a third time interval value, wherein the third time interval value is indicative of the time interval between the point in time wherein the frequency update is applied in the TRP and the point in time wherein the pre-compensation is performed; - estimating (627) a phase offset of each TRP during the third time interval; - applying (629) a scheduling restriction to the WD from the two or more TRPs during a fourth time interval, if the estimated phase offset in the third time interval is higher than a threshold.

10. The method (600) according to any of claims 1-9, wherein the frequency update of a TRP comprises one or more of a value of the frequency update of the TRP, a timestamp associated with the frequency update indicating the point in time the frequency update is applied to the TRP, a value indicating time and occurrence of further frequency updates.

11. A network node (101, 900) for supporting coherent joint transmission, CJT, in a distributed-multiple input multiple output, D-MIMO, system, wherein the D-MIMO system  comprises two or more transmit-and-receive points, TRPs, configured for CJT of a downlink, DL, communications channel to a wireless device, WD, the network node comprising a processor and a memory, the memory having stored thereon instructions executable by the processor, wherein the instructions, when executed by the processor, cause the network node to: - obtain information on frequency update of each of the two or more TRPs; - predict one or more phase offsets of between TRPs, based on historical information on the DL communications channels between the TRPs and the WD and the corresponding frequency updates; and - perform a pre-compensation of the DL communications channel transmitted from the TRPs, based on the one or more predicted phase offset.

12. The network node (101, 900) according to claim 11, wherein the instructions cause the network node to predict one or more phase offsets comprises predicting a phase offset by DL communications channel prediction.

13. The network node (101, 900) according to any of claims 11 or 12, wherein the instructions cause the network node to predict one or more phase offsets by - selecting, as a reference TRP, one of the two or more TRPs; - for each of the remaining TRPs, determining a relative communications channel associated with the reference TRP and each of the remaining TRPs, wherein a relative communications channel is based on the DL communications channel of the reference TRP and the DL communications channel of the corresponding remaining TRP; - predicting the phase offset for each of the relative communications channels.

14. The network node (101, 900) according to claim 13, wherein the instructions cause the network node to perform a pre-compensation of the DL channel transmitted to the WD based on the phase offset by - performing the pre-compensation of the DL communications channel transmitted from the reference TRP.

15. The network node (101, 900) according to claim 13, wherein the instructions cause the network node to perform a pre-compensation of the DL channel transmitted to the WD based on the phase offset by  - performing the pre-compensation of the DL communications channel transmitted from the remaining TRPs.

16. The network node (101, 900) according to any of claims 11-15, wherein the instructions cause the network node to obtain information on frequency update from each of the two or more TRPs by receiving the frequency update of a TRP from the corresponding TRP.

17. The network node (101, 900) according to any of claims 11-16, wherein the instructions further cause the network node to: - for each TRP, obtain a first time interval, wherein the first time interval is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time the frequency update is compensated in the network node; - send the determined first time interval value to the corresponding TRP.

18. The network node (101, 900) according to claim 17, wherein the instructions further cause the network node to - obtain as first time interval value for all the two or more TRPs the highest between the first time intervals of the two or more TRPs.

19. The network node (101, 900) according to any of claims 11-16 wherein the instructions further cause the network node to: - determine a third time interval value, wherein the third time interval value is indicative of the time interval between the point in time wherein the frequency update is applied in the TRP and the point in time wherein the pre-compensation is performed; - estimate a phase offset of each TRP during the third time interval; - apply a scheduling restriction to the WD from the two or more TRPs during a fourth time interval, if the estimated phase offset in the third time interval is higher than a threshold.

20. The network node (101, 900) according to any of claims 11-19, wherein the frequency update of a TRP comprises one or more of a value of the frequency update of the TRP, a timestamp associated with the frequency update indicating the point in time the frequency update is applied to the TRP, a value indicating time and occurrence of further frequency updates.

21. A system for supporting coherent joint transmission, CJT, the system comprising a network node (101, 900) for supporting coherent joint transmission, CJT, and two or more transmit-and-receive points, TRPs, (103a-b) the system comprising: - the network node configured to obtain information on frequency update from each of the two or more TRPs; - the network node configured to predict one or more phase offsets between TRPs, based on historical information on the DL communications channels transmitted from the TRPs to the WD and the corresponding frequency updates; and - the network node configured to send to the two or more TRP the corresponding predicted phase offset; - the two or more TRPs configured to perform a pre-compensation of the corresponding DL communications channel, based on the received phase offsets.

22. The system according to claim 21, further comprising: - the two or more TRPs configured to obtain information on frequency update; - the two or more TRPs configured to perform a pre-compensation of the corresponding DL communications channel, based on the information on the frequency update.

23. The system according to any of claim 21 or 22, wherein the network node is configured to predict a phase offset by DL communications channel prediction.

24. The system according to any of claims 21-23, wherein the network node is configured to predict a phase offset by - selecting, as a reference TRP, one of the two or more TRPs; - for each of the remaining TRPs, determining a relative communications channel associated with the reference TRP and each of the remaining TRPs, wherein a relative communications channel is based on the DL communications channel of the reference TRP and the DL communications channel of the corresponding remaining TRP; - predicting the phase offset for each of the relative communications channels.

25. The system according to any of claims 21-24, further comprising: - the network node configured to obtain, for each TRP, a first time interval, wherein the first time interval is indicative of the time interval between the point in time wherein the  frequency update is determined and the point in time the frequency update is compensated in the network node; - the network node configured to send the determined first time interval value to the corresponding TRP; - each TRP configured to determine a fifth time interval, wherein the fifth time interval value is equal or higher than the first interval, wherein the fifth time interval value is indicative of the time interval between the point in time wherein the frequency update is determined and the point in time wherein the frequency update is applied.

26. The system according to claim 25, further comprising: - the network node configured to obtain as first time interval value for all the two or more TRPs the highest between the first time intervals of the two or more TRPs.

27. The system according to any of claims 21-24, wherein the network node is configured to: - determine a third time interval value, wherein the third time interval value is indicative of the time interval between the point in time wherein the frequency update is applied in the TRP and the point in time wherein the pre-compensation is performed; - estimate a phase offset of each TRP during the third time interval; - apply a scheduling restriction to the WD from the two or more TRPs during a fourth time interval, if the estimated phase offset in the third time interval is higher than a threshold.

28. The system according to any of claims 21-27, wherein the frequency update of a TRP comprises one or more of a value of the frequency update of the TRP, a timestamp associated with the frequency update indicating the point in time the frequency update is applied to the TRP, a value indicating time and occurrence of further frequency updates.

29. A computer program comprising instructions which, when run in a processing unit of a network node, cause the network node to perform the method according to any of claims 1 to 10.

30. A computer-readable data carrier having stored thereon the computer program according to claim 29.

31. A data carrier signal carrying the computer program according to claim 29.

Images