WO2020019336A1

WO2020019336A1 - Method and apparatus for transmiting demodulation reference signal

Info

Publication number: WO2020019336A1
Application number: PCT/CN2018/097636
Authority: WO
Inventors: Nuan SONG
Original assignee: Nokia Shanghai Bell Co., Ltd.
Priority date: 2018-07-27
Filing date: 2018-07-27
Publication date: 2020-01-30
Also published as: CN112544046B; CN112544046A

Abstract

Embodiments of the present disclosure provide apparatus, methods and computer readable storage medium for transmission of a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. A transmitting apparatus comprises a first precoder configured to perform a non-linear precoding operation on a plurality of streams of data signals addressed to a plurality of receiving apparatus; and a second precoder configured to perform a linear precoding operation on respective demodulation reference signals for the plurality of streams of data signals. The second precoder is configured to perform a feedback filtering operation with a same feedback filter as the first precoder, and perform a feedforward filtering operation with a same feedforward filter as the first precoder. The second precoder is further configured to add tilting phases to the respective demodulation reference signals in the feedback filtering operation.

Description

METHOD AND APPARATUS FOR TRANSMITING DEMODULATION REFERENCE SIGNAL

TECHNICAL FIELD

The non-limiting and exemplary embodiments of the present disclosure generally relate to a signal transmission in multi-user multiple-input multiple-output (MIMO) system, and specifically to methods, apparatus and computer readable storage medium for a pre-coding procedure for DeModulation Reference Signal (DMRS) .

BACKGROUND

Non-linear precoding has been identified as a promising interference cancellation technique for new radio (NR) system. For example, with full CSI (Channel State Information) at the transmitter side, “Dirty-Paper” Coding (DPC) technique that relies on a pre-subtraction of the non-causally known interference can achieve the maximum sum rate of the system and provide the maximum diversity order. Tomlinson-Harashima Precoding (THP) is a simplified and efficient version of DPC, which is less computationally demanding and thus more attractive for practical implementation. Non-linear precoding such as THP is able to provide a significantly enhanced system performance as compared to linear precoding, especially for correlated channels where the subspaces of user equipments (UEs) are overlapped.

However, one of the key technical challenges to implement non-linear precoding is to design specific DMRS to facilitate appropriate receive combining at the UE. In a traditional way, a sequence of DMRS is transmitted together with data using the same precoding, so that the sequence of DMRS would go through the same channel as the data and the UE is able to estimate and recover the channel using such DMRS to design receive combining as well as to demodulate the data. However, it is no longer suitable for non-linear precoding, as non-linear processing at the transmitter leads to a corrupted DMRS and the receiver cannot estimate the channel directly. Additionally, in NR MIMO systems, the gNB is always mounted with a large array of antennas and tries to support a large number of UEs. The DMRS with orthogonal resources may lead to a large overhead. Therefore, specific DMRS with low overhead is a key problem to be solved for non-linear precoding.

SUMMARY

The present disclosure is going to solve the aforementioned problems by proposing a tilted non-orthogonal DMRS scheme for the non-linear precoding procedure in NR MIMO systems, in order to reduce the DMRS overhead and enhance the channel estimation performance for the data demodulation. Other features and advantages of embodiments of the present disclosure will also be understood from the following description of specific embodiments when read in conjunction with the accompanying drawings, which illustrate, by way of example, the principles of embodiments of the present disclosure.

In a first aspect of the disclosure, there is provided an apparatus for transmitting a demodulation reference signal (DMRS) in a multi-user multiple-input multiple-output (MIMO) system. The apparatus comprises a first precoder configured to perform a non-linear precoding operation on a plurality of streams of data signals addressed to a plurality of receiving apparatus. The apparatus further comprises a second precoder configured to perform a linear precoding operation on respective demodulation reference signals for the plurality of streams of data signals. The second precoder is configured to perform a feedback filtering operation with a same feedback filter as the first precoder, and perform a feedforward filtering operation with a same feedforward filter as the first precoder. The second precoder is further configured to add tilting phases to the respective demodulation reference signals in the feedback filtering operation.

In one embodiment, the apparatus can further comprise a transmitter, which is configured to multiplex the precoded respective demodulation reference signals with the precoded plurality of streams of data signals, and transmit the multiplexed demodulation reference signals and streams of data signals to the plurality of receiving apparatus.

In one embodiment, the first precoder can be configured to perform the non-linear precoding operation according to a Tomlinson-Harashima (THP) scheme.

In one embodiment, the apparatus can further comprise a tilting phases determining unit, which is configured to determine tilting phases for respective demodulation reference signals through an iterative algorithm, to minimize power increases of respective demodulation reference signals.

In one embodiment, the apparatus can further comprise a transmitting unit, which is configured to send an indication that a phase tilting operation is applied to the respective demodulation reference signals.

In one embodiment, the apparatus can further comprise a transmitting unit, configured to send an indication of signs of receive weights of the respective demodulation reference signals.

In one embodiment, the apparatus can further comprise a determining unit, which is configured to determine parameters of the feedback filter and the feedforward filter, according to channel status information from the receiving apparatus.

In a second aspect of the disclosure, there is provided an apparatus for receiving a demodulation reference signal in a multiple-input multiple-output (MIMO) system. The apparatus comprises a receiver configured to receive a demodulation reference signal from a transmitting apparatus. The apparatus further comprises a estimator configured to derive a channel estimation based on the received demodulation reference signal and a corresponding original demodulation reference signal. The estimator is further configured to recover receive weights for data signals multiplexed with the received demodulation reference signal from the channel estimation, based on a tilted phase of the received demodulation reference signal.

In one embodiment, the estimator can be configured to abstract amplitudes of the channel estimation as the receive weights.

In one embodiment, the apparatus can further comprise a demodulator, which is configured to demodulate the data signal according to the receive weights.

In one embodiment, the apparatus can further comprise a receiving unit, which is configured to receive an indication of a phase tilting operation applied to the demodulation reference signal at the transmitting apparatus.

In one embodiment, the receiving unit can be further configured to receive an indication of signs of the receive weights, and the estimator is further configured to recover the receive weights based on the sign of the receive weights.

In one embodiment, the estimator can be further configured to estimate the tilted phase applied the demodulation reference signal, and recover the receive weight based on the estimated tilted phase.

In one embodiment, the apparatus can further comprise a transmitter unit, which is configured to transmit channel status information to the receiving apparatus.

In a third aspect of the disclosure, there is provided a method for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. The method comprises performing a non-linear precoding operation on a plurality of streams of data signals addressed to a plurality of receiving apparatus. The method further comprises performing a linear precoding operation on respective demodulation reference signals for the plurality of streams of data signals. Performing the linear precoding operation comprises performing a feedback filtering operation with a same feedback filter as the non-linear precoding operation, and performing a feedforward filtering operation with a same feedforward filter as the non-linear precoding operation. Performing the linear precoding operation further comprises performing a phases tilting operation to the respective demodulation reference signals in the feedback filtering operation.

In a fourth aspect of the disclosure, there is provided a method for receiving a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. The method comprises receiving a demodulation reference signal from a transmitting apparatus. The method further comprises deriving a channel estimation based on the received demodulation reference signal and a corresponding original demodulation reference signal. The method further comprises recovering receive weights for data signals multiplexed with the received demodulation reference signal from the channel estimation, based on a tilted phase of the received demodulation reference signal.

In a fifth aspect of the disclosure, there is provided an apparatus for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. The apparatus may comprise a processor and a memory communicatively associated with the processor. The memory may have computer coded instructions stored therein, said instructions when executed by the processor causing the apparatus to perform the method according to the third aspect of the present disclosure.

In a sixth aspect of the disclosure, there is provided an apparatus for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. The apparatus may comprise a processor and a memory communicatively associated with the processor. The memory may have computer coded instructions stored therein, said instructions when executed by the processor causing the apparatus to perform the method according to the fourth aspect of the present disclosure.

In a seventh aspect of the disclosure, there is provided an apparatus for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. The apparatus may comprise process means adapted to perform any method in accordance with the third aspect of the disclosure.

In an eighth aspect of the disclosure, there is provided an apparatus for receiving a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system. The apparatus may comprise process means adapted to perform any method in accordance with the fourth aspect of the disclosure.

In a ninth aspect of the disclosure, there is provided a computer readable storage medium, on which stored computer code instructions. When the computer code instructions are executed on at least one processor, at least one processor is caused to carry out the method according to the third aspect of the disclosure.

In a tenth aspect of the disclosure, there is provided a computer readable storage medium, on which stored computer code instructions. When the computer code instructions are executed on at least one processor, at least one processor is caused to carry out the method according to the fourth aspect of the disclosure.

According to the various aspects and embodiments as mentioned above, an issue of DMRS transmission in a MIMO system can be resolved.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other aspects, features, and benefits of various embodiments of the present disclosure will become more fully apparent, by way of example, from the following detailed description with reference to the accompanying drawings, in which like reference numerals or letters are used to designate like or equivalent elements. The drawings are illustrated for facilitating better understanding of the embodiments of the disclosure and not necessarily drawn to scale, in which:

FIG. 1 illustrates a block diagram of an orthogonal DMRS scheme;

FIG. 2 illustrates a block diagram of a spatial multiplexing DMRS scheme;

FIG. 3 illustrates a block diagram of a DMRS scheme with power back-off;

FIG. 4 illustrates a block diagram of tilted non-orthogonal DMRS scheme according to at least part of embodiments of the present disclosure;

FIG. 5 illustrates a block diagram of another model of non-linear precoding;

FIG. 6 illustrates a flowchart of a signal transmission procedure with a tilted non-orthogonal DMRS scheme according to at least part of embodiments of the present disclosure;

FIG. 7 illustrates a flowchart of a procedure for finding tilting phases for a DMRS to be transmitted according to at least part of embodiments of the present disclosure;

FIGs. 8-10 depict results of simulations of the proposed scheme in comparison with existing solutions;

FIG. 11 depicts DMRS patterns of a simulation of the proposed scheme in comparison with existing solutions;

FIGs. 12-13 depict results of another simulations of the proposed scheme in comparison with existing solutions;

FIG. 14 depicts a proposed non-orthogonal DMRS pattern to support more than 8 ports throughput according to the proposed scheme of this disclosure;

FIG. 15 illustrates a flowchart of a method according to at least part of embodiments of the present disclosure;

FIG. 16 illustrates a flowchart of a method according to at least part of embodiments of the present disclosure; and

FIG. 17 illustrates a simplified block diagram of an apparatus according to at least part of embodiments of the present disclosure.

DETAILED DESCRIPTION

Hereinafter, the principle and spirit of the present disclosure will be described with reference to illustrative embodiments. It should be understood, all these embodiments are given merely for one skilled in the art to better understand and further practice the present disclosure, but not for limiting the scope of the present disclosure. For example, features illustrated or described as part of one embodiment may be used with another embodiment to yield still a further embodiment. In the interest of clarity, not all features of an actual implementation are described in this specification.

References in the specification to “one embodiment, ” “an embodiment, ” “an example embodiment, ” etc. indicate that the embodiment described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs. For example, the term “receiving apparatus” used herein may refer to any terminal device or user equipment (UE) having wireless communication capabilities, including but not limited to, mobile phones, cellular phones, smart phones, or personal digital assistants (PDAs) , portable computers, and the like. Furthermore, user equipment that is not mobile may also readily employ embodiments of the present invention. In the following description, the terms “user equipment” , “UE” and “terminal device” may be used interchangeably. Similarly, the term “base-station apparatus” may represent a base station (BS) , a node B (NodeB or NB) , an evolved NodeB (eNodeB or eNB) , a gNodeB (gNB) and a relay node (RN) and so forth.

For illustrative purposes, several embodiments of the present disclosure will be described in the context of a NR MU MIMO system. Those skilled in the art will appreciate, however, that the concept and principle of the several embodiments of the present disclosure may be more generally applicable to other wireless networks, for example a third generation Long Term Evolution (3G-LTE) network, a fifth generation (4G) network, 4.5G LTE, or a future network (e.g. 5G network) .

The non-limiting and exemplary embodiments of the present disclosure relate to DMRS precoding in an MU-MIMO system. As mentioned above, traditional DMRS precoded together with data will be corrupted and cannot be used directly to recover the receive weighting. There are several existing schemes to solve the problem. One scheme proposes to design DMRS with linear precoding. But in this scheme, DMRS should use orthogonal resources (RS) , which imposes a large overhead for the system. Figure 1 illustrates an exemplary block diagram of such orthogonal DMRS scheme. As shown in Figure 1, the data signal 110 goes through a non-linear precoding procedure comprising a feedforward filter 150 and a feedback loop, which comprises a feedback filter 130 for interference cancelation and a moduler 140 for power increasing. The DMRS sequence 120 does not go through a feedback loop, but go through the same feedforward filter 150 as the data signal 110. The RS used for the DMRS is orthogonal to the RS used for data signal.

Another scheme is designed to reduce the overhead. It proposes that the spatial multiplexing DMRS (can be transmitted in the same resource as the data signal) undergoes a same non-linear precoding together with a data signal. Figure 2 illustrates an exemplary block diagram of a spatial multiplexing DMRS scheme. As shown in Figure 2, the DMRS shares a same feedforward filter 250 and a feedback loop as the data signal. But in this scheme, the gNB and the UE require a DMRS corrector and a perturbation vector adder, respectively, which imposes a high implementation complexity at the both ends.

A further another scheme proposes that the DMRS goes through a feedback loop similar as the feedback in non-linear precoding of data signals for interference cancelation, without being processed by a modulo operation. Figure 3 illustrates an exemplary block diagram of such scheme. In this scheme, non-orthogonal DMRS is possible. However, the feedback loop results in a significant power increase of the DMRS and accordingly a power back-off is required. Using DMRS with a reduced transmit power leads to a degraded performance in channel estimation and data demodulation.

There are some schemes applying phase rotations on both DMRS and data signals to control the power of the DMRS. A fast phase search method is proposed, which, however, does not ensure a convergence. The phase rotation is designed for the DMRS but should not for the data. Thus, the phase rotation implemented on the data signal may cause power imbalance due to a rotated constellation. How the phase can be recovered at the UE has not been discussed and is not clear in these schemes.

There are some other schemes suggesting that the gNB directly indicates the UE with respect to the receive combining or weighting, so that a specific DMRS would not be required. However, this requires significant overhead for transmitting such indicators, especially for the massive MIMO systems with many UEs.

The non-limiting and exemplary embodiments of the present disclosure propose a tilted non-orthogonal DMRS scheme for non-linear precoding. The main concept of the invention is shown in Figure 4. The gNB generates tilted non-orthogonal DMRS to reduce the transmission power and alleviate the performance loss due to the power back-off. The tilted DMRS is generated by adding tilting phases (represented as a tilting matrix T in Figure 4) to the original DMRS symbols. As the tilted DMRS will go through a feedback loop, which is an interference cancellation part, the original DMRS can be non-orthogonal to the data signal.

In order to ensure a best estimation performance, the gNB may partially indicates the UE the receive weighting D _k, k=1, ...K, i.e., the signs of the receive weights. This procedure can be named as a partial weight indication. For example, due to a nature of the THP precoding scheme, the receive weights for data streams are always of a real number. Indication for the signs of the receive weights would not occupy much resources.

Additionally, the gNB may indicate UE in a downlink control channel on the transmission of the tilted DMRS, so that the UE is able to carry out a specific estimation processing for data demodulation. The UE can have a simple estimation block to recover the receive weights via the original sequence of DMRS symbols based on tilted phases of the received sequence of DMRS symbols. Optionally, the UE can recover the tilting phases.

System model

Figure 4 shows a transceiver block diagram of the tilted DMRS scheme for the non-linear THP precoder. It is assumed there are K UEs in a MU MIMO system 400, and each UE has

antennas

480, 482. There are M _T antennas 470 at the gNB and in total

streams multiplexed via the anntennas 470, where the gNB transmits r _k streams of data signals to a UE k. The matrix

(of M _T row and r columns) corresponds to the feedforward filter 450, and the matrix

(of r row and r columns) corresponds to the feedback filter 430 for interference preprocessing.

According to the proposal of the present disclosure, DRMS for respective data signals can undergo a precoding procedure different from that for their respective data signals. In exemplary embodiments, the data signals 410 of the r streams (represented as

) can undergo a non-linear precoding operation, such as a regular THP procedure. For example, as shown in Figure 4, the THP procedure can comprise a feedback loop involving a feedback filtering operation with the feedback filter 430 and a modulo operation with a moduler 440, and a feedforward filtering operation with the feedforward filter 450. The r sequences of DMRS symbol 420 (represented as

) for the respective r streams of data signals can undergo a linear precoding operation. For example, as shown in Figure 4, the DMRS 420 can go through a feedback loop involving a same feedback filter 430, but without being processed by the modulo operation 440. Instead, the DMRS 420 go through a phase tilting operation in the feedback loop, in order to reduce the power increase of the DMRS. For example, the feedback loop may comprise a phase tilting function entity 460 of T and an inverse phase tilting function entity 462 of T ^-1.

At the receiver, each UE k receives applies specific DMRS d _k for its data signal s _k, to estimate receive weights D _k, so that the non-linearly precoded data signal can be correctly demodulated.

In this example, each sequence of DMRS symbols after phase tilting can be written as

where

and θ _k∈ [0, 2π) is a tilting phase for specific d _k. The phase tilting operation at the transmitter side can be constructed as a tilting matrix T (of r row and r columns) , which is diagonal and can be denoted by

Then, the data model for each sequence of DMRS symbols after the feedback loop involving the phase tilting can be represented as a matrix form as

As the feedforward filter P is (semi) -unitary and does not impose any power increase, the key problem to be solved is to design the tilting phases in T so that each transmitted sequence of DMRS symbols x _k has a minimum power increase. This design will be discussed in detail later.

It should be noted that although a THP precoder is applied in the system of Figure 4, the system can apply other equivalent precoding model. For example, another model of non-linear precoding as shown in Figure 5 can be applied for the precoding for the data signal. As illustrated in Figure 5, the THP can be reformulated into an equivalently linearized model, where v is the perturbed vector and its k-th element are defined as follows:

Figure 6 illustrates an exemplary procedure of the tilted DMRS scheme. As shown in Figure 6, a UE 602 can report CSI information to a gNB 601, at step 610. After the gNB 601 acquires the CSI from UE 602, the gNB 601 can design non-linear precoding matrices B and P based on the CSI, at step 620. With the CSI, the gNB 601 can learn the current channel status for channels between the UE 602 and a gNB 601. It should be appreciated that, in addition to or alternative to the CSI, the non-linear precoding matrices B and P can be designed based on other information, such as system information.

Then, the gNB 601 can generate a tilted DMRS for the UE 602 at step 630, multiplex the tilted DMRS with data at step 640, and then transmit the multiplexed DMRS with data to the UE 602 at step 660. In some embodiments, the gNB 601 can further send an indication to the UE 602, to inform the UE of the transmission of the tilted DMRS, as shown at step 650. In some embodiments, the gNB 601 can further send indications of partial weights of the receive processing to the UE 602.

After the UE 602 received the tilted DMRS multiplexed with data transmitted from the eNB 601 at step 660, the UE can estimate a channel function and recover the receive weights from the channel estimation by virtue of the received DMRS at step 670. In some embodiments, the UE 602 may receive an indication of the transmission of the tilted DMRS, and perform the estimation and recovering in connection with phase tilting of the transmitted DMRS. In some embodiments, the UE 602 may receive indications of partial weights, and perform the estimation and recovering further based on the indications of partial weights. The UE 602 may further recover the tilting phases of the transmitted DMRS to correct the recovered receive weights, at step 680. Then, as shown at step 690, the UE 602 weights and demodulates the received data signals according to the recovered receive weights.

Now, details of these steps will be described with respect to some particular embodiments. In the generation of tilted DMRS at gNB 601, one target is to design tilted non-orthogonal DMRS that goes through the feedback loop of the THP with a minimum power increase. As mentioned above, the gNB can be configured to design the tilting phases in T, so that each transmitted sequence of DMRS symbols x _k for respective stream of data signal has a minimum power increase. In some embodiments, the gNB 601 can determine tilting phases jointly for sequences of DMRS symbols for all streams multiplexed via its anntennas (such as 470) . An optimization problem as to jointly minimize the maximum transmission power of the DMRS layer can be defined as,

where x (m) refers to the m-th element of the vector x. A DMRS layer corresponds to a sequence of DMRS symbols for a stream of transmitted data signal. For example, the n-th layer DMRS refers to the n-th sequence of DMRS symbol for the n-th stream of data signal.

This is a non-linear least squares problem. Such a problem is non-convex and has multiple local minima. Therefore, this disclosure proposes to apply an iterative algorithm with a fast convergence to obtain the desired tilting phase, by fixing other sequences of DMRS symbols in each iteration.

Figure 7 illustrates a process for finding such tilting phases for the transmitted DMRS with an iterative algorithm. As shown at block 730, after the gNB designs the feedback filter B at 710, the process initializes all phases of the sequences of DMRS to 0. In an outer iteration defined by p, for q=1, ..., r, the optimization problem in the Equation 5) can be solved by firstly formulating

where

is the k-th row of the matrix

For the tilting phases Tat the current iteration defined by p, the layer of the DMRS symbol that maximizes g _k can be obtained at block 750, which is denoted by n. In other words, at block 750, the process can calculate eachg _k, k=1, ..., r, with the equation 6) , and find the maximum among g _k. The max g _k can be represented as g _n. Then, a Lagrange multiplier can be applied to solve the minimization of the g _n and obtain θ _q as

where x (m) refers to the m-th element of a vector x, and ^*is the complex conjugate.

The calculation of the equation 7) can be carried out iteratively for all q=1, ..., r with updated θ _q for the n-th layer DMRS with the largest g _k, k=1, ..., r (which is actually g _n) , as shown at

blocks

760, 770 and 780. In this inner iteration, the tilted phase for the n-th layer DMRS can be calculated firstly, by fixing tilted phases of other layers of DMRS. Then, the titled phase θ _q of each of the other q-th layers of DMRS (q=1, ..., r and q≠n) can be calculated and updated iteratively, by fixing the tilted phase of the n-th layer DMRS and other layers of DMRS.

Once all the inner iteration of q is completed at step 780 ( “Yes” ) , an outer iteration p=p+1 is continued, from finding the layer that maximizes g _k at block 750. Usually the number of outer iterations N can be pre-defined. Then, an optimized tilting phase matrix T can be determined. By use of this process, an efficient scheme can be provided to solve the non-linear least squares problem ensures a fast convergence, which delivers closed-form solutions iteratively and does not requires any search procedure for tilted phases.

As the original DMRS is tilted, UE needs to know such a behavior at the gNB, and carries out appropriate receive processing accordingly. In some embodiments, the gNB may form one indication, e.g., in DCI (Downlink Control Information) , to inform the UE that a tilted DMRS is transmitted. Depending on network where embodiments of the disclosure are applied, different channels or signaling can be used for this purpose. For example, the indication may be carried by other messages addressed to specific UE. In some other examples, the indication may be broadcasted by the eNB in a system message or be multicasted to a group of UEs. In any case, the tilted DMRS scheme of this disclosure bypasses the direct indication of receive weights to UE, which reduces the utilization of the control channel.

Additionally, to avoid phase ambiguity during the channel estimation via DMRS at the UE side, signs of the receive weights for streams of data signals may also be indicated to UEs. In embodiments where a non-linear precoding is configured for data signals according to the THP scheme, the receive weights would be real numbers. As such, it would occupy little resources for informing the signs of the receive weights, especially in comparison with the direct indication of receive weights to UE.

At the UE side, a receiver of a UE can receive a DMRS multiplexed with a data signal specific to the UE. The DMRS can be de-multiplexed for channel estimation. In the embodiments which take THP scheme for the non-linear precoding for data signals, the total received DMRS (denoted as y _d) can be represented as,

where d is the original DMRS, H is the channel response, and n _d is the Additive White Gaussian Noise (AWGN) .

According to a design theory of THP, the inverse of the receive weights D can be derived from the channel response, a filtering function of the feedforward filter and a filtering function of the feedback filter, as HPB ^-1=D ^-1, where the diagonal elements of D ^-1 are real and correspond to the inverse of receive weights D. As the original DMRS d is known, D ^-1T can be firstly estimated from the equivalent expression

where D ^-1T is defined as Λ.

Then, Least Squares (LS) channel estimation can be applied and each diagonal element λ _k of Λ can be obtained. The data model for each DMRS layer can be rewritten as

where

is the total receivedDMRS sequence specific to the k-th stream of data signals, d ^(k) is the sequence of original DMRS symbols for the k-th stream of data signals with length N _RS, and

is AWGN for the k-th stream of data signals. The LS solution is

Since λ _k consists of the receive weight of the k-th stream and the tilted phase, UE need to correctly recover the receive weights for data demodulation. The receive weights can be recovered based on tilted phases of the DMRS, to avoid phase ambiguity.

According to the definition of Λ in Equation 9) , the estimation of its each element λ _k can thus be written as

For a general complex variable, λ _k can also be written in the form of phase and amplitude as

where |λ _k| denotes its amplitude and

is the angle of λ _k. Thus, the amplitude of λ _k can be abstracted directly as the receive weight.

In some embodiments, in order to link the estimate λ _k to the receive weight

(which is real) and θ _k as well as to avoid phase ambiguity, the signs of the receive weights sign (D _k) can be indicated to a UE as discussed above. Then, based on the received signs of the receive weights sign (D _k) , the UE can recover the receive weight by deriving

and mapping the estimated phase

to θ _k in the range of [0, 2π) .

By applying the tilted DMRS scheme as proposed in this disclosure, the resources used by the tilted DMRS can be non-orthogonal to the resources used by corresponding data signals. This reduces the overhead for DMRS, especially in comparison with traditional orthogonal DMRS scheme. Meanwhile, it reduces the transmission power from the DMRS by virtue of the phase tilting operation. The transceiver implementation in both gNB side and UE side can be simplified.

The performance of the proposed tilted DMRS scheme for MU-MIMO can be evaluated using Monte-Carlo simulations and compared with the existing solutions. The simulation setup is shown in Table 1. The proposed scheme is named as “Tilted DMRS” . The scheme “Orthogonal DMRS” refers to the scheme shown in Figure 1, and the scheme “DMRS with power back-off” refers to the scheme shown Figure 3. For the cases of “Tilted DMRS” and “DMRS with power back-off” , non-orthogonal (random) sequences are applied for DMRS. The scheme “Non-orthogonal DMRS” corresponds to the case where non-orthogonal DMRS are applied in the structure of Figure 1, which corresponds to the worst case, in order to show the effect from reduced resources of DMRS at the cost of performance degradation.

Table 1: Simulation setup

Figure 8 depicts the Cumulative Distribution function (CDF) of the power reduction in percentage using the “tilted DMRS” scheme as compared to the “DMRS with power back-off” . The power of the “tilted DMRS” scheme is calculated by

where N _RS is the length of the DMRS sequence, and d _n corresponds to the n-th layer of the DMRS symbols for all DMRS layers. Similarly, The power of the classic “DMRS with power back-off” can be written as

Then the power reduction in percentage is obtained by

From Figure 8, it can be observed that the tilted DMRS scheme effectively reduces the transmission power of the original DMRS.

Figure 9 shows the Normalized Mean Square Error (NMSE) of the estimated receive weights as a function of SNR for different schemes. It can be observed that the “Orthogonal DMRS” scheme shows the best performance. But as discussed above, the “Orthogonal DMRS” scheme requires orthogonal resources and DMRS sequences to ensure zero interference, which may impose a large overhead for a massive number of users in the system. The proposed “Tilted DMRS” scheme that applies non-orthogonal DMRS is inferior to the “orthogonal DMRS” scheme, but greatly reduces the overhead. For the “Orthogonal DMRS” scheme with the number of transmitted antennas 32, a DMRS sequence length of a least 32 has to be applied to ensure a full-rank channel estimation. The proposed “Tilted DMRS” scheme with a reduced DMRS sequence length (16 as compared to 32) performs almost the same as the use of the length-32 DMRS in the “Orthogonal DMRS” scheme. If the scheme shown in Figure 1 applies the same non-orthogonal DMRS of length 32 as in the proposed “Tilted DMRS” scheme, a large performance degradation is observed. If the sequences of DMRS symbols to be transmitted are not tilted, i.e., the scheme “DMRS with power back-off” , a performance degradation due to the increased DMRS power can be observed.

The throughput performance using the tilted DMRS scheme and compare to the case with power back-off is further evaluated, where different SNRs of DMRS are considered. From Figure 10, it can be observed that by controlling the transmit power using the “Tilted DMRS” scheme, a much better throughput performance than the case of the “DMRS with power back-off” scheme can be obtained, because of the more accurate estimation at the UE side. The performance can be quite close to the ideal DMRS case, i.e. the “Orthogonal DMRS” scheme.

To illustrate the DMRS overhead reduction of the proposed “Tilted DMRS” scheme, the LTE-A (Long Term Evolution-Advanced) DMRS design can be taken as a baseline. In LTE-A, each RB (Resource Block) consists of 12 subcarriers and 14 symbols, i.e., 168 Res (Resource Elements) . It is able to support 8 ports and accordingly the transmission rank of 8. To achieve the number of layers 8, 24 REs are required, which leads to an overhead of 14.28%. To be specified, the corresponding REs have hybrid CDM (Code Division Multiplexing) + FDM (Frequency Division Multiplexing) DMRS patterns, including 2 frequency slots (forming 2 CDM groups) and 4 length-4 OCC (Orthogonal Cover Code) sequences, which span 3 subcarriers and 4 symbols. The OCC is generated by Walsh codes, i.e., [1 1 1 1; 1 -1 1 -1; 1 1 -1 -1; 1 -1 -1 1] for length 4, namely (OCC-4 a) . It can be observed from the upper DMRS pattern figure of Figure 11 that the CDM group 1 is orthogonal in frequency to CDM group 2.

According to the proposed “Tilted DMRS” scheme, non-orthogonal code sequences are supported, and the interference caused among DMRS layers can be largely mitigated by the feedback loop filter. In one example to support 8 ports, two groups of OCC-4 codes could be applied, i.e., [1 1 1 1; 1 -1 1 -1; 1 1 -1 -1; 1 -1 -1 1] , namely (OCC-4 a) ) , and [1 -1 -1 -1; -1 1 -1 -1; -1 -1 1 -1; -1 -1 -1 1 ] , namely (OCC-4 b) ) . The OCC-4 a) and OCC-4 b) are applied in the same time and frequency resources, which are non-orthogonal between them. An example of such DMRS patterns can be shown in the lower DMRS pattern figure of Figure 10, where OCC-4a) corresponds to the LTE case, and OCC-4b) refers to the proposed non-orthogonal case. As the CDM group 1 and the CDM group 2 occupy the same time and frequency resources, i.e., non-orthogonal to each other, the above the proposed DMRS design can also support 8 layers but with an overhead 7.14% (50%reduction) .

Two schemes ( “Orthogonal DMRS” and the proposed “Tilted DMRS” ) are further evaluated with a simulation set up different from the previous part. Figure 12 shows the NMSE performance of the channel estimation using “Orthogonal DMRS” and “Tilted DMRS” . Figure 12 shows a throughput performance of the proposed “Tilted DMRS” scheme. It can be observed that there is a slight performance degradation with 50%overhead reduction.

Table 2: Simulation setup to test DMRS in the example of 8 ports

If more than 8 ports are supported, without sacrificing the density, more REs are required. As shown in Figure 14, by using the proposed “Tilted DMRS” scheme, 24 REs can be applied to support 16 ports. Two frequency slots are used to support additional layers. To be specified, there may be 4 CDM groups using two OCC-4 codes. CDM group 1 and CDM group 2 are orthogonal in frequency and reuse the OCC-4 a) . The CDM group 3 and CDM group 4 are orthogonal and apply OCC-4 b) . The

CDM groups

3 and 4 use the same time and frequency resources as the

CDM groups

1 and 2, i.e., non-orthogonal to each other. In this case, the same density and the overhead can be maintained as in LTE-A, but more ports are supported.

Therefore, it can be concluded that the proposed “Tilted DMRS” scheme not only provides a robust channel estimation performance, but also reduces the transmission overhead.

Reference is now made to Figure 15, which illustrates a flowchart of a method for transmitting DMRS to a plurality of receiving apparatus according to some embodiments of the present disclosure. The method 1500 can be performed at a base station provided with multiple antennas or antenna arrays, such as a gNB 401. The plurality receiving apparatus can be multiple UEs, such as UE ₁, …, UE _K which is denoted as 402 in common in Figure 4. As shown in Figure 15, the method 1500 comprises performing a non-linear precoding operation on a plurality of streams of data signals addressed to the plurality of receiving apparatus, at block 1510. The method 1500 further comprises performing a linear precoding operation on respective DMRS for the plurality of streams of data signals, at block 1520. The linear precoding operation comprises a feedback filtering operation with a same feedback filter as the non-linear precoding operation, and a feedforward filtering operation with a same feedforward filter as the non-linear precoding operation. The feedback filtering operation in the linear precoding operation comprises a phases tilting operation to the respective DMRS. It would be appreciated that the steps of the method 1500 do not have to be performed in the exact order disclosed. For example, the precoding operation on the data signals and the precoding operation on the DMRS can be performed at the same time. In an embodiment, the non-linear precoding operation can be implemented as the non-linear precoding operation for the data signals 410 of the r streams, as shown in Figure 4. In the embodiment, the linear precoding operation can be implemented as the linear precoding operation for the r sequences of DMRS symbol 420, as shown in Figure 4.

Reference is now made to Figure 16, which illustrates a flowchart of a method for receiving DMRS according to some embodiments of the present disclosure. The method 1600 can be performed at a UE provided with multiple antennas or antenna arrays, such as one UE 402 as shown in Figure 4. As shown in Figure 14, the method 1600 comprises receiving a demodulation reference signal from a transmitting apparatus at block 1610. The transmitting apparatus can be a base station, such as the gNB 401 as shown in Figure 4. Then, a channel estimation can be derived based on the received DMRS and a corresponding original DMRS, at block 1620. From the derived channel estimation, receive weights for data signals multiplexed with the received DMRS can be recovered based on a tilted phase of the received DMRS, at block 1630.

Reference is now made to Figure 17, which illustrates a simplified block diagram of an apparatus 1700 according to some embodiments of the present disclosure. The apparatus may be embodied in/as a base station in a MIMO system, which can communicate with multiple UEs in a same channel simultaneously. For example, the base station may be a gNB operating in an MU-MIMO system as shown in Figure 4. In another embodiment, the apparatus 1700 may be embodied in/as another entity at a user side, such as a UE, which can be communicatively connected to the base station. The apparatus 1700 is operable to carry out the exemplary methods 1500 and/or 1600 as described with reference to Figure 15 and/or Figure 16, and possibly any other processes or methods. It is also to be understood that any one of the

methods

1500 and 1600 are not necessarily carried out completely by the apparatus 1700. Some steps of the

methods

1500 and 1600 may be performed by one or more other entities.

The apparatus 1700 may comprise at least one processor 1701, such as a data processor (DP) and at least one memory (MEM) 1702 coupled to the processor 1701. The apparatus 1700 may further comprise a transmitter TX and receiver RX 1703 coupled to the processor 1701. The MEM 1702 stores a program (PROG) 1704. The PROG 1704 may include instructions that, when executed on the associated processor 1701, enable the apparatus 1700 to operate in accordance with the embodiments of the present disclosure, for example to perform the

method

1500 or 1600. A combination of the at least one processor 1701 and the at least one MEM 1702 may form processing means 1705 adapted to implement various embodiments of the present disclosure.

Various embodiments of the present disclosure may be implemented by computer program executable by the processor 1701, software, firmware, hardware or in a combination thereof. The processors 1701 may be of any type suitable to the local technical environment, and may include one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors DSPs and processors based on multicore processor architecture, as non-limiting examples. The MEMs 1702 may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as semiconductor based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory, as non-limiting examples.

The transmitter TX and receiver RX 1703 can have multiple antennas that utilize various transmission diversity schemes for supporting the MU-MIMO technology. For example, the apparatus 1700 can comprise two transmit antennas, or four transmit antennas that support beamforming as illustrated in Figure 4.

In addition, the present disclosure may also provide a carrier containing the computer program as mentioned above, wherein the carrier is one of an electronic signal, optical signal, radio signal, or computer readable storage medium. The computer readable storage medium can be, for example, an optical compact disk or an electronic memory device like a RAM (random access memory) , a ROM (read only memory) , Flash memory, magnetic tape, CD-ROM, DVD, Blue-ray disc and the like.

The techniques described herein may be implemented by various means so that an apparatus implementing one or more functions of a corresponding apparatus described with an embodiment comprises not only prior art means, but also means for implementing the one or more functions of the corresponding apparatus described with the embodiment and it may comprise separate means for each separate function, or means that may be configured to perform two or more functions. For example, these techniques may be implemented in hardware (one or more apparatuses) , firmware (one or more apparatuses) , software (one or more modules) , or combinations thereof. For a firmware or software, implementation may be made through modules (e.g., procedures, functions, and so on) that perform the functions described herein.

Exemplary embodiments herein have been described above with reference to block diagrams and flowchart illustrations of methods and apparatuses. It will be understood that each block of the block diagrams and flowchart illustrations, and combinations of blocks in the block diagrams and flowchart illustrations, respectively, can be implemented by various means including computer program instructions. These computer program instructions may be loaded onto a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions which execute on the computer or other programmable data processing apparatus create means for implementing the functions specified in the flowchart block or blocks.

Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the subject matter described herein, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment may also be implemented in multiple embodiments separately or in any suitable sub-combination.

It will be obvious to a person skilled in the art that, as the technology advances, the inventive concept can be implemented in various ways. The above described embodiments are given for describing rather than limiting the disclosure, and it is to be understood that modifications and variations may be resorted to without departing from the spirit and scope of the disclosure as those skilled in the art readily understand. Such modifications and variations are considered to be within the scope of the disclosure and the appended claims. The protection scope of the disclosure is defined by the accompanying claims.

Claims

An apparatus for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, the apparatus comprising:

a first precoder configured to perform a non-linear precoding operation on a plurality of streams of data signals addressed to a plurality of receiving apparatus; and

a second precoder configured to perform a linear precoding operation on respective demodulation reference signals for the plurality of streams of data signals,

wherein, the second precoder is configured to perform a feedback filtering operation with a same feedback filter as the first precoder, and perform a feedforward filtering operation with a same feedforward filter as the first precoder, and

wherein, the second precoder is further configured to add tilting phases to the respective demodulation reference signals in the feedback filtering operation.
The apparatus according to Claim 1, further comprises:

a transmitter configured to multiplex the precoded respective demodulation reference signals with the precoded plurality of streams of data signals, and transmit the multiplexed demodulation reference signals and streams of data signals to the plurality of receiving apparatus.
The apparatus according to Claim 1, wherein the first precoder is configured to perform the non-linear precoding operation according to a Tomlinson-Harashima (THP) scheme.
The apparatus according to Claim 1, further comprises:

a tilting phases determining unit, configured to determine tilting phases for respective demodulation reference signals through an iterative algorithm, to minimize power increases of respective demodulation reference signals.
The apparatus according to Claim 1, further comprises:

a transmitting unit, configured to send an indication that a phase tilting operation is applied to the respective demodulation reference signals.
The apparatus according to Claim 1, further comprises:

a transmitting unit, configured to send an indication of signs of receive weights of the respective demodulation reference signals.
The apparatus according to Claim 1, further comprises:

a determining unit configured to determine parameters of the feedback filter and the feedforward filter, according to channel status information from the receiving apparatus.
An apparatus for receiving a demodulation reference signal in a multiple-input multiple-output (MIMO) system, the apparatus comprising:

a receiver configured to receive a demodulation reference signal from a transmitting apparatus; and

a estimator configured to derive a channel estimation based on the received demodulation reference signal and a corresponding original demodulation reference signal,

wherein the estimator is further configured to recover receive weights for data signals multiplexed with the received demodulation reference signal from the channel estimation, based on a tilted phase of the received demodulation reference signal.
The apparatus according to Claim 8, wherein the estimator is configured to abstract amplitudes of the channel estimation as the receive weights.
The apparatus according to Claim 8, further comprises:

a demodulator configured to demodulate the data signal according to the receive weights.
The apparatus according to Claim 8, further comprises:

a receiving unit, configured to receive an indication of a phase tilting operation applied to the demodulation reference signal at the transmitting apparatus.
The apparatus according to Claim 8, wherein the receiving unit is further configured to receive an indication of signs of the receive weights, and

the estimator is further configured to recover the receive weights based on the sign of the receive weights.
The apparatus according to Claim 8, wherein the estimator is further configured to estimate the tilted phase applied the demodulation reference signal, and recover the receive weight based on the estimated tilted phase.
The apparatus according to Claim 8, further comprises:

a transmitter unit configured to transmit channel status information to the receiving apparatus.
A method for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, the method comprising:

performing a non-linear precoding operation on a plurality of streams of data signals addressed to a plurality of receiving apparatus;

performing a linear precoding operation on respective demodulation reference signals for the plurality of streams of data signals,

wherein, performing the linear precoding operation comprises performing a feedback filtering operation with a same feedback filter as the non-linear precoding operation, and performing a feedforward filtering operation with a same feedforward filter as the non-linear precoding operation, and

wherein, performing the linear precoding operation further comprises performing a phases tilting operation to the respective demodulation reference signals in the feedback filtering operation.
The method according to Claim 15, further comprises:

multiplex the precoded respective demodulation reference signals with the precoded plurality of streams of data signals; and

transmitting the respective demodulation reference signals multiplexed with the plurality of streams of data signals to a plurality of receiving apparatus.
The method according to Claim 15, wherein the non-linear precoding operation is performed according to a Tomlinson-Harashima (THP) scheme.
The method according to Claim 15, further comprises:

determining tilting phases for respective demodulation reference signals through an iterative algorithm, to minimize power increases of respective demodulation reference signals.
The method according to Claim 15, further comprises:

sending an indication that a phase tilting operation is applied to the respective demodulation reference signals.
The method according to Claim 15, further comprises:

sending an indication of signs of receive weights of the respective demodulation reference signals.
The method according to Claim 15, further comprises:

determine parameters of the feedback filter and the feedforward filter, according to channel status information from the receiving apparatus.
A method for receiving a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, the method comprising:

receiving a demodulation reference signal from a transmitting apparatus;

deriving a channel estimation based on the received demodulation reference signal and a corresponding original demodulation reference signal; and

recovering receive weights for data signals multiplexed with the received demodulation reference signal from the channel estimation, based on a tilted phase of the received demodulation reference signal.
The method according to Claim 22, wherein the recovering comprises abstracting amplitudes of the channel estimation as the receive weights.
The method according to Claim 22, further comprises:

demodulating the data signal according to the receive weights.
The method according to Claim 22, further comprises:

receiving an indication of a phase tilting operation applied to the demodulation reference signal at the transmitting apparatus.
The method according to Claim 22, further comprises:

receiving an indication of signs of the receive weights; and

recovering the receive weight based on the signs of the receive weights.
The method according to Claim 22, further comprises:

estimating the tilted phase applied the demodulation reference signal; and

recovering the receive weight is based on the estimated tilted phase.
An apparatus for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, the apparatus comprising:

at least a processor, and

a memory, communicatively associated with the processor with computer coded instructions stored therein, said instructions when executed by the processor causing the apparatus to:

perform a non-linear precoding operation on a plurality of streams of data signals addressed to a plurality of receiving apparatus; and

perform a linear precoding operation on respective demodulation reference signals for the plurality of streams of data signals,

wherein , performing the linear precoding operation comprises performing a feedback filtering operation with a same feedback filter as the non-linear precoding operation, and performing a feedforward filtering operation with a same feedforward filter as the non-linear precoding operation, and

wherein, performing the linear precoding operation further comprises performing phases tilting operation to the respective demodulation reference signals in the feedback filtering operation.

#
An apparatus for receiving a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, the apparatus comprising:

at least a processor, and

a memory, communicatively associated with the processor with computer coded instructions stored therein, said instructions when executed by the processor causing the apparatus to:

receive a demodulation reference signal from a transmitting apparatus;

derive a channel estimation based on the received demodulation reference signal and a corresponding original demodulation reference signal; and

recover receive weights for data signals multiplexed with the received demodulation reference signal from the channel estimation, based on a tilted phase of the received demodulation reference signal.

#
An apparatus for transmitting a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, comprising means operative to perform the method according to any of claims 15 to 21.
An apparatus for receiving a demodulation reference signal in a multi-user multiple-input multiple-output (MIMO) system, comprising means operative to perform the method according to any of claims 22 to 27.
A computer readable storage medium, on which stored computer coded instructions which, when executed on at least one processor, cause at least one processor to carry out the method according to any of claims 15 to 21.
A computer readable storage medium, on which stored computer coded instructions which, when executed on at least one processor, cause at least one processor to carry out the method according to any of claims 22 to 27.