Method for compressing wireless channel state information feedback

Abstract

This disclosure relates to channel state information reference signal configuration, precoding and transmission, and channel state information feedback configuration, compression and transmission in multiple-input multiple-output (MIMO) wireless communication systems. Various embodiments provide channel state information reference signal and feedback schemes that help reduce the amount of information that needs to be transmitted and reduce the communication resource overhead of channel state information configuration and feedback.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to channel reference signal transmission and channel state information feedback in a wireless communication system. BACKGROUND

[0002] Reliability in a wireless communication system employing multiple-input multiple-output (MIMO) technology mainly relies on signal processing, such as precoding and beamforming before signal transmission over a wireless channel. Parameters for such signal processing can be derived from measurements of reference signals transmitted over a wireless channel and fed back between wireless devices. Efficient transmission of reference signals and feedback information helps to reduce overall communication resource consumption of a wireless communication system. SUMMARY

[0003] The present disclosure relates to channel state information reference signal configuration, precoding and transmission, and channel state information feedback configuration, compression and transmission in a MIMO wireless communication system. Various embodiments provide channel state information reference signal and feedback schemes that help to reduce the amount of information that needs to be transmitted and reduce communication resource overhead for channel state information configuration and feedback.

[0004] In one implementation, a method for processing wireless channel state information (CSI) reporting is disclosed. The method can be performed by a wireless terminal device. The method can include receiving a CSI reporting time configuration from a wireless access network node; receiving a series of multiple CSI reference signals (CSI-RSs) from the wireless access network node within a time window of a time length determined according to the CSI reporting time configuration; generating CSI feedback based on the multiple CSI-RSs; and reporting the CSI feedback by the wireless terminal device to the wireless access network node.

[0005] In another implementation, a method for reporting wireless CSI is disclosed. The method can be performed by a wireless terminal device. The method can include receiving CSI reference signals (CSI-RSs) transmitted from a wireless access network node; generating one or more indicators associated with one or more non-zero elements in CSI feedback information in a vector space determined by a first set of basis vectors, a second set of basis vectors, a third set of basis vectors based on measuring the CSI-RSs; and transmitting information about the one or more non-zero elements in the CSI feedback information and the one or more indicators to the wireless access network node.

[0006] In another embodiment, a method for processing wireless CSI reporting is disclosed. The method can be performed by a radio access network node. The method can include transmitting a CSI reporting time configuration from a wireless terminal device; transmitting a series of multiple CSI reference signals (CSI-RSs) to the wireless terminal device within a time window of a time length determined according to the CSI reporting time configuration; and receiving a CSI feedback generated by the wireless terminal device based on the multiple CSI-RSs.

[0007] In another embodiment, a method for reporting wireless CSI is disclosed. The method can be performed by a radio access network node. The method can include transmitting a CSI reference signal (CSI-RS) to a wireless terminal device; and receiving a CSI report by the wireless terminal device based on a measurement of the CSI-RS. The CSI report can include information about one or more indicators associated with one or more non-zero elements in CSI feedback information generated by the wireless terminal device in a vector space determined by a first set of basis vectors, a second set of basis vectors, and a third set of basis vectors; and information about the one or more non-zero elements.

[0008] In some other embodiments, a communication device is disclosed. The communication device can mainly include one or more processors and one or more memories, wherein the one or more processors are configured to read computer codes from the one or more memories to implement any of the above methods.

[0009] In some other embodiments, a computer program product is disclosed. The computer program product can include a non-transitory computer-readable program medium having computer codes stored thereon, which, when executed by one or more processors, cause the one or more processors to implement any of the above methods.

[0010] The above embodiments and other aspects and alternatives thereof are explained in more detail in the drawings, the specification, and the claims. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 A radio access network including a radio access network node and a user equipment is shown.

[0012] Figure 2 A principle of precoding a channel state information reference signal by a radio access network node to improve channel state information feedback by a user equipment is shown. Figure 1 Figure 1 A principle of precoding a channel state information reference signal by a radio access network node to improve channel state information feedback by a user equipment is shown.

[0013] Figure 3 ​An exemplary mapping between elements of the channel state information reference sequence and multiple radio resource blocks is shown.

[0014] Figure 4 Another exemplary mapping between the elements of the channel state information reference sequence and multiple radio resource blocks is shown.

[0015] Figure 5 Another exemplary mapping between the elements of the channel state information reference sequence and multiple radio resource blocks is shown.

[0016] Figure 6 The resource block configuration of the channel state information reference signal is shown.

[0017] Figure 7 The compression principle of the weighting coefficients in channel state information is shown.

[0018] Figure 8 An exemplary channel state information triggering and feedback scheme is shown.

[0019] Figure 9 Another exemplary channel state information triggering and feedback scheme is shown.

[0020] Figure 10 The bit sequence configuration for the weighting coefficients used in the feedback channel state information is shown. Detailed Implementation

[0021] like Figure 1 As shown, the radio access network 100 may include a wireless user equipment (UE) or wireless terminal equipment 104 and a radio access network node or base station 102. The radio access network node 102 may be backhauled to the core network (not shown). The UE 104 may communicate with the radio access network node 102 via air channel 110. The radio access network node 102 and the core network may be configured to transmit and route voice, data, and other information between the UE 104 and other UEs, data networks, or other radio core networks terminating at the edge of the core network. The UE 104 may include, but is not limited to, mobile phones, tablets, laptops, smart home electronic devices or appliances, remote sensor devices, etc. The air channel 110 may include a downlink channel 120 for transmitting data and control signals from the radio access network node 102 to the UE 104, and an uplink channel 130 for transmitting data and control signals from the UE 104 to the radio access network node 102. The air channel 110 may be implemented by allocating frequency radio resources and time radio resources on one or more transmit antennas 140 and receive antennas 150. Radio resources can be organized into radio resource blocks (RBs), each containing multiple radio resource elements (REs).

[0022] Figure 1 The multi-transmit and receive antenna systems described in the background can be referred to as MIMO wireless access systems. MIMO systems can be used to achieve data transmission diversity for increased transmission accuracy, or data multiplexing for increased transmission throughput, or a combination thereof. MIMO systems can also be used to achieve diversity-based multi-stream or multi-layer RF beamforming to enhance wireless signals at specific wireless UEs, e.g., in a multi-UE environment. Data can be precoded and mapped to wireless resources and multiple transmit antennas for transmission via an over-the-air channel characterized by, e.g., a channel transmission matrix. Precoding can be achieved, e.g., by a data transformation using a linear combination of a set of basis vectors selected from one or more predetermined codebooks.

[0023] To improve the efficiency and reliability of signal transmission, a set of precoding basis vectors for a data channel of the transmitting end can be selected to match the over-the-air channel transformation characteristics. Thereby, the transmitting end can need to acquire channel state information (CSI) in order to determine the selection of precoding basis vectors and their linear combination coefficients. The CSI can be acquired as feedback from a receiving end of the wireless channel. In particular, the transmitting end can first transmit a reference signal (which is referred to as a CSI reference signal or CSI-RS) to the receiving end via the over-the-air channel. The receiving end can receive and measure the transmitted reference signal and generate a set of feedback parameters and include the set of feedback parameters in a CSI feedback report (or, for simplicity, CSI report) transmitted back to the transmitting end. The set of feedback parameters can include, e.g., a recommended selection of precoding basis codebook vectors and their coefficients. The basis codebook vector coefficients can include, e.g., the amplitude and phase of the coefficients. These parameters in the CSI report can be further quantized before being fed back in order to reduce the amount of information that needs to be transmitted (e.g., only the quantization level indices in a predetermined quantization table can need to be transmitted).

[0024] The transmitting end can be a wireless access network node 102 or a UE 104. Figure 1 The receiving end can be a UE 104 or a wireless access network node 102. For the sake of clarity, the following description refers to the wireless access network node 102 as the transmitting end and to the UE 104 as the receiving end. Thereby, the terms "transmitting end" and "wireless access network node" / "base station" can be used interchangeably. Likewise, the terms "receiving end" and "UE" / "wireless end device" / "terminal" can be used interchangeably. However, the basic principles described in the following apply similarly to scenarios where the UE 104 is the transmitting end and the wireless access network 102 is the receiving end.

[0025] Thus, the design of CSI-RS, the configuration of CSI-RS, the transmission of CSI-RS, the measurement accuracy of the transmitted CSI-RS, and the feedback generation and transmission of CSI are all key components of a high performance MIMO system. For example, the transmission of CSI-RS signals by the transmitting end and the feedback of CSI reports by the receiving end consume wireless resources. Thus, improving the design of CSI-RS and CSI reports and increasing the transmission efficiency of CSI-RS signals and CSI feedback reports help to reduce the downlink and uplink wireless resource consumption and lower the overhead of a wireless communication system.

[0026] As described in detail below in various embodiments, with respect to CSI-RS design and transmission and CSI feedback procedures, the efficiency of a MIMO system can be enhanced in several ways. In one exemplary embodiment, by exploiting frequency domain correlation, a set of selected spatial domain basis vectors (from a predetermined spatial domain codebook) and their coefficients at various frequencies or subcarriers in, for example, orthogonal frequency division multiplexing (OFDM) can be jointly transformed into a set of frequency domain basis vectors (from a predetermined frequency domain codebook) and corresponding frequency domain coefficients to reduce the number of basis vectors and coefficients that can need to be transmitted in a CSI feedback report.

[0027] In another example embodiment, the transmitting end (e.g., a radio access network node) can obtain in some way a certain candidate codebook basis vector for downlink transmission, for the configuration CSI of the receiving end. For example, the radio access node can obtain a set of candidate codebook basis vectors by additionally measuring the uplink channel in some way based on, for example, channel reciprocity. Thereby, the transmitting end can use this candidate codebook basis vector information to improve the performance of the CSI feedback from the receiving end (e.g., a wireless terminal device), or to reduce the CSI feedback overhead and reference signal overhead. More specifically, by giving the candidate precoding parameters containing the candidate codebook basis vector selected by the transmitting end, the CSI feedback mechanism can be enhanced to provide improved measurement of the CSI-RS by the receiving end and more accurate determination of the CSI parameters. Such identification information of the candidate basis vector can be carried in the CSI-RS signal when transmitting the CSI-RS signal, or via some other channel transmission. In this way, the receiving end can obtain the candidate basis vector selection information from the received CSI-RS. Such information can facilitate improved determination of the CSI feedback parameters by the receiving end. For example, by determining the selection information of the candidate basis vector via the information carried in the CSI-RS signal or transmitted by the transmitting end via some other channel, the receiving end can focus on deriving and determining more accurate linear combination coefficients of these candidate basis vectors. Under the principle or assumption of reciprocity of the uplink and downlink channel characteristics, the determination of the candidate basis vector by the transmitting end can be based on other reference signals received by the transmitting end from the receiving end, for example, sounding reference signals (SRS) transmitted from the receiving end to the transmitting end.

[0028] In another embodiment, the UE can be mobile and moving at an estimable speed (e.g., the UE can be in a moving car or train). In this case, the receiving end can need to frequently provide CSI feedback to the transmitting end to keep track of the changes in the channel characteristics and to adjust the precoding parameters in time accordingly, resulting in a large amount of wireless resource consumption over time. In some embodiments, instead of sending a CSI feedback report every time a CSI-RS is received, the receiving end can compress multiple CSI feedback information in the time domain. In this way, the temporal correlation between the CSI feedback information of multiple CSI-RSs can be exploited to enable transmission of a reduced number of CSI feedback reports for multiple CSI-RSs, thereby reducing the cumulative wireless resource consumption and overhead in transmitting the CSI reports over time. After receiving the compressed CSI feedback, the transmitting end can use, for example, an extrapolation method to anticipate or predict the precoding parameters at future times (due to the motion of the receiving end).

[0029] In yet another embodiment, the information included in the CSI feedback can be further compressed based on the sparsity of the information. For example, it can be only necessary to feedback the non-zero elements. In particular, for the precoding coefficients in the vector space formed by the multiple basis vector sets selected from the corresponding codebook, only the non-zero elements and their position information in the vector space can be transmitted. For example, such position information can be encoded and represented by various indices of various dimensions.

[0030] Precoding based on spatial codebook basis vectors and frequency domain codebook basis vectors

[0031] For high-precision CSI feedback embodiments, the terminal feeds back precoding information by measuring the CSI-RS. The precoding information for each transmission stream or layer can be organized in the form of a precoding matrix. The number of columns of the precoding matrix fed back by the terminal can be referred to as the channel rank or rank indicator, RI. The precoding vector for each transmission layer can be represented as a linear combination of a set of spatial codebook basis vectors from a predetermined spatial codebook. The term "codebook basis vector" can alternatively be referred to as "basis vector". This set of spatial codebook basis vectors can be referred to as a first basis vector set. The linear combination of the first basis vector set can be represented by a set of weighting coefficients each including an amplitude and a phase. The terminal can determine the first basis vector set and calculate the set of weighting coefficients based on the measured CSI-RS and the first basis vector set, and further quantize the amplitude and phase information of the weighting coefficients. The weighting coefficients of the first basis vector set can be referred to as a first coefficient set.

[0032] To improve the performance of the CSI feedback, it is generally necessary to report the amplitude and phase information of the first coefficient set for each frequency subband. The frequency subband represents the frequency domain granularity in the precoding process. The frequency subband for CSI feedback can be configured by the wireless network node via resource allocation signaling. For all RBs included in the CSI feedback resource bandwidth, a plurality of (M) contiguous RBs form a subband. The CSI feedback bandwidth can include N subbands of size M. For example, in a 5G wireless system, such a subband can be referred to as a PMI (Precoding Matrix Indicator) subband, which can be equal to a CQI (Channel Quality Information) subband, or 1 / R of the CQI subband, where R is a positive integer.

[0033] For the above CSI feedback method by the terminal, for the nthsubband (n = 1,..., N), the precoding vector of a certain transmission stream or layer can be represented as:

[0034]

[0035] wherein, includes a first set of basis vectors (spatial domain basis vectors), and is a vector formed by the first set of coefficients. Generally speaking, the information in may be the same. Specifically, the first set of basis vectors included in may include L spatial codebook basis vectors. Accordingly, the first set of coefficient vectors may be L-dimensional vectors. The L spatial codebook basis vectors are selected from a predetermined spatial codebook. In other words, the number of columns of may be L. For example, may be a block diagonal matrix, and the vectors included in are the first set of basis vectors. For a precoding vector , the first set of coefficient vectors for all subbands can be represented by the following first coefficient matrix

[0036]

[0037] Each column of the first coefficient matrix includes one of the vectors. If the amplitude and phase information of the elements in matrix are directly quantized, it can result in a large feedback overhead. By exploiting the frequency domain correlation among the weighting coefficients, compressing matrix along the subband dimension can help reduce the feedback overhead of the weighting coefficients without too much sacrifice in performance, as shown below.

[0038] Specifically, the matrix can be transformed by conjugate transpose to obtain . The column vectors of (l = 1, …, L) can be represented by Then, the terminal can select another set of basis vectors from a second codebook (a frequency domain codebook), referred to herein as a second set of basis vectors (a frequency domain basis vector set), and express as a linear combination of the basis vectors of the second set of basis vectors and a second set of coefficients

[0039]

[0040] where is an N-dimensional vector. The second set of basis vectors includes a total of K vectors, and the second set of coefficients is a K-dimensional vector. Each element includes amplitude and phase. The terminal calculates and identifies... and And report back to the wireless access network node in the CSI report. , and . For all l, they can be the same, or they can be different. When When all l are the same, it can be simply represented by D.

[0041] Due to correlation in the frequency domain, The dimension K will be significantly smaller than N. Therefore, for The amount of feedback required will be less than the above. Information, thereby improving feedback overhead.

[0042] In some implementations, the first set of basis vectors above may include the set of column vectors in the identity matrix, or orthogonal DFT vectors or the Kronecher product of DFT vectors. The second set of basis vectors may include DFT vectors.

[0043] Precoding of CSI-RS using a set of candidate first basis vectors

[0044] In some implementations, the transmitting end (e.g., a wireless access network node) can estimate a candidate set of first basis vectors(s) before transmitting the CSI-RS. Then, before transmitting the CSI-RS to the receiving end for CSI measurement and feedback, the transmitting side processes (precodes or maps) the CSI-RS as described below. This allows the transmitting end to provide this candidate codebook basis vector information to improve the performance of CSI feedback from the receiving end (e.g., a wireless terminal device) and / or reduce CSI feedback overhead. More specifically, the CSI feedback mechanism can be enhanced to provide improved measurement of the CSI-RS and more accurate determination of CSI feedback (selection of basis vectors, determination of the various matrices and weighting coefficients mentioned above). For example, a more suitable precoding matrix (set of basis vectors) can be identified (e.g., selected from candidate basis vectors) by the receiving end. Furthermore, a more accurate weighting coefficient vector can be computed by the receiving end. The determination of candidate basis vectors by the transmitting end based on a predetermined codebook can be based on various means. For example, based on the principle and assumption of reciprocity of uplink and downlink channel characteristics, it can be based on the analysis of one or more other reference signals (e.g., sounding reference signals (SRS) transmitted from the receiver to the transmitter) received by the transmitter from the receiver.

[0045] In some embodiments, the CSI-RS can be processed or mapped by the transmitting end using a candidate first basis vector set (spatial domain basis vectors) before transmission. In particular, if the first basis vector set includes a set of column vectors in an identity matrix, the CSI-RS can be precoded by the candidate first basis vector set. In other words, the wireless access network node can obtain the candidate first basis vector information in some way and load the candidate first basis vector information into the CSI-RS during transmission. For example, the CSI-RS of L transmitting ports can be mapped on the antennas by precoding and transmitted to the terminal. The terminal can feed back the transmitting port selection information through the fine first basis vector set selection (spatial or port basis vector selection).

[0046] In some other embodiments, the wireless access network node can obtain a candidate set of first basis vectors (spatial domain basis vectors) and a candidate set of second basis vectors (frequency domain basis vectors) in some way. Further, the wireless access network node can transmit to the terminal a precoded or virtualized CSI-RS precoded / mapped with the information of the two candidate basis vector sets. The terminal receives and measures the CSI-RS and can determine the candidate basis vector sets recommended by the wireless access network node and select a fine or optimal first basis vector set and a second basis vector set from the candidate basis vector sets. The terminal can also determine the second set of weighting coefficients (see above) and feed back the best second weighting coefficient information to the wireless access network node. Under the principle and assumption of reciprocity of the uplink and downlink channel characteristics, the determination of the candidate sets of first and second basis vectors can be based on the measurements of one or more other reference signals (e.g., SRS) received by the wireless access network node from the terminal by the wireless access network node. In these examples, the first basis vector set and the second basis vector set can include a set of column vectors in an identity matrix.

[0047] The basic principle of the above embodiments is illustrated in Figure 2 . In particular, Figure 2 The CSI feedback mechanism 200 illustrated in

[0048] Figure 2 The precoding / mapping step 204 of mapA sequence of CSI-RS for L transmission ports is mapped to a set of groups of frequency domain radio resources. The set of mapped frequency radio resources within the resources allocated to the CSI-RS can be referred to as first radio resources (or radio resource allocation(s)). The first radio resources can be configured as groups of first radio resources. Each group of first radio resources can include a plurality of radio resource blocks (RBs). The distribution of the groups of first radio resources in the radio resource allocation can be determined by mapping W map . Because the mapping is determined as W map , which can include CSI-RS precoding / mapping information (e.g., a set of candidate basis vectors information), the candidate basis vector information will be precoded / mapped in the manner that the groups of first radio resources are distributed in the radio resource allocation(s) for the CSI-RS. The following embodiments describe various exemplary ways for mapping between the CSI-RS sequence and the radio frequency resources.

[0049] In a first exemplary mapping embodiment for a particular CSI-RS port / in the CSI-RS for L transmission ports, the elements of the CSI-RS sequence may be mapped to N radio resource elements (REs) as follows:

[0050]

[0051] In other words, the port / CSI-RS sequence is mapped to N REs by , where the symbol represents an element-wise multiplication. The vector is an N-dimensional column vector representing a candidate basis vector (e.g., a second basis vector or a frequency domain basis vector). M candidate vectors can be provided (m = 1,..., M). The signal transmitted by the transmitting end on these N CSI-RS REs can be represented as:

[0052]

[0053] where represents a precoding vector that maps the CSI-RS on port / to a plurality of transmit antennas.

[0054] Ignoring noise and interference, the signal received and measured by the terminal can be represented as:

[0055]

[0056] where represents the channel matrix on the nth CSI-RS RE. Given and , The terminal can solve it. Further, the feedback precoding matrix can be determined by

[0057] .

[0058] In other words, the optimal precoding is a linear combination of . Thus, the strongest eigenvector of will be the optimal solution for the weighted coefficients , where is a matrix containing the column vectors .

[0059] Thus, in the above embodiments, the vector is used as the mapping matrix W map for each of the M groups of first wireless resources, linearly transforming or mapping the sequence of CSI-RS over L transmission ports to the wireless resources. The index m = (1,..., M) denotes the M groups of first wireless resources.

[0060] Figure 3 This resource matching within a group of first wireless resources is illustrated. Figure 3 A first wireless resource 300 for transmitting a CSI-RS comprising a plurality of RBs is illustrated, the plurality of RBs being illustrated as variously shaded or unshaded horizontally elongated boxes. The plurality of RBs comprises, by way of example, M = 4 groups of RBs, wherein each group of RBs is represented by a type of shading. For example, RBs 310, 320 and 330 belong to one group, RBs 312, 322 and 332 belong to another group, RBs 314, 324 and 334 belong to a third group, and RBs 316, 326 and 336 belong to a fourth group. Each group can comprise N RBs, as indicated by 302, 304 and 306. The mapping between elements of the CSI-RS sequence and the resource blocks is illustrated by 340, 350, 360 and 370. For example, 340 illustrates that the 0th element of the CSI-RS sequence is mapped via the N components of to the N RBs indicated by 310, 320,..., 330; 350 illustrates that the 1st element of the CSI-RS sequence is mapped via the N components of to the N RBs indicated by 312, 322,..., 332; 360 illustrates that the 2nd element of the CSI-RS sequence is mapped via the N components of to the N RBs indicated by 314, 324,..., 334; and 370 illustrates that the 3rd element of the CSI-RS sequence is mapped via the N components of to the N RBs indicated by 316, 326,..., 336.

[0061] ​Each port corresponds to one RE in an RB. As shown in Figure 3 Resource element 380 is spatial port 0, and resource element 390 is for spatial port 1, and so on. In some embodiments, not all of the RBs in the mapped RBs are mapped to CSI-RS sequence elements.

[0062] As shown above and Figure 3 , the first example mapping embodiment implements mapping of a CSI-RS sequence for a particular spatial port to RBs. In the example shown, each of the boxes shown as representing an RB can alternatively represent any other number of RBs. Figure 3

[0063] In a second example mapping embodiment for a particular CSI-RS port l in a CSI-RS for L transmission ports, the CSI-RS sequence for the L spatial transmission ports (e.g., after spatial virtualization) can be transmitted on N RBs (rather than RBs in the above first example mapping embodiment). In contrast to the above first example mapping embodiment, the CSI-RS sequence for the L spatial ports is not mapped on the N RBs in a straightforward manner, but rather is mapped to the N RBs after performing a frequency domain precoding process by a frequency domain basis vector (second basis vector).

[0064] In particular, the CSI-RS sequence of length M for the lth spatial port can be mapped on the N RBs in the following manner.

[0065]

[0066] where may be a DFT vector (e.g., second basis vector), and the wireless access network node can use the SRS to perform PDP detection to obtain (e.g., included in the FD basis information).

[0067] For each spatial port l, the CSI-RS signal transmitted on the N RBs and P antennas is:

[0068]

[0069] Ignoring interference and noise, the signal received by the terminal is:

[0070]

[0071] From the above two equations, it can be seen that the estimated equivalent channel matrix after signal processing of the transmitted frequency domain virtualized CSI-RS is:​

[0072] .

[0073] which is the channel matrix on the new domain after the two-dimensional transformation of the channel matrix H.

[0074] For such an equivalent channel matrix, the optimal weighting coefficient information can be derived by the terminal where C denotes the coefficient matrix.

[0075]

[0076] Further, the optimal coefficient matrix is included on the new domain after the two-dimensional transformation that best matches the channel characteristics. Thus, in some embodiments, the can be SVD-decomposed to obtain the strongest eigenvector as the optimal vec( ), which can be quantized for feedback to the radio access network node.

[0077] Figure 4 The above second exemplary mapping embodiment is further illustrated in

[0078] In a third exemplary mapping embodiment for a specific CSI-RS port l in the CSI-RS for L transmission ports, the CSI-RS sequence (e.g., after spatial virtualization) for the L spatial transmission ports can be transmitted on M RBs (instead of N*M RBs in the above first exemplary mapping embodiment or N RBs in the above second exemplary mapping embodiment). For the CSI-RS port 1, the CSI-RS sequence of length N can be mapped to the CSI-RS resource elements using the following formula:

[0079] .

[0080] The CSI-RS signal transmitted on the actual P antennas is

[0081]

[0082] Thus, on M CSI-RS RBs (M < N), one RE is used to transmit the above signal on each RB, and the CSI-RS signal mapped on the M RBs is

[0083]

[0084] In this way, the frequency domain signal is transformed by a linear transformation to the delay domain, and only M RBs are used to estimate the CSI on N RBs, thereby further reducing the CSI overhead.

[0085] In fact, the final precoding matrix for each RB can be written as a linear combination of

[0086]

[0087] Hence, if only power detection is used, the above CSI-RS configuration can provide detection of the strength of different spatial and frequency domain basis vectors with minimal resource consumption.

[0088] Figure 5 The above third exemplary mapping implementation is further illustrated in FIG. 5. In particular, Figure 5 The mapping 506 is shown to map the N-element CSI-RS sequence to M resource blocks 502 out of N resource blocks 504. Each spatial port corresponds to resource elements in the M resource blocks 502, as shown by 508 and 510.

[0089] In the above three exemplary mapping implementations, the CSI-RS sequence (or elements in the sequence) is mapped to the first resource in the frequency domain by a linear transformation (e.g., a matrix multiplication transformation or an element-wise multiplication involving a mapping matrix W map The resources mapped to the CSI-RS corresponding to W map may be referred to as a group of first frequency domain resources. Further, for various implementations, the number of rows or columns of W map may be as follows:

[0090] the number of RBs (e.g., N above) included in the CSI-RS bandwidth (or associated / corresponding CSI bandwidth),

[0091] the number of groups of frequency domain resources (e.g., M)

[0092] one of the number of rows or columns of W map may be equal to 1.

[0093] In particular, for a certain CSI-RS port / , the way the corresponding CSI-RS sequence is mapped to the video resources can be based on at least one of the following:

[0094] Exemplary mapping 1: where the CSI-RS sequence of length M is mapped to ​W is an N x M matrix. map is an N x M matrix.

[0095] Exemplary mapping 2: where a CSI-RS sequence of length M is mapped to N RBs, and W map is an N x M matrix.

[0096] Exemplary mapping 3: where a CSI-RS sequence of length N is mapped to M RBs, and W map is an N x M matrix.

[0097] To achieve better measurement of the second basis vector information across the entire bandwidth, the CSI-RS sequence pattern in each of the above exemplary mappings, in particular the distribution of the group of first frequency domain resources across the entire CSI-RS allocation bandwidth, can follow at least one of the following patterns:

[0098] Pattern 1 : Each of the one or more groups of first frequency domain resources occupies consecutive RBs in the entire CSI-RS bandwidth.

[0099] Pattern 2: The one or more groups of first frequency domain resources are distributed in the form of a comb across the entire CSI-RS bandwidth.

[0100] Pattern 3: Each of the one or more groups of first frequency domain resources occupies n2 RBs in every n1 RBs in the CSI-RS resource and allocation(s). Specifically, assuming the entire CSI-RS bandwidth includes C RBs, denoted as RB{1,...,C}, and a certain group of first frequency domain resources occupies where n0 and n1 are positive integers greater than or equal to 1, i is an integer between 0 and n2 inclusive, j is an integer greater than or equal to 0, and n2 is a positive integer. In other words, a certain group of first frequency domain resources occupies consecutive n2 RBs in every n1 RBs. In some embodiments, as shown in Figure 6 M groups of first frequency domain resources (e.g., different line types of resource blocks in Figure 6 represent different groups) can be configured, and the mapping matrix on each group of frequency domain resources can be an N-dimensional column vector. Then, the Mth group of frequency domain resources occupies (e.g., in Figure 6In the table, group m=1 occupies 602 and 604, and group m=M occupies 606 and 608), where i=1, ..., n2, j=0, ..., N-1. For the m-th frequency domain resource group, in the total occupied... In each of the RBs, the k-th element s of the CSI-RS sequence is obtained by... Indicated Mapped to Among the RBs So that it can be transmitted to the terminal, where k = 1, ..., n2. In other words, the k-th element s of the CSI-RS sequence is mapped to the entire bandwidth. The values ​​of n1 and n2 can be set to fixed numbers. Alternatively, n1 and n2 can be configured via signaling, such as Radio Resource Control (RRC) signaling.

[0101] After the CSI-RS is transmitted to the terminal by the wireless access network node, the terminal measures the CSI-RS and measures the CSI according to the corresponding CSI settings, wherein the CSI includes information for selecting the first frequency domain resource group as discussed above and the weighting coefficient amplitude and / or phase information corresponding to the selected first frequency domain resource group.

[0102] The parameters related to the first frequency domain resource groups and some parameters set for CSI can be correlated with each other. For example, the number of first frequency domain resource groups associated with CSI reporting can be correlated with the size of the subband (or CSI subband) used for CSI reporting, where the subband can be a CQI subband or a PMI subband. Specifically, each CQI subband can correspond to a subband CQI value, and each PMI subband can correspond to a precoding matrix, and the size of the CQI subband can be an integer multiple of the size of the PMI subband. For example, the number of RBs included in each subband can be equal to the number of first frequency domain resource groups multiplied by the number of RBs in each first frequency domain resource group within the subband. Furthermore, within each subband, the RBs occupied by the first frequency domain resource groups can be contiguous. Alternatively, the RBs partially occupied by the first frequency domain resource groups within the subband can be distributed in a comb-like manner within the subband. For a comb-like distribution within a certain subband, the m-th first frequency domain resource group out of M first frequency domain resource groups occupies a portion of that subband. There are RBs, where i = 0, ..., R-1, and R is a positive integer greater than or equal to 1, representing the number of RBs occupied by a portion of each first frequency domain resource group within each subband. In special cases, the subband size can be equal to the number of frequency domain resource groups, as shown in some of the exemplary embodiments above.

[0103] In some embodiments, the size of the first frequency domain resource group selected by the terminal and the size of the subband included in the CSI report can be related. Specifically, the size of the subband can be greater than or equal to the number of the selected first frequency domain resource group. For example, if the number of subbands included in the CSI feedback bandwidth or the CSI-RS bandwidth is S, the size of the subband can be required to be greater than or equal to where p is a positive real number less than 1. Further, the size of the subband can be required to be less than or equal to K, where K is the number of RBs contained in the subband, and B is the number of RBs contained in the CSI-RS or CSI bandwidth; or where M is the number of the configured first frequency domain resource groups, and R is the number of RBs occupied by each of the first frequency domain resource groups in each subband.

[0104] In some embodiments, the mapping matrix W map corresponding to a certain first frequency domain resource group has the number of rows or columns equal to the number of the first frequency domain resource groups selected by the terminal for the CSI report. For example, the number of rows or columns of the mapping matrix can be p times the number of subbands, rounded to the nearest integer, where p is a positive real number less than 1.

[0105] The above solution can be extended to the CSI feedback using multiple CSI-RSs. Specifically, in each of the CSI-RSs, a CSI-RS sequence is mapped to the frequency domain resources by a linear transformation (e.g., using a mapping matrix W map ). Each of the CSI-RSs corresponds to a W map . Further, in this type of CSI-RS, the density of the CSI-RS RBs in the whole bandwidth can be low, e.g., less than 1 / 2.

[0106] In some embodiments, the terminal feeds back a CSI report of one or more CSIs corresponding to the one or more CSI-RSs. The terminal can calculate the weighting coefficient information according to the selected CSI and / or the port information in each selected CSI, and feed back the weighting coefficient information back to the radio access network node.

[0107] In some embodiments, in the CSI report, the related CSI-RSs can satisfy certain constraints. For example, the number of RBs included in each subband can be required to be equal to the number of CSI-RSs multiplied by the number of RBs occupied by each CSI-RS in each subband, where the latter can be determined by the CSI-RS resource density multiplied by the number of RBs in the subband, where the density of all CSI-RSs is the same. It can also be required that where K is the number of CSI-RSs and d is the CSI-RS resource density. In a particular example, the subband size can be equal to the number of CSI-RSs, in which case the CSI-RS resource density is 1 / subband size. In another example, if the reported number of CSI-RSs is p times the number of subbands, then the subband size can be required to be greater than or equal to p times the number of subbands, where p is a positive real number less than 1. Further, the subband size can be required to be greater than or equal to the number of CSI-RSs, where the number of CSI-RSs is p times the number of subbands, where p is a positive real number less than 1. where "size" is the number of RBs included in the subband and B is the number of RBs included in the CSI-RS or CSI bandwidth. In one example, the mapping matrix W map has a number of rows or columns equal to the number of CRIs selected for reporting, which can be p times the number of subbands, where p is a positive real number less than 1.

[0108] To summarize, the above mapping implementations are merely examples. In general, a CSI-RS can be configured and provided with a CSI-RS resource allocation for transmitting the CSI-RS. Each CSI-RS resource allocation can include a CSI-RS resource block. Each CSI-RS resource block can include a CSI-RS resource element. For mapping of a CSI-RS sequence, one or more groups of CSI-RS resources can be selected from the CSI-RS allocation. Each of the one or more groups of CSI-RS resources can include one of: a group of CSI-RS resource allocations, a group of CSI-RS resource blocks, a group of CSI-RS resource elements. The CSI-RS sequence can be mapped to the one or more groups of CSI-RS resources.

[0109] In some implementations, the CSI-RS resources within one of the one or more groups of CSI-RS resources and mapped to the CSI-RS sequence corresponding to one CSI-RS port are associated with a same channel characteristic. In some other implementations, the one or more groups of CSI-RS resources are associated with different channel characteristics.

[0110] In some implementations, via the above mapping, a group of the one or more groups of CSI-RS resources can be distributed in one or more CSI-RS resource allocations as a contiguous set of resource blocks, a comb structure of resource blocks, or a plurality of sets of a first predetermined number of contiguous resource blocks separated by a second predetermined number of resource blocks.

[0111] In some embodiments, the above one or more CSI-RS resource allocations can be associated with one or more CSI reporting settings that determine the number of CSI subbands (as described above, which can be CQI subbands or PMI subbands), and wherein the number of resource blocks in one CSI subband is determined by the number of groups of CSI-RS resources multiplied by the number of resource blocks of a group of CSI-RS resources within one CSI subband. Further, via the above mapping, the resource blocks of a group of CSI-RS resources within one CSI subband are distributed within one CSI subband as a contiguous set of resource blocks, a comb structure, or a plurality of sets of a first predetermined number of contiguous resource blocks separated by a second predetermined number of resource blocks. The values of the first and second predetermined numbers can be set to fixed numbers. Alternatively, they can be configured via signaling, such as radio resource control (RRC) signaling. The above distribution of resource blocks of a group of CSI-RS resources within a CSI subband or across the entire CSI-RS resource allocation can be combined in any manner. The two sets of the first and second predetermined numbers can be the same or can be independent.

[0112] The term resource block (RB) is used herein to refer generically to a resource block in a resource allocation, such as a resource block allocated for any one of the CSI / CSI-RS purposes. The term "CSI-RS resource block" is used herein to refer to a resource block within a resource block allocated for a CSI-RS allocation that is mapped to a CSI-RS sequence. Likewise, the term resource element (RE) is used herein to refer generally to a resource element in any resource block. The term "CSI-RS resource element" is used herein to refer to a resource element within a resource element in a CSI-RS resource block that is mapped to a CSI-RS sequence.

[0113] Transmission of candidate basis vector information via independent channels

[0114] As an alternative to the above CSI-RS precoding or mapping, the candidate basis vector information can be transmitted to the terminal using explicit or implicit signaling. For example, the wireless access network node can use explicit or implicit signaling to inform the terminal of the candidate second basis vector set determined by the wireless access network node via a corresponding signaling channel . The explicit signaling can include, but is not limited to, MAC CE signaling, while the implicit signaling can include, but is not limited to, embedding the candidate basis vector information in CSI-RS related configuration parameters, such as sequence initialization parameters, cyclic shift parameters, etc.

[0115] In some embodiments, the wireless access network node can configure the candidate second basis vector information in the trigger signal of the aperiodic CSI-RS for the terminal to select. For example, for a predefined set of second basis vectors (including M second basis vectors), the information about the candidate second basis vectors selected by the wireless access network node can be configured in the CSI-RS trigger signal to inform the terminal in at least one of the following ways, including but not limited to: (1) configuring the selection of the candidate second basis vectors as a bitmap, and (2) using a combination number encoding scheme of M-choose K to configure the selection of the candidate second basis vectors, where K is a positive integer less than or equal to M (denoting the number of candidate basis vectors).

[0116] Although the above description refers to the second basis vectors and the candidate second basis vectors (frequency domain basis vectors), the basic principles apply to other sets of basis vectors.

[0117] CSI compression

[0118] In practical application scenarios, Figure 1 The UE 104 can be a mobile terminal device and can move relative to the wireless access network node 102. Such movement can cause time variation of the characteristics of the wireless channel. The wireless channel is thus affected by the Doppler effect. In the case of a UE moving at an estimable medium to high speed, the time variation of the characteristics of the wireless channel can need to be taken into account for the design and transmission of the CSI-RS and the CSI report. For example, the CSI-RS can need to be transmitted by the wireless access network node 102 and measured by the UE 104 respectively and frequently, so that the channel variation can be tracked throughout the movement of the UE 104. For example, the frequent transmission of the CSI report can significantly increase the transmission overhead and reduce the radio resource utilization efficiency of the wireless system.

[0119] In some embodiments, CSI reporting by a UE can be compressed in time, such that not every CSI needs to be fed back. For example, a single CSI report can be fed back for multiple CSI-RSs. As another example, the number of CSI reports can be less than the number of multiple CSI-RSs. In particular, a delay-Doppler approach can be utilized by a UE to compress the characteristic parameters of a wireless channel in time domain to reduce the time domain overhead of CSI feedback. For example, a UE can only need to send a reduced number of CSI reports (e.g., a single CSI report) corresponding to multiple CSI-RSs received by the UE from a wireless access network node within a CSI-RS measurement window, where the CSIs in the multiple CSI-RSs are aggregated and compressed in time. After receiving the time-compressed CSI report(s), the wireless access network node recovers the CSI information of the measurement time window and extrapolates (or periodically predicts) the channel information in a subsequent future time period. In this way, high-precision CSI can be achieved with lower transmission overhead.

[0120] Figure 7 The above principles are further illustrated in FIG. 7. In particular, Figure 7 Compression of subband-based CSI feedback is illustrated. The basic principles apply to other frequency-domain CSI feedback. Figure 7 The cube 702 of compressed CSI feedback information, e.g., a weighted coefficient matrix in a basis vector space determined by a terminal from multiple CSI-RSs received in time. The dimensions 704 of the compressed coefficient cube 702 illustrate a compressed time dimension representing the multiple CSI-RSs that can be transmitted in respective time slots 706. The original time dimension can be of size T, and the compressed time dimension can be denoted by N. The dimensions 708 of the compressed coefficient cube 702 illustrate a subband dimension 710 (e.g., which relates to the number of the second set of basis vectors or the set of frequency-domain basis vectors discussed above). The original subband dimension can be of size S, and it can be compressed to M. The dimensions 712 of the compressed coefficient cube 702 illustrate a transmit port (or spatial) dimension 714 (which relates to the parameter L of the weighted coefficients of the second set of basis vectors or the set of frequency-domain basis vectors discussed above). The spatial dimension 712 is illustrated as L (in some embodiments, polarization can be included in the spatial dimension. For example, with two possible polarization bases under consideration, the combined spatial and polarization dimension can have size 2L). As described in more detail below, the task of compression is to reduce the dimensionality of the original weighted coefficient cube to the compressed weighted coefficient cube 702 of dimensions In particular, time dimension compression can be achieved by exploiting the time-domain correlation of the weighted coefficients. In this way, the information that needs to be transmitted for CSI can be reduced, and the efficiency of CSI transmission in wireless resource consumption can be improved.

[0121] In some exemplary implementations based on the above principles, spatial and frequency domain compression of the original weighting coefficients can be performed as follows to obtain spatially and frequency compressed weighting coefficients C:

[0122]

[0123] .

[0124] The weighted coefficients C formed by spatial and frequency compression over T time slots are... The three-dimensional matrix is ​​then used to compress the weighted coefficient vector of size T in the time dimension to size N using DFT vectors (TD, or time-domain basis). Specifically, for a specific spatial dimension (port or beam) l and frequency-domain basis m, a weighted coefficient time vector of size T can be compressed to size N, as shown below:

[0125]

[0126] The basis vectors used above for performing time-domain compression This can be referred to as the third basis vector. Therefore, the time-compressed weighting coefficients of the CSI layer in the space of a certain first basis vector (space) and / or a certain second basis vector (frequency) over T time slots are a linear weighted combination of N third basis vectors. The terminal determines or obtains the third basis vector information and the amplitude and / or phase information of the compressed weighting coefficients in the third basis vector space, and feeds it back to the radio access network node.

[0127] As discussed earlier, the set of first basis vectors, the set of second basis vectors, and the set of third basis vectors can be drawn from the first codebook, the second codebook, and the third codebook, respectively. Furthermore, at least one of the sets can be selected by the terminal from the corresponding set of candidate basis vectors. This set of candidate basis vectors can be configured by the radio access network node.

[0128] To compress time-varying CSI by utilizing the correlation of CSI parameters in the time domain, multiple time-domain samples of CSI-RS can be used, such as multi-slot CSI-RS, like... Figure 8 and Figure 9 As shown.

[0129] Figure 8 A multi-slot CSI-RS configuration 800 is shown for periodic CSI-RS transmission and for triggering compressed CSI reports. The measurement time window T (820) can be configured for the terminal by the wireless access network node. CSI reports can be triggered by DCI or a specific CSI trigger signal, as shown in 830. Figure 8In this system, the terminal receives periodic CSI-RS data, as shown in Figures 802 to 812. Figure 8 In the example, CSI-RS is periodically transmitted and received by the terminal in two-slot intervals. The time slots are shown by empty boxes 801. Shaded boxes within the time slots represent CSI-RS. The time window T value can be measured in periods. Time window 820 contains the CSI-RS, and the terminal is configured to measure and compress the CSI-RS when a CSI report is triggered by a DCI or a specific CSI trigger signal 830. Specifically, once trigger 830 is received, the terminal can begin compressing the CSI information measured during time window 830. Once the compression result is obtained, the terminal then transmits the CSI report to the radio access network node, as shown in 840.

[0130] Although Figure 8 In the example, the terminal is configured to compress CSI-RS measurements within a T window that occurs just before CSI trigger 830. However, in some other implementations, the terminal may alternatively be configured to use a different T time window that does not occur later than CSI trigger 830.

[0131] Furthermore, in order to ensure that the first CSI feedback has a sufficient number of CSI-RS to compress, it may be necessary to limit the CSI trigger 830 to transmission at least after the time window T.

[0132] The compression schemes described above increase the signal processing complexity of the terminal. Specifically, the terminal is limited by the number of CSI-RS it can process within a reasonable timeframe. Therefore, the above schemes can also include mechanisms to ensure that the terminal does not operate beyond its capabilities. For example, the terminal can notify the radio access network node of its capabilities by reporting a maximum time window T0 based on its processing capacity. Thus, the radio access network can configure the terminal's time window T to be at most the value of T0. In some implementations, the CSI reference resource can be specified to represent a reference time, and the triggering of the CSI report can be configured such that there are no other CSI-RS between the CSI reference resource and trigger 830, or between (or including) the last symbol of the CSI reference resource and, for example, the first symbol of trigger 830. In this way, the terminal only needs to keep track of the T most recent CSI-RS preceding the CSI reference resource, rather than more CSI-RS, thereby reducing the terminal's processing requirements.

[0133] Figure 9 An exemplary multi-slot CSI-RS configuration 900 is further illustrated for periodic CSI-RS transmission and for triggering compressed CSI reports. Figure 9In some embodiments, as shown in 930, CSI can be triggered by, for example, DCI or other triggering signal, including triggering CSI-RS that can be sent from a radio access network node to a terminal. The CSI trigger 930 can initiate CSI measurement of the following CSI-RS (shown as 902-908) for compression and CSI reporting (shown as 940).

[0134] Such a triggering signal (e.g., triggering CSI-RS) can be configured to include CSI reporting time information or configuration that the terminal can use to configure its CSI measurement, compression, and reporting. For example, the triggering information can include, but is not limited to:

[0135] A slot offset indicating the number of slots between the DCI or other signaling triggering the CSI feedback and the first CSI-RS after the triggering signal.

[0136] The number of CSI-RS for measurement and compression (e.g., Figure 9 of 920, showing an exemplary number of CSI-RS configured for measurement and compression for reporting at 940). In particular, the number of CSI-RS can be configured to be less than the above-mentioned T0 (T0 is reported by the terminal as the maximum number of CSI-RS that the terminal can process for compression according to the processing capability of the terminal).

[0137] The number of slots between two adjacent CSI-RS.

[0138] In the case involving aperiodic CSI-RS, a CSI reference resource can also be specified to indicate the reference time. The triggering of the CSI report can be configured such that there is no other CSI-RS between the trigger 930 and the CSI reference resource, or there is no other CSI-RS between the last symbol of the trigger 930 and the first symbol of the CSI reference resource (or between and including them).

[0139] In some other alternative embodiments, a CSI trigger (such as a DCI trigger) can trigger more than one multi-CSI-RS set. Such a multi-set configuration can be specified by a bitmap.

[0140] In Figure 9In the embodiments shown in FIG. 9, because the CSI-RS is aperiodic or non-cyclic, the time required to receive T number of CSI-RS can be too long. In some embodiments, a separate threshold number T1may be configured such that if the time from the CSI trigger to the CSI report transmission is less than or equal to T1, the CSI is transmitted directly. However, if the time from the CSI trigger to the CSI report transmission is greater than T1, the CSI is reported multiple times. The content of each report is different. The DCI can trigger multiple such reports. The number of multiple CSI reports and the time interval between CSI reports can be configured by the radio access network node.

[0141] In some other embodiments, the above measurement time window T can include aperiodic CSI-RS before the trigger 930 and CSI-RS after the CSI trigger 930. In other words, the CSI-RS before and after the CSI trigger can be combined to generate the CSI report. These CSI-RS can even be transmitted using different resource allocations, but can be related in terms of the transmission ports.

[0142] In some embodiments, the above T number of time slots of rank indicator (RI) can be the same. Thereby, only one RI is fed back by the terminal. In some other exemplary embodiments, each of the T number of time slots can have a different RI, and these different RIs are independently fed back in the compressed CSI report. In this case, the PMI can be determined by the maximum of the multiple RIs, and then fed back to the radio access network node.

[0143] Enhanced transmission of compressed CSI reports

[0144] In the above compressed CSI feedback mechanism, after the third basis vector is determined, the weighting coefficient matrix of size may be compressed into a weighting coefficient matrix of size The amplitude and phase information in the compressed weighting coefficient matrix need to be fed back to the radio access network node. Generally, due to the sparsity of the transformed channel domain, most of the elements in the compressed coefficient matrix are zero elements. Thereby, the entire matrix does not need to be reported, only the position information and amplitude / phase information of the non-zero elements need to be fed back, which can effectively reduce the CSI feedback overhead.

[0145] Several exemplary embodiments of the position information feedback of the non-zero elements in the dimensional weighting coefficient matrix are described below. In some exemplary embodiments, the position of the non-zero elements can be fed back using a bitmap. The dimension of the bitmap can be .

[0146] In some other implementations, the coordinate value (x, y, z) of each non-zero element in the three-dimensional coordinate system (L, M, N) can be used as three independent position indices for position feedback. The value (x, y, z) can be used as an index or indicator to suggest the weighting coefficients of the non-zero coefficients for the x-th first basis vector, the y-th second basis vector, and the z-th third basis vector. The specific encoding of this position information (e.g., (x, y, z)) can be related to... The bit-width joint encoding. Specifically, the value of the linearization index of the joint encoding can be reported. ,or ,or Alternatively, (x, y, z) can be reported independently. Specifically, the bit width of x (log2(L)), the bit width of y (log2M), and the bit width of z (log2N) can be reported.

[0147] In some other implementations, The weighted coefficients can be divided into K groups, where each group has, for example, LMN / K coefficients. Accordingly, The vector space can be divided into K sub-vector spaces. The positional information of non-zero elements in a group can include three separate indices / indicators indicating the position of the non-zero element in the corresponding sub-vector space. For the linearized index of the joint encoding i above, the value i' = mod (i, LMN / K) of the non-zero element within each group can be fed back instead. The weighted coefficients in different groups can be mapped in a predefined order. A fourth index can be used to identify the sub-vector space. In some implementations, for a total of K0 non-zero coefficients, a third index is used. Each bit is used as a fourth index to identify the group and indicate the starting position of the coefficients in each group. For example, if... If the weighted coefficients are divided into two groups, then each group includes There are 10 coefficients. The first and second groups include weighted coefficients corresponding to the first half and the second half of the CSI-RS port. The number of feedback bits at the starting position of the weighted coefficients in each group is log2 K0.

[0148] The magnitude, phase, and position of the above non-zero coefficients can be mapped to groups of uplink control information (UCI) bit sequences in various exemplary ways. Figure 10 An example is shown below. Figure 10 As shown in figures 1002 to 1012, after the bit sequences corresponding to the amplitude, phase, and position values ​​of each coefficient are concatenated, they are mapped into the UCI bit sequence in coefficient order. In some implementations, a total of K totThe coefficients can be divided into A groups (shown as follows) Figure 10 The two groups (segments) 1020 and 1030 in the data contain each corresponding to K. tot / A coefficients' amplitude, phase, and position values. For example... Figure 10 As further illustrated in bit sequence 1000, these coefficient groups and their positions can be mapped in the UCI bit sequence in the order of the coefficients and the order of the groups. Additionally, depending on the resources and parameters allocated by PUSCH (such as beta_offset), the terminal can ignore one or more of the A groups and then perform a UCI transfer for CSI feedback. The order of the coefficients or the groups can be based on priority.

[0149] The above description and accompanying drawings provide specific example embodiments and implementations. However, the described subject matter can be embodied in a variety of different forms, and therefore, the covered or claimed subject matter is intended to be construed as not being limited to any of the example embodiments set forth herein. A reasonably broad scope of the claimed or covered subject matter is contemplated. For example, the subject matter can be embodied as a method, apparatus, component, system, or non-transitory computer-readable medium for storing computer code. Therefore, embodiments can take the form of, for example, hardware, software, firmware, storage media, or any combination thereof. For example, the above-described method embodiments can be implemented by a component, apparatus, or system including a memory and a processor by executing computer code stored in the memory.

[0150] Throughout the specification and claims, terms may have suggestive or implied nuances in context, in addition to their expressly stated meanings. Similarly, the phrase "in one embodiment / implementation" as used herein does not necessarily refer to the same embodiment, and the phrase "in another embodiment / implementation" as used herein does not necessarily refer to different embodiments. For example, the claimed subject matter is intended to include combinations of exemplary embodiments, in whole or in part.

[0151] Generally, the terminology can be understood at least in part from usage of the terms in the context in which they are used. For example, terms, such as "and", "or", or "and / or", as used herein, can include a variety of meanings that can depend at least in part upon the context in which they are used. Typically, "or" if used to associate a list, such as A, B or C, is intended to mean A, B, and C, here used in the inclusive sense, as well as A, B or C, here used in the exclusive sense. In addition, the term "one or more" as used herein, at least depending on the context, can be used to describe any feature, structure, or characteristic in the singular or can be used to describe combinations of features, structures or characteristics. Similarly, terms, such as "a", "an" or "the", again, can be understood to convey a singular usage or to convey a plural usage, at least depending on the context in which they are used. Also, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, and can instead allow for additional factors to be present, whether or not explicitly described. Additionally, such terms as "comprising", "including", "containing", etc. can be understood to be used

[0152] Reference throughout this specification to features, advantages, or similar language does not mean that all of the features and advantages that can be achieved with the present solutions should or must be present in any single implementation of the present solutions. Rather, languages

[0153] Additionally, in one or more embodiments, the features, advantages, and characteristics of the present solutions can be combined in any suitable manner. Persons skilled in the relevant art will recognize, in light of the description herein, that the present solutions can be practiced without one or more of the specific features or advantages of a particular embodiment. In other instances, additional features and advantages can be recognized in certain embodiments that can not be present in all embodiments of the present solutions.

Claims

1. A method for reporting wireless channel state information (CSI), comprising: receiving, by a wireless terminal device, a CSI reference signal (CSI-RS) transmitted from a wireless access network node; generating, by the wireless terminal device, one or more indicators associated with one or more non-zero elements in CSI feedback information in a vector space determined by a set of first basis vectors in spatial domain, a set of second basis vectors in frequency domain, a set of third basis vectors in time domain, based on measuring the CSI-RS; and transmitting, by the wireless terminal device, information about the one or more non-zero elements in the CSI feedback information and the one or more indicators to the wireless access network node, wherein the information about the one or more indicators comprises location information of the one or more non-zero elements in the vector space, wherein the location information comprises linearized indices representing first, second and third indices in the set of first basis vectors, the set of second basis vectors and the set of third basis vectors, respectively. 2.The method of claim 1, wherein the set of first basis vectors, the set of second basis vectors and the set of third basis vectors are selected by the wireless terminal device from a first codebook, a second codebook and a third codebook, respectively. 3.The method of claim 1, wherein at least one of the set of first basis vectors, the set of second basis vectors and the set of third basis vectors is selected from a set of candidate basis vectors configured by the wireless access network node. 4.The method of claim 3, wherein the set of candidate basis vectors is selected from a predetermined codebook. 5.The method of claim 1, wherein the location information of a non-zero element in the one or more non-zero elements comprises first, second and third indices in the set of first basis vectors, the set of second basis vectors and the set of third basis vectors, respectively. 6.The method of claim 1, wherein: the vector space is divided into a plurality of sub-vector spaces; and the location information of a non-zero element in a sub-vector space comprises: at least one of: first, second and third indices of the non-zero element in the sub-vector space, or linearized indices representing the first, second and third indices of the non-zero element in the sub-vector space; and a fourth index identifying the sub-vector space among the plurality of sub-vector spaces. 7.The method of claim 1, wherein the information about a non-zero element in the CSI feedback information and the information about an indicator associated with the non-zero element are mapped to adjacent locations in uplink control information when transmitted by the wireless terminal device to the wireless access network node. 8.The method of claim 1, wherein: the one or more non-zero elements are divided into a plurality of groups based on a priority ordering; and the information about the one or more non-zero elements and the information about the one or more indicators are mapped to uplink control information according to the priority ordering. ​ ​ ​ ​ 9. A method for reporting wireless channel state information, CSI, comprising: transmitting, by a radio access network node, a CSI reference signal, CSI-RS, to a wireless terminal device; and receiving, by the radio access network node, a CSI report based on measurements of the CSI-RS made by the wireless terminal device, wherein the CSI report includes: information about one or more indicators associated with one or more non-zero elements of CSI feedback information generated by the wireless terminal device in a vector space determined by a first set of basis vectors in a spatial domain, a second set of basis vectors in a frequency domain, and a third set of basis vectors in a time domain, wherein the information about the one or more indicators includes location information of the one or more non-zero elements in the vector space, and wherein the location information includes linearized indices representing first, second, and third indices in the first, second, and third sets of basis vectors, respectively; and information about the one or more non-zero elements.

10. The method of claim 9, wherein the first, second, and third sets of basis vectors are selected by the wireless terminal device from first, second, and third codebooks, respectively.

11. The method of claim 9, wherein at least one of the first, second, and third sets of basis vectors is selected from a set of candidate basis vectors configured by the radio access network node.

12. The method of claim 11, wherein the set of candidate basis vectors is selected from a predetermined codebook.

13. The method of claim 9, wherein the location information of a non-zero element of the one or more non-zero elements includes first, second, and third indices in the first, second, and third sets of basis vectors, respectively.

14. The method of claim 9, wherein: the vector space is divided into a plurality of sub-vector spaces; and the location information of a non-zero element in a sub-vector space includes: at least one of: first, second, and third indices of the non-zero element in the sub-vector space, or linearized indices representing first, second, and third indices of the non-zero element in the sub-vector space; and a fourth index identifying the sub-vector space among the plurality of sub-vector spaces.

15. The method of claim 9, wherein the information about a non-zero element in the CSI feedback information and the information about an indicator associated with the non-zero element are received by the radio access network node in adjacent locations in uplink control information.

16. The method of claim 9, wherein: the one or more non-zero elements are divided into a plurality of groups based on a priority ordering; and The information related to the one or more non-zero elements and the information related to the one or more indicators are received by the radio access network node when mapped in uplink control information according to a priority ordering.

17. A wireless communication device comprising one or more processors and one or more memories, wherein the one or more processors are configured to read computer code from the one or more memories to implement the method of any one of claims 1 to 16.

18. A computer program product comprising a non-transitory computer-readable program medium having computer code stored thereon, the computer code, when executed by one or more processors, causing the one or more processors to implement the method of any one of claims 1 to 16.

Images