RU2779138C1 - Channel state information feedback in wireless communication - Google Patents

Abstract

FIELD: wireless communication.

SUBSTANCE: technical result is achieved by compressing the pre-coding vector coefficients and communicating a subset of the compressed coefficients based on multiple parameters, including parameters signalled via the network.

EFFECT: reduced service load on the channel state information (CSI) feedback channel.

14 cl, 5 dwg

Description

FIELD OF TECHNOLOGY TO WHICH THE INVENTION RELATES

[0001] This document relates generally to wireless communications.

BACKGROUND OF THE INVENTION

[0002] Wireless technologies are making the world community increasingly connected and networked. A key factor in meeting the requirements of enhanced mobile broadband in emerging 5th generation New Radio (NR) (5G) networks is the massive adoption of multiple input multiple output (MIMO) and beamforming techniques where multiple transmit antennas and/or multiple receive antennas are used in wireless nodes. The use of these base stations and user devices can increase the performance, efficiency and reliability of the wireless communication channel between them. In such systems, it is important to accurately estimate and report channel state information (CSI). However, the CSI message overhead increases with the number of used frequency bands and transmit/receive antennas, which is especially problematic given the large bandwidths and multiple spatial streams from many transmit/receive antennas in 5G NR networks.

SUMMARY OF THE INVENTION

[0003] This document relates to methods, systems, and apparatuses for reducing overhead in reporting channel state information (CSI), such as precoding matrix pointers, by compressing coefficients of precoding vectors. In some embodiments, the message is performed using only a subset of the compressed coefficients based on various criteria.

[0004] In one illustrative aspect, a method for wireless communication of a wireless communication device is disclosed. The method includes determining spatial basis vectors and coefficients of the spatial basis vectors, where the spatial channel precoding vector is given by a linear combination of the spatial basis vectors and the coefficients of the spatial basis vectors. The precoding vector is a vector that can be used to precode a transmitted data stream to counter wireless channel violations. The method further includes compressing the coefficients of the spatial basis vectors by determining unit frequency (FD) basis vectors and the coefficients of the FD basis vectors, whereby the combination of the FD basis vectors and the coefficients of the FD basis vectors defines the coefficients of the spatial basis vectors, but with reduced overhead. load. The wireless communication device may then generate a CSI (and a corresponding CSI feedback report) based on the spatial basis vectors, the FD basis vectors, and the coefficients of the FD basis vectors, among other parameters.

[0005] In another illustrative aspect, a wireless communication device comprising a processor is disclosed. The processor is configured to implement the method described above.

[0006] In another illustrative aspect, a computer program product is disclosed. The computer program product includes a computer-readable medium storing processor-executable instructions implementing the method described above.

[0007] The above and other aspects and their implementation are described in more detail in the drawings, descriptions and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1 shows an example of a base station (BS) and a user equipment (UE) in wireless communication.

[0009] FIG. 2 shows an exemplary workflow showing a method for compressing coefficients of precoding vectors to determine CSI.

[0010] FIG. 3 shows an exemplary flowchart showing a method for compressing coefficients of precoding vectors.

[0011] FIG. 4 shows an exemplary flowchart showing a method for compressing coefficients of precoding vectors based on some constraints.

[0012] FIG. 5 shows a block diagram representation of a portion of the device.

DETAILED DESCRIPTION

[0013] There is a growing need for fourth generation mobile communication technology (4G, 4th Generation mobile communication technology), "project for the long-term development of communication systems" (LTE, Long Term Evolution) and fifth generation mobile communication technology (5G, 5th Generation mobile communication technology ), also referred to as NR (New Radio).

[0014] In MIMO wireless communication systems, multiple antennas are used to carry out signaling. Such implementations include processing at the transmitter side, such as precoding or beamforming, to improve performance, including transmission efficiency and reliability. To achieve high performance precoding or beamforming, a precoding matrix or beamforming vector is selected according to the wireless channel. Therefore, the transmitter needs to determine the channel state information (CSI) for accurate precoding or beamforming of the transmitted signal. The receiver may determine the CSI based on the received reference signals (eg, CSI reference signals (CSI-RS), sounding reference signals (SRS), etc.) or pilot signals, and may then return the CSI to the transmitter (eg, the UE may report CSI to BS). Accurate CSI feedback enables high performance MIMO transmission.

[0015] However, high resolution CSI feedback is costly in terms of the overhead required on the feedback channel. This is especially true when the transmitter needs CSI across multiple subbands (frequency segments) or transmission levels (spatial streams). The optimal trade-off between performance and overhead for CSI feedback is the main metric for realizing high definition CSI performance.

[0016] In MIMO systems, a user equipment (e.g., a user equipment (UE)) typically reports to a wireless node (e.g., a base station) a CSI that includes a precoding matrix indicator (PMI), a rank indicator (RI), and a channel quality indicator (CQI), among other parameters. RI indicates the number of layers (which is related to the rank of the channel matrix), while PMI indicates the precoding vector. The precoding vector is used by the wireless node to precode each layer and is expressed as a linear combination of a plurality of spatial basis vectors. The UE quantizes (eg, digitizes) the amplitudes and phases of the linear combination coefficients and the selected spatial basis vectors and reports the quantized values to the base station.

[0017] To enable frequency selective scheduling, for example, quantized phases and amplitudes are reported for each subband (or subset of subbands representing a frequency unit) if multiple subbands are contained within the range of the CSI message. This high resolution CSI feedback results in high throughput MIMO transmission. However, in order to determine or report CSI in multiple frequency units (or subbands) or across multiple layers, the overhead can be quite large, resulting in a large resource consumption on the reverse channel. Additionally, such a high definition CSI increases the complexity of the UE and results in increased power consumption. Therefore, it is useful to have a CSI reporting method that provides high resolution CSI for multiple subbands and/or layers, provides high performance MIMO or beamforming, but at the same time reduces CSI reporting overhead.

[0018] This document uses section headings and subheadings to facilitate understanding, but not to limit the scope of the disclosed methods and embodiments to certain sections. Accordingly, the embodiments disclosed in the various sections may be used with each other. In addition, examples from the 3GPP New Radio (NR) network architecture and 5G protocol are used herein only to facilitate understanding, and the disclosed methods and embodiments can be practiced in other wireless systems that use non-3GPP communication protocols.

[0019] FIG. 1 shows an example of a wireless communication system (e.g., LTE, 5G, or new radio (NR)) that includes a BS 120 and one or more user equipments (UEs) 111, 112, and 113. Uplink transmissions (131, 132, 133) may, according to this document, include CSI feedback reports. The UE may be, for example, a smartphone, tablet, mobile computer, machine-to-machine (M2M) device, terminal, mobile device, internet of things (IoT) device, and so on.

для пространственного канала. В некоторых вариантах осуществления L векторов пространственного базиса могут формироваться на основании векторов дискретного преобразования Фурье (DFT) или кронекеровского произведения векторов DFT. [0020] FIG. 2 shows an exemplary flowchart showing a method for compressing coefficients of precoding vectors. At block 210, the wireless device (eg, UE) determines L spatial basis vectors

for the spatial channel. In some embodiments, the L spatial basis vectors may be generated based on the discrete Fourier transform (DFT) vectors or the Kronecker product of the DFT vectors.

} (т.е. коэффициенты, имеющие модуль и фазу), где l=1, 2, ..., L - номера векторов пространственного базиса, s=1, 2, ..., S - номера частотных (FD) единиц или поддиапазонов (т.е. частотные сегменты, по которым нужно сообщать CSI), и r=1, 2, ..., R - индексы уровней для пространственного канала ранга R (т.е. уровней ранга R). Комплексные коэффициенты {

} (упрощенно обозначаемые как {a}) выбираются таким образом, чтобы комбинация (например, линейная комбинация) комплексных коэффициентов {a} с векторами пространственного базиса определяла вектор предварительного кодирования для пространственного канала по всем уровням r

{1, 2,...,R} и всем FD-единицам s

{1, 2,...,S}. UE может определять векторы пространственного базиса и коэффициенты векторов пространственного базиса из принятых опорных сигналов или пилот-сигналов, передаваемых беспроводным узлом (например, базовой станцией), например, из CSI-RS (опорных сигналов CSI) с известными амплитудами и фазными сдвигами.[0021] At block 220, the wireless device determines the complex coefficients {

} (i.e. coefficients having modulus and phase), where l=1, 2, ..., L - numbers of spatial basis vectors, s=1, 2, ..., S - numbers of frequency (FD) units or subbands (ie, frequency segments on which CSI is to be reported), and r=1, 2, ..., R are the layer indices for the spatial channel of rank R (ie, the levels of rank R). Complex coefficients {

} (simplistically denoted as {a}) are chosen such that the combination (eg, linear combination) of the complex coefficients {a} with the spatial basis vectors determines the precoding vector for the spatial channel over all layers r

{1, 2,...,R} and all FD-units s

{1, 2,...,S}. The UE may determine the spatial basis vectors and coefficients of the spatial basis vectors from received reference signals or pilot signals transmitted by a wireless node (eg, base station), for example, from CSI-RS (CSI Reference Signals) with known amplitudes and phase shifts.

над L, и по всем уровням r ранга R, и на блоке 240 UE определяет комплексные коэффициенты

(упрощенно обозначаемые как {c}) над L, M и по всем уровням r ранга R. UE выбирает векторы FD-базиса и комплексные коэффициенты {c} таким образом, чтобы комбинация векторов FD-базиса и комплексных коэффициентов {c} (например, линейная комбинация) определяла комплексные коэффициенты {a} блока 220. Кроме того, как дополнительно описано ниже, сеть и UE гарантируют, что количество битов, необходимых для сообщения векторов FD-базиса и комплексных коэффициентов {c}, меньше, чем количество битов, необходимых для сообщения {c}, что позволяет снизить служебную нагрузку при сообщении обратной связи по CSI после сжатия (например, благодаря выбору M). Различные методы дополнительного снижения этой служебной нагрузки описаны ниже в отношении вариантов осуществления раскрытой технологии. [0022] At blocks 230 and 240, the wireless device compresses the complex coefficients {a} to reduce overhead when reporting CSI to the base station. At block 230, the UE determines M FD basis vectors

over L, and over all levels r of rank R, and at block 240, the UE determines the complex coefficients

(simplistically denoted as {c}) over L, M, and over all levels r of rank R. The UE chooses the FD basis vectors and complex coefficients {c} such that the combination of the FD basis vectors and complex coefficients {c} (for example, linear combination) determined the complex coefficients {a} of block 220. In addition, as further described below, the network and UE ensure that the number of bits required to report the FD basis vectors and complex coefficients {c} is less than the number of bits required for the {c} message, thus reducing the overhead of reporting the CSI feedback after compression (eg, due to the choice of M). Various methods for further reducing this overhead are described below with respect to embodiments of the disclosed technology.

[0023] At block 250, the wireless device generates a CSI feedback report based on the spatial basis vectors, FD basis vectors, complex coefficients {c}, and other additional parameters described below. Thus, instead of reporting the complex coefficients {a}, the UE may report, with reduced overhead, the complex coefficients {c} together with the FD basis vectors.

[0024] At block 260, the wireless device sends CSI feedback reports to the wireless node (eg, base station on a Physical Uplink Shared Channel (PUSCH) or a Physical Uplink Control Channel (PUCCH)). The base station may apply this feedback information in deciding whether to precode or beamform the downlink data streams and generate transmission waveforms. In some embodiments, the wireless device may be a base station and may exclude the generation and transmission of a CSI feedback report.

[0025] FIG. 3 shows a flowchart showing a method for compressing the coefficients of precoding vectors (eg, MIMO precoding vectors or beamforming weights). Such precoding vectors may be used, for example, in CSI feedback reports sent from UEs to base stations to support high performance MIMO/beamforming transmissions.

[0026] 1.0 Exemplary Embodiments for Compressing Coefficients of a Precoding Vector and Reporting a Subset of Coefficients for Each Layer

[0027] In some embodiments, user devices (eg, UEs) return channel state information (CSI) to the wireless node (eg, base stations) in the form of a precoding matrix indicator (PMI). PMI can be represented as a linear combination of spatial basis vectors at each level of r and frequency (FD) units. For example, for L spatial basis vectors [ v ₁ , v ₂ , ..., v _L ], the precoding vector can be represented as:

} - коэффициенты вектора предварительного кодирования, используемого для формирования линейной комбинации с векторами пространственного базиса. В некоторых вариантах осуществления L векторов пространственного базиса могут формироваться на основании векторов дискретного преобразования Фурье (DFT) или кронекеровского произведения векторов DFT. Коэффициенты {

} являются комплексными переменными, включающими в себя модуль и фазу, и квантуются пользовательским устройством и сообщаются на базовую станцию как часть сообщаемой обратной связи по CSI. Поскольку коэффициенты {

} могут различаться для разных частотных единиц и/или уровней, служебная нагрузка при сообщении квантованных амплитуд и фаз {

} может быть довольно большой в особенности для большого количества FD-единиц (например, для широкого диапазона, разделенного на множественные малые поддиапазоны или FD-единицы, где CSI сообщается по всем таким FD-единицам), и для большого количества уровней (например, для массовых систем MIMO с большим количеством передающих и приемных антенн по полноранговому каналу). where r is the level index, s is the index of the FD-unit { v ₁ , v ₂ , ..., v _L } - L spatial basis vectors, and {

} - coefficients of the precoding vector used to form a linear combination with the spatial basis vectors. In some embodiments, the L spatial basis vectors may be generated based on the discrete Fourier transform (DFT) vectors or the Kronecker product of the DFT vectors. Coefficients {

} are complex variables including magnitude and phase and are quantized by the user device and reported to the base station as part of the reported CSI feedback. Since the coefficients {

} may differ for different frequency units and/or levels, the overhead of reporting quantized amplitudes and phases {

} can be quite large, especially for a large number of FD units (for example, for a wide range divided into multiple small subbands or FD units, where CSI is reported over all such FD units), and for a large number of levels (for example, for mass MIMO systems with a large number of transmit and receive antennas over a full rank channel).

}, UE сжимает коэффициенты {

} с использованием модуля 350 сжатия коэффициентов, как показано на фиг. 3. Таким образом, для каждого уровня r коэффициент вектора пространственного базиса l (v _l ) по всем S FD-единицам (блок 352 на фиг. 3) можно выразить в виде:[0028] Therefore, in some illustrative embodiments, in order to reduce the overhead in reporting the quantized amplitudes and phases {

}, the UE compresses the coefficients {

} using coefficient despreader 350 as shown in FIG. 3. Thus, for each level r, the coefficient of the spatial basis vector l (v _l ) over all S FD units (block 352 in FIG. 3) can be expressed as:

(блок 354 на фиг. 3) - M векторов FD-базиса длиной N3, и

в блоке 356 - коэффициент для l-го луча и m-го вектора FD-базиса после сжатия модулем 350 сжатия коэффициентов. Модуль 350 сжатия коэффициентов может генерировать вектор FD-базиса

на основании векторов DFT. where

(block 354 in FIG. 3) are M FD basis vectors of length N3, and

in block 356, the coefficient for the l-th beam and the m-th FD-basis vector after compression by the coefficient despreader 350 . Coefficient compression module 350 may generate an FD basis vector

based on DFT vectors.

(выход 312) для первой FD-единицы и уровня r, в виде линейной комбинации векторов пространственного базиса

(блок 314) и коэффициентов {

} (блок 316). Это можно выразить в виде:[0029] Prior to compression, the 1st precoder FD unit (block 310) generates a precoding vector

(output 312) for the first FD unit and level r, as a linear combination of spatial basis vectors

(block 314) and coefficients {

} (block 316). This can be expressed as:

для второй FD-единицы и уровня r, в виде линейной комбинации векторов пространственного базиса

и коэффициентов {

} (блок 326), что можно выразить в виде:[0030] Similarly, the 2nd precoder FD unit (block 320) generates a precoding vector

for the second FD-unit and level r, as a linear combination of spatial basis vectors

and coefficients {

} (block 326), which can be expressed as:

[0031] The sth precoder FD unit (block 330) generates a precoding vector, which can be expressed as:

(блок 354), и коэффициенты вектора FD-базиса

(блок 356), генерируемые модулем 350 сжатия коэффициентов вместо коэффициентов

. Ввиду корреляции коэффициентов

до сжатия, M может задаваться меньшим, чем 2L, приводя к снижению служебной нагрузки при передаче векторов FD-базиса

и коэффициентов

по сравнению со служебной нагрузкой при передаче не сжатых коэффициентов

. Преимущество сжатия может достигаться для M < S. [0032] In some embodiments, the UE reports the FD basis vector

(block 354), and the FD basis vector coefficients

(block 356) generated by coefficient despreader 350 instead of coefficients

. In view of the correlation of the coefficients

before compression, M may be set to less than 2L, resulting in reduced overhead when transmitting FD-basis vectors

and coefficients

compared to the overhead when transmitting uncompressed coefficients

. The compression advantage can be achieved for M < S.

может дополнительно снижаться, например, путем выбора, для каждого уровня, подмножество K0 из L*M коэффициентов. Таким образом, количество ненулевых коэффициентов ограничивается значением K0 (т.е. верхняя граница для количества коэффициентов в подмножестве равна K0), и UE сообщает местоположение или позицию K1 ненулевых коэффициентов среди L*M коэффициентов, и амплитуды и фазы K1 коэффициентов. [0033] In some embodiments, the overhead of reporting L*M coefficients

may be further reduced, for example, by selecting, for each level, a subset K0 of L*M coefficients. Thus, the number of non-zero coefficients is limited to the value of K0 (i.e., the upper bound for the number of coefficients in the subset is K0), and the UE reports the location or position of K1 of non-zero coefficients among the L*M coefficients, and the magnitude and phase of the K1 coefficients.

[0034] In some embodiments, the value of M may be based on N3 values and/or network settings (eg, N3=S). For example, M= ceil ( p N ₃ ), where p is based on the network setting, and M is the smallest integer greater than or equal to the product of p and the value of N3 (upper round function) ( ceil (x) = smallest integer greater than or equal to equal to x). In other embodiments, M may be limited to other parameters configured on the UE or derived by the UE.

[0035] In some embodiments, the implementation of the UE or the network may set the value of K0 depending on L, M and/or parameters configured by the network. For example, K0 may be given as K0= ceil ( β LM), where β may be based on configured parameters.

[0036] 2.0 Exemplary Embodiments for Limiting Overhead for a CSI Message

) по количеству уровней (например, по R уровням, где R - значение ранга), помимо прочих параметров. Битовая ширина указания количества ненулевых коэффициентов по количеству уровней может определяться максимальным количеством уровней. Максимальное количество уровней может определяться максимальным рангом, который может поддерживать кодовая книга (для предварительного кодирования на основе кодовой книги), параметром конфигурации сети, значением способности, которое UE сообщает на базовую станцию, и т.д. Во 2-й части CSI UE сообщает другие параметры CSI, например, указатель матрицы предварительного кодирования (PMI). Сообщаемый PMI может включать в себя указатель для векторов пространственного базиса (например, векторов 314 на фиг. 3), указатель для векторов FD-базиса (например, векторов 354), указатель амплитуды и фазы сжатых коэффициентов (например, коэффициентов 356), и указание местоположения ненулевых коэффициентов среди L*M коэффициентов. [0037] In some embodiments, the CSI may be divided into two parts, for example, a 1st CSI part and a 2nd CSI part. In the 1st part of the CSI, the UE may report a channel quality indicator (CQI), a rank indicator (RI), and a value indicating the number of non-zero coefficients (

) by the number of levels (for example, by R levels, where R is the rank value), among other parameters. The bit width of the indication of the number of non-zero coefficients by the number of levels may be determined by the maximum number of levels. The maximum number of layers may be determined by the maximum rank that the codebook can support (for codebook-based precoding), a network configuration parameter, a capability value that the UE reports to the base station, and so on. In the 2nd part of the CSI, the UE reports other CSI parameters, such as a precoding matrix indicator (PMI). The reported PMI may include an index for spatial basis vectors (eg, vectors 314 in FIG. 3), an index for FD basis vectors (eg, vectors 354), an amplitude and phase index for compressed coefficients (eg, coefficients 356), and an indication locations of non-zero coefficients among L*M coefficients.

[0038] In some embodiments, the CSI parameters in each part are jointly channel-coded, but the different parts are channel-coded independently. Additionally, the payload of the 2nd part of the CSI may depend on the value of the 1st part of the CSI, whereby the network does not always need to reserve the maximum overhead for transmitting the CSI. Thus, the 1st part of the CSI may have a fixed allocation and may contain parameters that always need to be returned, and the 2nd part of the CSI may include parameters that do not always need to be returned or that may have a variable size (for example, a variable bit width). By using this approach, when the 2nd part of the CSI does not use the maximum possible overhead, the unused overhead is saved from allocation, resulting in an increase in efficiency. For example, if the maximum number of layers that the UE reports in a capability value is R1 (i.e., the UE is capable of handling at most R1 levels), and the number of coefficients to report per layer is limited to K0, the minimum number of bits required to report K0 *R1 of non-zero coefficients, equals ceil (log ₂ (R1×K0)). Alternatively, if the maximum number of layers is limited by a network configuration parameter to a value of R2 (e.g., based on a parameter limiting the RI values that a UE can report), then the minimum number of bits for reporting K0*R2 non-zero coefficients is ceil (log ₂ (R2×K0 )). Alternatively, if the number of RI values that the codebook in use can support is R3, then the minimum number of bits to report R3*K0 non-zero coefficients is ceil (log ₂ (R3×K0)). In some embodiments, when the UE has multiple restrictions on the number or levels (e.g., the number of levels is limited to two or more of R1, R2, and R3), the bit width for reporting non-zero coefficients may be set to ceil (log ₂ (R4×K0) ), where R4 is the minimum of {R1, R2, R3}.

[0039] 3.0 Exemplary Embodiments for Limiting Overhead for a CSI Message for Large Ranks

(например, блок 356 на фиг. 3) по мере увеличения количества уровней R, (например, с увеличением ранга пространственного канала), аналогично увеличивается служебная нагрузка CSI. Поэтому, в некоторых вариантах осуществления, количество векторов пространственного базиса (например, векторов 314), количество векторов FD-базиса (например, векторов 354) или количество коэффициентов в подмножестве K0 из L*M коэффициентов может ограничиваться, чтобы не расти слишком сильно с повышением R (или чтобы расти медленнее), например, чтобы быть одинаковым при большом R и при малом R.[0040] With an increase in the number of compressed coefficients

(eg, block 356 in FIG. 3) as the number of R layers increases (eg, as the rank of the spatial channel increases), the CSI overhead increases similarly. Therefore, in some embodiments, the number of spatial basis vectors (e.g., 314 vectors), the number of FD basis vectors (e.g., 354 vectors), or the number of coefficients in the K0 subset of L*M coefficients may be limited so as not to grow too much with increasing R (or to grow slower), for example, to be the same for large R and for small R.

[0041] For a CSI report of rank R, if rank R is selected from candidate set S1 of rank values, the bit width for reporting non-zero coefficients depends on the maximum number of coefficients in the subset K0 that can be reported across all levels for any of the rank values in candidate set S1 . For example, if the rank R _max in S1 gives the maximum number of coefficients in the subset K0 across all levels, the bit width for reporting non-zero coefficients reaches ceil (log ₂ (R _max ×K0)). Thus, the bit width for reporting non-zero coefficients is ceil (log ₂ (max({S1}) × K0)) where max({S1}) is the rank (R _max ) in S1 giving the maximum number of coefficients in the subset K0. In some embodiments, the possible set of S1 may be determined by a network configuration parameter and/or a capability value reported by the UE.

R0

1. Дополнительно, в некоторых вариантах осуществления, битовая ширина, используемая для сообщения количества ненулевых коэффициентов в 1-й части CSI, может базироваться на количестве векторов пространственного базиса на каждом уровне, когда ранг равен R0. Например, битовая ширина может ограничиваться величиной ceil(log₂(R0 × ceil(β×M*2L*R0))), где L*R0 - количество векторов пространственного базиса на каждом уровне, когда ранг равен R0.[0042] In some embodiments, the total number of spatial basis vectors of all R levels for a CSI report of rank R may be limited to a value less than or equal to the total number of spatial basis vectors of all R0 levels for a CSI report of rank R0, where R

R0

1. Additionally, in some embodiments, the bit width used to report the number of non-zero coefficients in the 1st part of the CSI may be based on the number of spatial basis vectors in each level when the rank is R0. For example, the bit width may be limited to ceil (log ₂ (R0 × ceil (β×M*2L*R0))), where L*R0 is the number of spatial basis vectors at each level when the rank is R0.

R0

1. Дополнительно, в некоторых вариантах осуществления, битовая ширина, используемая для сообщения количества ненулевых коэффициентов в 1-й части CSI, может базироваться на количестве векторов FD-базиса на каждом уровне, когда ранг равен R0. Например, битовая ширина может ограничиваться величиной ceil(log₂(R0×ceil(β×M*2L*R0))), где M*R0 - количество векторов FD-базиса на каждом уровне, когда ранг равен R0. [0043] In some embodiments, the total number of FD basis vectors of all R levels for a CSI report of rank R may be limited to a value less than or equal to the total number of FD basis vectors of all R0 levels for a CSI report of rank R0, where R

R0

1. Additionally, in some embodiments, the bit width used to report the number of non-zero coefficients in the 1st part of the CSI may be based on the number of FD-basis vectors at each level when the rank is R0. For example, the bit width may be limited to ceil (log ₂ (R0× ceil (β×M*2L*R0))), where M*R0 is the number of FD basis vectors at each level when the rank is R0.

R0

1. Дополнительно, в некоторых вариантах осуществления, битовая ширина, используемая для сообщения количества ненулевых коэффициентов в 1-й части CSI, может базироваться на значении K0, когда ранг равен R0. Например, битовая ширина может ограничиваться величиной ceil(log₂(R0×K0*R0)), где K0*R0 - значение K0 на каждом уровне, когда ранг равен R0.[0044] In some embodiments, the total number of coefficients in the subset K0 of all R levels may be limited to a value less than or equal to the total number of coefficients in the subset K0 of all R0 levels for a CSI report of rank R0, where R

R0

1. Additionally, in some embodiments, the bit width used to report the number of non-zero coefficients in the 1st part of the CSI may be based on the value of K0 when the rank is R0. For example, the bit width may be limited to ceil (log ₂ (R0×K0*R0)), where K0*R0 is the value of K0 at each level when the rank is R0.

[0045] 3.1 Exemplary embodiments for limiting the number of FD basis vectors per level, or K0 values per level, for higher ranks to a predetermined fraction of values for lower ranks:

. То есть M может быть задано равным наибольшему целому числу, меньшему или равному

(floor(x) = наибольшее целое число, меньшее или равное x). [0046] In some embodiments, the number of FD basis vectors for each level for a CSI report of rank R may depend on the value of rank R. For example, if for ranks less than or equal to 2, the number of FD basis vectors for each level is M0, then the number of FD-basis vectors per level for any rank R>2 can be set equal to M= floor

. That is, M can be given equal to the largest integer less than or equal to

( floor (x) = largest integer less than or equal to x).

. Таким образом, отчет CSI ранга 4 должен иметь такое же суммарное количество векторов FD-базиса, как отчет CSI ранга 2.[0047] In some embodiments, the range of ranks may be limited such that the total number of FD basis vectors remains constant for the ranks in the range. For example, if the number of FD-basis vectors per level for rank 2 is equal to M0, then the number of FD-basis vectors per level for rank 3 and rank 4 can be limited to M= floor

. Thus, a rank 4 CSI report must have the same total number of FD basis vectors as a rank 2 CSI report.

.[0048] In some embodiments, the K0 value may be based on a rank value. For example, if the value of K0 for ranks less than or equal to 2 is K0 ₁ , then the value of K0 for rank R (R>2) can be set to K0= floor

.

.[0049] In some embodiments, the range of ranks may be limited to the same fraction of K0 values per level as the specified lower rank K0 value. For example, if ranks less than or equal to 2 have a K0 value of K0 ₁ , then rank 3 and rank 4 can be limited to a K0 value of floor

.

[0050] 3.2 Exemplary embodiments for limiting the number of spatial basis vectors per layer, the number of FD basis vectors per layer, and the value of K0 per layer based on the maximum rank given by the network configuration parameters:

, где R_cfg - максимальное значение ранга, заданное параметром конфигурации сети.[0051] In some embodiments, for a CSI report of rank R, the number of FD basis vectors for each layer may be based on a maximum rank value given by a network configuration parameter. For example, if the number of FD basis vectors per level for ranks less than or equal to 2 is M0, then the number of FD basis vectors per level for ranks greater than 2 can be set to M= floor

, where R_cfg is the maximum rank value specified by the network configuration parameter.

, где R_cfg - максимальное значение ранга, заданное параметром конфигурации сети.[0052] In some embodiments, for a rank R CSI report, each layer's K0 value may be based on a maximum rank value specified by a network configuration parameter. For example, if the value of K0 for each level for ranks less than or equal to 2 is K0 ₁ , then the value of K0 for each level for ranks greater than 2 can be set to K0= floor

, where R_cfg is the maximum rank value specified by the network configuration parameter.

, где R_cfg - максимальное значение ранга, заданное параметром конфигурации сети.[0053] In some embodiments, for a rank R CSI report, the number of spatial basis vectors for each layer may be based on a maximum rank value specified by a network configuration parameter. For example, if the number of spatial basis vectors per level for ranks less than or equal to 2 is L0, then the number of spatial basis vectors per level for ranks greater than 2 can be set to L=floor

, where R_cfg is the maximum rank value specified by the network configuration parameter.

[0054] 3.3 Exemplary embodiments for limiting the number of spatial basis vectors per layer, the number of FD basis vectors per layer, and the K0 value per layer based on UE capability:

, где R_UE - максимальное значение ранга, заданное значением способности UE.[0055] In some embodiments, for a rank R CSI report, the number of FD basis vectors for each layer depends on the maximum rank value that the UE can support, or that the UE reports in the capability parameter (i.e., the UE's capability value). For example, if the number of FD basis vectors per level for ranks less than or equal to 2 is M0, then the number of FD basis vectors per level for ranks greater than 2 can be set to M= floor

where R_UE is the maximum rank value given by the UE's capability value.

, где R_UE - максимальное значение ранга, заданное значением способности UE.[0056] In some embodiments, for a CSI report of rank R, the K0 value of each layer may be based on the reported capability value of the UE. For example, if the value of K0 for each level for ranks less than or equal to 2 is K0 ₁ , then the value of K0 for each level for ranks greater than 2 can be set to K0= floor

where R_UE is the maximum rank value given by the UE's capability value.

, где R_UE - максимальное значение ранга, заданное значением способности UE.[0057] In some embodiments, for a rank R CSI report, the number of spatial basis vectors for each layer may be based on the reported capability value of the UE. For example, if the number of spatial basis vectors per level for ranks less than or equal to 2 is L0, then the number of spatial basis vectors per level for ranks greater than 2 can be set to L= floor

where R_UE is the maximum rank value given by the UE's capability value.

[0058] FIG. 4 shows a flowchart showing a method for compressing coefficients of precoding vectors where M is set to 2L. That is, all 2L coefficients (eg, coefficients 416, 426, and 436) for each FD unit (a total of S*2L coefficients across all S FD units) are compressed to M*2L coefficients (for each rank) by coefficient compression module 450. This compression can be performed by decomposing the coefficients of the spatial basis vectors (eg, coefficients 452) into a linear combination of coefficients of FD basis vectors (eg, coefficients 456) and FD basis vectors (eg, vectors 454).

[0059] Some illustrative embodiments can be described using the following paragraphs.

[0060] Clause 1. The wireless device (eg, UE) generates channel state information (CSI) based at least in part on a first set of L basis vectors (eg, spatial basis vectors), a second set of M basis vectors (eg, , FD basis vectors) and a set of coefficients (e.g., coefficients for FD basis vectors), where the first basis vector set, the second basis vector set, and the coefficient set indicate information regarding the precoding vector (i.e., basis vectors and coefficients are selected to provide a wireless node, such as a base station, with channel-related information, allowing the wireless node to use this information in precoding downlink data streams). In some embodiments, a wireless device may determine a first set of basis vectors (e.g., spatial basis vectors) and/or a second set of basis vectors (e.g., FD basis vectors) by selecting basis vectors from a set of basis vectors provided in one or more codebases. books. Thus, the wireless can determine which basis vectors from a predetermined set of basis vectors in the codebook more closely match the basis vector needed to accurately estimate the wireless channel. When the wireless device is a UE, it may generate a CSI report based on the CSI, and generate a transmission waveform based on the CSI report and transmit it to the base station (for example, on a physical uplink control channel (PUCCH) or a physical uplink shared channel (PUSH)). Additional details are given in section 1.0 above.

[0061] Item 2. For each level r in the set of R levels, select from the set of coefficients a subset of non-zero coefficients in an amount not exceeding K0 and generate CSI at least partially based on the first set of basis vectors, the second set of basis vectors and the subset of non-zero coefficients for each of the levels r in the set of R levels. Additional details are given in section 1.0 above.

[0062] Item 3. A first part of the CSI (or the first part of a CSI feedback report) is generated, where the first part includes, for example, a channel quality indicator (CQI), a rank indicator (RI), and an indication of the number of non-zero coefficients over R layers (R is the rank value).

[0063] Item 4. A second CSI part (or CSI feedback report) is generated, where the second part includes, for example, a precoding matrix indicator (PMI). Additional details are given in section 2.0 above.

[0064] Item 5. The PMI includes specifying a first set of L basis vectors, specifying a second set of M basis vectors, specifying the amplitude and phase of the coefficient set, and indicating locations of non-zero coefficients from the coefficient set (e.g., a set of L*M coefficients at each level r of R). Additional details are given in section 2.0 above.

[0065] Clause 6. One or more CSI parameters in the first CSI part (or CSI feedback report) are jointly channel-coded, one or more CSI parameters in the second part are also jointly channel-coded, and the first part is channel-coded independently of the second part ( i.e., the CSI parameters in each part are jointly channel-coded, while the different parts are channel-coded independently). Additional details are given in section 2.0 above.

. Дополнительные детали приведены выше в разделе 2.0.[0066] Item 7. Restrict at least one of L, M, and K0 such that when the rank of R is greater than the rank of R0, the value of K0 for rank R is less than or equal to the value of K0 for rank R0 (R

. Additional details are given in section 2.0 above.

[0067] Item 8. Restrict at least one of L, M, and K0 such that when the rank of R is greater than the rank of R0, the value of K0 for rank R is a predetermined fraction of the value of K0 for rank R0, where R0 is greater than or is equal to one. Additional details are given in section 2.0 above.

1, суммарное количество коэффициентов в подмножестве всех R уровней меньше или равно суммарному количеству коэффициентов в подмножестве всех R0 уровней в CSI ранга R0. Битовая ширина указания количества ненулевых коэффициентов зависит от значения L/M/K0, когда ранг равен R0, а также значения R0. В результате, хотя указание количества ненулевых коэффициентов означает количество ненулевых коэффициентов среди максимального количества уровней, битовая ширина этого указания может зависеть от чуть меньшего количества значения ранга, например, R0. Дополнительно, в некоторых вариантах осуществления, K0 зависит от значения L и M (например, K0=p*L*M). Поэтому в некоторых вариантах осуществления K0 может ограничиваться ограничением L, M или другими параметрами, используемыми для определения K0 (например, p). K0 обозначает верхнюю границу количества ненулевых коэффициентов. Дополнительные детали приведены выше в разделах 2.0 и 3.0.[0068] Item 9. When R>R0

1, the total number of coefficients in a subset of all R levels is less than or equal to the total number of coefficients in a subset of all R0 levels in a CSI of rank R0. The bit width of the indication of the number of non-zero coefficients depends on the value of L/M/K0 when the rank is R0, as well as the value of R0. As a result, although the indication of the number of non-zero coefficients means the number of non-zero coefficients among the maximum number of levels, the bit width of this indication may depend on a slightly smaller number of rank values, such as R0. Additionally, in some embodiments, K0 depends on the value of L and M (eg, K0=p*L*M). Therefore, in some embodiments, K0 may be limited by the constraint L, M, or other parameters used to determine K0 (eg, p). K0 denotes an upper bound on the number of non-zero coefficients. Additional details are given in sections 2.0 and 3.0 above.

[0069] Clause 10. At least one of L, M, and K0 may be limited based on the maximum allowable number of layers, where the maximum allowable number of layers is determined by, for example, a network configuration parameter, wireless device capability, the maximum rank that the codebook can support , used to determine the precoding vector, etc. Additional details are given in section 3.0 above.

[0070] Clause 11: The wireless device may be a UE or a BS. Thus, in some embodiments, the UE or BS may generate the CSI according to any of the previous paragraphs. Additionally or alternatively, the UE may generate a CSI report and send it to the wireless node (eg, BS). When the BS receives a CSI feedback report from a UE, a CSI feedback report will be generated according to one or more of the previous clauses. Thus, the BS receives a first set of L basis vectors, a second set of M basis vectors, and a set of coefficients. Additionally, the CSI feedback report includes a first part and a second part as described above. In some embodiments, the BS may only receive an indication of the basis vectors and coefficients without receiving the actual vectors (e.g., the BS may receive an indication of the codebook index associated with the respective basis vectors, or look up the actual coefficients from a lookup table using the received coefficient indication). Additionally, the BS need not always receive the second part of the CSI report (or need not receive the maximum number of bits that can be associated with the second part). On the contrary, the BS may allocate a variable amount of resources for receiving the second part based on the payload received in the first part of the CSI report, which further reduces the overhead of the CSI message. The BS may (but is not required to) use the received information to generate a precoding vector for precoding the downlink data streams and generate a downlink transmission waveform based on the precoded data.

[0071] A wireless device (eg, UE) or a wireless node (eg, base station) may include a processor configured to perform the method mentioned in any one or more of the previous paragraphs. Additionally, the UE or base station may include a computer program product comprising a computer-readable program medium that stores processor-executable instructions that, when executed by the processor, cause the processor to perform the method mentioned in any one or more of the preceding claims.

[0072] FIG. 5 shows a block diagram representation of a portion of a device, in accordance with some embodiments of the technology disclosed herein. Device 505, such as a base station or wireless device (or UE), may include processor electronics 510, such as a microprocessor, that implements one or more of the techniques presented herein. Device 505 may include transceiver electronics 515 for sending and/or receiving wireless signals over one or more communication interfaces, such as antennas 520 and 522. Device 505 may include other communication interfaces for transmitting and receiving data. Device 505 may include one or more memory units (not explicitly shown) configured to store information, such as data and/or instructions. In some implementations, processor electronics 510 may include at least a portion of transceiver electronics 515. In some embodiments, at least some of the disclosed methods, modules, or functions are implemented using device 505.

[0073] The description of the invention, taken in conjunction with the drawings, is intended to be illustrative only, where illustrative means an example and, unless otherwise indicated, does not suggest an ideal or preferred embodiment. The word "or" as used herein is intended to include "and/or" unless the context clearly implies otherwise.

[0074] Some of the embodiments provided herein are described in the general context of methods or processes that may be performed in one embodiment by a computer program product embodied in a computer-readable medium including computer-executable instructions, such as program code executable by computers in linked over network environments. Computer readable media may include removable and non-removable storage devices including, but not limited to, read only memory (ROM), random access memory (RAM), compact discs (CDs), digital versatile disks (DVDs), etc. d. Therefore, computer-readable media may include nontransitory storage media. In general, program modules may include procedures, programs, objects, components, data structures, etc. that perform specific tasks or implement certain abstract data types. Computer or processor-executable instructions, associated data structures, and program modules provide examples of program code for performing the steps of the methods disclosed herein. The specific sequence of such executable instructions or associated data structures expresses examples of appropriate actions to perform the functions described in such steps or processes.

[0075] Some of the disclosed embodiments may be implemented as devices or modules using hardware circuitry, software, or combinations thereof. For example, a hardware circuit implementation may include discrete analog and/or digital components combined, for example, on a printed circuit board. Alternatively or additionally, the disclosed components or modules may be implemented as an application specific integrated circuit (ASIC) and/or as a field programmable gate array (FPGA). Some implementations may additionally or alternatively include a digital signal processor (DSP), which is a specialized microprocessor whose architecture is optimized for the digital signal processing operational needs associated with the disclosed functionality of this application. Likewise, various components or subcomponents within each module may be implemented in software, hardware, or firmware. The ability to communicate between modules and/or components of modules can be provided using any communication methods and media that are known in the art, including, but not limited to, communication over the Internet, wired or wireless networks, where appropriate protocols are used.

[0076] Although this document contains many specific details, they should not be construed as limiting the scope of the invention that is or may be claimed, but rather, as descriptions of features specific to particular embodiments. Some of the features described in this document in the context of separate embodiments may also be implemented together in a single embodiment. In contrast, the various features described in the context of a single embodiment may also be implemented in multiple embodiments individually or in any suitable subcombination. In addition, although the features may be described above as operating in certain combinations and even originally claimed as such, one or more features from the claimed combination may in some cases be excluded from the combination, and the claimed combination may refer to a subcombination or a variation of the subcombination. Similarly, although the operations are shown in the drawings in a specific order, this does not mean that such operations must be performed in the specific order shown or in sequential order, or that all the illustrated operations must be performed in order to achieve the desired results.

[0077] Only a few implementations and examples are described here, and other implementations, improvements, and changes can be suggested based on what is described and illustrated in this application.

Claims (31)

1. A wireless communication method, comprising the steps of: generate, by the wireless device, a Channel State Information (CSI) feedback report, wherein the first part of the CSI feedback report includes a channel quality indicator (CQI), a rank indicator (RI), and an indication of the number of non-zero coefficients across multiple layers, wherein the second part of the CSI feedback report includes a precoding matrix index (PMI) determined based on a first set of L basis vectors, a second set of M basis vectors, and a number of non-zero coefficients for each layer of the layer set, wherein the number of non-zero coefficients for each level is less than or equal to K0 non-zero coefficients, where K0 is determined based on at least one of L and M, where L, M and K0 are integers; wherein the bit width of indicating the number of non-zero coefficients across the plurality of levels is based on the maximum number of levels and K0; and transmitting, by the wireless device, a CSI feedback report to the wireless node. 2. The method of claim 1, wherein at least one of L, M, and K0 is determined by a network configuration parameter. 3. The method according to any one of paragraphs. 1 or 2, in which the value of K0 for the rank level R is equal to the value of K0 for the rank level R0, where R > R0 and where R0 is greater than or equal to one. 4. A wireless communication method, comprising the step of: receiving, by the wireless node, a Channel State Information (CSI) feedback report from the wireless device, wherein the first part of the CSI feedback report includes a channel quality indicator (CQI), a rank indicator (RI), and an indication of the number of non-zero coefficients across multiple layers, wherein the second part of the CSI feedback report includes a precoding matrix index (PMI) determined based on a first set of L basis vectors, a second set of M basis vectors, and a number of non-zero coefficients for each layer of the layer set, wherein the number of non-zero coefficients for each level is less than or equal to K0 non-zero coefficients, where K0 is determined based on at least one of L and M, and where L, M and K0 are integers; and wherein the bit width of indicating the number of non-zero coefficients over the plurality of levels is based on the maximum number of levels and K0. 5. The method of claim 4, wherein at least one of L, M, and K0 is determined by a network configuration parameter. 6. The method according to any one of paragraphs. 4 or 5, in which the value of K0 for the rank level R is equal to the value of K0 for the rank level R0, where R > R0 and where R0 is greater than or equal to one. 7. A wireless communication device containing a processor configured to: generate a link state information (CSI) feedback report, wherein the first part of the CSI feedback report includes a channel quality indicator (CQI), a rank indicator (RI), and an indication of the number of non-zero coefficients across multiple layers, wherein the second part of the CSI feedback report includes a precoding matrix index (PMI) determined based on a first set of L basis vectors, a second set of M basis vectors, and a number of non-zero coefficients for each layer of the layer set, wherein the number of non-zero coefficients for each level is less than or equal to K0 non-zero coefficients, where K0 is determined based on at least one of L and M, where L, M and K0 are integers; wherein the bit width of indicating the number of non-zero coefficients across the plurality of levels is based on the maximum number of levels and K0; and transmit a CSI feedback report to the wireless node. 8. The apparatus of claim 7, wherein at least one of L, M, and K0 is determined by a network configuration parameter. 9. The apparatus of claim 7, wherein the value of K0 for the rank level R is equal to the value of K0 for the rank level R0, where R > R0 and where R0 is greater than or equal to one. 10. A wireless communication device containing a processor configured to: receive a Channel State Information (CSI) feedback report from the wireless device, wherein the first part of the CSI feedback report includes a channel quality indicator (CQI), a rank indicator (RI), and an indication of the number of non-zero coefficients across multiple layers, wherein the second part of the CSI feedback report includes a precoding matrix index (PMI) determined based on a first set of L basis vectors, a second set of M basis vectors, and a number of non-zero coefficients for each layer of the layer set, wherein the number of non-zero coefficients for each level is less than or equal to K0 non-zero coefficients, where K0 is determined based on at least one of L and M, and where L, M and K0 are integers. 11. The apparatus of claim 10, wherein at least one of L, M, and K0 is determined by a network configuration parameter. 12. The apparatus of claim 10, wherein the value of K0 for the rank level R is equal to the value of K0 for the rank level R0, where R > R0 and where R0 is greater than or equal to one. 13. A computer-readable program medium that stores processor-executable instructions that, when executed by the processor, cause the processor to carry out the method of any one of claims. 1, 2 or 3. 14. A computer-readable program medium that stores processor-executable instructions that, when executed by the processor, cause the processor to carry out the method of any one of claims. 4, 5 or 6.

Images