WO2020192790A1 - Système et procédé de rétroaction et de signalement de csi réduits utilisant des tenseurs et une décomposition de tenseur - Google Patents

Système et procédé de rétroaction et de signalement de csi réduits utilisant des tenseurs et une décomposition de tenseur Download PDF

Info

Publication number
WO2020192790A1
WO2020192790A1 PCT/CN2020/082096 CN2020082096W WO2020192790A1 WO 2020192790 A1 WO2020192790 A1 WO 2020192790A1 CN 2020082096 W CN2020082096 W CN 2020082096W WO 2020192790 A1 WO2020192790 A1 WO 2020192790A1
Authority
WO
WIPO (PCT)
Prior art keywords
vectors
channel
csi
recited
tensor
Prior art date
Application number
PCT/CN2020/082096
Other languages
English (en)
Inventor
Diana MAAMARI
Jialing Liu
Qian CHENG
Weimin Xiao
Original Assignee
Huawei Technologies Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co., Ltd. filed Critical Huawei Technologies Co., Ltd.
Publication of WO2020192790A1 publication Critical patent/WO2020192790A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H04L25/0242—Channel estimation channel estimation algorithms using matrix methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0417—Feedback systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0658—Feedback reduction
    • H04B7/0663—Feedback reduction using vector or matrix manipulations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/0224—Channel estimation using sounding signals

Definitions

  • the present invention relates generally to wireless communications, and, in particular embodiments, to a system and method for reduced channel state information (CSI) feedback and reporting using tensors and tensor decomposition.
  • CSI channel state information
  • Wireless communication systems typically include base stations communicating with user equipments (UEs) .
  • a communication channel from a base station to a UE is generally referred to as a downlink channel, and a transmission from the base station to the UE is a downlink transmission.
  • a communication channel from a UE to a base station is generally referred to an uplink channel, and a transmission from the UE to the base station is an uplink transmission.
  • Channel estimation may be performed to improve communications between base stations and UEs.
  • TDD time division duplex
  • a base station can exploit the channel reciprocity and measure the uplink channel to infer the downlink channel.
  • a frequency division duplex (FDD) system separation between uplink transmissions and downlink transmissions is in frequency, and feedback may be required from receivers (UEs) to convey information regarding the downlink channel.
  • the feedback would be substantial when the users are highly mobile and the communications system has a massive number of antennas. Therefore, knowledge of channel station information (CSI) at the base station with low feedback overhead is becoming increasingly important in wireless communication systems for achieving higher spectral efficiency.
  • CSI channel station information
  • a method includes: receiving, through M antenna ports of a user equipment (UE) , a reference signal (RS) resource set, the RS resource set being transmitted from N antenna ports of a base station (BS) over K1 subcarriers on L1 OFDM symbols, wherein each of the M, N, K1 and L1 is an integer greater than 0, and reporting, by the UE to the BS, R sets of vectors that are obtained based on measurement of the RS resource set, wherein each set of vectors comprises one or more vectors, the one or more vectors comprising a first Nx1 vector, a second Mx1 vector, a third K2x1 vector, a fourth L2x1 vector, or any combination thereof, wherein a sum of R outer products of the respective R sets of vectors represents channel state information of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on L2 OFDM symbols, and wherein each of the R, K2
  • the method further includes receiving, by the UE, resources carrying the RS resource set.
  • the R sets of vectors are reported according to a configuration signaled by the BS.
  • the method further includes: constructing, by the UE, a multi-dimensional tensor representing the channel based on the measurement of the RS resource set, and performing, by the UE, decomposition on the multi-dimensional tensor using a tensor decomposition technique, whereby generating factor matrices, wherein the R sets of vectors are obtained from the factor matrices.
  • the multi-dimensional tensor is constructed based on the measurement of the RS resource set in a set of measurement domains, the set of measurement domains comprising a transmit domain, a receive domain, a time domain, a frequency domain, a delay domain, or any combination thereof.
  • the number of measurement domains in the set of measurement domains is configured by the BS.
  • the set of measurement domains is configured by the BS.
  • a factor matrix comprises a plurality of Nx1 vectors, a plurality of Mx1 vectors, a plurality of K2x1 vectors, or a plurality of L2x1 vector.
  • the factor matrices are generated from Canonical Polyadic decomposition (CPD) of the multi-dimensional tensor.
  • CPD Canonical Polyadic decomposition
  • the R sets of vectors are a subset of a plurality of Nx1 vectors, a plurality of Mx1 vectors, a plurality of K2x1 vectors, and a plurality of L2x1 vectors that make up the factor matrices.
  • information about R is configured by the BS.
  • the method further includes receiving, by the UE, signaling indicating the one or more vectors to be reported.
  • resources for transmitting the RS resource set are non-zero-power CSI-RS resources that are configured by the BS for CSI reporting.
  • the R sets of vectors that are reported comprises quantized amplitudes and phases of elements of vectors.
  • the method further includes receiving, by the UE, configuration information for reporting the R sets of vectors in higher layer signaling, or a downlink control information (DCI) message.
  • DCI downlink control information
  • the configuration information comprises information triggering reporting of the R sets of vectors.
  • a method includes: transmitting, through N antenna ports of a base station (BS) to M antenna ports of a user equipment (UE) , a reference signal (RS) resource set, the RS resource set being transmitted over K1 subcarriers on L1 OFDM symbols, wherein each of the M, N, K1 and L1 is an integer greater than 0, receiving, by the BS from the UE, a channel state information (CSI) report measured based on the RS resource set, the CSI report comprising information of R sets of vectors, wherein each set of vectors comprises one or more vectors, the one or more vectors comprising a first Nx1 vector, a second Mx1 vector, a third K2x1 vector, a fourth L2x1 vector, or any combination thereof, and wherein a sum of R outer products of the respective R sets of vectors represents CSI of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on
  • CSI channel state information
  • the method further includes transmitting, by the BS to the UE, resources carrying the RS resource set.
  • the method further includes transmitting, by the BS to the UE, information about R.
  • the method further includes transmitting, by the BS to the UE, a set of measurement domains for the UE to measure the RS resource set in the set of measurement domains, the set of measurement domains comprising a transmit domain, a receive domain, a time domain, a frequency domain, a delay domain, or any combination thereof.
  • the method further includes transmitting, by the BS to the UE, an order of the set of measurement domains for the UE to measure the RS resource set in the order of the set of measurement domains.
  • the method further includes transmitting, by the BS to the UE, a number of measurement domains in the set of measurement domains.
  • the method further includes configuring, by the BS for the UE, one or more measurement domains comprised in the set of measurement domains.
  • the method further includes transmitting, by the BS to the UE, signaling indicating the one or more vectors to be reported.
  • resources for transmitting the RS resource set are non-zero-power CSI-RS resources that are configured by the BS for CSI reporting.
  • the R sets of vectors that are reported comprises quantized amplitudes and phases of elements of vectors.
  • the method further includes transmitting, by the BS to the UE, a trigger triggering reporting of the CSI of the channel.
  • the method further includes generating, by the BS to the UE, a transmission filter based on the CSI report.
  • a UE feeds back R sets of vectors to a BS, where a sum of R outer products of the respective R sets of vectors represents channel state information of a channel. Feeding back the R sets of vectors by the UE involves less CSI feedback overhead compared with feeding back all information about the channel, while the R sets of vectors that are fed back can be used by the BS to reconstruct the channel, which may further be used for performing downlink transmission over the channel.
  • the forgoing aspects have an advantage of providing improved downlink beamforming, e.g., in single user (SU) -MIMO systems and multi-user (MU) -MIMO systems, with less CSI feedback overhead.
  • FIG. 1 illustrates a diagram of an embodiment wireless network
  • FIG. 2 illustrates a diagram of an embodiment multi-input multi-output (MIMO) communications system
  • FIG. 3 illustrates a diagram showing a communication channel between a transmitter and a receiver modeled using multipath components (MPCs) arriving in clusters;
  • MPCs multipath components
  • FIG. 4 illustrates a graph showing single user (SU) -MIMO capacity curves of a MIMO system
  • FIG. 5 illustrates a diagram of an embodiment method for channel state information (CSI) feedback in a MIMO communications system
  • FIG. 6 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 7 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 8 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 9 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 10 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 11 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 12 illustrates a flowchart of another embodiment method for CSI feedback in a MIMO communications system
  • FIG. 13 illustrates a flowchart of an embodiment method for CSI feedback in a communications system
  • FIG. 14 illustrates a flowchart of another embodiment method for CSI feedback in a communications system
  • FIG. 15 illustrates a diagram of an embodiment processing system
  • FIG. 16 illustrates a diagram of an embodiment transceiver.
  • a base station transmits reference signals to user equipment (UEs) for measuring channels between the BS and the UEs.
  • UEs user equipment
  • a UE measures a channel based on the reference signals, and feeds back channel state information (CSI) of the channel to the BS. If all information of the estimated channel can be fed back to the BS, the BS may obtain a more complete understanding of the channel condition and quality, and consequently perform generally optimal downlink transmissions.
  • CSI feedback overhead e.g., transmission resources
  • MIMO multi-input and multi-output
  • Embodiments of the present disclosure provide methods for feeding back CSI by UEs based on tensors and tensor decomposition techniques.
  • a MIMO channel or a MIMO channel covariance may be modelled as a multi-dimensional tensor, which is decomposed using a tensor decomposition technique, such as Canonical Polyadic decomposition.
  • the decomposition results in factor matrices. Because the factor matrices represent CSI of the MIMO channel or the MIMO channel covariance, information about all or partial of the factor matrices may be fed back to a BS, and the BS may reconstruct the MIMO channel based on the feedback, and perform downlink transmissions accordingly.
  • the overhead for feeding back the factor matrices depends on the size of each factor matrix, which is much less compared with feeding back all measurement information of the MIMO channel.
  • the embodiments reduce CSI feedback overhead, while still providing sufficient amount of information for a more accurate estimation of the MINO channel.
  • a UE may receive, through M antenna ports, a reference signal (RS) resource set transmitted from N antenna ports of a BS over K1 subcarriers on L1 OFDM symbols, and report, to the BS, R sets of vectors that are obtained based on measurement of the RS resource set.
  • Each set of vectors may include one or more vectors selected from a Nx1 vector, a Mx1 vector, a K2x1 vector, and a L2x1 vector, and the sum of R outer products of the respective R sets of vectors represents channel state information of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on L2 OFDM symbols.
  • the BS may reconstruct the channel by calculating an outer product of each set of vectors and summing up outer products of the respective R sets of vector. Details of the embodiments will be provided in the following description.
  • FIG. 1 illustrates a network 100 for communicating data.
  • the network 100 comprises a base station 110 having a coverage area 101, a plurality of mobile devices 120, and a backhaul network 130.
  • the base station 110 establishes uplink (dashed line) and/or downlink (dotted line) connections with the mobile devices 120, which serve to carry data from the mobile devices 120 to the base station 110, and vice-versa.
  • Data carried over the uplink/downlink connections may include data communicated between the mobile devices 120, as well as data communicated to/from a remote-end (not shown) by way of the backhaul network 130.
  • the term “base station” refers to any component (or collection of components) configured to provide wireless access to a network, such as, an access node, a Node B, an evolved Node B (eNB) , a next generation (NG) Node B (gNB) , a master eNB (MeNB) , a secondary eNB (SeNB) , a master gNB (MgNB) , a secondary gNB (SgNB) , a network controller, a control node, an access point, a transmission point (TP) , a transmission-reception point (TRP) , a cell, a carrier, a macro cell, a femto cell, a pico cell, a relay, a customer premises equipment (CPE) , a Wi-Fi access point (AP) , or other wirelessly enabled devices.
  • TP transmission point
  • TRP transmission-reception point
  • Base stations may provide wireless access in accordance with one or more wireless communication protocols, e.g., long term evolution (LTE) , LTE advanced (LTE-A) , 3GPP Rel. 15 and subsequent release, high speed packet access (HSPA) , Wi-Fi 802.11a/b/g/n/ac, ax and other 802.11xx standards.
  • LTE long term evolution
  • LTE-A LTE advanced
  • HPP Rel. 15 3GPP Rel. 15
  • High speed packet access HSPA
  • Wi-Fi 802.11a/b/g/n/ac Wi-Fi 802.11a/b/g/n/ac
  • ax 802.11xx standards.
  • the term “mobile device” refers to any component (or collection of components) capable of establishing a wireless connection with a base station, such as a user equipment (UE) , a mobile station (STA) , a mobile, a terminal, a user, a subscriber, a stations, a communication device, a CPE, a relay, an integrated access and backhaul (IAB) relay, and other wirelessly enabled devices.
  • a base station such as a user equipment (UE) , a mobile station (STA) , a mobile, a terminal, a user, a subscriber, a stations, a communication device, a CPE, a relay, an integrated access and backhaul (IAB) relay, and other wirelessly enabled devices.
  • the network 100 may comprise various other wireless devices, such as relays, low power nodes, In the device to device communication mode, such as the proximity services operating mode, direct communication between UEs is possible.
  • the base station 110 may utilize channel state information (CSI) to adapt downlink transmission to a downlink channel between the base station 110 and a UE 120.
  • CSI channel state information
  • the CSI may be estimated by the UE 120 and fed back to the base station 110.
  • MIMO multi-input multi-output
  • the design of a multi-input multi-output (MIMO) communication system depends on, at least in part, the accuracy of the knowledge of CSI. The theoretically best performance in terms of spectral efficiency may be achieved when perfect CSI is available at both communication nodes. However, imperfect CSI may arise due to channel estimation errors, quantization errors, and/or outdated channel estimates.
  • the base station may take advantage of channel reciprocity to obtain downlink channel (also called forward channel) estimation from pilots received over the uplink channel (also called reverse channel) .
  • downlink channel also called forward channel
  • pilots received over the uplink channel also called reverse channel
  • CSI may not need to be estimated and fed back by UEs to the base station.
  • measurement at the base station may not capture downlink interferences from co-scheduled UEs or from neighboring cells, and the downlink channel estimation may not accurately reflect the downlink channel conditions.
  • UEs may still need to feedback estimated CSI to the base station even for a TDD system.
  • FDD frequency division duplex
  • a channel between two devices is estimated by having a first device transmit a known signal on a known time or frequency resource (s) to a second device, the received signal at the second device is expressible as
  • y is the received signal at the second device
  • s is the known signal (which may be a reference signal, a pilot, or a pilot signal)
  • H is the channel model or channel impulse response (CIR)
  • n is the noise (and interference for some communication channels) .
  • a base station may transmit reference signals s to a UE, and the UE may estimate the channel H, and feed back CSI information based on the estimated H to the base station.
  • antenna element to element channel estimation and feedback may be expensive in terms of computational resources, as well as communications overhead.
  • antenna, antenna element, and antenna port may be generally interchangeable, but in some specific scenarios, they can mean different but related subjects.
  • one transmit (Tx) antenna port may be formed (or virtualized) by multiple antenna elements or antennas, and the receiver sees only the one Tx antenna port instead of each of the multiple antenna elements or antennas.
  • the virtualization may be achieved via beamforming, for example.
  • the received signal at the second device may be expressed as
  • y1 is the received signal at the second device
  • v is a transmit filter (or precoder)
  • s is the known signal
  • H is the channel model or response
  • n is the noise.
  • the signal s is precoded by v and transmitted.
  • the transmit filter v of dimension N tx x Ns enables the transmitter to precode or beamform the transmitted signal, where Ns is the number of layers, streams, symbols, pilots, messages, or known sequences transmitted.
  • the precoder i.e., the transmit filter v
  • the precoder can be a digital precoder, analog precoder, or hybrid digital-analog precoder, in which case Ns is replaced by N RF representing the number of RF chains and is either equal to or smaller than the number of transmit antennas or transmit antenna ports
  • a MIMO orthogonal frequency division multiplexing (OFDM) system provides an additional mode (or a dimension, or a domain) to the N Rx ⁇ N Tx matrix of a MIMO system that may lead to a multi-dimensional (e.g., a tensor, a multi-mode, a multi-domain, an array with multiple dimensions) .
  • the channel mode may be represented by
  • the third additional dimension represents a subcarrier dimension (e.g., subbands, tones, resource blocks, frequencies) of an OFDM system.
  • the number of subcarriers, tones, subbands, or resource blocks is represented by N K .
  • CSI feedback includes feedback of a channel quality indicator (CQI) , a precoding matrix indicator (PMI) , a rank indicator (RI) and a precoding type indication (PTI) .
  • CQI channel quality indicator
  • PMI precoding matrix indicator
  • RI rank indicator
  • PTI precoding type indication
  • the transmitter uses the CQI to select one of 15 modulation alphabet and code rate combinations for transmission.
  • the RI informs the transmitter about the number of useful transmission layers for the current MIMO channel.
  • the PMI signals the codebook index of the preferred digital precoding matrix.
  • CSI-RS is the channel state information reference signal and is used by a UE to estimate and report the CQI to a base station.
  • CQI transmission There are two types of CQI transmission: Periodic and Aperiodic.
  • Periodic CQI is transmitted periodically with a certain interval specified by a higher layer message, e.g., radio resource control (RRC)
  • RRC radio resource control
  • Aperiodic CQI is transmitted by a special trigger (e.g.,: DCI, or a RACH Response) .
  • RRC radio resource control
  • DCI or a RACH Response
  • the CSI feedback parameters are transmitted on the physical uplink control channel (PUCCH) or physical uplink shared channel (PUSCH) depending on the CSI Report Type (Periodic or Aperiodic) , and used for closed loop link adaptation. Both wideband and subband CQI are supported.
  • Codebook based feedback is adopted by 3GPP.
  • Other methods of CSI feedback include feedback of the MIMO matrix for each tone (e.g., each subband, or each subcarrier) , a group of tones (or group of subbands) , or feedback of a wideband MIMO matrix, after channel estimation is performed by the UE.
  • Another method includes having the UE perform singular value decomposition (SVD) on the matrix, decompose the channel into parallel single output single output (SISO) subchannel with non-equal gains, and feedback the right singular matrix or a subset of the right singular matrix. Efficient quantization/encoding of the unitary matrix may be adopted.
  • SSD singular value decomposition
  • the base station may be able to have a more complete understanding of the channel condition and quality, and thus perform generally optimal downlink transmission.
  • the base station may perform Eigen based beamforming, or SVD based beamforming, or compute the optimal downlink multi-user MIMO precoder with interference cancellation.
  • this type of feedback may incur a large overhead (e.g., transmission resources) for feeding back the CSI, especially in cases of multi-input and multi-output (MIMO) systems or massive MIMO systems, where the CSI scales with the large number of transmit antennas.
  • MIMO multi-input and multi-output
  • precoding is codebook based
  • the UE may feedback a codebook index of a codebook that maximizes a certain criteria such as throughput, or spectral efficiency.
  • Embodiments of the present disclosure provide methods for feeding back CSI by UEs based on tensors and tensor decomposition techniques.
  • the MIMO channel or the MIMO channel covariance may be modelled as a tensor, and tensor decomposition techniques may be applied to the channel or the channel covariance tensor.
  • the decomposition results in factor matrices.
  • the factor matrices (or a subset of the factor matrices depending on configurations from the network) may be fed back to a base station. That is, instead of feeding back the multi-dimensional downlink MIMO channel, the constituent factor matrices obtained from the tensor decomposition are fed back to the network.
  • the network may reduce the feedback overhead by requiring feedback of codeword (s) indices from codebook (basis) .
  • the feedback may be over a subset of the domains, instead of all configured domains.
  • the embodiments are beneficial for providing improved downlink beamforming in single user (SU) -MIMO systems and multi-user (MU) -MIMO systems with less feedback overhead compared with conventional schemes, such as the conventional scheme where a UE feedbacks the N Rx ⁇ N Tx ⁇ N K downlink MIMO channel, or the conventional scheme where a UE feedbacks the right singular matrix N Tx ⁇ N Tx for each of the N K subcarriers (tones, subbands, frequencies, or resource blocks) .
  • the embodiments also provide an improved performance in downlink beamforming compared with codebook based beamforming, such as that adopted by 3GPP.
  • a tensor is referred to as a multi-dimensional matrix, e.g., a 3-dimensional array, or a 4-dimensional array, or a higher dimensional array, which is used to mathematically represent a downlink MIMO channel between a base station and a UE in a MIMO system, or a covariance matrix of the downlink MIMO channel.
  • CSI feedback is generally referred to transmitting or reporting channel state information to a base station or a network entity.
  • FIG. 2 illustrates a diagram of an embodiment MIMO communications system 200.
  • the MIMO communications system 200 includes a first device 210 and a second device 220.
  • the first device may be a base station (or a UE)
  • the second device may be a UE (or a base station) .
  • FIG. 2 only shows two devices for illustrative convenience.
  • the MIMO communications system 200 may have more than 2 devices for wireless communications.
  • the first device 210 includes four (5) antenna for transmitting signals, i.e., Ax1, Ax2, Ax3 and Ax4, while the second device 220 includes three (4) antenna for receiving signals, i.e., Ar1, Ar2, and Ar3.
  • the downlink channel from the first device 210 to the second device 220 may be generally referred to as a MIMO channel 230.
  • the MIMO channel 230 includes multiple paths from each of the antennas at the first device 210 to each of the antennas at the second device 220.
  • the MIMO channel 230 with four inputs and three outputs may be modeled as a 4x3 matrix
  • the MIMO channel between the first device 210 and the second device 220 may be represented by the following N Tx x N Rx matrix
  • the N Tx x N Rx matrix may be referred to as a MIMO channel matrix.
  • the first device 210 and the second device 220 communicate using multiple subbands (or subcarriers or tones of an orthogonal frequency division multiplexing (OFDM) system) , e.g., N K subbands, for each subband, there may be a corresponding MIMO channel matrix estimated and formed.
  • N K MIMO channel matrices there may be N K MIMO channel matrices.
  • the N K MIMO channel matrices form a 3-dimensional tensor for the MIMO channel. The three dimensions include a transmit domain, a receive domain and a subband domain (or frequency domain) .
  • the MIMO channel is estimated multiple times during a measurement time window, for each measurement time point, there may be a MIMO channel matrix estimated and formed.
  • N T MIMO channel matrices may be obtained.
  • the N T MIMO channel matrices form a 3-dimensional tensor for the MIMO channel.
  • the three dimensions include a transmit domain, a receive domain and a time domain.
  • (N K x N T ) MIMO channel matrices may be obtained, and form a 4-dimensional tensor.
  • the four dimensions include transmit domain, a receive domain, a frequency domain and a time domain.
  • the downlink MIMO channel from the first device 210 to the second device 220 may be represented by a multi-dimensional tensor constructed based on channel measurements or estimation.
  • Other domains (or dimensions) that may be used to construct the multi-dimensional tensor may include multiple base stations, distributed antennas, a uniform planar array, or a base station with multiple carriers.
  • Those of ordinary skill in the art would recognize that various tensors using different domains may be formed to represent a MIMO channel without departing from the principle of the present disclosure.
  • the terms of “domain” , “dimension” and “mode” may be used interchangeably in the disclosure in connection with forming a multi-dimensional tensor.
  • the domains may also be referred to as measurement domains in which reference signals are measured and used to form a multi-dimensional tensor.
  • Nth order tensor The order of the tensor refers to the number of dimensions.
  • a 3-dimensional tensor (formed using three different domains) representing a downlink MIMO channel may be represented as
  • the first dimension refers to the receive antenna dimension
  • the second dimension refers to the transmit antenna dimension
  • the third dimension refers to the subband (subcarrier, tone, frequency, resource block) dimension.
  • a tensor may be formed using four domains (or dimensions, modes) including the transmit domain, receive domain, frequency domain and a time domain.
  • the 4-dimensional tensor may be represented as
  • N T represents the number of channel measurements in the time domain which may be related to the Doppler of a moving UE.
  • the channel between a transmitter (e.g., the first device 210) and a receiver (e.g., the second device 220) may be modeled using multipath components (MPCs) arriving in clusters, which are formed by multiple reflections from objects in the vicinity of the receiver and the transmitter.
  • MPCs multipath components
  • FIG. 3 illustrates a diagram showing this channel model graphically.
  • the frontal slabs of the tensor representing the channel may be mathematically represented as
  • u is the location vector of receive antenna element u ,
  • s is the location vector of transmit antenna element s
  • n is the angle of arrival and ⁇ AOD, n is the angle of departure for cluster n,
  • - N is the number of clusters in the MIMO channel.
  • a MIMO channel tensor such as the tensor represented by the mathematical expression (3) or (4) , may be decomposed using a tensor decomposition technique, such as the Canonical Polyadic Decomposition (CPD) , multilinear SVD (MLSVD) or the Higher Order SVD (HOSVD) , to obtain the tensor’s factor matrices.
  • a tensor decomposition technique such as the Canonical Polyadic Decomposition (CPD) , multilinear SVD (MLSVD) or the Higher Order SVD (HOSVD)
  • CPD Canonical Polyadic Decomposition
  • MLSVD multilinear SVD
  • HSVD Higher Order SVD
  • F stands for Frobenius norm
  • ⁇ represents outer product
  • a T , A R , and C are factor matrices of the tensor.
  • Each factor matrix includes L vectors.
  • L represents the CPD decomposition rank.
  • a tensor may be formulated by stacking the frontal slabs in Equation (5) .
  • a three dimensional tensor may be formulated in this way as shown below, which is expressed similarly to the expression (3) :
  • N Rx represents the number of receiver antennas
  • N Tx represents the number of transmit antennas
  • N K represents the number of subcarriers or tones.
  • This tensor may be represented in the CPD form, i.e. as the sum of rank one outer product, as provided in the following:
  • the channel in Equation (8) is a tensor of rank at most N.
  • the CPD decomposition of the channel in Equation (8) may be performed by solving the following optimization problem:
  • the optimization problem may be solved by alternating the least square (ALS) procedure.
  • the original factor matrices A, B, C and the estimated factor matrices obtained using (9) from the CPD operation may have a relationship that can be expressed as follows:
  • ⁇ 1 , ⁇ 2 , ⁇ 3 are scaling matrices, and ⁇ is permutation matrix.
  • the CPD decomposition as represented by expression (6) may be referred to as a rank L CPD decomposition (or least square approximation) .
  • the decomposition decomposes the tensor into a sum of rank-one component tensors.
  • the rank L may be referred to as a tensor decomposition rank.
  • a T , A R , and C may be referred to as CPD representation of the tensor or factor matrices of the tensor. Consequently, the MIMO channel may be represented using the CPD representation, i.e., A T , A R , and C, of the tensor of the MIMO channel.
  • the CPD representation of a channel is a low rank representation of the channel.
  • the factor matrices (or subsets of the factor matrices) of the tensor may be used to reconstruct the tensor or to approximate the tensor, i.e., the MIMO channel or covariance of the MIMO channel, and are thus useful, if fed back by UEs as CSI, for a base station to schedule and perform downlink transmission and compute downlink precoders, e.g., for SU and MU-MIMO based communications.
  • the factor matrices of the tensor may include information about channel parameters that are useful for a base station.
  • the matrix A T may include information about angles of departure
  • the matrix A R may include information about angles of arrival
  • the matrix C may include information about cluster delay and cluster gain.
  • the decomposition of the tensor results in compact representation of the channel, which may be used to represent the channel with reduced amount of information, compared with using the tensor to represent the channel.
  • CSI may be reported based on decomposition of the tensor, and accordingly, CSI feedback overhead may be reduced, while sufficient channel state information is fed back for the transmitter to estimate the downlink channel and perform downlink transmission.
  • An embodiment method may generally include a configuration step, a measurement step and a feedback step.
  • the network e.g., a base station
  • the UE measures the downlink MIMO channel (or the covariance matrix) , constructs downlink MIMO channel tensor, and decomposes the tensor into factor matrices.
  • CSI is fed back based on the decomposed tensor and CSI feedback configurations. Having received the feedback from the UE, which may include matrices or vectors, the network may estimate (and/or create, compute, reassemble) the downlink MIMO channel by taking the sum of rank 1 outer product of the feedback vectors. More details about these steps will be provided in the following.
  • the network may use the feedback information to construct (estimate, reassemble) the downlink MIMO channel, i.e., e.g., by taking the sum of rank 1 outer product of the feedback vectors, as shown below:
  • the network may use a portion or a subset of the feedback information to construct the downlink MIMO channel.
  • the configuration step may include configuring the structure of the tensor.
  • the different dimensions that constitute the tensor may be configured by the network. Dimensions may also be referred to as domains or modes.
  • the dimensions configured for constructing a tensor may also be referred to as tensor dimensions or tensor domains, or measurement domains in the following description, for the purpose of illustration convenience only.
  • the different dimensions that may be used to construct the tensor may not be limited to the transmit dimension, receive dimension, and frequency dimension, as discussed above. Other applicable dimensions, such as the time dimension, may also be used. Any permutation of the dimension is applicable. Any combination of the time, transmit, receive, frequency dimensions is also applicable in the present disclosure.
  • reference signals may be measured in one or more configured measurement domains to construct the tensor.
  • a number (quantity) of the measurement domains configured, and/or an order of the configured measurement domains in which reference signals are to be measured, may also be configured for constructing the tensor.
  • a tensor may be formed using two (tensor) domains (e.g., the transmit and frequency domains) .
  • the tensor may be referred to as a covariance tensor.
  • a tensor may be formed based on the transmit domain, the receive domain, the time domain, and a channel impulse response delay domain (i.e., path delay domain) .
  • a channel impulse response delay domain i.e., path delay domain
  • an additional dimension may be added to form the tensor to model the cooperation.
  • the configuration step may include configuring measurement resources for channel or spatial covariance measurement.
  • the measurement resources indicate resources that carry a reference signal resource set, or reference signals, e.g., CSI-RSs, for the UE to perform the measurement.
  • the measurement resources associated with signal reception may include at least non-zero power (NZP) channel state information reference signal (CSI-RS) resources configured for channel and interference measurement, or demodulation reference signal (DMRS) resources.
  • NZP non-zero power
  • CSI-RS channel state information reference signal
  • DMRS demodulation reference signal
  • the network may also configure feedback resources, which indicate resources that carry CSI that is fed back to the network.
  • the network may further configure trigger information triggering the UE to feedback CSI. For example, the network may send a DCI message to trigger dynamic CSI feedback.
  • the configuration step may include configuring the required feedback granularity in the tensor dimensions.
  • the frequency dimension may be configured to include a certain number of resource blocks, or a number of subcarriers.
  • the granularity in the frequency dimension may be configured by the network.
  • the granularity configured may not be limited to the frequency dimension, and feedback granularity may be configured for any tensor dimension.
  • a tensor may be formed using multiple dimensions such as the transmit, receive, time and frequency dimensions.
  • the network configures the tensor domains with a feedback granularity for each domain.
  • the network then transmits reference signals to enable the UE to perform channel estimation.
  • Reference signals such as CSI-RS may be used.
  • reference signals such as DMRS, or CRS may be used.
  • the reference signals are transmitted.
  • the reference signals may not be evenly spaced in the frequency (subcarrier, subband, tones) domain and/or in the time (OFDM, slot, symbol) domain.
  • the reference signals may be evenly spaced on resource block basis but not on subcarrier (tone) basis.
  • the UE may then perform some interpolation to obtain the channel estimation over resource elements that may not contain the reference signals.
  • the UE may down sample received signals such that the channel estimate may be on smaller granularity than that used for transmitting the reference signals.
  • the reference signals may be transmitted on subcarrier basis, but the feedback may be over resource block granularity.
  • the UE may interpolate over the symbols (time domain) to obtain the channel estimate in resources which may not contain any reference signals.
  • the UE may search for codewords from a predefined codebook (codebook basis) to represent the time domain. In these described embodiments, it may be viewed that the granularity over which the reference signals are transmitted may be different then the granularity over which the feedback for the different dimensions is performed. The granularity may be signaled by the network.
  • the reference signals are transmitted over K1 subcarrier and L1 symbols within a resource block.
  • the K2 subcarriers and L2 OFDM symbols are configured by the network or specified by standard specification. The subcarriers and symbols are not necessarily contiguous.
  • the K2 subcarriers and the L2 OFDM symbols are a CSI reference (according to TS 38.214, 5.2.2.1.1) associated with reporting.
  • the configuration step may include, configuring the UE with dimensions (modes, or domains) in which CSI feedback is required.
  • the CSI to be fed back in a domain includes information related to or about the domain.
  • the configured dimensions may also be referred to as feedback dimensions or domains.
  • the feedback dimensions include a subset of or all the tensor dimensions that are configured for constructing a tensor. For example, for the channel in the expression (4) , the network may require feedback over the second dimension only, i.e., the transmit domain only. In this case, the network may configure the UE with necessary signaling for feeding back CSI in this particular domain.
  • the network may use the feedback as a basis for downlink CSI-RS transmission.
  • the network may require substantial channel information for better downlink precoding or for downlink beamforming training for MU-MIMO settings.
  • the network may require feedback over all the tensor dimensions.
  • the network may configure the UE with necessary signaling for feeding back CSI in all the tensor domains that form the tensor.
  • the network may use the feedback to compute a downlink precoder, such as a zero forcing precoder. Any combination of the feedback dimensions is applicable.
  • the network may require feedback over the transmit and frequency dimensions, but not including the receive dimension.
  • the network may configure the UE with necessary signaling to feedback CSI over these two dimensions.
  • the network may configure the UE to report a number (quantity) of vectors of the factor matrix A T (transmit domain) and A R (receive domain) in the expression (6) , respectively.
  • the quantity of vectors from each factor matrix to be reported may also be configured, and details will be provided later in the description.
  • E quantity may be greater than or equal to L in the expression (6) .
  • Those of ordinary skill in the art would recognize that various combinations of feedback dimensions may be configured for reporting CSI in the configured feedback dimensions.
  • the configuration step may include configuring the rank L (may also be referred to as a CPD rank, an approximate tensor rank, an approximate low rank, a decomposition rank, a least square rank) in the optimization problem of (6) .
  • the terms of “rank L” and “CPD rank” are used interchangeably in the present disclosure.
  • the network may use the estimated uplink channel, which may be estimated from the uplink SRS, to compute and set the CPD rank, when assuming there is some degree of reciprocity between the uplink and downlink channels.
  • the network may use a previous feedback of downlink CSI to determine the CPD rank.
  • the UE may choose a CPD rank.
  • the CPD rank is related to the number of clusters in the downlink.
  • the configuration step may include configuring the number of vectors R f to be fed back out of the L vectors in each factor matrix.
  • the network may configure the UE for feeding back R f ⁇ L vectors in each of the factor matrices A T , A R , and C. The UE may choose R f vectors according to criteria specified the network.
  • the UE may determine the R f (the number of vectors) according to an optimization criteria that maximizes a utility, such as capacity, signal to noise ratio (SNR) , signal to interference to noise ratio (SINR) , throughput, or spectral efficiency.
  • the number of vectors R f may also be related to the number of layers for transmission (i.e., the transmission rank, or the rank of transmission) .
  • the UE may determine the feedback rank based on the transmission rank.
  • the number of vectors R f may be referred to as a feedback rank, or a feedback length.
  • the network may configure different feedback ranks for different factor matrices. That is, in this case, the network may configure different feedback lengths for different tensor domains, in which CSI is to be fed back to the network, because the factor matrices correspond to different tensor domains. For example, the network may configure a feedback length R f, 1 for A T , a feedback length R f,2 for A R , and a feedback length R f, 3 for C.
  • a base station may configure a UE with a greater feedback rank R f , e.g., in an initial communication period of time or for multi-user (MU) communications, and thus, the base station may receive greater amount of CSI and obtain more knowledge of the downlink MIMO channel. This may help the base station improve communication performance in the downlink MIMO channel during a succeeding communication period or for MU communications.
  • a base station may configure a smaller feedback R f for SU communications after the initial communication period.
  • the network may configure which factor matrix is to be fed back by the UE.
  • the base station may pre-assign a matrix ID to each of the factor matrices, e.g., 0, 1, 2 and 3 are assigned to the factor matrices.
  • the base station may predefine that a first matrix is related to the transmit domain, i.e., A T , a second matrix is related to the receive domain, i.e., A R , a third matrix is related to the frequency domain, i.e., C, and a fourth matrix is related to the time domain, i.e., D.
  • the base station may use 2-bit to indicate a factor matrix that is to be fed back. For example, “00” , “01” , “10” and “11” indicate the four respective factor matrices.
  • the base station does not specify how the factor matrices are generated, i.e., using what tensor decomposition techniques.
  • the base station may also configure what channel parameters are to be fed back.
  • One of ordinary skill in the art would recognize that various methods may be used for the base station to indicate or specify the CSI to be fed back to the base station.
  • the network may configure the UE to report different CSI for different ranks of MIMO transmissions.
  • the rank of transmission represents the number of layers to be transmitted
  • the CPD rank represents the number of vectors in the factor matrices obtained from the CPD decomposition.
  • the network may configure the CPD rank to be different for the two ranks of transmission. Based on the received multiple CSI reports, the network may choose a rank of transmission, and configures the UE with the chosen rank of transmission.
  • the network may configure the periodicity of the feedback of the domains. That is, the feedback over certain dimensions may be on a larger time scale, while the feedback over other dimensions may be on a smaller time scale. For example, the transmit and receive dimensions may be fed back on a longer time scale, while feedback over the frequency and time domain may be more dynamic. In this case, the time between two CSI reports for the is longer than that in the former case.
  • the feedback periodicity of CSI of a domain may be configured by higher layer signaling (e.g., RRC signaling, or RRC Connection Setup) .
  • a more dynamic configuration may be employed for a UE to aperiodically update the CSI feedback of a domain using a special trigger, e.g., triggering the UE to report updated CSI in the domain by sending a downlink control information (DCI) message.
  • a special trigger e.g., triggering the UE to report updated CSI in the domain by sending a downlink control information (DCI) message.
  • DCI downlink control information
  • the network may specify different update periods for different domains.
  • the network may configure the UE with different reporting periods for different domains.
  • the transmit domain may be reported on longer period, while the frequency domain may be reported on smaller time scale.
  • the factor matrices of the tensor may include information about channel parameters, such as angles of departure, angles of arrival, cluster delay and cluster gain. These channel parameters may be derived from the factor matrices.
  • the network may configure the UE to feedback a combination of analog feedback, in addition to tensor feedback as described with respect to the configuration step above.
  • the analog feedback herein refers to feedback of analog angles of arrival, angles of departure, delays, gains, or other channel parameters.
  • the tensor feedback refers to feedback of factor matrices, a subset of factor matrices, vectors, etc.
  • the rank R f represents the number of values of a channel parameter related to a domain. For example, a number R f of AOD values estimated from the factor matrix A T may be fed back.
  • the network may also configure the UE to feed back CSI parameters including RI, CQI, or PMI.
  • the network may configure the UE with information, for the UE to measure the downlink MIMO channel, to form a tensor representing the downlink MIMO channel, and to feedback CSI of the downlink MIMO channel based on decomposition of the tensor.
  • the information may be generally referred to as CSI measurement and feedback configuration information. (or CSI measurement and feedback parameters) .
  • the CSI measurement and feedback configuration information may include configuration of the domains for constructing a MIMO channel tensor, configuration of the domains to provide feedback upon, a decomposition rank (CPD rank L) indicating the number of vectors in each factor matrix of the tensor, feedback rank R f , one or more channel parameters (some type of analog feedback) , measurement resources, feedback resources, or any combination thereof.
  • CPD rank L decomposition rank
  • the CSI measurement and feedback configuration information may include configuration parameters.
  • These configuration parameters may include, e.g., tensor domains (or dimensions) , feedback granularity of a tensor domain, measurement resources, a feedback dimension, the rank L, the feedback rank, feedback periodicity, a channel parameter, and/or a CSI parameter, e.g., RI, CQI, PMI.
  • the network may configure a UE with one, or more, or all of these configuration parameters.
  • the network may configure RI, CQI, and/or PMI in the same configuration as or separate configuration from tensor based feedback.
  • the network may configure the CPD rank and the feedback rank in the same configuration or in separate configurations.
  • the base station may signal the CSI measurement and feedback configuration information using higher layer signaling, e.g., radio resource control (RRC) signaling, medium access control (MAC) signaling, e.g., a MAC control element (CE) , or downlink control signaling, e.g., downlink control information (DCI) signaling.
  • RRC radio resource control
  • MAC medium access control
  • CE MAC control element
  • DCI downlink control information
  • the base station may trigger, using DCI, a UE to measure and feed back CSI according to the signaled CSI measurement and feedback configuration.
  • the DCI may or may not include the decomposition rank L.
  • the UE may be configured to feed back CSI periodically according to the signaled CSI measurement and feedback configuration.
  • the base station may signal the CSI measurement and feedback configuration information in one or more messages to a UE.
  • the base station may configure a UE for CSI measurement and feedback dynamically on demand.
  • the UE may measure the downlink MIMO channel between the UE and the base station or covariance of the downlink MIMO channel from the measured downlink reference signals.
  • the UE may construct a tensor representing the measured downlink MIMO channel after the channel measurement and estimation, e.g., the tensor represented by expression (3) or (4) , and decompose the tensor using a tensor decomposition technique, e.g., CPD in expression (6) , generating factor matrices.
  • the UE may then, in the feedback step, feedback (or report) CSI based on the resulted factor matrices, e.g., A T , A R , and C and based on the configuration specified by the network.
  • this approach may be generally referred to as a tensor-based CSI feedback scheme.
  • the UE may use any applicable tensor decomposition technique, such as the Canonical Polyadic decomposition (CPD) , the Tucker decomposition, the Higher Order SVD (HOSVD) , or other tensor decomposition techniques, to decompose the MIMO channel tensor or the spatial covariance matrix of the MIMO channel into its constituent factor matrices.
  • CPD Canonical Polyadic decomposition
  • HSVD Higher Order SVD
  • Those of ordinary skill in the art would recognize that various decomposition techniques may be used to decompose the tensor.
  • the Tucker decomposition or the multi-linear decomposition may be used to decompose the MIMO channel into factor matrices and a core tensor.
  • the factor matrices may include information related to the domains of time, frequency, transmit, receive, and/or delay.
  • the core tensor includes the strength of the interaction between the columns of the factor matrices.
  • the UE may feedback the factor matrices and the core tensor, which may be used to reconstruct the MIMO channel at the network.
  • the UE may feedback a subset of vectors of the factor matrices and the corresponding relevant values from the core tensor.
  • the factor matrices resulted from the tensor decomposition or the multi-linear decomposition may be used as basis (referred to as basis matrices) for a codebook.
  • basis matrices the core tensor resulted from the Tucker decomposition or the multi-linear decomposition may be quantized, and the resulting quantized coefficients (after quantization) may be fed back to the network.
  • the network may use the feedback information to compute a downlink codebook.
  • the network may use the quantized feedback information of the core tensor and the feedback representing the different dimensions to compute the downlink precoding.
  • the feedback representing the different dimensions may be an index of codebook (may also be referred to as basis) that is suitable for each dimension. That is, the UE may not feedback an actual vector or vectors of factor matrices, but a codeword index from a codebook. The codeword may then be selected based on the index. An optimal codeword may also be selected based on RSRP, capacity, or throughput optimization.
  • the UE transmits a CSI feedback report including feedback parameters that are configured by the network.
  • the feedback parameters may include a combination of CQI, RI, PMI, factor matrices or subsets of the factor matrices, a CPD rank, a feedback rank, and/or a MIMO transmission rank.
  • the periodicity for transmitting the CSI feedback report may be configured by the network.
  • the UE may feedback all the factor matrices of the tensor according to the configuration signaled by the network.
  • the network may receive the factor matrices that is reported, and reconstruct (e.g., compute, reassemble, estimate) the downlink MIMO channel by taking the sum of the outer product of the vectors of the factor matrices as shown in the following:
  • the factor matrices coefficients may be quantized and the quantized coefficients may be fed back to the network.
  • the amplitude and phase values of the factor matrices coefficients may be quantized.
  • vector quantization of the vectors themselves may be performed.
  • the network may directly use the factor matrices to compute the downlink precoder without computing (or reconstructing) the downlink MIMO channel. For example, if the feedback includes the factor matrix A T , the network may use this matrix as basis for downlink precoding of a CSI-RS.
  • the UE may feedback a single vector from the factor the matrix A T , a single vector from the factor matrix A R , and a single vector c 1 from the factor matrix C.
  • the network may be able to reconstruct part of the original downlink MIMO channel by taking the outer product of the three vectors
  • the base station may compute the downlink precoder per subcarrier, or per tone, or per subband.
  • feeding back a single vector from each domain may not be sufficient to capture a good representation of the channel (e.g., to closely approximate the channel in terms of minimizing the least square error, or maximize the throughput) . In this case, more vectors need to be fed back.
  • the network may configure how many vectors from each domain (i.e., one or more feedback ranks) may need to be fed back.
  • FIG. 4 illustrates a graph 400 showing example SU-MIMO capacity curves of a MIMO system when using different CSI to reconstruct a downlink MIMO channel.
  • the graph 400 includes curves 402, 404 and 406, each shows the SU-MIMO capacity, i.e., number of bits per second and per tone that varies with signal and noise ratio (SNR) .
  • the curve 402 shows the SU-MIMO capacity when a subband scheme is used, where precoding per subband is based on SVD decomposition of a MIMO channel matrix H per subband (e.g., the matrix H in the Equation (1) ) .
  • the precoder for each subband is determined based on the SVD decomposition, specifically, based on V (k) * .
  • the cure 404 shows the SU-MIMO capacity when a wideband scheme is used, where a wideband precoding is applied based on an average covariance matrix over all subbands.
  • the covariance matrix is represented by
  • the curve 406 shows the SU-MIMO capacity when a CPD reconstruction (CPDR) scheme is used, where the downlink MIMO channel is reconstructed using CSI generated based on the CPD decomposition of a MIMO channel tensor.
  • CPDR CPD reconstruction
  • a UE forms a tensor as discussed above with respect to expressions (2) - (4) , decomposes the tensor using CPD generating three factor matrices A T , A R , and C, and feeds back a subset of each factor matrices.
  • a base station reconstructs the downlink MIMO channel using the fed back CSI, and applies per subband precoding to transmission signals.
  • the CPDR scheme (curve 406) has about the same performance as the subband scheme (curve 402) , but better than the wideband scheme (curve 404) . However, the CPDR requires less CSI feedback overhead than the subband scheme.
  • the subband scheme For feeding back the downlink MIMO tensor, the subband scheme needs to feed back N Tx *N Rx *N K amount of information for reconstructing the downlink MIMO.
  • the wideband scheme needs to feed back N Tx *N K amount of information.
  • R is the rank of MIMO transmission, representing a number of streams to be transmitted
  • N Tx +N Rx + N K *R f amount of CSI information needs to be fed back, while achieving similar performance as the subband scheme.
  • FIG. 5 illustrates a diagram of an embodiment method 500 for CSI feedback in a MIMO communications system.
  • a base station 502 including multiple transmit antennas communicates with a UE 504 including multiple receive antennas.
  • the base station 502 determines a CSI measurement and feedback configuration, with which the base station 502 configures the UE 504 for measuring a downlink MIMO channel between the base station 502 and the UE 504, and feeding back CSI of the downlink MIMO channel.
  • the CSI measurement and feedback configuration may include information as discussed above, such as domains for constructing a MIMO channel tensor, decomposition rank L indicating the number of vectors in each factor matrix of the tensor, feedback rank Rf, one or more factor matrices, one or more channel parameters, measurement resources, feedback resources, periodicity for CSI feedback, or any combination thereof.
  • the UE 502 configures the UE 504 with the CSI measurement and feedback configuration.
  • the base station 502 may signal the CSI measurement and feedback configuration to the UE 504 using higher layer signaling, DCI signaling, or MAC signaling.
  • the base station 502 transmits reference signals to the UE 504 for the UE 504 to measure the downlink MIMO channel.
  • the UE 504 measures the downlink MIMO channel using the reference signals, forms a multi-dimensional tensor according to the signaled CSI measurement and feedback configuration, and decomposes the tensor to generate factor matrices according to the signaled CSI measurement and feedback configuration.
  • the UE 504 forms the tensor using the domains specified by the CSI measurement and feedback configuration. Decomposition of the tensor is performed based on the decomposition rank L specified by the CSI measurement and feedback configuration. Thus, each of the factor matrices includes L sets of data.
  • the UE 504 generates CSI that is to be fed back to the base station 502 according to the signaled CSI measurement and feedback configuration.
  • the CSI measurement and feedback configuration specifies feeding back all factor matrices
  • the CSI includes all the factor matrices that are generated. If a subset of each factor matrix is specified by the CSI measurement and feedback configuration, e.g., if a feedback rank Rf is specified, the CSI includes Rf sets of data in the L sets of data of each factor matrix. If the CSI measurement and feedback configuration specifies feeding back a channel parameter, the CSI includes L (or Rf, if specified) sets of the channel parameter values.
  • the CSI may also include RI, PMI or CQI.
  • the UE 504 transmits the generated CSI to the base station.
  • the base station performs downlink transmission with the UE 504 based on the received CSI.
  • the base station 502 may reconstruct the downlink MIMO channel using all the factor matrices fed back, or reconstruct a portion of the downlink MIMO channel using subsets of the factor matrices. For example, the base station 502 may take outer product of all the factor matrices, or the subsets of the factor matrices. The result is a reconstructed channel or a reconstructed portion of the channel.
  • the base station 502 may also estimate one or more channel parameters using the CSI that including information about the one or more channel parameters.
  • the base station 502 may determine precoding parameters, modulation parameters, and other parameters used for transmitting signals to the UE 504 according to the reconstructed channel, the reconstructed portion of the channel, or the estimated one or more channel parameters.
  • FIG. 6 illustrates a flowchart of an embodiment method 600 for CSI feedback in a MIMO communications system.
  • the embodiment method 600 may be performed by a UE communicating with a base station in the MIMO communications system.
  • the UE forms a tensor representing a measured downlink MIMO channel between the base station and the UE.
  • the UE may form the tensor according to domains or dimensions or modes specified by a received CSI measurement and feedback configuration. For example, the UE may form a tensor as described above.
  • the UE performs a rank L CPD on the tensor after channel measurement and estimation.
  • the rank L CPD results in factor matrices.
  • Each factor matrix includes L sets of data.
  • the UE generates a measurement report based on the factor matrices.
  • the measurement report includes information indicating the downlink MIMO channel status.
  • the measurement report may include the factor matrices, a subset of each factor matrix, and/or one or more parameters estimated based on the factor matrices.
  • the UE transmits the measurement report to the base station.
  • FIG. 7 illustrates a flowchart of another embodiment method 700 for CSI feedback in a MIMO communications system.
  • the embodiment method 700 may be performed by a UE communicating with a base station in the MIMO communications system.
  • the UE receives a configuration for CSI measurement and feedback.
  • the configuration may include quantities (i.e., information) as discussed above with respect to the CSI measurement and feedback configuration information.
  • the quantities include CSI-RS resources, CSI reporting resources and trigger information.
  • the quantities may also include the decomposition rank L, and tensor domains, e.g., the transmit domain, the receive domain, the frequency domain, the time domain, or the delay domain , for constructing a tensor.
  • the UE performs channel measurement on the CSI-RS resources. This includes measuring CSI-RSs, estimating and forming a tensor representing a downlink MIMO channel between the base station and the UE. Step 704 may be triggered by a DCI message, or may be performed periodically as configured by the network.
  • the UE generates quantities representing the downlink MIMO channel according to the decomposition rank L.
  • the UE may generate factor matrices by performing rank L decomposition on the tensor.
  • the quantities may include at least one factor matrix or a subset of one factor matrix in one of the domains.
  • the UE generates a report based on the quantities.
  • the report may be in a predefined format, and includes CSI to be reported according to the configuration.
  • the UE transmits the report.
  • FIG. 8 illustrates a flowchart of another embodiment method 800 for CSI feedback in a MIMO communications system.
  • the embodiment method 800 may be performed by a UE communicating with a base station in the MIMO communications system.
  • the UE forms a channel tensor representing a downlink MIMO channel between the UE and the base station.
  • the UE performs rank L CPD on the tensor.
  • the UE may alternatively perform rank L MLSVD on the tensor.
  • the UE obtains factor matrices resulted from decomposition of the tensor.
  • the UE generates a report including the obtained factor matrices.
  • the UE transmits the report.
  • FIG. 9 illustrates a flowchart of another embodiment method 900 for CSI feedback in a MIMO communications system.
  • the embodiment method 900 may be performed by a UE communicating with a base station in the MIMO communications system.
  • the UE forms a channel tensor representing a downlink MIMO channel between the UE and the base station.
  • the UE performs rank L CPD on the tensor.
  • the UE may alternatively perform rank L MLSVD on the tensor.
  • the UE obtains factor matrices resulted from decomposition of the tensor.
  • the UE generates a report including a subset of vectors of each of the obtained factor matrices.
  • the UE transmits the report.
  • FIG. 10 illustrates a flowchart of another embodiment method 1000 for CSI feedback in a MIMO communications system.
  • the embodiment method 1000 may be performed by a UE communicating with a base station in the MIMO communications system.
  • the UE forms a channel tensor representing a downlink MIMO channel between the UE and the base station.
  • the UE performs rank L CPD on the tensor.
  • the UE may alternatively perform rank L MLSVD on the tensor.
  • the UE obtains factor matrices resulted from decomposition of the tensor.
  • the UE obtains values of one or more channel parameters estimated based on the factor matrices.
  • the UE generates a report including the values of the one or more channel parameters.
  • the UE transmits the report.
  • FIG. 11 illustrates a flowchart of another embodiment method 1100 for CSI feedback in a MIMO communications system.
  • the embodiment method 1100 may be performed by a base station communicating in the MIMO communications system.
  • the base station configures UEs for CSI measurement and feedback.
  • the base station may signal, to UEs, CSI measurement and feedback configuration information, as discussed above.
  • the base station transmits reference signals on measurement resources for downlink MIMO channel measurements.
  • the base station receives reports about downlink MIMO channel measurements.
  • Each report may include information about factor matrices that is generated and fed back by a UE.
  • Each report is associated with a downlink MIMO channel between the base station and a UE feeding back the report.
  • Each of the factor matrices has a rank L.
  • Each of the factor matrices has L sets of data.
  • the base station reconstructs downlink MIMO channels using the reports. For example, the base station may reconstruct a downlink MIMO channel between the base station and a UE based on factor matrices in a report sent by the UE, e.g., by taking the outer product of the factor matrices.
  • the base station performs downlink transmission to the UEs according to the reconstructed downlink MIMO channels.
  • FIG. 12 illustrates a diagram of another embodiment method 1200 for CSI feedback in a MIMO communications system, with emphasis on configuration of feedback rank Rf.
  • the method 1200 may be performed by a network device, such as a base station.
  • a base station 1202 signals the feedback rank Rf to a UE 1204 for reporting CSI in a tensor-based CSI feedback scheme.
  • the base station may have had signals other configuration information for the UE to form a tensor (e.g., domains) and decompose the tensor (e.g., the decomposition rank L) .
  • a tensor e.g., domains
  • decompose the tensor e.g., the decomposition rank L
  • the UE 1204 generates factor matrices of rank L by decomposing the tensor using a tensor decomposition technique.
  • the UE 1204 generates a CSI report according to the feedback rank Rf.
  • the CSI report may include the first Rf vectors in each factor matrix.
  • the CSI report may include Rf values of a channel parameter related to a domain (i.e., estimated from a factor matrix related to the domain) .
  • the UE 1204 transmits the CSI report to the base station 1202.
  • the base station 1202 transmits signals based on the CSI report.
  • FIG. 13 illustrates a flowchart of an embodiment method 1300 for CSI feedback in a communications system.
  • the method 1300 may be performed by a communication device, such as a user equipment (UE) .
  • a communication device such as a user equipment (UE) .
  • the UE receives, through M antenna ports of the UE, a reference signal (RS) resource set, where the RS resource set is transmitted from N antenna ports of a base station (BS) over K1 subcarriers on L OFDM symbols.
  • BS base station
  • K1 subcarriers on L OFDM symbols.
  • Each of the M, N, K1 and L1 is an integer greater than 0.
  • the UE reports, to the BS, R sets of vectors that are obtained based on measurement of the RS resource set, where each set of vectors comprises one or more vectors, and the one or more vectors comprise a first Nx1 vector, a second Mx1 vector, a third K2x1 vector, a fourth L2x1 vector, or any combination thereof, and where a sum of R outer products of the respective R sets of vectors represents channel state information of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on L2 OFDM symbols.
  • Each of the R, K2 and L2 is an integer greater than o.
  • FIG. 14 illustrates a flowchart of an embodiment method 1400 for CSI feedback in a communications system.
  • the method 1400 may be performed by a network device, such as a base station (BS) .
  • the BS transmits, through N antenna ports of the BS to M antenna ports of a user equipment (UE) , a reference signal (RS) resource set, where the RS resource set is transmitted over K1 subcarriers on L OFDM symbols.
  • RS reference signal
  • the BS receives, from the UE, a channel state information (CSI) report measured based on the RS resource set, where the CSI report comprises information of R sets of vectors, where each set of vectors comprises one or more vectors, and the one or more vectors comprises a first Nx1 vector, a second Mx1 vector, a third K2x1 vector, a fourth L2x1 vector, or any combination thereof, and where a sum of R outer products of the respective R sets of vectors represents CSI of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on L2 OFDM symbols.
  • Each of the R, K2 and L2 being an integer greater than o.
  • the BS reconstructs the channel based on the R sets of vectors by calculating an outer product of each set of vectors and summing up outer products of the respective R sets of vector.
  • a method includes: receiving, through M antenna ports of a user equipment (UE) , a reference signal (RS) resource set, the RS resource set being transmitted from N antenna ports of a base station (BS) over K1 subcarriers on L OFDM symbols, wherein each of the M, N, K1 and L1 is an integer greater than 0, and reporting, by the UE to the BS, R sets of vectors that are obtained based on measurement of the RS resource set, wherein each set of vectors comprises one or more vectors selected from a first Nx1 vector, a second Mx1 vector, a third K2x1 vector and a fourth L2x1 vector, wherein a sum of R outer products of the respective R sets of vectors represents channel state information of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on L2 OFDM symbols, wherein each of the R, K2 and L2 is an integer greater than o.
  • the method further includes receiving, by the UE, resources carrying the RS resource set.
  • the R sets of vectors are reported according to a configuration signaled by the BS.
  • the method further includes: constructing, by the UE, a multi-dimensional tensor representing the channel based on the measurement of the RS resource set, and performing, by the UE, decomposition on the multi-dimensional tensor using a tensor decomposition technique, whereby generating factor matrices, wherein the R sets of vectors are obtained from the factor matrices.
  • the R sets of vectors are a subset of the factor matrices.
  • the R sets of vectors comprise all vectors of the factor matrices.
  • the tensor decomposition technique comprises Canonical Polyadic decomposition (CPD) , Tucker decomposition, higher order singular vector decomposition (SVD) , or multilinear SVD.
  • CPD Canonical Polyadic decomposition
  • SVD singular vector decomposition
  • multilinear SVD multilinear SVD
  • the factor matrices are generated from Canonical Polyadic decomposition (CPD) of the multi-dimensional tensor.
  • CPD Canonical Polyadic decomposition
  • the factor matrices are generated from Tucker Decomposition or multilinear decomposition of the multi-dimensional tensor.
  • the multi-dimensional tensor is constructed based on the measurement of the RS resource set in a set of measurement domains, the set of measurement domains comprising a transmit domain, a receive domain, a time domain, a frequency domain, a delay domain, or any combination thereof.
  • the measurement of the RS resource set is performed in the set of measurement domains in an order of the set of measurement domains, wherein the order is configured by the BS.
  • a number of measurement domains in the set of measurement domains is configured by the BS.
  • one or more measurement domains included in the set of measurement domains are configured by the BS.
  • a factor matrix comprises a plurality of Nx1 vectors, a plurality of Mx1 vectors, a plurality of K2x1 vectors, or a plurality of L2x1 vector.
  • each set of vectors comprises the first Nx1 vector, the second Mx1 vector, the third K2x1 vector, and the fourth L2x1 vector.
  • the method further includes receiving, by the UE, signaling indicating the one or more vectors to be reported.
  • resources for transmitting the RS resource set are non-zero-power CSI-RS resources that are configured by the BS for CSI reporting.
  • the R sets of vectors that are reported comprises quantized amplitudes and phases of a vector in a set of vectors.
  • the method further includes receiving, by the UE, configuration information for reporting the R sets of vectors in higher layer signaling, or a downlink control information (DCI) message.
  • DCI downlink control information
  • the configuration information comprises information triggering reporting of the R sets of vectors.
  • a method includes: transmitting, through N antenna ports of a base station (BS) to M antenna ports of a user equipment (UE) , a reference signal (RS) resource set, the RS resource set being transmitted over K1 subcarriers on L OFDM symbols, wherein each of the M, N, K1 and L1 is an integer greater than 0, receiving, by the BS from the UE, a channel state information (CSI) report measured based on the RS resource set, the CSI report comprising information of R sets of vectors, wherein each set of vectors comprises one or more vectors selected from a first Nx1 vector, a second Mx1 vector, a third K2x1 vector, and a fourth L2x1 vector, and a sum of R outer products of the respective R sets of vectors represents CSI of a channel between the N antenna ports of the BS and the M antenna ports of the UE over K2 subcarriers on L2 OFDM symbols, each of the R, K2 and
  • the method further includes transmitting, by the BS to the UE, resources carrying the RS resource set.
  • the method further includes transmitting, by the BS to the UE, information about R.
  • the method further includes transmitting, by the BS to the UE, a set of measurement domains for the UE to measure the RS resource set in the set of measurement domains, the set of measurement domains comprising a transmit domain, a receive domain, a time domain, a frequency domain, a delay domain, or any combination thereof.
  • the method further includes transmitting, by the BS to the UE, an order of the set of measurement domains for the UE to measure the RS resource set in the order of the set of measurement domains.
  • the method further includes transmitting, by the BS to the UE, a number of measurement domains in the set of measurement domains.
  • the method further includes configuring, by the BS for the UE, one or more measurement domains comprised in the set of measurement domains.
  • the method further includes transmitting, by the BS to the UE, signaling indicating the one or more vectors to be reported.
  • resources for transmitting the RS resource set are non-zero-power CSI-RS resources that are configured by the BS for CSI reporting.
  • the R sets of vectors that are reported comprises quantized amplitudes and phases of a vector in a set of vectors.
  • the method further includes transmitting, by the BS to the UE, a trigger triggering reporting of the CSI of the channel.
  • the method further includes generating, by the BS to the UE, a transmission filter based on the CSI report.
  • FIG. 15 illustrates a block diagram of an embodiment processing system 1500 for performing methods described herein, which may be installed in a host device.
  • the processing system 1500 includes a processor 1504, a memory 1506, and interfaces 1510-1514, which may (or may not) be arranged as shown in FIG. 15.
  • the processor 1504 may be any component or collection of components adapted to perform computations and/or other processing related tasks
  • the memory 1506 may be any component or collection of components adapted to store programming and/or instructions for execution by the processor 1504.
  • the memory 1506 includes a non-transitory computer readable medium.
  • the interfaces 1510, 1512, 1514 may be any component or collection of components that allow the processing system 1500 to communicate with other devices/components and/or a user.
  • one or more of the interfaces 1510, 1512, 1514 may be adapted to communicate data, control, or management messages from the processor 1504 to applications installed on the host device and/or a remote device.
  • one or more of the interfaces 1510, 1512, 1514 may be adapted to allow a user or user device (e.g., personal computer (PC) , etc. ) to interact/communicate with the processing system 1500.
  • the processing system 1500 may include additional components not depicted in FIG. 15, such as long term storage (e.g., non-volatile memory, etc. ) .
  • the processing system 1500 is included in a network device that is accessing, or part otherwise of, a telecommunications network.
  • the processing system 1500 is in a network-side device in a wireless or wireline telecommunications network, such as a base station, a relay station, a scheduler, a controller, a gateway, a router, an applications server, or any other device in the telecommunications network.
  • the processing system 1500 is in a user-side device accessing a wireless or wireline telecommunications network, such as a mobile station, a user equipment (UE) , a personal computer (PC) , a tablet, a wearable communications device (e.g., a smartwatch, etc. ) , or any other device adapted to access a telecommunications network.
  • UE user equipment
  • PC personal computer
  • tablet a wearable communications device
  • FIG. 16 illustrates a block diagram of a transceiver 1600 adapted to transmit and receive signaling over a telecommunications network.
  • the transceiver 1600 may be installed in a host device. As shown, the transceiver 1600 comprises a network-side interface 1602, a coupler 1604, a transmitter 1606, a receiver 1608, a signal processor 1610, and a device-side interface 1612.
  • the network-side interface 1602 may include any component or collection of components adapted to transmit or receive signaling over a wireless or wireline telecommunications network.
  • the coupler 1604 may include any component or collection of components adapted to facilitate bi-directional communication over the network-side interface 1602.
  • the transmitter 1606 may include any component or collection of components (e.g., up-converter, power amplifier, etc. ) adapted to convert a baseband signal into a modulated carrier signal suitable for transmission over the network-side interface 1602.
  • the receiver 1608 may include any component or collection of components (e.g., down-converter, low noise amplifier, etc. ) adapted to convert a carrier signal received over the network-side interface 1602 into a baseband signal.
  • the signal processor 1610 may include any component or collection of components adapted to convert a baseband signal into a data signal suitable for communication over the device-side interface (s) 1612, or vice-versa.
  • the device-side interface (s) 1612 may include any component or collection of components adapted to communicate data-signals between the signal processor 1610 and components within the host device (e.g., the processing system 1500, local area network (LAN)
  • the transceiver 1600 may transmit and receive signaling over any type of communications medium.
  • the transceiver 1600 transmits and receives signaling over a wireless medium.
  • the transceiver 1600 may be a wireless transceiver adapted to communicate in accordance with a wireless telecommunications protocol, such as a cellular protocol (e.g., long-term evolution (LTE) , etc. ) , a wireless local area network (WLAN) protocol (e.g., Wi-Fi, etc. ) , or any other type of wireless protocol (e.g., Bluetooth, near field communication (NFC) , etc. ) .
  • the network-side interface 1602 comprises one or more antenna/radiating elements.
  • the network-side interface 1602 may include a single antenna, multiple separate antennas, or a multi-antenna array configured for multi-layer communication, e.g., single input multiple output (SIMO) , multiple input single output (MISO) , multiple input multiple output (MIMO) , etc.
  • the transceiver 1600 transmits and receives signaling over a wireline medium, e.g., twisted-pair cable, coaxial cable, optical fiber, etc.
  • Specific processing systems and/or transceivers may utilize all of the components shown, or only a subset of the components, and levels of integration may vary from device to device.
  • a signal may be transmitted by a transmitting unit or a transmitting module.
  • a signal may be received by a receiving unit or a receiving module.
  • a signal may be processed by a processing unit or a processing module.
  • a generating unit/module may be performed by a generating unit/module, a calculating unit/module, a reporting unit/module, a measuring unit/module, a reconstructing unit/module, a decomposing unit/module, a constructing unit/module, a performing unit/module, a quantizing unit/module, a triggering unit/module, a signaling unit/module, and/or a setting unit/module.
  • the respective units/modules may be hardware, software, or a combination thereof.
  • one or more of the units/modules may be an integrated circuit, such as field programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs) .
  • FPGAs field programmable gate arrays
  • ASICs application-specific integrated circuits

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Power Engineering (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

Selon l'invention, un équipement d'utilisateur (UE) reçoit, au travers de M ports d'antenne, un ensemble de ressources de signal de référence (RS) transmis depuis N ports d'antenne d'une station de base (BS) sur K1 sous-porteuses sur L1 symboles OFDM, et signale, à la BS, R ensembles de vecteurs qui sont obtenus sur la base de la mesure de l'ensemble de ressources RS. Chaque ensemble de vecteurs comprend un ou plusieurs vecteurs sélectionnés parmi un vecteur Nx1, un vecteur Mx1, un vecteur K2x1 et un vecteur L2x1, et la somme des R produits externes des R ensembles respectifs de vecteurs représente des informations d'état de canal d'un canal entre les N ports d'antenne de la BS et les M ports d'antenne de L'UE sur K2 sous-porteuses sur L2 symboles OFDM. La BS reconstruit le canal en calculant un produit externe de chaque ensemble de vecteurs et en totalisant des produits externes des R ensembles respectifs de vecteurs.
PCT/CN2020/082096 2019-03-28 2020-03-30 Système et procédé de rétroaction et de signalement de csi réduits utilisant des tenseurs et une décomposition de tenseur WO2020192790A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201962825624P 2019-03-28 2019-03-28
US62/825,624 2019-03-28

Publications (1)

Publication Number Publication Date
WO2020192790A1 true WO2020192790A1 (fr) 2020-10-01

Family

ID=72608509

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/082096 WO2020192790A1 (fr) 2019-03-28 2020-03-30 Système et procédé de rétroaction et de signalement de csi réduits utilisant des tenseurs et une décomposition de tenseur

Country Status (1)

Country Link
WO (1) WO2020192790A1 (fr)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11251845B2 (en) * 2016-05-13 2022-02-15 Huawei Technologies Co., Ltd. Channel information sending method, data sending method, and device
CN114268904A (zh) * 2021-11-29 2022-04-01 西安电子科技大学 室内移动目标三维定位方法、介质、计算机设备及应用
WO2022170500A1 (fr) * 2021-02-09 2022-08-18 北京小米移动软件有限公司 Procédé et appareil de traitement de capacité d'équipement utilisateur, dispositif de communication et support d'enregistrement
WO2023150299A1 (fr) * 2022-02-04 2023-08-10 Interdigital Patent Holdings, Inc. Dispersion de groupe et régularisation implicite pour estimation de canal mimo
WO2023147681A1 (fr) * 2022-02-01 2023-08-10 Qualcomm Incorporated Rapport de base de domaine temporel pour informations d'état de canal
WO2023184538A1 (fr) * 2022-04-02 2023-10-05 Oppo广东移动通信有限公司 Procédé de traitement d'informations, et dispositif
WO2024031679A1 (fr) * 2022-08-12 2024-02-15 Shenzhen Tcl New Technology Co., Ltd. Dispositif de communication et procédé de rapport de réalités de terrain groupées pour intelligence artificielle/apprentissage automatique
WO2024130611A1 (fr) * 2022-12-21 2024-06-27 Nec Corporation Procédé, dispositif et support de stockage informatique de communication

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102388628A (zh) * 2010-01-12 2012-03-21 中兴通讯美国公司 基于克罗内克积的空间信道状态信息反馈的方法和***
US20180076871A1 (en) * 2016-09-14 2018-03-15 Samsung Electronics Co., Ltd. Method and apparatus to enable channel compression in advanced wireless communication systems
EP3459439A1 (fr) * 2017-09-22 2019-03-27 Nokia Technologies Oy Détection d'activité ou de presence
EP3576361A1 (fr) * 2018-06-01 2019-12-04 FRAUNHOFER-GESELLSCHAFT zur Förderung der angewandten Forschung e.V. Rétroaction d'informations de canal explicite sur la base d'une décomposition de pca ou d'une composition de pca d'ordre supérieur

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102388628A (zh) * 2010-01-12 2012-03-21 中兴通讯美国公司 基于克罗内克积的空间信道状态信息反馈的方法和***
US20180076871A1 (en) * 2016-09-14 2018-03-15 Samsung Electronics Co., Ltd. Method and apparatus to enable channel compression in advanced wireless communication systems
EP3459439A1 (fr) * 2017-09-22 2019-03-27 Nokia Technologies Oy Détection d'activité ou de presence
EP3576361A1 (fr) * 2018-06-01 2019-12-04 FRAUNHOFER-GESELLSCHAFT zur Förderung der angewandten Forschung e.V. Rétroaction d'informations de canal explicite sur la base d'une décomposition de pca ou d'une composition de pca d'ordre supérieur

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
QIAN, CHENG ET AL.: "Tensor-Based Channel Estimation for Dual-Polarized Massive MIMO Systems", IEEE TRANSACTIONS ON SIGNAL PROCESSING, vol. 66, no. 24, 15 December 2018 (2018-12-15), XP081421459, DOI: 20200609140910A *
ZHU, YU ET AL.: "HOSVD-based Limited Feedback and Precoding Design for Massive MIMO Systems", 2016 IEEE/CIC INTERNATIONAL CONFERENCE ON COMMUNICATIONS IN CHINA (ICCC WORKSHOPS), 10 October 2016 (2016-10-10), XP032976331, DOI: 20200609140706A *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11251845B2 (en) * 2016-05-13 2022-02-15 Huawei Technologies Co., Ltd. Channel information sending method, data sending method, and device
WO2022170500A1 (fr) * 2021-02-09 2022-08-18 北京小米移动软件有限公司 Procédé et appareil de traitement de capacité d'équipement utilisateur, dispositif de communication et support d'enregistrement
CN114268904A (zh) * 2021-11-29 2022-04-01 西安电子科技大学 室内移动目标三维定位方法、介质、计算机设备及应用
CN114268904B (zh) * 2021-11-29 2022-09-30 西安电子科技大学 室内移动目标三维定位方法、介质、计算机设备及应用
WO2023147681A1 (fr) * 2022-02-01 2023-08-10 Qualcomm Incorporated Rapport de base de domaine temporel pour informations d'état de canal
WO2023150299A1 (fr) * 2022-02-04 2023-08-10 Interdigital Patent Holdings, Inc. Dispersion de groupe et régularisation implicite pour estimation de canal mimo
WO2023184538A1 (fr) * 2022-04-02 2023-10-05 Oppo广东移动通信有限公司 Procédé de traitement d'informations, et dispositif
WO2024031679A1 (fr) * 2022-08-12 2024-02-15 Shenzhen Tcl New Technology Co., Ltd. Dispositif de communication et procédé de rapport de réalités de terrain groupées pour intelligence artificielle/apprentissage automatique
WO2024130611A1 (fr) * 2022-12-21 2024-06-27 Nec Corporation Procédé, dispositif et support de stockage informatique de communication

Similar Documents

Publication Publication Date Title
WO2020192790A1 (fr) Système et procédé de rétroaction et de signalement de csi réduits utilisant des tenseurs et une décomposition de tenseur
US10804993B2 (en) Method and apparatus for downlink and uplink CSI acquisition
KR102618282B1 (ko) Mimo 측정 기준 신호 및 피드백을 동작시키기 위한 방법 및 장치
US10567066B2 (en) Apparatus and method for performing precoding in wireless communication system using massive antenna
KR102402529B1 (ko) 감소된 밀도 csi-rs를 위한 메커니즘들
EP3504854B1 (fr) Système et procédé de transmission d'une sélection d'un sous-espace
WO2019218967A1 (fr) Système et procédé pour un apprentissage de système de communication
EP3228021B1 (fr) Procédé et appareil de signalisation de liaison descendante pour retour d'informations de csi-rs et de csi partiellement précodées
KR102525602B1 (ko) 다중 안테나를 이용하는 무선 통신 시스템에서 채널 상태 정보를 송수신하는 방법 및 장치
JP6019502B2 (ja) 協調マルチポイント送信のためのチャネルフィードバック
US10374836B2 (en) Method and apparatus for downlink channel estimation in massive MIMO
US20230361842A1 (en) Improving precoding
JP2016040946A (ja) 無線通信システムにおいてチャネル状態情報をフィードバックする端末装置及びその方法
CN109075904A (zh) 一种预编码矩阵指示的反馈方法及装置
AU2021316062B2 (en) Localization and auto-calibration in a wireless network
WO2018024157A1 (fr) Procédé de transmission d'informations d'état de canal, procédé de réception, et dispositif et système
WO2022009151A1 (fr) Csi-rs partagé pour retour de csi basé sur une réciprocité partielle
CN109478948A (zh) 一种信道信息传输装置、方法和***
JPWO2016207929A1 (ja) 無線通信制御方法、無線通信システム、受信装置および送信装置
CN114499608A (zh) 用信号发送端口信息
CN114600384B (zh) 一种信道测量方法和通信装置
US20120057645A1 (en) method for communicating in a network
EP4304237A1 (fr) Procédé et appareil de rapport d'informations, premier dispositif et second dispositif
US20230051245A1 (en) Methods and Apparatus for Channel Estimation and Precoding with Incomplete Channel Observation and Channel State Information Feedback
WO2023170655A1 (fr) Amélioration d'indicateur de matrice de précodeur (pmi) de type ii pour une transmission conjointe cohérente (cjt)

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20779465

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20779465

Country of ref document: EP

Kind code of ref document: A1