Method and apparatus for receiving channel state information in a wireless communication system

Abstract

The present invention relates to a method and apparatus for receiving channel state information (CSI) in a wireless communication system, and the method may include the following steps: measuring a CSI-reference signal (RS) sent from a base station through multiple panels; and reporting CSI generated based on the measured CSI-RS to the base station, wherein, when a terminal reports a wideband (WB) panel corrector and a subband (SB) panel corrector for multiple panels as CSI, the WB panel corrector and the SB panel corrector are reported using different bit widths.

Description

[0001] This application is a divisional application of the patent application with the international application date of January 8, 2018, application number 201880006166.0 (PCT / KR2018 / 000345) filed with the China Patent Office on July 8, 2019, and the invention name is “Method and device for reporting channel state information in a wireless communication system”. Technical Field

[0002] The present invention relates to a wireless communication system, and more particularly, to a method for reporting channel state information and a device for executing / supporting the method. Background Art

[0003] Mobile communication systems have evolved to provide voice services while ensuring user mobility. However, their service coverage has expanded to include data services as well as voice services. Today, the explosive growth of services has led to resource shortages and user demand for high-speed services, necessitating the need for more advanced mobile communication systems.

[0004] The requirements for next-generation mobile communication systems may include supporting huge data traffic, a significant increase in the transmission rate per user, accommodating a significantly increased number of connected devices, very low end-to-end latency, and high energy efficiency. To this end, various technologies such as dual connectivity, massive multiple-input multiple-output (MIMO), in-band full-duplex, non-orthogonal multiple access (NOMA), support for ultra-wideband, and device networking have been studied. Summary of the Invention

[0005] Technical issues

[0006] An object of the present invention is to provide a method for transmitting / receiving channel state information (CSI).

[0007] Furthermore, an object of the present invention is to provide a method for transmitting / receiving control configuration information, which is transmitted to support transmission / reception of channel state information of a terminal.

[0008] In addition, an object of the present invention is to provide various codebooks for CSI reporting / feedback. In particular, an object of the present invention is to provide a new codebook for supporting beamforming through multiple panels newly introduced in NR.

[0009] The technical problems to be solved by the present invention are not limited to the above-mentioned technical problems, and those skilled in the art can clearly understand other technical problems not mentioned herein from the following description.

[0010] Technical Solutions

[0011] In one aspect of the present invention, a method for reporting channel state information (CSI) of a terminal in a wireless communication system is provided. The method includes: measuring a CSI-RS (CSI-Reference Signal) transmitted from a base station using multiple panels; and reporting CSI generated based on the CSI-RS measurement to the base station. When the terminal reports a WB (wideband) panel corrector and an SB (subband) panel corrector for the multiple panels as CSI, the WB panel corrector and the SB panel corrector are reported using different bit widths.

[0012] In addition, the WB panel corrector and SB panel corrector are used for phase correction between multiple panels.

[0013] In addition, the bit width of the SB panel corrector is smaller than that of the WB panel corrector.

[0014] Furthermore, the bit width of the SB panel corrector is 1 bit, and the bit width of the WB panel corrector is 2 bits.

[0015] Furthermore, the WB panel corrector is reported based on QPSK (Quadrature Phase Shift Keying), and the SB panel corrector is reported based on BPSK (Binary Phase Shift Keying).

[0016] In addition, when only the WB panel compensator is reported as CSI, the WB panel corrector is reported using a bit width of 2 bits.

[0017] Furthermore, the number of panels is set by higher layer signaling.

[0018] In addition, the reporting WB panel corrector and / or SB panel corrector is set by higher layer signaling.

[0019] In addition, the WB panel corrector and the SB panel corrector are reported in the PMI (Precoding Matrix Index) in the CSI.

[0020] In addition, the WB panel corrector and the SB panel corrector are reported independently for each of the multiple panels.

[0021] In another aspect of the present invention, a terminal for receiving a channel state information reference signal (CSI-RS) in a wireless communication system is provided, the terminal including: a radio frequency (RF) unit for sending and receiving radio signals; and a processor for controlling the RF unit, wherein the processor measures a CSI-RS (reference signal) sent from a base station through multiple panels, and reports the CSI generated based on the CSI-RS measurement to the base station, wherein when the terminal reports a WB (wideband) panel corrector and an SB (subband) panel corrector for multiple panels as CSI, the WB panel corrector and the SB panel corrector are reported using different bit widths.

[0022] In addition, the WB panel corrector and SB panel corrector are used for phase correction between multiple panels.

[0023] In addition, the bit width of the SB panel corrector is smaller than that of the WB panel corrector.

[0024] Furthermore, the bit width of the SB panel corrector is 1 bit, and the bit width of the WB panel corrector is 2 bits.

[0025] Furthermore, the WB panel corrector is reported based on QPSK (Quadrature Phase Shift Keying), and the SB panel corrector is reported based on BPSK (Binary Phase Shift Keying).

[0026] Effects of the present invention

[0027] According to exemplary embodiments of the present invention, the terminal can smoothly derive CSI and provide feedback thereon to the base station.

[0028] Furthermore, according to exemplary embodiments of the present invention, a codebook for NR that is newly adapted to a multi-panel array is defined, thereby resolving the ambiguity of which codebook to apply to NR.

[0029] Furthermore, according to an exemplary embodiment of the present invention, different bit widths are defined for the WB panel corrector and the SB panel corrector by taking into account the SB characteristics. Therefore, accurate CSI can be reported to the base station without significantly increasing signaling overhead.

[0030] Advantages of the following embodiments are not limited to the aforementioned advantages, and various other advantages may be apparently understood from the following description by those skilled in the art to which the embodiments pertain. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings are included as part of the specification to help understand the present invention, provide embodiments of the present invention, and describe the technical features of the present invention with the help of the following description.

[0032] Figure 1 The diagram illustrates a structure of a radio frame in a wireless communication system to which the present invention can be applied.

[0033] Figure 2 is a diagram illustrating a resource grid for a downlink slot in a wireless communication system to which the present invention can be applied.

[0034] Figure 3 FIG: 1 illustrates a structure of a downlink subframe in a wireless communication system to which the present invention can be applied.

[0035] Figure 4 FIG. 1 illustrates a structure of an uplink subframe in a wireless communication system to which the present invention can be applied.

[0036] Figure 5 The configuration of a known MIMO communication system is shown.

[0037] Figure 6 is a diagram illustrating channels from multiple transmit antennas to a single receive antenna.

[0038] Figure 7 Illustrations show reference signal patterns mapped to downlink resource block pairs in a wireless communication system to which the present invention can be applied.

[0039] Figure 8 is a diagram illustrating resources to which reference signals are mapped in a wireless communication system to which the present invention can be applied.

[0040] Figure 9 The diagram illustrates resources to which reference signals are mapped in a wireless communication system to which the present invention is applicable.

[0041] Figure 10 A two-dimensional (2D) active antenna system with 64 antenna elements is shown in the figure in the adaptable wireless communication system of the present invention.

[0042] Figure 11 FIG. 1 illustrates a system in which a base station or a UE has multiple transmission / reception antennas capable of forming a three-dimensional (3D) beam based on AAS in a wireless communication system to which the present invention is applicable.

[0043] Figure 12 FIG. 1 shows a 2D antenna system with cross-polarization in a wireless communication system to which the present invention is applicable.

[0044] Figure 13 FIG. 1 illustrates a transceiver unit model in a wireless communication system to which the present invention is applicable.

[0045] Figure 14 The figure shows a self-contained subframe structure to which the present invention is applicable.

[0046] Figure 15 Schematic diagram of the hybrid beamforming structure from the perspective of the TXRU and physical antenna.

[0047] Figure 16 FIG. 4 is a schematic diagram of a beam scanning operation for synchronization signals and system information during DL transmission. FIG.

[0048] Figure 17 The figure shows a flat panel antenna array to which the present invention is applicable.

[0049] Figure 18 Candidate beam group patterns for L=2 in a 2D port layout to which the present invention is applicable are shown.

[0050] Figure 19Candidate beam group patterns for L=4 in a 2D port layout to which the present invention is applicable are shown.

[0051] Figure 20 is a view illustrating a non-uniform port array according to an exemplary embodiment of the present invention.

[0052] Figure 21 is a view illustrating a GoB for N1=4, O1=4, N2=2, and O2=4 according to an exemplary embodiment of the present invention.

[0053] Figure 22 is a view illustrating a window configuration method for N1=4, O1=4, N2=2, and O2=4 according to an exemplary embodiment of the present invention.

[0054] Figure 23 is a flowchart illustrating a method for a UE to report CSI according to an exemplary embodiment of the present invention.

[0055] Figure 24 A block diagram of a wireless communication system according to an exemplary embodiment of the present invention is illustrated. DETAILED DESCRIPTION

[0056] Some embodiments of the present invention will be described in detail with reference to the accompanying drawings. Figure 1 The detailed description disclosed above is intended to describe some embodiments of the present invention and is not intended to describe the only embodiment of the present invention. The following detailed description includes further details in order to provide a complete understanding of the present invention. However, it will be understood by those skilled in the art that the present invention can be practiced without such further details.

[0057] In some cases, in order to avoid obscuring the concept of the present invention, known structures and devices may be omitted or may be shown in a block diagram format based on the core functions of each structure and device.

[0058] In this specification, a base station has the meaning of a terminal node of a network that communicates directly with a device through its base station. In this document, specific operations described as being performed by a base station may be performed by an upper node of the base station according to the circumstances. That is, it is obvious that in a network consisting of a plurality of network nodes including a base station, various operations performed for communication with a device may be performed by a base station or other network nodes other than the base station. A base station (BS) may be replaced by other terms such as a fixed station, a node B, an eNB (evolved node B), a base transceiver system (BTS), or an access point (AP). In addition, a device may be fixed or may have mobility and may be replaced by other terms such as a user equipment (UE), a mobile station (MS), a user terminal (UT), a mobile subscriber station (MSS), a subscriber station (SS), an advanced mobile station (AMS), a wireless terminal (WT), a machine type communication (MTC) device, a machine to machine (M2M) device, or a device to device (D2D) device.

[0059] Hereinafter, downlink (DL) refers to communication from an eNB to a UE, and uplink (UL) refers to communication from a UE to an eNB. In the DL, the transmitter may be part of the eNB and the receiver may be part of the UE. In the UL, the transmitter may be part of the UE and the receiver may be part of the eNB.

[0060] Specific terms used in the following description have been provided to help understanding of the present invention, and the use of such specific terms may be changed in various forms without departing from the technical spirit of the present invention.

[0061] The following technologies can be used in various wireless access systems such as code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), orthogonal frequency division multiple access (OFDMA), single carrier frequency division multiple access (SC-FDMA), and non-orthogonal multiple access (NOMA). CDMA can be implemented using radio technologies such as Universal Terrestrial Radio Access (UTRA) or CDMA2000. TDMA can be implemented using radio technologies such as Global System for Mobile Communications (GSM) / General Packet Radio Service (GPRS) / Enhanced Data Rates for GSM Evolution (EDGE). OFDMA can be implemented using radio technologies such as IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802-20, or Evolved UTRA (E-UTRA). UTRA is part of the Universal Mobile Telecommunications System (UMTS). 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) is a part of Evolved UMTS (E-UMTS) that uses Evolved UMTS Terrestrial Radio Access (E-UTRA) and adopts OFDMA in the downlink and SC-FDMA in the uplink. LTE-Advanced (LTE-AA) is an evolution of 3GPP LTE.

[0062] The embodiments of the present invention may be supported by standard documents disclosed in at least one of IEEE 802, 3GPP, and 3GPP2, i.e., wireless access systems. In other words, steps or parts that fall within the embodiments of the present invention and are not described to clearly reveal the technical spirit of the present invention may be supported by these documents. Furthermore, all terms disclosed in this document may be described in standard documents.

[0063] For a clearer description, 3GPP LTE / LTE-A is mainly described, but the technical features of the present invention are not limited thereto.

[0064] General systems to which the present invention can be applied

[0065] Figure 1 Shown is a structure of a radio frame in a wireless communication system to which embodiments of the present invention can be applied.

[0066] 3GPP LTE / LTE-A supports a radio frame structure type 1, which may be applied to frequency division duplex (FDD), and a radio frame structure type 2, which may be applied to time division duplex (TDD).

[0067] The size of a radio frame in the time domain is expressed as a multiple of a time unit of T_s = 1 / (15000*2048). UL and DL transmissions comprise a radio frame of duration T_f = 307200*T_s = 10 ms.

[0068] Figure 1 (a) illustrates a radio frame structure type 1. The type 1 radio frame can be applied to both full-duplex FDD and half-duplex FDD.

[0069] A radio frame consists of 10 subframes. A radio frame consists of 20 time slots of length T_slot = 15360 * T_s = 0.5 ms, and each time slot is indexed from 0 to 19. A subframe consists of two consecutive time slots in the time domain, and subframe i consists of time slot 2i and time slot 2i+1. The time required to transmit a subframe is called a transmission time interval (TTI). For example, the length of subframe i can be 1 ms, and the length of a time slot can be 0.5 ms.

[0070] In FDD, UL transmission and DL transmission are separated in the frequency domain. While there is no restriction in full-duplex FDD, a UE cannot transmit and receive simultaneously in half-duplex FDD operation.

[0071] A slot includes multiple orthogonal frequency division multiplexing (OFDM) symbols in the time domain and multiple resource blocks (RBs) in the frequency domain. In 3GPP LTE, because OFDMA is used in the downlink, an OFDM symbol is used to represent one symbol period. An OFDM symbol may be referred to as an SC-FDMA symbol or symbol period. An RB is a resource allocation unit and includes multiple consecutive subcarriers in a slot.

[0072] Figure 1 (b) shows frame structure type 2.

[0073] A type 2 radio frame includes two half-frames each having a length of 153600*T_s=5 ms, and each half-frame includes five subframes each having a length of 30720*T_s=1 ms.

[0074] In the frame structure type 2 of the TDD system, uplink-downlink configuration is a rule indicating whether uplink and downlink are allocated (or reserved) to all subframes.

[0075] Table 1 shows the uplink-downlink configuration.

[0076] [Table 1]

[0077]

[0078] Referring to Table 1, in each subframe of a radio frame, 'D' represents a subframe for DL ​​transmission, 'U' represents a subframe for UL transmission, and 'S' represents a special subframe including three types of fields: a downlink pilot time slot (DwPTS), a guard period (GP), and an uplink pilot time slot (UpPTS).

[0079] DwPTS is used for initial cell search, synchronization, or channel estimation in the UE. UpPTS is used for channel estimation in the eNB and for synchronizing the UE's UL transmission. GP is the duration used to remove interference that occurs in the UL due to multipath delay of DL signals between the UL and DL.

[0080] Each subframe i includes a time slot 2i and a time slot 2i+1 of T_slot=15360*T_s=0.5 ms.

[0081] UL-DL configurations may be classified into 7 types, and for each configuration, the positions and / or numbers of DL subframes, special subframes, and UL subframes are different.

[0082] Table 2 shows the configuration of the special subframe (length of DwPTS / GP / UpPTS).

[0083] [Table 2]

[0084]

[0085] according to Figure 1 The structure of the radio frame of the example is just an example, and the number of subcarriers included in the radio frame, the number of slots included in the subframe, and the number of OFDM symbols included in the slot may be changed in various ways.

[0086] Figure 2 is a diagram illustrating a resource grid of one downlink slot in a wireless communication system to which embodiments of the present invention can be applied.

[0087] refer to Figure 2 , one downlink slot includes multiple OFDM symbols in the time domain. For exemplary purposes only, it is described herein that one downlink slot includes 7 OFDM symbols and one resource block includes 12 subcarriers in the frequency domain, and the present invention is not limited thereto.

[0088] Each element on the resource grid is called a resource element, and one resource block includes 12×7 resource elements. The number N of resource blocks included in a downlink slot DL Depends on the downlink transmission bandwidth.

[0089] The structure of the uplink timeslot may be the same as that of the downlink timeslot.

[0090] Figure 3 Shown is a structure of a downlink subframe in a wireless communication system to which embodiments of the present invention can be applied.

[0091] refer to Figure 3 , up to three OFDM symbols located in the first part of the first time slot of a subframe correspond to a control region in which control channels are allocated, and the remaining OFDM symbols correspond to a data region in which the physical downlink shared channel (PDSCH) is allocated. Downlink control channels used in 3GPP LTE include, for example, the Physical Control Format Indicator Channel (PCFICH), the Physical Downlink Control Channel (PDCCH), and the Physical Hybrid ARQ Indicator Channel (PHICH).

[0092] The PCFICH is transmitted in the first OFDM symbol of a subframe and carries information about the number of OFDM symbols used to transmit control channels in the subframe (i.e., the size of the control region). The PHICH is a response channel for the uplink and carries positive acknowledgement (ACK) / negative acknowledgement (NACK) signals for hybrid automatic repeat request (HARQ). The control information transmitted in the PDCCH is called downlink control information (DCI). DCI includes uplink resource allocation information, downlink resource allocation information, or uplink transmit (Tx) power control commands for a specific UE group.

[0093] Figure 4 Shown is a structure of an uplink subframe in a wireless communication system to which embodiments of the present invention can be applied.

[0094] refer to Figure 4 , the uplink subframe can be divided into a control region and a data region in the frequency domain. The physical uplink control channel (PUCCH) carrying uplink control information is allocated to the control region. The physical uplink shared channel (PUSCH) carrying user data is allocated to the data region. To maintain the single-carrier characteristic, a UE does not transmit both the PUCCH and PUSCH simultaneously.

[0095] Within a subframe, a resource block (RB) pair is allocated to the PUCCH for one UE. The RBs belonging to the RB pair occupy different subcarriers in each of the two time slots. This is called frequency hopping of the RB pair allocated to the PUCCH at the time slot boundary.

[0096] Multiple Input Multiple Output (MIMO)

[0097] MIMO technology uses multiple transmit (Tx) and receive (Rx) antennas, rather than the single transmit and receive antennas typically used to date. In other words, MIMO technology uses multiple input / output antennas at the transmit or receive end of a wireless communication system to increase capacity or enhance performance. Hereinafter, MIMO is referred to as "multiple input / output antennas."

[0098] More specifically, MIMO antenna technology does not rely on a single antenna path to receive a single overall message, but rather completes the overall data by combining multiple data blocks received via multiple antennas. Therefore, MIMO antenna technology can increase the data transfer rate within a specific system range, and can also increase the system range at a specific data transfer rate.

[0099] It is expected that efficient multiple-input / output antenna technology will be used because next-generation mobile communications require higher data transmission rates than existing mobile communications. Under such circumstances, MIMO communication technology is a next-generation mobile communication technology that can be widely used in mobile communication terminals and relay nodes, and has attracted public attention as a technology that can overcome the limitations of mobile communication transmission rates caused by the expansion of data communications.

[0100] Meanwhile, multiple input / output (MIMO) technology, among various transmission efficiency improvement technologies being developed, has attracted widespread attention as a method that can significantly improve communication capacity and transmission / reception performance even without additional frequency allocation or power increase.

[0101] Figure 5 The configuration of a known MIMO communication system is shown.

[0102] refer to Figure 5 , if the number of transmit (Tx) antennas is increased to N_T and the number of receive (Rx) antennas is simultaneously increased to N_R, then unlike the case where multiple antennas are used only in the transmitter or receiver, the theoretical channel transmission capacity increases in proportion to the number of antennas. Therefore, the transmission rate can be increased and the frequency efficiency can be significantly improved. In this case, the transmission rate according to the increase in channel transmission capacity can theoretically increase by the value obtained by multiplying the following rate increment R_i by the maximum transmission rate R_o when using one antenna.

[0103] [Equation 1]

[0104] R i =min(N T ,N R )

[0105] That is, for example, in a MIMO communication system using four transmission antennas and four reception antennas, a transfer rate four times higher can theoretically be obtained compared to a single antenna system.

[0106] Such multi-input / output antenna technologies can be divided into spatial diversity, which uses symbols that have passed through various channel paths to increase transmission reliability, and spatial multiplexing, which uses multiple transmit antennas to simultaneously transmit multiple data symbols to increase the transmission rate. Furthermore, research is currently underway into combining these two methods to optimally leverage their respective advantages.

[0107] Each of these methods will be described in more detail below.

[0108] First, spatial diversity methods include space-time block code families and space-time Trelis code families, which utilize both diversity gain and coding gain. Generally, Trelis code families offer superior bit error rate improvement performance and code generation freedom, while space-time block code families offer lower computational complexity. Such spatial diversity gain can correspond to an amount corresponding to the product (N_T × N_R) of the number of transmit antennas (N_T) and the number of receive antennas (N_R).

[0109] Secondly, the spatial multiplexing scheme is a method of sending different data streams in the transmitting antenna. In this case, in the receiver, mutual interference occurs between the data sent simultaneously by the transmitter. The receiver removes the interference using an appropriate signal processing scheme and receives the data. The noise removal method used in this case may include a maximum likelihood detection (MLD) receiver, a zero forcing (ZF) receiver, a minimum mean square error (MMSE) receiver, a diagonal Bell Labs layered space-time code (D-BLAST) and a vertical Bell Labs layered space-time code (V-BLAST). In particular, if the transmitting end is able to know the channel information, a singular value decomposition (SVD) method can be used.

[0110] Third, there are methods that use a combination of spatial diversity and spatial multiplexing. If only spatial diversity gain is to be obtained, the performance improvement gain due to the increase in diversity difference gradually saturates. If only spatial multiplexing gain is used, the transmission reliability in the radio channel is degraded. Methods for solving this problem and obtaining both gains have been studied and may include dual space-time transmit diversity (dual STTD) methods and space-time bit-interleaved coded modulation (STBICM).

[0111] In order to describe the communication method in the multiple-input / output antenna system, as described above, in more detail, the communication method can be expressed as follows via mathematical modeling.

[0112] First, if Figure 5As shown, it is assumed that there are N_T transmitting antennas and N_R receiving antennas.

[0113] First, a transmission signal is described below: If there are N_T transmitting antennas as described above, the maximum number of items of information that can be transmitted is N_T, which can be represented using the following vector.

[0114] [Equation 2]

[0115]

[0116] Meanwhile, the transmission power may be different in each piece of transmission information s_1, s_2, ..., s_NT. In this case, if the respective transmission powers are P_1, P_2, ..., P_NT, the transmission information with the controlled transmission power may be represented using the following vector.

[0117] [Equation 3]

[0118]

[0119] In addition, the transmission information with controlled transmission power in Equation 3 can be expressed as follows using a diagonal matrix P of transmission power.

[0120] [Equation 4]

[0121]

[0122] Meanwhile, the information vector with controlled transmit power in Equation 4 is multiplied by the weighting matrix W, thereby forming N_T transmission signals x_1, x_2, ..., x_NT that are actually transmitted. In this case, the weighting matrix is ​​used to appropriately distribute the transmission information to the antennas according to the transmission channel conditions. The following equation can be expressed using the transmission signals x_1, x_2, ..., x_NT.

[0123] [Equation 5]

[0124]

[0125] In this case, w_ij represents the weight between the i-th transmit antenna and the j-th transmission information, and W is an expression of the matrix of weights. Such a matrix W is called a weighting matrix or a precoding matrix.

[0126] Meanwhile, the transmission signal x such as described above may be considered for use in a case where spatial diversity is used and in a case where spatial multiplexing is used.

[0127] If spatial multiplexing is used, all elements of the information vector s have different values ​​because different signals are multiplexed and transmitted. In contrast, if spatial diversity is used, all elements of the information vector s have the same value because the same signal is transmitted through several channel paths.

[0128] A method of mixing spatial multiplexing and spatial diversity can be considered. In other words, for example, the same signal can be transmitted through three transmit antennas using spatial diversity, and the remaining different signals can be spatially multiplexed and transmitted.

[0129] If there are N_R receiving antennas, the received signals y_1, y_2, ..., y_NR of the respective antennas are expressed using a vector y as follows.

[0130] [Equation 6]

[0131]

[0132] Meanwhile, if the channels in a multi-input / output antenna communication system are modeled, the channels can be classified by transmit / receive antenna index. The channel from transmit antenna j through receive antenna i is represented as h_ij. In this case, it should be noted that in the order of the index of h_ij, the index of the receive antenna appears first, and the index of the transmit antenna appears later.

[0133] Several channels can be grouped and expressed in vector and matrix form. For example, the vector expression is described below.

[0134] Figure 6 is a diagram illustrating channels from multiple transmit antennas to a single receive antenna.

[0135] like Figure 6 As shown, the channel from a total of N_T transmitting antennas to receiving antenna i can be expressed as follows.

[0136] [Equation 7]

[0137]

[0138] Furthermore, if all channels from N_T transmit antennas to N_R receive antennas are expressed by a matrix, such as Equation 7, they can be expressed as follows.

[0139] [Equation 8]

[0140]

[0141] Meanwhile, additive white Gaussian noise (AWGN) is added to the actual channel after the actual channel undergoes the channel matrix H. Therefore, AWGN n_1, n_2, ..., n_NR respectively added to N_R receiving antennas are expressed as follows using vectors.

[0142] [Equation 9]

[0143]

[0144] A transmission signal, a reception signal, a channel, and AWGN in a multiple-input / output antenna communication system can be expressed as having the following relationship through modeling of the transmission signal, reception signal, channel, and AWGN such as described above.

[0145] [Equation 10]

[0146]

[0147] Meanwhile, the number of rows and columns of the channel matrix H, which indicates the state of the channel, is determined by the number of transmit / receive antennas. In the channel matrix H, as described above, the number of rows becomes equal to the number of receive antennas N_R, and the number of columns becomes equal to the number of transmit antennas N_T. That is, the channel matrix H becomes an N_R×N_T matrix.

[0148] Typically, the rank of a matrix is ​​defined as the minimum number of independent rows or columns. Therefore, the rank of a matrix is ​​no greater than the number of rows or columns. In terms of representation, for example, the rank of the channel matrix H is limited as follows.

[0149] [Equation 11]

[0150] rank(H)≤min(N T ,N R )

[0151] Furthermore, if the matrix undergoes eigenvalue decomposition, the rank can be defined as the number of eigenvalues ​​that belong to the eigenvalue and are not 0. Similarly, if the rank undergoes singular value decomposition (SVD), it can be defined as the number of singular values ​​other than 0. Therefore, the physical meaning of the rank in the channel matrix can be said to be the maximum number of different information that can be transmitted in a given channel.

[0152] In this specification, the term "rank" used for MIMO transmission refers to the number of paths through which signals can be independently transmitted at a specific time point and on specific frequency resources. The term "number of layers" refers to the number of signal streams transmitted through each path. Generally, unless otherwise specified, the term "rank" has the same meaning as the number of layers, as the transmitting end transmits the number of layers corresponding to the number of ranks used in signal transmission.

[0153] Reference Signal (RS)

[0154] In wireless communication systems, because data is transmitted over a radio channel, signals may be distorted during transmission. In order for the receiving end to accurately receive distorted signals, it is necessary to use channel information to correct the distortion of the received signal. To detect channel information, a method is typically used that utilizes the degree of distortion of the signal transmission method when transmitting a signal known to both the transmitting and receiving sides over a channel, as well as a signal known to both the transmitting and receiving sides. This signal is called a pilot signal or reference signal (RS).

[0155] Furthermore, most mobile communication systems recently use methods that improve data transmission and reception efficiency by employing multiple transmit and receive antennas, rather than the single transmit and receive antennas typically used. When using multiple input and output antennas to transmit and receive data, it is necessary to monitor the channel conditions between the transmit and receive antennas in order to accurately receive signals. Therefore, each transmit antenna must have a separate reference signal.

[0156] In a mobile communication system, RSs can be basically divided into two types according to their purpose. There are RSs with the purpose of obtaining channel state information and RSs for data demodulation. The former has the purpose of obtaining channel state information in the downlink by the UE. Therefore, the corresponding RS must be sent in a wideband, and the UE must be able to receive and measure the RS, although the UE does not receive downlink data in a specific subframe. In addition, the former is also used for radio resource management (RRM) measurements such as handover. The latter is an RS sent together with the corresponding resources when the eNB sends a downlink. The UE can perform channel estimation by receiving the corresponding RS and can therefore demodulate the data. The corresponding RS must be sent in the area where the data is transmitted.

[0157] The downlink RS includes a common RS (CRS) for acquiring and measuring information about channel states shared by all UEs in a cell (such as handover) and a dedicated RS (DRS) for data demodulation only for a specific UE. Such RSs can be used to provide information for demodulation and channel measurement. In other words, the DRS is only used for data demodulation, while the CRS is used for both channel information acquisition and data demodulation.

[0158] The receiving side (i.e., the UE) measures the channel state based on the CRS and feeds back indicators related to the channel quality, such as the channel quality indicator (CQI), precoding matrix index (PMI), and / or rank indicator (RI), to the transmitting side (i.e., the eNB). CRS is also called cell-specific RS. In contrast, a reference signal related to the feedback of channel state information (CSI) can be defined as CSI-RS.

[0159] If the data on the PDSCH needs to be demodulated, a DRS can be sent through a resource element. The UE can receive information about whether a DRS exists through a higher layer, and the DRS is only valid when the corresponding PDSCH has been mapped. DRS can also be called UE-specific RS or demodulation RS (DMRS).

[0160] Figure 7 Illustration shows reference signal patterns mapped to downlink resource block pairs in a wireless communication system to which the present invention can be applied.

[0161] refer to Figure 7 , a downlink resource block pair (i.e., a unit mapped with a reference signal) can be represented in the form of one subframe in the time domain × 12 subcarriers in the frequency domain. That is, on the time axis (x-axis), one resource block pair is represented in the normal cyclic prefix (CP) ( Figure 7 a) has a length of 14 OFDM symbols, while the extended cyclic prefix (CP) ( Figure 7 b) has a length of 12 OFDM symbols. In the resource block grid, the resource elements (REs) indicated by "0", "1", "2", and "3" respectively mean the positions of the CRSs of the antenna port indexes "0", "1", "2", and "3", and the RE indicated by "D" means the position of the DRS.

[0162] If the eNB uses a single transmit antenna, the reference signal for the single antenna port is arranged.

[0163] If the eNB uses two transmit antennas, the reference signals for the two transmit antenna ports are arranged using a time division multiplexing (TDM) scheme and / or a frequency division multiplexing (FDM) scheme. That is, to distinguish the reference signals for the two antenna ports, different time resources and / or different frequency resources are allocated.

[0164] In addition, if the eNB uses four transmit antennas, the reference signals for the four transmit antenna ports are arranged using a TDM and / or FDM scheme. The channel information measured by the receiving side of the downlink signal (i.e., the UE) can be used to demodulate data sent using transmission schemes such as single transmit antenna transmission, transmit diversity, closed-loop spatial multiplexing, open-loop spatial multiplexing, or multi-user multiple input / output (MIMO) antennas.

[0165] If multiple-input multiple-output antennas are supported, when RSs are transmitted through a specific antenna port, the RSs are transmitted in the positions of resource elements specified according to the RS pattern, and not in the positions of resource elements specified for other antenna ports. In other words, RSs between different antennas do not overlap.

[0166] In the LTE-A system (i.e., an advanced and developed form of the LTE system), it is necessary to design the eNB to support up to eight transmit antennas in the downlink. Therefore, RSs for up to eight transmit antennas must also be supported. In the LTE system, downlink RSs are defined only for up to four antenna ports. Therefore, if the eNB has four to a maximum of eight downlink transmit antennas in the LTE-A system, RSs for these antenna ports must be additionally defined and designed. Regarding the RSs for up to eight transmit antenna ports, the aforementioned RSs for channel measurement and the aforementioned RSs for data demodulation must be designed.

[0167] An important factor that must be considered when designing LTE-A systems is backward compatibility. This means that LTE UEs must operate well even in LTE-A systems, which the system must support. From the perspective of RS transmission, in the time-frequency domain, where the CRS defined in LTE is transmitted across the full frequency band every subframe, RSs for up to eight transmit antenna ports must also be defined. In LTE-A systems, if the same method as existing LTE CRS is used to add RS patterns for up to eight transmit antennas across the full frequency band every subframe, RS overhead would increase excessively.

[0168] Therefore, the RS redesigned in the LTE-A system is basically divided into two types, which include an RS with a channel measurement purpose for selecting MCS or PMI (Channel State Information-RS or Channel State Indication-RS (CSI-RS)) and an RS for demodulation of data transmitted through eight transmit antennas (Data Demodulation-RS (DM-RS)).

[0169] The characteristic of the CSI-RS for channel measurement is that, unlike the existing CRS used for measurement (such as channel measurement and handover) and for data demodulation, it is designed to focus on the purpose of channel measurement. In addition, the CSI-RS can also be used for measurement purposes such as handover. Unlike the CRS, it is not necessary to transmit the CSI-RS every subframe because it is transmitted for the purpose of obtaining information about the channel state. In order to reduce the overhead of the CSI-RS, the CSI-RS is transmitted intermittently on the time axis.

[0170] In the LTE-A system, a maximum of eight transmit antennas are supported in the downlink of the eNB. In the LTE-A system, if the RS for up to eight transmit antennas is transmitted in the full band per subframe using the same method as the CRS in the existing LTE, the RS overhead will increase excessively. Therefore, in the LTE-A system, the RS has been divided into CSI-RS for CSI measurement purposes for selecting MCS or PMI and DM-RS for data demodulation, and thus two RSs have been added. CSI-RS can also be used for purposes such as RRM measurement, but has been designed for the main purpose of obtaining CSI. There is no need to transmit CSI-RS in every subframe because it is not used for data demodulation. Therefore, in order to reduce the overhead of CSI-RS, CSI-RS is transmitted intermittently on the time axis. That is, CSI-RS has a period corresponding to an integer multiple of one subframe and can be transmitted periodically or in a specific transmission pattern. In this case, the period or pattern of transmitting CSI-RS can be set by the eNB.

[0171] In order to measure CSI-RS, the UE must know information about the CSI-RS transmission subframe index, the time-frequency position of the CSI-RS resource element (RE) within the transmission subframe, and the CSI-RS sequence for each CSI-RS antenna port of the cell to which the UE belongs.

[0172] In the LTE-A system, the eNB must transmit CSI-RS for each of up to eight antenna ports. The resources used for CSI-RS transmission for different antenna ports must be orthogonal. When an eNB transmits CSI-RS for different antenna ports, it can allocate resources orthogonally using an FDM / TDM scheme by mapping the CSI-RS for each antenna port to different REs. Alternatively, the CSI-RS for different antenna ports can be transmitted using a CDM scheme that maps the CSI-RS to orthogonal codes.

[0173] When an eNB notifies a UE belonging to the eNB of CSI-RS information, it must first inform the UE of the time-frequency information to which the CSI-RS for each antenna port is mapped. Specifically, this information includes the subframe number or period for transmitting the CSI-RS, the subframe offset for transmitting the CSI-RS, the OFDM symbol number of the RE transmitting the CSI-RS for a specific antenna, the frequency spacing, and the offset or shift value of the RE on the frequency axis.

[0174] The CSI-RS is transmitted through one, two, four or eight antenna ports. The antenna ports used in this case are p=15, p=15, 16, p=15, ..., 18 and p=15, ..., 22, respectively. The CSI-RS can be defined for a subcarrier spacing Δf=15 kHz.

[0175] In a subframe configured for CSI-RS transmission, the CSI-RS sequence is mapped to a complex-valued modulation symbol a_k,l^(p) used as a reference symbol on each antenna port p as in Equation 12.

[0176] [Equation 12]

[0177]

[0178] In Equation 12, the conditions of (k', l') (where k' is a subcarrier index within a resource block and l' indicates an OFDM symbol index within a slot) and n_s are determined according to the CSI-RS configuration, such as Table 3 or Table 4.

[0179] Table 3 illustrates mapping of (k', l') from CSI-RS configuration in normal CP.

[0180] [Table 3]

[0181]

[0182] Table 4 illustrates mapping of (k', l') from CSI-RS configuration in the extended CP.

[0183] [Table 4]

[0184]

[0185] Referring to Tables 3 and 4, in the transmission of CSI-RS, in order to reduce inter-cell interference (ICI) in a multi-cell environment including a heterogeneous network (HetNet) environment, up to 32 different configurations (in the case of normal CP) or up to 28 different configurations (in the case of extended CP) are defined.

[0186] The CSI-RS configuration varies depending on the number of antenna ports and the CP within a cell, and adjacent cells may have the most different configurations. In addition, the CSI-RS configuration can be divided into the case where it is applied to both FDD and TDD frames and the case where it is applied to only TDD frames according to the frame structure.

[0187] (k', l') and n_s are determined according to the CSI-RS configuration based on Table 3 and Table 4, and the time-frequency resources for CSI-RS transmission are determined according to each CSI-RS antenna port.

[0188] Figure 8 : is a diagram illustrating resources to which reference signals are mapped in a wireless communication system to which the present invention can be applied. In particular, Figure 8 CSI-RS patterns in cases where the number of CSI-RS antenna ports is 1 or 2, 4, and 8 in a subframe to which normal CP is applied are illustrated.

[0189] Figure 8 (a) shows twenty types of CSI-RS configurations that can be used for CSI-RS transmission by one or two CSI-RS antenna ports, Figure 8 (b) shows ten types of CSI-RS configurations that can be used for four CSI-RS antenna ports, and Figure 8 (c) shows five types of CSI-RS configurations available for eight CSI-RS antenna ports.

[0190] As described above, the radio resources (ie, RE pairs) for transmitting the CSI-RS are determined according to each CSI-RS configuration.

[0191] If one or two antenna ports are configured for CSI-RS transmission for a specific cell, Figure 8 The CSI-RS is transmitted on a radio resource on a configured CSI-RS configuration among the twenty types of CSI-RS configurations shown in (a).

[0192] Likewise, when four antenna ports are configured for CSI-RS transmission for a specific cell, Figure 8 The CSI-RS is transmitted on the radio resources on the configured CSI-RS configuration among the ten types of CSI-RS configurations shown in (b). In addition, when eight antenna ports are configured for CSI-RS transmission for a specific cell, Figure 8 The CSI-RS is transmitted on a radio resource on a configured CSI-RS configuration among the five types of CSI-RS configurations shown in (c).

[0193] The CSI-RS for each antenna port is CDM (code division multiplexed) on the same radio resource for every two antenna ports (i.e., {15, 16}, {17, 18}, {19, 20}, and {21, 22}) and transmitted. For example, in the case of antenna ports 15 and 16, the CSI-RS complex symbols for the corresponding antenna ports 15 and 16 are the same, but are multiplied by different types of orthogonal codes (e.g., Walsh codes) and mapped to the same radio resource. The complex symbols of the CSI-RS for antenna port 15 are multiplied by [1, 1], and the complex symbols of the CSI-RS for antenna port 16 are multiplied by [1-1] and mapped to the same radio resource. The same is true for antenna ports {17, 18}, {19, 20}, and {21, 22}.

[0194] The UE can detect the CSI-RS for a specific antenna port by multiplying the transmitted symbol by the code by which it has been multiplied. That is, to detect the CSI-RS for antenna port 15, the transmitted symbol is multiplied by the multiplied code [1 1], and to detect the CSI-RS for antenna port 16, the transmitted symbol is multiplied by the multiplied code [1 -1].

[0195] refer to Figure 8 (a) to Figure 8 (c) In the case of the same CSI-RS configuration index, radio resources according to a CSI-RS configuration with a large number of antenna ports include radio resources with a small number of CSI-RS antenna ports. For example, in the case of CSI-RS configuration 0, radio resources for 8 antenna ports include radio resources for 4 antenna ports and radio resources for one or two antenna ports.

[0196] Figure 9 The diagram illustrates resources to which reference signals are mapped in a wireless communication system to which the present invention is applicable.

[0197] In particular, Figure 9 CSI-RS patterns are shown for cases where the number of CSI-RS antenna ports is 1 or 2, 4, and 8 in a subframe to which the extended CP is applied.

[0198] Figure 9 (a) shows 16 CSI-RS configurations, which can be used for CSI-RS transmission through 1 or 2 CSI-RS antenna ports. Figure 9 (b) shows 8 CSI-RS configurations that can be used for CSI-RS transmission through 4 CSI-RS antenna ports, and Figure 9 (c) shows 4 CSI-RS configurations that can be used for CSI-RS transmission through 8 CSI-RS antenna ports.

[0199] In this manner, radio resources (ie, RE pairs) used for CSI-RS transmission are determined according to each CSI-RS configuration.

[0200] When one or two antenna ports are set for CSI-RS transmission for a specific cell, Figure 9 The CSI-RS is transmitted on the radio resource using a set CSI-RS configuration among the 16 CSI-RS configurations shown in (a).

[0201] Similarly, when 4 antenna ports are set for CSI-RS transmission for a specific cell, Figure 9 In the example shown in (b), the CSI-RS configuration is configured to transmit CSI-RS on the radio resource. In addition, when 8 antenna ports are configured for CSI-RS transmission for a specific cell, according to Figure 9 The CSI-RS configuration is set among the four CSI-RS configurations shown in (c) to transmit CSI-RS on the radio resource. Multiple CSI-RS configurations can be used in a single cell. Only zero or one CSI-RS configuration can be used for non-zero power (NZP) CSI-RS, and zero or multiple CSI-RS configurations can be used for zero power (ZP) CSI-RS.

[0202] For each bit set to 1 in the zero power (ZP) CSI-RS ("ZeroPowerCSI-RS"), which is a 16-bit bitmap configured by a higher layer, the UE assumes zero transmit power in the REs corresponding to the four CSI-RS columns of Tables 3 and 4 (except when the REs overlap with REs assuming NZP CSI-RS configured by a higher layer). The most significant bit (MSB) corresponds to the lowest CSI-RS configuration index, and the next bit in the bitmap corresponds sequentially to the next CSI-RS configuration index.

[0203] The CSI-RS is transmitted only in downlink slots satisfying the conditions of (n_s mod 2) in Tables 3 and 4 and subframes satisfying the CSI-RS subframe configuration.

[0204] In the case of frame structure type 2 (TDD), CSI-RS is not transmitted in special subframes, synchronization signals (SS), subframes that conflict with PBCH or System Information Block Type 1 (SIB1) message transmissions, or subframes configured for paging message transmissions.

[0205] In addition, the RS for transmitting the CSI-RS for any antenna port belonging to the antenna port set S (S = {15}, S = {15, 16}, S = {17, 18}, S = {19, 20} or S = {21, 22}) is not used for PDSCH transmission or CSI-RS transmission for another antenna port.

[0206] The time-frequency resources used for CSI-RS transmission cannot be used for data transmission. Therefore, data throughput decreases as CSI-RS overhead increases. Taking this into account, the CSI-RS is not configured to be transmitted every subframe, but rather to be transmitted in each transmission period corresponding to multiple subframes. In this case, the CSI-RS transmission overhead can be significantly reduced compared to the case where the CSI-RS is transmitted every subframe.

[0207] Table 5 shows a subframe period (hereinafter referred to as a 'CSI transmission period') T_CSI-RS and a subframe offset Δ_CSI-RS for CSI-RS transmission.

[0208] Table 5 illustrates the CSI-RS subframe configuration.

[0209] [Table 5]

[0210]

[0211] Referring to Table 5, the CSI-RS period TCSI-RS and the subframe offset ΔCSI-RS are determined according to the CSI-RS subframe configuration ICSI-RS.

[0212] The CSI-RS subframe configuration in Table 5 may be set to one of the aforementioned 'SubframeConfig' field and the 'zeroTxPowerSubframeConfig' field. The CSI-RS subframe configuration may be set separately for NZP CSI-RS and ZP CSI-RS.

[0213] A subframe including a CSI-RS satisfies Equation 13.

[0214] [Equation 13]

[0215]

[0216] In Equation 13, TCSI-RS indicates CSI-RS periodicity, ΔCSI-RS indicates a subframe offset value, nf denotes a system frame number, and ns denotes a slot number.

[0217] When a UE with transmission mode 9 is configured for a serving cell, a single CSI-RS resource configuration may be configured for the UE. When a UE with transmission mode 10 is configured for a serving cell, one or more CSI-RS resource configurations may be configured for the UE.

[0218] CSI-RS configuration

[0219] In the current LTE standard, CSI-RS configuration parameters include antennaPortsCount, subframeConfig, resourceConfig, etc. These parameters indicate the number of antenna ports used to transmit CSI-RS, the period and offset of the subframe used to transmit CSI-RS, and the RE position (frequency and OFDM symbol index) transmitted in the corresponding subframe. Specifically, when the base station delivers a specific CSI-RS configuration to the UE, it delivers the following parameters / information.

[0220] -antennaPortsCount: A parameter indicating the number of antenna ports used to transmit CSI reference signals (e.g., 1 CSI-RS port, 2 CSI-RS ports, 4 CSI-RS ports, 8 CSI-RS ports, etc.).

[0221] -resourceConfig: Parameters associated with the location of resources allocated for CSI-RS.

[0222] -subframeConfig: Parameters associated with the period and offset of the subframe used to transmit the CSI-RS.

[0223] -pC: Regarding the UE assumption of the reference PDSCH transmit power for CSI feedback CSI-RS, Pc is the assumed ratio of PDSCH EPRE to CSI-RS EPRE when the UE derives CSI feedback and takes a value in the range of [-8, 15] with a 1 dB step size.

[0224] -zeroTxPowerResourceConfigList: Parameters associated with zero-power CSI-RS configuration.

[0225] -zeroTxPowerSubframeConfig: Parameters associated with the period and offset of the subframe used to transmit zero-power CSI-RS.

[0226] Massive MIMO

[0227] A MIMO system with multiple antennas may be referred to as a massive MIMO system, and has attracted attention as a means for improving spectral efficiency, energy efficiency, and processing complexity.

[0228] Recently, a massive MIMO system has been discussed in order to meet the spectrum efficiency requirements of future mobile communication systems in 3GPP. Massive MIMO is also called full-dimensional MIMO (FD-MIMO).

[0229] LTE Release 12 and subsequent wireless communication systems consider introducing active antenna systems (AAS).

[0230] Unlike a conventional passive antenna system in which an amplifier capable of adjusting the phase and amplitude of a signal is separated from an antenna, the AAS is configured in such a way that each antenna includes an active element such as an amplifier.

[0231] AAS does not require additional cables, connectors, and hardware for connecting amplifiers and antennas, and is therefore highly energy-efficient and has low operating costs. In particular, AAS supports electronic beam steering for each antenna, and is therefore able to implement enhanced MIMO to form precise beam patterns taking into account beam direction and beam width or 3D beam patterns.

[0232] With the introduction of enhanced antenna systems such as AAS, massive MIMO with multiple input / output antennas and multi-dimensional antenna structures is also being considered. For example, by forming a 2D antenna array instead of a traditional linear antenna array, it is possible to form a 3D beam pattern using the active antenna of the AAS.

[0233] Figure 10 FIG. 1 shows a 2D AAS having 64 antenna elements in a wireless communication system to which the present invention is applicable.

[0234] Figure 10 The figure shows a common 2D antenna array. Consider the case where Nt = Nv·Nh antennas are arranged in a square, as shown in Figure 10 Here, Nh indicates the number of antenna columns in the horizontal direction, and Nv indicates the number of antenna rows in the vertical direction.

[0235] When using the above-mentioned 2D antenna array, it is possible to steer radio waves in both the vertical direction (elevation angle) and the horizontal direction (azimuth angle) to control the transmission beam in 3D space. This wavelength control mechanism can be called 3D beamforming.

[0236] Figure 11 FIG. 1 illustrates a system in which an eNB or a UE has multiple transmit / receive antennas capable of forming a 3D beam based on AAS in a wireless communication system to which the present invention is applicable.

[0237] Figure 11The above example is illustrated and a 3D MIMO system using a 2D antenna array (ie, 2D-AAS) is illustrated.

[0238] From the perspective of the transmission antenna, when using a 3D beam pattern, quasi-static or dynamic beamforming can be performed in the vertical and horizontal directions of the beam. For example, applications such as sector formation in the vertical direction can be considered.

[0239] From the perspective of receive antennas, when massive receive antennas are used to form receive beams, signal power can be increased based on the antenna array gain. Therefore, in the uplink, the eNB can receive signals transmitted from the UE using multiple antennas, and the UE can set its transmission power to a very low level, taking into account the gain of the massive receive antennas.

[0240] Figure 12 FIG. 1 shows a 2D antenna system with cross-polarization in a wireless communication system to which the present invention is applicable.

[0241] The 2D planar antenna array model considering polarization can be Figure 12 Schematized as shown.

[0242] Unlike traditional MIMO systems that use passive antennas, active antenna-based systems can dynamically control the gain of antenna elements by applying weights to active elements (e.g., amplifiers) attached to (or included in) each antenna element. Because the radiation pattern depends on the antenna arrangement, such as the number of antenna elements and antenna spacing, the antenna system can be modeled at the antenna element level.

[0243] like Figure 12 The antenna arrangement model shown in can be represented by (M, N, P), which corresponds to parameters characterizing the antenna arrangement structure.

[0244] M indicates the number of antenna elements with the same polarization in each column (ie, in the vertical direction) (ie, the number of antenna elements with +45° tilt in each column or the number of antenna elements with −45° tilt in each column).

[0245] N indicates the number of columns in the horizontal direction (ie, the number of antenna elements in the horizontal direction).

[0246] P indicates the dimension of polarization. Figure 11 In the case of cross-polarization shown in FIG, P=2. In the case of co-polarization, P=1.

[0247] Antenna ports can be mapped to physical antenna elements. Antenna ports can be defined by the reference signals associated with them. For example, antenna port 0 can be associated with the cell-specific reference signal (CRS), and antenna port 6 can be associated with the positioning reference signal (PRS) in LTE systems.

[0248] For example, antenna ports and physical antenna elements can be mapped one-to-one. This can correspond to the case where a single cross-polarized antenna element is used for downlink MIMO or downlink transmit diversity. For example, antenna port 0 can be mapped to a single physical antenna element, while antenna port 1 can be mapped to another physical antenna element. In this case, there are two downlink transmissions as far as the UE is concerned. One is associated with the reference signal of antenna port 0, and the other is associated with the reference signal of antenna port 1.

[0249] Alternatively, a single antenna port can be mapped to multiple physical antenna elements. This can correspond to the case where a single antenna port is used for beamforming. Beamforming enables downlink transmissions to be directed to a specific UE by using multiple physical antenna elements. This can typically be achieved using an antenna array consisting of multiple columns of cross-polarized antenna elements. In this case, as far as the UE is concerned, there is a single downlink transmission derived from a single antenna port. One is associated with the CRS of antenna port 0, and the other is associated with the CRS of antenna port 1.

[0250] That is, the antenna ports represent downlink transmissions on the UE side, rather than the actual downlink transmissions from the physical antenna elements in the eNB.

[0251] Alternatively, multiple antenna ports can be used for downlink transmissions, and each antenna port can be multiple physical antenna ports. This can correspond to the case where the antenna arrangement is used for downlink MIMO or downlink diversity. For example, antenna port 0 can be mapped to multiple physical antenna ports, and antenna port 1 can be mapped to multiple physical antenna ports. In this case, there are two downlink transmissions as far as the UE is concerned. One is associated with the reference signal of antenna port 0, and the other is associated with the reference signal of antenna port 1.

[0252] In FD-MIMO, MIMO precoding of data streams may undergo antenna port virtualization, transceiver unit (TXRU) virtualization, and antenna element patterning.

[0253] In antenna port virtualization, the streams on the antenna ports are precoded at the TXRU. In TXRU virtualization, the TXRU signals are precoded at the antenna elements. In the antenna element pattern, the signals radiated from the antenna elements can have a directional gain pattern.

[0254] In traditional transceiver modeling, a static one-to-one mapping between antenna ports and TXRUs is assumed, and the TXRU virtualization effect is integrated into the (TXRU) antenna pattern, including the effects of TXRU virtualization and antenna element patterns.

[0255] Antenna port virtualization can be performed by a frequency selective method. In LTE, antenna ports are defined together with reference signals (or pilots). For example, in order to transmit data precoded on an antenna port, DMRS is sent in the same bandwidth as the data signal, and the DMRS and the data signal are precoded by the same precoder (or the same TXRU virtualization precoding). For CSI measurement, CSI-RS is sent through multiple antenna ports. In CSI-RS transmission, the precoder that characterizes the mapping between the CSI-RS port and the TXRU can be designed as a characteristic matrix so that the UE can estimate the TXRU virtualization precoding matrix for the data precoding vector.

[0256] 1D TXRU virtualization and 2D TXRU virtualization are discussed as TXRU virtualization methods, which will be described below with reference to the accompanying drawings.

[0257] Figure 13 FIG. 1 illustrates a transceiver unit model in a wireless communication system to which the present invention is applicable.

[0258] In 1D TXRU virtualization, an M_TXRU TXRU is associated with M antenna elements in a single column antenna arrangement with the same polarization.

[0259] In 2D TXRU virtualization, the corresponding Figure 12 The TXRU model of the antenna arrangement model (M, N, P) can be expressed as (M_TXRU, N, P). Here, M_TXRU represents the number of 2D TXRs in the same column and the same polarization, and M_TXRU ≤ M. That is, the total number of TXRUs is M_TXRU × N × P.

[0260] According to the correlation between antenna elements and TXRU, the TXRU virtualization model can be divided into TXRU virtualization model option-1: Figure 13 Sub-array partitioning model and TXRU virtualization model option-2 shown in (a): Figure 13 The fully connected model shown in (b).

[0261] refer to Figure 13 (a), In the case of the subarray partitioning model, the antenna elements are divided into multiple antenna element groups, and each TXRU is connected to one of the groups.

[0262] refer to Figure 13(b) In the case of the fully connected model, multiple TXRU signals are combined and delivered to a single antenna element (or antenna element array).

[0263] exist Figure 13 Where q is the transmitted signal vector of the M co-polarized antenna elements in a single column, w is the wideband TXRU virtualization weight vector, W is the wideband TXRU virtualization weight matrix, and x is the signal vector of the M_TXRU TXRU.

[0264] Here, the mapping between antenna ports and TXRUs can be 1-to-1 or 1-to-many mapping.

[0265] Figure 13 The example of TXRU to antenna element mapping is shown, and the present invention is not limited thereto. The present invention can also be applied to the mapping between TXRU and antenna elements implemented in various ways in hardware.

[0266] Definition of CSI (Channel State Information)-Reference Signal (CSI-RS)

[0267] For a serving cell and UE configured in transmission mode 9, the UE can be configured with one CSI-RS resource configuration. For a serving cell and UE configured in transmission mode 10, the UE can be configured with one or more CSI-RS resource configurations. The following parameters that the UE shall assume non-zero transmit power for CSI-RS are configured via higher layer signaling for each CSI-RS resource configuration:

[0268] -CSI-RS resource configuration identifier (if the UE is configured in transmission mode 10)

[0269] -Number of CSI-RS ports.

[0270] -CSIRS configuration

[0271] -CSIRS subframe configuration I_(CSI-RS).

[0272] - UE assumes reference PDSCH transmit power (P_c) for CSI feedback (if the UE is configured in transmission mode 9).

[0273] - If the UE is configured in transmission mode 10, the UE assumes a reference PDSCH transmit power for CSI feedback (P_c) for each CSI process. If the CSI subframe sets C_(CSI, 0) and C_(CSI, 1) are configured by higher layers for a CSI process, P_c is configured for each CSI subframe set for the CSI process.

[0274] - Pseudo-random sequence generator parameters (n_ID)

[0275] -CDM type parameter: If the UE is configured with the higher-layer parameter CSI-Reporting-Type, the CSI-reporting-Type is set to "CLASS A" for the CSI process.

[0276] If the UE is configured in transmission mode 10, the higher layer parameter qcl-CRS-Info-r11 for QCL type B UE assumes CRS antenna ports and CSI-RS antenna ports with the following parameters:

[0277] -qcl-ScramblingIdentity-r11.

[0278] -crs-PortsCount-r11.

[0279] -mbsfn-SubframeConfigList-r11.

[0280] P_c is the assumed ratio of PDSCH EPRE to CSI-RS EPRE (energy per resource element) when the UE derives CSI feedback and takes a value in the range of [-8, 15] dB with a 1 dB step size, where PDSCH EPRE corresponds to the number of symbols of the ratio of PDSCH EPRE to cell-specific RS EPRE.

[0281] The UE should not expect CSI-RS and PMCH to be configured in the same subframe of the serving cell.

[0282] For a frame structure type 2 serving cell and 4 CRS ports, the UE is not expected to receive a CSIRS configuration index belonging to the set [20-31] for the normal CP case or the set [16-27] for the extended CP case.

[0283] The UE may assume that the CSI-RS antenna ports of the CSI-RS resource configuration are quasi-co-located (QCL) with respect to delay spread, Doppler spread, Doppler shift, average gain, and average delay.

[0284] A UE configured in transmission mode 10 and with QCL type B may assume that antenna ports 0-3 associated with qcl-CRS-Info-r11 corresponding to the CSI-RS resource configuration and antenna ports 15-22 corresponding to the CSI-RS resource configuration are quasi-co-located (QCL) with respect to Doppler shift and Doppler spread.

[0285] If the UE is configured in transmission mode 10 with the higher-layer parameter CSI-Reporting-Type, the CSI-Reporting-Type is set to "Type B", and the number of CSI-RS resources configured for the CSI process is 1 or more, and QCL Type B is configured, the UE is not expected to receive a CSI-RS resource configuration for a CSI process with a different value from the higher-layer parameter qcl-CRS-Info-r11.

[0286] In the subframe configured for CSI-RS transmission, the reference signal sequence should be mapped to the complex-valued modulation symbol used as the reference symbol on antenna port p The mapping depends on the higher-level parameter CDMType.

[0287] If the CDMType does not correspond to CDM4, it can be mapped according to the following equation 14:

[0288] [Equation 14]

[0289]

[0290] If CDMType corresponds to CDM4, mapping may be done according to the following Equation 15.

[0291] [Equation 15]

[0292]

[0293] where w p' (i) Determined by the following Table 6. Table 6 shows the sequence of CDM4 p' (i).

[0294] [Table 6]

[0295]

[0296] OFDM numerology

[0297] As more and more communication devices require greater communication capacity, improved mobile broadband communication is needed compared to existing RATs (radio access technologies). In addition, large-scale MTC (machine type communication) that provides various services by connecting many devices and objects is one of the main issues to be considered in next-generation communications. In addition, the design of communication systems that take into account services / UEs that are sensitive to reliability and latency in next-generation communications is under discussion. The introduction of a next-generation RAT that takes into account enhanced mobile broadband communication, large-scale MTC, and URLLC (ultra-reliable and low-latency communication) is under discussion. In the present invention, for simplicity, this technology is referred to as a new RAT.

[0298] The new RAT system uses an OFDM transmission scheme or a similar transmission scheme. Generally, it has the OFDM parameter set shown in Table 3 below.

[0299] [Table 3]

[0300] parameter value Subcarrier spacing (△f) 60kHz OFDM symbol length 16.33μs Cyclic prefix (CP) length 1.30μs / 1.17μs s System bandwidth 80MHz (Number of available subcarriers) 1200 Subframe length 0.25ms The number of OFDM symbols per subframe 14 symbols

[0301] Self-contained subframe structure

[0302] In order to minimize the latency of data transmission in a TDD system in the new fifth generation RAT, a self-contained subframe structure is considered in which a control channel and a data channel are time division multiplexed (TDM).

[0303] Figure 14 The figure shows a self-contained subframe structure to which the present invention is applicable.

[0304] exist Figure 14 In FIG, the shaded area represents a transmission area of ​​a physical channel PDCCH carrying DCI, and the black area represents a transmission area of ​​a physical channel PUCCH carrying UCI (uplink control information).

[0305] Here, DCI is control information sent by the eNB to the UE. DCI may include information about the cell configuration that the UE should be aware of, DL-specific information such as DL scheduling, and UL-specific information such as UL grants. UCI is control information sent by the UE to the eNB. UCI may include HARQ ACK / NACK reports for DL ​​data, CSI reports on DL channel status, and / or Scheduling Requests (SRs).

[0306] exist Figure 14 In the figure, areas without hatching or black can be used to transmit the physical channel PDSCH carrying downlink data, or can be used to transmit the physical channel PUSCH carrying uplink data. Due to the self-contained subframe structure, DL transmission and UL transmission can be performed sequentially within a subframe, allowing DL data to be sent and UPACK / NACK to be received within the subframe. As a result, the time spent on retransmitting data when a data transmission error occurs can be reduced, thereby minimizing the latency of the final data transmission.

[0307] In a self-contained subframe structure, the eNB and UE require a time interval for switching from transmission mode to reception mode, or vice versa. To this end, some OFDM symbols when switching from DL to UL in the subframe structure are set as a guard period (GP). This subframe type may be referred to as a "self-contained SF."

[0308] Analog beamforming

[0309] In millimeter waves (mmW), wavelengths are shortened, and therefore multiple antenna elements can be installed in the same area. For example, a total of 64 (8×8) antenna elements can be installed in a 5×5cm panel in a 2D array at intervals of 0.5λ (wavelength) in a 30GHz band with a wavelength of approximately 1cm. Therefore, in mmW, the use of multiple antenna elements to increase beamforming (BF) gain is considered to increase coverage or throughput.

[0310] If a transceiver unit (TXRU) is provided for each antenna element to enable adjustment of transmit power and phase, independent beamforming can be performed for each frequency resource. However, installing TXRUs in all approximately 100 antenna elements is not feasible in terms of cost. Therefore, a method of mapping multiple antenna elements to one TXRU using an analog phase shifter and adjusting the beam direction is considered. This analog beamforming method can form only one beam direction in the entire frequency band, and therefore frequency selective beamforming (BF) may not be performed, which is disadvantageous.

[0311] A hybrid BF with fewer than Q antenna elements, B TXRUs, can be considered as an intermediate form between digital BF and analog BF. In the case of a hybrid BF, the number of directions in which beams can be transmitted simultaneously is limited to B or less, depending on the method of connecting B TXRUs and Q antenna elements.

[0312] In addition, in the case of using multiple antennas in the new RAT system, hybrid beamforming, that is, a combination of digital beamforming and analog beamforming, is recommended. Here, analog beamforming (or RF beamforming) refers to performing precoding (or combining) at the RF end. In hybrid beamforming, both the baseband end and the RF end perform precoding (or combining), which has the benefit of achieving performance close to that of digital beamforming while reducing the number of RF chains and the number of D (digital) / A (analog) (or A / D) converters. For convenience, the hybrid beamforming structure can be represented by N transceiver units (TXRUs) and M physical antennas. Then, the digital beamforming of the L data layers sent by the transmitting end can be represented by an N×L matrix, and then the N converted digital signals are converted into analog signals by the TXRU, and then analog beamforming is applied to be represented by an M×N matrix.

[0313] Figure 15 This is a schematic diagram of the hybrid beamforming structure from the perspective of the TXRU and physical antenna. Figure 15 , the number of digital beams is L, and the number of analog beams is N.

[0314] The new RAT system is designed in such a way that the base station changes the analog beamforming for each symbol, thereby supporting more efficient beamforming for UEs located in a specific area. Figure 15 In the embodiment, when N specific TXRUs and M RF antennas are defined by a single antenna panel, the new RAT system can deploy multiple antenna panels that can individually apply hybrid beamforming.

[0315] When a base station uses multiple analog beams, each UE may require a different analog beam for its signal reception. Therefore, for synchronization signals, system information, and paging, beam scanning can be considered to change the multiple analog beams used by the base station in a specific subframe (SF) for each symbol to allow each UE to have a chance to receive.

[0316] Figure 16 FIG. 4 is a schematic diagram of a beam scanning operation for synchronization signals and system information during DL transmission. FIG.

[0317] exist Figure 16 In the IEEE 802.11 specification, the physical resource (or physical channel) for transmitting system information in the new RAT system by broadcasting is called xPBCH (Physical Broadcast Channel).

[0318] refer to Figure 16 , the simulated beams belonging to different antenna panels can be sent simultaneously within one symbol. In order to measure the channel of each simulated beam, such as Figure 16 As shown in [1], a method for introducing beamformed RS (BRS) is being discussed. This RS is transmitted using a single analog beam (corresponding to a specific antenna panel). BRS can be defined for multiple antenna ports, and each antenna port of the BRS can correspond to a single analog beam. In this case, unlike BRS, all analog beams in the analog beam group can be used to transmit a synchronization signal or xPBCH so that a certain UE can properly receive it.

[0319] RRM measurements in LTE

[0320] The LTE system supports RRM operations for power control, scheduling, cell search, cell reselection, handover, radio link or connection monitoring, connection establishment / reestablishment, etc. In this case, the serving cell can request the UE to use RRM measurement information for RRM operations. For example, the UE can measure cell search information, reference signal received power (RSRP), reference signal received quality (RSRQ), etc. of each cell, and can report the measurement results. Specifically, in the LTE system, the UE can receive "measConfig" from the serving cell as a higher layer signal for RRM measurement. In this case, the UE can measure RSRP or RSRQ according to the "measConfig" information. In this case, RSRP, RSRQ and RSSI according to the TS 36.214 document for the LTE system can be defined as follows:

[0321] [RSRP]

[0322] The reference signal received power (RSRP) is defined as the linear average of the power contributions (in [W]) of the resource elements carrying the cell-specific RS (CR) within the considered measurement frequency bandwidth. For RSRP determination, CRS R0 shall be used according to TS 36.211 [3]. If the UE can reliably detect that R1 is available, R1 may be used in addition to R0 to determine RSRP.

[0323] The reference point for RSRP should be the UE's antenna connector.

[0324] If the UE is using receiver diversity, the reported value should not be lower than the corresponding RSRP of any individual diversity branch.

[0325] [RSRQ]

[0326] Reference Signal Received Quality (RSRQ) is defined as the ratio N × RSRP / (E-UTRA carrier RSSI), where N is the number of RBs in the E-UTRA carrier RSSI measurement bandwidth. The measurements in the numerator and denominator should be made on the same set of resource blocks.

[0327] The E-UTRA carrier received signal strength indicator (RSSI) consists of the linear average of the total received power (in [W]) observed / measured by the UE over N resource blocks from all sources (including co-channel serving and non-serving cells) in the measurement bandwidth, including channel interference, thermal noise, etc., only in OFDM symbols containing reference symbols for antenna port 0. If higher layer signaling indicates certain subframes for performing RSRQ measurement, the RSSI is measured over all OFDM symbols in the indicated subframes.

[0328] The reference point for RSRQ shall be the UE's antenna connector.

[0329] If the UE is using receiver diversity, the reported value should not be lower than the corresponding RSRQ of any individual diversity branch.

[0330] [RSSI]

[0331] The received wideband power, including thermal noise and noise generated in the receiver, is within the bandwidth limited by the receiver pulse-shaping filter.

[0332] The reference point for the measurement shall be the UE's antenna connector.

[0333] If the UE is using receiver diversity, the reported value shall not be lower than the corresponding UTRA carrier RSSI for any individual receive antenna branch.

[0334] According to the above definition, a UE configured to operate in an LTE system can measure RSRP using an IE (Information Element) associated with the Allowed Measurement Bandwidth (AMB) transmitted in SIB3 (System Information Block Type 3) in the case of intra-frequency measurement, or can measure RSRP at a bandwidth selected from 6 RBs (Resource Blocks), 15 RBs, 25 RBs, 50 RBs, 75 RBs, and 100 RBs using the Allowed Measurement Bandwidth (AMB) transmitted in SIBS (System Information Block Type 5) in the case of inter-frequency measurement. Alternatively, if the information element (IE) is not present, a UE configured to operate in an LTE system can measure RSRP across the entire DL system frequency bandwidth as a default value. In this case, if the UE receives an Allowed Measurement Bandwidth, it can assume that the corresponding value is the maximum measurement bandwidth, allowing the UE to freely measure RSRP values ​​within the corresponding value. However, if the serving cell transmits an IE defined as WB (Wideband)-RSRQ and sets the Allowed Measurement Bandwidth to 50 RBs or higher, the UE must calculate the RSRP value for the entire Allowed Measurement Bandwidth. Meanwhile, the RSSI may be measured in a frequency bandwidth allocated to a receiver of the UE according to an RSSI bandwidth definition.

[0335] Figure 17 The figure shows a flat panel antenna array to which the present invention is applicable.

[0336] refer to Figure 17 , the flat panel antenna array consists of Mg panels in the horizontal domain and Ng panels in the vertical domain, and each panel can be composed of M columns and N rows. In particular, the panels used here are illustrated with respect to X-pol (cross-polarization) antennas. Figure 17 The total number of antenna elements in can be 2*M*N*Mg*Ng.

[0337] Port Layout

[0338] Codebooks can be defined as various types. In NR (New RAT), there are two main types of codebooks: Type 1 codebooks and Type 2 codebooks. In addition, each type can be further subdivided according to whether it is a codebook for a single panel or a codebook for multiple panels (e.g., Type 1 single / multi-panel codebooks and Type 2 single / multi-panel codebooks).

[0339] For a type 1 single-panel codebook, W1 may be defined by the following Equation 16. Here, W1 represents a first PMI having long-term, wideband, and beam selection properties.

[0340] [Equation 16]

[0341]

[0342] At least for rank 1 and rank 2, the number (L) of candidate DFT (Discrete Fourier Transform) beams in B (or Bi) in W1 can be 1, 2, 4 and / or 7. The value of L can be configured by the network (e.g., base station).

[0343] For L>1, L beams can be freely selected by the UE. Alternatively, at least one beam group pattern can be defined, and reference will be made to Figure 18 and Figure 19 This section describes an example of a beam group pattern. The beam group pattern can be configured by the network (e.g., base station). The UE can report the beam pattern. Alternatively, the gNB can freely select L beams.

[0344] The selection of L beams may be applied similarly or differently to rank 1 and rank 2. For L=1, W1 may be defined by the following equation 17:

[0345] [Equation 17]

[0346]

[0347] Figure 18 The figure shows candidate beam group patterns for L=2 in a 2D port layout suitable for the present invention. In this figure, the patterned squares represent L selected beams.

[0348] Figure 19 The figure shows candidate beam group patterns for L=4 in a 2D port layout suitable for the present invention. In this figure, the patterned squares represent L selected beams.

[0349] In a 1D port layout, the beam group pattern includes a row of beams for L>1, which are uniformly and / or non-uniformly separated by d. For L>1, d1 and d2 can support a single value or multiple values.

[0350] Proposal of codebook in NR

[0351] In wireless communication systems using panel array antennas, including those for new RATs, narrow beams are formed because beamforming using massive antennas is performed, and the implementation of panel antenna arrays can eliminate linear deltas between antenna ports. Consequently, the performance of DFT-based codebooks used in LTE, LTE-A, and other technologies can degrade. Therefore, the present invention proposes a codebook structure suitable for panel array antennas.

[0352] First, the 2D DFT beam to be applied to a 2D antenna array within one panel can be defined by Equation 18:

[0353] [Equation 18]

[0354]

[0355] Where m1 and m2 represent the indices of the 1D-DFT codebook in the first and second domains, respectively, N1 and N2 represent the number of antenna ports for each pol in the first and second dimensions in the panel, and o1 and o2 represent the oversampling factors of the first and second dimensions in the panel.

[0356] exist Figure 17 In the figure, for ease of explanation, M and N represent antenna elements (hereinafter, M is referred to as a first domain (horizontal) parameter, and N is referred to as a second domain (vertical) parameter). According to the result of performing antenna virtualization on multiple antenna elements according to a specific vector and then performing antenna element-to-port mapping, the number of ports in the first domain and the second domain are defined by N1 and N2, respectively. When N1' and N2' are defined as the number of ports per panel, the total number of antenna ports to be considered in the present invention (Ntot) is defined as P*Mg*Ng*N1'*N2', and P can be set to 2 in the case of an X-pol antenna and to 1 in the case of a co-pol antenna.

[0357] Figure 20 is a view illustrating a non-uniform array according to an exemplary embodiment of the present invention.

[0358] refer to Figure 20 , vertical virtualization on a panel array with 32 elements per panel (i.e., M=4, N=2, P=2) results in P=2, N1'=4, N2'=1, Mg=2, Ng=2, for a total of 32 ports. Although antenna ports can correspond to antenna elements through antenna virtualization, for ease of explanation, the antenna ports in the present invention are generally referred to as "antenna ports" after virtualization of a single antenna element or multiple antenna elements. Antenna port information for beamforming (e.g., {N1, N2, O1 and O2} and / or {Mg, Ng, N1', N2', O1 and / or O2}) can be signaled through higher layer signaling or agreed in advance between the UE and the network.

[0359] The Ntot value may vary, but should conform to a codebook structure that is generally applicable to the antenna ports supported in the LTE system, such as 2, 4, 8, 12, 16, 20, 24, 28, and 32 ports. To this end, the present invention considers a multi-level codebook structure, and an example of three levels is shown in the following equation 19:

[0360] [Equation 19]

[0361] W=W1*W2*W3

[0362] In the dual-stage codebook structure used in LTE and LTE-A, a specific codebook matrix can be replaced by W1 (first PMI) or W2 (second PMI).

[0363] The 3GPP Release 13 codebook follows the dual structure of the Release 10 and Release 12 codebooks. That is, the final codebook is formed by multiplying W1 and W2, where W1 has long-term, wideband, and beam group selection properties, and W2 has short-term, subband and beam selection, and in-phase properties.

[0364] The difference from the Release 10 and Release 12 codebooks is that, because the antenna port layout to be considered includes two dimensions, the beams in the codebook are described as the Kronecker product of the vertical beam and the horizontal beam. The 3GPP Release 13 1-2 codebook can be expressed by the following equation 20:

[0365] [Equation 20]

[0366] W=W1W2

[0367]

[0368] where W^(1) represents the final form of the rank-1 codebook, and W^(2) represents the final form of the rank-2 codebook.

[0369] Here, N1 and N2 are the number of antenna ports per polarization in the first and second dimensions, and o1 and o2 are the oversampling factors in the first and second dimensions.

[0370] m1 and m2 represent the method of selecting DFT vectors in the horizontal and vertical (or first and second) domains. A specific W1 (i.e., first PMI) 2D beam group (i.e., codebook configurations 1 to 4) can be created by m1 (for rank 2, m1 and m1') and m2 (for rank 2, m2 and m2'). The subscript n indicates in-phase.

[0371] That is, the 3GPP Release 13 codebook can be viewed as a two-dimensional extension of the Release 10 8Tx (8-port transmit) codebook using Kroneckerk products.

[0372] Proposal 1) Simulate codebook

[0373] This proposal proposes a method of reporting CSI information for emulating beamforming using a codebook.

[0374] In an embodiment, one of the multiple stages of Equation 19 (eg, W1) performs a function / role of codeword selection corresponding to Tx / Rx analog beamforming, or an analog codebook may be generated by a single codebook matrix.

[0375] In analog beamforming, the analog codebook can be configured by the weight vector used for TXRU virtualization. In FD-MIMO, using a 2D subarray model, it can be configured by the following equation 21:

[0376] [Equation 21]

[0377]

[0378] where dv and dH are the spacing between each antenna element, λ is the carrier frequency, K is the number of antenna elements per TXRU in the N1 domain, L is the number of antenna elements per TXRU in the N2 domain, O_1TXRU and O_2TXRU are the oversampling factors of the 1-DFT beam formed by the elements in each domain of the TXRU, the length of wo is given by K=M / N1′, and the length of vi is given by L=N / N2′. θ etilt，o are specific direction angles in the N1 and N2 domains, respectively, and if N1 is the horizontal domain and N2 is the vertical domain, they can be expressed by the scan angle and the tilt angle.

[0379] Therefore, the final form of the Tx analog beam can be determined as in Equation 22:

[0380] [Equation 22]

[0381]

[0382] Equation 22 corresponds to when the antenna element indexing for virtualization is first performed in the N2 direction. If the antenna element indexing for virtualization is first performed in the N2 direction, Equation 22 can be transformed into the following Equation 23:

[0383] [Equation 23]

[0384]

[0385] As described above, the simulated beam can be steered in 2D, or can be steered in 1D direction only by using vectors for horizontal or vertical virtualization. In the present invention, for ease of explanation, exemplary embodiments will be described with respect to, but not limited to, a 2D simulated beam based on Equation 22.

[0386] Each vector in Equation 21 can be expressed in the same manner by the DFT beam in Equation 18 through a mathematical relationship. For example, by expressing each vector by tilt, Equation 21 can be transformed into the following Equation 24.

[0387] [Equation 24]

[0388]

[0389] Using Equation 24, Equation 21 can be expressed as follows in Equation 25:

[0390] [Equation 25]

[0391]

[0392] where k = 0, ..., o_(K-1), l = 0, ..., o_(L-1). The maximum size of the simulated codebook can then be expressed by multiplying L*O_1TXRU and K*O_2TXRU. In the simulated codebook, the resulting tilt and scan angles can be obtained by uniformly configuring all azimuth and zenith angles (e.g., Equation 26, and the above example assumes that the zenith angle ranges from -pi to pi and the azimuth angle ranges from -pi / 3 to pi / 3) and by uniformly dividing the boundaries of the simulated beams by the number of simulated beams as in θ1≤θ tilt ≤θ2 and In this case, the base station may notify the UE of the number of simulated beams used and / or the boundary value of the angle of the simulated beam through RRC.

[0393] [Equation 26]

[0394]

[0395] The analog codebooks explained above for antenna virtualization can be divided into two types of codebooks:

[0396] Select codebook

[0397] Simulation codebook based on NP (non-precoded) CSI-RS

[0398] In the following, a selection codebook and a simulation codebook based on NP CSI-RS will be proposed.

[0399] In the analog beam selection codebook, a specific N_A analog beamforming beam (for example, the N_A value can be set to L*O_1TXRU*K*O_2TXRU or set / defined as a specific value notified to the UE by the base station) can be mapped to the N_A CSI-RS port (or a specific port for analog beamforming), and the UE can report the (selected) PMI using the port selection codebook.

[0400] The UE may report multiple beams pre-agreed with the base station (or indicated by the base station), including the best beam, the first and second best beams, or the best and worst beams. To this end, the base station may indicate information such as K, O_1TXRU, L, O_2TXRU, etc. or N_A values ​​to the UE through high-layer signaling, or may pre-agreed with the UE on this. The tilt angle or scanning angle primarily used for the UE may be limited depending on the channel environment of the UE. Therefore, in order to reduce the overhead of analog beam scanning, the base station may notify the UE of the number of analog beams used for beam scanning and / or the number of analog beams reported by high-layer signaling or pre-agreed with the UE on this.

[0401] When using a selection codebook, a single analog beam is mapped to a single antenna port and transmitted, and the UE configures a selection vector using the codebook and reports it to the base station. That is, in this case, the codebook is configured by the analog beam selection vector, and the codeword is as shown in Equation 27, and the UE reports the i index of Equation 27 to the base station.

[0402] [Equation 27]

[0403]

[0404] Using Equation 27, the optimal Tx analog beam reporting codebook can be expressed as Equation 28:

[0405] [Equation 28]

[0406]

[0407] In this case, the number of feedback bits in the codebook is For example, for N_A=32, a total of 5 bits of feedback payload are required.

[0408] When the UE additionally reports the best beam or the worst beam, the UE may be redefined and used as an indicator indicating the number of beams to be reported, or the RI in the LTE system may be redefined and used as an indicator indicating the number of beams. For example, when the first beam is selected as the best beam and the fourth beam is selected as the worst beam, the UE may report RI=2 and the RI corresponding to the RI obtained by applying the first beam and the fourth beam to Equation 28 to the base station. And / or, the UE may assume rank 1 restriction for each beam and report PMI with different period and / or offset. In this case, the number of feedback bits to be reported is And / or, the UE may use a codebook to indicate the range of the TX analog beam (e.g., for tilt ≤θ2 indicates the purpose of θ1 and θ2).

[0409] Although this embodiment has been illustrated with respect to vertical tilt / domain, the present invention is not limited thereto, and this codebook can be used to indicate horizontal tilt / domain or 2-D tilt / domain in which horizontal and vertical tilt / domain are used. In addition, in this embodiment, the UE can be understood / interpreted as providing information about analog codebook subset restrictions to the base station, and this can also be applied to digital codebooks.

[0410] In analog beam scanning, payload size may not be a big issue due to its long-term and wideband nature. Therefore, when using an analog beam selection codebook, the analog beams can be fed back in a linear combination, as shown in Equation 29, which can be considered for more accurate feedback.

[0411] [Equation 29]

[0412]

[0413] Among them S l is the set of the first beams participating in the beam combination, and ci is a complex value that can have a specific value and is determined by c i,j,k =a i,j exp(jφ i,k ) configuration combination coefficient. S can be pre-agreed between the base station and the UE l ,a i,j ,φ i,k At least some elements of S can be established, and the base station can indicate these elements to the UE through RRC. For example, if the total number of simulated beams used for Tx beam scanning is 4 and the number of beams participating in the combination is 2, S l ∈{(1,2),(1,3),(1,4),(2,3),(2,4),(3,4)}, a i,j ={1,0.5,0.25,0},φ i,k = {1, j, -1, -j}. In the above example, the number of required feedback bits is 3 + 2 + 2 = 7, and at least some feedback elements / contents can be jointly encoded and fed back / reported. Also, / or, each element / content can be fed back with different time periods and / or different feedback granularity / units (e.g., wideband (WB) / subband (SB)). Compared to using only the codebook selected by Equation 27, the combined codebook has the advantage of being able to implement an analog codebook with relatively higher granularity.

[0414] If the UE is located in an environment with a lot of interference (e.g., at a cell border), performance degradation may become severe due to interference from simulated beams transmitted from interfering TRPs (transmission reception points). In this case, the UE can measure the interference of the codebook and report the information to the base station at a reduced power level due to the interference together with the corresponding codeword / simultaneously (e.g., {0.5, 0.25, 0.125, 0}*P, where P is the transmit power).

[0415] The above embodiments have been described with respect to the Tx beam scanning operation of the base station. However, if the UE performs Rx beam scanning, the UE may report information about this to the base station so that the base station can learn the UE's UL beamforming information. That is, similar to Equation 25, the Rx analog beam can be expressed by Equation 30:

[0416] [Equation 30]

[0417]

[0418]

[0419] Where A and B represent the number of antenna elements in the first and second domains of the TXRU of the UE, and o 1r,TXRU and o 2r,TXRU represents the oversampling factors in the first and second domains of the analog DFT codebook.

[0420] The final 2D (or 1D) DFT beam can be expressed as the Kronecker product using Equation 30 or As shown in Equation 22.

[0421] In order to configure the UL codebook, the UE may additionally provide the base station with information about A, B, and o 1r,TXRU and o 2r,TXRU The UE may also provide feedback to the base station regarding the port index direction (i.e., or ) or pre-agreed with the base station about it. The total size of the Rx analog beamforming codebook (Nrx,tot) can be “A*B*o 1r,TXRU *o 2r,TXRU ”, and the UE’s Rx beamforming selection codebook can be expressed by Equation 31:

[0422] [Equation 31]

[0423]

[0424] Where N_(A,RX) represents the number of Rx beamforming.

[0425] exist For example, the Tx-Rx beam pair codebooks can be reported collectively or independently and can have different feedback periods, offsets, and / or feedback granularity / units (e.g., wideband / subband / partial band). Alternatively, the RI can be used so that the UE indicates Rx beamforming. For example, if RI=2 is reported, the base station can identify that Rx beamforming (as well as Tx beamforming) is reported and can be reported by The codebook is used to calculate each Tx-Rx beamforming.

[0426] In the above-mentioned analog beamforming selection codebook, the complexity increases linearly with the number of analog beamforming ports. That is, for N_A=128, the number of CSI-RS ports required for one resource is 128, and sending all of these many ports in each PRB pair may be inefficient. Therefore, CSI-Rs comb transmission can be considered, in which all analog beamforming CSI-RS ports are divided into N sub-port groups, and the N sub-port groups are mapped one-to-one to N PRB pairs, so that N sub-port groups are sent to each N PRB pair (i.e., the CSI-RS ports required for one resource are divided and sent across N PRB pairs). For example, for N_A=128 and N=4, the ports corresponding to the 0th to 31st beams can be transmitted in 0, 4, 8, ... PRB pairs (i.e., 4n PRB pairs (n=0, 1, 2...)), the ports corresponding to the 32nd to 63rd beams can be transmitted in 1, 5, 9, ... PRB pairs (i.e., 4n+1 PRB pairs (n=0, 1, 2...)), the ports corresponding to the 64th to 95th beams can be transmitted in 2, 6, 10, ... PRB pairs (i.e., 4n+2 PRB pairs (n=0, 1, 2...)), and the ports corresponding to the 96th to 127th beams can be transmitted in 3, 7, 11, ... PRB pairs (i.e., 4n+3 PRB pairs (n=0, 1, 2...)). Alternatively, the ports corresponding to / included in each subport group (32 ports in the above example) may be transmitted at different time offsets and / or periods for each subport group (and / or each port in each subport group).

[0427] In order to reduce the overhead in terms of UE reporting, port selection can be performed by using RI to indicate the above time offset and / or frequency offset. For example, if the best simulated beam is the 64th beam in the above example, the UE can report RI=3 and PMI=1 to the base station (W analog,Tx,1 =[e1]∈C 32×1 ).

[0428] The above-mentioned selection-based codebook may be used only for beam management purposes and may have a higher priority level than other CSI (e.g., i1 (first PMI), i2 (second PMI), RI, CQI, and / or CRI). In addition, if the beam gain is lower than or equal to a specific threshold, the UE may trigger CSI-RS port transmission for codebook selection or report a beam index different from the beam index reported immediately before the reference resource is received (e.g., the second best beam index).

[0429] Hereinafter, a simulation codebook based on NP CSI-RS will be described.

[0430] In beam scanning, as the number of beams increases (i.e., when K, L, o1, and o2 increase), a larger number of OFDM symbols for beam scanning and / or more CSI-RS ports are required, and the computational complexity of the UE increases significantly. If the total number of antenna elements or K*L is equal to the number of CSI-RS supported in NR, the UE can measure the channel and report the best analog beam and / or digital beam by using NP CSI-RS (i.e., through 1:1 element-to-port mapping).

[0431] In one example, Equation 32 may be configured as a final codebook by using Equation 19. In this case, when the UE reports the analog codebook, the analog codebook may be reported based on a multi-stage codebook (e.g., three stages as in Equation 32), and the analog codebook may be used as one component of the multi-stage codebook.

[0432] [Equation 32]

[0433]

[0434] The simulated codebook ( where N port is the number of digital ports) can be configured by equation 22 or 23, and W1 and W2 can be an LTE codebook or a digital codebook to be described later. In addition, for Because the same analog beam is applied to all digital ports, in this case, the UE can only give feedback / report the PMI of the representative analog beam for all ports. However, it should be noted that for more accurate CSI feedback, the UE can assume Feedback / reporting is performed using a different analog beam for each port. In this case, there may be a disadvantage that the number of feedback bits increases by N_ports compared to using the same analog beam for each port. However, PMI (i.e., Wa) feedback for analog beams has very long-term characteristics (e.g., integer multiples of digital W1 or RI), and from the perspective of the entire system, the increase in overhead may not be that significant.

[0435] Therefore, in order to use the codebook efficiently, the base station can send NP CSI-RS in K*L ports in the first CSI-RS resource according to the analog codebook feedback period of the UE, assuming that the same analog beam is applied to each port. In this case, the UE can report the best analog beam index to the base station, and through this, the base station can use the analog beamforming for the second CSI-RS resource (corresponding to the analog beam index reported by the UE) to send N_ports CSI-RS to the UE. The UE can give a report / feedback (i.e., digital codebook feedback) about RI, PMI and / or CQI for N_ports / corresponding to N_port to the base station. The above two resources (i.e., the first and second CSI-RS resources) can have different time periods and / or offsets. If a conflict occurs between the two resources, the resource used for analog beamforming (i.e., the resource used to determine the analog beam; the first CSI-RS resource in the above example) has a higher priority level.

[0436] Alternatively, in order to apply a codebook with high granularity, the base station can use K*L*N_ports NP CSI-RS ports in one resource to send CSI-RS, and the UE can report the best PMI, CQI and / or RI to the base station based on the CSI-RS.

[0437] Proposal 2) Digital Codebook

[0438] In the new RAT, the LTE codebook or the Class A codebook can be reused. This codebook has a two-stage structure, and examples of this structure include Release 10 8Tx, Release 12 4Tx, Release 13 12Tx, 16Tx, and Release 14 20-, 24-, 28-, and 32Tx codebooks. In the two-stage structure (i.e., W = W1 * W2), W1 is used to determine a specific number of beam groups with long-term / wideband characteristics, and W2 is used to select beams within the beam group determined as W1 with short-term / subband characteristics, and performs cophasing in the case of X-pol antennas.

[0439] Preferably, the codebook used in the new RAT is configured within one framework, and it is expected that configuring the codebook with configuration information such as parameters N1 and N2 for configuring the TX port and o1 and o2 for configuring the codebook will make it easy to maintain scalability and implement UE.

[0440] In the LTE 2-port codebook, rank 1 is configured by QPSK (Quadrature Phase Shift Keying) (indexes 0, 1, 2, and 3 in Table 4), and rank 2 is configured by QPSK (indexes 0, 1, and 2 in Table 4). However, if analog beams are applied to the ports to make the beams sharper, increasing the beam granularity may be better in terms of performance.

[0441] Therefore, the present invention proposes to configure rank 1 and rank 2 2-port codebooks using 8-PSK for co-phasing, as shown in Table 4, in order to increase the 2-port granularity.

[0442] [Table 4]

[0443]

[0444] And / or, the base station can configure the codebook bit field for the UE to set whether the final codebook is QPSK or 8-PSK. For example, if the UE is given a 2-bit field from the base station, the UE can use the codebooks with indices 0 to 3 in Table 4, and if the UE is given a 3-bit field from the base station, the UE can use the codebooks with indices 0 to 7 in Table 4. This can be used for purposes similar to codebook subset restriction. Although existing codebook subset restriction cannot reduce feedback bits, the method proposed above can reduce feedback bits, thereby reducing uplink overhead.

[0445] In another embodiment, if different analog beamforming is performed for each port and there are many antenna elements for virtualization that form a single analog beam to form a very sharp beam, the codebook performance improvement caused by the application of the digital codebook is not expected to be very high. In this case, it may be more efficient to apply different beams to the two ports and select only specific ports. In this case, a 2-port codebook configuration can be proposed as shown in Table 5.

[0446] [Table 5]

[0447]

[0448] In the proposal according to Table 5, no PMI feedback (ie, beam selection) is required for rank 2.

[0449] In another embodiment, a codebook having codewords with different magnitudes for different ports may be configured, and an example of this is shown in Equation 33:

[0450] [Equation 33]

[0451]

[0452] As illustrated in Equation 33, in a 2-port codebook, a port may have a specific magnitude equal to or smaller than another port. For example, in Equation 33, α = {1, 0.5, 0.25, 0}. If α is 1, the codebook has the characteristics of the codebook illustrated in Table 4, and if α is 0, the codebook is similar to the port selection codebook illustrated in Table 5. α can be applied to each broadband or partial band, and the reporting period is long-term. In Equation 33, φ corresponding to the phase n=exp(-j2πn / 4)forn=0,1,2,3,φ n =exp(-j2πn / 8)forn=0,1,...,7 can be set to QPSK or 8-PSK according to the range of n value.

[0453] Therefore, the base station may notify the UE of the value of α and / or φ through RRC. n The base station may specify specific information corresponding to the in-phase size or pre-agreed with the UE. Alternatively, the base station may signal / configure the bit fields of the amplitude and in-phase of the codebook for the UE individually or as a whole and set them for the UE. For example, if the bit field size of the amplitude is set to 1 bit, α = {1, 0.5} or α = {1, 0}, and the base station may set the in-phase and in-phase bit field size to 2 bits and notify the UE of information about the in-phase based on QPSK.

[0454] The above codebook assumes X-pol and is more suitable when the same simulated beam is configured / applied to each port. On the other hand, if a different simulated beam is configured / applied to each port, it may be unclear which port is given the better beam gain due to differences in beam gain. Therefore, a codebook with the structure shown in Equation 34 is proposed, which is a more general form of the proposed codebook. Based on this codebook, the power amplitude codebook for each port can be independently configured, thereby improving performance gains.

[0455] [Equation 34]

[0456]

[0457] In the codebook according to Equation 34, the parameters (α, β, φ n ) (and / or each parameter set and / or set size) may be set by RRC or pre-agreed between the base station and the UE. Alternatively, when the UE reports a port index with a relatively high gain of 1 bit to the base station to reduce the feedback bits of α, β, the base station may set the amplitude coefficient of the corresponding reporting port to "1". In this case, the UE only reports the amplitude coefficient information of the other port to the base station, and as a result, the feedback bits are reduced. For example, in the case where the UE gives feedback / report of a port index with a high gain as the second port, β is determined / set to 1, and α may be determined / set to a value reported by the UE to the base station within an amplitude set (e.g., α={1,0.5,0.25,0}) pre-agreed between the base station and the UE.

[0458] In order to apply the above-mentioned 2-port codebook to a unified framework of a two-stage structure, W1 (matrix) can be assumed to be a square matrix (I), and the codebook of Table 4 or Table 5 can be applied as W2 (matrix) (i.e., W=W1*W2=I*W2). In another method, the above-mentioned analog beam selection codebook can be used for W1, and W2 can be configured as in Tables 4 and 5, so that the codebook is defined / applied in the form of W=W1*W2=Wa*W2. For Wa, the aforementioned Tx analog codebook configuration can be used. In another method, Wa can be the N_(a, Tx) analog beam selection codebook. For example, for N_(a, Tx)=4, the selection codebook of Wa can be configured / defined as (e i ,e j )∈{(i,j)|(1,1),(2,2),(3,3),(4,4),(1,2),(1,3),(1,4),(2,3),(2,4),(3,4)} or a combination of some of them to adjust to the payload size - for example, (e i ,e j )∈{(i,j)|(1,1),(2,2),(3,3),(4,4),(1,2),(1,4),(2,3),(2,4)}, beam selection combination for LTE-A rank 2. Alternatively, for N_(a,Tx)=4, the selection codebook of Wa can be dedicated to Table 5, and by using different beams for different ports (in the above example, (e i ,e j )∈{(i,j)|(1,2),(1,3),(1,4),(2,3),(2,4),(3,4)}) to configure / define / send.

[0459] In another method for configuring a 2-port codebook, W2 in W=Wa*W2 can be configured by a linear combination codebook. For example, W2 can be configured as And c1 and c2 have complex values.

[0460] The base station can configure the UE which of the above codebooks will be used / applied by RRC.

[0461] Proposal 3) Panel-based codebook

[0462] One of the new features of the new RAT is to support multi-panel antenna arrays consisting of multiple antennas, such as Figure 20 In this case, as Figure 20 As shown in , unless the intervals between panels are set in such a manner that the intervals between all antenna elements are constant, the characteristics of the DFT codebook on which the existing LTE is based (ie, uniform increment) are not satisfied, resulting in performance degradation.

[0463] To solve this problem, the present invention proposes a method of performing compensation between each panel (Proposal 3-1) and / or a method of selecting a specific panel and configuring a digital codebook (Proposal 3-2).

[0464] 3-1) Compensation between panels

[0465] For ease of explanation, reference will be made to Figure 20 This embodiment is described.

[0466] exist Figure 20 In

[15] , ports can be configured for each panel through 4-element vertical antenna virtualization, with each panel (panels 1 to 4) including 8 ports, and thus a total of 32 digital ports can be configured. In eFD-MIMO, 32 ports are supported and a type A codebook can be used. In this case, the final codebook can be as shown in Equation 35:

[0467] [Equation 35]

[0468]

[0469] in, is a diagonal matrix and is used to perform codebook compensation control (i.e., compensation matrix / codebook), is W1 of the two-stage codebook in the LTE system, N_W1 corresponds to the number of beam groups of W1, and It is W2 of the two-stage codebook in the LTE system and is used to perform beam selection and phasing.

[0470] refer to Figure 20 , assuming (N1′=4, N2′=1) that the N2 direction is the vertical direction, Wc can be configured as in Equation 36:

[0471] [Equation 36]

[0472]

[0473] Where α, β, γ refer to the compensation terms / compensators / correctors (hereinafter referred to as "correctors") for panels 2, 3 and 4 (relative to panel 1) - for example, they may have specific complex values ​​such as QPSK {1, -1, j, -j} (and / or BPSK (binary phase shift keying)). These correctors can be used to compensate for phase and / or amplitude between panels, and the UE can send the correctors (e.g., α, β, γ) with a signal and provide a report / feedback to the base station via CSI (e.g., PMI in CSI). In this case, the UE can provide a report / feedback indicating whether the corrector is WB (wideband) and / or SB (subband) according to the RRC mode setting of the base station (e.g., mode 1 and 2) (i.e., the UE reports to the base station the correctors selected / derived / obtained for wideband and / or subband according to the mode setting (hereinafter described as "WB and / or SB panel correctors").

[0474] If γ can be expressed by a function of α and β due to the characteristics of a linear planar array (eg, γ=α*β), the UE may not give feedback on γ, and thus feedback overhead is reduced.

[0475] The above concept can be extended to specify a representative panel corrector value for each domain. In the above example, α can be specified as the vertical panel reference corrector, and β can be specified as the horizontal panel reference corrector, and the correctors for other panels can be expressed as functions of α and / or β. For example, if panel 5 exists in Figure 20 At the right side of panel 3, the phase corrector (compensation value) of panel 5 can be represented by a function of α, for example, as in f(a)=α 2 middle.

[0476] Although the corrector achieves maximum performance when feedback is given in SB and / or short-term periods, feedback overhead can be saved by giving reports / feedback in the same period as the W1 PMI or an integer multiple thereof. The configuration of the corrector matrix (compensation matrix) may affect the indexing method between all panels and ports. Therefore, the port indexing direction can be pre-agreed between the base station and the UE, or indicated to the UE via higher-layer signaling.

[0477] In another compensation method, compensation can be performed on ports in a panel or on subgroups of panels that maintains a linear delta between antenna ports. This can be expressed mathematically by Equation 37:

[0478] [Equation 37]

[0479]

[0480] in, It is defined as a W1 fat matrix configured by the number of ports set in one panel. In addition, W2 is a matrix that performs beam selection and phase matching for each panel. It has been described assuming rank 1 in the above example, but is not limited to this and can be expanded to a general W2 expression.

[0481] In the above examples and examples described later, for ease of explanation, the phase is described as being configured / reported identically for each panel, but it goes without saying that the phase may be configured / reported independently for each panel for performance improvement.

[0482] According to the method of Equation 37, the following two cases can be considered: i) and ii)

[0483] i) Such as Figure 20 As illustrated in , for example, an 8-port codebook is used for each panel, and a different W1 beam group is assumed for each panel. Therefore, the codebook granularity is significantly increased, thereby increasing the performance gain. However, compared with ii), the computational complexity and the number of feedback bits may increase in proportion to the number of panels. Since a representative W1 beam group is used for each panel, the advantages of reducing complexity and the number of feedback bits can be achieved. In this case, as in the case of the codebook of Equations 35 and / or 36, α, β, and γ used for panel compensation may have specific complex values, such as QPSK {1, -1, j, -j}, and their feedback period may be equal to the W1 PMI period or an integer multiple of the W1 PMI period.

[0484] exist Figure 20 In the example, according to the method of Equation 37, an 8-port codebook is assumed for each panel, and Equation 37 can also be applied to two 16-port panel subgroups consisting of panels 1 and 2 and panels 3 and 4, respectively. To this end, the UE can additionally report information about the panel subgroup to the base station. For example, if the base station notifies the UE of the number of panels in each panel subgroup and the number of panels in each panel subgroup in the horizontal or vertical direction, the UE can select a specific panel subgroup and report it to the base station. This sub-panel group (i.e., the sub-panel group reported by the UE) can be used for each digital codebook application, or can be used to indicate the purpose of a group to which the same analog beam is applied.

[0485] In the case where digital precoding is configured for each panel or each sub-panel group as in Equation 37, it can be expected that port indexing is preferentially performed on "ports with the same polarization within one panel" to facilitate codebook configuration.

[0486] 3-2) Panel / sub-panel group selection codebook

[0487] When different analog beamforming is performed / applied for each panel or for each sub-panel group, it is desirable to select the panel or sub-panel group corresponding to the best analog beam and give CSI feedback. To this end, the Wc matrix in Equation 37 can be modified to a selection matrix as shown in Equation 38 and used.

[0488] [Equation 38]

[0489]

[0490] in, If only one of p, α, β, and γ has a value of 1, single-panel selection is performed (requiring 2 bits of feedback in the above example), and if two or more of p, α, β, and γ have a value of 1, multi-panel selection is performed (requiring 4 bits of feedback in the above example). In the latter case, the UE can expect that the same simulated beam will be transmitted through the selected multiple panels.

[0491] Therefore, when the PMI for panel selection is reported to the base station, the base station will find out that the UE only uses the ports in the panel corresponding to the reported PMI, and will activate the corresponding port and deactivate other ports for the corresponding UE and use them for transmission of other UEs.

[0492] In the above example, assuming two panels are selected, a total of 2*2*N1'*N2' ports are activated. In this case, the UE can apply the digital codebook corresponding to the 2*2*N1'*N2' ports and provide PMI / CQI / RI reports. If a non-uniform port layout is configured through the UE's panel selection, a codebook combined with the above-mentioned inter-port compensation method can be applied / used to improve performance.

[0493] When describing this embodiment in conjunction with the capabilities of the base station, if the base station has good calibration between panels, the base station may be expected to perform digital beamforming using all 2*2*N1'*N2' ports. In contrast, if the base station does not have good calibration between panels, the base station may be expected to perform digital beamforming on 2*N1'*N2' ports, or NP*2*N1'*N2' ports corresponding to one panel, or specific NP panels. That is, the uncalibrated base station may instruct the UE to configure / apply a panel selection codebook in order to prevent digital beamforming by port aggregation between panels. Alternatively, if the UE has a sufficiently high gain by analog beamforming due to good geometry, digital beamforming may not be required, and the UE may select a preferred panel by using the panel selection codebook in order to reduce the complexity of the codebook calculation.

[0494] To ensure the normal operation of the aforementioned panel selection codebook, the base station may notify the UE of information regarding at least one of N1, N2, N1', and N2' via RRC, or pre-negotiate with the UE. Furthermore, the aforementioned codebooks may be used individually or in combination. In the latter case, for example, the analog beam selection codebook and the panel selection codebook may be used in combination. This example is applicable when different analog beams are applied to different panels.

[0495] 3-3) Panel / sub-panel combination codebook

[0496] When the panel linear combination codebook is configured by modifying the above-mentioned selection codebook, Wc can be configured by Equation 39.

[0497] [Equation 39]

[0498]

[0499] Referring to Equation 39, the length of a column in the dimension of the final codebook is set to 2*N1'*N2', and this can be understood / interpreted as analog and digital beamforming vectors having a length of 2*N1'*N2' for each port are combined.

[0500] In the above method, the values ​​of ρ, α, β and γ can be given by, for example, ρ = a a exp(jψ a ),α=a b exp(jψ b ),β=a c exp(jψ c ),γ=a d exp(jψ d ) expression. In this case, the amplitude components (a a ,a b ,a c ,a d ) and the phase component (ψ a ,ψ b ,ψ c ,ψ d ). For example, in a separate report, the UE may separately report the amplitude component as wideband (or partial band) / long term and the phase component as wideband / subband. Because the combination is done, the amplitude (a) may be set to one of the values ​​{1, 0.5, 0.25, 0}, and the phase (ψ) may be set to one of the values ​​QPSK {1, -1, j, -j}.

[0501] In order to save payload size, the number of combined panels can be limited to a specific number, and this number of panels can be signaled via RRC (or MAC (Media Access Control) CE (Control Element)) or can be pre-agreed between the UE and the base station. That is, in the above example, if it is assumed that the number of combined panels is 2, the UE can report a power index of 0 for the two least preferred panel indices, or two panels can be selected first at the front end of the panel combination codebook. That is, in the case of combining two of the four panel beams, if an index is assigned to each panel combination, such as {(1,2), (1,3), (1,4), (2,3), (2,4), (3,4)}, the UE can first report the specific index they selected to the base station, and then execute the panel combination codebook for the selected panel. In addition, reporting each value corresponding to all power combining coefficients may be inefficient. Therefore, the power combining coefficient of a specific panel (i.e., the panel with the highest beam gain or the default first panel) can be assumed / set to a specific value, and the UE can only report the power combining coefficients for other combined beams. For example, if it is assumed that the power of the first panel is "1", the UE may report the amplitude values ​​of α, β, and γ corresponding to another panel.

[0502] In the panel combination codebook proposed above, the base station can configure whether the UE will use the same codebook or phase compensation (WB and / or SB) for each panel or use a different codebook or phase combination for each panel.

[0503] In the case of a codebook with different beam groups for each polarization, if the above-mentioned panel compensation codebook is applied, the values ​​of ρ, α, β, and γ can be set / applied independently for each polarization. That is, ρ_1, α_1, β_1, and γ_1 for the first polarization and ρ_2, α_2, β_2, and γ_2 for the second polarization - that is, a total of 8 independent variables - can be used to perform panel compensation.

[0504] Similar to the method proposed above, a codebook for indicating / reporting the difference in compensation value between a panel and a reference panel may be considered / proposed.

[0505] The proposed compensation codebook can be widely applied as an SB panel correction codebook, as well as a WB panel correction codebook. This can increase the payload for SB CSI feedback, but a panel calibration codebook can be applied for each SB, which better reflects the frequency selectivity and thus leads to a very large improvement in performance. However, in order to address the problem of increased payload, the feedback granularity / unit / size / bitwidth of the SB panel corrector can be set / defined differently from the feedback granularity / unit / size / bitwidth of the WB panel corrector. In particular, in order to reduce feedback overhead, the feedback granularity / unit / size / bitwidth of the SB panel corrector can be set / defined to be smaller than the feedback granularity / unit / size / bitwidth of the WB panel corrector (i.e., the feedback granularity of the SB panel corrector is lower than the feedback granularity of the WB panel corrector). For example, the feedback granularity / unit / size / bitwidth of the WB panel corrector can be set to 2 bits (in units of QPSK), and the feedback granularity / unit / size / bitwidth of the SB panel corrector can be set to 1 bit (in units of BPSK).

[0506] In this case, the UE can give a recommendation / feedback to the base station on whether to use the WB and / or SB panel correction codebook. And / or, the base station can configure the UE whether to use the WB and / or SB panel correction codebook through RRC configuration. For example, the application of the WB panel correction codebook can be defined as a first mode, and the application of the WB and SB panel correction codebooks can be defined as a second mode, and the base station can indicate to the UE which mode to apply through specific RRC signaling (e.g., "CodebookMode"). When the first mode is set, the UE can report to the base station through CSI (particularly, PMI) the WB panel corrector selected / derived based on QPSK with a size of 2 bits. When the second mode is set, the UE can report to the base station through CSI (particularly, PMI) the WB panel corrector selected / derived based on QPSK with a size of 2 bits and the SB panel corrector selected / derived based on BPSK with a size of 1 bit. In the second mode, the WB panel corrector can be used to compensate for the entire in-phase, and the SB panel corrector can be used to compensate for the entire in-phase, and the SB panel corrector can be used to finely compensate for the in-phase.

[0507] Alternatively, whether to use the WB panel correction codebook or the SB panel correction codebook can be associated with the number of panels of the base station (=Mg*Ng). For example, for Mg*Ng=4, the number of beams of W1 in the digital codebook, N_W1, can be set / applied to 1, and for Mg*Ng=2, the number of beams of W1 in the digital codebook, N_W2, can be set / applied to 2 (N_W1=2).

[0508] Although the proposed codebook has been described with respect to DL, it is not limited thereto but may be easily and widely applied to UL codebook configuration.

[0509] Hereinafter, a type 1 codebook configuration assuming a single panel will be described.

[0510] First, the use of the same beam group will be described, which can be expressed as Equation 40:

[0511] [Equation 40]

[0512]

[0513] Where W1 performs beam grouping with WB / long-term characteristics in a two-stage codebook. In this case, And B can have L values ​​(eg, L=1, 2, 4, 7, ...). Although N_W1 has been used before to indicate the number of beam groups of W1, it is replaced with L hereinafter. Now, a description will be given of a case where the UE selects L beams.

[0514] The UE can freely and unambiguously indicate the L beams used for the base station in an explicit manner (e.g., in the form of a bitmap or by indicating a beam index). In this case, the number of bits required is L*N1*N2*O1*O2 or And there is a problem that the number of feedback bits increases with L and the number of Tx antenna ports increases. Therefore, the UE can freely select a beam within a specific GoB (grid of beams) as a way to reduce the number of feedback bits. Figure 21 Describe an example thereof.

[0515] Figure 21 is a view illustrating a GoB where N1=4, O1=4, N2=2, and O2=4 according to an exemplary embodiment of the present invention.

[0516] refer to Figure 21 When a 4x6 beam selection window is configured, the UE can freely select L-1 beams within this window. In this case, the UE can provide feedback on the position of the main / pilot beam 2101 and the 4x6 window size.

[0517] In another approach, Figure 22 The exemplary embodiments may be applied.

[0518] Figure 22 is a view illustrating a window configuration method for N1=4, O1=4, N2=2, and O2=4 according to an exemplary embodiment of the present invention.

[0519] refer to Figure 22, the entire GoB is divided into windows of a size recommended / feedback by the base station or UE, and the UE can give feedback about the index (position) of the window and / or information about the selection of L beams freely selected within the window. Figure 22 The figure shows that there are 8 4 by 6 windows. According to this configuration, adjacent windows may overlap. In this case, the base station can configure information about the position and / or size of the UE's window.

[0520] If the UE selects L beams for W1, as proposed above, a high feedback bit rate is required. Therefore, the feedback information (e.g., information about the selection of L beams) can be limited to using PUCCH reporting instead of PUSCH reporting.

[0521] Hereinafter, different diagonal matrices forming W1 (ie, different beam groups for each polarization) will be described, which can be expressed as Equation 41.

[0522] [Equation 41]

[0523]

[0524] Where B1≠B2, and B1 and B2 can have different sizes. L i The number of beams in the beam group with i-tilt is represented (e.g., i=1 for H tilt and i=2 for V tilt), and L1 and L2 can have different values ​​(e.g., L1=1 and L2=2). The base station can pre-negotiate the L1 and L2 values ​​with the UE, or these values ​​can be configured for the UE via higher layers (e.g., RRC or MAC CE). Alternatively, the UE can provide a recommendation / feedback to the base station regarding information on the L1 and L2 values.

[0525] Configuring W1 as described above has the advantage of applying the optimal codeword for each polarization, but has the disadvantage of significantly increasing the feedback overhead of W1. Therefore, exemplary embodiments for resolving these disadvantages will be proposed.

[0526] First, L1=L2 (ie, if the number of vertical beams and the number of horizontal beams are equal) will be described.

[0527] In this case, W2 for rank 1 codebook configuration will be proposed as in Equation 42.

[0528] [Equation 42]

[0529]

[0530] where i≠j,i,j∈{1,...,L} and φ n ={1,j,-1,-j} is defined, and ei Denotes a selection vector whose length is L and whose i-th element has a value of 1 and the other elements have a value of 0. In this case, i and j must be reported separately, and more feedback overhead is consumed for beam selection compared to when the same beam group is used. For this design, i, j, and the in-phase value can be reported as SB. To reduce the SB feedback overhead for beam selection, L1 = L2 = 1 should be satisfied. In this case, W2 can be set to

[0531] In another method, the UE may give a report / feedback on i11 and i12 of B1 and an additional report / feedback on the difference between B1 and B2. Here, i11 and i12 represent the first and second domain indices of the W1 PMI as in the LTE codebook. That is, the UE may provide feedback / report to the base station on how far B2 is spaced from the leading beam indices i11 and i12 of B1 in the first and second domains. For example, when the leading beam index (i11, i12) of B1 is (10, 2) and a value corresponding to (2.4) is additionally reported / feedback to the base station as the difference from B1, the base station may identify the leading beam (i11, i12) of B2 as (12, 6) and configure B2.

[0532] In this method of indicating the difference between B1 (index) and B2 (index), the difference can be agreed upon between the UE and the base station as a specific value for each domain, or the base station can configure the difference for the UE, or the UE can provide report / feedback to the base station. To reduce reporting / feedback overhead, the UE can only provide feedback on information related to a specific domain (e.g., the first domain or the second domain). In this case, the base station can configure the specific domain for the UE, or the UE can notify the base station of the specific domain.

[0533] The rank 2 codebook configuration can be expressed as in Equation 43:

[0534] [Equation 43]

[0535]

[0536] It can be seen from Equation 43 that the variables of i, j, k, and l should satisfy the following conditions to maintain the orthogonality of each layer in the rank-2 codebook.

[0537] 1.e i =e k ,e j =e l : In this case, the codebook indicates that the same beam is selected for each polarization when configuring the layer. When configuring the codebook, φ n Can be limited to, for example, φ n ={1,j}. In this case, the beam forming the codebook can be normalized to 1.

[0538] 2.{e i ≠e k},{e j ≠e l}: In this case, B1 of W1 should be configured for each polarization so that the beams selected by i and k are orthogonal to each other, and B2 of W1 should be configured for each polarization so that the beams selected by j and l are orthogonal to each other. That is, the beam groups of B1 and B2 of W1 should consist of beams that are orthogonal to each other. Alternatively, if there are some non-orthogonal beams, the codebook can be configured by pairing orthogonal beams in the above method. For example, for Assumptions are orthogonal to each other and are orthogonal to each other and can be paired according to the second method above and That is, a total of two pairings can be performed. In this method, φ n = the same phase of {1, j, -1, -j}.

[0539] Hereinafter, L1≠L2 will be discussed. In this case, the codebook can be configured by widely applying the method proposed above for L1=L2.

[0540] As a specific example of L1≠L2, we will first describe L1=1. In this case, the rank 1 configuration of W2 is as shown in Equation 44:

[0541] [Equation 44]

[0542]

[0543] In this case, beam selection and co-phasing are possible for the beam corresponding to one tilt. Therefore, the PMI can be determined / indicated independently for each polarization, thereby increasing the codebook granularity and improving performance. In this case, φ n ={1,j,-1,-j}.

[0544] When the codebook is designed as above, it can be established The codebook is configured in a manner (i.e., the L beams of B2 always include B1) to form a superset of LTE Class A codebook configuration 1. Alternatively, the UE may recommend information about B2 to the base station.

[0545] Similarly, the rank 2 codebook can be configured as in Equation 45:

[0546] [Equation 45]

[0547]

[0548] For ei =e j , you can use φ n ={1,j}, and for e i ≠e j , if the beams chosen as i and j are orthogonal to each other, then φ n ={1,j,-1,-j}. Alternatively, the same phase with the same granularity can be used for both cases.

[0549] In the above method, the WB phase of the beam used for B2 can be reported together with the B2 index. That is, it is possible to set And W1 can be configured as in Equation 46.

[0550] [Equation 46]

[0551]

[0552] For example, ψ n is the WB in-phase value, for example, ψ n ={1,j,-1,-j}. In this case, SB phase can be as in In order to save SB feedback bits, the UE can use 1 bit of phase (e.g., ) to report the 2nd level in phase.

[0553] The proposed method can be easily applied to B1=B2, and B1 and B2 of W1 can be configured / applied independently for each frequency band (or frequency band group).

[0554] As described above, similar to the method of using different beam sets for each polarization, the SB size can be reduced to increase the accuracy of the SB PMI. Once the SB size is reduced, the PMI per SB can be more accurate, but the feedback overhead increases. Therefore, the base station can configure for the UE whether to reduce the SB size and / or use a codebook with B1 ≠ B2.

[0555] A new codebook can be configured by combining the codebook designs proposed above.

[0556] Figure 23 1 is a flowchart illustrating a method for a UE to report CSI according to an exemplary embodiment of the present invention. With respect to this flowchart, the aforementioned embodiments / descriptions may be applied identically or similarly, and redundant explanations will be omitted.

[0557] First, the UE may measure a CSI-RS transmitted from a base station through a plurality of panels ( S2310 ).

[0558] Next, the UE may report the CSI generated based on the CSI-RS measurement to the base station CSI ( S2320 ).

[0559] In this case, if the UE reports the WB panel corrector and the SB panel corrector for multiple panels as CSI (according to the CSI setting of the base station), the WB panel corrector and the SB panel corrector can be reported with different bit widths. Here, the WB panel corrector can correspond to the beam / codebook phase corrector of each panel derived / determined / selected based on the measurement of the CSI-RS (resource) of the WB, and the SB panel corrector can correspond to the beam / codebook phase corrector of each panel derived / determined / selected based on the measurement of the CSI-RS (resource) of the SB (or each SB). That is, the WB panel corrector and the SB panel corrector can be used for phase correction between multiple panels. The number of panels can be set through high-layer signaling.

[0560] In particular, the bit width of the SB panel corrector can be shorter than the bit width of the WB panel corrector - for example, the bit width of the SB panel corrector can be set to 1 bit, and the bit width of the WB panel corrector can be set to 2 bits. Therefore, the WB panel corrector can be reported based on QPSK, and the SB panel corrector can be reported based on BPSK. If the UE reports only the WB panel corrector as CSI, the WB panel corrector can be reported with a bit width of 2 bits. Whether to report both the WB panel corrector and the SB panel corrector or only the WB panel corrector can be determined according to a mode set by the base station (for example, a mode set by RRC signaling). For example, if the base station indicates mode "1" to the UE, the UE can recognize that only the WB panel corrector is reported, and if the base station indicates mode "2", the UE can recognize that both the WB panel corrector and the SB panel corrector are reported.

[0561] When reported, the WB panel corrector and the SB panel corrector may be included in the PMI within the CSI. In addition, the WB panel corrector and the SB panel corrector may be reported independently for each of the multiple panels.

[0562] General equipment to which the present invention can be applied

[0563] Figure 24 A block diagram of a wireless communication system according to an exemplary embodiment of the present invention is illustrated.

[0564] refer to Figure 24 , the wireless communication system includes a base station (eNB) 2410 and a plurality of user equipments (UEs) 1420 located within the area of ​​the base station 2410 .

[0565] Base station 2410 includes a processor 2411, a memory 2412, and an RF (Radio Frequency) unit 2413. Processor 2411 performs the functions, processes, and / or methods described above. The layers of the wireless interface protocol can be executed by processor 2411. Memory 2412 is connected to processor 2411 and stores various types of information for driving processor 2411. RF unit 2413 is connected to processor 2411 and transmits and / or receives radio signals.

[0566] The UE 2420 includes a processor 2421, a memory 2422, and a radio frequency unit 2423. The processor 2421 performs the functions, processes, and / or methods described above. The layers of the radio interface protocol may be executed by the processor 2421. The memory 2422 is connected to the processor 2421 and stores various types of information for driving the processor 2421. The radio frequency unit 2423 is connected to the processor 2421 and transmits and / or receives radio signals.

[0567] The memories 2412 and 2422 may be located inside or outside the processors 2411 and 2421 and may be connected to the processors 2411 and 2421 through well-known means. In addition, the base station 2410 and / or the UE 2420 may have a single antenna or multiple antennas.

[0568] The above-described embodiments correspond to the combination of elements and features of the present invention in a prescribed form. Furthermore, unless explicitly mentioned, each element or feature may be considered to be selective. Each element or feature can be implemented in a form that cannot be combined with other elements or features. In addition, by partially combining elements and / or features together, embodiments of the present invention can be implemented. The series of operations described for each embodiment of the present invention can be modified. Some configurations or features of one embodiment can be included in another embodiment, or can replace the corresponding configurations or features of another embodiment. Furthermore, it is clearly understood that embodiments can be configured by combining claims that do not have a clear reference relationship in the appended claims, or can be included as new claims by amendment after submitting the application.

[0569] In this specification, “A and / or B” can be interpreted as “at least one of A and B”.

[0570] The embodiments of the present invention may be implemented by various means, such as hardware, firmware, software, and combinations thereof. In the case of hardware implementation, the embodiments of the present invention may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, and the like.

[0571] In the case of being implemented by firmware or software, the embodiments of the present invention may be implemented in the form of modules, procedures, or functions that perform the functions or operations described so far. The software code may be stored in a memory and driven by a processor. The memory may be located inside or outside the processor and may exchange data with the processor via various well-known means.

[0572] It will be understood by those skilled in the art that various modifications and variations can be made without departing from the essential features of the present invention. Therefore, the detailed description is not limited to the above-mentioned embodiments, but should be regarded as examples. The scope of the present invention should be determined by a reasonable interpretation of the appended claims, and all modifications within the scope of equivalents should be included within the scope of the present invention.

[0573] Mode for the Invention

[0574] Various embodiments of the present invention have been described in the best mode for carrying out the invention.

[0575] Industrial Applicability

[0576] Although the present invention has been described mainly with regard to an example of application to the 3GPP LTE / LTE-A system, it may be applied to various wireless communication systems other than the 3GPP LTE / LTE-A system.

Claims

1. A method for receiving channel state information (CSI) by a base station, the method comprising: Sending a configuration related to a multi-panel codebook to a user equipment UE, wherein the configuration includes mode information; Sending a channel state information reference signal CSI-RS to the UE; and receiving the CSI from the UE based on the pattern information, Wherein, based on the mode information being set to the first mode, the CSI includes a first panel phase compensation parameter for wideband WB, Wherein, based on the mode information being set to the second mode, the CSI includes the first panel phase compensation parameter for WB and the second panel phase compensation parameter for sub-band SB.

2. The method for receiving channel state information (CSI) by a base station according to claim 1, wherein: The first panel phase compensation parameters for WB and the second panel phase compensation parameters for SB are used to compensate for the phases of multiple panels of a multi-panel based base station.

3. The method for receiving channel state information (CSI) by a base station according to claim 2, wherein: Phase compensation is applied to the plurality of panels by compensating for a phase difference between a reference panel that is one of the plurality of panels and each panel except the reference panel.

4. The method for receiving channel state information (CSI) by a base station according to claim 3, wherein: The first panel phase compensation parameter for WB included in the CSI is configured to be based on 2 bits, and The second panel phase compensation parameter for SB included in the CSI is configured to be based on 1 bit.

5. The method for receiving channel state information (CSI) by a base station according to claim 1, wherein: The mode information is related to the number of the plurality of panels of the base station.

6. The method for receiving channel state information (CSI) by a base station according to claim 1, wherein: The first panel phase compensation parameter for WB and the second panel phase compensation parameter for SB are reported in a precoding matrix index PMI in the CSI.

7. The method for receiving channel state information (CSI) by a base station according to claim 1, wherein: The first panel phase compensation parameter for WB and the second panel phase compensation parameter for SB are reported independently at least for each of a plurality of panels or for each of a plurality of polarizations.

8. The method for receiving channel state information (CSI) by a base station according to claim 1, wherein: The first panel phase compensation parameters for WB provide a first compensation amount for overall in-phase, and the second panel phase compensation parameters for SB provide a second compensation amount for overall in-phase in addition to the first compensation amount.

9. The method for receiving channel state information (CSI) by a base station according to claim 1, wherein: Based on that only the first panel phase compensation parameter for WB is to be reported by being included in the CSI, the CSI includes a third phase compensation parameter used for phase compensation between cross-polarization antenna ports included in each of a plurality of panels, and The value of the third phase compensation parameter applied to each of the plurality of panels is the same.

10. A base station, comprising: Radio frequency RF unit; at least one processor, and At least one computer memory operatively connected to the at least one processor and storing therein instructions that, when executed by the at least one processor, perform operations comprising: Sending a configuration related to a multi-panel codebook to a user equipment UE, wherein the configuration includes mode information; Sending a channel state information reference signal CSI-RS to the UE; and receiving channel state information (CSI) from the UE based on the pattern information, Wherein, based on the mode information being set to the first mode, the CSI includes a first panel phase compensation parameter for wideband WB, Wherein, based on the mode information being set to the second mode, the CSI includes a first panel phase compensation parameter for WB and a second panel phase compensation parameter for sub-band SB.

11. An apparatus comprising one or more memories and one or more processors operatively coupled to the one or more memories, in, The one or more processors control the device to: Sending a configuration related to a multi-panel codebook to a user equipment UE, wherein the configuration includes mode information; Sending a channel state information reference signal CSI-RS to the UE; and receiving channel state information (CSI) from the UE based on the pattern information, Wherein, based on the mode information being set to the first mode, the CSI includes a first panel phase compensation parameter for wideband WB, Wherein, based on the mode information being set to the second mode, the CSI includes a first panel phase compensation parameter for WB and a second panel phase compensation parameter for sub-band SB.

12. One or more non-transitory computer-readable media storing one or more instructions, wherein the one or more instructions executable by one or more processors control a base station: Sending a configuration related to the multi-panel codebook to the user equipment UE, wherein: The configuration includes mode information; Sending a channel state information reference signal CSI-RS to the UE; as well as receiving channel state information (CSI) from the UE based on the pattern information, Wherein, based on the mode information being set to the first mode, the CSI includes a first panel phase compensation parameter for wideband WB, Wherein, based on the mode information being set to the second mode, the CSI includes a first panel phase compensation parameter for WB and a second panel phase compensation parameter for sub-band SB.

Images