KR20230066275A - Multi-antenna based precoding method and apparatus in wireless communication system - Google Patents

Abstract

The present disclosure relates to a 5G or 6G communication system for supporting a higher data rate after a 4G communication system such as LTE.
According to one embodiment of the present disclosure, a method of a network entity in a communication system is provided. Whether or not the method of the network entity performs scheduling to adjust a ratio between a plurality of first transmission points (TPs) included in a first group serving at least one terminal and a plurality of second TPs included in a second group determining; determining whether to switch at least one second TP among the plurality of second TPs to a first TP based on information associated with the at least one terminal when it is determined to perform the scheduling; and transmitting an indicator instructing the at least one second TP to operate as the first TP when it is determined to switch the at least one second TP to the first TP, wherein the first When operating as a TP, the network entity performs precoding on signals, and when operating as the second TP, the second TP performs precoding on signals. According to this, a more improved service can be provided in an environment where the computational complexity and front haul capacity that can be tolerated by the CPU controlling the plurality of TPs are limited.

Description

Multi-antenna based precoding method and apparatus in wireless communication system

The present disclosure relates to a wireless communication system, and more particularly, in a communication system considering function split into a central unit (CU) and a distributed unit (DU), a data transmission/reception path between a CU and a DU A precoding method considering the load of a fronthaul and the computational complexity of a CU controlling a DU, and a device capable of performing the same.

Looking back at the process of development through successive generations of wireless communication, technologies for human-targeted services, such as voice, multimedia, and data, have been developed. After the commercialization of 5G (5 ^th -generation) communication systems, it is expected that connected devices, which have been explosively increasing, will be connected to communication networks. Examples of objects connected to the network may include vehicles, robots, drones, home appliances, displays, smart sensors installed in various infrastructures, construction machinery, and factory equipment. Mobile devices are expected to evolve into various form factors such as augmented reality glasses, virtual reality headsets, and hologram devices. In the ^6th generation (6G) era, efforts are being made to develop an improved 6G communication system in order to provide various services by connecting hundreds of billions of devices and objects. For this reason, the 6G communication system is being called a Beyond 5G system.

In the 6G communication system expected to be realized around 2030, the maximum transmission speed is tera (i.e., 1,000 gigabytes) bps, and the wireless delay time is 100 microseconds (μsec). That is, the transmission speed in the 6G communication system compared to the 5G communication system is 50 times faster and the wireless delay time is reduced to 1/10.

To achieve these high data rates and ultra-low latency, 6G communication systems use terahertz bands (e.g., 95 GHz to 3 terahertz (3 THz) bands). An implementation in is being considered. In the terahertz band, it is expected that the importance of technology that can guarantee signal reach, that is, coverage, will increase due to more serious path loss and atmospheric absorption compared to the mmWave band introduced in 5G. As the main technologies to ensure coverage, RF (radio frequency) devices, antennas, new waveforms that are superior in terms of coverage than orthogonal frequency division multiplexing (OFDM), beamforming, and massive multiple- Multi-antenna transmission technologies such as input and multiple-output (MIMO), full dimensional MIMO (FD-MIMO), array antenna, and large scale antenna should be developed. In addition, new technologies such as metamaterial-based lenses and antennas, high-dimensional spatial multiplexing technology using orbital angular momentum (OAM), and reconfigurable intelligent surface (RIS) are being discussed to improve coverage of terahertz band signals.

In addition, in order to improve frequency efficiency and system network, in the 6G communication system, full duplex technology in which uplink and downlink simultaneously utilize the same frequency resource at the same time, satellite and HAPS (high-altitude platform stations), network structure innovation technology that supports mobile base stations and enables network operation optimization and automation, dynamic frequency through collision avoidance based on spectrum usage prediction AI-based communication technology that utilizes dynamic spectrum sharing technology and AI (artificial intelligence) from the design stage and internalizes end-to-end AI support functions to realize system optimization. Development of next-generation distributed computing technology, etc., which realizes a service of a complexity beyond that by utilizing ultra-high-performance communication and computing resources (mobile edge computing (MEC), cloud, etc.) is being carried out. In addition, through the design of a new protocol to be used in the 6G communication system, the implementation of a hardware-based security environment, the development of a mechanism for safe use of data, and the development of technology for maintaining privacy, connectivity between devices is further strengthened and networks are further strengthened. Attempts are ongoing to optimize, promote softwareization of network entities, and increase the openness of wireless communications.

Due to the research and development of these 6G communication systems, a new level of hyper-connected experience (the next hyper-connected experience) is expected to be possible. Specifically, it is expected that services such as truly immersive extended reality (XR), high-fidelity mobile hologram, and digital replica will be available through the 6G communication system. In addition, services such as remote surgery, industrial automation, and emergency response through security and reliability enhancement are provided through the 6G communication system, which can be applied in various fields such as industry, medical care, automobiles, and home appliances. It will be.

Keeping pace with the development of the communication system as described above, there is an increasing demand for a technology that provides more improved transmission and reception performance to user equipment (UE). As a solution to meet this demand, a structure of a communication system capable of transmitting and receiving data according to a coordinated multi-point (CoMP) method, which is a multi-cell cooperative communication technology, has been proposed. The CoMP method is a method in which a plurality of transmission points (TPs) cooperate to transmit and receive data with at least one UE, and through this, a UE located relatively far from the TP can have improved data transmission and reception performance. Efficient communication service can be supported. On the other hand, as the number of TPs cooperating to transmit data to the UE increases, the computational complexity of the central processing unit (CPU) (or network entity) controlling the TPs increases greatly, and the signal between the CPU and the TPs increases. There is a problem in that the load of the fronthaul, which is a transmission and reception passage, increases.

In a communication system composed of multiple TPs (for example, a cell-free massive multi-input multi-output (MIMO) system in which data transmission and reception by multiple TPs is possible and the boundary between cells is broken down), the CoMP method Accordingly, when uplink data or downlink data is transmitted/received between the UE and a plurality of TPs, problems such as increased calculation complexity of the CPU and increased load on the front haul, which is a data transmission/reception path between the CPU and the TPs, may occur. Therefore, it is necessary to devise a precoding method capable of solving this problem and a system structure to which the method can be applied.

In order to solve the foregoing problem, according to an embodiment of the present disclosure, a method of a network entity in a communication system is provided. The method of the network entity is for adjusting a ratio between a plurality of first transmission points (TPs) included in a first group serving at least one terminal and a plurality of second TPs included in a second group determining whether to perform scheduling; When it is determined to perform the scheduling, determining whether to switch at least one second TP among the plurality of second TPs in the second group to a first TP based on information associated with the at least one terminal doing; and transmitting an indicator instructing the at least one second TP to operate as the first TP when it is determined to switch the at least one second TP to the first TP, wherein the first TP When operating as , precoding of a signal is performed by the network entity, and when operating as a second TP, precoding of a signal is performed by the second TP.

Also, according to an embodiment of the present disclosure, a network entity of a communication system is provided. The network entity determines whether to perform scheduling for adjusting a ratio between a plurality of first TPs included in a first group serving a transceiver and at least one terminal and a plurality of second TPs included in a second group, and , When it is determined to perform the scheduling, whether to switch at least one second TP among the plurality of second TPs in the second group to a first TP based on information associated with the at least one terminal and, if it is determined to switch the at least one second TP to the first TP, transmitting an indicator instructing the at least one second TP to operate as the first TP through the transceiver unit Including, when operating as a first TP, the network entity performs precoding on signals, and when operating as a second TP, the second TP performs precoding on signals. .

According to various embodiments of the present disclosure, in a communication system composed of a plurality of TPs, a precoding method considering the computational complexity allowed to a CPU controlling the plurality of TPs and the capacity of a front haul, which is a signal transmission and reception path between the CPU and the TPs, is provided. . According to this, when transmitting and receiving uplink data or downlink data between a UE and a plurality of TPs according to the CoMP scheme, the UE can have optimal data transmission and reception performance.

Effects obtainable in the present disclosure are not limited to the effects mentioned above, and other effects not mentioned may be clearly understood by those skilled in the art from the description below. will be.

Effects obtainable in the present disclosure are not limited to the effects mentioned above, and other effects not mentioned may be clearly understood by those skilled in the art from the description below. will be.

The above and other objects, features and advantages of the present disclosure will become clearer through the following description of embodiments of the present disclosure with reference to the accompanying drawings.
1 is a diagram showing the structure of a cellular mobile communication system to which the present disclosure can be applied.
2 is a diagram illustrating a long term evolution (LTE) system structure of a 3GPP standard to which the present disclosure may be applied.
3 is a diagram illustrating a radio protocol structure in an LTE system of the 3GPP standard to which the present disclosure can be applied.
4 is a diagram illustrating a system structure of next-generation mobile communication of the 3GPP standard to which the present disclosure can be applied.
5 is a diagram illustrating a radio protocol structure in a next-generation mobile communication system of the 3GPP standard to which the present disclosure can be applied.
6 is a diagram illustrating the structure of a system to which a non-coordinated multi point (non-CoMP) scheme is applied according to an embodiment of the present disclosure.
7 is a diagram illustrating the structure of a system to which a Coordinated Multi Point (CoMP) method according to an embodiment of the present disclosure is applied.
8 is a diagram illustrating classification of CoMP schemes in an LTE system or a next-generation mobile communication system to which the present disclosure can be applied.
9 is a diagram illustrating coherent joint transmission (CJT) and non-coherent joint transmission (NCJT) schemes in an LTE system or a next-generation mobile communication system to which the present disclosure may be applied.
10 is a diagram illustrating a distributed antenna system in which a CJT scheme of CoMP according to an embodiment of the present disclosure can be implemented.
11 is a diagram illustrating an example of a common public radio interface (CPRI) between a remote radio header (RRH) (or radio unit (RU)) and a base band unit (BBU) according to an embodiment of the present disclosure.
12 is a diagram showing a model of a cell-free massive MIMO-based system to which the present disclosure can be applied.
13 is a diagram illustrating a sequence in which a CPU performs complexity-shared scheduling (CSS) according to an embodiment of the present disclosure.
14 is a diagram illustrating a location of a function separation point according to an embodiment of the present disclosure.
15A is a diagram illustrating a system structure in which a terminal transmits and receives uplink data with a TP instructed to operate as a central TP or a local TP according to a CSS execution result according to an embodiment of the present disclosure.
15B is a diagram illustrating a system structure in which a terminal transmits and receives downlink data with a TP instructed to operate as a central TP or a local TP according to a CSS execution result according to an embodiment of the present disclosure.
16 is a diagram illustrating a procedure for transmitting and receiving downlink data when multiple demodulation-reference signals (DM-RSs) are allocated to a central TP and a local TP determined according to a CSS execution result according to an embodiment of the present disclosure.
17 is a diagram illustrating a sequence in which a UE operates according to whether a coherent joint reception indicator (CJRI) is received according to an embodiment of the present disclosure.
18 is a diagram illustrating an example of a CSS algorithm that can be performed in a CPU according to an embodiment of the present disclosure.
19 is a diagram illustrating a CSS framework based on a TP with embedded enhanced CPRI (eCPRI) according to an embodiment of the present disclosure.
20 is a diagram illustrating a result of evaluating average user yield while increasing a ratio of central TPs when supporting two UEs per TP, when an embodiment of the present disclosure is applied.
21 is a diagram illustrating a result of evaluating average user yield while increasing a ratio of central TPs when supporting 4 UEs per TP, when an embodiment of the present disclosure is applied.
22 is a diagram illustrating a result of evaluating a 10% lower performance user yield while increasing a ratio of a central TP when supporting two UEs per TP when an embodiment of the present disclosure is applied.
23 is a diagram illustrating a result of evaluating a user yield of 10% lower performance while increasing a ratio of a central TP when supporting 4 UEs per TP, when an embodiment of the present disclosure is applied.
FIG. 24 is a diagram illustrating results of evaluating the yield of users with a lower performance of 10% while increasing the ratio of central TPs when supporting two UEs per TP when an embodiment of the present disclosure is applied.
FIG. 25 is a diagram illustrating results of evaluating the yield of users with a lower performance of 10% while increasing the ratio of central TPs when supporting 4 UEs per TP, when an embodiment of the present disclosure is applied.
26 is a diagram showing the results of comparing user bit error performance according to whether or not there is a channel correlation between groups determined by performing CSS when one or two DM-RSs are used when an embodiment of the present disclosure is applied. .
27 is a diagram illustrating a structure of a UE to which an embodiment of the present disclosure may be applied.
28 is a diagram showing the structure of a TP to which an embodiment of the present disclosure can be applied.
29 is a diagram illustrating a structure of a network entity to which an embodiment of the present disclosure may be applied.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

In describing the embodiments, descriptions of technical contents that are well known in the technical field to which the present disclosure pertains and are not directly related to the present disclosure will be omitted.

This is to more clearly convey the gist of the present disclosure without obscuring it by omitting unnecessary description.

For the same reason, in the accompanying drawings, some components are exaggerated, omitted, or schematically illustrated. Also, the size of each component does not entirely reflect the actual size. Identical or corresponding elements in each figure are assigned the same reference numerals.

Advantages and features of the present disclosure, and methods for achieving them, will become clear with reference to embodiments described below in detail in conjunction with the accompanying drawings.

However, the present disclosure is not limited to the embodiments disclosed below and may be implemented in various different forms, only the present embodiments complete the configuration of the present disclosure and are common knowledge in the art to which the present disclosure belongs. It is provided to completely inform the person who has the scope of the invention, and the present disclosure is only defined by the scope of the claims. Like reference numbers designate like elements throughout the specification.

At this time, it will be understood that each block of the process flow chart diagrams and combinations of the flow chart diagrams can be performed by computer program instructions. These computer program instructions may be embodied in a processor of a general purpose computer, special purpose computer, or other programmable data processing equipment, so that the instructions executed by the processor of the computer or other programmable data processing equipment are described in the flowchart block(s). It creates means to perform functions. These computer program instructions may also be stored in a computer usable or computer readable memory that can be directed to a computer or other programmable data processing equipment to implement functionality in a particular way, such that the computer usable or computer readable memory The instructions stored in are also capable of producing an article of manufacture containing instruction means that perform the functions described in the flowchart block(s). The computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operational steps are performed on the computer or other programmable data processing equipment to create a computer-executed process to generate computer or other programmable data processing equipment. Instructions for performing processing equipment may also provide steps for performing the functions described in the flowchart block(s).

Additionally, each block may represent a module, segment, or portion of code that includes one or more executable instructions for executing specified logical function(s). It should also be noted that in some alternative implementations it is possible for the functions mentioned in the blocks to occur out of order. For example, two blocks shown in succession may in fact be executed substantially concurrently, or the blocks may sometimes be executed in reverse order depending on their function.

At this time, the term '~unit' used in the present disclosure means software or a hardware component such as FPGA or ASIC, and '~unit' performs certain roles. However, '~ part' is not limited to software or hardware. '~bu' may be configured to be in an addressable storage medium and may be configured to reproduce one or more processors. Therefore, as an example, '~unit' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. Functions provided within components and '~units' may be combined into smaller numbers of components and '~units' or further separated into additional components and '~units'. In addition, components and '~units' may be implemented to play one or more CPUs in a device or a secure multimedia card.

For convenience of explanation, some terms and names defined in the 3rd generation partnership project (3GPP) specifications (5G, NR, LTE, or similar system specifications) may be used below. The use of these terms is not limited by the terms and names of the present disclosure, and may equally be applied to systems conforming to other standards, and may be changed into other forms without departing from the technical spirit of the present disclosure.

In addition, in the present disclosure, the expression of more than or less than is used to determine whether a specific condition is satisfied or fulfilled, but this is only a description to express an example and excludes more or less description. It's not about doing it. Conditions described as 'above' may be replaced with 'exceeds', conditions described as 'below' may be replaced with 'below', and conditions described as 'above and below' may be replaced with 'above and below'.

In the present disclosure, a UE may be referred to as a terminal, a mobile station, or a wireless transmit/receive unit (WTRU), but is not limited thereto and may be referred to by a term having the same or similar meaning.

In addition, in the present disclosure, a TP (transmission point) means a point capable of transmitting and receiving a radio signal by having one or more antennas, and various types of base stations will be referred to as TPs regardless of their names. can In addition, a TP may be referred to as a transmission reception point (TRP), an access point (AP), a distributed antenna, a radio unit (RU), or a distributed unit (DU) according to an embodiment of the present disclosure, It is not limited thereto, and may be referred to by terms having the same or similar meaning.

Also, in the present disclosure, a central processing unit (CPU) may be referred to as a separate entity or a network entity that performs an operation proposed in the present disclosure. Alternatively, various types of base stations may be referred to as CPUs regardless of their names. In addition, the CPU may be referred to as a central unit (CU) according to an embodiment of the present disclosure, but is not limited thereto, and may be referred to by terms having the same or similar meaning.

In addition, the data described in this disclosure may be referred to as a data stream, a signal, or a signal conveying control information, but are not limited thereto, and terms having the same or similar meanings may be referred to by

Also, in the present disclosure, precoding may refer to an operation of multiplying a modulated transmission signal by a precoder (or a precoding vector or a beamforming weight). For example, in the present disclosure, precoding may refer to an operation of changing an amplitude and a phase of a corresponding complex symbol by multiplying a complex symbol by a precoder after layer mapping. is a diagram showing the structure of a cellular mobile communication system to which the present disclosure can be applied.

Referring to FIG. 1, the mobile communication system shown in FIG. 1 includes a first cell 100, a second cell 110, and a third cell 120, and each of the cells 100, 110, and 120 At the center, a TP (here, it may mean a base station) performing (or providing) mobile communication within each cell may be disposed. Among them, the first cell 100 includes a TP 130, a first UE 140, and a second UE 150. The TP 130 may provide a mobile communication service to two UEs 140 and 150 located in the first cell 100 . The first UE 140 receiving the mobile communication service through the TP 130 has a relatively long distance to the TP 130 compared to the second UE 150 . Accordingly, data transmission/reception performance that can be supported by the first UE 140 may be relatively lower than data transmission/reception performance that can be supported by the second UE 150 .

Meanwhile, in the mobile communication system of FIG. 1, a reference signal (RS) may be transmitted to measure a downlink channel state of each cell or estimate a downlink channel. The reference signal may also be referred to as a pilot. According to the 3GPP standard, the reference signal transmitted by the TP to the UE includes CSI-RS channel status information-reference signal), DM-RS (demodulation-reference signal), and the like. The UE may measure the channel state between the TP and the UE using the CSI-RS and provide feedback on the channel state information. In addition, the UE can estimate a downlink channel using DM-RS, and can demodulate resources allocated to it based on the estimated channel.

2 is a diagram illustrating a structure of an LTE system of the 3GPP standard to which an embodiment of the present disclosure may be applied.

Referring to FIG. 2, as shown, the radio access network of the LTE system is a next-generation base station (evolved node B, hereinafter eNB, Node B or base station, where eNB may refer to a TP according to the present disclosure) (2-05 , 2-10, 2-15, 2-20), MME (2-25, mobility management entity), and S-GW (2-30, serving-gateway). The UE 2-35 accesses an external network through the eNBs 2-05 - 2-20 and the S-GW 2-30.

In FIG. 2, eNBs 2-05 - 2-20 correspond to existing Node Bs in the UMTS system. The eNB is connected to the UE 2-35 through a radio channel and performs a more complex role than the existing Node B. In the LTE system, since all user traffic, including real-time services such as VoIP (voice over IP) through the Internet protocol, is serviced through a shared channel, status information such as buffer status, available transmission power status, and channel status of UEs A device for scheduling by collecting is required, and the eNB (2-05 - 2-20) is in charge of this. One eNB typically controls multiple cells. For example, in order to implement a transmission rate of 100Mps, the LTE system uses orthogonal frequency division multiplexing (hereinafter referred to as OFDM) as a radio access technology in a 20Mhz bandwidth, for example. In addition, an adaptive modulation & coding (hereinafter referred to as AMC) method for determining a modulation scheme and a channel coding rate according to the channel condition of the terminal is applied. The S-GW 2-30 is a device that provides a data bearer, and creates or removes a data bearer under the control of the MME 2-25. The MME is a device in charge of various control functions as well as a mobility management function for a terminal, and is connected to a plurality of base stations.

3 is a diagram illustrating a radio protocol structure in an LTE system of the 3GPP standard to which the present disclosure can be applied.

Referring to FIG. 3, the radio protocols of the LTE system are PDCP (packet data convergence protocol 3-05, 3-40), RLC (radio link control 3-10, 3-35), MAC (medium access) in a terminal and an eNB, respectively. controls 3-15, 3-30).

The PDCPs 3-05 and 3-40 take charge of operations such as IP header compression/restoration. The main functions of PDCP are summarized as follows.

- Header compression and decompression (ROHC only)

- Transfer of user data

- Sequence delivery of upper layer PDUs at PDCP re-establishment procedure for RLC AM

- Order reordering function (for split beareres in DC (only supprt for RLC AM): PDCP PDU routing for transmission and PDCP PDU reordering for reception)

- Duplicate detection of lower layer SDUs at PDCP re-establishment procedure for RLC AM

- Retransmission of PDCP SDUs at handover and, for split bearers in DC, of PDCP PDUs at PDCP data-recovery procedure, for RLC AM)

- Encryption and deciphering function (ciphering and deciphering)

- Timer-based SDU discard in uplink

Radio link control (hereinafter referred to as RLC) (3-10, 3-35) reconstructs PDCP packet data units (PDUs) into appropriate sizes and performs an ARQ operation or the like. The main functions of RLC are summarized as follows.

- Data transfer function (transfer of upper layer PDUs)

- ARQ function (error correction through ARQ (only for AM data transfer))

- Concatenation, segmentation and reassembly of RLC SDUs (only for UM and AM data transfer)

- Re-segmentation of RLC data PDUs (only for AM data transfer)

- Reordering of RLC data PDUs (only for UM and AM data transfer)

- Duplicate detection (only for UM and AM data transfer)

- Error detection function (protocol error detection (only for AM data transfer))

- RLC SDU discard function (RLC SDU discard (only for UM and AM data transfer))

- RLC re-establishment

The MACs 3-15 and 3-30 are connected to several RLC layer devices configured in one terminal, and perform operations of multiplexing RLC PDUs to MAC PDUs and demultiplexing RLC PDUs from MAC PDUs. The main functions of MAC are summarized as follows.

- Mapping between logical channels and transport channels

- Multiplexing/demultiplexing of MAC SDUs belonging to one or different logical channels into/from transport blocks (TB) delivered to/from the physical layer on transport channels)

- Scheduling information reporting

- HARQ function (Error correction through HARQ)

- Priority handling between logical channels of one UE

- Priority handling between UEs by means of dynamic scheduling

- MBMS service identification

- Transport format selection

- Padding function (Padding)

The physical layer (3-20, 3-25) channel-codes and modulates upper layer data, transforms it into OFDM symbols and transmits it through a radio channel, or demodulates and channel-decodes OFDM symbols received through a radio channel and transmits them to the upper layer do the action

4 is a diagram illustrating a structure of a next-generation mobile communication system of the 3GPP standard to which the present disclosure can be applied.

Referring to FIG. 4, as shown, the radio access network of the next-generation mobile communication system is a next-generation base station (New Radio Node B, hereinafter referred to as NR gNB or NR base station, where NR gNB may refer to a TP according to the present disclosure.) (4-10) and NR CN (4-05, New Radio Core Network). A New Radio User Equipment (UE) 4-15 accesses an external network through the NR gNB 4-10 and the NR CN 4-05.

In FIG. 4, NR gNBs 4-10 correspond to evolved Node Bs (eNBs) of the existing LTE system. The NR gNB is connected to the NR UEs 4-15 through a radio channel and can provide a superior service to the existing Node B. In the next-generation mobile communication system, since all user traffic is serviced through a shared channel, a device that performs scheduling by collecting status information such as buffer status, available transmit power status, and channel status of UEs is required, which is called NR NB (4-10) is in charge. One NR gNB usually controls multiple cells. In order to implement high-speed data transmission compared to the current LTE, it may have more than the existing maximum bandwidth, and additional beamforming technology may be applied using orthogonal frequency division multiplexing (hereinafter referred to as OFDM) as a radio access technology. . In addition, an adaptive modulation & coding (AMC) method for determining a modulation scheme and a channel coding rate according to the channel condition of the terminal is applied. The NR CN 4-05 performs functions such as mobility support, bearer setup, and QoS setup. The NR CN is a device in charge of various control functions as well as mobility management functions for a terminal, and is connected to a plurality of base stations. In addition, the next-generation mobile communication system can interwork with the existing LTE system, and the NR CN is connected to the MME (4-25) through a network interface. The MME is connected to the eNB 4-30, which is an existing base station.

5 is a diagram illustrating a radio protocol structure in a next-generation mobile communication system of the 3GPP standard to which the present disclosure can be applied.

Referring to FIG. 5, the radio protocols of the next-generation mobile communication system are NR SDAP (5-01, 5-45), NR PDCP (5-05, 5-40), and NR RLC (5-10) in a terminal and an NR base station, respectively. , 5-35), and NR MACs (5-15, 5-30).

The main functions of the NR SDAPs 5-01 and 5-45 may include some of the following functions.

- Transfer of user plane data

- A mapping function between a QoS flow and a data bearer for uplink and downlink (mapping between a QoS flow and a DRB for both DL and UL)

- Marking QoS flow ID for uplink and downlink (marking QoS flow ID in both DL and UL packets)

- A function of mapping a relective QoS flow to a data bearer for uplink SDAP PDUs (reflective QoS flow to DRB mapping for the UL SDAP PDUs).

Regarding the SDAP layer device, the terminal may receive a RRC message to set whether to use the header of the SDAP layer device or the function of the SDAP layer device for each PDCP layer device, each bearer, or each logical channel, and SDAP header is set, the NAS QoS reflection setting 1-bit indicator (NAS reflective QoS) and the AS QoS reflection setting 1-bit indicator (AS reflective QoS) in the SDAP header allow the terminal to provide uplink and downlink QoS flows and mapping information for data bearers can be instructed to update or reset. The SDAP header may include QoS flow ID information indicating QoS. The QoS information may be used as data processing priority and scheduling information to support smooth service.

The main functions of the NR PDCPs 5-05 and 5-40 may include some of the following functions.

-Header compression and decompression (ROHC only)

- Transfer of user data

- In-sequence delivery of upper layer PDUs

- Out-of-sequence delivery of upper layer PDUs

- PDCP PDU reordering for reception

- Duplicate detection of lower layer SDUs

- Retransmission of PDCP SDUs

- Ciphering and deciphering

- Timer-based SDU discard in uplink.

In the above, the reordering function of the NR PDCP device refers to a function of rearranging PDCP PDUs received from a lower layer in order based on a PDCP SN (sequence number), and a function of transmitting data to an upper layer in the rearranged order Alternatively, it may include a function of immediately forwarding without considering the order, and may include a function of rearranging the order to record lost PDCP PDUs, and reporting the status of lost PDCP PDUs. to the transmitting side, and may include a function of requesting retransmission of lost PDCP PDUs.

The main functions of the NR RLCs 5-10 and 5-35 may include some of the following functions.

- Transfer of upper layer PDUs

- In-sequence delivery of upper layer PDUs

- Out-of-sequence delivery of upper layer PDUs

- ARQ function (Error Correction through ARQ)

- Concatenation, segmentation and reassembly of RLC SDUs

- Re-segmentation of RLC data PDUs

- Reordering of RLC data PDUs

- Duplicate detection

- Error detection function (Protocol error detection)

- RLC SDU discard function (RLC SDU discard)

- RLC re-establishment

In the above, the in-sequence delivery function of the NR RLC device refers to a function of sequentially delivering RLC SDUs received from a lower layer to an upper layer, and originally one RLC SDU is divided into several RLC SDUs and received , it may include a function of reassembling and forwarding the received RLC PDUs, and may include a function of rearranging the received RLC PDUs based on RLC SN (sequence number) or PDCP SN (sequence number). It may include a function of recording lost RLC PDUs, a function of reporting the status of lost RLC PDUs to the transmitting side, and a function of requesting retransmission of lost RLC PDUs. and, if there is a lost RLC SDU, may include a function of delivering only RLC SDUs prior to the lost RLC SDU to the upper layer in order, or if a predetermined timer expires even if there is a lost RLC SDU, a timer may be included. may include a function of forwarding all received RLC SDUs to the upper layer in order before the start of the RLC SDU, or if a predetermined timer expires even if there is a lost RLC SDU, all RLC SDUs received so far are sequentially transmitted to the upper layer It may include a forwarding function. In addition, RLC PDUs may be processed in the order in which they are received (regardless of the order of serial numbers and sequence numbers, in the order of arrival) and delivered to the PDCP device regardless of order (out-of sequence delivery). In , segments stored in a buffer or to be received later may be received, reconstructed into one complete RLC PDU, processed, and transmitted to the PDCP device. The NR RLC layer may not include a concatenation function, and the function may be performed in the NR MAC layer or replaced with a multiplexing function of the NR MAC layer.

In the above, the out-of-sequence delivery function of the NR RLC device refers to a function of immediately delivering RLC SDUs received from a lower layer to an upper layer regardless of the order, and originally one RLC SDU is multiple RLC When received divided into SDUs, it may include a function of reassembling and forwarding them, and may include a function of storing RLC SNs or PDCP SNs of received RLC PDUs and arranging them in order to record lost RLC PDUs. can

NR MACs (5-15, 5-30) may be connected to several NR RLC layer devices configured in one terminal, and the main functions of the NR MAC may include some of the following functions.

- Mapping between logical channels and transport channels

- Multiplexing/demultiplexing of MAC SDUs

- Scheduling information reporting

- HARQ function (Error correction through HARQ)

- Priority handling between logical channels of one UE

- Priority handling between UEs by means of dynamic scheduling

- MBMS service identification

- Transport format selection

- Padding function (Padding)

The NR PHY layers (5-20, 5-25) channel code and modulate higher layer data, make OFDM symbols and transmit them through a radio channel, or demodulate OFDM symbols received through a radio channel and channel decode them to a higher layer. You can perform forwarding operations.

Meanwhile, in the 3GPP standard LTE system and the next-generation mobile communication system described in FIGS. 2 and 4, a plurality of cells are generally built in a limited area, and one UE in the cell serves itself. It may be implemented through a non-coordinated multi point (non-CoMP) data transmission/reception method in which data is received from one TP. A more detailed description of the non-CoMP scheme will be described with reference to FIG. 6 .

6 is a diagram illustrating the structure of a system to which a non-coordinated multi point (non-CoMP) scheme is applied according to an embodiment of the present disclosure.

Referring to FIG. 6, in FIG. 6, a mobile communication system including three cells and having a TP at the center of each cell is taken as an example.

The TP of each cell may transmit data to a UE existing in the corresponding cell. That is, the TP of Cell 0 may transmit data 600 to UE0 existing in the service area of Cell 0. In addition, using time and frequency resources different from those used in Cell 0, the TP of Cell 1 transmits data 610 to UE1 existing in the service area of Cell 1, and the TP of Cell 2 transmits data 610 Data 620 may be transmitted to UE2 existing in the service area of #2. Referring to a radio resource 630 used in Cell 0, a radio resource 650 used in Cell 1, and a radio resource 640 used in Cell 2, Cell 0 to Cell 2 use different time and frequency resources. It can be seen that the data can be transmitted. In this way, a transmission method in which TPs of each cell transmit data only to UEs in a corresponding cell using different time and frequency resources, and there is no cooperation between the TPs, may be referred to as a non-CoMP method.

On the other hand, in the non-CoMP data transmission/reception method, since transmission/reception antennas of each TP are concentrated in the center of a cell, high data transmission/reception performance is provided to a UE located far from the center of a cell (ie, a cell boundary). There are limitations that cannot be supported. Accordingly, various system structures to which the CoMP method, which is a multi-cell cooperative communication technology, can be applied have been proposed. A detailed description of the CoMP scheme will be described with reference to FIG. 7 .

7 is a diagram illustrating the structure of a system to which a Coordinated Multi Point (CoMP) method according to an embodiment of the present disclosure is applied.

Unlike the non-CoMP scheme in which data transmission/reception is performed in a state where there is no cooperation between each TP as described in FIG. 6, in FIG. shown

Referring to FIG. 7 , it is exemplified that the UE receives data transmitted through a physical downlink shared channel (PDSCH) in Cell 0 and Cell 2. That is, the UE simultaneously receives data 700 and 710 transmitted from two TPs. In a mobile communication system in which a plurality of TPs exist, one or more TPs cooperate to transmit the same data to one UE, so that a UE located relatively far from the TP (e.g., cell boundary) can obtain more improved It is now possible to support data transmission and reception performance.

On the other hand, the CoMP scheme to provide more advanced services to users at cell boundaries can be implemented in various ways in a communication system, and will be described below with reference to FIG. 8 .

8 is a diagram illustrating classification of CoMP schemes in an LTE system or a next-generation mobile communication system to which an embodiment of the present disclosure can be applied.

Referring to FIG. 8, the CoMP scheme is an uplink CoMP scheme for improving uplink data transmission/reception performance, which is a radio connection from a UE to a TP, and a downlink CoMP scheme for improving downlink data transmission/reception performance, which is a radio connection from a TP to a UE. It can be classified as a CoMP method.

First, the uplink CoMP scheme may be implemented by joint reception (JR), coordinated scheduling (CS), or a combination of the above schemes.

JR is a method of receiving data transmitted from a terminal together in several TPs. CS is a method in which a plurality of TPs cooperate with each other to perform scheduling and precoding.

The downlink CoMP scheme may be implemented by joint processing (JP), coordinated scheduling/coordinated beamforming (CS/CB), or a combination of the above schemes.

CS/CB is a method in which one TP transmits data to the UE, but a plurality of TPs cooperate to perform scheduling and beamforming. Through this, interference to a UE located at the boundary of the TP can be reduced.

JP is a sharing of data to be transmitted to a UE by a plurality of TPs, and may be further classified into dynamic point selection (DPS) and joint transmission (JT). The DPS may refer to DCS (dynamic cell selection), and one of a plurality of TPs transmits downlink data to the UE, but the TP transmitting data to the UE dynamically changes. That is, it means a method of transmitting downlink data through a TP selected from among a plurality of TPs according to a specific rule.

JT is a method in which several TPs cooperate to transmit the same data to the UE.

Meanwhile, JT may be further classified into CJT (coherent JT) and NCJT (non-coherent JT) depending on whether TPs transmit data streams in synchronization. do it with

9 is a diagram illustrating coherent joint transmission (CJT) and non-coherent joint transmission (NCJT) schemes in an LTE system or a next-generation mobile communication system to which the present disclosure may be applied.

A of FIG. 9 illustrates a system for transmitting data to the UE according to the CJT method, and B of FIG. 9 illustrates a system for transmitting data to the UE according to the NCJT method.

In A of FIG. 9, according to the CJT scheme, a central processing unit (CPU) centrally performs scheduling (which may be referred to as central scheduling), and TPs controlled by the CPU are synchronized according to the scheduling. transmits data to (coherent transmission).

In B of FIG. 9, according to the NCJT scheme, each TP performs independent scheduling (which may be referred to as local scheduling) and independently transmits data to the UE without synchronizing with other TPs (non-coherent transmission). .

On the other hand, in coherent transmission, a plurality of TPs transmit data streams synchronized in time by compensating for delay to the UE, and the UE transmits from the plurality of TPs By receiving the sum of the data streams, the reception performance of the UE can be improved accordingly.

According to the CJT method, in a system where multiple users exist through cooperation between a plurality of TPs, interference caused by cells belonging to other users can be reduced, and UE throughput and data reception performance can be increased at the cell boundary. .

Meanwhile, the CJT scheme of CoMP considered in the present disclosure can be implemented in a communication system based on a distributed antenna system composed of a plurality of TPs and a CPU connected to the plurality of TPs through a front hall. The CPU can transmit/receive data to and from multiple TPs through the front haul, and can control the TPs. A more detailed description will be given with reference to FIG. 10 .

10 is a diagram illustrating a distributed antenna system in which a CJT scheme of CoMP according to an embodiment of the present disclosure can be implemented. However, the embodiment of the present disclosure is not limited thereto, and the JR method may be implemented in the distributed antenna system of FIG. 10 .

Referring to FIG. 10 , a distributed antenna system including one cell 1000 is shown as an example. A DU (as described above, the DU of the present disclosure may refer to the above-described TP) and a CU connected through a front hall (as described above, the CU of the present disclosure may refer to the above-described CPU) (1030 ) is disposed at the center of the cell 1000, and DUs 1060, 1070, 1080, and 1090 in each cell may be disposed at different positions. The CU 1030 controls DUs 1060, 1070, 1080, and 1090 in the cell, and each DU may transmit a downlink signal to a UE located in the cell 1000 under the control of the CU 1030 ( JT). In this case, each of the DUs 1060, 1070, 1080, and 1090 may include one or a plurality of antennas. Meanwhile, in such a distributed antenna system, like the distributed MIMO method, downlink data can be coherently transmitted to the UE through beamforming using an antenna located in a TP (CJT). To this end, the CPU may generate a centralized precoder (or may refer to an operation of obtaining or verifying a precoder) and apply the data to be transmitted to the UE. In addition, the CPU transmits the precoder-applied data to each TP connected to the CPU through the front haul, and each TP can transmit a data stream to the UE in synchronization (coherent transmission).

On the other hand, in the CJT scheme of CoMP, which can be implemented through the distributed antenna system structure as described above, the amount of data transmitted and received through the front hall between each TP and the CPU is excessive, and as the number of TPs controlled by the CPU increases, the number of TPs controlled by the CPU increases. There may be a problem in that the calculation complexity of the CPU greatly increases. Details will be described with reference to FIG. 11 .

11 illustrates a common public radio interface (CPRI) between a remote radio header (RRH) (or radio unit (RU), DU) and a base band unit (BBU) (or CU) according to an embodiment of the present disclosure. It is a drawing showing an example.

Referring to FIG. 11, when a distributed antenna system is used in a 4G (or LTE) system, the CPRI standard (hereinafter referred to as CPRI) may be used. For example, based on option 8 shown in FIG. 11, functions may be separated into CUs and DUs, and CPRI may be used as a communication interface between CUs and DUs. At this time, if the functions are separated based on option 8 as described above, only the RF stage exists in the DU, and operations related to the remaining signal processing can be performed by the CU. On the other hand, since the CU must sample an RF signal and transmit it to the DU through the front hall when transmitting downlink data, a lot of load is generated in the front hall. In addition, in the case of uplink in which the RF signal sampled in the DU must be transmitted to the CU, the load of the front haul becomes larger. In addition, in the case of the CJT method described above, not only user data but also accurate channel information for generating a precoder (or precoding vector) must be transmitted to the CPU (CU) through front haul, In a situation where transmission capacity is limited, there is a limit to implementing the above-described CJT scheme.

In the next-generation mobile communication system, an enhanced common public radio interface (eCPRI) standard is defined to solve the above-mentioned limitations of the front haul, so that functions can be separated by selecting different options. For example, based on option 2 or option 7 shown in FIG. 11, functions may be separated into CU and DU. If functions are separated from the upper layer as in option 2, precoding must be distributed in DUs, and thus the effect of improving data transmission/reception performance obtained through the CJT method may be reduced. On the other hand, if the function is separated in the lower layer as in option 7, the front haul transmission amount may still be excessive compared to the existing case. That is, a trade-off may occur between CJT performance and front haul transmission amount according to a function separation point. In this respect, there are limitations in implementing the CJT method through eCPRI.

On the other hand, in addition to the above-described problems, in order to generate a precoder matrix for CJT in the CPU, it is necessary to accurately know the channel, but there is also a problem that a large amount of overhead may occur depending on the operation for this.

In addition, when the number of UEs is greater than the number of TPs, there is a limit to implementing the CJT scheme. For example, in a cellular network as shown in FIG. 1, it is possible to estimate a downlink channel based on a downlink pilot of each TP. When the number of TPs is less than the number of UEs, the overhead due to channel estimation is not large. Since channel estimation must be performed by using the channel estimation method, there may be a problem in that overhead due to channel estimation may greatly increase.

In addition, due to the overhead of uplink, each UE cannot accurately feedback channel information, or generally quantizes channel information to obtain channel quality information (CQI), precoding matrix indicator (PMI), Feedback is given in the form of a rank indicator (RI), etc. At this time, due to limitations due to quantization of channel information, accurate channel information may not be reflected in the feedback. Therefore, in existing communication systems, there are various limitations in implementing the CJT method of CoMP.

On the other hand, in a recent UDN environment, as a method of eliminating the cell boundary of a terminal, a cell-free massive MIMO-based system (distributed MIMO, or distributed antenna system, etc.), which is a system structure in which the number of TPs supporting one terminal is large. this has been suggested A detailed description of the cell-free massive MIMO-based system will be described with reference to FIG. 12 .

12 is a diagram showing a model of a cell-free massive MIMO-based system to which the present disclosure can be applied.

Referring to FIG. 12, the cell-free massive MIMO-based system can estimate a downlink channel based on an uplink pilot by operating with time division duplexing (TDD). That is, unlike a downlink channel estimated based on a downlink pilot transmitted from a TP to a UE, a downlink channel may be estimated based on an uplink pilot transmitted from a UE to a TP. The TP 1210 may generate channel information for the terminal based on the uplink pilot received from the UE 1220 and transmit the channel information to the CPU 1200. Through this, it is possible to overcome the above-described limitations due to quantization of channel information and overhead due to channel estimation.

Then, the CPU 1200 may perform precoding to control multi-user interference based on the channel information. Accordingly, the TP 1210 coherently transmits data to the UE 1220, so that the CoMP CJT scheme considered in the present disclosure can be implemented.

On the other hand, since the number of TPs that need to cooperate with each other in the cell-free massive MIMO system is greatly increased compared to the existing system, in implementing the CJT method in the cell-free massive MIMO system, the computational complexity required to control the numerous TPs in the CPU There may be a problem that greatly increases. For example, the CPU of a cell-free massive MIMO-based system must accurately synchronize data transmission and reception among numerous TPs, and since numerous TPs dynamically cooperate, the precoder creation process (or precoding process) considering these cooperative TPs The computational complexity of the CPU can be very high. In addition, there is a limitation that the load of data or channel information transmitted through the front haul is still large. Therefore, problems may arise in implementing the CJT method of CoMP even in a cell-free massive MIMO-based system.

Accordingly, the present disclosure proposes a method capable of solving the problems of front hall load and CPU computational complexity that may occur in implementing the CoMP CJT method in a cell-free massive MIMO-based wireless communication system. Specifically, according to the present disclosure, a plurality of TPs controlled by a CPU are divided into a Central TP group in which precoding is performed centrally in the CPU and a Local TP group in which precoding is performed in each TP in a distributed manner, and the Central TP or We propose a precoding method that allows each TP to operate as a local TP. In addition, according to the precoding method proposed in the present disclosure, a system structure capable of transmitting and receiving uplink data or downlink data between a plurality of TPs and a UE is proposed.

According to the present disclosure, scheduling may be performed to determine whether a TP operates as a central TP or a local TP based on a given front haul capacity and calculation complexity according to UE and network performance. In this case, since computational complexity is distributed and shared among TPs according to a scheduling result based on computational complexity constraints, scheduling for determining an operating method of such TPs may be referred to as CSS (complexity-shared scheduling). CSS may also mean that a plurality of TPs are classified (determined or classified) to perform one of the operations performed by the central TP group and the operation performed by the local TP group. In addition, CSS may mean a process of classifying all TPs into a central TP group and a local TP group, or a process of determining a ratio of a central TP group and a local TP group among all TPs. In addition, since it is determined whether precoding is performed in the CPU (Central TP) or in the TP (Local TP) according to the CSS execution result, CSS may mean a precoding method. Meanwhile, in the present disclosure, for convenience of description, the central TP group may be referred to as a first group and the local TP group may be referred to as a second group.

라고 할 수 있다.)와 계산 복잡도(예를 들어,

라고 할 수 있다.)가 고려된 하기와 같은 [수학식 1]의 목적함수를 고려할 수 있다.In the cell-free massive MIMO system described above, in order to obtain optimal data transmission and reception performance, the front hall capacity (eg,

) and computational complexity (e.g.

) may be considered, and the objective function of [Equation 1] as follows.

는

의 수율(throughput),

는 각 TP 단위로 수행할 계산량이며,

은 프론트 홀 링크로 전송될 정보의 양을 나타낸 것이다. 상기 목적함수에 대한 최적화 문제를 해결하기 위해서는 모든 TP가 중앙집중적으로 동작하는 경우(즉, Central TP) 또는 분산적으로 동작하는 경우(즉, Local TP), 두 가지 상황을 가정해볼 수 있다. 만약, 모든 TP가 Central TP로서 동작하는 경우에는 프론트 홀 링크로 전송될 정보의 양(

) 및 총 계산량(

)의 제한을 받게 된다. 이와 달리, 모든 TP가 Local TP로서 경우에는 TP들 간 높은 간섭으로 인해 수신 성능이 저하될 수 있다. 이러한 특징을 고려하여, CPU는 CSS를 통해

를 지원하는 TP 그룹(예들 들어,

라 할 수 있다.)을 Central TP 그룹(예를 들어,

라 할 수 있다.)과 Local TP 그룹(예를 들어,

라 할 수 있다.)로 분류(또는, 분할, 구분)할 수 있다.In the above [Equation 1],

Is

the yield of,

is the amount of calculation to be performed in each TP unit,

represents the amount of information to be transmitted through the front haul link. In order to solve the optimization problem for the objective function, two situations can be assumed: when all TPs operate centrally (ie, Central TP) or when they operate in a distributed manner (ie, Local TPs). If all TPs operate as central TPs, the amount of information to be transmitted to the front haul link (

) and the total amount of computation (

) will be limited. In contrast, when all TPs are local TPs, reception performance may deteriorate due to high interference between TPs. Considering these characteristics, the CPU via CSS

TP group that supports (eg,

) to the Central TP group (eg,

) and Local TP group (for example,

) can be classified (or divided, divided).

More specifically, the objective function of the CSS considering the trade-off between transmission and reception performance and constraints according to front haul capacity and computational complexity can be expressed by [Equation 2] below.

)은 안테나의 수(예를 들어,

라 할 수 있다.)와 심볼 크기(예를 들어,

라 할 수 있다.), 그리고 quantization level(Q)에 기반하여 결정될 수 있다. 이때 CPU는 상기 주어진 제약에서, 목적 함수에 의해 주어진 성능을 최적화할 수 있도록 각 TP가 Central TP 또는 Local TP로 동작할지를 결정한다. CPU에서 상기 CSS를 수행하는 구체적인 내용은 도 13을 참조하여 설명하기로 한다.In [Equation 2], the amount of information to be transmitted through the front haul link (

) is the number of antennas (eg,

) and symbol size (eg,

), and can be determined based on the quantization level ( Q) . At this time, the CPU determines whether each TP operates as a central TP or a local TP so that the performance given by the objective function can be optimized in the given constraint. Details of performing the CSS in the CPU will be described with reference to FIG. 13 .

13 is a diagram illustrating a sequence in which a CPU performs CSS according to an embodiment of the present disclosure.

Referring to FIG. 13 , in step 13-100, the CPU may check whether TPs controlled by the CPU perform CSS for determining whether to operate as a central TP or a local TP. The CPU may determine whether CSS is performed based on at least one of several criteria. For example, the CPU may receive a report of channel information on a channel between the TP and the UE, and determine whether to perform CSS based on the channel information and various criteria. For example, the CPU may initiate CSS in the following cases. However, the following conditions are only examples of the present disclosure, and the scope of the present disclosure is not limited thereto.

- When the CPU periodically performs CSS

- When the CPU performs CSS based on certain criteria (threshold, etc.)

- When the UE requests the CPU to perform CSS

First, when the CPU periodically performs CSS, CSS is performed first when clustering (which may mean determining a set of TPs to serve the UE) is performed, or CSS is performed at random intervals It can be. For example, CSS can be performed at a shorter period than the large-scale coherence time. Specifically, clustering is performed every period of large-scale coherence time, which is a longer period than CSS, or when handover of a terminal occurs. When channel information changes rapidly, such as the mobility of a UE, the CPU CSS may be performed at a period shorter than the clustering period (eg, T_CSS = T_Clustering/2 may be set). In this case, CSS is performed more frequently than clustering, but because CSS is a procedure that consumes less cost than clustering, which requires complex procedures, by operating CSS, the network costs less than a network that only performs clustering without CSS. has the advantage of being able to optimize the performance of

Second, when the CPU performs CSS based on a certain criterion (threshold), the UE reports its own downlink throughput information or its own channel reception state (RSRP) to the CPU through uplink, and the CPU can know the uplink and / or downlink throughput of the UE through this. At this time, when the throughput of a specific UE falls below a certain threshold, the difference in throughput between the UE with the highest throughput and the lowest UE exceeds a certain threshold, or the dispersion of the UE's throughput distribution exceeds a certain threshold. , the CPU may perform CSS.

Also, the UE may request the CPU to perform CSS. In this case, the UE checks its own downlink throughput or channel state, and if the UE's throughput or channel state falls below a certain threshold, the CPU can be requested to perform CSS so that more central TPs can be allocated to the UE. there is.

), Central TP의 그룹(

)으로 나눌 때, 상기 Central TP의 집합(

)에 더 많은 TP가 속하도록 CSS를 수행할 수 있다. 만약, 파일롯 개수의 제한 등으로 인해 클러스터링 행렬에서 TP가 지원할 수 있는 UE의 수가 감소하고, 상기 TP의 수가 고정되어 정해져 있는 경우, CPU는 UE를 지원하는 TP중 Central TP의 비율을 늘릴 수 있다. 한편, 복잡도 제한 등으로 상기 Central TP의 비율을 늘릴 수 없다면, CPU는 수신 성능이 이미 충분히 좋은 UE의 계산 자원을 하위 사용자에게 할당해 줄 수 있다(즉, CPU는 수신 성능이 충분히 좋은 UE의 Central TP를 Local TP로 동작하도록 하고, 하위 성능인 UE의 Local TP를 Central TP로 동작하도록 할 수 있다). 한편, UE가 Central TP를 더 많이 할당해달라고 CPU에 요청한 경우, 이를 반영하여 CSS를 수행할 수도 있다.In step 13-100, when the CPU determines to perform CSS, in step 13-200, the CPU converts TPs into central TPs based on at least one of front haul capacity and calculation complexity that the CPU can tolerate. Alternatively, you can decide to operate as a local TP. For example, the CPU may preferentially operate as a central TP for TPs supporting a UE with a low yield (which may refer to throughput) in order to maximize performance for cell edge users through CSS. can do That is, the CPU assigns at least one TP group supporting the UE to a group of Local TPs (

), the group of the Central TP (

), the set of the Central TPs (

) can perform CSS so that more TPs belong to it. If the number of UEs that can be supported by a TP in the clustering matrix decreases due to a limitation on the number of pilots, etc., and the number of TPs is fixed, the CPU can increase the ratio of central TPs among TPs supporting UEs. On the other hand, if the ratio of the central TP cannot be increased due to complexity limitations, etc., the CPU can allocate computational resources of a UE with sufficiently good reception performance to a lower user (ie, the CPU can allocate the computational resources of a UE with sufficiently good reception performance to a lower user) The TP can be operated as a Local TP, and the Local TP of the UE, which is a lower performance, can be operated as a Central TP). Meanwhile, when the UE requests the CPU to allocate more central TPs, CSS may be performed by reflecting this request.

In addition, embodiments of the present disclosure are not limited to the above examples, and the CPU may perform CSS based on various criteria to maximize reception performance of the UE.

When the CSS execution is completed, in step 13-300, the CPU may transmit the CSS execution result (eg, an indicator instructing to operate as a central TP or a local TP) to TPs that were CSS execution targets. TPs that have received CSS execution results may operate as Central TPs or Local TPs according to the CSS results.

Meanwhile, CSS according to an embodiment of the present disclosure may be implemented through TPs having eCPRI shown in FIG. 14 . At this time, the TP instructed to operate as a local TP is instructed by the CPU to operate as a central TP according to the CSS execution result, or the TP instructed to operate as a central TP is determined to operate as a local TP according to the CSS execution result. Since it can be instructed, CSS execution target TPs must be able to freely select split 1 or split 2 shown in FIG. 14 . That is, when layer splitting occurs in split 1 or split 2, all procedures for split 1 or split 2 may be embedded in the TP so that the TP can perform all necessary procedures according to the split selected by T. If the precoding procedure is performed in a distributed unit (DU), layer splitting is performed in split 1, and at this time, the DU may be referred to as a local TP. In contrast, if the precoding procedure is performed in a central unit (CU), layer splitting is performed in split 2, and in this case, the DU may be referred to as a central TP.

The present disclosure is a system in which local TPs that perform precoding distributedly in DUs (or RUs) and central TPs that centrally perform precoding in CUs coexist according to eCPRI function separation points among communication systems utilizing eCPRI can be applied to As described above, a TP corresponding to a central TP or a local TP may be changed to operate as a local TP or a central TP as needed. That is, according to an embodiment of the present disclosure, it may be determined whether precoding for each TP is performed in a corresponding TP in a distributed manner or in a CU that centrally controls the corresponding TP.

As shown in FIG. 13, the CoMP-type data transmission/reception system can be implemented based on TPs instructed to operate as a central TP or a local TP according to the result of the CPU performing CSS and performing the CSS. A more detailed description will refer to FIGS. 15A and 15B.

15A and 15B are diagrams illustrating a system structure in which a TP instructed to operate as a central TP or a local TP according to a CSS execution result according to an embodiment of the present disclosure transmits and receives uplink data and downlink data, respectively. .

First, in FIGS. 15A and 15B, for convenience of description, the following assumptions may be applied.

개의 안테나를 장착하고 있고, 상기 TP들은 프론트 홀을 통해 CPU와 연결되어 있다. - L TPs are distributed, and each TP

Two antennas are mounted, and the TPs are connected to the CPU through a front hole.

개의 UE가 있으며, 이때 TP의 수가 UE 수 보다 많다는 cell-free massive MIMO시스템의 기본 전제에 따라 상기 TP의 수(L)는 UE의 수(K)보다 더 많다. -

There are two UEs, and in this case, according to the basic premise of the cell-free massive MIMO system that the number of TPs is greater than the number of UEs, the number of TPs ( L) is greater than the number of UEs ( K) .

- Each UE is equipped with one antenna, and the channel between the UE and the TP has a Rayleigh fading distribution.

번째 TP와

번째 단말 간의 채널은 벡터

, 각 UE별로 TP의 안테나 간 공간적 상관성은 상관 행렬

, 채널의 경로 손실과 shadowing을 나타내는 large-scale fading 계수를

로 정의하였다.- At this time,

with the second TP

The channel between the th terminal is a vector

, The spatial correlation between the antennas of the TP for each UE is a correlation matrix

, the large-scale fading coefficient representing the path loss and shadowing of the channel

defined as

On the other hand, the assumption as described above is only for convenience of explanation, and the present disclosure is not limited thereto. That is, the above assumptions are only one embodiment for carrying out the present disclosure, and some of the assumptions may not be applied.

FIG. 15A illustrates a system structure in which a UE transmits and receives uplink data to and from a CPU through a central TP and a local TP serving itself as a result of CSS being performed according to an embodiment of the present disclosure. The central TP estimates a channel based on the uplink pilot received from the UE and transmits information about the estimated channel to the CPU. Thereafter, the CPU calculates a central combining vector based on the estimated channel and estimates the signal by applying the combined vector to the signal transmitted from the UE. In contrast, the local TP estimates a channel based on an uplink pilot received from the UE and calculates a local combining vector based on the estimated channel. Thereafter, the local TP estimates a symbol by applying the combined vector to the signal transmitted from the UE, and sends the estimated symbol to the CPU, and the CPU combines them to estimate the signal transmitted from the UE. According to the method described above, since a signal transmitted by a UE can be received through a plurality of TPs composed of a central TP and a local TP, the JR method considered in the present disclosure can be implemented.

개 일 수 있다.)의 상향링크 파일롯을 전송할 수 있으며, 상기 상향링크 파일롯은 하기 [수학식 3]로 나타낼 수 있다. More specifically, referring to FIG. 15A, the terminal has an arbitrary number (eg, orthogonality) with each other during the coherence time interval.

may be) can transmit uplink pilots, and the uplink pilots can be represented by [Equation 3] below.

)의 길이는 가질 수 있다. 이때,

이다. 상향링크 파일롯은 각 UE에 할당되며, 일반적으로 UE의 수가 상향링크 파일롯의 수보다 많으므로, 하나의 파일롯이 다수의 UE에 할당될 수 있다. UE로부터 상향링크 파일롯을 수신한 Central TP 또는 Local TP는 UE와 TP 간 채널(

)을 추정할 수 있다. 이때 채널 추정은 예를 들어, MMSE(minimum mean square error) ZF(Zero-forcing) 또는 LS Square(Least Square) 등에 기반하여 수행될 수 있다.In Equation 3, each uplink pilot has an arbitrary length (eg,

) can have a length of At this time,

am. Uplink pilots are allocated to each UE, and since the number of UEs is generally greater than the number of uplink pilots, one pilot may be allocated to multiple UEs. The central TP or local TP receiving the uplink pilot from the UE establishes a channel between the UE and the TP (

) can be estimated. At this time, channel estimation may be performed based on, for example, minimum mean square error (MMSE) zero-forcing (ZF) or least square (LS Square).

라고 하면, UE와 모든 TP 간 채널의 추정치는 하기와 같은 [수학식 4]으로 나타낼 수 있다.Channel estimation value between UE and TP

In this case, the estimated value of the channel between the UE and all TPs can be expressed as [Equation 4] below.

이며,

번째 UE의 신호를

라고 하면,

번째 TP에서 상향 링크의 데이터 수신 신호는 하기의 [수학식 5]으로 나타낼 수 있다.At this time,

is,

the signal of the second UE

If you say

The uplink data reception signal at the TP 1 can be expressed by Equation 5 below.

는

번째 TP의 각 안테나에서 독립적인 수신 잡음이다.here,

Is

It is the independent reception noise at each antenna of the th TP.

번째 UE의 심볼을 검출하기 위해 L개의 TP로 수신된 상향 링크 신호를 결합하며, 이때 각 TP에서 결합 벡터(예를 들어,

라 나타낼 수 있다.)를 가중치를 곱한 후에 모든 TP에 대해서 결합하여 상향링크 데이터를 추정하며, 이는 하기와 같은 [수학식 6]으로 나타낼 수 있다. in the CPU

In order to detect the symbol of the th UE, uplink signals received by L TPs are combined, and in each TP, a combined vector (eg,

) is multiplied by a weight and then combined for all TPs to estimate uplink data, which can be expressed as [Equation 6] below.

는 각 TP에 대한 결합 벡터를 통합한 것으로서

와 같이 나타낼 수 있으며,

은 각 TP에 대한 잡음(noise)를 나타낸 것이다.here,

Is the integration of the binding vector for each TP

can be expressed as,

represents the noise for each TP.

는 MR또는 MMSE에 기반하여 계산될 수 있다. MR에 기반하여 각 TP에 대한 결합 벡터를 계산하는 경우,

로 나타낼 수 있으며, 각 TP에서 심볼을 추정하는데 적용할 수도 있다. 한편, 중앙 집중적 MMSE 결합 벡터(central combining vector)는 하기 [수학식 7]와 같이 나타낼 수 있다. On the other hand, the binding vector for each TP

can be calculated based on MR or MMSE. When calculating the binding vector for each TP based on MR,

It can be expressed as , and it can be applied to estimate symbols in each TP. Meanwhile, the central combining vector can be expressed as in [Equation 7] below.

이며,

이다.here,

is,

am.

번째 UE의 심볼을 추정하고, CPU에서 이들을 모아서 결합하는 방법을 고려할 수 있다. 이때

번째 TP에서 추정된 심볼을

라고 하면, 이는 하기 [수학식 8]에 기반하여 수신 신호를 결합함으로써 추정될 수 있다.The above method is a method of centrally combining and estimating received signals in a CPU. On the other hand, unlike this, in each TP

A method of estimating the symbols of the th UE and collecting and combining them in the CPU may be considered. At this time

The symbol estimated at the th TP

, this can be estimated by combining the received signals based on the following [Equation 8].

이며, MMSE에 기반하여 계산된 경우 하기와 같은 [수학식 9]으로 나타낼 수 있다.At this time, if the joint vector is calculated based on MR,

, and when calculated based on MMSE, it can be represented by [Equation 9] as follows.

은 CPU에서

의 형태로 결합되며, 이를 large-scale fading decoding(LSFD)이라고 칭할 수 있다. estimated at each TP

on the CPU

It is combined in the form of , which can be referred to as large-scale fading decoding (LSFD).

과 TP index

에 대하여 TP의 안테나 개수가

일 때

의 집합으로 이루어진 diagonal matrix를 정의함으로써 어떤 TP의 안테나가 어떤 UE를 지원하는지 나타낼 수 있다. diagonal matrix(

)는 하기 [수학식 10]으로 나타낼 수 있다.Meanwhile, according to an embodiment of the present disclosure, clustering may be assumed such that a TP supports (or serves) only some UEs belonging to the system rather than supporting all UEs in the system. For example, UE index

and TP index

For , the number of antennas of the TP is

when

By defining a diagonal matrix consisting of a set of TPs, it is possible to indicate which UEs are supported by antennas of which TPs. diagonal matrix(

) can be represented by the following [Equation 10].

번째 TP가

를 지원하는 TP의 집합인

에 속하면 TP의

개 안테나에 대하여 단위행렬을 부여하고,

번째 TP가

에 속하지 않으면 같은 크기의 모든 entry가 0인 matrix를 부여한다. 이에 따라,

를 지원하는 clustering matrix는 하기의 [수학식 11]으로 나타낼 수 있다.In [Equation 10],

second TP

is a set of TPs that support

of TP if it belongs to

An identity matrix is given for each antenna,

second TP

If it does not belong to , a matrix with all entries of the same size being 0 is given. Accordingly,

A clustering matrix that supports can be represented by [Equation 11] below.

를 이용하여,

를 서빙 하는 것으로 클러스터링 된 TP들로만 이루어진 경우의 상향링크 전송 데이터는 하기 [수학식 12]으로 나타낼 수 있다.[Equation 11]

using,

Uplink transmission data in the case of serving only clustered TPs can be represented by [Equation 12] below.

는 상향 링크 데이터를 추정하기 위한 결합 벡터이고,

는

의 상향 링크 전송 심볼을 나타낸다.At this time,

Is a combined vector for estimating uplink data,

Is

Indicates an uplink transmission symbol of

번째 TP가 지원하는 UE의 집합을

이라고 하면, 상기 클러스터링은 하기 [수학식 13]으로 정의될 수도 있다.Unlike this

The set of UEs supported by the th TP

, the clustering may be defined by the following [Equation 13].

에게 간섭을 미치는 UE의 집합을

로 정의하면, 그러한 UE의 집합은

를 서빙(또는, 지원)하는 TP가 서빙하는 모든 UE를 포함할 수 있으며, 하기 [수학식 14]으로 나타낼 수 있다.Meanwhile, in the clustering situation, the UE may determine an interfering UE strongly interfering with itself in order to consider interference affecting the calculation of the combining vector. for example,

A set of UEs interfering with

Defined as , the set of such UEs is

may include all UEs served by a TP serving (or supporting), and may be represented by Equation 14 below.

When interfering UEs are considered in this clustering situation, the existing centralized MMSE combining vector (Equation 7) or distributive MMSE combining vector (Equation 9) is represented by [Equation 15] and [Equation 16], respectively. can

At this time, as an MMSE combining vector considering interference of some UEs among all UEs, [Equation 15] is referred to as a P-MMSE (partial MMSE) combined vector, and [Equation 16] is LP-MMSE (local partial MMSE) can be referred to as The CPU may estimate the uplink data transmitted by the UE as described above based on the combined vector according to [Equation 15] and [Equation 16].

Hereinafter, a system structure for transmitting and receiving downlink data according to an embodiment of the present disclosure will be described with reference to FIG. 15B.

In FIG. 15B, in transmitting and receiving downlink data, the central TP receives data precoded by the CPU from the CPU, processes the data and transmits it to the UE, and the local TP precodes data to be transmitted to the UE. Then, the precoded data is processed and transmitted to the UE. According to this, a plurality of TPs composed of a central TP and a local TP can transmit downlink data to the UE, so that the CJT method considered in the present disclosure can be implemented.

More specifically, if the assumption in FIG. 15A is applied for convenience of explanation, the k-th received signal of the UE in the downlink of FIG. 15B can be expressed by [Equation 17] below.

는

번째 UE에 전송될 하향링크 데이터 신호이며(

),

는 L개의 TP를 각각 고려한 프리코딩 벡터(precoding vector)를 나타낸다. 한편, 상기 프리코딩 벡터에 대해

번째 UE의 할당 전력이

일 때, 각 TP에서는 하기 [수학식 18]과 같이 MR 프리코딩을 적용할 수 있다.At this time,

Is

A downlink data signal to be transmitted to the th UE (

),

Denotes a precoding vector considering L TPs, respectively. On the other hand, for the precoding vector

The allocated power of the th UE is

When , MR precoding may be applied to each TP as shown in [Equation 18] below.

If channel compatibility (reciprocity, or duality) is assumed between uplink and downlink, the power of the combined vector for uplink data may be normalized and used as a precoding vector for downlink transmission there is. In this case, the precoding vector may be used as a central precoder applied by the CPU or a local precoder applied by the TP. Meanwhile, in determining the precoding vector, a vector obtained by normalizing the power of a combination vector may be used, or a combination vector may be used, and the present embodiment is not limited thereto.

Meanwhile, even when downlink data is transmitted/received, as described in FIG. 15A, clustering may be assumed such that the TPs support only some UEs belonging to the system, rather than all UEs in the system. When the clustering is assumed, downlink transmission data can be expressed by [Equation 19] as follows.

는 하향 링크 데이터를 위한 프리코딩 벡터,

는 하향 링크 전송 심볼을 나타낸다.At this time,

Is a precoding vector for downlink data,

represents a downlink transmission symbol.

에게 간섭을 미치는 UE의 집합을

로 정의할 수 있다. 그러한 UE의 집합은

를 서빙하는 TP가 서빙하는 모든 UE를 포함할 수 있으며, 하기 [수학식 20]으로 나타낼 수 있다.Meanwhile, as described in FIG. 15A, the clustering

A set of UEs interfering with

can be defined as The set of such UEs is

A TP serving may include all UEs serving, and may be represented by Equation 20 below.

When interfering UEs are considered in a cluster situation, the central MMSE precoder and the distributed MMSE precoder (local precoder) are a heuristic solution, and can be referred to as reciprocity or duality between uplink and downlink. .), the power of the uplink data combination vector can be normalized and used as a normalized precoding vector for downlink transmission.

)을 할당할 수 있다. UE는 DM-RS에 기반하여 하향링크 채널을 정확히 추정하고, 이에 따른 수신 성능의 이득을 얻을 수 있다. 이를 위해 각각 하나의 안테나를 갖는 TP와 UE가 결합 빔포밍(conjugate beamforming, CB)을 통해 하향링크 신호를 송수신하는 상황을 가정하면,

가 normalized SNR이고,

가

및

의 전력 제어(power control) 계수일때, conjugate beamforming(

)를 통해

이 전송하는 신호는 하기와 같은 [수학식 22]으로 나타낼 수 있다.On the other hand, in the downlink of a cell-free massive MIMO-based system to which an embodiment of the present disclosure can be applied, a unique downlink DM-RS (

) can be assigned. The UE can accurately estimate the downlink channel based on the DM-RS and obtain a reception performance gain accordingly. For this purpose, assuming a situation in which a TP and a UE each having one antenna transmit and receive downlink signals through conjugate beamforming (CB),

is the normalized SNR,

go

and

When the power control coefficient of , conjugate beamforming (

)Through the

This transmitted signal can be represented by [Equation 22] as follows.

가 수신하는 신호는 하기와 같은 [수학식 23]으로 나타낼 수 있다.At this time, downlink

The signal received by can be represented by the following [Equation 23].

는 잡음을 나타낸 것이고, 이때 결합 빔포밍 된 유효 채널 이득(effective channel gain,

)는 하기와 같은 [수학식 24]으로 나타낼 수 있다.In [Equation 23],

Denotes noise, where the joint beamformed effective channel gain (effective channel gain,

) can be represented by [Equation 24] as follows.

번째 UE는 유효 채널(

)을 알기 위해서는 충분한 정보를 갖고 있어야 하고, UE가

를 아는 경우,

를 통해 하향링크 데이터(

)를 신뢰성 있게 추정할 수 있다.

를 추정하기 위해 빔포밍을 통해 하향링크 파일롯(하향링크 기준신호, 또는 DM-RS 등을 칭할 수 있다.)을 전송하는 경우와 전송하지 않는 경우로 구분할 수 있다.

The th UE is an effective channel (

), the UE must have sufficient information to know

If you know

Through the downlink data (

) can be reliably estimated.

It can be divided into a case where a downlink pilot (downlink reference signal, or DM-RS, etc.) is transmitted through beamforming and a case where it is not transmitted.

If the DM-RS is not transmitted, the channel is perfectly estimated through the uplink file in a massive MIMO environment with a large number of antennas, and transmission precoding is perfect in an ideal environment such as no mobility. Due to the channel hardening effect, the downlink It is possible to receive only the statistical characteristics of the channel without a link pilot. However, in a real environment, the number of antennas is insufficient to achieve sufficient channel hardening due to various reasons such as channel estimation error, interference, and mobility, so that the UE cannot decode only the statistical characteristics of the channel. Therefore, in order to coherently receive downlink data in a real environment, a DM-RS for estimating a downlink channel through which data is transmitted is required.

)을 할당하고, TP는 동일한 하나의 DM-RS 자원을 통해 UE-specific DM-RS(

)을 전송하며, 이에 따라 UE가 수신하는 신호는 하기 [수학식 25]로 나타낼 수 있다.Accordingly, a reception procedure of a UE through one DM-RS, which is an existing method, is as follows. First, in the network, a unique downlink DM-RS (

) is allocated, and the TP transmits the UE-specific DM-RS (through the same DM-RS resource).

), and thus the signal received by the UE can be represented by [Equation 25] below.

는 수신기의 잡음을 나타낸 것이다. 그 후,

을

로 사영(projection) 시키면 하기 [수학식 26]으로 나타낼 수 있다.In [Equation 25],

represents the noise of the receiver. After that,

second

When projected as , it can be expressed as [Equation 26] below.

를 통해 MMSE에 기반하여 유효 채널(

)를 추정한다. 이때 추정된 유효 채널(

)은 하기와 같은 [수학식 27]으로 나타낼 수 있다.At the receiving end,

Based on the MMSE through the effective channel (

) is estimated. At this time, the estimated effective channel (

) can be represented by [Equation 27] as follows.

In this case, the values of [Equation 27] in the Rayleigh fading channel performing combined beamforming can be expressed by the following equations.

)을 통해

는 하향링크 신호

를 coherent하게 수신할 수 있다.The effective channel estimated in this way (

)Through

is the downlink signal

can receive coherently.

)를

라고 할 때, 하향링크의 용량은 하기와 같은 [수학식 31]에 기반하여 계산될 수 있다.On the other hand, effective channel estimation error (estimation error,

)cast

When it is said, downlink capacity can be calculated based on the following [Equation 31].

)를 추정하여, 유효 채널의 추정 오류(

)가 감소할수록 수신 성능이 향상될 수 있다는 점을 도출할 수 있다.Referring to [Equation 31], the exact effective channel (

), the estimation error of the effective channel (

It can be derived that the reception performance can be improved as ) decreases.

개의 안테나를 갖는 다중 안테나 TP가

개의 안테나를 갖는 다중 안테나

의 상황으로 확장할 수도 있다.The situation described above assumes that the TP and the UE each use one antenna and the UE receives downlink data through one DM-RS. On the other hand, this

A multi-antenna TP having two antennas

Multiple antennas with two antennas

can be extended to the situation of

과

사이의 채널(

)을 통해 전송되는 하향링크 파일롯을 각각

라고 정의하고, 각 하향링크 파일롯에 동일한 프리코더(

)가 적용된다고 가정하면, 특정 시간(예를 들어,

시간) 동안 UE가 수신하는 신호(

)는 하기와 같은 [수학식 32]으로 나타낼 수 있다.

class

channel between

), each downlink pilot transmitted through

, and the same precoder for each downlink pilot (

) is applied, a specific time (e.g.,

time) signal received by the UE during (

) can be represented by the following [Equation 32].

는 같은 DM-RS자원에서 전송된 다른 사용자의 DM-RS에 의한 간섭을 나타낸다.In [Equation 32],

Indicates interference by another user's DM-RS transmitted from the same DM-RS resource.

가 상기 수신 신호(

)로부터 각각 유효 채널(

)를 추정할 수 있다. 그 후, UE는 추정된 유효 채널에 기반하여 하향링크 결합 벡터(

)를 생성할 수 있다. UE는 하향 링크 수신 신호(

)에 대해서 하향링크 결합 벡터를 적용하여 결합함으로써 coherent하게 심볼을 추정할 수 있다.(예를 들어,

, 여기서

는 추정된 심볼의 의미할 수 있다.)

is the received signal (

) from each effective channel (

) can be estimated. Then, the UE based on the estimated effective channel downlink combination vector (

) can be created. The UE receives a downlink signal (

), symbols can be coherently estimated by combining them by applying a downlink combining vector. (For example,

, here

may mean an estimated symbol.)

Meanwhile, according to an embodiment of the present disclosure, several DM-RSs may be used to obtain a reception diversity gain. In this case, different DM-RS resources may be allocated to the central TP and the local TP determined as a result of CSS execution. For example, one DM-RS may be allocated to a central TP, and the remaining DM-RSs may be allocated to local TPs. On the other hand, when multiple DM-RSs are used, when grouping TPs, since the correlation between effective channels of the group must be low to ensure diversity gain, the CPU uses a central TP to secure the diversity gain when performing CSS. can be selected (or can be scheduled).

과

간 유효 채널(

) 간 높은 상관 관계가 존재하므로, 앞서 설명한 바와 같이 여러 TP가 동일한 DM-RS 자원을 할당하여 UE에 파일롯을 전송할 경우(즉, 1개의 DM-RS를 이용하여 파일롯을 전송) UE가 여러 TP에서 전송된 채널의 합인 유효 채널을 추정해도 perfect CSIR(channel state information reception)과 비교하여 수신 성능차이가 크게 나지 않는다. 그러나, 각

과

간 유효 채널(

)의 분포 간에 상관 값이 매우 적다면(예를 들어, 분산 안테나에서 독립적 또는 분산적으로 프리코딩 하는 경우 등), 각 TP별로 DM-RS자원을 따로 할당하여 유효 채널을 각각 추정하는 것이 더 perfect CSIR에 가까운 정확한 채널을 추정할 수 있고, 이렇게 더 정확히 추정한 채널을 통해 수신단을 구성하면, 더 높은 다이버시티 이득을 얻을 수 있다. More specifically, in the existing cellular network, several TPs are gathered in one place, so

class

effective channel between

), as described above, when multiple TPs allocate the same DM-RS resources and transmit pilots to the UE (i.e., transmit pilots using one DM-RS), the UE transmits pilots from multiple TPs. Even if the effective channel, which is the sum of the transmitted channels, is estimated, there is no significant difference in reception performance compared to perfect channel state information reception (CSIR). However, each

class

effective channel between

), if the correlation value between the distributions is very small (for example, in the case of independent or distributed precoding in a distributed antenna, etc.), it is more perfect to estimate the effective channels by separately allocating DM-RS resources for each TP. An accurate channel closer to the CSIR can be estimated, and a higher diversity gain can be obtained if a receiving end is configured using the more accurately estimated channel.

Therefore, the present disclosure proposes a structure for estimating a channel close to the perfect CSIR by using a plurality of DM-RSs in order to increase reception diversity performance through accurate effective channel estimation of the user.

According to the present disclosure, a CPU allocates different DM-RS resources to a central TP and a local TP, and each TP transmits data to a UE based on the allocated DM-RS resources. The reason for allocating DM-RS resources by dividing the Central TP and Local TP is that in the case of the Central TP, the channel correlation of the Central TPs is reflected in the effective channel through centralized precoding. It is to allocate a DM-RS that can be estimated. The remaining DM-RSs can be allocated to local TPs, and as the number of DM-RSs increases, the UE can accurately estimate the effective channels and improve reception performance.

On the other hand, since the overhead increases as the number of DM-RS increases, the number of DM-RSs should be limited. More DM-RSs can be used.

개의 안테나를 갖는 다중 안테나 TP 및

개의 안테나를 갖는 다중 안테나

를 가정하면,

과

간 채널(

)을 통해 전송되는 2개의 햐향링크 파일롯은 예를 들어 각각

와

으로 나타낼 수 있다. 상기 각 파일롯에는 동일한 프리코더(예를 들어,

라 나타낼 수 있다.)가 적용된다. 이때, 특정 시간(예를 들어,

라 할 수 있다.) UE가 수신하는 신호들(예를 들어,

,

와 같이 나타낼 수 있다.)은 각각 하기와 같은 [수학식 33] 및 [수학식 34]으로 나타낼 수 있다.In order to examine the performance gain in the case of using two DM-RSs, it can be assumed that one DM-RS resource is allocated to each of the central TP group and the local TP group by extending the MIMO pilot allocation structure described above. there is. At this time,

A multi-antenna TP having two antennas and

Multiple antennas with two antennas

Assuming

class

Inter-channel (

The two downlink pilots transmitted through ) are, for example, respectively

and

can be expressed as Each pilot has the same precoder (e.g.,

can be expressed as) is applied. At this time, a specific time (eg,

) Signals received by the UE (e.g.,

,

) can be represented by the following [Equation 33] and [Equation 34], respectively.

와

는 각각 Local TP와 Central TP의 DM-RS자원에서 전송된 다른 UE의 DM-RS에 의한 간섭을 나타낸다.

는 수신 신호

와

로부터 각각 유효 채널

과

을 추정한다. 이후, UE는 상기 추정된 2개의 유효 채널 각각에 기반하여 결합 벡터

와

를 생성하고, 상기 생성된 결합벡터의 coherent combining를 통하여 수신 신호(

)를 추정(또는, 검출)할 수 있다.In [Equation 33] and [Equation 34],

and

Indicates interference by DM-RS of another UE transmitted from DM-RS resources of Local TP and Central TP, respectively.

is the received signal

and

each effective channel from

class

to estimate Then, the UE uses a combined vector based on each of the estimated two effective channels.

and

, and the received signal (through coherent combining of the generated combination vector)

) can be estimated (or detected).

과

간에 낮은 상관성을 가질 수 있다. 만약, 두 유효 채널의 분포가 비슷하여 상관성 높은 경우, 한 개의 DM-RS에 기반하여 유효 채널을 추정할 때 보다 여러 개의 DM-RS에 기반하여 정확한 채널 추정을 통해 얻는 다이버시티 이득은 적다. 이와 달리, 만약 두 유효채널의 분포가 달라 채널간 상관성이 낮은 경우, 2개의 DM-RS에 기반하여 각각 채널을 정확하게 추정하여 얻는 성능 이득은 크다.At this time, in order to obtain sufficient receive diversity gain through two DM-RSs, two effective channels to be estimated

class

may have low correlations. If the distributions of two effective channels are similar and the correlation is high, the diversity gain obtained through accurate channel estimation based on multiple DM-RSs is smaller than when estimating an effective channel based on one DM-RS. In contrast, if the distributions of the two effective channels are different and the inter-channel correlation is low, the performance gain obtained by accurately estimating each channel based on the two DM-RSs is large.

Therefore, in order to obtain the diversity gain as described above, the CPU according to an embodiment of the present disclosure determines whether each TP operates as a central TP or a local TP through CSS, and then the central TP and local Multiple DM-RSs can be assigned to a TP.

On the other hand, as described above, when multiple DM-RSs are allocated to the central TP and the local TP, it is necessary to instruct the UE to perform coherent reception combining, and a separate indicator for this (eg, coherent Joint Reception Indicator (CJRI) may be included in a message and transmitted to the UE. A detailed procedure considering such CSS and multiple DM-RS allocation will be described with reference to FIG. 16 .

16 is a diagram illustrating a procedure for transmitting and receiving downlink data when multiple DM-RSs are allocated to a central TP and a local TP determined according to a CSS execution result according to an embodiment of the present disclosure.

Meanwhile, steps 16-100, 16-110, 16-120, 16-130, 16-140, 16-150, 16-160, 16-170, and 16-180 shown in FIG. Some of steps 16-190, 16-200, 16-210, 16-220, and 16-230 may be omitted, may be performed sequentially, or may be performed simultaneously. Hereinafter, it will be described in detail.

In step 16-100, the UE 1220 transmits an uplink pilot (uplink reference signal or SRS (sounding A reference signal) can be transmitted.

Upon receiving the uplink pilot from the UE, the TP 1210 generates channel information (or may be channel state information) between the UE 1220 and the TP 1210 based on the uplink pilot in steps 16-110. can do. And, the TP 1210 may transmit the channel information to the CPU 1200 in steps 16-120. In this case, the channel information may mean channel information (large-scale channel information) for clustering or CSS rather than information on all channels for precoding.

Upon receiving the channel information from each TP (which may be a TP group or a set of TPs) 1210, the CPU 1200 may check whether CSS is performed based on at least one of various criteria in steps 16 to 130. 16 illustrates an example in which the CPU 1200 receives channel information from the TP 1210 and determines whether CSS is performed based on the channel information and various criteria, but the present disclosure is not limited thereto. For example, as described in FIG. 13, CSS may also be initiated by a predetermined period, a specific threshold value, or a CSS performance request of the UE 1220, in which case steps 16-100 to 16-120 are performed. can be omitted.

If the CPU 1200 determines to perform the CSS, the CPU 1200 performs the CSS based on a preset criterion (eg, a criterion set by a network operator) in steps 16-140. can be done Meanwhile, in step 16-130, the CPU 1200 may determine clustering based on the channel information (or channel state information) received in step 16-120. If clustering is determined in steps 16-130, the CPU 1200 may perform clustering in steps 16-140. In this case, when the CPU 1200 determines to perform clustering, CSS may not be performed, but is not limited thereto, and the clustering and CSS may be performed simultaneously or sequentially.

Meanwhile, when it is determined to perform CSS in step 16-140 and the CSS execution is completed, the CPU 1200 transmits the CSS execution result to each TP 1210 in step 16-150. Meanwhile, the CSS execution result is about whether each TP 1210 operates as a Central TP or a Local TP. According to an embodiment of the present disclosure, the CPU provides CSS execution results to each TP 1210 In order to notify the TP, a central/local indicator indicating operation as a central TP or a local TP may be transmitted to each TP 1210, and the indicator may consist of 1 bit. For example, when the indicator indicates a value of '0', it may mean instructing to operate as a local TP, and when indicating a value of '1', instructing to operate as a central TP can mean Meanwhile, each of the Central TP or Local TP may be included in a Central TP group (or Central TP set) composed of a plurality of Central TPs or a Local TP group (or Local TP set) composed of a plurality of Local TPs. In this case, CPU Step 1200 may instruct one of the TPs included in the central TP group or the local TP group to operate as a central TP or a local TP. At this time, the TPs of the group to which the TPs instructed by the CPU 1200 to operate as Central TPs or Local TPs belong may operate as Central TPs or Local TPs as instructed (that is, the CPU may operate as Central TPs or Local TPs in units of TP groups). It can also be instructed to operate as a TP or local TP).

Upon receiving the indicator transmitted from the CPU 1200 in steps 16-150, the TP 1210 can determine whether it will operate as a central TP or a local TP based on the indicator. After confirming which TP it will operate in this way, the central TP processing process (steps 16-160) or the local TP processing process (steps 16-170) as described above is performed to obtain a centralized precoder (central TP). precoder) or distributed precoder (local precoder). Since the detailed description of the processing of each TP has been described above, it will be omitted here.

Meanwhile, in order to obtain a receive diversity gain through the central precoder and the local precoder generated as described above, the CPU 1200 allocates several DM-RS resources to the central TP and the local TP, and the UE 1220 Signals received from each TP 1210 may be coherently combined and received based on effective channels estimated through the DM-RS.

Therefore, in step 16-180, the CPU 1200 determines which TPs will share the multiple DM-RS resources to be transmitted to the UE 1220, and in step 16-190, the CPU 1200 determines the number of DM-RS resources. Information on allocation can be transmitted to the Central TP (or a set or group of Central TPs) and a Local TP (or a set or group of Local TPs) 1210 . Meanwhile, in steps 16-180, the CPU 1200 may allocate DM-RS resources so that the central TP shares one DM-RS resource. That is, in this case, the central TP may share the same DM-RS resource. The TP (which may be either a Central TP or a Local TP, or may be both) 1210 receiving information on DM-RS allocation from the CPU 1200 receives data in steps 16-200 Information on DM-RS allocation may be transmitted to the UE 1220 to be assigned.

Since the UE 1220 does not know whether to coherently receive signals transmitted based on multiple DM-RSs, it is necessary to instruct the UE 1220 to perform coherent reception combining before transmitting downlink data. To this end, the TP (which may be either a Central TP or a Local TP) 1210 coherently receives a data stream transmitted to the UE 1220 corresponding to a plurality of DM-RSs in steps 16-210. It may transmit an indicator (coherent joint reception indicator, CJRI) (or may be referred to as coherent reception indicator, CRI) indicating that. In other words, when a data stream precoded with the same precoder as the DM-RS assigned to each of the plurality of TPs is transmitted from each of the plurality of TPs 1210 to the UE 1220, the UE 1220 either prior to or at the same time An indicator instructing to coherently receive a data stream (eg, CJRI described above) may be received.

Meanwhile, in JT of CoMP, a local TP generally operates in a non-coherent joint transmission (NCJT) method that transmits non-coherently. That is, it is common for local TPs to transmit data by independent scheduling (local scheduling) without synchronizing transmission with cooperating TPs. Meanwhile, the local TP according to the present disclosure may transmit data coherently in synchronization with cooperative TPs such as the central TP in steps 16-220. In other words, the local TP may select either non-coherent data transmission or coherent data transmission, and may indicate this to the UE 1220 through CJRI. In steps 16-230, the UE 1220 may receive data coherently or non-coherently according to a method in which the TP 1210 transmits data (coherent transmission or non-coherent transmission). Therefore, the operation of the UE 1220 may vary depending on whether the UE 1220 receives the CJRI from the TP, which will be described with reference to FIG. 17 .

17 is a diagram illustrating a sequence in which a UE operates according to CJRI reception or not according to an embodiment of the present disclosure.

Referring to FIG. 17, in step 17-100 of FIG. 17, if the UE receives CJRI from the TP, this means that cooperating TPs synchronize and transmit data streams to the UE, so the UE transmits the data stream to the UE in step 16-230. Data streams transmitted from each TP and Local TP can be coherently received. In this case, since the resource allocation information of the local TP can be checked through the resource allocation information of the central TP, the local TP does not need to notify the UE of separate resource allocation indication information through downlink control information (DCI).

Meanwhile, when the UE receives the CJRI, the operation may be performed differently depending on whether the Local TP and the Central TP share the DM-RS.

In steps 17-110, if the Local TP shares a DM-RS with the Central TP (which may mean a case in which a DM-RS port is shared, or a case in which the UE is allocated only one DM-RS port), In step 17-130, the UE performs channel estimation through one DM-RS and generates a reception filter through the estimated channel. Thereafter, one DCI (which may mean a DCI transmitted from a central TP) may be received, a location of a resource allocated thereto may be confirmed through the DCI, and data may be received.

In contrast, in step 17-110, if the Local TP does not share the DM-RS with the Central TP (when each DM-RS port is used independently, or if the UE is assigned two or more DM-RS ports) case), in step 17-140, the UE estimates one channel based on several DM-RSs allocated to the UE and creates a reception filter through the estimated channel. Thereafter, one DCI among DCIs transmitted from several TPs may be received, a resource location allocated to the TP based on the received DCI may be confirmed, and data may be received.

On the other hand, if the UE does not receive CJRI in 17-100, this means that the Local TP transmits the data stream to the UE without synchronizing with other cooperative TPs. allocates independent DM-RS transmission and resources to the UE, and transmits a DCI indicating a resource allocation location. Accordingly, the UE non-coherently receives and combines non-coherently transmitted data streams. That is, the UE estimates the channel based on the DM-RS transmitted from each TP for demodulation in steps 17-120, and each TP independently transmits to the UE based on the DCI independently transmitted by each TP After finding out the locations of resources of one's own, the location can be demodulated based on the channel estimated through the DM-RS of the corresponding TP.

Meanwhile, some of steps 17-100, 17-110, 17-120, 17-130, and 17-140 of FIG. 17 may be omitted, may be performed sequentially, or may be performed simultaneously. .

Meanwhile, as described above, CSS performed to obtain optimal data transmission/reception performance may be implemented in the form of various algorithms, which will be described with reference to FIG. 18 .

18 is a diagram illustrating an example of a CSS algorithm that can be performed in a CPU according to an embodiment of the present disclosure.

(이때,

는 미리 정해진 조건에 따라 지정될 수 있고, 또는 CPU에 의해 임의로 지정될 수 있다.)를 지원하는 TP의 집합(예를 들어,

라 할 수 있다.)을 확인하고, 이에 속하는 TP를 모두(또는, 일부) Central TP로 바꾸는 절차를 도시하였다.Referring to FIG. 18, as an example of CSS algorithm implementation,

(At this time,

may be designated according to a predetermined condition, or may be arbitrarily designated by the CPU.) a set of TPs (eg,

It can be called.) and shows a procedure of changing all (or some) of the TPs belonging to it to the Central TP.

)와 계산 복잡도(예를 들어,

)를 고려하여 상기 CPU에서 지원 가능한 Central TP의 전체 수(예를 들어,

)를 계산할 수 있다.First, in steps 18-100, the CPU determines whether to perform CSS based on at least one of several criteria. For example, as described above, when a CSS period is set, the CPU may periodically perform CSS based on the period, or determine whether to perform CSS based on a certain criterion such as a threshold value, Alternatively, when a request for performing CSS is received from the UE, it may be determined whether or not to perform CSS. In step 18-100, if the CPU determines not to perform CSS, data can be transmitted/received based on the existing central TP or local TP. Meanwhile, if it is determined to perform CSS, operations of steps 18-110 are performed. At this time, the CPU has an available front hall capacity (eg,

) and computational complexity (e.g.

), the total number of central TPs that can be supported by the CPU (eg,

) can be calculated.

)을 확인할 수 있다. 예를 들어, CPU는 throughput 정보에 기반하여 수율(throughput 값)이 가장 낮은

를 확인하고, UE를 서빙 하는 TP 집합(

)을 확인할 수 있다. 또는, CPU는 수율이 가장 낮은 UE부터 수율이 낮은 순서대로 미리 정해진 수 또는 미리 정해진 조건에 따라 결정된 수 또는 상기 CPU에 의해 계산된 수에 상응하는 UE들을 확인하고, 확인된 UE들을 서빙 하는 TP 집합을 확인할 수 있다. 이때, CPU에 의해 계산된 수는 예를 들어 18-100 단계에서 계산된 상기

를 의미할 수 있다.In steps 18-110, the CPU may identify a UE that satisfies a predetermined condition based on throughput information for each of the at least one UE. In addition, the CPU sets the TP serving the identified UE (

)can confirm. For example, CPU has the lowest yield (throughput value) based on throughput information.

, and a set of TPs serving the UE (

)can confirm. Alternatively, the CPU identifies UEs corresponding to a predetermined number or a number determined according to a predetermined condition or a number calculated by the CPU in the order of the lowest yield from the UE with the lowest yield, and TP set serving the identified UEs can confirm. At this time, the number calculated by the CPU is, for example, the number calculated in steps 18-100.

can mean

)에 속한 TP 중 Local TP에 해당하는 TP를 모두 Central TP로 전환할 수 있다. 한편, 도 18의 설명에서는 Local TP에 해당하는 TP를 모두 Central TP로 전환할 수 있다고 하였으나, 이에 국한되지 않고, Local TP 중 적어도 하나를 Central TP로 전환할 수도 있다. 이때, Local TP에 해당하는 TP를 Central TP로 전환하는 것은 Local TP가 상기 Central TP가 속한 그룹으로 변경되는 것을 의미할 수 있다. 또는, Local TP에 해당했었던 TP를 Central TP로 전환하는 것은 Local TP가 상기 Central TP가 수행하는 동작을 수행하도록 지시하는 것을 의미할 수 있다. 또한, 도 18의 설명에서는 Local TP를 Central TP로 전환하는 것을 예로 들었으나, 이에 국한되지 않으며 CPU는 계산 복잡도 및 프론트 홀 용량 등 여러 가지 기준에 기반하여 확인된 TP 집합(

) 중 Central TP에 해당하는 TP를 Local TP로 전환하는 것도 가능하다. 한편, 18-120 단계는 CPU가 18-110 단계에서 확인된 TP 집합에서, Central TP 및 Local TP의 비율을 조절(adjust)하는 동작을 의미할 수도 있다. 즉, 미리 결정된 Central TP 및 Local TP 간 비율에 기반하여, CPU는 18-120 단계에서, 상술한 바와 같이 기존의 Central TP를 Local TP로 동작하도록 지시 또는 전환하거나, 기존의 Local TP를 Central TP로 동작하도록 지시 또는 전환함으로써 상기 TP 집합에서 Central TP 및 Local TP 간 비율을 조절할 수 있다. 이때, Central TP 및 Local TP 간 비율은 계산 복잡도 및 프론트 홀 용량 중 적어도 하나에 기반하여 결정되는 것을 특징으로 할 수 있다. In steps 18-120 thereafter, the CPU determines the identified TP set (

), all TPs corresponding to Local TPs can be converted to Central TPs. Meanwhile, in the description of FIG. 18, it is said that all TPs corresponding to local TPs can be converted to central TPs, but the present invention is not limited thereto, and at least one of local TPs may be converted to central TPs. In this case, switching a TP corresponding to a local TP to a central TP may mean that the local TP is changed to a group to which the central TP belongs. Alternatively, switching a TP corresponding to a local TP to a central TP may mean instructing the local TP to perform an operation performed by the central TP. In addition, in the description of FIG. 18, the conversion of the Local TP to the Central TP is taken as an example, but it is not limited thereto, and the CPU determines the TP set based on various criteria such as computational complexity and front haul capacity (

), it is also possible to convert the TP corresponding to the Central TP to a Local TP. Meanwhile, step 18-120 may mean an operation in which the CPU adjusts the ratio of the central TP and the local TP in the TP set identified in step 18-110. That is, based on the predetermined ratio between the central TP and the local TP, the CPU instructs or switches the existing central TP to operate as the local TP, or converts the existing local TP to the central TP, as described above, in steps 18-120. By instructing or switching to operate, the ratio between the central TP and the local TP in the TP set can be adjusted. In this case, the ratio between the central TP and the local TP may be determined based on at least one of computational complexity and front haul capacity.

)보다 작은 경우, 다시 18-110 단계로 돌아갈 수 있다. 이와 달리 만약 Central TP의 수가

와 같거나 또는 큰 경우, CPU는 CSS 수행을 종료할 수 있다.Then, in steps 18-130, the CPU may check whether the CSS has ended. The CPU can determine whether the CSS has ended based on various criteria. For example, if the number of central TPs is the total number of supportable central TPs calculated in steps 18-110 (

), it is possible to return to steps 18-110 again. In contrast, if the number of central TPs is

When equal to or greater than , the CPU may end CSS execution.

Meanwhile, some of steps 18-100, 18-110, 18-120, and 18-130 of FIG. 18 may be omitted and may be performed sequentially or concurrently.

As described above, by performing CSS according to an algorithm considering a given front haul capacity and computational complexity, a communication system providing more improved services can be implemented.

19 is a diagram illustrating a CSS framework based on a TP with embedded eCPRI according to an embodiment of the present disclosure.

Referring to FIG. 19, as a result of performing the CSS, where to perform layer splitting for each TP (split 1 or split 2 shown in FIG. 14) is determined, and accordingly, the CU and DU procedures change. Since the Central TP or the Local TP is determined as a result of the CSS execution, the TP processing process has been described above, so a detailed description thereof will be omitted.

On the other hand, each TP can perform all procedures corresponding to different layer splitting, and by performing different procedures according to the result of performing the CSS, optimal data transmission and reception performance is achieved within the constraints of the amount of CPU calculation and front haul usage. You can get it.

20 and 21 are diagrams showing the yield of all users when an embodiment of the present disclosure is applied.

20 and 21 show the yield of total users as CDF when 2 UEs or 4 UEs are supported per TP, respectively, in a system with 100 TPs and 20 UEs. The solid line is the case when MMSE is performed considering the interference of all users, and the dotted line is a precoder that considers only the interference of UEs belonging to its cluster (or the TP set to which it belongs) using TP clustering. How to create and transmit. At this time, the leftmost dotted line graph is the L-MMSE performance using a distributed scalable MMSE precoder, and the rightmost dotted line graph is the performance of the P-MMSE method using a centralized scalable MMSE precoder. The middle graph shows the CDF while trying to centrally transform the random TP generation. In both FIGS. 20 and 21, it can be seen that the performance is traded-off as the percentage of the central TP (Central TP) varies from 0% to 100%. Therefore, it is necessary to select a precoder generation method to have the maximum performance of the TP according to complexity and front haul load.

22 and 23 are diagrams illustrating the performance of users in the bottom 10% when an embodiment of the present disclosure is applied.

22 and 23 show the performance of the lower 10% user when supporting 2 UEs or 4 UEs per TP, respectively, in a system with 100 TPs and 20 UEs, as shown in FIGS. 20 and 21 .

At this time, it can be confirmed that the performance of the bottom 10% user can be improved by 2 times or 2.5 times in each embodiment through control of the centralized or distributed precoder generation method of the TP. In the conventional method, all TPs must operate only with distributed precoders in consideration of the constraints caused by the load of the front haul, whereas by applying the CSS structure of the present disclosure to control the TP precoder generation method according to the capacity constraints of the front haul, the bottom 10 % User's performance improved up to 2x to 2.5x.

24 and 25 are diagrams illustrating the performance of users in the bottom 10% when an embodiment of the present disclosure is applied.

24 and 25 are diagrams illustrating results of evaluating the yield of the lower 10% user when supporting 2 UEs and 4 UEs per TP, respectively, in a system with 100 TPs and 20 UEs. In FIG. 24, the average number of TPs serving the UE is 10, and in FIG. 25, the average number of TPs serving the UE is 20.

24 and 25, the Distributed-MMSE: Clustering graph shows the performance of the L-MMSE scheme in which all TPs are local TPs and distributed scalable MMSE precoders are used for the UEs they serve. The Partial_MMSE:Clustering graph shows the performance of the P-MMSE method using a centralized scalable MMSE precoder with all TPs performing clustering, the Central TP. The Multicell-MMSE: Full Clustering graph shows the maximum performance that can be theoretically obtained when all TPs are central TPs and serve all UEs.

The Hybrid-MMSE: Clustering + Random CSS graphs in FIGS. 24 and 25 show the yield of the bottom 10% users while randomly changing TPs to central TPs (Random CSS). According to the Hybrid-MMSE: Clustering+Random CSS graphs in FIGS. 24 and 25, it can be seen that the performance is traded-off as the percentage of the central TP varies from 0% to 100%.

In the Hybrid-MMSE: Clustering + Greedy CSS graphs in FIGS. 24 and 25, unlike the above-described Random CSS, a method of preemptively switching a TP serving a UE of lower performance to a central TP (Greedy CSS) is used. shows the performance of According to the Hybrid-MMSE: Clustering + Greedy CSS graphs in FIGS. 24 and 25, the performance of the lower 10% UE improves more rapidly than the case according to the Random CSS method while the percentage of the central TP changes from 0% to 100% can confirm that it is.

Therefore, referring to the graphs shown in FIGS. 24 and 25, it can be seen that it is necessary to select an optimal precoder generation method to obtain maximum performance according to CPU calculation complexity and front haul load.

26 is a diagram showing a result of comparing user bit error performance according to presence or absence of channel correlation between TP groups determined by performing CSS when one or two DM-RSs are used when an embodiment of the present disclosure is applied; FIG. am.

의 각 TP 그룹으로부터 오는 유효 채널 벡터를

라고 할 때, 유효 채널 간 송신 correlation

을 하기와 같은 수학식으로 나타낼 수 있다.26 assumes that the total number of antennas of the central TP and the local TP is 100, respectively, and a single antenna situation is assumed.

The effective channel vector from each TP group of

, transmission correlation between effective channels

can be represented by the following equation.

는 상관계수(correlation coefficient)로,

이면, Central TP와 Local TP간 correlation이 존재하지 않는 것이고, 0.9인 경우에는 correlation이 높은 환경이다. 따라서 송신 correlation이 반영된 그룹의 유효 채널 벡터는 하기와 같은 [수학식 36]으로 나타낼 수 있다.At this time,

is the correlation coefficient,

If , correlation between the central TP and local TP does not exist, and if 0.9, the correlation is high. Therefore, the effective channel vector of the group reflecting the transmission correlation can be expressed by [Equation 36].

가 DM-RS를 통해 유효 채널을 정확히 추정했다고 가정하면, 1개의 DM-RS를 사용하였을 때에는 한 개의 유효 채널이 생성되어(

)을 추정하게 되고, 2개의 DM-RS를 사용하였을 때는

와

를 각각 추정하게 된다. 결과적으로 유효 채널 이 하나가 될 때는 각 채널의 합을 추정하여 수신하고, 유효 채널이 2개가 될 때는 이들을 통해 통해 수신한 값을 결합(combining) 한다. 이때 1개의 DM-RS를 사용한 경우 안테나를 통한 diversity 이득을 충분히 얻을 수 없는 반면, 2개의 DM-RS를 사용한 경우에는

와

가 독립이면서 동일한 분포를 가질 때 최대의 diversity를 얻을 수 있다. 도 26에 따르면,

일수록, 즉 Central TP 및 Local TP 간 correlation이 없을 수록 비트오류 성능이 향상되는 것을 확인할 수 있다.

Assuming that the effective channel is accurately estimated through the DM-RS, when one DM-RS is used, one effective channel is generated (

) is estimated, and when two DM-RSs are used,

and

are estimated respectively. As a result, when there is one effective channel, the sum of each channel is estimated and received, and when there are two effective channels, values received through them are combined. At this time, when one DM-RS is used, the diversity gain through the antenna cannot be sufficiently obtained, whereas when two DM-RS are used,

and

The maximum diversity can be obtained when are independent and have the same distribution. According to Figure 26,

It can be seen that the bit error performance improves as the number of TPs increases, that is, as there is no correlation between the central TP and the local TP.

27 is a diagram illustrating a structure of a UE to which an embodiment of the present disclosure may be applied.

Referring to FIG. 27 , the UE may include a transceiver 2710, a controller 2720, and a memory 2730. In the present disclosure, the control unit may be defined as a circuit or an application-specific integrated circuit or at least one processor.

The transceiver 2710 may transmit and receive signals. The transceiver 2710 may receive a data stream (or data or signal) transmitted from, for example, a central TP and a local TP.

The control unit 2720 may control the overall operation of the UE according to an embodiment proposed in the present disclosure. For example, the control unit 2720 may control signal flow between blocks to perform an operation according to the above-described diagram (or flowchart or flowchart).

The memory 2730 may store at least one of information transmitted/received through the transceiver 2710 and information generated through the controller 2720.

28 is a diagram showing the structure of a TP to which an embodiment of the present disclosure can be applied.

Referring to FIG. 28 , a TP may include a transceiver 2810, a controller 2820, and a memory 2830. In the present disclosure, the control unit may be defined as a circuit or an application-specific integrated circuit or a single processor.

The transceiver 2810 may include a communication unit or a network interfacing unit, and the transceiver 2810 according to an embodiment of the present disclosure is connected to the CPU through a front hole. signals can be transmitted and received. In addition, the TP can transmit and receive signals with the UE through the transceiver 2810.

The control unit 2820 may control the overall operation of the TP according to the embodiment proposed in the present disclosure. For example, the control unit 2820 may control signal flow between blocks to perform an operation according to the above-described diagram (or flowchart or flowchart). For example, when an instruction is received from the CPU to operate as a central TP or a local TP, the TP may be controlled to perform an operation according to the instruction.

The memory 2830 may store at least one of information transmitted/received through the transceiver 2810 and information generated through the controller 2820 .

29 is a diagram illustrating a structure of a network entity to which an embodiment of the present disclosure may be applied.

Referring to FIG. 29 , a network entity (CPU, which may refer to a central unit (CU) according to an embodiment of the present disclosure) may include a transceiver 2910, a controller 2920, and a memory 2930. can In the present disclosure, the control unit may be defined as a circuit or an application-specific integrated circuit or a single processor.

The transceiver 2910 may include a communication unit or a network interfacing unit, and the transceiver 2910 according to an embodiment of the present disclosure is connected to the TP through a front hole. signals can be transmitted and received. In addition, signals can be transmitted and received with other network entities (eg, MME and gateway of LTE system, or AMF and SMF of 5G (NR) system) through the transceiver 2910. .

The controller 2920 may control overall operations of network entities according to an embodiment proposed in the present disclosure. For example, the control unit 2920 may control signal flow between blocks to perform an operation according to the above-described diagram (or flowchart or flowchart). For example, according to an embodiment of the present disclosure, it is possible to determine whether to perform a CSS for controlling a ratio between a central TP and a local TP, and control the network entity to perform the CSS.

The memory 2930 may store at least one of information transmitted/received through the transceiver 2910 and information generated through the controller 2920 .

Meanwhile, in the present disclosure, a network entity may mean an external server or an external network entity configured to control at least one TP.

According to various embodiments of the present disclosure as described above, in a communication system composed of a plurality of TPs (or DUs) and a CPU (or CU) controlling them, calculation complexity allowed for the CPU controlling the plurality of TPs and A precoding method considering the capacity of a front hall, which is a signal transmission and reception path between a CPU and a TP, is provided. According to this, when transmitting and receiving uplink data or downlink data between a UE and a plurality of TPs according to the CoMP scheme, the UE can have optimal data transmission and reception performance.

In the present disclosure, a communication system based on a star topology in which at least one TP (or DU) is directly connected to one CPU (or CU) by 1-hop has been described, but the present disclosure is not limited thereto. The embodiments proposed in the present disclosure can be applied to various communication systems such as a communication system in which a plurality of CUs exist, a multi-hop communication system in which DU is connected to another DU, or a communication system according to various topologies such as a bus or mesh. can

In the drawings for explaining the method of the present disclosure, the order of description does not necessarily correspond to the order of execution, and the order of precedence may be changed or executed in parallel.

Alternatively, drawings describing the method of the present disclosure may omit some components and include only some components within a range that does not impair the essence of the present invention.

On the other hand, in the present specification and drawings, preferred embodiments of the present disclosure are disclosed, and although specific terms are used, they are only used in a general sense to easily explain the technical content of the present disclosure and help understanding of the present invention, It is not intended to limit the scope of this disclosure. In addition to the embodiments disclosed herein, it is obvious to those skilled in the art that other modified examples based on the technical spirit of the present disclosure may be implemented.

That is, although specific embodiments have been described in the detailed description of the present disclosure, various modifications are possible without departing from the scope of the present disclosure, and the scope of the present disclosure should not be limited to the described embodiments. and should be defined by not only the scope of claims described later, but also those equivalent to the scope of these claims.

Claims (14)

In a method of a network entity in a communication system,
Determining whether to perform scheduling for adjusting a ratio between a plurality of first transmission points (TPs) included in a first group serving at least one terminal and a plurality of second TPs included in a second group;
determining whether to switch at least one second TP among the plurality of second TPs to a first TP based on information associated with the at least one terminal when it is determined to perform the scheduling; and
Transmitting an indicator instructing the at least one second TP to operate as the first TP when it is determined to switch the at least one second TP to the first TP;
When operating as the first TP, precoding is performed on a signal in the network entity;
In the case of operating as the second TP, precoding of a signal is performed in the second TP. According to claim 1,
determining whether to switch at least one first TP among the plurality of first TPs to a second TP based on information associated with the at least one terminal when it is determined to perform the scheduling; and
When it is determined to switch the at least one first TP to the second TP, transmitting an indicator instructing the at least one first TP to operate as the second TP How to. According to claim 1,
allocating a plurality of demodulation-reference signal (DM-RS) ports to each of the plurality of first TPs included in the first group and the plurality of second TPs included in the second group; and
Transmitting the information on the DM-RS port to the plurality of first TPs included in the first group and the plurality of second TPs included in the second group;
The plurality of first TPs included in the first group share the same DM-RS port. According to claim 3,
When the plurality of DM-RS ports are allocated to each of the plurality of first TPs included in the first group and the plurality of second TPs included in the second group, the plurality of Characterized in that a coherent joint reception indicator (CJRI) indicating to receive data to be transmitted from the first TP of and the plurality of second TPs included in the second group in synchronization is transmitted to the at least one terminal How to. According to claim 1,
Whether or not the scheduling is performed depends on at least one of channel state information for each of the at least one terminal, a predefined period, or whether a message requesting to perform the scheduling is received from the at least one terminal. A method characterized in that it is determined based on. The method of claim 1, wherein determining whether to switch the at least one second TP to the first TP comprises:
Identifying a first terminal having the lowest throughput value among the at least one terminal based on throughput information corresponding to each of the at least one terminal;
identifying a third TP serving the first terminal among the at least one second TP; and
and determining whether to switch the third TP to the first TP. According to claim 1,
In the communication system, the number of the plurality of first TPs included in the first group and the plurality of second TPs included in the second group is greater than the number of the at least one terminal,
The method of claim 1 , wherein the maximum number of TPs that can be included in the first group is determined based on at least one of a computational complexity and a fronthaul capacity that the network entity can allow. In a network entity of a communication system, the network entity comprises:
A transceiver for transmitting and receiving signals to and from a plurality of TPs through a fronthaul; and
Determining whether to perform scheduling to adjust a ratio between a plurality of first transmission points (TPs) included in a first group serving at least one terminal and a plurality of second TPs included in a second group;
When it is determined to perform the scheduling, determine whether to switch at least one second TP among the plurality of second TPs to a first TP based on information associated with the at least one terminal;
and a control unit for transmitting an indicator instructing the at least one second TP to operate as the first TP through the transceiver when it is determined to convert the at least one second TP into the first TP; ,
When operating as the first TP, precoding is performed on a signal in the network entity;
When operating as the second TP, the network entity characterized in that precoding for a signal is performed in the second TP. According to claim 8,
The control unit,
When it is determined to perform the scheduling, determining whether to switch at least one first TP among the plurality of first TPs to a second TP based on information associated with the at least one terminal;
When it is determined to switch the at least one first TP to the second TP, an indicator instructing the at least one first TP to operate as the second TP is transmitted through the transceiver unit. network entity. According to claim 8,
The control unit,
Allocating a plurality of demodulation-reference signal (DM-RS) ports to each of the plurality of first TPs included in the first group and the plurality of second TPs included in the second group;
transmits information on the DM-RS port to the plurality of first TPs included in the first group and the plurality of second TPs included in the second group through the transceiver;
The plurality of first TPs included in the first group share the same DM-RS port. According to claim 10,
When the plurality of DM-RS ports are allocated to each of the plurality of first TPs included in the first group and the plurality of second TPs included in the second group, the plurality of Characterized in that a coherent joint reception indicator (CJRI) indicating to receive data to be transmitted from the first TP of and the plurality of second TPs included in the second group in synchronization is transmitted to the at least one terminal A network entity that does. According to claim 8,
Whether or not the scheduling is performed depends on at least one of channel state information for each of the at least one terminal, a predefined period, or whether a message requesting to perform the scheduling is received from the at least one terminal. Network entity, characterized in that determined based on. According to claim 8,
The control unit,
Identifying a first terminal having the lowest throughput value among the at least one terminal based on throughput information corresponding to each of the at least one terminal;
Identifying a third TP serving the first terminal among the at least one second TP;
and determining whether to switch the third TP to the first TP. According to claim 8,
In the communication system, the number of the plurality of first TPs included in the first group and the plurality of second TPs included in the second group is greater than the number of the at least one terminal,
The network entity, characterized in that the maximum number of TPs that can be included in the first group is determined based on at least one of a computational complexity and a fronthaul capacity that the network entity can tolerate.

Images