KR102156192B1 - Method and apparatus of pdcp reordering considering multi-flow in dual connectivity system - Google Patents

Abstract

The present invention relates to a method and apparatus for reordering PDCP considering multi-flow in a wireless communication system supporting dual connectivity. According to the present invention, in a packet data convergence protocol (PDCP) entity of a terminal (UE) configured with a macro base station (Macro eNB) and a small base station (small eNB) and dual connectivity, multi-flow (multi- PDCP Service Data Units (SDUs) taking into account flow), provides a method for reordering. According to the present invention, when a UE is configured with dual connectivity with a macro base station and a small base station, in performing multi-flow downlink reception, it is driven when a PDCP PDU is received out of sequence by the PDCP entity of the UE. PDCP SDUs may be rearranged based on the standby timer, PDCP SDUs may be delivered to a higher layer in ascending order, and transmission efficiency may be improved.

Description

PPCP rearrangement method and apparatus considering multi-flow in a dual connection system {METHOD AND APPARATUS OF PDCP REORDERING CONSIDERING MULTI-FLOW IN DUAL CONNECTIVITY SYSTEM}

The present invention relates to wireless communication, and more particularly, to a method and apparatus for reordering PDCP considering multi-flow in a wireless communication system supporting dual connectivity.

In a specific area such as a hotspot inside a cell, a particularly high demand for communication occurs, and in a specific area such as a cell edge or a coverage hole, the reception sensitivity of radio waves may be deteriorated. With the development of wireless communication technology, small cells, for example, pico cells, in a macro cell for the purpose of enabling communication in areas such as hot spots, cell boundaries, and coverage holes. (Pico Cell), Femto Cell, Micro Cell, remote radio head (RRH), relay, and repeater are installed together. This network is called a Heterogeneous Network (HetNet). In a heterogeneous network environment, a macro cell is a cell having a relatively large coverage, and a small cell such as a femto cell and a pico cell is a cell having a small coverage. Compared to macro cells, small cells such as femto cells and pico cells use low power and are therefore referred to as low power networks (LPNs). Coverage overlap occurs between multiple macro cells and small cells in a heterogeneous network environment.

The terminal may configure dual connectivity through two or more of the base stations configuring at least one serving cell. Dual connectivity is an operation in which a corresponding terminal consumes radio resources provided by at least two different network points (eg, a macro base station and a small base station) in a radio resource control connection (RRC_CONNECTED) mode. In this case, the at least two different network points may be connected through a non-ideal backhaul.

In this case, one of the at least two different network points may be called a macro base station (or a master base station or an anchor base station), and the other may be called a small base station (or a secondary base station or an assisting base station or a slave base station).

In general, a wireless communication system has a single flow structure in which a service is provided to a terminal through one radio bearer (RB) for one EPS bearer service. However, in the case of a wireless communication system supporting dual connectivity, a service may be provided to a terminal through two RBs respectively configured for a macro cell and a small cell for one EPS bearer instead of one RB. That is, a service may be provided to the terminal through multi-flow. In the above, one RB is provided through only the macro cell, and the other RB may be configured through two base stations corresponding to the macro cell and the small cell. In other words, one RB may be configured in a single base station and the other RB may be configured in a form in which two base stations are split (Bearer split).

In the case of RLC AM (Acknowlegdged Mode), the RLC entity of the UE in downlink rearranges the RLC PDU when the received RLC Packet Data Unit (PDU) is sequentially received out of order. In the case of RLC AM, the RLC PDU that is missed at the receiving side may be retransmitted by the transmitting side. The RLC entity reassembles an RLC Service Date Unit (SDU) based on the rearranged RLC PDU, and sequentially transfers it to an upper layer (ie, a PDCP entity). In the case of RLC AM, sequential delivery is possible through reordering and retransmission of RLC PDUs. In other words, the PDCP entity must sequentially receive RLC SDUs except for re-establishment of the lower layer. However, in the case of a terminal in which multi-flow is configured, the RLC entity for the small base station and the RLC entity for the macro base station are divided to receive each RLC PDU, and can deliver the RLC SDU to a higher layer (ie, PDCP layer). In this case, the PDCP entity cannot expect sequential reception of RLC SDUs. Accordingly, in the case of a multi-flow-configured terminal, a PDCP rearrangement scheme is required for ascending delivery of PDCP SDUs from a PDCP entity to an upper layer.

An object of the present invention is to provide a method and apparatus for rearranging PDCP in consideration of multi-flow in a dual connection system.

Another technical problem of the present invention is to provide a method and apparatus for a receiving end of a PDCP entity to transmit a PDCP SDU to an upper layer in ascending order in a multi-flow structure.

Another technical problem of the present invention is to perform PDCP rearrangement based on a timer in a multi-flow structure.

According to an aspect of the present invention, in a packet data convergence protocol (PDCP) entity of a terminal (UE) configured with dual connectivity with a macro base station (Macro eNB) and a small base station (small eNB), multi-flow Provides a method for reordering PDCP Service Data Units (SDUs) considering (multi-flow). The PDCP SDU rearrangement method includes sequentially receiving PDCP packet data units (PDUs) of PDCP sequence number (SN) n, and PDCP SN n, not PDCP SN n+1, which is expected to receive the next sequential When a +k PDCP PDU is received, it characterized in that it includes the step of driving a wait timer.

According to another aspect of the present invention, the PDCP SDU rearrangement method, when the PDCP SN n+1 PDCP PDU is received before the waiting timer expires, starting from the PDCP SN n+1 (starting from ) Transmitting all stored PDCP SDUs having a continuously associated PDCP SN value to a higher layer in ascending order.

According to another aspect of the present invention, the PDCP SDU rearrangement method is, when the PDCP SN n+1 PDCP PDU is not received until the rearrangement timer expires, a PDCP smaller than the PDCP SN n+k It can be determined that PDCP SDUs that have not yet been received associated with the SN value have been removed.

According to another aspect of the present invention, the PDCP SDU rearrangement method removes the PDCP SN n+1 when the PDCP PDU of the PDCP SN n+1 is not received until the rearrangement timer expires. It can be judged that it has become.

According to another aspect of the present invention, the PDCP SDU rearrangement method, when the PDCP SN n+1 PDCP PDU is not received until the rearrangement timer expires, starting from the PDCP SN n+k It characterized in that it includes the step of transferring all stored PDCP SDUs of continuously associated PDCP SN values to an upper layer in ascending order.

According to the present invention, when a UE is configured with dual connectivity with a macro base station and a small base station, in performing multi-flow downlink reception, PDCP PDUs are non-sequentially transmitted to the PDCP entity of the UE due to a transmission path delay. Even if received, PDCP SDUs may be rearranged based on a standby timer, PDCP SDUs may be delivered to a higher layer in ascending order, and transmission efficiency may be improved.

In addition, since the standby timer is driven based on the PDCP PDUs received out of sequence, even if a time delay occurs while packets are being received to the macro base station due to service disconnection, PDCP SDU rearrangement can be smoothly performed.

1 shows a wireless communication system to which the present invention is applied.
2 is a block diagram showing a radio protocol architecture for a user plane.
3 is a block diagram showing a radio protocol structure for a control plane.
4 is a diagram showing an outline of an example of an RLC sub-layer model to which the present invention is applied.
5 is a diagram showing an overview of an example of a PDCP sub-layer model to which the present invention is applied.
6 shows an example of a dual connection situation of a terminal applied to the present invention.
7 shows an example of an EPS bearer structure when a single flow is configured.
8 shows an example of a network structure of a macro base station and a small base station in a single flow in a dual connectivity situation.
9 shows an example of an EPS bearer structure when multi-flow is configured in a dual connection situation.
10 shows an example of a network structure of a macro base station and a small base station in multi-flow.
11 shows a packet delivery process in the case of a single flow when considering dual connectivity.
12 shows a packet delivery process in case of multi-flow when considering dual connectivity.
13 shows an example of a PDCP PDU reception timing in a PDCP entity of a terminal.
14 is a diagram illustrating a case in which packets are intermittently received due to service disconnection.
15 illustrates a method of rearranging PDCP SDUs in consideration of multi-flow according to an example of the present invention.
16 is a flowchart of a method for rearranging PDCP SDUs using a standby timer according to an example of the present invention.
17 is a block diagram of a macro base station, a small base station, and a terminal according to the present invention.

Hereinafter, in the present specification, contents related to the present invention will be described in detail through exemplary drawings and embodiments along with the contents of the present invention. In adding reference numerals to constituent elements in each drawing, it should be noted that the same constituent elements are given the same reference numerals as much as possible even though they are indicated on different drawings. In addition, in describing an embodiment of the present specification, if it is determined that a detailed description of a related known configuration or function may obscure the subject matter of the present specification, a detailed description thereof will be omitted.

In addition, this specification describes a wireless communication network, and the work performed in the wireless communication network is performed in the process of controlling the network and transmitting data in a system (for example, a base station) that governs the wireless communication network, or The work can be done at a terminal coupled to the network.

1 shows a wireless communication system to which the present invention is applied. This may be a network structure of an E-UMTS system (Evolved-Universal Mobile Telecommunications System). The E-UMTS system may be an LTE (Long Term Evolution) or LTE-A (advanced) system. Wireless communication systems include Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single Carrier-FDMA (SC-FDMA), and OFDM-FDMA. , OFDM-TDMA, OFDM-CDMA, such as various multiple access techniques can be used.

Referring to FIG. 1, the E-UTRAN includes a base station (20; evolved NodeB:eNB) providing a control plane (CP) and a user plane (UP) to a user equipment (UE). Include.

The terminal 10 may be fixed or mobile, and may be referred to as other terms such as a mobile station (MS), an advanced MS (AMS), a user terminal (UT), a subscriber station (SS), and a wireless device. .

The base station 20 generally refers to a station that communicates with the terminal 10, and includes a base station (BS), a base transceiver system (BTS), an access point, and a femto-eNB. , Pico base station (pico-eNB), home base station (Home eNB), relay (relay) may be referred to as other terms. The base stations 20 may be connected to each other through an X2 interface. The base station 20 is connected to an Evolved Packet Core (EPC) 30 through an S1 interface, more specifically, a Mobility Management Entity (MME) through an S1-MME and a Serving Gateway (S-GW) through an S1-U. The S1 interface exchanges signals with the MME to exchange OAM (Operation and Management) information for supporting the movement of the terminal 10.

The EPC 30 includes MME, S-GW, and P-GW (Packet Data Network-Gateway). The MME has access information of the terminal 10 or information on the capabilities of the terminal 10, and this information is mainly used for mobility management of the terminal 10. S-GW is a gateway having an E-UTRAN as an endpoint, and P-GW is a gateway having a PDN (Packet Data Network) as an endpoint.

E-UTRAN and EPC 30 may be integrated to be called an EPS (Evoled Packet System), and all traffic flows from the radio link to which the UE 10 accesses the base station 20 to the PDN that connects to the service entity is IP Operates based on (Internet Protocol).

The air interface between the terminal and the base station is called a Uu interface. The layers of the Radio Interface Protocol between the terminal and the network are the first layer (L1), based on the lower three layers of the Open System Interconnection (OSI) standard model widely known in communication systems. It may be divided into a second layer (L2) and a third layer (L3). Among them, the physical layer belonging to the first layer provides an information transfer service using a physical channel, and the radio resource control (RRC) layer located in the third layer exchanges RRC messages It controls radio resources between the network and the network.

FIG. 2 is a block diagram showing a radio protocol architecture for a user plane, and FIG. 3 is a block diagram showing a radio protocol architecture for a control plane. The user plane is a protocol stack for transmitting user data, and the control plane is a protocol stack for transmitting control signals.

2 and 3, a physical layer (PHY) provides an information transfer service to an upper layer using a physical channel. The physical layer is connected to a medium access control (MAC) layer, which is an upper layer, through a transport channel. Data is transferred between the MAC layer and the physical layer through a transport channel. Transmission channels are classified according to how data is transmitted through the air interface.

In addition, data is transmitted between different physical layers (ie, between the physical layers of the transmitter and the receiver) through a physical channel. The physical channel may be modulated in an OFDM (Orthogonal Frequency Division Multiplexing) method, and time and frequency are used as radio resources.

As an example, the physical downlink control channel (PDCCH) of the physical channels informs the UE of resource allocation of a paging channel (PCH) and a downlink shared channel (DL-SCH) and hybrid automatic repeat request (HARQ) information related to the DL-SCH, The PDCCH may carry an uplink scheduling grant that informs the UE of resource allocation for uplink transmission. In addition, the PCFICH (physical control format indicator channel) informs the UE of the number of OFDM symbols used for PDCCHs and is transmitted every subframe. In addition, PHICH (physical Hybrid ARQ Indicator Channel) carries a HARQ ACK/NAK signal in response to uplink transmission. In addition, PUCCH (Physical uplink control channel) carries uplink control information such as HARQ ACK/NAK for downlink transmission, scheduling request, and CQI. In addition, a physical uplink shared channel (PUSCH) carries an uplink shared channel (UL-SCH).

The MAC layer may perform mapping between a logical channel and a transport channel and multiplexing or demultiplexing into a transport block provided as a physical channel on a transport channel of a MAC service data unit (SDU) belonging to the logical channel. The MAC layer provides a service to the Radio Link Control (RLC) layer through a logical channel. The logical channel can be divided into a control channel for transmitting control area information and a traffic channel for transmitting user area information.

The functions of the RLC layer include concatenation, segmentation, and reassembly of RLC SDUs. In order to ensure various QoS (Quality of Service) required by Radio Bearer (RB), the RLC layer has a Transparent Mode (TM), Unacknowledged Mode (UM), and Acknowledged Mode. , AM).

RLC SDUs are supported in various sizes, and for example, may be supported in units of bytes. RLC PDUs (protocol data units) are specified only when a transmission opportunity is notified from a lower layer (eg, MAC layer), and RLC PDUs are delivered to the lower layer when the transmission opportunity is notified. The transmission opportunity may be reported along with the size of the total RLC PDUs to be transmitted. Hereinafter, the RLC layer will be described in detail in FIG. 4.

Functions of the Packet Data Convergence Protocol (PDCP) layer in the user plane include transmission of user data, header compression, and ciphering. The functions of the PDCP layer in the user plane include transmission of control plane data and encryption/integrity protection.

The RRC layer is in charge of controlling logical channels, transport channels and physical channels in relation to configuration, re-configuration and release of RBs. RB refers to a logical path provided by the first layer (PHY layer) and the second layer (MAC layer, RLC layer, PDCP layer) for data transmission between the terminal and the network. Configuring the RB means a process of defining characteristics of a radio protocol layer and channel to provide a specific service, and setting specific parameters and operation methods for each. The RB may be further divided into SRB (Signaling RB) and DRB (Data RB). SRB is used as a path for transmitting RRC messages and non-access stratum (NAS) messages in the control plane, and DRB is used as a path for transmitting user data in the user plane.

The NAS layer is located above the RRC layer and performs functions such as session management and mobility management.

If there is an RRC connection between the RRC layer of the UE and the RRC layer of the E-UTRAN, the UE is in an RRC connected state, otherwise it is in an RRC idle state.

As a downlink transport channel for transmitting data from a network to a terminal, there are a broadcast channel (BCH) for transmitting system information, and a downlink shared channel (SCH) for transmitting user traffic or control messages. In the case of downlink multicast or broadcast service traffic or control messages, they may be transmitted through a downlink SCH or a separate downlink multicast channel (MCH). Meanwhile, as an uplink transport channel for transmitting data from a terminal to a network, there are a random access channel (RACH) for transmitting an initial control message, and an uplink shared channel (SCH) for transmitting user traffic or control messages.

It is located above the transport channel, and the logical channels mapped to the transport channel include BCCH (Broadcast Control Channel), PCCH (Paging Control Channel), CCCH (Common Control Channel), MCCH (Multicast Control Channel), MTCH (Multicast Traffic). Channel).

The physical channel is composed of several symbols in the time domain and several sub-carriers in the frequency domain. One sub-frame is composed of a plurality of OFDM symbols in the time domain. One subframe is composed of a plurality of resource blocks (Resource Block), one resource block is composed of a plurality of symbols and a plurality of sub-carriers (sub-carriers). In addition, each subframe may use specific subcarriers of specific symbols (eg, the first symbol) of the corresponding subframe for the PDCCH (Physical Downlink Control Channel). The transmission time interval (TTI), which is a unit time at which data is transmitted, is 1 ms corresponding to one subframe.

4 is a diagram showing an outline of an example of an RLC sub-layer model to which an embodiment of the present invention is applied.

Referring to FIG. 4, arbitrary RLC entities are classified into different RLC entities according to data transmission schemes. As an example, there are TM RLC entity 400, UM RLC entity 420, and AM RLC entity 440.

The UM RLC entity 400 may be configured to receive or deliver RLC PDUs through logical channels (eg, DL/UL DTCH, MCCH or MTCH). In addition, the UM RLC entity may transmit or receive an Unacknowledged Mode Data PDU (UMD PDU).

The UM RLC entity consists of a transmitting UM RLC entity or a receiving UM RLC entity.

The transmitting UM RLC entity receives RLC SDUs from the upper layer and transmits the RLC PDUs to the peer receiving UM RLC entity through the lower layer. When a transmitting UM RLC entity constructs UMD PDUs from RLC SDUs, when a specific transmission opportunity is notified by a lower layer, the RLC SDUs are segmented or concatenated to the total size of RLC PDUs indicated by the lower layer UMD PDUs are configured to be within the range, and related RLC headers are included in the UMD PDU.

The receiving UM RLC entity delivers RLC SDUs to an upper layer and receives RLC PDUs from a peer receiving UM RLC entity through a lower layer. When the receiving UM RLC entity receives the UMD PDUs, the receiving UM RLC entity detects whether the UMD PDUs have been repeatedly received, removes the duplicated UMD PDUs, and the UMD PDUs are received out of sequence. Reorder the order of UMD PDUs, avoid excessive rearrangement delays by detecting the loss of UMD PDUs at the lower layer, and reassemble RLC SDUs from rearranged UMD PDUs , The reassembled RLC SDUs are delivered to an upper layer in ascending order of RLC SN (sequence number), and a UMD PDU that cannot be reassembled into an RLC SDU due to the loss of UMD PDUs belonging to a specific RLC SDU in the lower layer. Can be removed. Upon RLC re-establishment, the receiving UM RLC entity reassembles RLC SDUs from the received UMD PDUs out of sequence if possible and delivers them to the upper layer, and the remaining UMD PDUs that could not be reassembled into RLC SDUs Remove all of them, initialize the associated state variables and stop the associated timers.

Meanwhile, the AM RLC entity 440 may be configured to receive or transmit RLC PDUs through logical channels (eg, DL/UL DCCH or DL/UL DTCH). The AM RLC entity transmits or receives an AMD PDU or ADM PDU segment, and transmits or receives an RLC control PDU (eg, STATUS PDU).

The AM RLC entity 440 delivers STATUS PDUs to the peer AM RLC entity in order to provide positive and/or negative ACK (akcnowledgement) of RLC PDUs (or a portion thereof). This can be called STATUS reporting. A polling procedure from the peer AM RLC entity may be followed to trigger STATUS reporting. That is, the AM RLC entity may poll the peer AM RLC entity to trigger STATUS reporting from its peer AM RLC entity.

If the STATUS report is triggered and the t-StatusProhibit is not running or has expired, the STATUS PDU is transmitted at the next transmission opportunity. Accordingly, the UE predicts the size of the STATUS PDU and considers the STATUS PDU as data available for transmission in the RLC layer.

The AM RLC entity consists of a transmitting side and a receiving side.

The transmitter of the AM RLC entity receives RLC SDUs from the upper layer and transmits the RLC PDUs to the peer AM RLC entity through the lower layer. When the transmitter of the AM RLC entity constructs AMD PDUs from RLC SDUs, when a specific transmission opportunity is notified by the lower layer, the RLC SDUs are divided to fit within the total size of the RLC PDU(s) indicated by the lower layer ( Segment) or concatenate to compose AMD PDUs. The transmitter of the AM RLC entity supports retransmission (ARQ) of RLC data PDUs. If the RLC data PDU to be retransmitted does not fit within the total size of the RLC PDU(s) indicated by the lower layer when a specific transmission opportunity is notified by the lower layer, the AM RLC entity subdivides the RLC data PDU into AMD PDU segments. (re-segment).

At this time, the number of re-segmentation is not limited. When the transmitter of the AM RLC entity makes AMD PDUs from RLC SDUs received from a higher layer or makes AMD PDU segments from RLC data PDUs to be retransmitted, the relevant RLC headers are included in the RLC data PDU.

The receiving unit of the AM RLC entity delivers RLC SDUs to the upper layer and receives RLC PDUs from the peer AM RLC entity through the lower layer.

When the reception unit of the AM RLC entity receives RLC data PDUs, it detects whether or not the RLC data PDUs have been redundantly received, removes the redundant RLC data PDUs, and the RLC data PDUs are received out of sequence. In this case, the order of the RLC data PDUs is rearranged, the loss of the RLC data PDUs generated in the lower layer is detected, retransmission is requested to the peer AM RLC entity, and RLC SDUs are reassembled from the rearranged RLC data PDUs ( reassemble), and transfer the reassembled RLC SDUs to the upper layer in the reassembled sequence.

Upon RLC reconfiguration, the receiver of the AM RLC entity reassembles RLC SDUs from the RLC data PDUs received out of sequence if possible, transfers them to the upper layer, and all the remaining RLC data PDUs that cannot be reassembled into RLC SDUs. Remove, initialize the associated state variables and stop the associated timers.

5 is a diagram showing an overview of an example of a PDCP sub-layer model to which the present invention is applied.

The PDCP sublayer includes at least one PDCP entity 500. Each RB (eg, DRB and SRB, except for SRB0) is associated with one PDCP entity 500. Each PDCP entity may be associated with one or two RLC entity(s) according to a characteristic of the RB and an RLC mode.

The PDCP entity 500 receives user data from an upper layer (eg, an application layer) or transmits user data to an upper layer. Here, the user data is an IP packet. User data can be delivered through a PDCP-SAP (Service Access Point). The PDCP layer receives a PDCP configuration request (PDCP_CONFIG_REQ) message, which is signaling data, from the RRC layer. The PDCP configuration request message may be delivered through a Control-Service Access Point (C-SAP). The PDCP configuration request message is a message requesting to configure the PDCP according to the PDCP configuration parameters.

The transmitting side of the PDCP entity 500 starts a discard timer upon reception of user data from an upper layer. User data (i.e., PDCP SDU) is subjected to header compression, flawless protection (in the control plane), and encryption (cipering) to add a PDCP header to become a PDCP PDU (i.e., RLC SDU). The transmitting end PDCP delivers the PDCP PDU to a lower layer (eg, RLC layer). The PDCP PDU may include a PDCP Data PDU and a PDCP Control PDU. The PDCP Data PDU carries user plane data, control plane data, etc., and carries the PDCP SDU SN (Sequence Number). The PDCP SDU SN may be referred to as a PDCP SN. The PDCP Control PDU carries a PDCP status report and header compression control information.

The RLC SDU may be delivered to the RLC layer through RLC-SAP. If user data is not transmitted until the removal timer expires, the transmitting end PDCP removes the user data (PDCP SDU including user data).

The receiving side of the PDCP entity 500 receives an RLC SDU (ie, PDCP PDU) from a lower layer. The PDCP PDU becomes a PDCP SDU through PDCP header decompression, deciphering, and integrity verification (in the control domain). The receiving end of the PDCP entity 500 delivers the PDCP SDU to an upper layer (eg, an application layer).

The receiving end of the PDCP entity 500 generally expects to receive RLC SDUs (ie, PDCP PDUs) sequentially, except for re-establishment of the lower layer. Accordingly, the receiving end of the PDCP entity 500 was able to deliver the PDCP SDU corresponding to the PDCP SDU to the upper layer in ascending order except for the case of receiving the RLC SDU through reconfiguration of the lower layer. If there is a stored PDCP SDU, it is delivered to the upper layer in ascending order. For example, when the PDCP entity 500 receives a PDCP PDU for reasons other than resetting of a lower layer, all stored PDCP SDUs with a count value lower than the count value of the received PDCP SDU are delivered to the upper layer in ascending order. And, starting from the count value of the received PDCP SDU, all stored PDCP SDU(s) of the count value continuously associated with each other are transferred to the upper layer in ascending order.

6 shows an example in which a dual connection is configured in a terminal to which an embodiment of the present invention is applied.

6, a terminal 650 located in a service area of a macro cell in a macro base station (or a master base station or an anchor base station, 600) is a small base station (or a secondary base station or an assisting base station or a slave base station, 610). ) In the case of entering an area that is over-laid with the service area of the small cell.

In order to support additional data service through the small cell in the small base station while maintaining the existing wireless connection and data service connection through the macro cell in the macro base station, the network configures a dual connection to the terminal.

Accordingly, the user data arriving at the macro cell may be transmitted to the terminal through the small cell in the small base station. Specifically, the F2 frequency band is assigned to the macro base station, and the F1 frequency band is assigned to the small base station. In this situation, the UE can receive a service through the F2 frequency band from the macro base station and at the same time receive the service through the F1 frequency band from the small base station.

7 shows an example of an EPS bearer structure when a single flow is configured.

Referring to FIG. 7, an RB is a bearer provided in a Uu interface to support a service of a user. In the wireless communication system, each bearer is defined for each interface, thereby ensuring independence between interfaces.

The bearer provided by the wireless communication system is collectively referred to as an EPS (Evolved Packet System) bearer. The EPS bearer is a transmission path created between the UE and the P-GW. The P-GW may receive an IP flow from the Internet or transmit an IP flow to the Internet. One or more EPS bearers may be configured per terminal, and each EPS bearer may be divided into E-RAB (E-UTRAN Radio Access Bearer) and S5/S8 bearers, and the E-RAB is RB (Radio Bearer), S1 Can be divided into bearers. That is, one EPS bearer corresponds to one RB, S1 bearer, and S5/S8 bearer, respectively. Depending on which service (or application) is used, IP flows may have different QoS (Quality of Service) characteristics, and IP flows having different QoS characteristics for each EPS bearer may be mapped and transmitted. EPS bearers may be identified based on EPS bearer identity. The EPS bearer identifier is assigned by the UE or MME.

A packet gateway (P-GW) is a network node that connects a wireless communication network (for example, an LTE network) according to the present invention and another network. EPS bearer is defined between the terminal and the P-GW. EPS bearers are further subdivided between nodes, and defined as RB between UE and base station, S1 bearer between base station and S-GW, and S5/S8 bearer between S-GW and P-GW inside EPC. do. Each bearer is defined through QoS. QoS is defined through data rate, error rate, delay, etc.

Therefore, after the QoS to be provided by the wireless communication system as a whole is defined as an EPS bearer, each QoS is determined for each interface. Each interface is to set up a bearer according to the QoS it must provide. Since the bearer of each interface divides the QoS of the entire EPS bearer for each interface, the EPS bearer, RB, S1 bearer, etc. are all basically in a one-to-one relationship.

That is, the LTE wireless communication system is basically a single flow structure, and one RB is configured for one EPS bearer. In other words, one EPS bearer is mapped to the S1 bearer through one RB. In the case of single flow, one EPS bearer is serviced through one RB. In this case, one RB (eg, PDCP entity, RLC entity, MAC entity, PHY layer) is configured in the base station for the corresponding EPS bearer, and one RB is also configured in the terminal.

8 shows an example of a network structure of a macro base station and a small base station in a single flow in a dual connectivity situation. 8 shows a case in which a service is provided to a terminal through two EPS bearers.

Referring to FIG. 8, a macro base station includes two PDCP entities, an RLC entity, a MAC entity, and a PHY layer, while a small base station includes an RLC entity, a MAC entity, and a PHY layer. EPS bearer #1 (800) provides a service to the terminal through the RB (PDCP/RLC/MAC/PHY) configured in the macro base station. On the other hand, EPS bearer #2 850 provides a service to a terminal through a PDCP entity configured in a macro base station and an RB (RLC/MAC/PHY) configured in a small base station. Therefore, a service is provided through one RB for each EPS bearer in a single flow.

9 shows an example of an EPS bearer structure when multi-flow is configured in a dual connection situation.

Referring to FIG. 9, when multi-flow is configured, a service is provided through two RBs each configured in a macro base station and a small base station instead of one RB for one EPS bearer. For one EPS bearer, the UE can receive a service simultaneously through the RB configured in the macro base station and the RB configured in the small base station. This is a form in which one EPS bearer provides services through two RBs. As described above, when one EPS bearer provides a service to a terminal through two or more RBs, it can be considered that a multi-flow is configured in the terminal. Alternatively, a case in which a service is provided to a terminal through a macro base station and a small base station by dividing one RB may be said to be configured with a multi-flow in the terminal. Alternatively, it can be considered that a multi-flow is configured when an RB providing service only through a macro base station and another RB provided by dividing one RB into a macro base station and a small base station are simultaneously provided to the terminal. In the above, a case in which a service is provided to a terminal through a macro base station and a small base station by dividing one RB may be referred to as a bearer split.

10 shows an example of a network structure of a macro base station and a small base station in multi-flow.

Referring to FIG. 10, a macro base station includes a PDCP entity, an RLC entity, a MAC entity, and a PHY layer, but a small base station includes an RLC entity, a MAC entity, and a PHY layer. In FIG. 10, unlike FIG. 8, RBs are respectively configured in a macro base station and a small base station for one EPS bearer 1000 to provide a service to a terminal. That is, for one EPS bearer, the macro base station and the small base station provide a service to the terminal through multiflow.

On the other hand, when considering dual connectivity, in the case of single flow and multi-flow, the packet delivery process may be expressed as follows.

11 shows a packet delivery process in the case of a single flow when considering dual connectivity.

Referring to FIG. 11, the macro base station 1130 receives packets for each of two EPS bearers through a P-GW and an S-GW. Here, the flow through which packets are transmitted is mapped to each EPS bearer. Packets transmitted through EPS bearer #1 are assumed to be packet 1, and packets transmitted through EPS bearer #2 are assumed to be packet 2.

The PDCP 1135-1 of the macro base station 1130 receives packet 1 from the S-GW, and the PDCP 1135-2 receives packet 2 from the S-GW. The PDCP 1135-1 generates PDCP PDU1 based on packet 1, and the PDCP PDU1 is transmitted to the RLC 1140 of the macro base station 1130, and each entity and each entity through the MAC 1145 and PHY 1150 It is transformed into a format suitable for the layer and transmitted to the terminal 1100.

PDCP 1135-2 of macro base station 1130 generates PDCP PDU2 based on packet 2, and the PDCP PDU2 is transferred to RLC 1170 of small base station 1160, and MAC 1175, PHY 1180 ) Is transformed into a format suitable for each entity and layer and transmitted to the terminal 1100.

In the terminal 1100, a radio protocol entity exists for each of EPS bearer #1 and EPS bearer #2. In other words, the UE 1100 has a PDCP/RLC/MAC/PHY entity (or layer) for EPS bearer #1, and a PDCP/RLC/MAC/PHY entity (or layer) for EPS bearer #2. . Specifically, for EPS bearer #1, PHY (1105-1), MAC (1110-1), RLC (1115-1), and PDCP (1120-1) exist, so that service data and packets for EPS bearer #1 are present. Process. For EPS bearer #2, there are PHY (1105-2), MAC (1110-2), RLC (1115-2), and PDCP (1120-2), and service data and packets for EPS bearer #2 Process.

Meanwhile, the macro base station 1130 and the small base station 1160 may be connected through an X2 interface. That is, the macro base station 1130 transmits the PDCP PDU2 of the PDCP 1135-2 to the RLC 1140 of the small base station 1160 through the X2 interface. Here, the X2 interface may be an X3 interface or other expression indicating an interface between a macro base station and a small base station. In this case, when the X2 interface between the macro base station 1130 and the small base station 1160 is configured with a non-ideal backhaul, a transmission delay of about 20 to 60 ms may occur. The size of the transmission delay may be changed according to a transmission line or method as an example.

However, even in this case, the RLC (1115-1), PDCP (1120-1) for EPS bearer #1, and RLC (1115-2), PDCP (1120-2) for EPS bearer #2 in this case. Since it is configured separately, no problem occurs even when the RLC entity of the AM performs the sequential delivery of the RLC SDU to the PDCP entity. In other words, each PDCP entity corresponding to the PDCP (1120-1) and the PDCP (1120-2) is sequentially transmitted from each RLC entity corresponding to the RLC (1115-1) and RLC (1115-2). There is no problem of changing or changing.

12 shows a packet delivery process in case of multi-flow when considering dual connectivity.

Referring to FIG. 12, a macro base station 1230 receives packets for one EPS bearer through P-GW and S-GW. For the single EPS bearer, the macro base station 1230 and the small base station 1260 each configure an RB. Specifically, the macro base station 1230 configures the PDCP 1235, RLC 1240, MAC 1245, and PHY 1250, and the small base station 1240 comprises RLC 1270, MAC 1275, and PHY 1280. ). The RB configured by the small base station 1240 shares the PDCP 1235 configured by the macro base station 1230. Accordingly, one RB is divided into a macro base station 1230 and a small base station 1260 and configured.

The PDCP 1235 of the macro base station 1230 receives a packet from the S-GW. The PDCP 1235 generates PDCP PDUs based on a packet, and converts the PDCP PDUs according to a predefined rule or an arbitrary method. The RLC 1240 of the macro base station 1230 and the RLC 1270 of the small base station 1260 Allocate and deliver appropriately. For example, among the PDCP PDUs, a PDCP PDU having an odd number of SNs is transmitted to the RLC 1240 of the macro base station 1230, and a PDCP PDU having an even number of SNs is transmitted to the RLC 1270 of the small base station 1260. I can.

The RLC 1240 generates RLC PDU1(s), and the RLC PDU1(s) is transformed into a format suitable for each entity and layer through the MAC 1245 and PHY 1250 and transmitted to the terminal 1200. In addition, the RLC 1270 generates RLC PDU2(s), and the RLC PDU2(s) is transformed into a format suitable for each entity and layer through the MAC 1275 and PHY 1280 and transmitted to the terminal 1200. do.

In the terminal 1200, there are two radio protocol entities for the EPS bearer. In other words, the terminal 1200 has a PDCP/RLC/MAC/PHY entity (or layer) as an RB corresponding to the macro base station 1230, and an RLC/MAC/PHY entity as an RB corresponding to the small base station 1260 ( Or layer) exists. Specifically, PHY (1205-1), MAC (1210-1), RLC (1215-1), and PDCP (1220) corresponding to the macro base station 1230 for the EPS bearer exist, and the small base station 1260 Corresponding PHY 1205-2, MAC 1210-2, and RLC 1215-2 exist. The PDCP 1220 is a PDCP entity corresponding to the macro base station 1230 and the small base station 1260 at the same time. That is, in this case, two RLC entities 1215-1 and 1215-2 exist at the terminal 1200 end, but the two RLC entities 1215-1 and 1215-2 are one PDCP entity 1220 Corresponds to

As described above, the macro base station 1230 and the small base station 1260 may be connected through an X2 (or Xn) interface. That is, the macro base station 1230 transfers some of the PDCP PDUs of the PDCP 1235-2 to the RLC 1240 of the small base station 1260 through the X2 interface. Here, the X2 interface may be an Xn interface or another expression indicating an interface between a macro base station and a small base station. In this case, when the X2 interface between the macro base station 1230 and the small base station 1260 is configured as a non-ideal backhaul, a transmission delay of about 20 to 60 ms may occur. The PDCP entity 1220 of the terminal 1200 must receive RLC SDUs (i.e., PDCP PDUs) from two RLC entities 1215-1 and 1215-2, respectively, and generate a PDCP SDU and deliver it to the upper layer, Due to the transmission delay, the RLC SDUs (i.e., PDCP PDUs) received by the PDCP entity 1220 have a time difference between those received from the RLC entity 1215-1 and the RLC entity 1215-2, The PDCP entity 1220 may have a problem in performing ascending transmission of the PDCP SDU to an upper layer.

As can be seen in FIG. 12, one PDCP 1235 exists in the macro base station 1230 and one PDCP entity 1220 exists in the terminal 1200 for multi-flow in a dual connectivity environment. In addition, RLC entities 1240 and 1270 exist in the macro base station 1230 and the small base station 1230, respectively, and the terminal 1200 also has two RLC entities 1215-1 and 1215-2 corresponding thereto. . That is, at the end of the RLC entities 1215-1 and 1215-2 of the terminal 1210, in-sequence delivery to an upper layer may be guaranteed. However, at the end of the PDCP entity 1220 of the terminal 1210, RLC SDUs (that is, PDCP PDUs) are transmitted from two RLC entities 1215-1 and 1215-2 instead of one RLC entity. Accordingly, sequential delivery at the end of the RLC entity 1215-1 and 1215-2 does not guarantee sequential reception of the PDCP PDU at the end of the PDCP entity. In addition, transmission of the PDCP PDU(s) from the PDCP entity 1235 of the macro base station 1230 to the RLC entity 1270 of the small base station 1260 may involve a transmission delay of 20 to 60 ms, and the macro base station 1230 A time delay may occur between the transmission of the PDCP PDU(s) to the RLC entity 1240 of) and the transmission of the PDCP PDU(s) to the RLC entity 1270 of the small base station 1230. In the end, even when the PDCP PDU(s) transmitted by the PDCP entity 1235 of the macro base station 1230 is received by the PDCP entity 1220 at the terminal 1200, transmission and small transmission through the RLC or lower end of the macro base station 1230 In transmission through the RLC or lower end of the base station 1260, a difference in reception time occurs, and the PDCP entity 1220 of the terminal 1200 is difficult to expect sequential reception of the PDCP PDU(s).

13 shows an example of timing of receiving PDCP PDUs in a PDCP entity of a terminal. 13 exemplarily shows a time when a PDCP PDU transmitted through a macro base station and a PDCP PDU transmitted through a small base station arrive at a PDCP entity of a terminal. The macro base station may determine a PDCP PDU to be transmitted through the macro base station and a PDCP PDU to be transmitted through the small base station for a service for one EPS bearer. In FIG. 13, PDCP PDUs associated with an odd number of PDCP Sequence Numbers (SNs) are transmitted to a terminal through a macro base station, and PDCP PDUs associated with an even number are transmitted to a terminal through a small base station.

Referring to FIG. 13, there is a difference in time delay between a reception time of a PDCP PDU transmitted through a macro base station at a terminal and a reception time of a PDCP PDU transmitted through a small base station at the terminal. A transmission delay of about 20 to 60 ms may occur in the PDCP PDU transmitted through the small base station. This is mainly caused by a transmission delay occurring in the X2 (or Xn) interface when the macro base station transmits the PDCP PDU to the small base station. In this case, the PDCP entity of the UE receives the PDCP PDU out of sequence due to the time difference between the RLC (AMD) SDUs transmitted from the two RLC entities, and the PDCP entity processes it to a higher layer (e.g., application layer). In the case of transmission, it is difficult to ensure transmission of PDCP SDUs in ascending order. That is, in the multi-flow structure, since PDCP PDUs transmitted from one PDCP entity of the macro base station are transmitted through the RLC entity of the macro base station and the RLC entity of the small base station, there is a time delay in receiving the PDCP PDU from the PDCP entity of the terminal. Therefore, a problem arises in performing ascending transmission of PDCP SDUs from the PDCP entity to the upper layer.

The PDCP entity of the terminal performs deciphering and header decompression of the received PDCP PDU, and transmits the PDCP SDU to a higher layer. At this time, if a PDCP SDU of an SN smaller than the sequence number (SN) of the current PDCP SDU is stored, the PDCP SDUs are transmitted to the upper layer in order from the smallest SN to the largest SN.

On the other hand, the transmission side of the PDCP entity may operate a discard timer. The duration of the removal timer may be configured from an upper layer, and a timer is started when a PDCP SDU is received from an upper layer. When the removal timer expires, the PDCP entity removes the corresponding PDCP SDU. Accordingly, due to the expiration of the removal timer, the PDCP SDU of a specific SN may be removed, and the receiving end of the PDCP entity may transmit all PDCP SDUs in ascending order without having to sequentially transmit all PDCP SDUs to the upper layer.

However, in the case of supporting multi-flow in the above-described dual connectivity situation, the PDCP entity may receive RLC SDUs (PDCP PDUs) from two associated RLC entities. In this case, PDCP PDUs (especially, RLC AMD SDUs) are not sequentially received by the PDCP entity, and a PDCP PDU having a larger PDCP SN may be received first due to a problem of transmission path reception delay, and the PDCP entity is transferred to a higher layer. PDCP SDUs may not be guaranteed to be transmitted in ascending order.

Meanwhile, various communication services provided to users may have a service gap due to various reasons. Service interruption refers to a phenomenon in which the flow of packet transmission between the user and the server does not continue to occur continuously in Internet service provision, and packet transmission is interrupted or discontinuously transmitted for a specific time. Services provided on the Internet may be transmitted from an application server to a user via an Internet network. In this case, the packet transmitted to the user may be discontinuously transmitted according to a network load condition. Incoming packets may be discontinuously introduced from the side of multiple nodes in the network. In addition, depending on the application, packets may be generated discontinuously rather than continuously.

Even in the network (for example, LTE network) of the wireless communication system according to the present invention, packets for a service flowing to a base station (for example, a macro base station) are not continuously received, and the packet flow is interrupted due to service disconnection. I can.

14 is a diagram illustrating a case in which packets are intermittently received due to service disconnection.

Referring to FIG. 14, a packet is transmitted from a gateway 1460 (GW) to a base station 1430 for a service provided to the terminal 1400. Here, the base station 1430 may be, for example, a macro base station. The application service may be provided through a server or the like, which is transmitted to the wireless network and transmitted to the base station 1430 in the form of a packet through the GW 1460. At this time, packet traffic is generally continuously transmitted from the GW 1460 to the base station 1430. However, a service interruption in which packets are not transmitted at regular or random intervals may occur depending on network conditions or applications.

Although FIG. 14 shows that there are some intervals from Packet 1 to Packet 4, this is for explicitly diagramming the flow of packets, and is actually transmitted with little or no interval between packets. However, between packet 4 and packet 5, packets are not transmitted during a relatively long time interval corresponding to Gap 1. Also, even between packet 8 and packet 9, packets are not transmitted during a relatively long time interval corresponding to Gap 2. Here, Gap1 and Gap2 may be treated as service disconnection. The base station 1430 receives the transmitted packets in order from packet #1. The received packets correspond to the PDCP SDU. That is, the received packet is stored in the PDCP buffer in the form of a PDCP SDU by the PDCP entity of the PDCP layer. At this time, the PDCP SDUs stored in the buffer are sequentially processed as PDCP PDUs and delivered to the lower layer. As described above, if a difference in time intervals at which packets are received by the base station 1430 due to service disconnection occurs, a time interval may also occur during a time interval at which PDCP PDUs are transmitted from the PDCP entity of the base station 1430 to a lower layer. That is, the PDCP PDUs processed by the PDCP SDUs corresponding to packets 4 and 5 and packets 8 and 9 received at the base station 1430 at the time intervals of Gap1 and Gap2 will also be delivered to the lower layer with a time difference. I can.

When dual connectivity is configured in the terminal, in the case of a single flow situation, generally one PDCP entity is associated with one RLC entity. Therefore, even if a service interruption occurs, there may be a time interval between PDCP PDUs received by the PDCP entity of the terminal, but this situation does not cause a problem. This is because the PDCP PDU generated by the PDCP entity is delivered through the RLC entity of the lower layer, and in particular, the RLC AM supports sequential delivery. Therefore, PDCP PDUs that are introduced at time intervals need only be processed after a certain period of time, and service interruption does not cause any special problems in system performance.

However, when a dual connection is configured in the terminal, in the case of a multi-flow situation, the terminal can receive the packets through two different transmission paths. In other words, for one EPS bearer configured in the terminal and the network, the terminal receives a service through the macro base station and the small base station. For example, when one PDCP entity exists in a macro base station for one EPS bearer, and an RLC entity exists in each of the macro base station and the small base station, the PDCP PDUs transmitted by the PDCP entity of the macro base station are RLC of the macro base station. It is transmitted to the PDCP entity of the terminal through the RLC entity of the entity or the small base station. Accordingly, the PDCP entity of the UE receives PDCP PDUs that have undergone different path delays. In this case, a problem such as a PDCP PDU having a large PDCP SN arriving earlier than a PDCP PDU having a small PDCP SN may occur, and thus, a method of rearranging PDCP SDUs in a PDCP entity should be considered. In order to rearrange the PDCP SDUs in the PDCP entity, a timer considering the path delay time may be used. However, PDCP SDUs can be rearranged only when additional PDCP PDU transmission delay time due to service disconnection is also considered.

In this case, when a time interval occurs due to service disconnection, the service disconnection time may be much larger than the path delay time. Accordingly, there is a need for a PDCP SDU rearrangement method capable of delivering PDCP SDUs from a PDCP entity to a higher layer in ascending order in consideration of not only path delay but also service disconnection.

Hereinafter, in the present invention, a wait timer that is driven when a PDCP entity receives a PDCP PDU out-of-sequence in consideration of situations such as a reception delay and a service disconnection through a multipath is used. As a basis, we propose a method of performing PDCP SDU rearrangement. Here, the standby timer may be referred to as an out-of-sequence timer. The present invention can be applied to both a downlink data transfer procedure and an uplink data transfer procedure, and the downlink data transfer procedure will be mainly described below.

15 illustrates a method of rearranging PDCP SDUs in consideration of multi-flow according to an example of the present invention. Figure 15 shows that PDCP PDUs of PDCP SNs 1 to 8 are delivered through the RLC entity of the macro base station, and PDCP PDUs of PDCP SNs 9, 10, 14, 15, 16, 23, and 24 are delivered through the RLC entity of the small base station. Is assumed. In addition, FIG. 15 assumes a case in which other PDCP PDUs are not delivered for a certain time due to service disconnection after the PDCP PDU of PDCP SN 7 is delivered.

15A is a case in which a PDCP PDU of PDCP SN 9 is received without receiving PDCP PDUs of PDCP SN 8 after receiving PDCP PDUs of PDCP SN 1,2,3,4,5,6,7 by a PDCP entity of a UE.

Referring to FIG. 15A, a PDCP entity of a UE sequentially receives PDCP PDUs from PDCP SNs 1 to 7. The PDCP entity of the terminal processes the received PDCP PDUs and delivers the corresponding PDCP SDUs to the upper layer. When the PDCP PDUs are sequentially received, the PDCP entity of the terminal does not drive the standby timer. The PDCP entity of the UE does not receive other PDCP PDUs for a certain period of time thereafter, and then receives the PDCP PDU of PDCP SN No. 9. Since the PDCP entity of the UE has not yet received the PDCP PDU of PDCP SN 8, it is determined that the PDCP PDU reception of PDCP SN9 is out of sequence, and drives the standby timer.

Here, the sequential reception may be determined according to, for example, the following criteria. If the PDCP SN of the PDCP SDU last delivered to the upper layer is defined as Last_Submitted_PDCP_RX_SN, and the PDCP SN of the PDCP SDUs expected to be sequentially received next is defined as Next_PDCP_RX_SN, Next_PDCP_RX_SN is one of the following Equations 1 and 2 You can follow one.

In Equation 2, Maximum_PDCP_SN represents the maximum value of the allowed PDCP SN. That is, Equation 2 indicates that the number starts again from 0 after the maximum value of PDCP SN.

In addition, the waiting timer may mean a timer waiting for a PDCP entity receiving a non-sequential PDCP PDU to receive a sequential PDCP PDU. The standby timer may be set to a value of, for example, 20 to 60 ms in consideration of the path delay time of the X2 (or Xn) interface between the macro base station and the small base station.

When the standby timer is driven based on the out of sequence reception of the PDCP PDU as described above, it is possible to prevent the timer from expiring due to a time delay due to service disconnection. For example, a timer for rearrangement of PDCP SDUs may be operated based on the sequential reception of PDCP PDUs, but in this case, the timer may undesirably expire in the case of service disconnection as described above, and may be received by waiting longer. PDCP PDUs can be mistreated as removed. However, by operating the standby timer based on the non-sequential reception of the PDCP PDU as described above, it is possible to solve a problem caused by a time delay due to service disconnection.

15B assumes a case in which a PDCP entity of a terminal receives PDCP PDUs of PDCP SN 8 after receiving PDCP PDUs of PDCP SN 10 and 14 after FIG. 15A and before expiration of the standby timer.

Referring to FIG. 15B, the PDCP entity of the UE determines that the PDCP PDUs of PDCP SNs 10 and 14 are still received as out of sequence. Subsequently, when the PDCP PDU of PDCP SN 8 is received, it is determined as sequential reception. The PDCP entity of the terminal stops the standby timer when the PDCP PDU of PDCP SN 8 is received. And, starting from PDCP SN number 8, all stored PDCP SDUs of the continuously associated PDCP SN values are delivered to the upper layer in ascending order. For example, the PDCP entity PDCP SNs 8, 9, and 10 PDCP SDUs of the UE are delivered to the upper layer in ascending order.

15C assumes a case in which the PDCP entity of the UE receives the PDCP PDU of PDCP SN 15 after FIG. 15B.

Referring to FIG. 15C, when a PDCP entity of a terminal receives a PDCP PDU of PDCP SN 15, it determines that it is non-sequential reception, and drives a standby timer again. The PDCP entity of the terminal waits for the PDCP PDU of PDCP SN 11, which is expected to be sequentially received, to be received until the waiting timer expires.

If the PDCP PDU of the PDCP SN 11 is not received until the standby timer expires, as an example, the PDCP entity of the terminal starts from the PDCP SN 11 when the standby timer expires and all of the associated PDCP SN values are not stored. PDCP SDUs that are not (e.g., PDCP SNs 11, 12, 13) are determined to be removed, and then all stored PDCP PDUs of the associated PDCP SN values (e.g. For example, PDCP SNs 14 and 15 PDCP SDUs) are transmitted to the upper layer.

As another example, when the PDCP entity of the terminal receives the PDCP PDU for the first time after expiration of the standby timer, it is determined that all unstored PDCP SDUs of the continuously associated PDCP SN value are removed, starting from the PDCP SN 11, and then the PDCP SN Starting from the first, all stored PDCP PDUs of the associated PDCP SN value are transmitted to the upper layer.

According to the present invention as described above, when the terminal is configured with dual connectivity with the macro base station and the small base station, in performing multi-flow downlink reception, due to a transmission path delay, the PDCP entity of the terminal Even if PDCP PDUs are received, PDCP SDUs may be rearranged based on a standby timer, PDCP SDUs may be delivered to a higher layer in ascending order, and transmission efficiency may be improved.

In addition, since the standby timer is driven based on the PDCP PDUs received out of sequence, even if a time delay occurs while a packet is received to the macro base station due to service disconnection, PDCP SDU rearrangement can be smoothly performed.

16 is a flowchart of a method of rearranging PDCP SDUs using a standby timer according to an example of the present invention.

Referring to FIG. 16, when a PDCP entity of a terminal sequentially receives PDCP PDUs of PDCP SN n times, it transmits a corresponding PDCP SDU to a higher layer (S1600). The PDCP entity of the terminal may check whether the corresponding PDCP PDUs are sequentially received based on the PDCP SN of the received PDCP PDU.

The PDCP entity of the UE drives a standby timer when a PDCP PDU of PDCP SN n+k (k is a natural number other than 1) other than PDCP SN n+1, which is expected to be sequentially received, is received (S1610). . The standby timer may mean a timer waiting for a PDCP entity receiving a sequential PDCP PDU to receive a sequential PDCP PDU.

For example, in FIG. 15A, after receiving the PDCP PDU corresponding to PDCP SN 7, the UE may be in a state in which it expects to receive the PDCP PDU of PDCP SN 7+1 number 8. In this case, after a certain period of time, the UE has received a PDCP PDU corresponding to SN 9 instead of PDCP SN 8 which is expected to be received. The UE runs a timer to wait for the reception of PDCP PDU 8 after PDCP PDU 9 reception. In the above, when the terminal receives the PDCP PDU corresponding to PDCP SN 9, the PDCP PDU corresponding to PDCP SN 7 is transmitted to the upper layer and is no longer stored in the terminal. However, the UE remembers the SN of the PDCP PDU transmitted to the upper layer and can know the PDCP SN that it expects to receive next.

In addition, referring to FIG. 15B, the UE transmits all PDCP PDUs corresponding to PDCP SNs 8, 9, and 10 to the upper layer when receiving PDCP PDU No. 8. Accordingly, the UE transmits PDCP PDUs corresponding to PDCP SN No. 10. It is understood that the PDU was transmitted to the upper layer. In this case, the UE is in a situation in which the PDCP SN that it expects to receive next is identified as number 11. In FIG. 15C, when the UE receives the PDCP PDU corresponding to PDCP SN 15, it is different from the PDCP SN 11 that was expected to be received next, so that the UE drives the standby timer to wait for reception of the PDCP PDU corresponding to PDCP SN 11. . In this case, the UE is storing PDCP PDUs 14 and 15. Although it expected reception of PDCP SN 11 and started the standby timer, it is reasonable to determine that PDCP PDUs 12 and 13 were transmitted through a macro or a small base station prior to transmission of PDCP PDUs 14 and 15. Therefore, when the terminal receives PDCP PDU 15, it is reasonable to receive it within the standby timer operation period. If not received, it can be considered discarded. Accordingly, it can be determined that PDCP PDUs that have not arrived at the time when the standby timer driven by the UE when receiving PDCP PDU 15 times expires have been removed.

The PDCP entity of the terminal checks whether the PDCP PDU of PDCP SN n+1 is received before the standby timer expires (S1620). The PDCP entity of the terminal does not start/restart the standby timer even if it receives other out-of-order PDCP PDUs while the standby timer is running.

If the PDCP entity of the terminal receives the PDCP PDU of PDCP SN n+1 before the standby timer expires in S1620, the PDCP entity of the terminal stops the standby timer and starts from PDCP SN n+1. All stored PDCP SDUs of the continuously associated PDCP SN values are delivered to the upper layer in ascending order (S1630).

If the standby timer expires in S1620, that is, if the PDCP entity of the terminal does not receive the PDCP SN n+1 PDCP PDUs until the standby timer expires, the PDCP entity of the terminal is smaller than PDCP SN n+k times. All PDCP SDUs associated with a PDCP SN value that has not been received are considered to have been removed (S1640). That is, the PDCP entity of the UE removes the PDCP SDUs excluding the currently stored PDCP SDUs among the associated PDCP SDUs with a PDCP SN value smaller than the PDCP SN value n+k starting from the PDCP SN value n+1 expected to be sequentially received. Can be considered. In this case, the PDCP entity of the UE transfers all stored PDCP SDUs associated with a PDCP SN value less than PDCP SN n+k to the upper layer (S1650), and starts from PDCP SN n+k. The stored PDCP SDUs are delivered to the upper layer (S1660). As an example, the PDCP entity of the terminal transmits all stored PDCP SDUs associated with a PDCP SN value smaller than PDCP SN n+k times to the upper layer in ascending order at the time of expiration of the waiting timer, and continuously associated PDCPs starting from n+k times. All stored PDCP SDUs of the SN value can be delivered to the upper layer in ascending order. As another example, the PDCP entity of the terminal transmits all stored PDCP SDUs associated with a PDCP SN value smaller than PDCP SN n+k to a higher layer in ascending order at the time of receiving a random PDCP PDU for the first time after the rearrangement timer expires, Starting from n+k, all stored PDCP SDUs of a PDCP SN value that are continuously associated can be delivered to a higher layer in ascending order.

Or, starting from PDCP SN n+1, all unsaved PDCP SDUs up to PDCP SN n+m, which is a continuously associated PDCP SN value, are considered to be removed, and the last (largest) PDCP SN value of the PDCP SDUs considered to be removed. Starting from +1 of n+m, all stored PDCP SDUs of continuously associated PDCP SN values may be delivered to the upper layer in ascending order.

17 is a block diagram of a macro base station, a small base station, and a terminal according to the present invention.

Referring to FIG. 17, a terminal 1700 according to the present invention may configure daul connectivity with a macro base station 1730 and a small base station 1760. In addition, the terminal 1700, the macro base station 1730, and the small base station 1760 according to the present invention support the above-described multi-flow.

The macro base station 1730 includes a macro transmission unit 1735, a macro reception unit 1740, and a macro processor 1750.

The macro receiving unit 1740 receives a packet for one EPS bearer from the S-GW. The macro processor 1750 controls the PDCP entity of the macro base station 1730 to process PDCP SDUs corresponding to the received packet and generate PDCP PDUs. The macro processor 1750 distributes the PDCP PDUs according to a reference, transfers (or transmits) some of the PDCP PDUs to the RLC entity of the macro base station 2140, and transmits the PDCP PDUs to the terminal through the macro transmission unit 1735. The macro processor 1750 transmits (or transmits) the remaining part to the RLC entity of the small base station 1760 through the macro transmission unit 1735. In this case, PDCP SDUs corresponding to PDCP PDUs may be classified and indicated as PDCP SNs.

In addition, the macro processor 1750 generates information on the waiting timer for the PDCP SDU and transmits the information to the terminal through the macro transmission unit 1735. The information on the standby timer may be signaled exclusively to the terminal 1700 or may be signaled in a broadcast manner. The macro transmission unit 1735 may transmit information on the standby timer to the terminal 1700 through an RRC message (eg, an RRC connection reconfiguration message).

The small base station 1760 includes a small transmission unit 1765, a small reception unit 1770, and a small processor 1780.

The small receiver 1770 receives the remaining part of the PDCP PDUs from the macro base station 1730.

The small processor 1780 processes the PDCP PDU by controlling the RLC entity, the MAC entity, and the PHY layer of the small base station 2130, and transmits the PDCP PDU to the terminal through the small transmission unit 1565.

The terminal 1700 includes a terminal receiving unit 1705, a terminal transmitting unit 1710, and a terminal processor 1720. The terminal processor 1720 performs functions and controls necessary to implement the features of the present invention as described above.

The terminal receiving unit 1705 receives information on a standby timer from the macro base station 1730. The information on the standby timer may be included in an RRC message (eg, an RRC connection reconfiguration message) and received by the terminal receiver 1705. In this case, the terminal transmission unit 1710 may transmit an RRC connection reconfiguration completion message to the macro base station 1730.

In addition, the terminal receiving unit 1705 receives data for PDCP PDUs from the macro base station 1730 and the small base station 1760, respectively.

The terminal processor 1720 interprets the data, and controls the PHY layer (s), the MAC entity (s), the RLC entity (s), and the PDCP entity of the terminal 1700 to obtain PDCP SDUs.

When the terminal processor 1720 sequentially receives PDCP PDUs of PDCP SN n times from the PDCP entity, the terminal processor 1720 delivers the corresponding PDCP SDU to the upper layer. Here, the terminal processor 1720 may check whether the corresponding PDCP PDU is sequentially received by the PDCP entity based on the PDCP SN of the received PDCP PDU. For example, a PDCP SN value of a PDCP SDU (or PDU) expected to be sequentially received may be determined based on Equation 1 or 2 described above.

The terminal processor 1720 drives a standby timer when a PDCP PDU of PDCP SN n+k (where k is a natural number other than 1) other than PDCP SN n+1, which are expected to be sequentially received, is received by the PDCP entity.

If a PDCP PDU of PDCP SN n+1 is received by the PDCP entity before the standby timer expires, the terminal processor 1720 stops the standby timer. In addition, the terminal processor 1720 transfers all stored PDCP SDUs of a PDCP SN value that are successively associated starting from PDCP SN n+1 to an upper layer of the PDCP entity in ascending order.

When the standby timer expires, that is, if the PDCP SN n+1 PDCP PDUs are not received by the PDCP entity until the standby timer expires, the terminal processor 1720 returns a PDCP SN value smaller than PDCP SN n+k times. All associated unreceived PDCP SDUs can be considered to have been removed.

In this case, the terminal processor 1720 delivers all stored PDCP SDUs associated with a PDCP SN value smaller than PDCP SN n+k times to a higher layer in ascending order. In addition, the terminal processor 1720 transfers all stored PDCP SDUs of the PDCP SN value continuously associated from the PDCP SN n+k to the upper layer in ascending order.

As an example, the terminal processor 1720 may deliver all stored PDCP SDUs associated with a PDCP SN value smaller than the PDCP SN n+k stored in the PDCP entity at the time of expiration of the rearrangement timer to the upper layer in ascending order, and PDCP SN n+k From, all PDCP SDUs of the PDCP SN value that are continuously associated can be delivered to the upper layer in ascending order. As another example, the terminal processor 1720 ranks all stored PDCP SDUs associated with a PDCP SN value less than the PDCP SN n+k stored in the PDCP entity at the time of receiving a random PDCP PDU for the first time after the reordering timer expires in ascending order. It can be delivered to a layer, and all PDCP SDUs of a PDCP SN value continuously associated from PDCP SN n+k can be delivered to a higher layer in ascending order.

The above description is merely illustrative of the technical idea of the present invention, and those of ordinary skill in the art to which the present invention pertains will be able to make various modifications and variations without departing from the essential characteristics of the present invention. Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention, but to explain the technical idea, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be interpreted as being included in the scope of the present invention.

Claims (11)

In the PDCP (Packet Data Convergence Protocol) entity of a terminal (UE) configured with dual connectivity with a macro base station (Macro eNB) and a small base station (Dual connectivity), PDCP considering multi-flow In a method of reordering Service Data Units (SDUs),
Sequentially receiving a PDCP packet data unit (PDU) of n PDCP sequence number (SN);
Driving a wait timer when a PDCP PDU of PDCP SN n+k other than PDCP SN n+1 for which next sequential reception is expected is received;
When the PDCP PDU of the PDCP SN n+1 is received before the standby timer expires, all stored PDCP SN values starting from the PDCP SN n+1 (starting from) and continuously (associated) Delivering PDCP SDUs to an upper layer in ascending order;
When the PDCP PDU of the PDCP SN n+1 is not received until the waiting timer expires, the PDCP SDUs not yet received associated with the PDCP SN value less than the PDCP SN n+k are removed. Judging; And
When the PDCP PDU of the PDCP SN n+1 is not received until the waiting timer expires, delivering all stored PDCP SDUs associated with a PDCP SN value smaller than PDCP SN n+k to a higher layer in ascending order
Characterized in that it comprises a, PDCP SDU rearrangement method. The method of claim 1,
The value of the standby timer is set to a random value within 20 ms to 60 ms corresponding to an Xn interface path delay between the macro base station and the small base station. The method of claim 1,
Even when a PDCP PDU of PDCP SN n+k, which is a non-sequential recipient, is received, the standby timer is driven only when the standby timer is not running. delete delete delete The method of claim 1,
When the standby timer expires, all stored PDCP SDUs having a PDCP SN value smaller than the PDCP SN n+k are delivered to an upper layer in ascending order. The method of claim 1,
At the time when a random PDCP PDU is received for the first time after the waiting timer expires, all stored PDCP SDUs of the PDCP SN value that are continuously related to the PDCP SN are delivered to the upper layer in ascending order starting from the PDCP SN n+k. How to rearrange the PDCP SDU. The method of claim 1,
When the PDCP PDU of the PDCP SN n+1 is not received until the waiting timer expires, all stored PDCP SDUs of the PDCP SN value that are continuously associated starting from PDCP SN n+k are delivered to the upper layer in ascending order. Characterized in that it comprises a step, PDCP SDU rearrangement method. The method of claim 9,
When the waiting timer expires, all stored PDCP SDUs associated with a PDCP SN value smaller than the PDCP SN n+k are delivered to a higher layer in ascending order. The method of claim 9,
At the time when a certain PDCP PDU is first received after the waiting timer expires, all stored PDCP SDUs of the PDCP SN value continuously associated starting from the PDCP SN n+k are delivered to the upper layer in ascending order. , PDCP SDU rearrangement method.

Images