WO2013042908A1

WO2013042908A1 - Apparatus and method for performing random access procedure in multiple component carrier system

Info

Publication number: WO2013042908A1
Application number: PCT/KR2012/007428
Authority: WO
Inventors: Ki Bum Kwon; Jae Hyun Ahn; Myung Cheul Jung; Kang Suk Huh
Original assignee: Pantech Co., Ltd.
Priority date: 2011-09-20
Filing date: 2012-09-18
Publication date: 2013-03-28

Abstract

An apparatus and method for performing a random access procedure in a multiple component carrier system are disclosed. The method includes starting a first random access procedure intended for a first serving cell, increasing a preamble transmission counter indicating a retransmission number when a random access preamble is retransmitted in the first random access procedure, receiving a random access start indicator that orders a start of a second random access procedure intended for a second serving cell different from the first serving cell from a base station, holding the start of the second random access procedure if it is checked that the first random access procedure is in progress, starting the second random access procedure when the first random access procedure is terminated.

Description

APPARATUS AND METHOD FOR PERFORMING RANDOM ACCESS PROCEDURE IN MULTIPLE COMPONENT CARRIER SYSTEM

The present invention relates to wireless communication and, more particularly, to an apparatus and method for performing a random access procedure in a multiple component carrier system.

A mobile station performs a random access procedure in order to access a network. The random access procedure may be divided into a contention-based random access procedure and a non-contention-based random access procedure. The greatest difference between the contention-based random access procedure and the non-contention-based random access procedure is whether a random access preamble is dedicated to one mobile station or not. In the non-contention-based random access procedure, contention (or collision) with other mobile station does not occur because a mobile station uses a dedicated random access preamble. Here, the term ‘contention’ means that two or more mobile stations attempt a random access procedure using the same random access preamble through the same resources. In the contention-based random access procedure, there is a possibility of contention because a mobile station uses a randomly selected random access preamble.

A random access procedure may be applied to the case where a mobile station is newly combined with a network through handover or may be performed in a variety of situations, such as the case where a synchronization state or a Radio Resource Control (RRC) state changes from an RRC_IDLE state to an RRC_CONNECTED state after a mobile station is combined with a network or the case where a mobile station requires uplink synchronization in order to transmit and receive data to and from a base station.

As multiple component carrier systems are introduced, there is a need for a clear definition of a random access procedure when the random access procedure is performed through a plurality of serving cells.

An object of the present invention is to provide an apparatus and method for performing a random access procedure in a multiple component carrier system.

Another object of the present invention is to provide an apparatus and method for controlling a random access procedure depending on the characteristics of a serving cell in a wireless communication system in which a plurality of component carriers is operated.

Yet another object of the present invention is to provide an apparatus and method for controlling a maximum number of times of retransmission of preambles in a random access procedure.

Further yet another object of the present invention is to provide an apparatus and method for transmitting an indicator that stops a random access procedure in a specific serving cell.

Still yet another object of the present invention is to provide an apparatus and method for providing a function of indicating the stop of a random access procedure to downlink control information that orders the start of the random access procedure.

Further yet another object of the present invention is to provide an apparatus and method for stopping a random access procedure in a specific serving cell based on an indicator that orders the stop of the random access procedure.

Still yet another object of the present invention is to provide an apparatus and method for transmitting an indicator that stops a random access procedure in a serving cell and orders the start of a random access procedure in another serving cell in a wireless communication system in which a plurality of component carriers is operated.

Further yet another object of the present invention is to provide an apparatus and method for successfully terminating a random access procedure in another serving cell that is now in progress based on an indicator that orders the start of a random access procedure in a serving cell.

Still yet another object of the present invention is to provide an apparatus and method for preventing a collision between random access procedures in different serving cells and midway terminating a random access procedure in a wireless communication system in which a plurality of component carriers is operated.

In accordance with an aspect of the present invention, there is provided a method of a mobile station performing a random access procedure in a multiple component carrier system. The method of performing a random access procedure includes receiving a random access stop indicator that orders the stop of the random access procedure in a secondary serving cell configured in the mobile station from a base station, determining whether a stop condition that the random access procedure is stopped is satisfied or not based on the random access stop indicator, and stopping the random access procedure in the secondary serving cell if, as a result of the determination, it is determined that the stop condition is satisfied.

In accordance with another aspect of the present invention, there is provided a mobile station configured to perform a random access procedure in a multiple component carrier system. The mobile station includes a mobile station reception unit configured to receive a random access stop indicator that orders the stop of the random access procedure in a secondary serving cell configured in the mobile station from a base station, a random access processing unit configured to determine whether a stop condition that the random access procedure is stopped is satisfied or not based on the random access stop indicator and to stop the random access procedure in the secondary serving cell if, as a result of the determination, it is determined that the stop condition is satisfied, and a mobile station transmission unit configured to transmit a preamble to the base station if, as a result of the determination, it is determined that the stop condition is not satisfied.

In accordance with yet another aspect of the present invention, there is provided a method of a base station performing a random access procedure in a multiple component carrier system. The method of performing a random access procedure includes transmitting a random access command that orders the start of a random access procedure in a secondary serving cell configured in a mobile station to the mobile station, determining whether or not to request to stop the random access procedure, and transmitting a random access stop indicator that orders the stop of the random access procedure to the mobile station if, as a result of the determination, it is determined to stop the random access procedure.

In accordance with yet another aspect of the present invention, there is provided a base station configured to perform a random access procedure in a multiple component carrier system. The base station includes a base station transmission unit configured to transmit a random access command that orders the start of a random access procedure in a secondary serving cell configured in a mobile station to the mobile station, a stop request unit configured to determine whether or not to request to stop the random access procedure, and a base station reception unit configured to receive a preamble from the mobile station.

The stop request unit transmits a random access stop indicator that orders the stop of the random access procedure to the mobile station if, as a result of the determination, it is determined to stop the random access procedure.

In accordance with still yet another aspect of the present invention, there is provided a method of a mobile station performing a random access procedure in a multiple component carrier system. The method of performing a random access procedure includes performing a first random access procedure in a first serving cell, receiving a random access start indicator that orders the start of a second random access procedure in a second serving cell different from the first serving cell from a base station, terminating the first random access procedure based on the random access start indicator, and transmitting a preamble for the second random access procedure to the base station.

In accordance with still yet another aspect of the present invention, there is provided a mobile station configured to perform a random access procedure in a multiple component carrier system. The mobile station includes a random access processing unit configured to perform a first random access procedure in a first serving cell, a mobile station reception unit configured to receive a random access start indicator that orders the start of a second random access procedure in a second serving cell different from the first serving cell from a base station, and a mobile station transmission unit configured to transmit a preamble for the second random access procedure to the base station.

The random access processing unit may terminate the first random access procedure based on the random access start indicator.

In accordance with still yet another aspect of the present invention, there is provided a method of a base station configured to perform a random access procedure in a multiple component carrier system. The method of performing a random access procedure includes performing a first random access procedure in a first serving cell, transmitting a random access start indicator that orders the start of a second random access procedure in a second serving cell different from the first serving cell to a mobile station, and receiving a preamble for the second random access procedure from the mobile station.

In accordance with still yet another aspect of the present invention, there is provided a base station configured to perform a random access procedure in a multiple component carrier system. The base station includes a message processing unit configured to generate a message related to a first random access procedure in a first serving cell and generate a random access start indicator that orders the start of a second random access procedure in a second serving cell different from the first serving cell, a base station transmission unit configured to transmit the random access start indicator to a mobile station, and a base station reception unit configured to receive a preamble for the second random access procedure from the mobile station.

In accordance with yet another aspect of the present invention, there is provided a method of a mobile station performing a random access procedure in a multiple component carrier system. The method of performing a random access procedure includes increasing a preamble transmission counter, indicating the number of times that a preamble is retransmitted in a current random access procedure intended for a first serving cell, by 1, determining whether a midway termination condition that the current random access procedure is midway terminated is satisfied or not, and terminating the current random access procedure as being unsuccessful if, as a result of the determination, the midway termination condition is satisfied.

In accordance with yet another aspect of the present invention, there is provided a mobile station configured to perform a random access procedure in a multiple component carrier system. The mobile station includes a random access processing unit configured to increase a preamble transmission counter, indicating the number of times that a preamble is retransmitted in a current random access procedure intended for a first serving cell, by 1, determine whether a midway termination condition that the current random access procedure is midway terminated is satisfied or not, and terminate the current random access procedure as being unsuccessful if, as a result of the determination, the midway termination condition is satisfied, a mobile station reception unit configured to receive a random access start indicator that orders the start of a new random access procedure intended for a second serving cell from a base station, and a mobile station transmission unit configured to transmit a preamble to the base station according to the new random access procedure.

In accordance with the present invention, a base station may forcedly stop a random access procedure that is in progress in a specific serving cell and start a random access procedure in another serving cell. Accordingly, the deterioration of system performance, such as the delay of random access that occurs in an environment in which a parallel random access procedure is not supported, can be reduced.

Furthermore, the end of a first random access procedure and the start of a second random access procedure in different serving cells are not performed by different signalings, but can be performed at once in response to a random access start indicator. Accordingly, a random access procedure can be simplified, and the deterioration of performance due to signaling delay can be reduced.

Furthermore, the stability of a random access procedure can be provided by compromising the profits of an anterior random access procedure and a posterior random access procedure in selecting any one of random access procedures regarding different serving cells in a situation in which the random access procedures occur frequently.

FIG. 1 shows a wireless communication system to which the present invention is applied;

FIG. 2 shows an example of a protocol structure for supporting multiple CCs to which the present invention is applied;

FIG. 3 shows an example of a frame structure for the operation of multiple CCs to which the present invention is applied;

FIG. 4 shows linkage between downlink component carriers and uplink component carriers in a multiple component carrier system to which the present invention is applied;

FIG. 5 is a diagram showing an example of a timing advance in a synchronization process to which the present invention is applied;

FIG. 6 is a diagram showing that an uplink timing alignment value is applied using the downlink timing alignment values of a primary serving cell and a secondary serving cell;

FIG. 7 is a flowchart illustrating a contention-based random access procedure to which the present invention is applied;

FIG. 8 is a flowchart illustrating a random access procedure according to the order of a base station to which the present invention is applied;

FIG. 9 is a flowchart illustrating a method of performing a random access procedure in accordance with an embodiment of the present invention;

FIG. 10 is a block diagram showing the sub-header of an MAC control element to which the present invention is applied;

FIG. 11 is a block diagram showing a random access stop indicator in accordance with an embodiment of the present invention;

FIG. 12 is a flowchart illustrating a method of performing a random access procedure in accordance with another embodiment of the present invention;

FIG. 13 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with an embodiment of the present invention;

FIG. 14 is a flowchart illustrating a random access procedure performed by a base station in accordance with an embodiment of the present invention;

FIG. 15 is a block diagram showing a mobile station and a base station which perform a random access procedure in accordance with an embodiment of the present invention;

FIG. 16 is a flowchart illustrating a method of performing a random access procedure in accordance with an embodiment of the present invention;

FIG. 17 is a block diagram showing an MAC Protocol Data Unit (PDU) including a random access start indicator in accordance with an embodiment of the present invention;

FIG. 18 is a block diagram showing an example of an MAC control element to which the present invention is applied;

FIG. 19 is a block diagram showing an example of an MAC sub-header to which the present invention is applied;

FIG. 20 is a block diagram showing another example of an MAC sub-header to which the present invention is applied;

FIG. 21 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with an embodiment of the present invention;

FIG. 22 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with another embodiment of the present invention;

FIG. 23 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with yet another embodiment of the present invention;

FIG. 24 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with further yet another embodiment of the present invention;

FIG. 25 is a flowchart illustrating a random access procedure performed by a base station in accordance with an embodiment of the present invention;

FIG. 26 is a block diagram showing a mobile station and a base station which perform a random access procedure in accordance with an embodiment of the present invention;

FIG. 27 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with an embodiment of the present invention;

FIG. 28 shows an example of a scenario in which a mobile station performs a random access procedure in accordance with the present invention;

FIG. 29 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with another embodiment of the present invention;

FIG. 30 is a flowchart illustrating a random access procedure performed by a mobile station in accordance with yet another embodiment of the present invention;

FIG. 31 is a flowchart illustrating a random access procedure in accordance with an embodiment of the present invention;

FIG. 32 is a flowchart illustrating a random access procedure performed by a base station in accordance with an embodiment of the present invention; and

FIG. 33 is a block diagram showing a mobile station and a base station which perform a random access procedure in accordance with an embodiment of the present invention.

Some embodiments of the present invention will be described in detail with reference to the accompanying drawings. It is to be noted that in assigning reference numerals to respective elements in each of the drawings, the same reference numerals designate the same elements although the elements are shown in different drawings. Furthermore, in describing the present invention, a detailed description of the known functions and constructions will be omitted if it is deemed to make the gist of the present invention unnecessarily vague

FIG. 1 shows a wireless communication system to which the present invention is applied.

Referring to FIG. 1, a plurality of the wireless communication systems 10 are widely deployed in order to provide a variety of communication services, such as voice and packet data. The wireless communication system 10 includes one or more Base Stations (BS) 11. The BSs 11 provide communication services to

specific cells

15a, 15b, and 15c. Each of the cells may be classified into a plurality of areas (called sectors).

User Equipment (UE) 12 may be fixed or mobile and may also be called another terminology, such as a Mobile Station (MS), a Mobile Terminal (MT), a User Terminal (UT), a Subscriber Station (SS), a wireless device, a Personal Digital Assistant (PDA), a wireless modem, or a handheld device. The BS 11 may also be called another terminology, such as an evolved-NodeB (eNB), a Base Transceiver System (BTS), an access point, a femto BS, a home NodeB, or a relay. The cell should be interpreted as a comprehensive meaning that indicates some area covered by the BS 11. The cell has a meaning that covers a variety of coverage areas, such as a mega cell, a macro cell, a micro cell, a pico cell, and a femto cell.

Hereinafter, downlink refers to communication from the BS 11 to the UE 12, and uplink refers to communication from the UE 12 to the BS 11. In downlink, a transmitter may be part of the BS 11, and a receiver may be part of the UE 12. In uplink, a transmitter may be part of the UE 12, and a receiver may be part of the BS 11. Multiple access schemes applied to the wireless communication system are not limited. A variety of multiple access schemes, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single Carrier-Frequency Division Multiple Access (SC-FDMA), OFDM-FDMA, OFDM-TDMA, and OFDM-CDMA, may be used. Uplink transmission and downlink transmission may be performed in accordance with a Time Division Duplex (TDD) scheme using different times or a Frequency Division Duplex (FDD) scheme using different frequencies.

A carrier aggregation (CA) supports a plurality of component carriers. The carrier aggregation is also called a spectrum aggregation or a bandwidth aggregation. Each of carriers aggregated by the carrier aggregation is called a Component Carrier (CC). Each of the CCs is defined by the bandwidth and the center frequency. The carrier aggregation is introduced to support an increased throughput, prevent an increase of costs due to the introduction of wideband Radio Frequency (RF) devices, and guarantee compatibility with the existing systems. For example, if 5 CCs are allocated as the granularity of a carrier unit having a 20 MHz bandwidth, a maximum of a 100 MHz bandwidth can be supported.

A carrier aggregation may be divided into a contiguous carrier aggregation performed between contiguous CCs and a discontiguous carrier aggregation performed between discontiguous CCs in the frequency domain. The number of carriers aggregated between downlink and uplink may be differently set. The case where the number of downlink CCs is equal to the number of uplink CCs is called a symmetric aggregation, and the case where the number of downlink CCs is different from the number of uplink CCs is called an asymmetric aggregation.

CCs may have different sizes (i.e., bandwidths). For example, assuming that 5 CCs are used to form a 70 MHz band, a resulting configuration may be, for example, 5 MHz CC (carrier #0) + 20 MHz CC (carrier #1) + 20 MHz CC (carrier #2) + 20 MHz CC (carrier #3) + 5 MHz CC (carrier #4).

Hereinafter, a multiple component carrier system refers to a system that supports a carrier aggregation. In a multiple component carrier system, a contiguous carrier aggregation and/or a discontiguous carrier aggregation may be used and a symmetric aggregation or an asymmetric aggregation may be used.

A UE that is in Radio Resource Control (RRC) idle mode cannot aggregate CCs, and only a UE that is in RRC connected mode can aggregate CCs. Accordingly, a UE has to select a cell for RRC connection prior to a carrier aggregation and perform an RRC connection establishment procedure for a BS through the selected cell. The RRC connection establishment procedure is performed in such a manner that the UE sends an RRC connection request message to the BS, the BS sends an RRC connection setup message to the UE, and the UE sends an RRC connection establishment complete message to the BS. The RRC connection establishment procedure includes the configuration of an SRB1.

FIG. 2 shows an example of a protocol structure for supporting multiple CCs to which the present invention is applied.

Referring to FIG. 2, the common Medium Access Control (MAC) entity 210 manages a physical layer 220 using a plurality of carriers. An MAC management message transmitted on a specific carrier can be applied to other carriers. That is, the MAC management message can control other carriers including the specific carrier. The physical layer 220 may operate in accordance with a Time Division Duplex (TDD) scheme and/or a Frequency Division Duplex (FDD) scheme.

There are several physical control channels used in the physical layer 220. A physical downlink control channel (PDCCH) informs a UE of the resource assignment 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 grant hat informs a UE of the assignment of resources for uplink transmission. A physical control format indicator channel (PCFICH) is used to inform a UE of the number of OFDM symbols used in PDCCHs and is transmitted for each subframe. A physical hybrid ARQ indicator channel (PHICH) carries an HARQ ACK/NAK signal in response to uplink transmission. A physical uplink control channel (PUCCH) carries HARQ ACK/NAK for downlink transmission, a scheduling request, and uplink control information, such as a Channel Quality Indicator (CQI). A physical uplink shared channel (PUSCH) carries an Uplink Shared channel (UL-SCH). A physical random access channel (PRACH) carries a Random Access Preamble (RAP).

FIG. 3 shows an example of a frame structure for the operation of multiple carriers to which the present invention is applied.

Referring to FIG. 3, a frame includes 10 subframes. Each of the subframes includes a plurality of OFDM symbols. Each carrier can carry its own control channel (e.g., a PDCCH). The multiple carriers may be contiguous to each other or may not be contiguous to each other. A UE may support one or more carriers depending on its capabilities.

A CC may be divided into a Primary CC (PCC) and a Secondary CC (SCC) depending on whether it has been activated or not. The PCC is always activated, and the SCC is activated or deactivated depending on a specific condition. The term ‘activation’ means that the transmission or reception of traffic data is in progress or is in a ready state. The term ‘deactivation’ means that the transmission or reception of traffic data is impossible, but measurement or the transmission or reception of minimum information is possible. A UE may use only one PCC or may use one or more SCCs along with a PCC. A BS may allocate a PCC and/or an SCC a UE.

FIG. 4 shows linkage between downlink CCs and uplink CCs in a multiple component carrier system to which the present invention is applied.

Referring to FIG. 4, in downlink, for example, downlink CCs (DL CCs) D1, D2, and D3 are aggregated. In uplink, uplink CCs (UL CCs) U1, U2, and U3 are aggregated. Here, Di is the index of a DL CC, and Ui is the index of a UL CC (where i=1, 2, 3). The index is not identical with the order of CCs or the position of the frequency band of the corresponding CC.

Meanwhile, at least one DL CC may be configured as a PCC, and the remaining CCs may be configured as SCCs. Likewise, at least one UL CC may be configured as a PCC, and the remaining CCs may be configured as SCCs. For example, D1 and U1 may be PCCs, and D2, U2, D3, and U3 may be SCCs.

Here, the index of the PCC may be set to 0, and one of other natural numbers may be the index of the SCC. For example, the index of a downlink or uplink CC may be set identically with the index of a CC (or a serving cell) including the downlink or uplink CC. For another example, only the index of a CC or the index of an SCC may be set, and the index of an uplink or uplink CC included in the CC may not exist.

In an FDD system, a DL CC and an UL CC may be linked in a one-to-one manner. For example, each of D1 and U1, D2 and U2, and D3 and U3 may be linked in a one-to-one manner. A UE performs linkage between DL CCs and UL CCs based on system information transmitted by a logical channel BCCH or a UE-dedicated RRC message transmitted by a DCCH. This connection is called System Information Block1 (SIB1) connection or SIB2 connection. Each linkage may be configured in a cell-specific way or in a UE-specific way. For example, a PCC may be configured in a cell-specific way, and an SCC may be configured in a US-specific way.

Here, the DL CC and the UL CC may have not only 1:1 linkage, but also 1:n or n:1 linkage.

A primary serving cell means one serving cell that provides security input and NAS mobility information in an RRC establishment or re-establishment state. One or more cells may be configured to form a set of serving cells along with a primary serving cell depending on the capabilities of a UE. The one or more cells are called secondary serving cells.

Accordingly, a set of serving cells configured for one UE may include only one primary serving cell or may include one primary serving cell and one or more secondary serving cells.

A DL CC corresponding to a primary serving cell is called a downlink PCC (DL PCC), and a UL CC corresponding to a primary serving cell is called an uplink PC (UL PCC). Furthermore, in downlink, a CC corresponding to a secondary serving cell is called a downlink SCC (DL SCC). In uplink, a CC corresponding to a secondary serving cell is called an uplink SCC (UL SCC). Only a DL CC or both a DL CC and a UL CC may correspond to one serving cell.

Accordingly, in a multiple component carrier system, a concept that communication between a UE and a BS is performed through a DL CC or a UL CC is the same as the concept that communication between a UE and a BS is performed through a serving cell. For example, in a method of performing a random access procedure in accordance with the present invention, a concept that a UE sends a preamble using a UL CC may be considered as the same concept that the UE sends the preamble using a primary serving cell or a secondary serving cell. Furthermore, a concept that a UE receives downlink information using a DL CC may be considered as the same concept that the UE receives the downlink information using a primary serving cell or a secondary serving cell.

The technical spirit of the present invention regarding the characteristics of a primary serving cell and a secondary serving cell is not limited to the above description, but is only illustrative. The present invention may include more examples.

In a wireless communication environment, while electric waves from a transmitter are propagated and they reach a receiver, the electric waves experience propagation delay. Accordingly, although both the transmitter and the receiver precisely know the time when the electric waves are propagated by the transmitter, the time when the signal reaches the receiver is influenced by the distance between the transmitter and the receiver and a surrounding electromagnetic environment. If the receiver moves, the time when the signal reaches the receiver is changed over time. If the receiver cannot know the time when the signal of the transmitter is reached precisely, the receiver fails in receiving the signal or receives a distorted signal, thereby making it difficult communication.

Accordingly, in a wireless communication system, synchronization between a BS and a UE must be performed in advance in order to receive an information signal both in downlink and uplink. The type of synchronization may be various, such as frame synchronization, information symbol synchronization, and sampling period synchronization. The sampling period synchronization is synchronization that must be obtained most basically in order to distinguish physical signals from each other.

Downlink synchronization is obtained by a UE in response to the signal of a BS. The BS sends an agreed and specific signal so that the UE can easily obtain downlink synchronization. The UE has to be able to precisely know the time when the BS sent a specific signal. In the case of downlink, a plurality of MSs can obtain synchronization independently because one BS sends a synchronization signal to the plurality of MSs at the same time.

In uplink, BSs receive signals transmitted by a plurality of MSs. If the distance between each UE and each BS is different, signals received by the BSs have different transmission delay times. If uplink information is transmitted based on obtained downlink synchronization, pieces of information of MSs are received by BSs at different times. In this case, the BS cannot obtain synchronization on the basis of any one UE. Accordingly, in obtaining uplink synchronization, a procedure different from that in downlink is necessary.

Meanwhile, the necessity for the acquisition of uplink synchronization may be different depending on each multiple access method. For example, in the case of a CDMA system, although a BS receives the uplink signals of different MSs at different times, the BS can separate the uplink signals from each other. In a wireless communication system based on an OFDMA or FDMA scheme, however, a BS receives the uplink signals of all MSs at the same time and demodulates the uplink signals at once. Accordingly, reception performance is increased as the uplink signals of a plurality of MSs are received on correct times, and reception performance is suddenly deteriorated as a difference between the reception times of UE signals increases. Accordingly, it is essentially necessary to obtain uplink synchronization.

A random access procedure may be performed in order to obtain uplink synchronization. For example, during a random access procedure, a UE obtains uplink synchronization based on a timing alignment value transmitted by a BS. The timing alignment value may also be called a timing advance value. When uplink synchronization is obtained, the UE starts a time alignment timer. While the time alignment timer operates, the UE and the BS are in an uplink synchronization state. If the time alignment timer expires or does not operate, the UE and the BS consider that they have not been synchronized, and thus the UE does not perform uplink transmission other than the transmission of a random access preamble.

FIG. 5 is a diagram showing an example of a timing advance in a synchronization process to which the present invention is applied.

Referring to FIG. 5, for the purpose of communication between a BS and a UE, an uplink radio frame 520 has to be transmitted at a point of time at which a downlink radio frame 510 is transmitted. The UE may send the uplink radio frame 520 earlier than the time when the downlink radio frame 510 is transmitted by taking a time lag, occurring due to propagation delay between the UE and the BS, into consideration by using a timing advance so that the BS and the UE are synchronized with each other.

An uplink timing TA adjusted by the UE can be calculated by Math Figure 1 below.

MathFigure 1

Here, N_TA is a timing alignment value and is variably controlled in response to the timing advance command of a BS. N_{TA offset} is a value fixed depending on a frame structure. T_s is a sampling period. Here, when the timing alignment value N_TA is positive (+), it means that uplink timing is adjusted so that it is advanced. When the timing alignment value N_TA is negative (-), it means that uplink timing is adjusted so that it is delayed.

For the purpose of uplink synchronization, a UE may receive a timing alignment value provided by a BS and use a timing advance based on the received timing alignment value. The UE can obtain synchronization for wireless communication with the BS.

The application of multiple timing advances is described below.

In a multiple component carrier system, one UE performs communication with a BS through a plurality of CCs or a plurality of serving cells. If the signals of the plurality of serving cells configured for the UE have different time delays, it is necessary for the UE to apply different uplink timings to the respective serving cells.

FIG. 6 is a diagram showing that an uplink timing alignment value is applied using the downlink timing alignment values of a primary serving cell and a secondary serving cell. A DL CC1 and a UL CC1 are primary serving cells, and a DL CC2 and a UL CC2 are secondary serving cells.

Referring to FIG. 6, when a BS sends frames through the DL CC1 and the DL CC2 at a point of time T_Send (610), a UE receives the frames through the DL CC1 and the DL CC2 (620). The UE receives the frames late by a propagation delay corresponding to the time after the point of time T_Send at which the BS sent the frames. Since a propagation delay T1 occurs in the DL CC1, the UE receives the frames late by the propagation time T1. Since a propagation delay T2 occurs in the DL CC2, the UE receives the frames late by the propagation delay T2.

Assuming that the propagation delay time of downlink transmission is equal to the propagation delay time of uplink transmission, the UE may adjust uplink timing so that the frames are transmitted in the UL CC1 and the UL CC2 by the respective propagation delays T1 and T2 and send the frames to the BS (630). As a result, the BS can receive the frames of the UE through the UL CC1 and the UL CC2 at a point of time T_Receive that has been set for uplink synchronization (640).

If multiple timing advances are applied in the state in which a plurality of serving cells has been configured between a UE and a BS, UL CCs may be configured for all or some of secondary serving cells, and the uplink timing of a primary serving cell and the uplink timings of some other secondary serving cells may be set differently. Uplink timings for the UL CCs of some secondary serving cells may be identical with the uplink timing of the primary serving cell. Furthermore, some other secondary serving cells may have different uplink timings from the primary serving cell and need the same uplink timing as other secondary serving cells.

Accordingly, serving cells having the same uplink synchronization may be configured as one group without adjusting uplink timings individually for serving cells in which UL CCs have been configured. This group is defined as a Timing Advanced Group (TAG). Accordingly, serving cells having different uplink synchronizations or uplink timings belong to different TAGs. The TAG may have an index value, and the index value of a TAG to which a primary serving cell belongs may be fixed to 0. Furthermore, a TAG may be changed in the case of a secondary serving cell, but a TAG cannot be changed in the case of a primary serving cell. Furthermore, a Timing Advanced Timer (TAT) may be configured for each group. Furthermore, the TAT of each TAG may have a different expiration value, and an operation may be different when each TAT expires. For example, when the TAT of a TAG to which a primary serving cell belongs expires, a UE flushes data within uplink HARQ buffers for all the serving cells of a TAG to which a primary serving cell belongs, informs an RRC layer L3 of the release of a PUCCH/Sounding Reference Signal (SRS) for all the serving cells, releases a type 0 SRS (non-triggering base), does not release a type 1 SRS (triggering base), and clears downlink and uplink resource assignments configured for all the serving cells.

Furthermore, if the TAT of a TAG including only secondary serving cells expires, a UE flushes data within uplink HARQ buffers for the secondary serving cells within the TAG, does not release a PUCCH/SRS for the secondary serving cells within the TAG, stops the transmission of an SRS through the uplink of the secondary serving cells within the TAG, and clears uplink resource assignments configured for the secondary serving cells within the TAG.

A BS may order a UE to perform a random access procedure in order to secure or update a timing alignment value for a TAG. For example, information about whether a UE has to perform a random access procedure in what serving cell using what time/frequency resources may be obtained in response to a random access command. A UE cannot start a random access procedure in a secondary serving cell spontaneously, but can start the random access procedure in response to signaling including an indicator (e.g., a Cell Indicator Field (CIF), a SCell index, or a secondary cell index) for the secondary serving cell transmitted by a BS.

A UE does not perform two or more random access procedures (i.e., a parallel random access procedure) at the same time. That is, two or more random access procedures are not synchronized and performed at the same time, and they are not performed at the same time in some of time when a random access procedure is performed. Furthermore, a UE does not perform random access procedures in two or more serving cells at the same time. In this case, there may be a problem in that a random access procedure is not started through a secondary serving cell while a UE waits for an RAR in a random access procedure through a primary serving cell. In contrast, if a random access procedure through a secondary serving cell is performed, although a BS sends signaling, requesting that a random access procedure be started, to a UE through a primary serving cell, there may be a problem in that the UE may not start a random access procedure through the primary serving cell.

Another problem is that the transmission of uplink data may be delayed. In a random access procedure started by a UE spontaneously, if there is data to be transmitted through uplink by the UE, the UE sends Buffer State Report (BSR) information to a BS in order to receive uplink resource assignment. Here, a random access procedure is used. If the BS is performing a first random access procedure in order to update a timing alignment value for a TAG, the UE cannot start a second random access procedure of transmitting the BSR information. Moreover, if 'random access progress failure and retransmission' is delayed by a maximum number of times of retransmission when performing the first random access procedure, the second random access procedure is delayed and the transmission of uplink data is also delayed. It leads to deteriorating uplink Quality of Service (QoS). The minimum number of times of retransmission that may be set in a random access procedure is 3 times, and the time necessary to perform a random access procedure in each number of times is about 10 ~ 64 ms. Accordingly, although the minimum number of times of retransmission is set, a maximum of a delay time of 192 ms may occur.

Accordingly, there is a need for a method of reducing the complexity of a random access procedure through a plurality of serving cells in which UL CCs are configured and of solving the deterioration of system performance, such as the delay of random access occurring in an environment in which a parallel random access procedure is not supported. To this end, a BS may forcedly stop a random access procedure that is in progress in a specific serving cell and start a random access procedure in another serving cell. A UE may determine whether a random access procedure has been stopped and take subsequent measures against a stopped random access procedure based on a result of the determination.

(1) The stop of a random access procedure

A random access procedure stopped by a BS may include both a contention-based random access procedure of FIG. 7 and a non-contention-based random access procedure of FIG. 8. A random access procedure may include both a random access procedure in response to the order of a BS and a spontaneous random access procedure of a UE.

FIG. 7 is a flowchart illustrating a contention-based random access procedure to which the present invention is applied.

Referring to FIG. 7, a UE randomly selects a preamble signature in order to generate a Random Access Preamble (RAP). Furthermore, the UE sends the selected RAP to a BS at step S700. When selecting the preamble signature, the UE may perform a random access procedure in a contention-based manner.

The UE may cognize a Random Access-Radio Network Temporary Identifier (RA-RNTI) by taking temporarily selected frequency resources and a point of time at which the frequency resources are transmitted into consideration in order to select the RAO or send a random access channel (RACH).

The BS performs a Random Access Response (RAR) in response to the reception of the RAP. Here, the BS sends a Random Access Response (RAR) message to the UE through a physical downlink shared channel (PDSCH) at step S705. Information transmitted through the RAR message may include, for example, information on the ID of the RAP, the ID of the BS, a Cell Radio Network Temporary Identifier (C-RNTI), information on a time slot in which the BS has received the RAP, and a timing advance command. Since timing information for uplink synchronization is received through the RAR message, the UE can perform uplink synchronization with the BS.

The UE performs scheduled transmission in response to a timing advance command at step S710. The UE may send synchronized data through a physical uplink shared channel (PUSCH) and perform a Hybrid Automatic Repeat reQuest (HARQ).

The message transmitted at the step S710 may include, for example, an RRC connection request, tracking area update, and a scheduling request. Furthermore, one of the messages may include a temporary C-RNTI, a C-RNTI (already included in the UE), or information on the ID of the UE.

Meanwhile, since a collision may occur in the process of the steps S700 to S710, the BS sends a Contention Resolution (CR) message to the UE at step S715. The UE checks that a received messages is for its own message and sends an ACK message or ii) checks that a received message is for another UE and does not send response data. If downlink assignment is missed or the message is not decoded, the UE does not send response data. Furthermore, the CR message may include a C-RNTI or information on the ID of a UE.

FIG. 8 is a flowchart illustrating a random access procedure according to the order of a BS to which the present invention is applied.

Referring to FIG. 8, the BS selects one of RAPs previously scheduled for a non-contention-based procedure, from among all available RAPs and informs a UE of available time/frequency resource information at step S800. This is called Random Access (RA) preamble assignment.

For example, the RA preamble assignment may be transferred through a higher layer message. For example, the RA preamble assignment may be transmitted to the UE through Mobility Control Information (MCI) within a handover command. In some embodiments, the RA preamble assignment may be transmitted to the UE through an RRC reconfiguration procedure of configuring secondary serving cells. In some embodiments, the RA preamble assignment may be transmitted to the UE through an RRC reconfiguration procedure of transmitting information on a TAG. In this case, preamble configuration information on all the serving cells within the TAG must be the same.

For another example, the RA preamble assignment may be mapped to a PDCCH as physical layer signaling (e.g., Format 1A Downlink Control Information (DCI)) and transferred to the UE. Format 1A DCI may be defined as in Table 1 below.

Table 1

- Carrier Indicator Field (CIF) - 0 or 3 bits.

- Flag - 1 bit for identifying Format 0/1A (0 indicates

Format

0 and 1 indicates Format 1A)

If Format 1A CRC is scrambled by a C-RNTI and the remaining fields are set as follows, Format 1A is used for a random access procedure initiated in response to a PDCCH order.

- Below-

- Localized/distributed VRB assignment flag - 1 bit. Set to 0.

- Resource block assignment -

bits. All bits are set to 1

- Preamble index - 6 bits

- PRACH mask index - 4 bits

- All the remaining bits of Format 1A for the temporary scheduling assignment of one PDSCH codeword is set to 0.

Referring to Table 1, a random access procedure according to the order of the BS may be based on contention or non-contention depending on the value of a preamble index. For example, when all 6 bits of the preamble index are set to '0', a contention-based random access procedure is performed. For example, when the preamble index='000000', the UE selects a specific preamble, sets the PRACH mask index value to '0', and then performs a contention-based random access procedure. When the preamble index all the 6 bits of which are set to '0' is received, the UE selects a preamble group based on the amount of uplink data to be transmitted and transmission power necessary to offset path attenuation. Furthermore, the UE randomly selects one of preambles within a selected preamble group. The UE sends the selected preamble to a BS using one of resources that is the closest in terms of the time in RACH time/frequency resources that have been allocated for an RACH in a corresponding serving cell at step S805.

For another example, if all the 6 bits of the preamble index are not ‘0', the UE sends a preamble, selected based on the received preamble index, to the BS at step S805. The BS can check that what UE has sent the preamble based on time/frequency resources according to the preamble and the PRACH mask index. Accordingly, since a UE having the same RA-RNTI is one UE only, a CR procedure is not necessary. The BS sends an RAR message to the UE at step S810, so that the random access procedure is completed.

While a contention-based random access procedure or a non-contention-based random access procedure through a primary serving cell is performed, the UE may report that there is a problem in the random access procedure to a higher layer (RRC layer). For example, the UE accumulatively increases a PREAMBLE_TRANSMISSION_COUNTER value for a primary serving cell by 1 whenever the transmission of an RAP fails in the primary serving cell. If a preamble is transmitted, but is unsuccessful after the number of times that an RAP was transmitted reaches a maximum retransmission number (preambleTransMax), the UE determines that there is a problem in a random access procedure and reports it to a higher layer (RRC layer). That is, when the PREAMBLE_TRANSMISSION_COUNTER=preambleTransMax+1, the UE reports that there is a problem in the random access procedure in the corresponding primary serving cell. Next, the UE starts a procedure of selecting RACH resources in order to restart a random access procedure.

While a contention-based random access procedure or a non-contention-based random access procedure through a secondary serving cell is performed, the UE may declare the failure of the random access procedure. For example, the UE accumulatively increases the PREAMBLE_TRANSMISSION_COUNTER value for a secondary serving cell by 1 whenever the transmission of a preamble fails in the secondary serving cell. If a preamble is transmitted, but is unsuccessful after the number of times that a preamble was transmitted reaches a maximum retransmission number (preambleTransMax), the UE declares the final failure of the random access procedure. That is, when the PREAMBLE_TRANSMISSION_COUNTER=preambleTransMax+1, the UE declares the final failure of the random access procedure in the corresponding secondary serving cell. Here, the UE does not start a random access procedure in the corresponding serving cell and does not perform all types of transmission, including the transmission of a Sounding Reference Signal (SRS) and data, through uplink.

FIG. 9 is a flowchart illustrating a method of performing a random access procedure in accordance with an embodiment of the present invention.

Referring to FIG. 9, a UE performs a random access procedure in a secondary serving cell at step S900. The random access procedure in the secondary serving cell may include the above-described contention-based random access procedure, the above-described non-contention-based random access procedure, and a random access procedure according to the order of a BS.

The BS determines whether or not to request the UE to stop the random access procedure that is in progress in the secondary serving cell at step S905.

For example, the BS may order the UE to perform the random access procedure in the secondary serving cell and then request the UE to stop the random access procedure after a lapse of some time elapses without receiving an RAP. The term ‘lapse of some time’ may mean that, for example, a counter becomes equal to a specific amount or a timer has expired after there is an order of a random access procedure. The counter or the timer may be, for example, 100 ms, 200 ms, or 500 ms.

For another example, the BS may request the UE to stop the random access procedure if a new random access procedure is started or a new random access procedure is scheduled to be started in a serving cell having higher priority than a secondary serving cell that is in progress in a random access procedure. A TAG including the largest number of secondary serving cells, from among a number of TAGs, may have the highest priority. For example, it is assumed that a random access procedure is in progress in a first secondary serving cell that belongs to a first TAG and a random access procedure is scheduled to be started in a second secondary serving cell that belongs to a second TAG. If the second TAG includes more secondary serving cells than the first TAG, the second TAG has higher priority. Accordingly, the BS may request the UE to stop the random access procedure in the first secondary serving cell. Here, a primary serving cell may be considered as always having higher priority than a secondary serving cell.

If, as a result of the determination at step S905, it is determined that the random access procedure does not need to be stopped, the BS continues to perform the random access procedure.

If, as a result of the determination at step S905, it is determined that the random access procedure needs to be stopped, the BS may request the UE to stop the random access procedure that is in progress in the secondary serving cell. In order to request to stop the random access procedure, the BS sends a Random Access (RA) stop indicator, requesting that the random access procedure in the secondary serving cell be stopped, to the UE at step S910.

For example, the random access stop indicator may be DCI of Format 1A. The DCI is mapped to a PDCCH, that is, a physical channel, and it may include fields listed in Table 2 below.

Table 2

- Carrier Indicator Field (CIF) - 0 or 3 bits.

- Flag - 1 bit for identifying Format 0/1A (0 indicates

Format

0 and 1 indicates Format 1A)

- Below-

- Localized/distributed VRB assignment flag - 1 bit. Set to 0.

- Resource block assignment -

bits. All bits are set to 1

- Preamble index - 6 bits

- PRACH mask index - 4 bits

Referring to Table 2, in order to indicate a random access stop indicator, specific fields are set to a specific value. For example, if cross-carrier scheduling is applied, the carrier indicator field indicates a specific carrier in order for DCI is DCI for the specific carrier. Both the localized/distributed VRB assignment flags are set to 0, and all the bits of the resource block assignment field are set to 1.

The PRACH mask index is information on available time/frequency resources. The information on the available time/frequency resources includes different resources depending on a Frequency Division Duplex (FDD) system and a Time Division Duplex (TDD) system as in Table 3.

Table 3

PRACH mask index	Permitted PRACH(FDD)	Permitted PRACH(TDD)
0	All	All
1	PRACH resource index 0	PRACH resource index 0
2	PRACH resource index 1	PRACH resource index 1
3	PRACH resource index 2	PRACH resource index 2
4	PRACH resource index 3	PRACH resource index 3
5	PRACH resource index 4	PRACH resource index 4
6	PRACH resource index 5	PRACH resource index 5
7	PRACH resource index 6	Reserved
8	PRACH resource index 7	Reserved
9	PRACH resource index 8	Reserved
10	PRACH resource index 9	Reserved
11	All even-numbered PRACH opportunities within the time domain, The first PRACH resource index within a subframe	All even-numbered PRACH opportunities within the time domain,The first PRACH resource index within a subframe
12	All odd-numbered PRACH opportunities within the time domain, The first PRACH resource index within a subframe	All even-numbered PRACH opportunities within the time domain,The first PRACH resource index within a subframe
13	Reserved	The first PRACH resource index within a subframe
14	Reserved	The second PRACH resource index within a subframe
15	Reserved	The third PRACH resource index within a subframe

Referring to Table 3, if the preamble index indicates a dedicated RAP that has been previously scheduled in order to start a random access procedure, the PRACH mask index indicates information on available time/frequency resources.

Meanwhile, the preamble index can provide two functions. The first function of the preamble index is to indicate a dedicated RAP scheduled to start a random access procedure in the state in which the random access procedure has not been performed in a specific serving cell. The second function of the preamble index is to indicate the stop of a random access procedure in the state in which the random access procedure is in progress in a specific serving cell. Here, 6 bits corresponding to the preamble index may be set to a specific value, for example, '000000'. If the preamble index indicates the stop of a random access procedure, the PRACH mask index is fixed to a specific value and it may become a meaningless field.

The DCI has the same form as a form that orders the start of a random access procedure in a secondary serving cell. A UE recognizes the DCI as a request to stop a random access procedure and stops a random access procedure in a secondary serving cell.

In some embodiments, the DCI may have the same form as a form that orders the start of a contention-based random access procedure in a secondary serving cell. Here, a UE recognizes the DCI as a request to stop a random access procedure and stops a random access procedure in a secondary serving cell.

If only a non-contention-based random access procedure is permitted, a message that requests a contention-based random access procedure may have the function of a random access stop indicator. In this case, a random access procedure is stopped irrespective of whether a random access procedure is performed in a secondary serving cell or not.

As described above, the meaning of the preamble index may be different depending on whether a UE performs a random access procedure or not. The UE has to determine a stop condition as in step S915 in order to determine whether a preamble index included in a random access stop indicator orders the start of a random access procedure or the stop of a random access procedure.

The UE determines whether a stop condition that the random access procedure in the secondary serving cell is stopped is satisfied or not at step S915. The stop condition includes i) that a random access stop indicator must be received and ii) that a random access procedure must be in progress in a corresponding serving cell. The condition i) is satisfied at step S910, and thus whether the condition ii) is satisfied or not has to be determined.

The meaning that the random access procedure is in progress is that the random access procedure has not been completed. The condition ii) may be used to determine whether a preamble index included in the random access stop indicator orders the stop of a random access procedure in a corresponding serving cell or not. For example, if a random access procedure is in progress in the corresponding serving cell, the preamble index indicates the stop of the random access procedure. If a random access procedure is not in progress in the corresponding serving cell, the preamble index indicates the start of a new random access procedure. Accordingly, the meaning that a 'random access procedure is in progress' must be clearly defined.

A random access procedure may be said to be in progress before the random access procedure is completed. In other words, when the random access procedure is completed, the random access procedure is no longer in progress. The state in which the random access procedure has been completed may be differently defined depending on whether the random access procedure is based on contention or non-contention.

First, the non-contention-based random access procedure is described below. For example, if a random access procedure is sought to be completed, the random access procedure has to be started in response to the order of a BS, a preamble indicated by the BS has to be received from a UE through designated time/frequency resources, and an RAR message has to be sent to the UE. Here, the RAR message is an MAC control element, and the sub-header of the MAC control element includes a preamble index indicated by the BS as in FIG. 10.

Referring to FIG. 10, a sub-header 1000 includes an E field 1005, a T field 1010, and a Random Access Preamble ID (RAPID) field 1015. The E field 1005 is a bit that informs whether another MAC header field exists or not. When the E field 105 is '1', it indicates that one or more header fields exists subsequently. When the E field 105 is '', it indicates that an MAC RAR or padding is started from a next byte. The T field 1010 is a bit that informs whether the MAC sub-header is for an RAPID or a backoff indicator. When the T field 1010 is '0', it means a backoff indicator. When the T field 1010 is '1', it means an RAPID.

For another example, if a BS has transmitted an RAR message to a UE in an nth frame and checked that an RAP has not been received from the UE in the earliest frame ((n+1)th or (n+2)th) that arrives after the time (3 ms or 4 ms) during which it can be checked that the UE has received the RAR message, the BS considers that a random access procedure has been terminated. Here, it is a precondition that the number of times that an RAP was transmitted has not reached a maximum retransmission number. If it is determined that the number of times that an RAP was transmitted has reached the maximum retransmission number, the BS determines that the random access procedure has been terminated without a procedure of checking whether an RAP has been received or not.

Next, the contention-based random access procedure is described below. For example, a BS receives scheduled data from a UE and checks that an RAP has been received from what UE based on a Cell-Radio Network Temporary Identifier (C-RNTI) value within the scheduled data. Furthermore, the BS determines that a random access procedure has been completed when a CR message is transmitted or a PDCCH scrambled to a C-RNTI is transmitted by a secondary serving cell. Here, it is a precondition that the number of times that an RAP was transmitted has not reached a maximum retransmission number.

For yet another example, if a BS has received ACK information on a PDCCH from a UE after sending the PDCCH scrambled to a C-RNTI in an nth subframe, the BS determines that a random access procedure has been completed.

The BS determines that the random access procedure is in progress other than the three cases in which a random access procedure is considered as having been completed. In this case, the condition ii) is satisfied. In some embodiments, the cases where a random access procedure is considered as having been completed may include more cases other than the three cases.

Referring back to step S915 of FIG. 9, if, as a result of the determination, it is determined that all the stop conditions are satisfied, the UE stops the random access procedure at step S920. The stop of the random access procedure may be the same as an operation when a random access procedure in a secondary serving cell fails. For example, the UE does not start a random access procedure in the corresponding secondary serving cell again and does not perform all types of transmission, including the transmission of an SRS and data, through uplink.

If, as a result of the determination at step S915, it is determined that the stop condition ii) is not satisfied, that is, if the random access procedure is not in progress, the UE may ignore the random access stop indicator received from the BS. In some embodiments, the UE starts a random access procedure by sending an RAP to the BS in a secondary serving cell at step S925. Here, the random access procedure may be based on contention.

A variety of embodiments regarding a random access stop indicator are described below. A serving cell in which a random access procedure is stopped in response to a random access stop indicator may be limited to a secondary serving cell. Accordingly, when a UE receives signaling (e.g., a PDCCH order) that orders the start of a random access procedure in a primary serving cell, the UE may 1) continue to perform an on-going procedure and ignore signaling or 2) stop an on-going procedure and start a new procedure. Here, a criterion for selecting 1) or 2) is not specially limited and may be set in various ways depending on a method of implementing a UE.

The random access stop indicator may be physical layer signaling or may be a higher layer message, such as an MAC layer or an RRC layer.

For example, if the random access stop indicator is physical layer signaling, the random access stop indicator is DCI mapped to a PDCCH. The DCI may have the same form as a form that orders the start of a random access procedure in a secondary serving cell. A UE recognizes the DCI as a request to stop a random access procedure and stops a random access procedure in a secondary serving cell.

In some embodiments, the DCI may have the same form as a form that orders the start of a contention-based random access procedure in a secondary serving cell. In this case, a UE recognizes the DCI as a request to stop a random access procedure and stops a random access procedure in a secondary serving cell.

If only a non-contention-based random access procedure is permitted, a message that requests a contention-based random access procedure may have the function of a random access stop indicator. In this case, if a random access procedure in a secondary serving cell is in progress, the random access procedure is stopped. If a random access procedure is not in progress, a random access stop indicator is ignored when the random access stop indicator is received.

From a viewpoint of the priority of a random access procedure, a primary serving cell has priority over a secondary serving cell. For example, if a UE that is performing a random access procedure in a secondary serving cell receives DCI for performing a contention-based or non-contention-based random access procedure in a primary serving cell through a PDCCH, the random access procedure that is in progress in the secondary serving cell is stopped (fails) and the random access procedure is started in the primary serving cell.

For another example, if a random access stop indicator is an MAC layer message, it may be represented as in FIG. 11.

FIG. 11 is a block diagram showing an MAC PDU including a random access stop indicator in accordance with an embodiment of the present invention.

Referring to FIG. 11, the MAC PDU 1100 includes an MAC header 1110, one or more MAC control elements 1120 to 1125, one or more MAC Service Data Units (SDUs) 1130-1 to 1130-m, and padding 1140.

The MAC header 1110 includes one or more sub-header 1110-1, 1110-2 to 1110-k. Each of the sub-headers 1110-1, 1110-2 to 1110-k correspond to one MAC SDU or one or more MAC control elements 1120 to 1125 or the padding 1140. The sub-headers 1110-1, 1110-2 to 1110-k are arranged in the same order as the MAC SDU, the MAC control elements 1120 to 1125, or the padding 1140 within the MAC PDU 1100.

Each of the sub-headers 1110-1, 1110-2 to 1110-k may include 4 fields: R, R, E, and LCID or 6 fields: R, R, E, LCID, F, and L. A sub-header including the 4 fields corresponds to the MAC control elements 1120 to 1125 or the padding 1140, and a sub-header including the 6 fields corresponds to the MAC SDUs 1130-1 to 1130-m.

The LCID field is an ID field that identifies a logical channel corresponding to an MAC SDU, orders the start of a random access procedure, orders the stop of a random access procedure, and indicates the type of MAC control elements 1120 to 1125 or the padding 1140. When each of the sub-headers 1110-1, 1110-2 to 1110-k has an octet structure, the LCID field may have 5 bits.

For example, the LCID field may indicate whether an MAC control element orders the start of a random access procedure in a current serving cell or not as in Table 4 below.

Table 4

LCID index	LCID value
00000	CCCH
00001-01010	ID of logical channel
01011-11010	Reserved
11011	Activated or deactivated
11100	UE CR ID
11101	Order to start RA procedure
11110	DRX order
11111	Padding

Referring to Table 4, when the LCID field has a value of 11101, the MAC control element orders the start of a Random Access (RA) procedure. For example, the MAC control element 1125 is an MAC control element for ordering the start of a random access procedure, and it may include an R field, a preamble ID field, a cell index field, and a mask index field.

For another example, the LCID field may indicate the stop of a random access procedure in a current serving cell as in Table 5.

Table 5

LCID index	LCID value
00000	CCCH
00001-01010	ID of logical channel
01011-11010	Reserved
11011	Activated or deactivated
11100	UE CR ID
11101	Order to stop RA procedure
11110	DRX order
11111	Padding

Referring to Table 5, if the LCID field has a value of 11101, a corresponding MAC control element is an MAC control element that orders the stop of a Random Access (RA) procedure in a specific serving cell. In some embodiments, a payload corresponding to the LCID field is set to 0 bit, and thus it may do not exist. That is, the LDIC field itself may order the stop of the random access procedure.

Next, the MAC control elements 1120 to 1125 are control messages generated by an MAC layer. The padding 1140 has a specific number of bits that are added to make constant the size of the MAC PDU. The MAC control elements 1120 to 1125, the MAC SDUs 1130-1 to 1130-m, and the padding 1140 are collectively called an MAC payload.

For yet another example, if a random access stop indicator is an RRC layer message, the random access stop indicator may be represented by RACH-dedicated configuration information as in Table 6.

Table 6

RACH-ConfigDedicated ::= SEQMSNCE {

SCellIndex INTEGER (1..7)

or

ServCellIndex INTEGER (0..7)

ra-PreambleIndex INTEGER (0..63)

ra-PRACH-MaskIndex INTEGER (0..15)

}

Referring to Table 6, RACH-ConfigDedicated is a message used in an RRC reconfiguration procedure, and it provides a function of stopping a random access procedure. In particular, the message RACH-ConfigDedicated includes a secondary serving cell (SCell) index or an SCell index. A BS sends the message RACH-ConfigDedicated to a UE on a serving cell that is indicated by a secondary SCell index or an SCell index. When the message RACH-ConfigDedicated is received while a random access procedure is performed, the UE stops a random access procedure in a serving cell indicated by a secondary SCell index or an SCell index.

FIG. 12 is a flowchart illustrating a method of performing a random access procedure in accordance with another embodiment of the present invention. This method is performed in order to control a maximum retransmission number of an RAP in a random access procedure in order to prevent the delay of the random access procedure.

Referring to FIG. 12, a BS sets a maximum retransmission number, from among random access parameters, to 0 at step S1200 and sends RACH configuration information to a UE at step S1205. Table below relates to RACH configuration information RACH-ConfigCommon that configures random access parameters, and the random access parameters are generated by an RRC layer.

Table 7

RACH-ConfigCommon ::= SEQMSNCE {

....

ra-SupervisionInfo SEQMSNCE {

preambleTransMax ENUMERATED {

n0, n3, n4, n5, n6, n7, n8, n10, n20, n50, n100, n200},

...

}

-- ASN1STOP

Referring to Table 7, preambleTransMax is a maximum retransmission number of a preamble, and it may have a value of 0, 3, 4, 5, 6 to 200. Here, n0, n3, n4, n5, …, nk, …, n200 means preambleTransMax=k.

Referring back to FIG. 12, when determining that a random access procedure in a secondary serving cell needs to be started, the BS maps DCI indicative of the start of the random access procedure to a PDCCH and sends the PDCCH to the UE at step S1210. For example, the random access procedure may be necessary when the BS tries to update an uplink timing advance value.

The BS checks whether the random access procedure in the secondary serving cell has been completed within a maximum delay time preset by the BS or not at step S1215. The maximum delay time may be 50 ms, 70 ms, or 100 ms.

If, as a result of the check, it is checked that the random access procedure has not been completed within the maximum delay time, the BS considers that the random access procedure in the secondary serving cell has failed and sends DCI indicative of the start of a random access procedure in the secondary serving cell to the UE at step S1220. The BS may hold the transmission of the DCI for a preset time. Here, the transmission holding time may be 10 ms, 20 ms, 40 ms, 80 ms, or 160 ms.

The UE ignores the DCI indicative of the start of the random access procedure in the secondary serving cell or starts the random access procedure in the secondary serving cell based on the progress state of the random access procedure in the primary serving cell at step S1225. For example, if the UE starts the random access procedure through a primary serving cell within the transmission holding time and performs the random access procedure, the UE ignores the DCI. For another example, if the random access procedure has been terminated after the UE started the random access procedure through a primary serving cell within the transmission holding time, the UE starts the random access procedure in the secondary serving cell.

If the BS wants to indicate the start of a random access procedure in a serving cell having higher priority, DCI indicative of the start of a random access procedure in a secondary serving cell in which a random access procedure has failed may not be transmitted. Furthermore, a time alignment group including the largest number of secondary serving cells, from among a plurality of time alignment groups, may have higher priority the remaining time alignment groups. For example, it is assumed that a random access procedure is in progress in a first secondary serving cell that belongs to a first time alignment group and a random access procedure is scheduled to be started in a second secondary serving cell that belongs to a second time alignment group. If the second time alignment group includes more secondary serving cells than the first time alignment group, the second time alignment group has higher priority than the first time alignment group. A BS may request a UE to stop a random access procedure in the first secondary serving cell. Here, a primary serving cell may be considered as always having higher priority than a secondary serving cell.

FIG. 13 is a flowchart illustrating a random access procedure performed by a UE in accordance with an embodiment of the present invention.

Referring to FIG. 13, the UE, together with a BS, performs a random access procedure in a secondary serving cell a BS at step S1300. The random access procedure in the secondary serving cell includes the above-described contention-based random access procedure, the above-described non-contention-based random access procedure, and a random access procedure according to the order of a BS.

The UE receives a random access stop indicator, requesting that the random access procedures in the secondary serving cell be stopped, from the BS at step S1305.

For example, the random access stop indicator may be DCI of Format 1A and be the same as that of Table 2. For another example, the random access stop indicator may include an LCID field indicative of an RA procedure stop order in Table 5. For yet another example, the random access stop indicator may be an RRC message including a secondary serving cell or SCell index that will stop the random access procedure as in Table 6.

The UE determines whether a stop condition that the random access procedure in the secondary serving cell is stopped is satisfied or not at step S1310. The stop condition includes i) that a random access stop indicator must be received and ii) that the random access procedure is in progress in a corresponding serving cell. Since the condition i) is satisfied at step S1305, the UE must determine whether the condition ii) is satisfied or not.

The meaning that a random access procedure is in progress is that the random access procedure has not been completed. In relation to the condition ii), a preamble index included in the random access stop indicator may be used to determine whether or not to stop the random access procedure. For example, the preamble index indicates the stop of a random access procedure if the random access procedure is in progress in a corresponding serving cell. If a random access procedure is not in progress in a corresponding serving cell, the preamble index indicates the start of a new random access procedure.

If, as a result of the determination at step S1310, it is determined that the stop condition is satisfied, the UE stops the random access procedure in the secondary serving cell at step S1315. The stop of the random access procedure may be the same as an operation when the random access procedure in the secondary serving cell fails. That is, the UE does not start a random access procedure in the corresponding secondary serving cell and does not perform all types of transmission, including the transmission of an SRS and data through uplink.

If, as a result of the determination at step S1310, it is determined that the stop condition is not satisfied, the UE starts a random access procedure by sending an RAP to the BS in a secondary serving cell at step S1320. Here, the random access procedure may be based on contention.

FIG. 14 is a flowchart illustrating a random access procedure performed by a BS in accordance with an embodiment of the present invention.

Referring to FIG. 14, the BS, together with a UE, performs a random access procedure in a secondary serving cell at step S1400. The random access procedure in the secondary serving cell includes the above-described contention-based random access procedure, the above-described non-contention-based random access procedure, and a random access procedure according to the order of a BS.

The BS determines whether or not to request to stop the random access procedure that is in progress in the secondary serving cell at step S1405. For example, the BS may request the UE to stop a random access procedure after a lapse of some time if a preamble is not received after ordering the UE to perform the random access procedure in a secondary serving cell.

For another example, the BS may request the UE to stop a random access procedure if a random access procedure is started in a serving cell having higher priority than the secondary serving cell or a random access procedure is scheduled to be started. For example, a time alignment group including the largest number of secondary serving cells, from among a plurality of time alignment groups, may have higher priority than the remaining time alignment groups. Here, a primary serving cell may be considered as always having higher priority than a secondary serving cell. If, as a result of the determination at step S1405, it is determined that the random access procedure does not need to be stopped, the BS continues to perform the random access procedure.

If, as a result of the determination at step S1405, it is determined that the random access procedure needs to be stopped, the BS may request the UE to stop the random access procedure in the secondary serving cell. In order to request to stop the random access procedure, the BS sends a random access stop indicator, requesting that the random access procedure in the secondary serving cell be stopped, to the UE at step S1410.

In response to the random access stop indicator, the UE determines whether a stop condition that the random access procedure in the secondary serving cell is stopped or not. If, as a result of the determination, it is determined that the stop condition is satisfied, the UE sends a preamble to the BS. If, as a result of the determination, it is determined that the stop condition is not satisfied, the UE stops a random access procedure in a secondary serving cell that is in progress. Accordingly, the BS receives the preamble from the UE or receives no response at step S1415.

FIG. 15 is a block diagram showing a UE and a BS that perform a random access procedure in accordance with an embodiment of the present invention.

Referring to FIG. 15, the UE 1500 includes a UE reception unit 1505, a UE processing unit 1510, and a UE transmission unit 1515.

The UE reception unit 1505 receives a random access stop indicator, requesting that a random access procedure in a secondary serving cell be stopped, from the BS 1550. For example, the random access stop indicator may be DCI of Format 1A and be the same as that of Table 2.

Furthermore, the UE reception unit 1505 receives RACH configuration information in which a maximum retransmission number is set to 0, from among random access parameters, from the BS 1550. The RACH configuration information may be the same as, for example, Table 7.

The UE processing unit 1510 generates a signal or message necessary for the random access procedure, analyzes a message related to the random access procedure, received from the BS 1550, and stops or starts the random access procedure based on a result of the analysis. The random access procedure in the secondary serving cell includes the above-described contention-based random access procedure, the above-described non-contention-based random access procedure, and a random access procedure according to the order of the BS 1550. For example, the UE processing unit 1510 determines time/frequency resources through which an RAP will be transmitted and performs control so that the UE transmission unit 1515 sends the RAP. Furthermore, the UE processing unit 1510 determines whether a stop condition that the random access procedure in the secondary serving cell has been satisfied or not. The stop condition includes i) that a random access stop indicator must be received and ii) that a random access procedure is in progress in a corresponding serving cell.

The meaning that the random access procedure is in progress is that the random access procedure has not been completed. The condition ii) may be used to determine whether a preamble index included in the random access stop indicator orders the stop of a random access procedure that is in progress in a corresponding serving cell or not. The preamble index indicates the stop of a random access procedure if the random access procedure is in progress in a corresponding serving cell, whereas the preamble index indicates the start of a new random access procedure if the random access procedure is not in progress in the corresponding serving cell.

If, as a result of the determination, it is determined that the stop condition is satisfied, the UE processing unit 1510 stops the random access procedure in the secondary serving cell. The stop of the random access procedure may be the same as an operation when a random access procedure in a secondary serving cell fails. That is, the UE processing unit 1510 does not start a random access procedure in the corresponding secondary serving cell again and does not perform all types of transmission, including the transmission of an SRS and data, through uplink.

If, as a result of the determination, it is determined that the stop condition is not satisfied, the UE processing unit 1510 starts the random access procedure. For example, the UE processing unit 1510 generates an RAP using a preamble sequence and sends the RAP to the UE transmission unit 1515. The UE transmission unit 1515 sends the RAP to the BS 1550 in the secondary serving cell. Here, the random access procedure may be based on contention.

Meanwhile, the UE processing unit 1510 may ignore DCI, indicating the start of the random access procedure in the secondary serving cell, based on a random access procedure in a primary serving cell or start the random access procedure in the secondary serving cell although the DCI is received from the BS. For example, if the UE processing unit 1510 starts the random access procedure through the primary serving cell within a transmission holding time and is performing the random access procedure, the UE processing unit 1510 ignores the DCI. Furthermore, if the random access procedure has been terminated after the UE processing unit 1510 started the random access procedure through a primary serving cell within a transmission holding time, the UE processing unit 1510 starts the random access procedure in the secondary serving cell.

The BS 1550 includes a BS transmission unit 1555, a stop processing unit 1560, and a BS reception unit 1565.

The BS transmission unit 1555 sends a random access stop indicator, requesting that a random access procedure in a secondary serving cell be stopped, to the UE 1500. The BS transmission unit 1555 sets a maximum retransmission number to 0, from among random access parameters, and sends RACH configuration information, such as that of Table 7, to the UE 1500.

The stop processing unit 1560 generates the random access stop indicator. Furthermore, the stop processing unit 1560 determines whether or not to request the UE 1500 to stop the random access procedure that is in progress in the secondary serving cell. For example, the stop processing unit 1560 may request the UE 1500 to stop the random access procedure in the secondary serving cell after a lapse of some time if a preamble is not received after ordering the UE 1500 to stop the random access procedure in the secondary serving cell.

For another example, the stop processing unit 1560 may request the UE 1500 to stop the random access procedure if a random access procedure is started in a serving cell having higher priority than the secondary serving cell or a new random access procedure is scheduled to be started. For example, a time alignment group including the largest number of secondary serving cells, from among a plurality of time alignment groups, may have higher priority than the remaining time alignment groups. Here, a primary serving cell may be considered as always having higher priority than a secondary serving cell. For example, the stop processing unit 1560 may selectively perform any one of a first random access procedure and a second random access procedure based on priority between the first random access procedure and the second random access procedure.

If, as a result of the determination, it is determined that the random access procedure does not need to be stopped, the stop processing unit 1560 continues to perform an on-going random access procedure.

If, as a result of the determination, it is determined that the random access procedure needs to be stopped, the stop processing unit 1560 may request the UE 1500 to stop the random access procedure in the secondary serving cell. In order to request to stop the random access procedure, the stop processing unit 1560 generates a random access stop indicator, requesting that the random access procedure in the secondary serving cell be stopped, and sends the generated random access stop indicator to the BS transmission unit 1555.

If it is determined that the random access procedure in the secondary serving cell needs to be started, the stop processing unit 1560 generates DCI indicative of the start of the random access procedure and transmits the DCI to the BS transmission unit 1555. For example, the random access procedure may be necessary when the BS 1550 tries to update an uplink timing advance value.

The stop processing unit 1560 checks whether the random access procedure in the secondary serving cell has been completed within a preset maximum delay time or not. The maximum delay time may be 50 ms, 70 ms, or 100 ms. If, as a result of the check, it is checked that the random access procedure has not been completed within the maximum delay time, the stop processing unit 1560 considers that the random access procedure in the secondary serving cell has failed and determines whether or not to send DCI, indicating the start of the random access procedure in the secondary serving cell to the UE 1500. Before determining whether or not to send the DCI, the stop processing unit 1560 may hold the transmission of the DCI for a set time. Here, the transmission holding time may be 10 ms, 20 ms, 40 ms, 80 ms, or 160 ms.

If, as a result of the determination, it is determined that the stop condition is satisfied, the UE 1500 sends the RAP to the BS 1550. If, as a result of the determination, it is determined that the stop condition is not satisfied, the UE 1500 stops the random access procedure in the secondary serving cell that is in progress. Accordingly, the BS 1500 receives the RAP from the UE 1500 or receives no response.

(2) Method of determining whether a random access procedure has been terminated or not or will be successful

Meanwhile, if a random access procedure is already in progress in a specific serving cell and a UE tries to start a new random access procedure through another serving cell spontaneously or the UE receives information indicative of the start of a new random access procedure from a BS, the UE has to determine whether or not to continue to perform the existing random access procedure or whether or not to terminate the existing random access procedure and then start a new random access procedure. If the UE determines to terminate the existing random access procedure and start a new random access procedure, conditions that this procedure is performed and the steps of the procedure need to be defined in detail. That is, the start and end (or a success, failure, or stop) of a random access procedure need to be clearly defined. If not, there may be a problem in that a random access procedure is not started through another serving cell while a UE waits for a Random Access Response (RAR) in a random access procedure in a specific serving cell. Accordingly, the present invention discloses a possible condition that a random access procedure is ended and also discloses the operations of a BS and a UE.

FIG. 16 is a flowchart illustrating a method of performing a random access procedure in accordance with an embodiment of the present invention.

Referring to FIG. 16, a UE, together with a BS, performs a first random access procedure in a first serving cell at step S1600. The first random access procedure may be performed in response to a PDCCH order issued by the BS in order to indicate a contention-based random access procedure as in FIG. 7 or may be a non-contention-based random access procedure spontaneously started by the MAC layer of the UE as in FIG. 8. In some embodiments, the first random access procedure may be performed in response to a PDCCH order issued by the BS in order to indicate a non-contention-based random access procedure. Here, a first serving cell may be a primary serving cell, and a second serving cell may be a secondary serving cell. In some embodiments, the first serving cell may be a secondary serving cell, and the second serving cell may be a primary serving cell. In some embodiments, both the first serving cell and the second serving cell may be secondary serving cells.

If the first random access procedure is a contention-based random access procedure, the UE may wait to receive a CR message from a BS. The state in which the UE waist to receive the CR message (i.e., the state in which a CR timer is operating) is formally the state in which the first random access procedure is in progress without being stopped, but is substantially the case where the first random access procedure is considered as having been finished successfully. If the transmission of a PDCCH is informed by a lower layer of the UE and any one of the following conditions 1), 2), and 3) is satisfied, the UE considers that a CR is successful, terminates the CR timer, deletes a temporary C-RNTI, and recognizes that the first random access procedure has been successfully terminated. In order to terminate the first random access procedure successfully, the UE determines whether the three conditions below are satisfied or not.

1) Whether the first random access procedure has been started by the MAC layer of the UE, the transmission of a PDCCH has been designated by a C-RNTI, and the PDCCH includes an uplink grant for new transmission or not

2) Whether the first random access procedure in a serving cell has been started by a PDCCH order and the transmission of a PDCCH has been designated by a C-RNTI or not

3) Whether a random access start indicator indicative of the start of a second random access procedure in a second serving cell has been received while the first random access procedure is performed

In all the conditions 1), 2), and 3), it is a precondition that the first random access procedure that is in progress is in a success-enabled state. For example, when the UE receives a random access start indicator as in the condition 3) in the success-enabled state of the first random access procedure, the UE may determine that the first random access procedure is successful and terminate the first random access procedure.

The first random access procedure at step S1600 may be placed in a success-enabled state or a success-unknown state. The reason why the first random access procedure is divided into the two states is that whether the first random access procedure is substantially successful or not may depend on each of the two states.

In a contention-based random access procedure, i) the success-enabled state: the state in which a CR timer (mac-ContentionResolutionTimer) is operating. The CR timer is started when the UE sends scheduled data to the BS or is restarted when the UE sends the scheduled data again according to a Hybrid Automatic Repeat request (HARQ) operation regarding the scheduled data. ii) the success-unknown state: the state in which the UE has received a random access start indicator indicative of the start of the contention-based random access procedure from the BS, but has not transmitted a preamble, the state in which the UE has transmitted the preamble, but has not received an RAR, or the state in which the UE has received the RAR, but has not transmitted scheduled data.

In a non-contention-based random access procedure, i) the success-enabled state: the state in which the UE is placed in an RAR window section (3 ms to 10 ms) after a lapse of 3 ms since the UE transmitted a dedicated preamble to the BS, and ii) the success-unknown state: the state in which the UE has received a message, ordering that the non-contention-based random access procedure be started, from the BS, but has not transmitted a preamble.

While the first random access procedure is in progress, the BS transmits a random access start indicator, ordering that a second random access procedure in a second serving cell to be started, to the UE at step S1605. At a point of time at which the random access start indicator is received, the first random access procedure may be in the success-enabled state or the success-unknown state.

When the random access start indicator is received, the UE terminates the first random access procedure at step S1610. The end of the first random access procedure includes a success or a failure. The UE has to start the second random access procedure in response to the order of the random access start indicator in the second serving cell. Here, the UE cannot start the second random access procedure if the first random access procedure is not finished because the first random access procedure continues to be performed. This is because a parallel random access procedure is not taken into consideration as described above. Since the condition 3) is satisfied, however, the UE considers the random access start indicator as signaling to terminate the first random access procedure and successfully terminates the first random access procedure. In other words, if the random access start indicator ordering that the second random access procedure in the second serving cell be started is received from the BS while the first random access procedure is performed in the first serving cell, the UE terminates the first random access procedure and performs the second random access procedure.

In this case, anterior signaling for terminating the first random access procedure is not necessary. That is, the end of the first random access procedure and the start of the second random access procedure are performed at once in response to the random access start indicator without being performed by different signalings. Accordingly, a procedure can be simplified, and the deterioration of performance due to signaling delay can be reduced. Furthermore, a condition that a UE starts a random access procedure, and a criterion that an on-going random access procedure is processed based on a serving cell that is performing a random access procedure and a serving cell that receives a random access procedure start indicator are proposed.

An example in which the first random access procedure is successfully terminated has been described. However, an example in which the first random access procedure is unsuccessfully terminated may exist. For example, if a first serving cell is a secondary serving cell, a second serving cell is a primary serving cell, and a first random access procedure is now in the success-unknown state, a UE determines that the first random access procedure has failed, terminates the first random access procedure, and starts a second random access procedure. For another example, if a first random access procedure is started by a random access start indicator that orders a non-contention-based random access procedure, the first random access procedure is in the success-enabled state, a first serving cell is a secondary serving cell, and a second serving cell is a primary serving cell, a UE determines that the first random access procedure has failed and starts a second random access procedure in response to the random access start indicator.

After terminating the first random access procedure, the UE transmits an RAP to the BS in order to start the second random access procedure at step S1615. The second random access procedure may be based on contention or non-contention. If the second random access procedure is based on non-contention, the UE transmits a preamble index defined in Table 1 and a preamble designated by a PRACH mask index to the BS.

In FIG. 16, it has been assumed that the second random access procedure is related to the second serving cell, but the second random access procedure may be performed in relation to the first serving cell. That is, the technical spirit of the present invention in FIG. 16 may also be applied to the case where a random access procedure that is already in progress in the same serving cell is terminated and a new random access procedure is started. In this case, a UE performs a basic operation. The basic operation includes an operation of the UE continuing to perform the first random access procedure and ignoring signaling or an operation of the UE stopping the first random access procedure that is in progress and starting the second random access procedure newly. The UE may perform only one of the two basic operations. A criterion for selecting one of the two basic operations may be determined in various ways depending on a method of implementing a UE.

A random access start indicator is described in more detail below. The random access start indicator may be physical layer signaling or may be a higher layer message, such as an MAC layer or an RRC layer.

For example, if a random access start indicator is physical layer signaling, the random access start indicator is DCI mapped to a PDCCH. The DCI orders the start of a random access procedure in a second serving cell and may have the same form as that of Table 1. Furthermore, the random access start indicator may order the start of a non-contention-based random access procedure or the start of a contention-based random access procedure.

For another example, if a random access start indicator is an MAC layer message, the random access start indicator may be represented as in FIG. 17.

FIG. 17 is a block diagram showing an MAC Protocol Data Unit (PDU) including a random access start indicator in accordance with an embodiment of the present invention.

Referring to FIG. 17, the MAC PDU 1700 includes an MAC header 1710, one or more MAC control elements 1720 to 1725, one or more MAC Service Data Units (SDUs) 1730-1 to 1730-m, and padding 1740.

The MAC header 1710 includes one or more sub-headers 1710-1, 1710-2 to 1710-k. Each of the sub-header 1710-1, 1710-2 to 1710-k corresponds to one MAC SDU, one or more MAC control elements 1720 to 1725, or the padding 1740. The sub-headers 1710-1, 1710-2 to 1710-k are arranged in the same order as the MAC SDU, the MAC control elements 1720 to 1725, or the padding 1740 within the MAC PDU 1700.

Each of the sub-headers 1710-1, 1710-2 to 1710-k may include 4 fields: R, R, E, and LCID or 6 fields: R, R, E, LCID, F, and L. The sub-header including 4 fields corresponds to the MAC control elements 1720 to 1725 or the padding 1740, and the sub-header including 6 fields corresponds to the MAC SDU.

The Logical Channel ID (LCID) field is an ID field that identifies a logical channel corresponding to the MAC SDU, orders the start of a random access procedure, or identifies the type of MAC control elements 1720 to 1725 or the padding 140. When each of the sub-headers 1710-1, 1710-2 to 1710-k has an octet structure, the LCID field may have 5 bits.

The LCID field indicates whether an MAC control element orders that a random access procedure be started in a current serving cell or not as in Table 8.

Table 8

LCID index	LCID value
00000	CCCH
00001-01010	ID of logical channel
01011-11010	Reserved
11011	Activated or deactivated
11100	UE CR ID
11101	RA procedure start order
11110	DRX order
11111	Padding

Referring to Table 8, when the LCID field has a value of 11101, an MAC control element is an MAC control element that orders the start of a Random Access (RA) procedure.

The MAC control elements 1720 to 1725 are control messages generated by an MAC layer. The padding 1740 has a specific number of bits that are added in order to make constant the size of the MAC PDU.

In a non-contention-based random access procedure, the MAC control element 1725 that orders the starts of a Random Access (RA) procedure may include two R fields, a preamble ID field of 6 bits, a cell index field of 3 bits, and a mask index field of 4 bits, as in FIG. 17. If the MAC control element 1725 is transmitted through the PDSCH of a serving cell that always orders the starts of a random access procedure, the MAC control element 1725 may not include an SCell index or a TAG index. The size of the TAG index field may be any one of 1, 2, and 3 bits. If the mask index field is defined as 2 bits the MAC control element may be represented as in FIG. 18.

Referring to FIG. 18, the former two bits of a preamble ID field 1810 in an MAC control element 1800 that forms an octet structure may be used as a mask index field 1805. In this case, the existing R fields are used as the mask index field 1805. Accordingly, a payload size is 8 bits. Here, a PRACH mask index may be defined as in Table 9.

Table 9

PRACH mask index	Permitted PRACH(FDD)	Permitted PRACH(TDD)
0	All	All
1	All even-numbered PRACH opportunities within the time domain,The first PRACH resource index within a subframe	All even-numbered PRACH opportunities within the time domain,The first PRACH resource index within a subframe
2	All odd-numbered PRACH opportunities within the time domain, The first PRACH resource index within subframe	All even-numbered PRACH opportunities within the time domain,The first PRACH resource index within a subframe
3	Reserved	The first PRACH resource index within a subframe

In a contention-based random access procedure, if the MAC control element 1725 is transmitted through the PDSCH of a serving cell that always orders the start of a random access procedure, an MAC sub-header corresponding to the MAC control element 1725 may not include an SCell index or a TAG index. Furthermore, if the MAC control element 1725 may be transmitted through the PDSCH of another serving cell not through the PDSCH of a serving cell that orders the start of a random access procedure (e.g., Cross-Carrier Scheduling (CCS)), an MAC sub-header corresponding to the MAC control element 1725 may include an SCell index or a TAG index.

If an MAC sub-header corresponding to the MAC control element 1725 includes a TAG index, the MAC sub-header may be represented as in FIG. 19 or FIG. 20. Referring to FIGS. 19 and 20, an MAC sub-header 1900 may include one R field 1905 and a TAG index field 1910 of 1 bit on the left side of an E field 1915 of 1 bit, and an LCID field 1920 of 5 bits on the right side of the E field 1915, as shown in FIG. 19. An MAC sub-header 2000 may include a TAG index field 2005 of 2 bits on the left side of an E field 2010 and an LCID field 2015 of 5 bits on the right side of the E field 2010, as shown in FIG. 20. Here, each of the LCID fields 1920 and 2015 has a value corresponding to an MAC control element that orders the start of a Contention-Based Random Access (BRA) procedure.

If the TAG index field 1910 is defined as 1 bit, the TAG index field 1910 may define two TAGs. For example, when the TAG index is '0', the TAG index indicates a TAG including a primary serving cell. When the TAG index is '1', the TAG index indicates a TAG including only secondary serving cells. If the TAG index field 2005 is defined as 2 bits, the TAG index field 2005 may define a total of four TAGs. For example, when the TAG index is '0', the TAG index indicates a TAG including a primary serving cell. When the TAG index is ''1', '2', or '3', the TAG index indicates a TAG including only secondary serving cells. A method of configuring the TAG field in the MAC sub-header as described above may also be configured in an MAC control element that is used in a non-contention-based random access procedure.

For yet another example, if a random access start indicator is the message of an RRC layer, the random access start indicator may be represented as in Table 10.

Table 10

RACH-ConfigDedicated ::= SEQMSNCE {

SCellIndex INTEGER (1..7)

or

ServCellIndex INTEGER (0..7)

ra-PreambleIndex INTEGER (0..63),

ra-PRACH-MaskIndex INTEGER (0..15)

}

Referring to Table 10, RACH-ConfigDedicated is a message used in an RRC reconfiguration procedure, and the message provides a function for starting a random access procedure. In particular, the message RACH-ConfigDedicated includes a secondary SCell index or an SCell index. A BS transmits the message RACH-ConfigDedicated to a UE on a serving cell that is indicated by a secondary SCell index or an SCell index. When the message RACH-ConfigDedicated is received while a random access procedure is performed in a specific serving cell, the UE may start a random access procedure in a second serving cell that is indicated by a secondary SCell index or an SCell index within the message RACH-ConfigDedicated. If a value ra-PreambleIndex is 0, the UE starts a contention-based random access procedure. In this case, a value ra-PRACH-MaskIndex is set to 1 by the UE.

A detailed operation of a UE that performs a random access procedure in accordance with the present invention is described below.

FIG. 21 is a flowchart illustrating a random access procedure performed by a UE in accordance with an embodiment of the present invention. In FIG. 21, it is a precondition that a first random access procedure is in progress is in a primary serving cell, the first random access procedure has been started by the MAC layer of the UE or in response to a message that orders a contention-based random access procedure, and the first random access procedure is now in the success-enabled state.

Referring to FIG. 21, the UE performs the first random access procedure in the primary serving cell at step S2100. While the first random access procedure is performed, the UE receives a random access start indicator, indicating the start of a second random access procedure based on contention or non-contention, from a BS at step S2105.

The UE determines whether a serving cell in which the second random access procedure is performed is a primary serving cell or not at step S2110. If, as a result of the determination, it is determined that a serving cell in which the second random access procedure is a primary serving cell, the UE performs a basic operation at step S2115. If, as a result of the determination, it is determined that a serving cell in which the second random access procedure is not a primary serving cell, but a secondary serving cell, the UE determines that the first random access procedure in the primary serving cell has been successfully terminated and performs the second random access procedure in the secondary serving cell at step S2120.

FIG. 22 is a flowchart illustrating a random access procedure performed by a UE in accordance with another embodiment of the present invention. It is a precondition that a first random access procedure is in progress in a primary serving cell or a secondary serving cell, the first random access procedure has been stared in response to a message that orders a non-contention-based random access procedure, and the first random access procedure is now in a success-enabled state.

Referring to FIG. 22, the UE performs the first random access procedure in a first serving cell at step S2200. While performing the first random access procedure, the UE receives a random access start indicator that orders the start of a second random access procedure based on contention or non-contention in a second serving cell from a BS at step S2205.

The UE determines whether the first serving cell is identical with the second serving cell or not at step S2210. If, as a result of the determination, it is determined that the first serving cell is identical with the second serving cell, the UE performs a basic operation at step S2215. If, as a result of the determination, it is determined that the first serving cell is not identical with the second serving cell, the UE determines whether the second serving cell is a secondary serving cell at step S2220.

If, as a result of the determination at step S2220, it is determined that the second serving cell is a secondary serving cell, the first serving cell becomes a primary serving cell because the first serving cell is not identical with the second serving cell. Since the primary serving cell has priority over the secondary serving cell, the first serving cell has priority over the second serving cell. Accordingly, the UE ignores the random access start indicator regarding the second secondary serving cell at step S2225.

In contrast, if, as a result of the determination at step S2220, it is determined that the second serving cell is not a secondary serving cell, it means that the first serving cell is a secondary serving cell and the second serving cell is a primary serving cell. Accordingly, the UE determines that the first random access procedure in the first serving cell has failed and performs the second random access procedure at step S2230.

FIG. 23 is a flowchart illustrating a random access procedure performed by a UE in accordance with yet another embodiment of the present invention. It is a precondition that a first random access procedure is in progress in a secondary serving cell, the first random access procedure has been started in response to a message that orders a contention-based random access procedure, and the first random access procedure is now in a success-enabled state.

Referring to FIG. 23, the UE performs the first random access procedure in a first secondary serving cell at step S2300. While performing the first random access procedure, the UE receives a random access start indicator, indicating the start of a second random access procedure based on contention or non-contention, from a BS at step S2305.

The UE determines whether a serving cell in which the second random access procedure is in progress is the first secondary serving cell or not at step S2310. If, as a result of the determination, it is determined that a serving cell in which the second random access procedure is in progress is the first secondary serving cell, the UE performs a basic operation at step S2315. If, as a result of the determination at step S2310, it is determined that a serving cell in which the second random access procedure is in progress is not the first secondary serving cell, the UE determines that the first random access procedure in the first secondary serving cell has been successful and performs the second random access procedure at step S2320. Here, the serving cell in which the second random access procedure is in progress may be a primary serving cell or a second secondary serving cell.

FIG. 24 is a flowchart illustrating a random access procedure performed by a UE in accordance with further yet another embodiment of the present invention. It is a precondition that a first random access procedure is now in a success-unknown state.

Referring to FIG. 24, the UE performs the first random access procedure in a first serving cell at step S2400. While performing the first random access procedure, the UE receives a random access start indicator, indicating the start of a second random access procedure based on contention or non-contention in a second serving cell, from a BS at step S2405.

The UE determines whether the first serving cell is identical with the second serving cell or not at step S2410. If, as a result of the determination, it is determined that the first serving cell is identical with the second serving cell, the UE performs a basic operation at step S2415. If, as a result of the determination, it is determined that the first serving cell is not identical with the second serving cell, the UE determines whether the second serving cell is a secondary serving cell or not at step S2420.

If, as a result of the determination at step S2420, it is determined that the second serving cell is a secondary serving cell, the UE performs the basic operation at step S2415.

In contrast, if, as a result of the determination at step S2420, it is determined that the second serving cell is not a secondary serving cell, it means that the first serving cell is a secondary serving cell and the second serving cell is a primary serving cell. Accordingly, the UE determines that the first random access procedure in the first serving cell has failed and performs the second random access procedure at step S2425.

FIG. 25 is a flowchart illustrating a random access procedure performed by a BS in accordance with an embodiment of the present invention.

Referring to FIG. 25, the BS, together with a UE, performs a first random access procedure in a first serving cell at step S2500. The BS terminates the first random access procedure successfully or unsuccessfully and transmits a random access start indicator, indicating the start of a second random access procedure in a second serving cell, to the UE at step S2505. The second random access procedure may be based on contention or non-contention.

When the second random access procedure is started, the BS receives a preamble from the UE or receives no response at step S2510.

FIG. 26 is a block diagram showing a UE and a BS that perform a random access procedure in accordance with an embodiment of the present invention.

Referring to FIG. 26, the UE 2600 includes a UE reception unit 2605, a random access processing unit 2610, and a UE transmission unit 2615.

The UE reception unit 2605 receives a random access start indicator that terminates a first random access procedure that is already in progress in a first serving cell successfully or unsuccessfully and that orders the start of a second random access procedure in a second serving cell from a BS 2650.

The first random access procedure may be performed in response to a PDCCH order issued by the BS in order to indicate a contention-based random access procedure as in FIG. 7 or may be a non-contention-based random access procedure spontaneously started by the MAC layer of the UE as in FIG. 8. In some embodiments, the first random access procedure may be performed in response to a PDCCH order issued by the BS in order to indicate a non-contention-based random access procedure. Here, a first serving cell may be a primary serving cell, and a second serving cell may be a secondary serving cell. In some embodiments, the first serving cell may be a secondary serving cell, and the second serving cell may be a primary serving cell. In some embodiments, both the first serving cell and the second serving cell may be secondary serving cells.

The random access processing unit 2610 generates a message or preamble necessary for a random access procedure. Furthermore, the random access processing unit 2610 determines time/frequency resources through which the preamble will be transmitted and performs control so that the UE transmission unit 2615 transmits the preamble. Furthermore, the random access processing unit 2610 determines whether a first random access procedure is in a success-enabled state or a success-unknown state. Furthermore, the random access processing unit 2610 completes the first random access procedure successfully or unsuccessfully, performs a basic operation, or performs a second random access procedure newly depending on whether the first random access procedure is in the success-enabled state or the success-unknown state and whether the first random access procedure is based on contention or non-contention. The random access processing unit 2610 determines whether the first random access procedure is substantially successful or not according to the conditions 1 to 3.

The UE transmission unit 2615 transmits an RAP to the BS 2650 using time/frequency resources determined by the random access processing unit 2610 and a preamble index.

The BS 2650 includes a BS transmission unit 2655, a message processing unit 2660, and a BS reception unit 2665.

The BS transmission unit 2655 transmits a random access start indicator, ordering that a random access procedure be ended in a first serving cell, to the UE 2600.

The message processing unit 2660 generates a message related to a first random access procedure in a first serving cell, determines whether the first random access procedure needs to be ended or not or a second random access procedure needs to be started or not, and generates a random access start indicator that orders the start of a second random access procedure in a second serving cell different from the first serving cell. For example, if a timer related to the uplink synchronization of the UE 2600 expires and the BS 2650 requires information related to new uplink synchronization, the message processing unit 2660 generates a random access start indicator in order to induce the start of a second random access procedure. In some embodiments, if the UE 2600 has to perform a Buffer State Report (BSR) in order to send uplink data, the BS 2650 may generate a random access start indicator in order to induce the start of the second random access procedure.

The message processing unit 2660 transfers the generated random access start indicator to the BS transmission unit 2655.

The BS reception unit 2665 receives a preamble from the UE 2600 or receives no response.

(3) Method of terminating a random access procedure by taking priority into consideration

FIG. 27 is a flowchart illustrating a random access procedure performed by a UE in accordance with an embodiment of the present invention.

Referring to FIG. 27, the UE perform a current random access procedure at step S2700. The current random access procedure is intended by a first serving cell and has been attempted n times (i.e., nth rounds) (0≤n≤maxRA). If n=0, the current random access procedure does not correspond to a reattempt, but the first attempt. maxRA is a maximum reattempt number that the UE can retransmit an RAP after the current random access procedure results in a short-term failure and is also called a maximum reattempt number. To retransmit the RAP may also be called a reattempt at a random access procedure. In the short-term failure, unlike in a final failure, the current random access procedure is not completely terminated, and the current random access procedure can be reattempted until the maximum reattempt number is reached. A preamble transmission counter PREAMBLE_TRANSMISSION_COUNTER is increased by 1 every short-term failure.

The short-term failure includes the case where a UE does not receives an RAR from a BS within an RAR window or does not receive a CR message from a BS in a contention-based random access procedure, such as that shown in FIG. 7. In some embodiments, the short-term failure may include the case where step S810 is not normally completed in a non-contention random access procedure, such as that shown in FIG. 8.

For example, if a current random access procedure that is firstly attempted (i.e., a 0th attempt) results in a short-term failure, a UE may firstly reattempt the current random access procedure. Furthermore, if the firstly reattempted current random access procedure results in a short-term failure again, the UE may secondly reattempt the current random access procedure. If a current random access procedure continues to experience a short-term failure until a maximum reattempt number and thus the UE is unsuccessful in the current random access procedure, the UE treats the current random access procedure as a final failure.

While a current random access procedure that is attempted nth times is in progress, the UE determines whether a midway termination condition is satisfied or not at step S2705. The midway termination is to make a current random access procedure a final failure midway after a lapse of some time. The midway termination condition includes 1) the case where a UE receives a random access start message, ordering that a new random access procedure intended for a second serving cell be started, from a BS or 2) the case where a contention-based random access procedure is scheduled to be spontaneously started in a primary serving cell in the MAC layer of the UE when a UE has to send a Buffer State Report (BSR) or for a reason of the expiration of a Time Alignment Timer (TAT). A first serving cell may be a secondary serving cell, and a second serving cell may be a primary serving cell. In some embodiments, both the first serving cell and the second serving cell may be secondary serving cells. If any one of the conditions 1) and 2) is satisfied, a UE may perform the midway termination.

If the nth attempted current random access procedure has not been terminated although the midway termination condition is satisfied, the UE continues to perform the current random access procedure and waits for results thereof without immediately terminating the current random access procedure at step S2710. If the current random access procedure is terminated right after the midway termination condition is satisfied, it results in an opportunity that the current random access procedure may be successful. For this reason, if the nth attempted current random access procedure is in progress, the UE performs midway termination and a new random access procedure while checking a result of a corresponding attempt.

The UE determines whether the nth attempted current random access procedure results in a short-term failure or not at step S2715. If, as a result of the determination, the nth attempted current random access procedure is successfully completed without resulting in a short-term failure, the UE terminates the random access procedure performed in the first serving cell and starts a new random access procedure in a second serving cell at step S2720.

If, as a result of the determination at step S2715, it is determined that the nth attempted current random access procedure results in a short-term failure, the UE increases a preamble transmission counter by 1, no longer reattempts the current random access procedure although the preamble transmission counter is not equal to a maximum reattempt number+1, and declares midway termination at step S2725. Furthermore, the UE starts a new random access procedure in a second serving cell. That is, the UE does not attempt an (n+1)th current random access procedure. When the UE declares the midway termination, a current random access procedure is considered as being a final failure although a reattempt number in a current random access procedure remains until a final reattempt number. A procedure performed after the midway termination is the same as a procedure performed after the final failure.

If, as a result of the determination at step S2705, it is determined that the midway termination condition is not satisfied, the UE continues to perform the current random access procedure until it is successful within a maximum reattempt number at step S2730.

The stability of a random access procedure can be provided by compromising the profits of an anterior random access procedure and a posterior random access procedure in selecting any one of random access procedures regarding different serving cells in a situation in which the random access procedures occur frequently.

The random access procedure of the UE described with reference to FIG. 27 may be embodied in a form, such as that shown in Table 11 below. In this case, it is a precondition that a current random access procedure results in a short-term failure because the UE has not received an RAR within an RAR window, a current random access procedure results in a short-term failure because any one of RARs received by the UE does not include the ID of an RA preamble corresponding to an RAP transmitted by the UE, or a current random access procedure results in a short-term failure because the UE has not received a CR message.

Table 11

a) If a current random access procedure is intended for a secondary serving cell,

b-1) if a preamble transmission counter=maxRA+1 or

b-2) if a new random access procedure intended for a primary serving cell is started by the MAC layer of a UE or a new random access procedure intended for a primary serving cell is started by a random access start indicator,

c-1) the UE determines that the current random access procedure has resulted in a final failure.

b-3) if the current random access procedure results in a final failure,

c-2) the UE terminates the current random access procedure.

b-4) or else,

c-3) If an RA preamble is selected by the MAC layer in the current random access procedure,

d-1) the UE selects a random backoff time according to a uniform distribution between 0 and a backoff parameter value based on the backoff parameter of the UE.

d-2) the UE delays subsequent or new random access transmission according to a backoff time.

c-4) the UE selects random access resources for reattempting a current random access procedure.

An example in which the step S2705 is performed earlier than the step S2715 has been illustrated in FIG. 27, but is only illustrative. The technical spirit of the present invention includes an embodiment in which the step S2715 is performed earlier than the step S2705.

FIG. 28 shows an example of a scenario in which a UE performs a random access procedure in accordance with the present invention.

Referring to FIG. 28, the UE starts a current random access procedure in a first serving cell at step S2800. This corresponds to the first reattempt. It is assumed that a maximum reattempt number is set to 3. The first attempt (i.e., a 0th attempt) is not shown in the drawings.

If the firstly reattempted current random access procedure results in a short-term failure at step S2805, the UE checks whether a backoff parameter has been received from a BS or not if the random access procedure is a contention-based random access procedure. If, as a result of the check, it is checked that the backoff parameter has not been received, the UE enters a secondly reattempted current random access procedure without a backoff.

If, as a result of the check, it is checked that the backoff parameter has been received, the UE checks the backoff parameter selects one of 0 to a time defined by the backoff parameter based on a uniform distribution. Furthermore, the UE waits for by the selected time enters the secondly reattempted current random access procedure. The backoff time is included in the firstly reattempted period.

While the secondly reattempted current random access procedure is performed, a midway termination condition is satisfied at step S2810. For example, when the UE receives a random access start indicator that orders the start of a new random access procedure in a second serving cell from the BS, the midway termination condition is satisfied. For another example, when the UE spontaneously tries to start a contention-based new random access procedure in a primary serving cell, the midway termination condition is satisfied.

When the midway termination condition is satisfied, the UE waits for a result of the secondly reattempted current random access procedure that is in progress at step S2815. Although the midway termination condition is satisfied, the UE does not terminate the secondly reattempted current random access procedure immediately and holds a new random access procedure until a result of a maximum of a (2+k)th current random access procedure is obtained.

Here, the value k may be set to 0 or a positive integer value and may be transmitted to the UE through a system information channel (BCCH) or RRC signaling. If the UE does not receive any value, an initial k value is set to 0.

Furthermore, a ‘Random Access (RA) procedure holding timer)’ may be used as a method of setting a period in which a new random access procedure is held. Accordingly, a new random access procedure is held until a maximum retransmission number of random access procedures that may be performed until the timer expires.

If, a result of the maximum of the (2+k)th current random access procedure results in a short-term failure at step S2820, the UE terminates the current random access procedure midway although the UE can make one reattempt at step S2825. Furthermore, the UE starts a new random access procedure in a second serving cell.

If a current random access procedure is in a backoff operation and the midway termination condition is satisfied, the UE stops the backoff operation and terminates the current random access procedure midway. Furthermore, the UE starts the new random access procedure in the second serving cell.

In FIG. 28, an example in which a point of time at which the midway termination condition is satisfied is limited to the time when a secondly reattempted current random access procedure is in progress has been described, but is only illustrative. The point of time may include the case where a current random access procedure is attempted for the first time (not shown) or the middle of a firstly reattempted current random access procedure.

A processing method in which priority is taken into consideration between different random access procedures is described below. The reason why priority is taken into consideration in processing different random access procedures is that to process a random access procedure having relatively higher importance with priority helps the stability of a system. Priority may be determined depending on whether a random access procedure is based on contention or non-contention or depending on that a random access procedure is in progress in what serving cell.

FIG. 29 is a flowchart illustrating a random access procedure performed by a UE in accordance with another embodiment of the present invention.

Referring to FIG. 29, the UE performs a current Random Access (RA) procedure in a first serving cell at step S2900.

The UE receives a random access indicator ordering that a new RA procedure be performed in a second serving cell at step S2905.

The UE determines whether the new RA procedure has higher priority than the current RA procedure or not at step S2910.

For example, priority is determined depending on whether a corresponding random access procedure is based on contention or non-contention. For example, a non-contention-based random access procedure has higher priority than a contention-based random access procedure. Accordingly, if the new RA procedure > the current RA procedure is sought to be satisfied, the new RA procedure must be based on non-contention and the current RA procedure must be based on contention. Here, it is a precondition that both the first serving cell and the second serving cell are secondary serving cells.

For another example, priority is determined depending on whether a corresponding random access procedure is in progress in a primary serving cell or a secondary serving cell. For example, a primary serving cell has higher priority than a secondary serving cell, and secondary serving cells have the same priority. Accordingly, if the new RA procedure > the current RA procedure is sought to be satisfied, the new RA procedure must be in progress in a primary serving cell and the current RA procedure must be in progress in a secondary serving cell. Furthermore, if the new RA procedure > the current RA procedure is sought to be satisfied, both the new RA procedure and the current RA procedure must be in progress in secondary serving cells.

For yet another example, priority is determined depending on whether a corresponding random access procedure is started by a UE or the order of a BS. For example, a random access procedure started by the order of a BS has higher priority than a random access procedure started by a UE. Accordingly, if the new RA procedure > the current RA procedure is sought to be satisfied, the new RA procedure must be started by the order of a BS, and the current RA procedure must be started by a UE.

For yet another example, priority is determined depending on that the reason why a corresponding random access procedure is started by a UE results from a Buffer Status Report (BSR) or the expiration of a Time Alignment Timer (TAT). For example, a random access procedure triggered by the expiration of a TAT has higher priority than a random access procedure that is triggered in order to transmit information on the state of an uplink data buffer within a UE. Here, all the reasons why the random access procedure is started can be checked by only a UE in a primary serving cell. Accordingly, random access procedures having different reasons in which the random access procedures are started cannot be performed in a plurality of serving cells.

If, as a result of the determination at step S2910, it is determined that the new RA procedure > the current RA procedure is satisfied, the UE determines that the current RA procedure results in a final failure at step S2915 and starts the new RA procedure in the second serving cell at step S2920.

In contrast, if, as a result of the determination at step S2910, it is determined that the new RA procedure > the current RA procedure is not satisfied, the UE cannot start the new RA procedure until the current RA procedure is terminated because the current RA procedure has higher priority than the new RA procedure. The current RA procedure is terminated when the current RA procedure is determined to be successful or determined to result in a final failure. Accordingly, the UE starts the new RA procedure after the current RA procedure is terminated at step S2925.

FIG. 30 is a flowchart illustrating a random access procedure performed by a UE in accordance with yet another embodiment of the present invention.

Referring to FIG. 30, if a condition that random access procedures are started through two or more serving cells at the same point of time is satisfied at step S3000, a method of performing a plurality of random access procedures is necessary. It is assumed that a condition that two random access procedures are started in a first serving cell and a second serving cell at the same time has been satisfied.

In this case, the two random access procedures may be considered as respective new RA procedures at the same point of time. At the present time, both the two random access procedures are treated as the new random access procedures. However, when one of the two procedures is started with first priority, a random access procedure started with first priority becomes a current RA procedure, and a random access procedure that has not been started remains as a new RA procedure. Accordingly, there is a need for a procedure for selecting a serving cell in which a random access procedure will be first started from random access procedures according to priority.

Next, the UE compares the priorities of the two random access procedures with each other at step S3005. Here, a method of determining priority is the same as a method of determining whether a new RA procedure has higher priority than a current RA procedure or not. That is, the random access procedure of a serving cell in which the random access procedure has not been started because it has low priority becomes the new RA procedure, and the random access procedure of a serving cell in which the random access procedure has been first started because it has high priority becomes the current RA procedure.

If, as a result of the comparison, the first serving cell has higher priority than the second serving cell, the UE determines the random access procedure in the first serving cell as a first priority random access procedure and first performs the random access procedure in the first serving cell at step S3010. Next, the UE determines the random access procedure in the second serving cell as a second priority random access procedure and subsequently performs the random access procedure in the second serving cell at step S3015.

If, as a result of the comparison at step S3005, the first serving cell does not have higher priority than the second serving cell, the UE determines the random access procedure in the second serving cell as a first priority random access procedure and first performs the random access procedure in the second serving cell at step S3020. Furthermore, the UE determines the random access procedure in the second serving cell as a second priority random access procedure and subsequently performs the random access procedure in the first serving cell at step S3025.

As an example of priority, if the random access procedure in the first serving cell is started by the order of a BS and the random access procedure in the second serving cell is started by a UE, the first serving cell has higher priority than the second serving cell.

As another example of priority, if the random access procedure in the first serving cell is triggered by the expiration of a TAT and the random access procedure in the second serving cell is a random access procedure in a secondary serving cell indicated by a BS, the first serving cell has higher priority than the second serving cell. Furthermore, if the random access procedure in the first serving cell is triggered in order to transmit information on the state of an uplink data buffer within a UE and the random access procedure in the second serving cell is a random access procedure in a secondary serving cell indicated by a BS, the second serving cell has higher priority than the first serving cell.

FIG. 31 is a flowchart illustrating a random access procedure in accordance with an embodiment of the present invention.

Referring to FIG. 31, a UE, together with a BS, performs an nth current RA procedure intended for a first serving cell at step S3100. The current RA procedure may be performed in response to a PDCCH order issued by the BS in order to indicate a contention-based random access procedure as in FIG. 7 or may be a non-contention-based random access procedure spontaneously started by the MAC layer of the UE as in FIG. 8. In some embodiments, the current RA procedure may be performed in response to a PDCCH order issued by the BS in order to indicate a non-contention-based random access procedure. Here, the first serving cell may be a primary serving cell, and a second serving cell may be a secondary serving cell. In some embodiments, the first serving cell may be a secondary serving cell, and the second serving cell may be a primary serving cell. In some embodiments, both the first serving cell and the second serving cell may be secondary serving cells.

While the current RA procedure is in progress, the BS transmits a random access start indicator, ordering the start of a new RA procedure in a second serving cell, to the UE at step S3105.

In response to the random access start indicator, the UE determines whether a midway termination condition is satisfied or not at step S3110. If, as a result of the determination, it is determined that the midway termination condition is satisfied, the UE holds the progress of the new RA procedure and determines whether an nth current RA procedure results in a short-term failure or not at step S3115. If, as a result of the determination, it is determined the nth current RA procedure results in a short-term failure, the UE terminates the current RA procedure midway at step S3120 and transmits an RAP for the new RA procedure to the BS at step S3125.

An example in which the step S3110 is performed earlier than the step S3115 has been illustrated in FIG. 31, but is only illustrative. The technical spirit of the present invention includes an embodiment in which the step S3115 is performed earlier than the step S3110.

FIG. 32 is a flowchart illustrating a random access procedure performed by a BS in accordance with an embodiment of the present invention.

Referring to FIG. 32, the BS, together with a UE, performs a current RA procedure intended for a first serving cell at step S3200. The BS transmits a random access start indicator, ordering the start of a new RA procedure intended for a second serving cell, to the UE or waits to receive an RAP spontaneously transmitted by the UE while the current RA procedure is in progress at step S3205. The current RA procedure may be based on contention or non-contention. The first serving cell may be a secondary serving cell, and the second serving cell may be a primary serving cell. In some embodiments, both the first serving cell and the second serving cell may be secondary serving cells.

The BS receives an RAP from the UE or receives no response at step S3210.

FIG. 33 is a block diagram showing a UE and a BS that perform a random access procedure in accordance with an embodiment of the present invention.

Referring to FIG. 33, the UE 3300 includes a UE reception unit 3305, a random access processing unit 3310, and a UE transmission unit 3315.

The UE reception unit 3305 receives a random access start indicator, ordering the start of a new RA procedure intended for a second serving cell, from the BS 3350.

The random access processing unit 3310 determines whether a midway termination condition is satisfied or not while a current RA procedure intended for a first serving cell is in progress. The midway termination condition includes 1) the case where the UE receives a random access start message, ordering that a new RA procedure intended for a second serving cell be started, from the BS or 2) the case where a contention-based random access procedure is scheduled to be spontaneously started in a second serving cell in the MAC layer of the UE. If any one of the conditions 1) and 2) is satisfied, the random access processing unit 3310 can perform midway termination.

If an nth attempted current RA procedure has not been finished although the midway termination condition is satisfied, the random access processing unit 3310 waits for a result of the nth attempted current RA procedure while continuing to perform a current RA procedure, without terminating the current RA procedure immediately.

The random access processing unit 3310 determines whether the current RA procedure results in a short-term failure or not. If, as a result of the determination, it is determined that the current RA procedure does not result in a short-term failure and is successfully completed, the random access processing unit 3310 starts the new RA procedure in the second serving cell.

In contrast, if, as a result of the determination, it is determined that the current RA procedure results in a short-term failure, the random access processing unit 3310 increases a preamble transmission counter by 1. Although the preamble transmission counter is not equal to a maximum retransmission number+1, the UE 3300 no longer reattempts the current RA procedure and declares midway termination. Furthermore, the random access processing unit 3310 starts the new RA procedure in the second serving cell. When the midway termination is declared, the current RA procedure is considered as being a final failure although a reattempt number for the current RA procedure remains until the final number of times of reattempts. If the midway termination condition is not satisfied, the random access processing unit 3310 may perform the current RA procedure until the current RA procedure is successful within a maximum reattempt number.

In a random access procedure based on priority, the random access processing unit 3310 determines whether the new RA procedure has higher priority than the current RA procedure.

For example, priority is determined depending on whether a corresponding random access procedure is based on contention or non-contention. If the new RA procedure > the current RA procedure is sought to be satisfied, the new RA procedure must be based on non-contention and the current RA procedure must be based on contention. Here, it is a precondition that both the first serving cell and the second serving cell are secondary serving cells.

For another example, priority is determined depending on whether a corresponding random access procedure is in progress in a primary serving cell or a secondary serving cell. If the new RA procedure > the current RA procedure is sought to be satisfied, the new RA procedure must be in progress in a primary serving cell and the current RA procedure must be in progress in a secondary serving cell. Furthermore, if the new RA procedure > the current RA procedure is sought to be satisfied, both the new RA procedure and the current RA procedure must be in progress in secondary serving cells.

If the new RA procedure > the current RA procedure is satisfied, the random access processing unit 3310 determines that the current RA procedure results in a final failure and starts the new RA procedure in the second serving cell. In contrast, if the new RA procedure > the current RA procedure is not satisfied, the random access processing unit 3310 cannot start the new RA procedure until the current RA procedure is terminated because the current RA procedure has higher priority than the new RA procedure. The current RA procedure is terminated when it is determined to be successful or to be as a final failure. Accordingly, the random access processing unit 3310 starts the new RA procedure after the current RA procedure is terminated.

The UE transmission unit 3315 transmits a preamble to the BS 3350.

The BS 3350 includes a BS transmission unit 3355, a message processing unit 3360, and a BS reception unit 3365. The BS transmission unit 3355 transmits a random access start indicator, ordering the start of a new random access procedure in a second serving cell, to the UE 3300. The message processing unit 3360 generates the random access start indicator and transfers the generated random access start indicator to the BS transmission unit 3355. The BS reception unit 3365 receives a preamble from the UE 3300 or receives no response.

A variety of exemplary logic blocks, modules, and circuits described in connection with the disclosed embodiments may be controlled by general-purpose processors, Digital Signal Processors (DSPs), Application-Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, or a combination of them designed to perform the above-described functions. The control steps of the methods and algorithms described in connection with the disclosed embodiments may be directly implemented by hardware, software modules executed by processors, or a combination of them. In one or more exemplary embodiments, the above-described control functions may be implemented by hardware, software, firmware, or a combination of them. In software implementations, corresponding functions may be stored in a computer-readable medium or transmitted in the form of one or more instructions or codes.

Claims

A method of a mobile station performing a random access procedure in a multiple component carrier system, the method comprising:

starting a first random access procedure intended for a first serving cell;

increasing a preamble transmission counter indicating a retransmission number when a random access preamble is retransmitted in the first random access procedure;

receiving a random access start indicator that orders a start of a second random access procedure intended for a second serving cell different from the first serving cell from a base station; and

holding the start of the second random access procedure if it is checked that the first random access procedure is in progress;

starting the second random access procedure when the first random access procedure is terminated.
The method of claim 1, further comprising determining priority between the first random access procedure and the second random access procedure, wherein when the first random access procedure has higher priority than the second random access procedure, the start of the second random access procedure is held.
The method of claim 2, wherein the priority comprises at least one of priority between a contention-based random access procedure and a non-contention-based random access procedure, priority between the first serving cell and the second serving cell and priority between a random access procedure started by the mobile station and a random access procedure started by the base station.
The method of claim 1, wherein the second random access procedure is:

i) held until a random access holding timer expires, and

ii) started when the first random access procedure is unsuccessfully terminated and although the preamble transmission counter is smaller than a maximum retransmission number.
The method of claim 1, wherein:

the random access start indicator is a sub-header within a Medium Access Control (MAC) Protocol Data Unit (PDU), and

the sub-header comprises an index field indicative of the second serving cell and a Logical Channel ID (LCID) field indicative of the start of the second random access procedure in the second serving cell.
A mobile station configured to perform a random access procedure in a multiple component carrier system, the mobile station comprising:

a random access processing unit configured to start a first random access procedure intended for a first serving cell and increase a preamble transmission counter indicating a retransmission number when a random access preamble is retransmitted in the first random access procedure;

a reception unit configured to receive a random access start indicator that orders a start of a second random access procedure intended for a second serving cell different from the first serving cell from a base station; and

a transmission unit configured to transmit the random access preamble to the base station in response to the random access start indicator,

wherein the random access processing unit holds the start of the second random access procedure if it is checked that the first random access procedure is in progress and starts the second random access procedure when the first random access procedure is terminated.
The mobile station of claim 6, wherein the random access processing unit is configured to:

determine priority between the first random access procedure and the second random access procedure, and

holds the start of the second random access procedure when the first random access procedure has higher priority than the second random access procedure.
The mobile station of claim 7, wherein the random access processing unit determines the priority between the first random access procedure and the second random access procedure based on at least one of priority between a contention-based random access procedure and a non-contention-based random access procedure, priority between the first serving cell and the second serving cell and priority between a random access procedure started by the mobile station and a random access procedure started by the base station.
The mobile station of claim 6, wherein the random access processing unit is configured to:

i) hold the second random access procedure until the random access procedure timer expires, and

ii) start the second random access procedure when the first random access procedure is unsuccessfully terminated and although the preamble transmission counter is smaller than a maximum retransmission number.
The mobile station of claim 6, wherein:

the random access start indicator is a sub-header within a Medium Access Control (MAC) Protocol Data Unit (PDU), and

the sub-header comprises an index field indicative of the second serving cell and a Logical Channel ID (LCID) field indicative of the start of the second random access procedure in the second serving cell.
A method of a base station performing a random access procedure in a multiple component carrier system, the method comprising:

transmitting a random access command, indicating a start of a first random access procedure intended for a first serving cell configured in a mobile station, to the mobile station;

determining whether or not to stop the first random access procedure;

generating a random access start indicator indicating the stop of the first random access procedure and a start of a second random access procedure intended for a second serving cell if, as a result of the determination, it is determined to stop the first random access procedure; and

transmitting the random access start indicator to the mobile station.
The method of claim 11, wherein determining whether or not to stop the first random access procedure comprises determining to stop the first random access procedure if some time elapses without receiving a random access preamble from the mobile station in the first random access procedure.
The method of claim 11, wherein:

the random access start indicator is downlink control information mapped to a physical downlink control channel,

the downlink control information comprises a preamble index, and

all bits of the preamble index are set to 0.
The method of claim 11, wherein:

the random access start indicator is a sub-header within a Medium Access Control (MAC) Protocol Data Unit (PDU), and

the sub-header comprises an index field indicative of the second serving cell and a Logical Channel ID (LCID) field indicative of the start of the second random access procedure.
The method of claim 11, wherein any one of the first random access procedure and the second random access procedure is selectively performed based on priority between the first random access procedure and the second random access procedure.
A base station configured to perform a random access procedure in a multiple component carrier system, the base station comprising:

a transmission unit configured to transmit a random access command, indicating a start of a first random access procedure intended for a first serving cell configured in a mobile station, to the mobile station;

a stop processing unit configured to determine whether or not to stop the first random access procedure and generate a random access start indicator indicating the stop of the first random access procedure and a start of a second random access procedure intended for a second serving cell if, as a result of the determination, it is determined to stop the first random access procedure; and

a reception unit configured to receive a random access preamble from the mobile station,

wherein the transmission unit transmits the random access start indicator to the mobile station.
The base station of claim 16, wherein the stop processing unit determines to stop the first random access procedure if some time elapses without receiving the random access preamble from the mobile station in the first random access procedure.
The base station of claim 16, wherein:

the random access start indicator is downlink control information mapped to a physical downlink control channel,

the downlink control information comprises a preamble index, and

all bits of the preamble index are set to 0.
The base station of claim 16, wherein:

the random access start indicator is a sub-header within a Medium Access Control (MAC) Protocol Data Unit (PDU), and

the sub-header comprises an index field indicative of the second serving cell and a Logical Channel ID (LCID) field indicative of the start of the second random access procedure.
The base station of claim 16, wherein any one of the first random access procedure and the second random access procedure is selectively performed based on priority between the first random access procedure and the second random access procedure.