JP2020065289A - Short latency fast retransmission triggering - Google Patents

Abstract

To provide a method of operating an improved transmission protocol for uplink data packet transmission in a communication system, user equipment, and a base station.SOLUTION: A receiver of user equipment receives a fast retransmission indicator called as FRI. The FRI indicates whether or not a base station requires a retransmission of a previously transmitted data packet. A transmitter of the user equipment re-transmits the data packet using a redundant version that is the same as that already used for the previous transmission of the data packet.SELECTED DRAWING: Figure 6

Description

  The present disclosure relates to a method of operating a transmission protocol for uplink data packet transmission in a communication system. The present disclosure also provides user equipment and base stations involved in the methods described herein.

[Long Term Evolution (LTE)]
Third generation mobile systems (3G) based on WCDMA® radio access technology are widely deployed throughout the world. The first stage of improvement or evolution of this technology is also referred to as High-Speed Downlink Packet Access (HSDPA) and Improved Uplink (High-Speed Uplink Packet Access (HSUPA)). ) Will inevitably be accompanied by the granting of competitive wireless access technology.

  In order to secure an advantage over new radio access technology in preparation for further increase in user demand, 3GPP has introduced a new mobile communication system called Long Term Evolution (LTE). LTE is designed to meet the carrier demands of the next decade for high-capacity voice support as well as high-speed data media transfer. The ability to provide high bit rates is a major measure of LTE.

  Evolutionary UMTS Terrestrial Radio Access (UTRA) and UMTS Terrestrial Radio Access Network (UTRAN) Long Term Evolution (LTE) work item (WI) specifications are the 8th edition. It has been completed as LTE Rel. 8). The LTE system represents an efficient packet-based radio access and radio access network that offers low latency and low cost fully IP-based functionality. In LTE, scalable multiples such as 1.4, 3.0, 5.0, 10.0, 15.0, and 20.0 MHz are available for flexible system deployment using a given spectrum. The transmission bandwidth of is specified. In the downlink, Orthogonal Frequency Division Multiplexing (OFDM) based radios due to their inherent immunity to multipath interference (MPI) due to low symbol rates, the use of cyclic prefix (CP), and their affinity for different transmission bandwidth configurations. Access was adopted. In the uplink, single carrier frequency division multiple access (SC) has been prioritized over the provision of a wide service area over the improvement of the peak data rate in consideration of the transmission power constraint of the user equipment (UE). -FDMA) based radio access was adopted. LTE Rel. In 8/9, many major packet radio access technologies including multi-input multi-output (MIMO) channel transmission technology are adopted to realize highly efficient control signaling structure.

[LTE architecture]
The overall LTE architecture is shown in FIG. E-UTRAN consists of eNodeB and provides E-UTRA user plane (PDCP / RLC / MAC / PHY) and control plane (RRC) protocol termination towards user equipment (UE). The eNodeB (eNB) hosts the physical (PHY), including the functionality of user plane header compression and encryption, media access control (MAC), radio link control (RLC), and packet data control protocol (PDCP) layers. It also provides a radio resource control (RRC) function compatible with the control plane. eNodeB is a radio resource management, admission control, scheduling, uplink quality of service (QoS) enforcement by negotiation, cell information broadcast, user / control plane data encryption / decryption, and downlink / uplink user plane packets. Performs many functions such as header compression / decompression. The eNodeBs are interconnected by an X2 interface.

  The eNodeB is also connected to the EPC (Evolved Packet Core) by the S1 interface, and more specifically, S1-MME to the MME (mobility management entity) and S1-U to the serving gateway (SGW). It is connected. The S1 interface corresponds to the many-to-many relationship between the MME / serving gateway and the eNodeB. The SGW acts as a user plane mobility anchor during inter-eNodeB handover, and also as a mobility anchor between LTE and other 3GPP technologies (terminating the S4 interface, and 2G / 3G system and PDN GW). Route and transfer user data packets while relaying traffic to and from. For idle user equipment, the SGW terminates the downlink data path and triggers paging when downlink data arrives at the user equipment. The SGW manages and stores the context of the user equipment (eg parameters of the IP bearer service) or network internal routing information. It also replicates user traffic in case of lawful interception.

  The MME is the main control node of the LTE access network. Responsible for tracking and paging procedures for user equipment in idle mode, including retransmissions. The MME is involved in the bearer enable / disable process and is responsible for the selection of the SGW of the user equipment during initial connection and during intra-LTE handover with core network (CN) node relocation. It is also responsible for user authentication (due to interaction with HSS). In the MME, non-access stratum (NAS) signaling terminates and is also responsible for the generation of temporary identification information and its distribution to user equipment. It verifies the user equipment's authentication and stays in the service provider's public land mobile network (PLMN) and enforces the user equipment's roaming restrictions. The MME is the termination point in the network for encryption / integrity protection of NAS signaling and handles the primary management of security. The MME also supports lawful interception of signaling. The MME also provides a mobility control plane function between the LTE and 2G / 3G access networks with the S3 interface terminating at the SGSN to the MME. The MME also terminates the S6a interface towards the home HSS and roams the user equipment.

[Component carrier structure in LTE]
The downlink component carrier of the 3GPP LTE system is subdivided into so-called subframes in the time-frequency domain. In 3GPP LTE, as shown in FIG. 2, each subframe is divided into two downlink slots, and the first downlink slot includes a control channel region (PDCCH region) in the first OFDM symbol. Each subframe consists of a given number of OFDM symbols in the time domain (12 or 14 OFDM symbols in 3GPP LTE (Rel. 8)), and each OFDM symbol spans the entire bandwidth of the component carrier. Therefore, each OFDM symbol consists of a number of modulation symbols transmitted on each subcarrier. In LTE, the transmission signal of each slot is described by a resource grid of N ^DL _RB N ^RB _SC subcarriers and N ^DL _symb OFDM symbols. N ^DL _RB is the number of resource blocks in the bandwidth. The quantity N ^DL _RB is determined by the downlink transmission bandwidth set in the cell and satisfies N ^{min, DL} _RB ≦ N ^DL _RB ≦ N ^{max, DL} _RB . Here, N ^{min, DL} _RB = 6 and N ^{max, DL} _RB = 110 are the minimum and maximum downlink bandwidths, respectively, which the current version of the specification corresponds to. N ^RB _SC is the number of subcarriers in one resource block. For a normal cyclic prefix subframe structure, N ^RB _SC = 12 and N ^DL _symb = 7.

  For example, assuming a multicarrier communication system adopting OFDM or the like used in 3GPP Long Term Evolution (LTE), the minimum unit of resources that can be allocated by the scheduler is one “resource block”. As illustrated in FIG. 2, a physical resource block (PRB) is a continuous OFDM symbol (eg, 7 OFDM symbols) in the time domain and a continuous subcarrier (eg, in the case of a component carrier in the frequency domain). 12 subcarriers). Therefore, in 3GPP LTE (Rel. 8), a physical resource block is composed of resource elements corresponding to 1 slot in the time domain and 180 kHz in the frequency domain (for details of the downlink resource grid, for example, http: // www. 3GPP TS 36.211 "Evolved Universal Terrestrial Radio Access (E-UTRA); Physical Channels and Modulation (Release 8)" (current version 2.0.0), available at 3 gpp.org. Reference, which is incorporated herein by reference).

  Since one subframe consists of two slots, there are 14 OFDM symbols in a subframe when a so-called “normal” CP (cyclic prefix) is used, and when a so-called “extended” CP is used. There are 12 OFDM symbols in a subframe. For convenience of terminology, in the following, time-frequency resources equivalent to the same consecutive subcarriers over the entire subframe are referred to as “resource block pairs” or equivalent “RB pairs” or “PRB pairs”.

  The term "component carrier" refers to the combination of multiple resource blocks in the frequency domain. In a future version of LTE, the term "component carrier" will no longer be used. Alternatively, the jargon is changed to “cell”, which represents a combination of downlink resources and optionally uplink resources. Cooperation between the carrier frequency of the downlink resource and the carrier frequency of the uplink resource is specified in the system information transmitted on the downlink resource.

  Similar assumptions about the component carrier structure apply to the successor version.

[Carrier aggregation in LTE-A to support wider bandwidth]
At the World Radiocommunication Conference 2007 (WRC-07), the frequency spectrum of IMT Advance was determined. Although the full frequency spectrum of IMT Advance has been determined, the actual available frequency bandwidth varies by region or country. However, following the determination of the outline of the available frequency spectrum, standardization of the air interface has started in the 3rd Generation Partnership Project (3GPP). At the 3GPP TSG RAN # 39 meeting, the discussion items regarding “Further E-UTRA Progress (LTE Advance)” were approved. This consideration covers the technical elements to be considered for the evolution of E-UTRA, such as the satisfaction of the requirements for IMT Advance.

  The bandwidth that the LTE advance system can support is 100 MHz. On the other hand, the LTE system can only support 20 MHz. Today, the lack of radio spectrum has become a bottleneck in the development of wireless networks, and as a result it is difficult to find a spectrum band wide enough for LTE advanced systems. As a result, it is an urgent task to find a method for acquiring a wider radio spectrum band, and a possible answer is a carrier aggregation function.

  In carrier aggregation, two or more component carriers are aggregated to support a wider transmission bandwidth of up to 100 MHz. The cells of the LTE system are aggregated into one wider channel of the LTE advance system, which is wide enough for 100 MHz even if these cells of LTE are in different frequency bands.

  At least the bandwidth of the component carrier is LTE Rel. If the corresponding bandwidth of 8/9 cells is not exceeded, LTE Rel. All component carriers can be set to be 8/9 compliant. All component carriers aggregated by the user equipment are not necessarily Rel. It does not have to be 8/9 compliant. Rel. Existing mechanisms (eg, barring) may be used to prevent the 8/9 user equipment from staying on the component carrier.

  The user equipment may simultaneously receive or transmit on one or more component carriers (corresponding to multiple serving cells) depending on its capabilities. LTE-A Rel. Having a reception and / or transmission capability of carrier aggregation. Ten user equipments may simultaneously receive and / or transmit on multiple serving cells. On the other hand, LTE Rel. In the 8/9 user equipment, the structure of the component carrier is Rel. Assuming that the 8/9 specification is followed, reception and transmission can be performed only on a single serving cell.

  Carrier aggregation is performed for both continuous and non-continuous component carriers, each limited to 110 resource blocks in the frequency domain (using 3GPP LTE (Rel.8 / 9) numerology).

  3GPP LTE-A (Rel. 10) compliant user equipment to aggregate different numbers of component carriers and different numbers of potentially different bandwidths from the same eNodeB (base station) on the uplink and downlink. Can be set. The number of configurable downlink component carriers depends on the downlink aggregation capability of the UE. On the contrary, the number of configurable uplink component carriers depends on the uplink aggregation capability of the UE. Currently, it may not be possible to configure mobile terminals with more uplink component carriers than downlink component carriers. In a normal TDD arrangement, the number of component carriers in the uplink and downlink and the bandwidth of each component carrier are the same. Component carriers from the same eNodeB do not have to provide the same coverage area.

  The spacing between center frequencies of consecutively aggregated component carriers shall be a multiple of 300 kHz. This is because it conforms to the 100 kHz frequency raster of 3GPP LTE (Rel. 8/9) and at the same time maintains the orthogonality of subcarriers at 15 kHz intervals. Depending on the aggregation scenario, a small number of unused subcarriers may be inserted between consecutive component carriers to facilitate n × 300 kHz spacing.

  The nature of aggregation of multiple carriers appears only up to the MAC layer. One HARQ entity is required for the MAC for each aggregated component carrier for both uplink and downlink. For one component carrier, there is at most one transport block (if there is no SU-MIMO on the uplink). The transport block and its potential HARQ retransmissions need to be mapped on the same component carrier.

  When carrier aggregation is set, there is only one RRC connection between the mobile terminal and the network. In establishing / re-establishing the RRC connection, LTE Rel. Similar to 8/9, one cell provides security input (one ECGI, one PCI, and one ARFCN) and non-access stratum mobility information (eg, TAI). After establishing / reestablishing the RRC connection, the component carrier corresponding to the cell is referred to as a downlink primary cell (PCell). There is always only one downlink PCell (DL PCell) and one uplink PCell (UL PCell) for one connected user equipment. In the set of configured component carriers, the other cells are referred to as secondary cells (SCell), and the carriers of the SCell are the downlink secondary component carrier (DL SCC) and the uplink secondary component carrier (UL SCC). . Up to 5 serving cells including PCell can be configured for one UE.

[MAC layer / entity, RRC layer, physical layer]
The LTE Layer 2 user plane / control plane protocol stack includes four sublayers: RRC, PDCP, RLC and MAC. The media access control (MAC) layer is the lowest sublayer of the layer 2 architecture of the LTE radio protocol stack, and is defined, for example, by 3GPP technical standard TS 36.321 (current version 13.0.0). The connection to the physical layer below it is via the transport channel, and the connection to the RLC layer above it is via the logical channel. Therefore, the MAC layer performs multiplexing and demultiplexing of logical channels and transport channels. The MAC layer on the transmitting side constructs a MAC PDU known as a transport block from the MAC SDU received through the logical channel, and the MAC layer on the receiving side restores the MAC SDU from the MAC PDU received through the transport channel.

  The MAC layer provides RLC layer data transfer services over logical channels (see TS 363.321, Sections 5.4 and 5.3, which are incorporated herein by reference). Are logical control channels carrying control data (eg, RRC signaling) or traffic logical channels carrying user plane data. On the other hand, data from the MAC layer is exchanged with the physical layer through a transport channel classified as downlink or uplink. The data is multiplexed onto the transport channel depending on the method of wireless transmission.

  The physical layer is responsible for the actual wireless transmission of data and control information. That is, the physical layer, on the transmit side, wirelessly carries all information from the MAC transport channel. Significant functions performed by the physical layer include coding and modulation, link adaptation (AMC), power control, cell search (for initial synchronization and handover), and RRC layer (inside LTE system and system). Other measurements (in between) are mentioned. The physical layer executes transmission based on a transmission method such as a modulation scheme, a coding rate (that is, modulation / coding scheme (MCS)), the number of physical resource blocks, and the like. Details regarding the functions of the physical layer can be found in 3GPP technical standard 36.213 (current version 13.0.0), which is incorporated herein by reference.

  The Radio Resource Control (RRC) layer controls the communication between the UE and the eNB on the radio interface and the mobility of the UE moving across multiple cells. The RRC protocol also supports transfer of NAS information. In the case of the RRC_IDLE UE, the RRC also supports notification of an incoming call from the network. RRC connection control includes paging, measurement setup and reporting, radio resource setup, initial security enablement, and establishment of signaling radio bearers (SRB) and radio bearers carrying user data (data radio bearer DRB), etc. It covers all procedures related to the establishment, modification and release of RRC connections.

The radio link control (RLC) sublayer mainly includes an ARQ function and supports data segmentation and concatenation. That is, the RLC layer performs framing of the RLC SDU to a size specified by the MAC layer. The latter two minimize protocol overhead independent of data rate. The RLC layer is connected to the MAC layer via a logical channel. Each logical channel carries different types of traffic. The upper layer of the RLC layer is usually the PDCP layer, but in some cases it is the RRC layer. That is, since RRC messages transmitted on the logical channels BCCH (broadcast control channel), PCCH (paging control channel), and CCCH (common control channel) do not require security protection, they bypass the PDCP layer and are directly linked to the RLC message. Head to the layer. The main services and functions of the RLC sublayer are:
Transfer of upper layer PDUs corresponding to AM, UM or TM data transfer,
Error correction through ARQ,
Division according to the size of TB,
Subdivision as needed (eg, when changing the radio quality or corresponding TB size),
Concatenation of SDUs when radio bearers are the same in FFS,
In-sequence delivery of upper layer PDUs,
Duplicate detection,
Protocol error detection and recovery,
Discard SDU,
reset,
Is mentioned.

  The ARQ functionality provided by the RLC layer is discussed in more detail below.

[LTE uplink access method]
For uplink transmission, power efficient user terminal transmission is required to maximize the service area. Single carrier transmission combined with FDMA with dynamic bandwidth allocation has been chosen as the evolved UTRA uplink transmission scheme. The main reason for prioritizing single carrier transmission is lower peak-to-average power ratio (PAPR) compared to multi-carrier signal (OFDMA), which correspondingly improves the efficiency of the power amplifier and improves the service area. (The data rate is high for a given terminal peak power). At each time interval, the eNodeB ensures in-cell orthogonality by assigning a unique time / frequency resource for transmitting user data to the user. Orthogonal access in the uplink is expected to improve spectral efficiency by eliminating intra-cell interference. Interference due to multipath propagation is handled at the base station (eNodeB) by inserting a cyclic prefix into the transmitted signal.

  The basic physical resource used for data transmission consists of a frequency resource of size BWgrant in one time interval (for example, subframe) to which coded information bits are mapped. It should be noted that the subframe, also called the transmission time interval (TTI), is the minimum time interval for user data transmission. However, by concatenating the subframes, it is possible to allocate the frequency resource BW grant for a time longer than one TTI to the user.

[Layer 1 / Layer 2 control signaling]
L1 / L2 control signaling needs to be sent along with the data on the downlink in order to inform the scheduled users of their respective allocation status, transport format and other data related information (eg HARQ). is there. Control signaling needs to be multiplexed with downlink data in subframes (assuming the user allocation may change from subframe to subframe). It has to be noted here that the user allocation may be performed on a TTI (Transmission Time Interval) basis and the TTI length is a multiple of the subframe. The TTI length may be fixed in the service area for all users, may be different for different users, or may be dynamic for each user. Generally, the L1 / L2 control signaling may be transmitted every TTI. L1 / L2 control signaling is transmitted on the Physical Downlink Control Channel (PDCCH). It should be noted that on the PDCCH, uplink data transmissions and UL grant assignments are also transmitted.

  In the following, detailed L1 / L2 control signaling information conveyed for each uplink allocation of DL allocation will be described.

[Send downlink data]
L1 / L2 control signaling is transmitted on a separate physical channel (PDCCH) in conjunction with downlink packet data transmission. This L1 / L2 control signaling typically contains information about:

  The physical resource on which the data is transmitted (eg the subcarrier or subcarrier block for OFDM, the code for CDMA): this information allows the UE (receiver) to identify the resource on which the data is transmitted can do.

  -Transport format used for transmission: This may be a transport block size (payload size, information bit size) of data, MCS (modulation / coding system) level, spectrum efficiency, coding rate, and the like. With this information (usually in conjunction with resource allocation), the UE (receiver) identifies the information bit size, modulation scheme, and code rate to initiate the demodulation, derate matching, and decoding processes. can do. In some cases, the modulation scheme may be explicitly transmitted.

・ Hybrid ARQ (HARQ) information:
Process Number: Allows the UE to identify the Hybrid ARQ process to which the data is mapped.
Sequence number or new data indicator: Allows the UE to identify whether the transmission is a new packet or a retransmitted packet.
Redundant and / or constellation version: informs the UE of the hybrid ARQ redundancy version (required for derate matching) and / or the modulation constellation version (required for demodulation) used.

  UE identification information (UE ID): L1 / L2 control signaling indicates the target UE. In a typical implementation, this information is used to prevent other UEs from reading this information by masking the CRC of the L1 / L2 control signaling.

[Uplink data transmission]
To enable uplink packet data transmission, the UE is informed of the details of the transmission by transmitting L1 / L2 control signaling on the downlink (PDCCH). This L1 / L2 control signaling typically contains information about:

  The physical resource in which the UE should transmit data (eg subcarrier or subcarrier block in case of OFDM, code in case of CDMA).

  -Transport format that the UE should use for transmission: This may be the transport block size (payload size, information bit size) of data, MCS (modulation / coding system) level, spectrum efficiency, coding rate, etc. . With this information (usually in conjunction with resource allocation), the UE (transmitter) gets the information bit size, modulation scheme, and coding rate to initiate the modulation, rate matching, and coding process. You can In some cases, the modulation scheme may be explicitly transmitted.

・ Hybrid ARQ information:
Process Number: Tells the UE which Hybrid ARQ process should get the data.
Sequence number or new data indicator: Informs the UE to send a new packet or retransmit a packet.
Redundant and / or constellation version: informs the UE which hybrid ARQ redundancy version to use (required for rate matching) and / or modulation signal point constellation version to use (required for modulation).
UE identification information (UE ID): Indicate the UE to which data should be transmitted. In a typical implementation, this information is used to prevent other UEs from reading this information by masking the CRC of the L1 / L2 control signaling.

  There are several different characteristics to the exact method of transmitting the above information. Furthermore, the L1 / L2 control information may include other information, or a part of the information may be omitted. For example:

  -For synchronous HARQ protocol, the HARQ process number may not be needed.

  No redundant and / or constellation version is required if chase combining is used (always the same redundant and / or constellation version) or if a sequence of redundant and / or constellation versions is predefined there is a possibility.

  -Power control information may be separately included in control signaling.

  -MIMO related control information such as precoding may be separately included in the control signaling.

  -For multi-codeword MIMO transmission, multiple codeword transport formats and / or HARQ information may be included.

  In the case of the uplink resource allocation (PUSCH) transmitted on the LTE PDCCH, the HARQ process number is not included in the L1 / L2 control information because the synchronous HARQ protocol is adopted for the LTE uplink. The HARQ process used for uplink transmission is given by the timing. In addition, redundant version (RV) information is encoded with the transport format information. That is, note that the RV information is embedded in the Transport Format (TF) field. Each MCS field of the TF has a size of 5 bits, for example, and corresponds to 32 entries. Three TF / MCS table entries are reserved for specifying RV1, 2, or 3. The other MCS table entries are used to convey the MCS level (TBS) that implicitly specifies RV0. The size of the CRC field of the PDCCH is 16 bits. Further details on the control information of the uplink resource allocation on PUSCH can be found in TS 5.36.2, Section 5.3.3 and TS 36.213, Section 8.6.

  For the downlink assignment (PDSCH) carried on the LTE PDCCH, the redundancy version (RV) is carried separately in the 2-bit field. Furthermore, the modulation order information is coded together with the transport format information. Similar to the uplink case, a 5-bit MCS field is carried on the PDCCH. Three of the entries are reserved to convey the explicit modulation order and no transport format (transport block) information is provided. Modulation order and transport block size information is conveyed for the other 29 entries. Further details regarding control information for uplink resource allocation on the PUSCH can be found in TS 36.212, section 5.3.3 and TS 36.213, section 7.1.7, which are hereby provided. Be used for.

[E-UTRAN measurement-measurement gaps]
The E-UTRAN reports the measurement information and can configure the UE to support, for example, control of UE mobility. Each measurement configuration element is signaled by the RRCConnectionReconfiguration message. For example, the UE may perform measurements because the measurement gap defines the time when no uplink or downlink transmissions are scheduled. The measurement gap is common to all measurements using the gap. Inter-frequency measurements may require the setting of measurement gaps depending on the capabilities of the UE (eg, have dual receivers). The UE identifies E-UTRA cells operating on carrier frequencies other than the serving cell. If the UE does not have more than one receiver, inter-frequency measurements including cell identification are performed in periodic measurement gaps. Two possible gap patterns, each 6 ms in length, can be set by the network. In the gap pattern # 0, a gap occurs every 40 ms, while in the gap pattern # 1, a gap occurs every 80 ms.

  For example, the reference signal received power (RSRP) is measured by the UE for a cell-specific reference signal within the measurement bandwidth over the measurement period.

[ARQ / Hybrid ARQ (HARQ) method]
In LTE, there are two retransmission levels that provide reliability: HARQ in the MAC layer and outer ARQ in the RLC layer. The RLC retransmission mechanism is responsible for providing error free data delivery to higher layers. To achieve this, a (re) transmission protocol operates between the RLC entities of the receiver and the transmitter, for example in response mode. Transmission errors due to noise, unpredictable channel fluctuations, etc. can be handled by the RLC layer, but in most cases they are handled by the MAC layer HARQ retransmission protocol. Therefore, the use of the RLC layer retransmission mechanism may initially seem unnecessary. However, the fact is different and in practice the difference in feedback signaling is a sufficient trigger for both RLC-based and MAC-based retransmission mechanisms. For example, the RLC-ARQ mechanism addresses potential NACK-to-ACK errors that can occur with the MAC HARQ mechanism.

  A common technique for error detection and correction in packet transmission systems over unreliable channels is called Hybrid Automatic Repeat Request (HARQ). Hybrid ARQ is a combination of Forward Error Correction (FEC) and ARQ. If an FEC encoded packet is sent and the receiver cannot decode the packet correctly (the error is typically checked by CRC (Cyclic Redundancy Check)), the receiver requests retransmission of the packet.

[RLC Retransmission Protocol]
The RLC is considered to be operating in Acknowledged Mode (AM) when it is set to request retransmission of lost PDUs. This is similar to the corresponding mechanism used in WCDMA / HSPA.

  As a whole, RLC has three operation modes: a transparent mode (TM), a non-acknowledged mode (UM), and an acknowledged mode (AM). Each RLC entity has been configured by RRC to operate in one of these modes.

  In transparent mode, no protocol overhead is added for RLC SDUs received from higher layers. In special cases, transmission with limited split / reassembly capability is feasible. In the radio bearer setup procedure, it is necessary to negotiate whether to use the division / reassembly. Transparent mode is used for services that are highly susceptible to delay, such as conversation.

  In non-responsive mode, no data delivery is guaranteed because no retransmission protocol is used. The PDU structure includes a sequence number for integrity observation in the upper layer. Based on the RLC sequence number, the receiving UM RLC entity can perform the reordering of the receiving RLC PDUs. Splitting and concatenation is provided by the header fields added to the data. The RLC entity in non-responsive mode is unidirectional as there is no defined association between uplink and downlink. If erroneous data is received, the corresponding PDU is discarded or marked depending on the setting. At the transmitter, RLC SDUs that are not transmitted within the fixed time specified by the timer are discarded and removed from the transmit buffer. The RLC SDU received from the upper layer is divided / concatenated into RLC PDU on the transmission side. On the receiving side, the reassembly is executed correspondingly. The unacknowledged mode is used for services where error-free delivery is less important compared to short delivery times, eg cell broadcast services such as MBMS and Voice over IP (VoIP) for certain RRC signaling procedures. .

  In response mode, the RLC layer supports error correction with Automatic Repeat Request (ARQ) protocol and is typically used for IP-based services such as file transfers that are of primary interest for error-free data delivery. . RLC retransmissions are based on, for example, RLC status reports or ACK / NACKs received from peer RLC receiving entities. Response mode is designed for reliable transmission of packet data by retransmission in the presence of high air interface bit error rates. In case of a PDU error or loss, the transmitter retransmits upon receipt of the RLC status report from the receiver.

  ARQ is used as a retransmission scheme for retransmission of erroneous or lost PDUs. For example, by monitoring the incoming sequence number, the receiving RLC entity can identify the lost PDU. Then, an RLC status report can be generated on the receiving RLC side and fed back to the transmitting RLC entity to request retransmission of lost PDUs or PDUs that were not successfully decoded. Also, the RLC status report can be polled by the transmitter. That is, the RLC transmitter uses the polling function to inform the RLC transmitter of the reception buffer status by obtaining the status report from the RLC receiver. The status report provides positive acknowledgment (ACK) or negative acknowledgment information (NACK) on the RLC data PDU or part thereof until the last RLC data PDU for which HARQ reordering is completed. The RLC receiver triggers a status report if the PDU with polling field is set to "1" or if the RLC data PDU is detected as lost. Section 5.2.3 of TS 36.322 (current version 13.0.0) defines a specific trigger that triggers polling of the RLC status report at the RLC transmitter and is described herein. Be used for. At the transmitter, only PDUs within the transmission window are allowed to transmit, and the transmission window is updated only by the RLC status report. Therefore, if the RLC status report is delayed, the transmission window cannot be advanced and the transmission may stall.

  The receiver sends an RLC status report to the transmitter when triggered.

  As already mentioned above, in addition to data PDU delivery, control PDUs can be conveyed between peer entities.

[MAC HARQ protocol]
The MAC layer includes a HARQ entity responsible for transmitting and receiving HARQ operations. Transmit HARQ operation includes transmission and retransmission of transport blocks and reception and processing of ACK / NACK signaling. Receive HARQ operations include receiving transport blocks, combining received data, and generating ACK / NACK signaling. Multi-process "stop and wait" (SAW) HARQ operation is supported by the use of up to eight HARQ parallel processes to allow continued transmission while the previous transport block is being decoded. Each HARQ process is responsible for a separate SAW operation and manages a separate buffer.

  The feedback provided by the HARQ protocol is a response (ACK: Acknowledgment) or a negative response (NACK: Negative Acknowledgment). ACKs and NACKs are generated depending on whether the transmission can be received correctly (eg, successful decoding). Furthermore, in HARQ operation, different coding from the original transport block may be used in the retransmission so that the UE may employ Incremental redundancy (IR) combining to obtain another coding gain due to the combining gain. The version can be sent by the eNB.

  If an FEC encoded packet is sent and the receiver cannot decode the packet correctly (the error is typically checked by CRC (Cyclic Redundancy Check)), the receiver requests retransmission of the packet. Generally (and throughout this specification) the transmission of additional information is referred to as a "retransmission (of a packet)", which may, but does not necessarily mean, the transmission of the same encoded information. Does not mean. It may also mean the transmission of any information belonging to the packet (eg additional redundancy information), eg by using different redundancy versions.

  In general, HARQ schemes can be classified as synchronous or asynchronous, and the retransmission in each case is adaptive or non-adaptive. Synchronous HARQ means that the retransmission of transport blocks for each HARQ process occurs at a predetermined (periodic) time for the initial transmission. Thus, no explicit signaling is needed to indicate the retransmission schedule or eg the HARQ process number to the receiver, which can be inferred from the transmission timing.

  In contrast, with asynchronous HARQ, retransmissions can occur at any time relative to the initial transmission. This gives the flexibility to schedule retransmissions based on air interface conditions. However, in this case, another explicit signaling is required, for example to indicate the HARQ process to the receiver, to allow correct combining and protocol operation. In 3GPP LTE system, HARQ operation with 8 processes is used.

  In LTE, asynchronous adaptive HARQ is used in the downlink and synchronous HARQ is used in the uplink. In the uplink, the retransmissions may be adaptive or non-adaptive, eg depending on whether new signaling of transmission attributes is provided in the uplink grant.

  In the uplink HARQ protocol operation (ie the response of the uplink data transmission), there are two different options regarding the retransmission scheduling method. The retransmission is performed by “scheduling” by NACK (also referred to as synchronous non-adaptive retransmission) or explicit scheduling by the network by transmission of PDCCH (also referred to as synchronous adaptive retransmission).

  For synchronous non-adaptive retransmissions, this retransmission will use the same parameters as the previous uplink transmission. That is, this retransmission will be carried on the same physical channel resource, each using the same modulation scheme / transport format. However, the redundant version changes or cycles in a predetermined sequence of redundant versions of 0, 2, 3, 1.

  Since the synchronous adaptive retransmissions are explicitly scheduled by the PDCCH, the eNodeB can change certain parameters of the retransmissions. To avoid fragmentation in the uplink, it is also possible to schedule retransmissions, eg on different frequency resources, change the modulation scheme by the eNodeB, or indicate to the user equipment the redundant version to use for retransmissions. It is also possible. Note that for UL HARQ FDD operation, HARQ feedback (ACK / NACK) and PDCCH signaling occur at the same timing. Therefore, the user equipment determines whether a synchronous non-adaptive retransmission is triggered (ie, only NACK is received) or the eNodeB is requesting synchronous adaptive retransmission (ie, PDCCH is also conveyed). You only have to check once.

  The PHICH carries HARQ feedback indicating whether the eNodeB has correctly received the transmission on the PUSCH. The HARQ indicator is set to 0 for positive acknowledgment (ACK) and 1 for negative acknowledgment (NACK). The PHICH carrying the ACK / NACK message for uplink data transmission may be transmitted to the same user terminal at the same time as the physical downlink control channel PDCCH. By such simultaneous transmission, the user terminal can perform new transmission (new UL grant for switching NDI) or retransmission (referred to as adaptive retransmission) (referred to as NDI) regardless of the instruction to the terminal by PDCCH, that is, the content of PHICH. It is possible to determine an instruction to execute a new UL grant that does not switch. If the PDCCH of the terminal is not detected, the content of PHICH determines the UL HARQ behavior of the terminal, which is summarized as follows.

  NACK: The terminal performs non-adaptive retransmissions, ie retransmissions on the same uplink resources that were previously used in the same HARQ process.

  ACK: The terminal does not perform any uplink retransmission and keeps the data in the HARQ buffer during the HARQ process. Separate transmission related to the HARQ process requires explicit scheduling by the subsequent grant by the PDCCH. The terminal is in the “suspended state” until such a grant is received.

This is shown in Table 1 below.

  The uplink HARQ protocol schedule timing in LTE is illustrated in FIG. The eNB sends the first uplink grant 301 on the PDCCH to the UE, and in response the UE sends the first data 302 on the PUSCH to the eNB. The timing between PDCCH uplink grant and PUSCH transmission is currently fixed at 4 ms. After receiving the first data transmission 302 from the UE, the eNB transmits feedback information (ACK / NACK) of the received transmission and, for example, UL grant 303 to the UE (or, if the UL transmission is successful, The eNB may have triggered a new uplink transmission by transmitting the appropriate second uplink ground). The time between a PUSCH transmission and the corresponding PHICH carrying the feedback information is also currently fixed at 4 ms. As a result, the round trip time (RTT) indicating the next (re) transmission opportunity in the uplink HARQ protocol is 8 ms. After this 8 ms, the UE may send a retransmission 304 of the previous data as instructed by the eNB. Against this separate operation, the retransmission 304 of the previously transmitted data packet was not successful again, as the eNodeB instructs the UE to perform another retransmission (eg, transmission of NACK 305 as feedback). Suppose. In response to this, the UE will perform a separate retransmission 306.

  In the upper part of FIG. 3, the numbering of the subframes as well as an exemplary association with the subframes of the HARQ process is shown. As can be seen, each of the eight available HARQ processes is cyclically associated with each subframe. In the exemplary scenario of FIG. 3, assume that initial transmission 302 and its corresponding retransmissions 304 and 306 are processed by the same HARQ process number 5.

  The measurement gap performing the measurement at the UE has higher priority than the HARQ retransmission. Thus, whenever a HARQ retransmission collides with a measurement gap, no HARQ retransmission will occur. On the other hand, whenever a HARQ feedback transmission on PHICH collides with a measurement gap, the UE assumes an ACK as expected HARQ feedback content.

  The downlink control information has a plurality of fields for assisting HARQ operation.

New Data Indicator (NDI): switched each time a transport block transmission is scheduled (ie, also referred to as initial transmission) (“switch” to the value in the previous transmission of this HARQ process). On the other hand, it means that the NDI bit given in the associated HARQ information has been changed / switched).
Redundant version (RV): indicates the RV selected for transmission or retransmission credit.
・ MCS: Modulation / encoding method

  HARQ operation is complex, not fully described / cannot be described herein, and is not necessary for a complete understanding of the invention. The relevant part of the HARQ operation is described, for example, in 3GPP TS 36.321 (current version 13.0.0), section 5.4.2, “HARQ operation”, which is incorporated herein by reference and Some of them are quoted below.

"5.4.2 HARQ operation 5.4.2.1 HARQ entity In the MAC entity, there is one HARQ entity for each serving cell with an uplink configured, and HARQ feedback regarding successful or unsuccessful reception of a previous transmission. , While maintaining many parallel HARQ processes that allow transmissions to occur continuously.
The number of parallel HARQ processes per HARQ entity is specified in Section 8 [2].
If the physical layer is configured for uplink spatial multiplexing [2], then two HARQ processes are associated with a given TTI. Otherwise, a given TTI is associated with one HARQ process.
For a given TTI, the HARQ entity identifies the HARQ process to transmit if the TTI indicates an uplink grant. It also routes the received HARQ feedback (ACK / NACK information), MCS, and resources relayed by the physical layer to the appropriate HARQ process.
If the TTI bundle is set, the parameter TTI_BUNDLE_SIZE gives the number of TTIs in the TTI bundle. The TTI Bundle operation relies on a HARQ entity that invokes the same HARQ process for each transmission that is part of the same bundle. Within the bundle, HARQ retransmissions are non-adaptive and are triggered according to TTI_BUNDLE_SIZE without waiting for feedback from previous transmissions. HARQ feedback for a bundle is only received in that TTI regardless of whether a transmission in the last TTI of the bundle (ie, the TTI corresponding to TTI_BUNDLE_SIZE) occurs (eg, if a measurement gap occurs). Retransmissions of TTI bundles are also TTI bundles. The TTI bundle is incompatible when one or more SCells with uplink configured are configured in the MAC entity.
The TTI bundle is not supported in RN communication where E-UTRAN is combined with RN subframe setting.
The TTI bundle does not apply to the transmission of Msg3 at random access (see Section 5.1.5).
For each TTI, the HARQ entity is
Identify the HARQ processes associated with this TTI, and for each identified HARQ process,
If an uplink grant is indicated for this process and this TTI,
If the receive grant is not addressed to the tentative C-RNTI on the PDCCH and the NDI switch given in the associated HARQ information is done to the previous in-transit value of this HARQ process, Or if an uplink grant is received on the PDCCH for C-RNTI and the HARQ buffer of the identification process is empty, or if an uplink grant is received in the random access response,
If the MAC PDU is present in the Msg3 buffer and the uplink grant is received in the random access response,
Acquire the MAC PDU to be transmitted from the Msg3 buffer.
Otherwise,
Acquire the MAC PDU to be transmitted from the "multiplexing / assembling" entity.
Deliver MAC PDU, uplink grant, and HARQ information to the identified HARQ process.
Instruct the identified HARQ process to trigger a new transmission.
Otherwise,
Deliver the uplink grant and HARQ information (redundant version) to the identified HARQ process.
Instructs the identification HARQ process to generate adaptive retransmissions.
Otherwise, if the HARQ buffer for this HARQ process is not empty,
Instructs the identification HARQ process to generate non-adaptive retransmissions.
When determining whether the NDI has been switched to the value of the previous transmission, the MAC entity shall ignore the NDI received in all uplink grants on the PDCCH for its tentative C-RNTI. "

[NB-IoT / eMTC Uplink HARQ Protocol]
For NB-IoT and eMTC (Rel.13), an asynchronous UL HARQ protocol has been introduced (also discussed for the ongoing Rel.14 work item for the uplink on unlicensed carriers). Unlike the synchronous uplink HARQ protocol used for legacy LTE, retransmissions for NB-IoT or eMTC UEs are adaptive and asynchronous. More specifically, retransmissions do not have to occur at fixed timing relative to previous HARQ transmissions of the same process, giving the flexibility of explicitly scheduling retransmissions. Furthermore, there will be no explicit HARQ feedback channel (PHICH). That is, retransmission / initial transmission is indicated by the PDCCH (NDI is distinguished between initial transmission and retransmission). In essence, the uplink HARQ protocol behavior for NB-IoT or eMTC UE will be very similar to the asynchronous HARQ protocol used for downlink Rel-8 and beyond.

  Note that for NB-IoT there is only one UL HARQ process.

  For details of the asynchronous uplink HARQ protocol introduced in the NB-IoT / eMTC UE, refer to Section 5.4.2 of 3GPP TS 36.321 (V13.1.0) (2016-03). Which is incorporated herein by reference.

[Short Latency Considerations (Study Item)]
Packet data latency is one of the performance metrics that vendors, operators and thus end users regularly measure (via speed test applications). Latency measurements are taken at all stages of the life of a radio access network system, whether to identify new software releases or system components, to deploy the system, or to operate the system commercially.

  The superior latency of the 3GPP RAT over previous generations was one performance metric that guided the design of LTE. LTE is also now recognized by end users as a system that provides faster Internet access and lower data latency than previous generations of mobile wireless technology.

  In the 3GPP community, much effort has been made to improve the data rate from the first release of LTE (Rel.8) to the latest release (Rel.12). Features such as Carrier Aggregation (CA), 8 × 8 MIMO, 256QAM, etc. have increased the technical potential of L1 data rates from 300 Mbps to 4 Gbps. Rel. In 13, 3GPP aims to introduce a higher bit rate by introducing a maximum of 32 component carriers in CA.

  However, almost nothing has been done about the specific improvement targeting system delay. Packet data latency is not only important to the perceived responsiveness of the system, it is also a parameter that indirectly affects throughput. HTTP / TCP is the predominant application and transport layer protocol suite used on the Internet today. According to HTTP Archive (http://httpparse.org/trends.php), the typical size of HTTP-based transactions on the Internet is in the range of tens of Kbytes to 1 Mbyte. In this size range, the TCP slow start period is the majority of the total transmission period of the packet stream. During TCP slow start, performance is limited by latency. Thus, for this type of TCP-based data transaction, the improved latency to improve average throughput can be fairly easily demonstrated. Moreover, in order to actually realize a high bit rate (range of Gbps in Rel. 13 CA), it is necessary to size the UE buffer accordingly. The longer the RTT, the larger the buffer required. The only way to reduce the buffering requirements on the UE and eNB side is to reduce latency.

  The reduced latency can also positively impact the efficiency of radio resources. A low packet data latency increases the number of possible transmission attempts within a certain delay range, so it is also possible to use a higher BLER target for data transmission, which eliminates the limitation of radio resources, but also inadequate radio conditions. Continue to maintain the same level of robustness for users. Increasing the number of possible transmissions within a certain delay range also allows for more robust transmission of real-time data streams (eg VoLTE) if the same BLER target is maintained. This will increase the capacity of the VoLTE voice system.

  In many existing applications, there are many others that benefit from low latency with respect to the perceived quality of experience. Examples include games, real-time applications such as VoLTE / OTT VoIP, and video telephony / conference.

  In the future, there will be many new applications where latency will become increasingly important. One example is remote control / driving of vehicles, for example augmented reality applications in smart glasses, or certain mechanical communications requiring low latency and critical communications.

  Various pre-scheduling methods can be used to reduce the latency to some extent, but Rel. Not all aspects of efficiency are addressed, as well as the low scheduling request (SR) interval introduced in 9.

  It should also be noted that the reduced user plane data latency may also indirectly result in shorter call setup / bearer setup times due to faster control signaling transmissions.

  Therefore, it is considered necessary to study and improve the packet data latency to ensure the evolution and competitiveness of LTE.

The purpose of this consideration is to consider strengthening the E-UTRAN radio system,
-Significantly reduces the packet data latency on the LTE-Uu air interface of a valid UE,
Significantly reduce packet data transport round trip latency for UEs that have been disabled for a long time (in connected state),
That is.

  The scope of consideration includes resource efficiencies such as air interface capacity, battery life, control channel resources, specification impact, and technical feasibility. Both FDD and TDD duplex modes are possible.

  As a first aspect, the potential benefits of reduced latency and improved TCP throughput due to improved latency in typical applications and use cases are identified and documented. In conclusion, this aspect of the study shows that it is desirable to reduce latency.

  As a second aspect, the following areas shall be considered and documented.

High Speed Uplink Access Solution For valid UEs and UEs that have been disabled for a long time but still have an RRC connection, the current standard is both for maintaining and not maintaining the current TTI length and processing time. It shall focus on obtaining a resource efficient solution with enhanced protocol and signaling while reducing the user plane latency of scheduled UL transmissions by comparison with possible pre-scheduling solutions.
Shortening of TTI and shortening of processing time. Evaluating the influence of specification of TTI length between 0.5 ms and OFDM symbol, study feasibility, and performance in consideration of the influence on reference signal and physical layer control signaling. To do.
Backward compatibility shall be maintained (this allows normal operation of UEs before Rel. 13 on the same carrier).

[Processing chain function for uplink]
For the processing chain shown in FIG. 4, reference is made to Section 5.2.2 of 3GPP TS 36.212 (V13.1.0) (2016-03), which is incorporated herein by reference.

  FIG. 4 is a block diagram including a coding chain function in a physical layer of a single codeword / transport block. The input consists of transport blocks inherited by the MAC layer. Due to the retransmission of transport blocks, the redundancy version (RV) is an input parameter in the "rate matching block". As a result, if different RVs are used for retransmissions, at least the blocks "rate matching", "code block concatenation", "data and control multiplexing", and "channel interleaver" need to be processed.

  The output of the block “Channel Interleaver” serves as the “codeword” input to the physical channel processing step shown in FIG. 5, for which 3GPP TS 36.211 (V13.1.0) (2016-03). ), Section 5.3, which is incorporated herein by reference.

  FIG. 5 is a block diagram including a physical channel processing function in the physical layer. The input consists of the resulting codewords of the coding chain shown in [36.212] Figure 5.2.2-1. For normal (non-MTC or NB-IoT) processing, the "scramble" block has in its input parameters the transmitted subframe index in the radio frame. Therefore, even if the codeword input is the same, the output of the "scramble" block will be different if the subframe index is different. Due to the retransmission of transport blocks, the redundancy version (RV) is an input parameter in the "rate matching block". As a result, if different RVs are used for retransmissions, at least the blocks "rate matching", "code block concatenation", "data and control multiplexing", and "channel interleaver" need to be processed.

  Non-limiting and exemplary embodiments provide improved transmission protocol operation for user equipment uplink data packet transmission.

  The independent claims provide non-limiting and exemplary embodiments. Advantageous embodiments are covered by the dependent claims.

  In accordance with aspects described herein, transmission protocol operation should be improved.

  According to one general aspect, a user equipment operating a transmission protocol for uplink data packet transmission in a communication system, the user equipment comprising a receiver for receiving a fast retransmission indicator is described. Thus, the fast retransmission indicator indicates whether the base station is requesting the retransmission of previously transmitted data packets. The user equipment comprises a transmitter that retransmits the data packet with the same redundant version already used for the previous transmission of the data packet.

  According to another general aspect, there is described a base station operating a transmission protocol for uplink data packet transmission in a communication system, the base station comprising a transmitter for transmitting a fast retransmission indicator. Thus, the fast retransmission indicator indicates to the user equipment whether or not the base station is requesting the retransmission of previously transmitted data packets. The base station comprises a receiver that receives from the user equipment the same redundant version of the retransmitted data packet that was already used by the user equipment for the previous transmission of the data packet.

  Correspondingly, in another general aspect, the techniques disclosed herein feature a method of operating at a user equipment a transmission protocol for uplink data packet transmission in a communication system. The method is the step of receiving a fast retransmit indicator, referred to as a FRI, wherein the FRI receives a fast retransmit indicator that indicates whether the base station requires retransmission of previously transmitted data packets. Including the step of performing. The method further comprises the step of retransmitting the data packet with the same redundant version already used for the previous transmission of the data packet.

  Correspondingly, in another general aspect, the techniques disclosed herein feature a method of operating a transmission protocol of uplink data packet transmission in a communication system at a base station. The method is a step of transmitting a fast retransmit indicator, referred to as a FRI, the FRI transmitting a fast retransmit indicator that indicates to a user equipment whether or not a retransmit of a previously transmitted data packet is requested. Including the step of performing. The method further comprises receiving from the user equipment the same redundant version of the retransmitted data packet that was already used by the user equipment for a previous transmission of the data packet.

  Other benefits and advantages of the disclosed embodiments will be apparent from the specification and drawings. These benefits and / or advantages may be provided individually, but not necessarily all, by one or more of the various embodiments and features of the disclosed specification and drawings. You may be able to obtain it.

  These general and specific aspects may be implemented using systems, methods, and computer programs, and any combination of systems, methods, and computer programs.

  Hereinafter, exemplary embodiments will be described in more detail with reference to the accompanying drawings.

FIG. 3 is a diagram illustrating an exemplary architecture of a 3GPP LTE system. It is the figure which showed the example downlink resource grid of the downlink slot of the sub-frame defined for 3GPP LTE (Rel.8 / 9). FIG. 6 is a diagram illustrating an exemplary transmission protocol operation between a UE and an eNodeB in the case of uplink transmission and its retransmission. FIG. 6 is a schematic block diagram including a coding chain function in a physical layer of a single codeword / transport block. It is a schematic block diagram including a physical channel processing function in a physical layer. 6 is a timing chart of a transmission request and corresponding transmission according to the present embodiment. 6 is a timing chart of a transmission request and corresponding transmission when the simultaneously requested retransmissions collide according to the present embodiment.

  As can be seen from FIG. 3 and its description in the “Background” section, there is currently a 4 ms delay between the PDCCH / PHICH and the corresponding PUSCH uplink transmission. This delay is mainly due to the PDCCH / PHICH detection as well as the processing required on the UE side including the coding chain and physical channel processing steps outlined above. This 4 ms latency can also be reduced by shortening the TTI, as discussed within the scope of the "short latency" consideration above, but the major time savings are reduced transport block size and faster processing. With potential improvements in hardware / software design. Nevertheless, such savings are still outlined in the above, even in the case of retransmissions, especially when different RVs are used for retransmissions compared to previous transmissions of the same transport block (data packet). As we have done, we are constrained by the need to handle all the functional blocks of the transmit chain.

  Another delay, currently 4 ms long, is the gap between the PUSCH transmission and the next potential trigger by the PDCCH / PHICH for the same HARQ process. This gap requires the eNB to process the PUSCH and attempt to decode it, and if the decoding attempt fails, re-determines proper scheduling and link adaptation procedures to determine the need for uplink transmissions of other users. Occurs because it is necessary to determine a suitable set of physical layer transmission parameters (including MCS, number and location of RBs, RV, transmit power) of retransmission credits that also need to be taken into account. Finally, once these parameters are determined, they need to be carried to the UE by DCI on (E) PDCCH (for adaptive retransmissions) and / or HI on PHICH (for non-adaptive retransmissions). .

  Although it is possible to think of PHICH as a compact way to trigger non-adaptive retransmissions, especially because of scrambling depending on different RV versions and subframes, the UE still takes quite a few steps before it can be transmitted. Need to do.

  The object of the invention is to reduce the delay between the transmission on the PUSCH from the UE and the corresponding retransmission indication by the eNodeB. Another purpose is also to suppress the delay between the retransmission instruction by the eNodeB and the corresponding retransmission on the PUSCH from the UE.

  In order to alleviate one or more of the problems described above, the following exemplary embodiments are contemplated by the inventors.

  Particular implementations of variants of this embodiment will be realized in a wide range of specifications set forth by the 3GPP standard, some of which are also described in the "Background" section, but certain major features. Will be added in the following description regarding the described embodiment. Although the present embodiment can be conveniently used in a mobile communication system such as, for example, the 3GPP LTE-A (Rel. 10/11/12/13) communication system described in the above-mentioned “Background Art” section, Note that it is not limited to use in any particular exemplary communication network.

  It should be understood that the following description does not limit the scope of the present disclosure, but is merely an example of an embodiment for a deeper understanding of the present disclosure. Those skilled in the art will recognize that the general principles of the present disclosure as claimed may be applied to various scenarios in ways not explicitly described herein. For the sake of convenience of explanation, a plurality of assumptions are made, but these shall not limit the scope of the following embodiments.

  Hereinafter, an embodiment for solving the above problem will be described in detail. Different implementations and modifications of the present embodiment will be described in the same manner.

  This embodiment provides a user equipment (UE) that operates a transmission protocol for uplink data packet transmission in a communication system. According to this transmission protocol, a Fast Retransmission Indicator (FRI) is used to trigger a fast retransmit in a short time at the UE when the eNodeB fails to try to decode the PUSCH. When this FRI is adopted, the retransmission request can be transmitted by the eNodeB earlier than when DCI / HI is used.

  Since the UE retransmits the data packet faster than the response to the DCI, the UE only uses the same radio resource like the non-adaptive retransmission trigger by the HI according to a variant of this embodiment. Instead, other identical parameters that were applicable to the latest transmitted data packet triggered by DCI or HI may be used for the retransmission of the data packet.

  Similarly, the present embodiment is a base station that operates a transmission protocol for uplink data packet transmission, and whether transmission of a FRI to a UE requires retransmission of a previously transmitted data packet. , A base station is provided. In response to such a request, the base station receives from the UE the same redundant version of the retransmitted data packet that was already used for the previous transmission of the data packet.

  As a general consideration, if the eNodeB is trying to trigger a fast retransmission of data packets, eg due to an urgent Quality-of-Service requirement, then non-optimal use of radio channel capacity should be used as much as possible. It is more important to sacrifice and retransmit as fast as possible. As a major aspect of achieving such fast retransmission of data packets, the eNodeB needs to perform a full link adaptation evaluation since all parameters have already been determined for the previous transmission of the data packet. There is no.

  As explained in the Background section, even if HI is used for retransmission, the redundant version will change for the retransmitted data block. In this case, the redundant versions cycle in a sequence of predetermined redundant versions, for example 0, 2, 3, 1. As shown in FIG. 4, the redundancy version specifically selected for retransmission is the input value of the "rate matching" block. Thus, for each retransmitted data block using a different redundancy version (RV), at least the blocks "rate matching", "code block concatenation", "data and control multiplexing", and "channel interleaver" are processed again. There is a need. The output of the "channel interleaver" block is input to the entire physical channel processing shown in FIG.

  In order to achieve a significant reduction in the time required for the transmission of the retransmission of the data packet, in one implementation of this embodiment, the UE uses the same redundant version as that already used for the previous transmission of the data packet. Used to retransmit. Since the UE uses the same subset of transmission parameters as the previous DCI triggered transmission to retransmit the data packet, ie the same redundancy version as the previously transmitted data packet, all The processing steps can be omitted.

  That is, even if only using the same RV as the previous DCI (or HI) triggered transmission of the data packet, a new “rate matching” and subsequent block (until the start of scrambling (shown in FIG. 5)). (Shown in FIG. 4) need not be processed. In other words, if the UE is buffering the most recently transmitted codewords, this is sufficient for fast retransmissions, and these buffered codewords are processed by the physical channel procedure as shown in FIG. Supply to.

  FIG. 6 shows a timing chart of a transmission request and corresponding transmission. As can be seen from this figure, the period between the transmission of DCI by the eNodeB and the corresponding transmission of the data packet by the UE is shown as period t0, which is referred to as the "third timing". It may be. In the case of the uplink HARQ protocol adopted from LTE Release 8, the period t0 corresponds to the period of 4 ms as shown in FIG. 3, and shows the conventional case regarding triggering of uplink data transmission by DCI. . As can be seen from FIG. 6, in contrast to FIG. 3, the period t1 (also shown in FIG. 6) between the transmission of the data packet by the UE (PUSCH) and the transmission of the FRI by the eNodeB is less than or equal to t0. Good. On the other hand, in the preferred modification of this embodiment, t1 is shorter than the period t0. Although the period t0 may be less than 4 ms due to further development in the future, it does not limit the scope of the present invention. In the description of the present embodiment, the period t0 is 4 ms unless otherwise specified. I assume.

  Therefore, as another modification of the present embodiment, the period t1 in which the name "first timing" is also possible is a fixed period or a period which can be set quasi-statically by the base station and is less than 4 ms. Is preferred.

  Note that FRIs can generally indicate at least two states. According to "state 1", the FRI is a "positive FRI" and triggers a fast retransmission, while in this case it can also be considered as a negative acknowledgment of the received data packet. According to "state 2" the FRI is a "negative FRI" and does not trigger a fast retransmission. In this case, it may be regarded as an acknowledgment of the received data packet. Therefore, as a functionally equivalent interpretation of a state, a "positive FRI" is equivalent to an FRI carrying a "negative acknowledgment (NACK)", and a "negative FRI" is a FRI carrying a "response (ACK)". Are equivalent. For simplification without limiting the scope of the present embodiment, only the technical terms “affirmative FRI” and “negative FRI” are used in the following description.

  As further derived from FIG. 6, the period t2 defined as the time between the positive FRI sent by the eNodeB and the corresponding PUSCH transmission sent by the UE is the DCI (or HI) and the corresponding PUSCH. It must be shorter than the period t0 which is the period between the transmission and the transmission. The shorter time period t2 compared to the time period t0 is due to the fact that the same subset of transmission parameters as the previous DCI / HI triggered transmission described above is used for the retransmission of the data packet, thus saving the computation time at the UE. It is the result. FIG. 6 shows the use of the same redundant version. For example, by using the redundancy version RV # 0, both DCI-initiated PUSCH transmission and FRI-initiated PUSCH transmission are performed. In other words, RV # 0 is determined by DCI-initiated PUSCH transmission and reused by FRI-initiated PUSCH transmission.

  Therefore, as another modification of the present embodiment, the period t2 in which the name "second timing" is also possible is included in the fixed period or the period that can be set quasi-statically by the base station, or the transmission / reception FRI. It is a variable based on each information. Preferably, the period t2 may be less than 4 ms.

  According to another implementation of this embodiment, a positive FRI indicates that the retransmission is performed with the same additional transmission parameters that were used for the previous transmission of the data packet, but these additional identical The transmission parameters are then used by the UE to retransmit the data packet and to receive the retransmitted data packet at the base station.

  According to another implementation of this embodiment, the additional identical transmission parameter used for the retransmission of the data packet is at least the scramble code of the previously transmitted data packet. As an advantage of separately having the same transmission parameter such as the same scramble code, in addition to skipping from the block “rate matching” to the “channel interleaver” as shown in FIG. There is no need to process the "scramble" block shown in FIG.

  In another variation of this embodiment, additional identical transmission parameters are re-established from the previous DCI initiated PUSCH transmission until the time when precoding information is available, ie after the block “precoding” in FIG. It may be used. For example, from the previous DCI initiated PUSCH transmission, the same additional transmission parameter may reuse the modulation scheme and the layer mapping, so that the same transmission scheme as the previous transmission is used for the retransmission. . That is, the same number of transmission layers and the same antenna ports may be used. Using the same precoding vector as in the previous transmission is most reasonable if the FRI does not indicate the use of a different precoder.

However, if the same transmission parameter is separately reused beyond the block “precoding” shown in FIG. 5 from the previous DCI-initiated PUSCH transmission, only some of the resources are used for retransmission. For example, in a "Resource Element Mapper" block, data will only be mapped to a portion of the resources used for previous transmissions (eg, a portion of the resource block, such as 50% of the resource block). Equivalently, for fast retransmission, only part of the output of the "Resource Element Mapper" block is used as input to the "SC-FDMA Signal Generation" block. Therefore, the UE buffers the output of the "resource element mapper" block of the previous transmission, and when triggering a partial retransmission, only the corresponding part is read from the buffer and used as input to the SC-FDMA signal generation. It is enough to apply these. The portion used for fast retransmission preferably comprises a non-negative integer multiple of a reference time or frequency resource unit, such as a "resource block" or "resource block group" defined in TS 36.213. This has the advantage that unused resources can be optimally allocated to other UEs without wasting resources by some resource blocks or resource block groups. Further, the relevant portion has a PUSCH bandwidth with respect to the resource block M ^PUSCH _RB , but here, M ^PUSCH _RB = 2 ^α2 · 3 ^α3 · 5 ^α5 and α ₂ , α ₃ , α ₅ Is a set of non-negative integers. Therefore, if the designated part is a non-integer number of resource blocks or groups of resource blocks or if the resulting bandwidth M ^PUSCH _RB does not satisfy the condition 2 ^α2 · 3 ^α3 · 5 ^α5 , the UE preferably ^Rounding up to the smallest integer value M ^PUSCH _RB that is larger than the specified part that satisfies the smallest integer number of resource blocks or resource block groups that exceed the part or M ^PUSCH _RB = ^2α2 · ^3α3 · ^5α5 .

  In another modified example of the present embodiment, the same “circular shift parameter” as that used in the generation of the reference signal for the retransmission trust of the data packet may be used as the additional identical transmission parameter. In this regard, see Section 5.5.2 of 3GPP technical standard 36.211. By using the same "cyclic shift parameter" for the generated reference signal, the total processing time for retransmission is further reduced. In another variation, the FRI sent by the eNodeB may further include information on the "cyclic shift parameter", which will be used by the UE to generate a reference signal for retransmission credit of the data packet.

  The latest transmission of a data block may not only consist of UL-SCH data, but may also include uplink control information (UCI) such as ACK / NACK, CSI. As can be seen from FIG. 4, such information is added to the data in the block “Data and Control Multiplexing”. In general, such information is preferably added to the FRI-triggered retransmissions as well as the most recent transmissions of the data block. However, it is not always rational to send the same ACK / NACK or CSI information. This is because the delay between the previous transmission and the triggered retransmission may invalidate the content. Therefore, in another embodiment, the UCI is not included in the retransmission, but these resources are reserved as if the information were present. As a result, the order of the data block bits can be invariant, thus eliminating the need for a separate step of reordering the bits for retransmission.

  Similarly, a part of the resources of the uplink subframe, preferably the end of the subframe, may include a sounding reference symbol (SRS). In such cases, the fast retransmission may include SRS, as in the previous transmission. That is, resources are reserved (eg, mute). As a result, the mapping of PUSCH to resource elements can remain unchanged, so that no separate procedure for reordering REs for retransmission is required.

  Referring to FIG. 6 and FIG. 7, when the data packet is not successfully received, the retransmission of the data packet from the UE is requested using FRI, DCI, or HI, and FRI, DCI, or Note that the eNodeB may flexibly determine which of the HIs to send to the UE. Similarly, the UE flexibly performs corresponding transmission / retransmission of the data packet based on the received FRI, DCI, or HI, as described above, upon receipt of either the FRI, DCI, or HI. You may do it.

  According to another implementation of this embodiment, the FRI indicates that retransmission of some of the previously transmitted data packets is performed, and optionally some of the previously transmitted data packets. Of 50% or 25%. In such a case, the UE retransmits the indicated portion of the previously transmitted data packet. The UE may also adapt the transmission power to the retransmissions such that the total transmission power of the retransmissions of said portion of the previously transmitted data packet is equal to the total transmission power of the previously transmitted data packet. Well, optionally, using 50% of the data packet doubles the transmit power of that portion of the previously transmitted data packet.

  If only some of the frequency resources of the previous transmission are utilized for retransmission, then the total power the UE will transmit for partial retransmission will also be some. However, in order to improve the quality of the partial retransmitted data, its power can be mutually increased up to the ratio of frequency resources. For example, if only 50% of the frequency resources are utilized for partial retransmissions, each RE of partial retransmissions can be increased by a factor of 2 and the full RE and full retransmissions of partial retransmissions can be increased. And the total transmission power of is equal. Such partial retransmission may be performed when a complete retransmission does not have to lead to successful decoding of the transport block or the eNodeB uses only a part of the frequency resources for the retransmission and the remaining part for another. It is particularly attractive if one wants to be able to schedule UEs.

The amount of frequency resources utilized for partial retransmissions can be determined according to the following:
1. Quasi-static configuration: Whenever a positive FRI triggers a fast retransmission, the UE looks up the configuration and applies it accordingly.
2. Specification within FRI: The FRI can carry an indicator that determines the amount of a partial resource. For example, a first FRI value triggers a partial retransmission of 50%, a second FRI value triggers a partial retransmission of 25%, and a third value is a complete retransmission (ie 100%). While the fourth FRI value does not trigger a fast retransmission. Therefore, in this example, there are three positive FRI values and one negative FRI value.

  Combinations of these are also possible. For example, after the eNodeB sets three different partial retransmit values (possibly including 100%), each positive FRI value points to a corresponding quasi-static partial retransmit value (one FRI value). Does not indicate fast retransmission (ie negative FRI value).

  In another implementation of this embodiment, the user equipment may comprise multiple transmit antennas for transmitting data packets. In this case, the receiving FRI triggers the retransmission of the data packet so that the UE retransmits the data packet to the eNodeB using multiple transmit antennas. That is, if the transmission includes two transport blocks (codewords), as in SU-MIMO, then, of course with respect to FIG. 5, an affirmative FRI directs the retransmission of both transport blocks, resulting in excess PHY. It is preferable to reuse the transmit buffer as much as possible without reprocessing. In this case, referring to FIG. 5, if the transmission buffer is reused when both transport blocks are retransmitted, there is no need to process the blocks shown in FIG. That is, since the retransmission of the two transport blocks occurs immediately after each "SC-FDMA signal generation" block, it can be directly transmitted to the eNodeB without additional processing.

  Similarly, on the side of the eNodeB capable of using a plurality of antennas, the retransmitted transport block is received by the eNodeB using the plurality of receiving antennas when transmitting the FRI that triggers the retransmission of the transport block.

  However, the triggering of the retransmission of both transport blocks by the (single) FRI comes at the expense of radio resource efficiency and signal to noise ratio. Therefore, in another implementation of this embodiment, per FRI such that retransmission of one transport block to eNodeB and reception of one transport block at eNodeB is performed using multiple transmit antennas. It will trigger the retransmission of one transport block. Note that in this case more processing is required at the UE until the SC-FDMA signal is available for transmission. That is, referring to FIG. 5, when the FRI triggers only the retransmission of one transport block, it involves the processing of the “layer mapping” block up to the “SC-FDMA signal generation” block.

  The above description relates to the behavior of retransmissions, according to which it is assumed that data of the same transport block is used for transmission and retransmission. That is, it implies that the retransmissions apply to the same HARQ process. However, there may be multiple HARQ processes that can be scheduled simultaneously according to a synchronous or asynchronous protocol.

  In either case, as shown in FIG. 7, fast retransmission will occur at the TTI at time “#t_pusch”. Thus, generally for different HARQ processes, a PUSCH transmission at time “#t_pusch” may be triggered at time “# t_pusch-t2” by a positive FRI, and by DCI (or HI) at time “#”. There is also a possibility of being triggered by "t_push-t0". Therefore, in FIG. 7, different HARQ processes P0 and P1 are shown. In the exemplary case shown in FIG. 7, HARQ process P0 relates to a FRI initiated retransmission at time “#t_pusch”, while HARQ process P1 relates to a DCI initiated retransmission at time “# t_pusch-t0”.

  As further shown in FIG. 7, the retransmission of both HARQ processes P0 and P1 results in a retransmission at time “#t_pusch”. However, in order to avoid transmission collisions, the UE needs to decide what to do at time “#t_push”. The first option is to continue the retransmission of HARQ process P0. That is, according to the FRI trigger received at time "# t_push-t2". The second option is to continue the retransmission of HARQ process P1. That is, the DCI (or HI) received at the time “# t_push-t0” is followed.

  In this regard, a preferred embodiment of the present embodiment relates to a specific behavior of the UE, and makes a request to perform retransmission of a data packet by FRI and a request to simultaneously perform transmission of another data packet by DCI or HI. If received, follow the FRI request (ie, the first option described above) and ignore the DCI or HI request.

  As shown in the “Background” section, when the HI and the DCI collide, the UE follows the DCI and ignores the HI. However, in contrast, DCI (or HI) shall be ignored according to the fast retransmission, as provided by another implementation of this embodiment. This is because the positive FRI was transmitted at a time later than the DCI corresponding to the same subframe. As a result, the eNodeB should assume that it only sends a positive FRI if the UE tries to obey the positive FRI instead of the DCI. If not, it means that the subframe is not triggering retransmission with a positive FRI.

  As described above with respect to FIG. 6, even when the UE avoids transmission collision according to the request by FRI, by using the same redundancy version RV # 0, PUSCH transmission and FRI initiated by DCI can be performed as shown in FIG. Both of the initiated PUSCH transmissions are performed. In other words, RV # 0 is determined by DCI-initiated PUSCH transmission and reused by FRI-initiated PUSCH transmission.

  Further, in another variation of this embodiment, the FRI further includes a HARQ process number indicator to indicate the particular HARQ process used by the transmitter for the previous transmission of the data packet.

In the table below, the UE behavior is shown for several cases regarding the content of the received FRI and DCI / HI.

In the table below, different UE behaviors are shown for several cases regarding the content of the received FRI and DCI / HI.

According to the modified example of the present embodiment with reference to the above description, the FRI may indicate at least one of the following elements.
Whether a fast retransmission was triggered (positive FRI or negative FRI or NACK or ACK)
If fast retransmission is triggered: HARQ process number indicator of triggered retransmissions. If fast retransmission is triggered: partial retransmission parameter indicating the requested part of the data block to be retransmitted. Fast retransmission. Is triggered: an indication about the period t2 until the UE transmits

  According to another implementation of this embodiment, the UE receives the FRI in the radio resource used to receive the HI, the FRI as the DCI (eg DCI format 7), or the common search space. The FRI is received on a preset radio resource, or the FRI is received on a preset radio resource of a user equipment specific search space.

Generally, the FRI can be sent in one of the following ways.
The same RE where the UE expects to discover PHICH (but different subframes if the period t1 is shorter than the time between the PUSCH transmission and the corresponding subframe carrying the HI), ie the subframe / TTI control channel Transmission in REs belonging to REG in the area, or
-Transmission in REs that belong to the common search space of DCI, ie REs where all UEs detect FRI: or-Transmission in DCI where the FRIs of multiple UEs and / or subframes are preferably multiplexed. For example, the DCI can include four FRIs, the first FRI can be applied to UE1, the second FRI can be applied to UE2, and so on. Especially in the case of a TDD system, it is also possible to multiplex or bundle multiple FRIs for one UE into DCI, so for example the first four FRIs are applicable to four PUSCH transmissions of UE1, the following three Others are similar, such that one FRI can be applied to the three PUSCH transmissions of UE2. If the FRI is multiplexed for multiple UEs, the DCI is preferably sent in the common search space. If the FRI is sent for only one UE, the DCI is preferably sent in the UE-specific search space.

  As a variation of this embodiment, instead of including one or more of the above contents in the FRI, one or more of the above may be used to determine the RE to which the FRI is sent. is there. For example, the HARQ process can also determine the RE to which the FRI is sent. The UE then preferably monitors multiple FRI resources and only evaluates the FRI received at maximum power.

  As described with respect to variants of the above embodiment, in contrast to the HI triggered retransmission described in the “Background” section, a positive FRI is an implicit or explicit change in RV for a retransmission. Does not imply. However, as another modification of the present embodiment, the FRI-triggered retransmission shall not affect the potential RV decision rule of non-adaptive retransmission by PHICH. As described above in the "Background" section, the PHICH-triggered retransmissions switch implicitly and cyclically between RV {0,2,3,1}. According to this variant, the positive FRI shall be ignored for the purpose of the RV decision of the later non-adaptive retransmission. That is, the RV switching / circulation shall take into account only the RV of the previous DCI / HI triggered (re) transmission.

  As another variation of this embodiment, in addition to using FRI as described above, if the PUSCH occupies only a short TTI (discussed in the "short latency" consideration above) rather than the entire 1 ms TTI, Since the port block is smaller than the TTI of 1 ms, the decoding result (OK / failure) in the eNodeB can be used immediately. Therefore, in this case, the FRI can be transmitted earlier than the DCI / HI in the conventional system.

  As another implementation of this embodiment, the previous transmission of the data packet may be an “initial transmission of the data packet” or a “retransmission of the data packet”.

[Hardware and software implementation of the present disclosure]
Other example embodiments relate to implementations of the various embodiments described above through the use of hardware, software, or software in conjunction with hardware. In this connection, user equipment (mobile terminal) is provided. User equipment is configured to perform the methods described herein, with corresponding entities such as receivers, transmitters, processors, etc. involved in these methods as appropriate.

  It is further appreciated that various embodiments may be implemented or performed using computer equipment (processors). The computer equipment or processor may be, for example, a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device. Also, various embodiments may be implemented or embodied by combinations of these devices. In particular, each functional block used in the description of each of the above-described embodiments can be realized by an LSI as an integrated circuit. These may be individually formed as chips, or one chip may be formed so as to include a part or all of the functional blocks. These may be combined with data input / output. Here, the LSI may be referred to as an IC, a system LSI, a super LSI, or an ultra super LSI depending on the degree of integration. However, the technology for mounting the integrated circuit is not limited to the LSI, and may be realized by using an individual circuit or a general-purpose processor. Further, a programmable FPGA (Field Programmable Gate Array) that is programmable after the manufacture of the LSI or a reconfigurable processor that can reconfigure the connection and setting of the circuit cells arranged inside the LSI may be used. .

  Furthermore, the various embodiments may be implemented by software modules that are executed by a processor or directly in hardware. It is also considered possible to combine software modules and hardware implementations. The software modules may be stored in any kind of computer readable storage medium, for example RAM, EPROM, EEPROM, flash memory, registers, hard disks, CD-ROM, DVD, etc. It is further noted that the individual features of the various embodiments, individually or in any combination, may be the subject of another embodiment.

  One of ordinary skill in the art will appreciate that many variations and / or modifications of the present disclosure are possible, as shown in particular embodiments. Therefore, the present embodiment is merely an example in all respects and should not be considered to be limiting.

Claims (14)

A user equipment for operating a transmission protocol for uplink data packet transmission in a communication system, comprising:
A receiver operable to receive a retransmission indicator, the retransmission indicator indicating whether the base station requests retransmission of a previously transmitted data packet, a receiver,
A transmitter operative to retransmit the data packet using the same redundant version already used for the previous transmission of the data packet if the retransmission indicator indicates an affirmative request;
Equipped with
The retransmission indicator indicates that a retransmission of the portion of the previously transmitted data packet is performed,
The transmission unit uses the transmission power in consideration of the total transmission power of the previously transmitted data packets for the transmission power of the retransmission of the part of the previously transmitted data packets, the user machine. The retransmission indicator indicates that the retransmission is performed with the same additional transmission parameters that were used for the previous transmission of the data packet,
The user of claim 1, wherein the transmitter is further operative to retransmit the data packet with the same additional transmission parameters that were used for the previous transmission of the data packet. machine.   The user equipment according to claim 2, wherein the same additional transmission parameter used for retransmission of the data packet is at least a scramble code of the previously transmitted data packet. A portion of the previously transmitted data packets is 50% or 25% of the previously transmitted data packets,
4. Use of 50% of the data packets in the use of the transmit power of the retransmission doubles the transmit power of the part of the previously transmitted data packets. User equipment according to one paragraph.   The receiver further operates to receive the retransmission indicator at a first timing after the transmission of the previously transmitted data packet, the first timing being fixed or the base station. 5. The user equipment according to any one of claims 1 to 4, which is quasi-statically configurable according to. The transmitter is further operative to transmit the retransmission of the data packet at a second timing after receipt of the retransmission indicator by the receiver,
The second timing may be fixed, quasi-statically set by the base station, or variable based on each piece of information included in the retransmission indicator. User equipment described in. Retransmission of data packets can also be triggered by downlink control information called DCI and / or HARQ indicators called HI,
A first time period between the previous transmission of the data packet and the reception of the retransmission indicator and / or a second time period between the reception of the retransmission indicator and the retransmission of the data packet. Is shorter than a third period between the reception of DCI or HI and the corresponding retransmission of the data packet, and at least one of the first period and the second period is shorter than 4 ms. User equipment according to any one of claims 1 to 6.   If the receiving unit receives a request to perform the retransmission of the data packet according to the retransmission indicator and a request to perform another transmission of the data packet according to the DCI or HI at the same time, the transmitting unit determines the retransmission 8. User equipment according to claim 7, further operative to ignore the request by the DCI or HI according to the request by the transmission indicator.   The user equipment according to any one of claims 1 to 8, wherein the retransmission indicator further comprises a HARQ process number indicator indicating a HARQ process used by the transmitter for the previous transmission of the data packet.   User equipment according to any one of the preceding claims, wherein the previous transmission of the data packet is an initial transmission or a retransmission of the data packet.   The receiving unit receives the retransmission indicator in a radio resource used for receiving a HARQ indicator called HI, receives the retransmission indicator as DCI, or in a preset radio resource of a common search space. 11. The operation of any one of claims 1-10, further receiving the retransmission indicator or receiving the retransmission indicator in a preset radio resource of a user equipment specific search space. User equipment. If the user equipment uses multiple transmit antennas for transmitting data packets,
The receiver is further operative to receive the retransmission indicator that triggers a retransmission of the data packet,
The transmitter is further operative to retransmit the data packet to the base station using the plurality of transmit antennas,
Or
The receiver is further operative to receive the retransmission indicator that triggers the retransmission of one of the data packets,
The user according to any one of claims 1 to 11, wherein the transmitter further operates to retransmit the one of the data packets to the base station using the plurality of transmit antennas. machine. A base station for operating a transmission protocol for uplink data packet transmission in a communication system,
A transmitter operable to transmit a retransmission indicator, the retransmission indicator indicating to a user equipment whether retransmission of a previously transmitted data packet is requested, a transmitter.
Receiving a same redundant version of a retransmitted data packet from the user equipment as that already used by the user equipment for the previous transmission of the data packet if the retransmit indicator indicates an affirmative request. A receiver that operates in
Equipped with
The retransmission indicator indicates that a retransmission of the portion of the previously transmitted data packet is performed,
The reception unit, in the user equipment, regarding the transmission power of the retransmission of the part of the previously transmitted data packets, the transmission power in which the total transmission power of the previously transmitted data packets is considered. A base station that receives the retransmitted data packet that was used. A method of operating a transmission protocol of uplink data packet transmission in a communication system in a user equipment, comprising:
Receiving a retransmission indicator, the retransmission indicator indicating whether the base station requests retransmission of a previously transmitted data packet,
Retransmitting the data packet with the same redundant version already used for the previous transmission of the data packet if the retransmission indicator indicates an affirmative request;
Including,
The retransmission indicator indicates that a retransmission of the portion of the previously transmitted data packet is performed,
The step of retransmitting the data packet may include the transmission power of the retransmission power of the part of the previously transmitted data packets considering the total transmission power of the previously transmitted data packets. Using a method.