WO2007104261A1 - A method and system for supporting packet retransmission segmentation cascading - Google Patents

Abstract

A method and system for supporting packet retransmission segmentation cascading. The method includes the following steps: the data unit of an acknowledged mode traffic is numbered to obtain a sequence number, the sequence number will continue to be used during the next processing procedure of the data unit; the packet header part of the data unit transmitted at least includes the sequence number and corresponding segmentation cascading information, the sequence number together with the segmentation cascading information indicates the definite location relationship of each data block in the data unit. The method and system support segmentation cascading once again when retransmitted, not only support numbering of the SDU, but also support numbering of the PDU, and effectively save the spending of the status report.

Description

 Method and system for supporting data packet retransmission and split cascading

Technical field

 The present invention relates to mobile communication technologies, and more particularly to a method and system for supporting data packet retransmission and split cascading. Background of the invention

 In the LTE (Long Term Evolution) system, the PDU split cascading scheme for AM (acknowledgement mode) and its corresponding PDU header design are mainly as follows.

 Method 1. Based on numbering the SDU, reuse this number for ARQ retransmission detection. The data packet format in Method 1 is shown in Figure 1. Method one is specifically as - 1. For each PDU that is down from the upper layer, the Type field indicates whether it is a control packet or a user packet.

 2. According to the air interface and the scheduling indication, the cascading is performed, and the packet header HI indicates the cascading information. The HI includes the SN (that is, the sequence number of the SDU), F indicates whether it is the end of the SDU, and SD is the segmentation depth (that is, the re-segmentation cascade is performed several times) ), SSN is the subsequence of the cascading cascade each time.

 3. The length of the SSN field is determined by the current SD, that is, the current splitting cascade is performed. The SD can be configured by the upper layer to configure the highest number of re-segmentation cascades, and LI indicates the end of the SDU.

 4. The status report needs to include the corresponding SN, SD, SSN to indicate which block needs to be retransmitted.

 Although the above method 1 can realize the split cascading at the time of retransmission, the header of the packet is relatively large, and the overhead of the status report is relatively large, which is not conducive to the improvement of the transmission efficiency.

 The packet format of method 2 is shown in Fig. 2A. In Fig. 2A, TSN indicates the first PDU sequence number of each SDU, so the TSN is not continuous. The process of re-segmenting the cascade is shown in Figure 2B. As can be seen from Figure 2B, the re-segmentation cascade can only occur at the SDU boundary. The disadvantages of Method 2 are: The restrictions on re-segmentation are more stringent and can only be done at the SDU boundary. If the TB (Transport Block) transmitted for the first time is part of an SDU, the cascading cannot be re-segmented. At the same time, the second method does not make the most use of the air interface.

The packet format of method three is shown in Figure 3. The third method proposes that the flexibility of downlink scheduling should be very high. Therefore, there should be very few cases where the data of the initial segmentation is too large and cannot be transmitted during retransmission. Therefore, the re-segmentation cascade mechanism can be used only for uplink. For the re-segmentation cascading mechanism only twice, the packet header is marked with a special identifier to be a quadratic split packet, and the retransmission split cascading TSN can extend the TSN at the initial transmission. The defect of the third method is similar to the second method, and the restriction on the re-segmentation cascade is large, which is not conducive to the full utilization of the air interface. The data packet format of Method 4 is shown in Figure 4. Method 4 is basically the same as Method 1, except that Method 4 can only support re-segmentation and does not support re-cascading. The defects caused by this are: There are more headers and the status report that needs to be supported is correspondingly larger. In addition, the method 4 only supports re-segmentation and does not support re-cascading. Therefore, Method 4 cannot fully utilize the air interface. ' Invention content

 The embodiments of the present invention provide a method and system for supporting packet retransmission and split cascading, which reduces the overhead of the packet header and the status report size as much as possible, and simplifies the processing process, and provides the implementation of the MAC entity for the RLC function. A choice.

 A method for supporting a data packet retransmission and split cascading according to an embodiment of the present invention includes: - numbering a data unit of an acknowledge mode service to obtain a sequence number, and the sequence number is extended to a subsequent process of the data unit The data unit is a transmission unit or a unit that needs to be split and cascaded;

 The header part of the transmitted transmission unit includes at least the serial number and the split concatenation information of each data block in the transmission unit, and the split concatenation information includes: a status identifier and a length identifier, where the status identifier indicates that the data block is The existing form of the transmission unit, the length identifier is a length indication field, and the status identifier and the length identifier are combined to indicate the positional relationship of the data block in the transmission unit, so that the peer receiving end can be correctly reorganized.

 The embodiment of the present invention further provides a system for supporting a data packet retransmission and splitting cascade, comprising: a transmitting end and a receiving end, wherein the transmitting end is provided with a determining splitting cascaded information module and a sending module, and the receiving The terminal is provided with a receiving module and a recombining module;

 Determining the split cascading information module: determining, according to the existence form of the transport unit according to the split data block, and the length indication corresponding to the data block, determining the status identifier and the length identifier of each data block, and determining the number and location of the data unit The status identifier and the length identifier are carried in the header portion of the transmission unit;

 • Transmitting module: used to send the transmission unit;

 Receiving module: for receiving a transmission unit transmitted by the sending module;

 The reassembly module is configured to reassemble the data blocks after the cascading according to the packet header information carried by the transmission unit received by the receiving module. '

The embodiment of the present invention provides a split cascading scheme of an AM service for ARQ retransmission and a corresponding PDU and a status report (status PDU), and a support packet retransmission and split cascading according to an embodiment of the present invention is provided. The method shares a sequence number in the process of split cascading, re-segment cascading, reassembly, and high-level encryption; the sequence number can be either a number for the SDU (Service Data Unit) or a target number. The number of the PDU (Packet Data Unit) after the cascading is first divided, that is, the re-segmentation cascade, reorganization, and the like reuse the sequence number. The packet header of the data packet needs to include a sequence number, split cascading information, a packet type indication, and the like. The split concatenation information indicates which part of a certain PDU of a certain PDU is a SDU (or a re-segmented PDU) by using the D field and a special ELI. Thus, for a case where a PDU contains multiple SDU partitions (or re-segmented PDU fragments), it can be uniquely represented by SN, D domain, special corresponding ELI domain, and corresponding DATA.

 Thus, embodiments of the present invention improve the deficiencies of the prior art, which have the following advantages:

 1. A valid header format supports re-segment cascading when retransmitting.

 2. The reuse of ELI saves the header of the header.

 3. Supports both SDU numbering and PDU numbering. BRIEF DESCRIPTION OF THE DRAWINGS

 FIG. 1 is a data packet format in a prior art split cascading method 1.

 2A and 2B are schematic diagrams of a prior art split cascading method 2 and a data packet format thereof.

 FIG. 3 is a data packet format in the prior art split cascading method 3.

 FIG. 4 is a data packet format in the prior art split cascading method 4.

 FIG. 5 is a format of a data packet numbered for an SDU according to an embodiment of the present invention.

 FIG. 6 is a format of an initial data packet for numbering a PDU according to an embodiment of the present invention.

 Figure 7 is a diagram of a retransmission packet format for numbering PDUs according to an embodiment of the present invention. Mode for carrying out the invention

 An embodiment of the present invention provides a method for supporting a data packet retransmission and split cascading, which includes: numbering a data unit of an acknowledge mode service to obtain a sequence number, and the sequence number is extended to subsequent processing on the data unit. The packet header portion of the transmitted data unit includes at least the sequence number and corresponding split concatenation information, the sequence number and the segmentation concatenation information collectively indicating a determined positional relationship of each data block in the data unit. The data unit here may be a transmission of an i|l element such as a PDU, or may be a unit that needs to be split and cascaded, such as an SDU.

 If the SDU is numbered, the split cascading information is required for both retransmission and initial transmission; if it is the initial split cascading Pt) U number, the split level link is needed for re-segmenting the cascading data packets, so it is necessary to indicate each Whether the data packet is an initial transmission PDU or a retransmission PDU.

 The split cascading information indicates which SDU of a certain PDU is by using a special length identifier ELI.

Which part of the PDU (or re-segmented PDU). According to the split cascading feature, an SDU is packaged in four forms. Included in a PDU, mode 1, a complete SDU in a PDU; mode 2, the beginning of the SDU to a certain byte (the beginning of the SDU) in a PDU; mode 3, a byte of the SDU to the end of the SDU ( The SDU end fragment is in a PDU; mode four, a certain byte of the SDU to a certain byte of the SDU (the middle segment of the SDU) is in a PDU.

 In the header, the two-bit status indicator D field is used to indicate the form in which the SDU exists in this PDU (one of the above four forms), and the special ELI length indication field and the D field are combined to reflect which part of the SDU this segment is. If the D field indicates that the fragment is a complete SDU, the ELI field indicates the location of the SDU at the end of the PDU or the length of the SDU; if the D field indicates that the fragment is from the beginning of an SDU to the end of a byte, the ELI The field indicates the total length of the SDU in the PDU, that is, the SDU end byte. If the D field indicates that the fragment is from a certain byte of an SDU to the end of the SDU, the ELI field indicates the SDU' start byte and the SDU end byte number. Or the end of the SDU is at the location of the PDU; if the D field indicates that the segment is an intermediate segment of an SDU, the ELI field indicates which byte the SDU starts with and which byte ends. ,

 The above D domain + ELI domain representation method can be applied to the case of numbering the first divided cascaded PDUs, and the ELI indication field indicates which parts of the original PDU are fragmented when the cascade is again divided.

 For a case where a PDU contains multiple SDU segments (or re-segmented PDU segments), it can be uniquely represented by SN, D domain, special corresponding ELI domain, and corresponding DATA.

 The SDU is buffered in the retransmission buffer (if for each SDU number), or the first concatenated PDUs are concatenated (if the PDUs are concatenated for each initial split). After receiving the status report of the peer entity, the corresponding update is performed. For the SDU cache, it may be confirmed that all the SDUs are correctly received by the peer entity, and then the entire SDU is deleted at one time, and the confirmed part may also be deleted. SDU deleted. The cache for the PDU is similar to this and will not be described again.

 The corresponding status report also needs to include split cascading information (ie, D domain and ELI domain): If the PDUs of the initial split cascading are numbered, the status report of the feedback after the initial transmission may only include the SN; if it is for the SDU, Then, whether the initial transmission or the retransmission needs to carry the SN (SDU sequence number), the D domain, and the ELI domain in the status report, the sender can find the corresponding segment through the information to perform re-segment cascading and retransmission.

 The header information for the foregoing description is only some identifiers necessary for the split cascading in the embodiment of the present invention. Those skilled in the art may understand that the header of the data packet may also include other indication bits, such as whether the indication is a data packet or a control packet. The next byte is data or header, etc., and is not shown here one by one.

 The foregoing PDU header structure, cache content and update mode, and corresponding status report format are described in detail below with reference to the accompanying drawings and specific embodiments.

Embodiment 1 If the SDUs are numbered, the packet format is as shown in Figure 5. In FIG. 5, the SN in the packet header is the SDU sequence number corresponding to the data segment, D indicates the manner in which the SDU exists in the PDU, and the combination of the D domain and the ELI domain specifically indicates which portions of the SDU the data segment contains. '

 D is a two-bit header field, which indicates the four existence modes of the SDU in the PDU. For example, 00 indicates that the data segment is a complete SDU, and ELI indicates that the SDU is at the end of the entire PDU (and LI in R6). Sample 01 indicates that the fragment is from the beginning of the SDU to a byte, ELI indicates the number of bytes; 10 indicates that the fragment is from a certain byte to the end of the SDU, and ELI indicates the number of bytes and the number of bytes at the end of the SDU 11 indicates that the fragment is from a certain byte of the SDU to a certain byte of data, and ELI indicates the two bytes. The following example shows: Considering that the IP packet is up to 1500 bytes, the ELI is 12 bits (or more) at D = 01, and the ELI requires 24 bits (or more) when D = l 1 or 10. For a specific example: If D-01 is the SDU from the beginning to 500bit, then ELI=500; if D-10, that is, SDU from 300bit to the end, SDU is 800bit, then ELI =300 800; if D=l 1, That is, the SDU ranges from 300 bits to 700 bits, and ELI=300 700.

 For this packet format, the corresponding status report format support is required. If D=00, that is, this data fragment is a complete SDU, the status report only needs to contain the SN, otherwise (D is the other three values) status report needs to contain three 3⁄4 values of SN, D, ELI to tell the peer entity Some of the data was received correctly.

 The retransmission buffer of the sending end may be buffered for the SDU, and the SDU is deleted until the received status report indicates that a complete SDU is correctly received by the receiving end; or the SDU divided fragment is cached, that is, the status report indication is received. If a segmented SDU segment is received correctly, the segment can be deleted.

 Embodiment 2

 In this embodiment, the split concatenation, sequence number allocation, and PDU header format are the same as those in the first embodiment, except that the status report is used.

 The status report of this embodiment is represented by the first SDU block in the PDU: if D=00 of the first SDU, the SN of the first SDU in the PDU is used; if the D of the first SDU is not If it is equal to 00, the PDU, D and ELI of the first SDU of this PDU are used to represent this PDU.

 Corresponding to the format of the foregoing status report, in this embodiment, the PDUs buffered in the retransmission buffer are partitioned PDUs, and each time the retransmission is performed to re-segment the PDUs, the cache is re-updated.

 'Example three

Different from the above two embodiments, this embodiment numbers the PDUs, and the format of the first transmission data packet is as shown in FIG. 6. In Figure 6, T is a single bit, Ί ) indicates the initial transmission PDU, T = 1 indicates that the PDU is a retransmission PDU; SN is the PDU sequence number; ELI indicates the end of each SDU 'block, ELI can be in the same as R6 Phase Similar, it will not be detailed here.

 If the PDUs are numbered, the retransmission packet format can be as shown in Figure 7. In FIG. 7, T is a single bit, T=0 represents an initial transmission PDU, T=l indicates that the PDU is a retransmission PDU; SN is a PDU sequence number; D is a double bit, similar to the D field in the first embodiment, specifically :

 D=00 indicates that this fragment of the retransmitted PDU is a first-pass complete PDU, and the ELI is at the end of the corresponding PDU;

D=01 means that the fragment is the beginning of the initial PD'U to a certain byte, ELI is the corresponding byte; D=10 means that the fragment is a byte to the end of the initial transmission PDU, and ELI is the corresponding word. The number of bytes at the end of the section and the initial PDU; D = ll indicates that the fragment is a byte of the initial transmission PDU to a certain byte, and ELI is the corresponding two bytes.

 In this embodiment, the status report also gives different formats according to different T and D domains: If T=0, the status report needs to include the T field and the SN number of the outgoing PDU; if T=l, D=00 The status report needs to include the T domain, the SN and D domain of the PDU. In other cases, the values of the four domains T, D, SN, and ELI need to be included to help the sender find the correct block.

 In this embodiment, the retransmission buffer of the sending end buffers the PDU after the initial transmission; after the retransmission occurs, the buffer may be a PDU, or the PDU may be re-segmented and concatenated, and each update is based on the weight. Depending on the content of the cache.

 Embodiment 4

 In this embodiment, the SN and the packet header are the same as the third embodiment, and the status report is the same as the third embodiment in the initial retransmission, but the retransmission status report only needs to use the first PDU in the PDU. SN, D, ELI indicate this retransmission PDU.

 Correspondingly, the PDU buffered in the retransmission buffer is the PDU that was originally transmitted.

 Thus, embodiments of the present invention improve the deficiencies of the prior art, which have the following advantages -

1. A valid header format supports re-segment cascading when retransmitting.

 2. The overhead of the header is saved by the reuse of the ELI.

 3. Supports both SDU numbering and PDU numbering.

 4. Valid ϊ The cost of the status report.

 The present invention has been disclosed in the above-described embodiments, and is not intended to limit the present invention. Any equivalent changes and modifications made by those skilled in the art without departing from the spirit and scope of the present invention should still be The scope covered by this patent.

Claims

Rights request  A method for supporting a packet retransmission and split cascading, comprising:  The data units of the acknowledge mode service are numbered to obtain a sequence number, and the sequence number is extended to a subsequent process performed on the data unit; the data unit is a transmission unit or a unit that needs to be split and cascaded;  The header part of the transmitted transmission unit includes at least the serial number and the split concatenation information of each data block in the transmission unit, and the split concatenation information includes: a status identifier and a length identifier, where the status identifier indicates that the data block is The existing form of the transmission unit, the length identifier is a length indication field, and the status identifier and the length identifier are combined to indicate a positional relationship of the data block in the transmission unit, so that the peer receiving end can be correctly reorganized.  2. The method according to claim 1, wherein the subsequent processing of the data unit comprises one or more of the following: a new split cascading, reassembly, and a high layer encryption process.  3. The method according to claim 1, wherein the method is to number the service data unit to obtain the serial number; or the method is to number the protocol data unit to obtain the serial number. - The method according to claim 1, wherein the status identifier is at least two bits, the transmission unit is a protocol data unit, and the unit to be split and cascaded is a service data unit, and the status identifier and the length identifier are The combination indicates that the location relationship of the data block in the transmission unit includes: the status identifier indicates that the protocol data unit includes the complete service data unit, and the length identifier indicates that the end of the service data unit is at the location of the protocol data unit or the entire service data. The number of bytes of the unit; the status identifier indicates that the protocol data unit contains the first fragment of the service data unit, and the length identifier indicates the number of bytes of the service data unit included; the status identifier indicates the end fragment of the protocol data unit that contains the service data unit And the length identifier indicates a starting byte number and a trailing byte number of the service data unit or a position of the end of the protocol data unit; the status identifier indicates an intermediate segment of the protocol data unit that includes the service data unit, and the length identifier indicates a corresponding The first and last words of this intermediate piece .  5. The method according to claim 1, wherein the transmission unit is a protocol data unit, the unit to be divided and cascaded is a service data unit, and the header portion of the protocol data unit further includes a T domain, where The T field is the initial or retransmission flag of the protocol data unit.  6. The method of claim 5, wherein the length identification of the initial transmission protocol data unit indicates that the corresponding service data unit block is at the end of the protocol data unit. The method according to claim 1, wherein the transmission unit is a protocol data unit, and the unit that needs to be split and cascaded is a service data unit, and the buffering manner of the sending end of the sending protocol data unit includes: for complete service data. The unit performs retransmission buffering until the complete service data unit is correctly received by the receiving end, and the service data unit is deleted from the cache; Cache the segment that wants to be split by the number of services, and if the segment of the divided service data unit is correctly received by the receiving end, the segment is deleted from the cache;  The protocol data unit is buffered, and when the protocol data unit is correctly received by the receiving end, the protocol data unit is deleted from the cache.  8. The method according to claim 4, wherein the status report defined by the receiving end is: if the protocol data unit includes a complete service data unit, the status report includes a sequence number; otherwise, the status report includes a sequence number, a status identifier And length identification; or  If the first service data unit is divided into a complete service data unit, the corresponding status report only needs to be represented by the sequence number of the first service data unit in the protocol data unit; if the first service data unit is not blocked For a complete service data unit, the corresponding status report needs to include the sequence number, status identifier and length identifier of the first service data unit block in the protocol data unit to represent the protocol data unit.  The method according to claim 1, wherein the retransmission buffering manner of the transmitting end comprises: retransmitting and buffering the protocol data unit; or performing retransmission buffering on the service data unit; or splitting the protocol data unit data block.  10. The method of claim 5, wherein the status report of the initial feedback includes a sequence number of the T domain and the protocol data unit, and the retransmission uses the following manner to define the status report: 'Status ί indication The retransmitted fragment is a complete protocol data unit, and the status report needs to include the τ field, the sequence number and status identifier of the protocol data unit; otherwise, it includes 1\ sequence number, status identifier, and length identifier; or the status report includes the retransmission protocol. The sequence number, status identifier, and length identifier of the first initiating protocol data unit in the data unit to indicate the retransmission protocol data unit.  A system for supporting a data packet retransmission and cascading cascade, the system comprising: a transmitting end and a receiving end, wherein the transmitting end is provided with a determining split cascading information module and a sending module, and the receiving end It is provided with a receiving module and a recombining module;  Determining the split cascading information module: determining, according to the existence form of the transport unit according to the split data block, and the length indication corresponding to the data block, determining the status identifier and the length identifier of each data block, and determining the number and location of the data unit The status identifier and the length identifier are carried in the header portion of the transmission unit;  ■ Sending module: used to send the transmission unit;  Receiving module: for receiving a transmission unit transmitted by the sending module; The reassembly module is configured to perform grouping on each data block after the cascading according to the packet header information carried by the transmission unit received by the receiving module. The system according to claim 11, wherein the sending end is further provided with a buffering unit, and the buffering unit is configured to perform retransmission buffering for the complete service data unit until the complete service data unit is If the receiving end is correctly received, the service data unit is deleted from the cache; or the fragment divided by the service data unit is buffered, and when the divided service data unit block is correctly received by the receiving end, the fragment is deleted from the cache; or The protocol data unit is buffered, and when the protocol data unit is correctly received by the receiving end, the protocol data unit is deleted from the cache.