WO2018176480A1 - 数据处理方法和设备 - Google Patents

数据处理方法和设备 Download PDF

Info

Publication number
WO2018176480A1
WO2018176480A1 PCT/CN2017/079327 CN2017079327W WO2018176480A1 WO 2018176480 A1 WO2018176480 A1 WO 2018176480A1 CN 2017079327 W CN2017079327 W CN 2017079327W WO 2018176480 A1 WO2018176480 A1 WO 2018176480A1
Authority
WO
WIPO (PCT)
Prior art keywords
pdu
rlc
pdcp
rlc pdu
mac
Prior art date
Application number
PCT/CN2017/079327
Other languages
English (en)
French (fr)
Inventor
李拟珺
杨颜博
魏岳军
李廉
唐欣
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to JP2019553840A priority Critical patent/JP7101702B2/ja
Priority to CN201780087344.2A priority patent/CN110326357B/zh
Priority to BR112019020346A priority patent/BR112019020346A2/pt
Priority to EP17903926.8A priority patent/EP3592101B1/en
Priority to KR1020197030561A priority patent/KR102255051B1/ko
Priority to PCT/CN2017/079327 priority patent/WO2018176480A1/zh
Publication of WO2018176480A1 publication Critical patent/WO2018176480A1/zh
Priority to US16/587,910 priority patent/US11290214B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/1607—Details of the supervisory signal
    • H04L1/1642—Formats specially adapted for sequence numbers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0045—Arrangements at the receiver end
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/003—Arrangements for allocating sub-channels of the transmission path
    • H04L5/0053—Allocation of signaling, i.e. of overhead other than pilot signals
    • H04L5/0055—Physical resource allocation for ACK/NACK
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/04—Error control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/34—Flow control; Congestion control ensuring sequence integrity, e.g. using sequence numbers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
    • H04W28/065—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information using assembly or disassembly of packets
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W80/00—Wireless network protocols or protocol adaptations to wireless operation
    • H04W80/02—Data link layer protocols

Definitions

  • the present invention relates to the field of communications, and in particular, to a data processing method and device.
  • LTE Long term evolution
  • PDCP Packet Data Convergence Protocol
  • RLC Radio Link Protocol
  • MAC Media Intervention Protocol
  • the RLC layer is located between the PDCP layer and the MAC layer. It communicates with the PDCP layer through a Service Access Point (SAP) and communicates with the MAC layer through a logical channel.
  • SAP Service Access Point
  • one PDCP PDU is composed of multiple RLC PDUs. If at least one RLC PDU constituting a PDCP PDU is lost, the MAC layer will retransmit. If the retransmission cannot recover the lost RLC PDU, all the PDCP PDUs are Related RLC PDU fragments are discarded, affecting the user's subjective experience.
  • the embodiment of the invention provides a data processing method and device for obtaining a complete PDCP PDU according to a correctly received RLC PDU of a PDCP PDU, thereby improving a subjective experience of the user.
  • an embodiment of the present invention provides a data processing method, including: acquiring N correctly received radio link control RLC PDUs of a packet data convergence protocol PDCP protocol data unit PDU; and according to N correctly received RLCs
  • the PDU generates a PDCP PDU, which includes M RLC PDUs, N and M being positive integers, N ⁇ M.
  • the correctly received RLC PDU is the RLC PDU obtained from the correctly received MAC PDU, and the correctly received MAC PDU is the corresponding MAC PDU with the correct CRC check.
  • the PDCP PDU can be generated by receiving the correct RLC PDU according to the part of the PDCP PDU, and it is not necessary to use all the RLC PDUs to combine and obtain one PDCP PDU, so that the RLC PDU that is not needed to be used is not received or lost, and does not affect. Generation of the PDCP PDU.
  • generating a PDCP PDU according to the N correctly received RLC PDUs includes: acquiring at least one compensated RLC PDU, and generating, according to the N correctly received RLC PDUs and the compensated RLC PDUs. PDCP PDU.
  • the correctly received RLC PDU is part, so that it is also necessary to acquire the compensated RLC PDU to combine with the correctly received RLC PDU to obtain the PDCP PDU.
  • the compensated RLC PDU is acquired to replace the lost RLC PDU, thereby implementing the combined PDCP PDU.
  • the device can also use the partially correctly received RLC PDU and the compensated RLC PDU to obtain the PDCP PDU regardless of whether the RLC PDU is lost or not.
  • the generating the PDCP PDU according to the N target RLC PDUs includes: determining whether the PDCP PDU has a lost RLC PDU; if the PDCP PDU to be combined is lost The RLC PDU acquires at least one compensated RLC PDU, the compensated RLC PDU is used to compensate for the lost RLC PDU; and the PDCP PDU is generated according to the N target RLC PDUs and the compensated RLC PDU. That is, when the device has a lost RLC PDU, the compensated RLC PDU is used as compensation for the lost RLC PDU.
  • the at least one compensation RLC PDU includes: acquiring data domain data, and the data domain data is data domain data of an RLC PDU type, and acquiring an RLC header. That is, the parts of the RLC PDU are separately obtained, thereby generating a compensated RLC PDU according to the data domain data and the RLC header.
  • the RLC PDU is lost due to the erroneously received MAC PDU, so that the voice data is obtained from the erroneously received MAC PDU.
  • the data of the RLC PDU is compensated, that is, the data in the compensated RLC PDU is the voice payload obtained from the erroneously received MAC PDU.
  • the ACK that is incorrectly received is the MAC PDU of the corresponding CRC check error.
  • the RLC PDU included in the erroneously received MAC PDU and the RLC PDU lost in the PDCP PDU may be the same RLC PDU.
  • the compensated RLC PDU can be obtained according to the error-containing data, thereby retaining the original PDCP PDU to be discarded according to the compensated RLC PDU, thereby reducing the loss of voice data and improving the loss.
  • User's voice subjective experience
  • the acquiring data domain data includes: acquiring a target MAC PDU, where the target MAC PDU is a CRC check MAC PDU that is checked incorrectly
  • the RLC PDU included in the target MAC PDU is the RLC PDU that is to be combined with the PDCP PDU to be combined.
  • the voice payload is obtained from the target MAC PDU.
  • the acquiring a voice payload from the target MAC PDU includes: when the lost RLC PDU is located at a non-end position of the PDCP PDU, and the quantity is One time, according to the voice data length of the target RLC PDU and the preset voice frame length, the voice payload length of the lost RLC PDU is calculated; for the target MAC PDU, the voice length of the voice payload length is cut from the back to the front. Lotus.
  • the voice data of the target MAC PDU corresponding to the lost RLC PDU is up to the end of the target MAC PDU, so that the voice data of the lost RLC PDU is obtained. After that, the voice data of the length can be obtained from the end of the target MAC PDU.
  • the acquiring a voice payload from the target MAC PDU includes: when the lost RLC PDU is located at a non-end position of the PDCP PDU to be combined Determining the MAC header length of the target MAC PDU; removing the MAC header length of the target MAC PDU and the preset RLC header length to obtain a voice payload. In this way, the voice payload of the target MAC PDU is removed, and the voice payload is obtained.
  • This method is especially suitable for scenarios in which multiple RLC PDUs are lost, because the voice payload of each lost RLC PDU is not well determined at this time.
  • the acquiring a voice payload from the target MAC PDU includes: determining, when the lost RLC PDU is located at an end position of the PDCP to be combined, End position and start position of the voice payload of the target MAC PDU; the voice payload is intercepted from the target MAC PDU according to the start position and the end position.
  • the voice MAC data of the target MAC PDU can be obtained because the MAC PDU corresponding to the last RLC PDU includes the MAC header and the RLC header, and the terminal further includes padding bits, so that the two types of data are removed one by one.
  • the acquiring the at least one compensation RLC PDU includes: determining, when the PDCP PDU loses the RLC PDU, the receiving times T1 and T2, T1 For the reception moment of the correctly received RLC PDU before the lost RLC PDU in the PDCP PDU, T2 is the reception moment of the correctly received RLC PDU after the lost RLC PDU in the PDCP PDU.
  • the target media access control MAC PDU whose reception time is between T1 and T2 is determined from the MAC PDU receiving the error, and the reception time of the RLC PDU is the reception time of the MAC PDU to which the RLC PDU belongs, and the reception time of the MAC PDU is The time recorded when the MAC PDU is received, that is, the time recorded when the receiving end receives the MAC PDU.
  • the voice payload is obtained from the target MAC PDU to generate a compensated RLC PDU based on the voice payload, and the compensated RLC PDU is used to replace the lost RLC PDU.
  • the correctly received RLC PDU is obtained by the first MAC PDU
  • the first MAC PDU is the MAC PDU of the MAC PDU that is correctly verified by the CRC check
  • the second MAC PDU is stored,
  • the second MAC PDU is a MAC PDU of the MAC PDU that is checked for erroneous CRC check; according to the target correspondence, the receiving times T1 and T2 are determined, and T1 is the correct reception before the lost RLC PDU in the PDCP PDU to be combined.
  • the receiving moment of the RLC PDU, T2 is the receiving moment of the correctly received RLC PDU after the lost RLC PDU in the PDCP PDU to be combined; according to the target correspondence, the receiving time is selected from the second MAC PDU between T1 and T2 A target MAC PDU between; a voice payload is obtained from the target MAC PDU; and a compensated RLC PDU is generated based on the voice payload.
  • the target MAC PDU corresponding to the lost RLL PDU is selected from the obtained multiple received error MAC PDUs, that is, the reception error of the target MAC PDU is the cause of the loss of the RLC PDU.
  • the data in the compensated RLC PDU is an all-zero sequence or a random bit sequence. In this way, data compensation can also be performed on the lost RLC PDU.
  • the data in the N correctly received RLC PDUs includes important bits in the PDCP PDU.
  • the correctly received RLC PDU includes the important bits of the PDCP PDU, so that even if the error compensation is performed on the lost RLC PDU, the overall data of the PDPC PDU is not affected much.
  • the important bit may be a bit whose voice quality parameter is greater than a preset voice quality threshold, for example, a bit belonging to the A substream.
  • the acquiring data domain data includes: calculating, according to a voice data length of the N target RLC PDUs and a preset voice frame length, The voice payload length of the lost RLC PDU; a random bit sequence whose length matches the voice payload length.
  • the data in the compensated RLC PDU can be implemented as a random bit sequence.
  • the RLC header includes the RLCSN and the FI, and the RLC header is obtained, including: according to the front and/or the rear of the lost RLC PDU.
  • the RLCSN of the target RLC PDU determines the RLCSN of the RLC header.
  • the FI of the RLC header is 10
  • the FI of the RLC header is 11. In this way, the acquisition of the RLC header can be achieved.
  • determining whether the PDCP has a lost RLC PDU includes: determining, to be combined, according to the FI of the RLC PDU in the pre-stored RLC group packet range The starting RLC PDU and the terminating RLC PDU of the PDCP PDU, wherein the RLC group packet range includes the target RLC PDU, or the RLC group packet range includes the starting RLC PDU and the target RLC PDU of the next PDCP of the PDCP PDU to be combined.
  • the PDCP PDU to be combined has a lost RLC PDU.
  • the RLC SN can be used to determine whether the RLC PDU of the PDCP PDU is lost.
  • the method of the implementation manner further includes:
  • the MAC PDU corresponding to the non-significant bit may be, for example, a MAC PDU whose voice data belongs to the B substream bit.
  • the maximum number of retransmissions of the MAC PDU can be a value set by the system.
  • the receiving end of the MAC PDU can terminate the retransmission of the MAC PDU corresponding to the non-significant bit, thereby implementing early termination of the MAC PDU, and does not need to wait until the maximum number of retransmissions is terminated.
  • the increase of the high-layer transmission delay is avoided, thereby reducing the high-level packet loss rate, and the early termination of the retransmission is the MAC PDU corresponding to the non-significant bit, and the bit of the type is less subjective to the user even if the error is included.
  • a data processing device in the embodiment of the present application, and the data processing device has the function of the data processing device in the foregoing method.
  • This function can be implemented in hardware or in hardware by executing the corresponding software.
  • the hardware or software includes one or more modules corresponding to the functions described above.
  • the data processing device includes:
  • An acquiring unit configured to acquire N correctly received radio link control RLC PDUs of a packet data convergence protocol PDCP protocol data unit PDU;
  • a generating unit configured to generate a PDCP PDU according to the N correctly received RLC PDUs, where the PDCP PDU includes M RLC PDUs, where N and M are positive integers, N ⁇ M.
  • the data processing device includes:
  • the transceiver performs the following actions: acquiring N correctly received radio link control RLC PDUs of the packet data convergence protocol PDCP protocol data unit PDU;
  • the processor performs the following actions: generating a PDCP PDU according to the N correctly received RLC PDUs, the PDCP PDU including M RLC PDUs, N and M being positive integers, N ⁇ M.
  • Yet another aspect of the present application provides a computer readable storage medium having instructions stored therein that, when executed on a computer, cause the computer to perform the methods described in the above aspects.
  • Yet another aspect of the present application provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the methods described in the various aspects above.
  • the device acquires N correctly received RLC PDUs of the PDCP PDU, and then generates PDCP PDUs according to the N correctly received RLC PDUs, where the PDCP PDU includes M RLC PDUs, where N and M are Positive integer, N ⁇ M.
  • the PDCP PDU can be generated by receiving the correct RLC PDU according to the part of the PDCP PDU, and it is not necessary to use all the RLC PDUs to combine and obtain one PDCP PDU, so that the RLC PDU that is not needed to be used is not received or lost, and does not affect. Generation of the PDCP PDU.
  • FIG. 1a is a block diagram of an existing LTE user plane protocol
  • FIG. 1b is a schematic diagram of a packet format of a voice frame in the existing VoLTE
  • FIG. 1c is a schematic diagram of segmentation and serial connection of an existing RLC SDU
  • FIG. 2 is a flowchart of a method for processing a data according to an embodiment of the present invention
  • FIG. 3 is a flowchart of a method for processing a data according to an embodiment of the present invention
  • FIG. 3b is a comparison diagram of a prior art and a data processing method according to an embodiment of the present invention.
  • FIG. 4 is a flowchart of a method for processing a data according to an embodiment of the present invention.
  • FIG. 5 is a flowchart of a method for data processing according to an embodiment of the present invention.
  • FIG. 6a is a schematic diagram of only one B substream slice lost in the embodiment shown in FIG. 5;
  • FIG. 6b is a schematic diagram of the loss and loss of multiple consecutive B substream fragments according to the embodiment shown in FIG. 5; FIG.
  • 6c is a schematic diagram of the embodiment shown in FIG. 5, in which only a plurality of non-contiguous B substream fragments are lost;
  • FIG. 7 is a flowchart of a method for terminating retransmission according to an embodiment of the present invention.
  • FIG. 8a is a schematic diagram of a B substream with an error bit MOS when the AMR-WB is 12.65 k according to an embodiment of the present invention
  • FIG. 8b is a schematic diagram of a B substream with an erroneous bit MOS according to an AMR-WB 23.85k according to an embodiment of the present invention
  • FIG. 8c is a schematic diagram of a MOS-vs-BER curve according to an embodiment of the present invention.
  • FIG. 9a is a gain diagram of a data processing method according to an embodiment of the present invention.
  • FIG. 9b is a schematic diagram showing performance comparison between a data processing method for acquiring data domain data based on an error-accepting MAC PDU and a data processing method for random compensation according to an embodiment of the present invention
  • FIG. 10 is a schematic structural diagram of a data processing device according to an embodiment of the present disclosure.
  • Figure 10b is a partial schematic structural view of the data processing device shown in Figure 10a;
  • FIG. 11 is a schematic structural diagram of hardware of a data processing device according to an embodiment of the present invention.
  • FIG. 1a is a block diagram of an LTE (Long Term Evolution) user plane protocol, which shows a schematic diagram of a related protocol layer involved in a user plane in an LTE transmission process.
  • LTE Long Term Evolution
  • RLC layer the functions of the RLC layer are as follows:
  • RLC in LTE can be summarized as segmentation or resegmentation, concatenation, reordering, and regrouping.
  • the segmentation is only performed at the transmitting end of the UM/AM entity.
  • the RLC (Radio Link Control) PDU Protocol Data Unit
  • the MAC Medium Access Control
  • the RLC SDU Service Data
  • the RLC entity performs a segmentation operation on the RLC SDU, so that the generated RLC PDU can be adapted to be re-segmented and only executed at the transmitting end of the AM entity.
  • the RLC AM entity supports ARQ retransmission.
  • the RLC layer When the retransmitted RLC PDU cannot be adapted to the total RLC PDU size indicated by the MAC layer, the RLC layer performs a segmentation operation on the retransmitted RLC PDU according to the situation, and generates the RLC PDU.
  • the RLC PDU segment can be adapted. Concatenation is only performed at the sender of the UM/AM.
  • the RLC layer When the total RLC PDU size indicated by the MAC layer is greater than the RLC SDU, the RLC layer performs a concatenation operation on the RLC SDU according to the situation, and concatenates one RLC SDU with other RLC SDUs or other RLC SDUs.
  • the generated RLC PDU can be adapted.
  • Segmentation, re-segmentation, and concatenation are performed by the transmitting end (the UE (User Equipment) side in Figure 1a), and the corresponding rearrangement and reassembly functions occur at the receiving end (eNodeB (Evolved Node in Figure 1a) B; Evolved Node B) side).
  • HARQ Hybrid Automatic Repeat reQuest
  • the eNodeB checks the RCLSN number and any repeatedly received RLC SDUs. It is discarded first, and the remaining RLC SDUs are sorted according to the RLCSN number.
  • the RLC SDU After the rearrangement, the RLC SDU recomposes a complete PDCP SDU according to the information of the SN and FI bits in the RLC header, and is sent to the upper layer. However, if at least one RLC PDU fragment is lost, all other RLC PDU fragments in the entire PDCP associated with it are discarded.
  • AMR-WB Adaptive Multi-Rate Wideband
  • GSM Global System for Mobile Communication
  • a new standard for speech coding is currently widely used in VoLTE (Voice over LTE). Therefore, it is the first speech coding system in communication history that can be used for both wired and wireless services.
  • AMR-WB is different from the conventional fixed rate speech coding method. Its code rate is not fixed, but varies with different transmission environments. It is called adaptive.
  • the AMR-WB has a voice bandwidth of 500-7000 Hz and a sampling rate of 16 kHz.
  • AMR-WB is a high quality digital wideband speech coding system.
  • the bits output by the AMR speech coder are reordered according to their subjective importance.
  • the so-called subjective importance refers to the influence of the bit on the speech quality when an error (bit flip or delete) occurs in the transmission. From the perspective of the subjective experience of the person, there may be a significant or weak difference due to the bit position. Therefore, the bit-to-voice quality at different locations is different in subjective experience.
  • the AMR encoder divides the bits generated by the encoding into three parts, called A, B, and C substreams.
  • the bit contained in the A substream is the most sensitive bit of all the coded bits.
  • the voice frame will be seriously damaged, not only in the subjective experience of the voice, but also possibly Causes a problem that cannot be correctly decoded on the receiving side. Therefore, in the protocol (3GPP TS 26.201), the A substream has an additional CRC to verify it.
  • the speech bits in the B and C substreams generate errors, the subjective experience of the speech is gradually decremented according to the order of the bits.
  • the receiving side decoder can The decoding is completed normally.
  • the sensitivity of the A, B, and C substreams to the error bits is:
  • Figure 1b depicts a typical packet format of a voice frame in VoLTE.
  • the RTP Payload in the upper part of Figure 1b represents the voice frame data.
  • the specific structure is shown in the lower part of Figure 1b and is divided into A, B, and C. Three substreams, the specific length value of each substream follows the requirements in 3GPP TS 26.201. Taking AMR-WB as an example, as shown in Table 1, where the length of the C substream is 0, the speech frame only contains the A substream and the B substream.
  • the RLC segmentation and reassembly mechanism determines that when at least one fragment in the SDP of the same PDCP (Packet Data Convergence Protocol) SDU is lost, all RLC fragments related to the PDCP are discarded.
  • PDCP Packet Data Convergence Protocol
  • the entire PDCP PDU is divided into [0,...,n-1] RLC SDUs. If the n-1th RLC SDU is lost during transmission, all n RLC SDUs are discarded according to the RLC layer processing mechanism.
  • n is a positive integer.
  • the n-1th RLC SDU is in the B substream or the C substream portion of the speech frame, and the coded bits in the portion have less influence on the subjective experience of the speech than the A substream.
  • Lose all relevant RLC SDUs eventually lose one PDCP voice packet and retain the error RLC SDU, form a PDCP with the wrong PDCP, and submit it to the PDCP layer.
  • the voice frame has a total length of 476 bits, and the current RLC SDU has an average TBSize of 100 bits. In this scenario, when an error occurs in 100 bits, the entire 476 bits are discarded, and the subjective experience in speech is It is far worse than retaining these 476 bits.
  • the embodiment of the present invention provides a data processing method for compensating for a lost RLC PDU to avoid discarding all RLC PDUs of the corresponding PDCP, thereby reducing frame data loss and improving the subjective experience of the user.
  • FIG. 2 is a flowchart of a method for processing a data according to an embodiment of the present invention, where the method includes:
  • Step 201 Acquire N correctly received RLC PDUs of the PDCP PDU.
  • the data processing device acquires a portion of the correctly received RLC PDU in the PDCP PDU before generating the PDCP PDU.
  • the correctly received RLC PDU is the correct RLC PDU for the CRC (Cyclic Redundancy Check) of the associated MAC PDU, that is, the RLC PDU extracted by the correct MAC PDU based on the CRC check, the MAC The PDU receives the correct MAC PDU for transmission.
  • CRC Cyclic Redundancy Check
  • the data processing device in the embodiment of the present invention may be a device such as a base station or a UE, which is not specifically limited in this embodiment of the present invention.
  • Step 202 Generate a PDCP PDU according to the N correctly received RLC PDUs.
  • the PDCP PDU includes M RLC PDUs, and N and M are positive integers, N ⁇ M.
  • the data processing device obtains a correctly received RLC PDU, it means that the valid data of the PDCP to which the RLC PDU belongs is obtained. Even if the PDCP contains incorrect data, the data processing device may retain the PDCP, or the PDCP does not acquire the complete data. The PDCP may be generated based on the acquired at least one correctly received RLC PDU, or the data processing device may also obtain the PDCP based on the obtained partial RLC PDU in the correctly received RLC PDU.
  • the step 202 of the embodiment of the present invention includes multiple implementation manners, for example, after obtaining the partially correctly received RLC PDU, the PDCP can be generated by using these correctly received RLC PDUs.
  • generating a PDCP PDU according to the N correctly received RLC PDUs includes: acquiring at least one compensated RLC PDU; and then generating a PDCP PDU according to the N correctly received RLC PDUs and the compensated RLC PDUs.
  • the PDCP PDU When the data data compensation device acquires the RLC PDU correctly received by the PDCP PDU, in order to generate The PDCP PDU needs to obtain the number of RLC PDUs that meet the requirements of the PDCP. In addition to the correctly received RLC PDUs, a preset number of compensated RLC PDUs are available. The compensated RLC PDUs and the correctly received RLC PDUs can be included as PDCP PDUs. All RLC PDUs, whereby the PDCP PDUs can be generated based on the N correctly received RLC PDUs and the compensated RLC PDUs.
  • the data domain data of the compensated RLC PDU has multiple sources or types, for example, the data in the compensated RLC PDU is the voice payload obtained from the erroneously received MAC PDU, or the data in the compensated RLC PDU is all zero. Sequence or random bit sequence.
  • the data in the N correctly received RLC PDUs includes important bits in the PDCP PDU, so that the PDCP PDU can be generated as long as the important bits are acquired, and the remaining non-significant bits may have Other ways to fill, such as using a preset bit sequence, or getting from the wrong MAC PDU.
  • the PDCP PDU to be combined in the embodiment of the present invention has a lost RLC PDU, and thus, after obtaining the partially received RLC PDU, the obtained compensated RLC PDU is used to compensate the lost RLC PDU.
  • the execution process of the data processing method of the embodiment of the present invention is as follows.
  • FIG. 3 is a flowchart of a method for processing a data according to an embodiment of the present invention.
  • a data processing method according to an embodiment of the present invention includes:
  • Step 301 Acquire at least one correctly received RLC PDU of the PDCP PDU to be combined.
  • the correctly received RLC PDU is the RLC PDU whose CRC of the MAC PDU is correct, that is, the RLC PDU extracted by the correct MAC PDU based on the CRC (Cyclic Redundancy Check), the MAC The PDU receives the correct MAC PDU for transmission.
  • Step 302 Determine whether the PDCP PDU to be combined has a lost RLC PDU. If the PDCP PDU to be combined has a lost RLC PDU, go to step 303.
  • the data compensation device After the data compensation device acquires at least one correctly received RLC PDU of the PDCP PDU to be combined, the data compensation device determines whether the PDCP PDU to be combined has a lost RLC PDU, and if so, attempts to compensate the lost RLC PDU, and the specific compensation method Reference may be made to the detailed description of steps 303 to 305. If the PDCP PDU to be combined does not lose the RLC PDU, the PDCP is obtained according to the correctly received RLC PDU combination.
  • the data compensation device After obtaining the partial RLC PDU of the PDCP packet, the data compensation device stops acquiring the remaining RLC PDUs, and determines that the PDCP PDU has lost the RLC PDU.
  • the reason for stopping may be that the device actively stops acquiring, or the device is faulty. Stop getting and so on.
  • the number of the lost RLC PDUs may be one or more, which is not specifically limited in this embodiment of the present invention, and the type of the lost RLC PDU is not specifically limited.
  • step 302 specifically determines whether the PDCP PDU to be combined has a missing preset RLC PDU, and the voice quality parameter of the voice data of the preset RLC PDU as the RLC PDU is less than a preset quality threshold.
  • the RLC PDU, or the preset RLC PDU is the RLC PDU whose voice data of the data domain data belongs to the A substream bit. That is, when a specific RLC PDU is lost, the data compensation device compensates the lost RLC PDU, otherwise it does not compensate, so that when the inconveniently compensated RLC PDU is lost, the loss of the lost RLC PDU is stopped.
  • the compensation directly discarding the PDCP PDU packet to be combined, and implementing the selective compensation for the lost RLC PDU of the PDCP PDU, improves the flexibility of the method of the embodiment of the present invention, thereby improving the execution efficiency.
  • Step 303 Acquire data domain data of the RLC PDU type.
  • the data domain data of the RLC PDU is acquired, and the data domain data of the lost RLC PDU is compensated by the data domain data.
  • the random bit sequence is used as the data domain data of the RLC PDU, and as long as the length of the random bit sequence matches the data domain length of the lost RLC PDU.
  • the preset bit sequence may be an all-zero sequence, or a bit sequence that has a beneficial compensation effect on data such as voice or video according to statistical analysis.
  • the method of the embodiment of the present invention caches the MAC PDU that accepts the error by using a preset buffer area. After determining that the RLC PDU is lost, the MAC PDU to which the lost RLC PDU belongs is obtained from the buffer area, and the MAC PDU to which the lost RLC PDU belongs is obtained.
  • the data payload such as the voice payload, can be used even if the data payload is erroneous to form a erroneous PDCP PDU, thereby preserving the PDCP PDU in the device.
  • Step 304 Acquire an RLC header.
  • the RLC header is also acquired to use the RLC header as an RLC header for compensating the RLC PDU.
  • the RLC header includes RLCSN and FI.
  • the RLCSN is an RLC serial number. When a long sequence number of 10 bits is used, the RLCSN takes a value from 0 to 1023.
  • FI is a 2-bit flag for recording the RLC segmentation.
  • the RLCSN of the lost RLC PDU that is, all the RLCSNs of the PDCP PDU to be combined, if the RLCSN does not lose the correctly received RLC PDU.
  • the RLCSN is the RLCSN of the lost RLC PDU.
  • the RLCSN of the correctly received RLC PDU correctly received before and after the lost RLC PDU may determine the RLCSN of the lost RLC PDU, which is especially applicable when there are multiple RLC PDUs that are lost.
  • the FI of the RLC PDU is determined according to the location of the lost RLC PDU in the PDCP PDU. When the lost RLC PDU is in the end position in the PDCP PDU to be combined, it is determined that the FI of the RLC header is 10, otherwise the FI is 11.
  • the RLC header is extracted from the MAC PDU, and the RLC header is directly used as the RLC header of the lost RLC PDU.
  • Step 305 Generate a compensated RLC PDU by using the data domain data and the RLC header;
  • the data compensation device may generate a compensated RLC PDU according to the data domain data and the RLC header, where the compensated RLC PDU includes the data domain data and the RLC header.
  • the compensated RLC PDU is used to compensate for the lost RLC PDU of the PDCP PDU to be combined, that is, to replace the lost RLC PDU. Therefore, the PDCP PDU to be combined includes all RLC PDUs.
  • steps 303 to 305 can obtain the compensated RLC PDU.
  • Step 306 Generate a PDCP PDU according to the correctly received RLC PDU and the compensated RLC PDU.
  • the data compensation device acquires the correct received RLC PDU, and compensates the lost RLC PDU by using the generated compensated RLC PDU by the above compensation method, so that the data compensation device includes all RLCs of the PDCP PDU to be combined. PDUs, whereby the PDCP PDUs can be composed based on these RLC PDUs.
  • FIG. 3b it shows one of the specific implementations of the data processing method of the embodiment of the present invention and a comparison diagram of the prior art.
  • the PDCP PDU is generated, but the embodiment of the present invention compensates the lost RLC PDU according to the correctly received RLC PDU and the compensated RLC PDU.
  • a PDCP PDU can be generated.
  • At least one correctly received RLC PDU of the PDCP PDU to be combined is obtained, where the correctly received RLC PDU is the correct RLC PDU of the CRC of the associated MAC PDU. Then, it is determined whether the PDCP PDU to be combined has a lost RLC PDU, and if the PDCP PDU to be combined has a lost RLC PDU, the data domain data of the RLC PDU is acquired, and the RLC header is obtained. Thus, after the data domain data and the RLC header are used to generate the compensated RLC PDU, the PDCP PDU is generated according to the correctly received RLC PDU and the compensated RLC PDU.
  • the RLC PDU when the RLC PDU is lost in the PDCP PDU, the RLC PDU is generated to compensate the lost RLC PDU, and the PDCP PDU can be combined according to the compensated RLC PDU and the correctly received RLC PDU, thereby avoiding the loss of the RLC.
  • the PDCP PDU of the PDU is discarded, and the PDCP PDU is retained, thereby reducing data loss and improving the integrity of the frame data.
  • the subjective experience of the user can be improved. That is, the RLC layer group packet is allowed to contain a certain BER, and the entire PDCP packet is discarded, thereby retaining more effective voice bits, thereby improving the subjective experience of the voice.
  • the following is a description of the method of the embodiment of the present invention performed on the base station side in the context of the VoLTE and AMR-WB scenarios. It can be understood that the method of the embodiment of the present invention can be used for other scenarios, such as a video transmission scenario, etc., in addition to the scenario that can be used for the AMR-WB.
  • the voice payload described below can also be other types of data.
  • FIG. 4 is a flowchart of a method for data processing according to an embodiment of the present invention.
  • a data processing method according to an embodiment of the present invention includes:
  • Step 401 Acquire a MAC PDU and a CRC for verifying the MAC PDU.
  • the base station acquires a MAC PDU and a CRC that verifies the MAC PDU.
  • the CRC is checked. If the CRC check is correct, the MAC PDU is transmitted correctly, that is, the base station receives the correct MAC PDU; if the CRC check is incorrect, the MAC PDU receives the error.
  • the physical layer of the base station obtains the transport block TB and the CRC through the Harq process, and the transport block carries a MAC PDU, which is used to check the MAC PDU.
  • the MAC PDU is a transport block TB before the turbo code is encoded, and the MAC PDU and the CRC are encoded and modulated, and finally the bearer is transmitted on the physical channel.
  • Step 402 Record the receiving moment of the MAC PDU.
  • the base station When the TB (Transport Block) carrying the MAC PDU is acquired, the base station records the reception time, which is used to record the time when the base station acquires the MAC PDU.
  • the receiving time is the initial transmission receiving time, and the initial transmission receiving time is used to record the time when the base station first acquires the MAC PDU. If the CRC check corresponding to the MAC PDU is incorrect, the MAC PDU is transmitted. When the error occurs, the MAC PDU is retransmitted by the retransmission mechanism of the base station to obtain the correct MAC PDU, and the receiving moment of the retransmitted MAC PDU may not be recorded.
  • the receiving moment of the specific scenario of the embodiment of the present invention is the initial transmission receiving time T. That is, T is the receiving time of the initial transmission data of each Harq process, and the initial transmission receiving time T corresponds to the MAC PDU, or the corresponding TB.
  • Step 403 Acquire a target correspondence between the MAC PDU and the receiving time.
  • the target correspondence between the MAC PDU and the initial transmission time is obtained, and the target correspondence is the correspondence between the MAC PDU and the initial transmission receiving time. Since the MAC PDU includes an RLC PDU, the target correspondence also includes a correspondence between the RLC PDU and the initial transmission time T.
  • Step 404 When the CRC check is correct, the correctly received RLC PDU is obtained by the first MAC PDU.
  • the first MAC PDU is a MAC PDU of the MAC PDU that is correctly verified by the CRC.
  • the correctly received RLC PDU is the correct RLC PDU for the CRC of the associated MAC PDU.
  • the base station After the base station obtains the CRC and the MAC PDU corresponding to each other, if the CRC check is obtained, the corresponding MAC PDU is the correct MAC PDU, that is, the first MAC PDU.
  • the RLC PDU extracted from the correct MAC PDU is transmitted with the correct RLC PDU, and the correct RLC PDU is the correctly received RLC PDU.
  • Step 404 is a specific method for the base station to obtain at least one correctly received RLC PDU of the PDCP PDU to be combined.
  • the base station obtains a correctly received RLC PDU of the PDCP PDU to be combined, and the number of correctly received RLC PDUs may be one. For multiple.
  • the physical layer of the base station transmits the data packet to the MAC layer according to the normal procedure, and simultaneously transmits the corresponding initial transmission receiving time T to the MAC layer.
  • the MAC layer transmits the correct MAC PDU and its corresponding initial transmission time T to the RLC layer according to the normal processing flow.
  • CRC is the cyclic redundancy check value of the transport block, "0" means the transmission is correct, and "1" means the transmission is wrong.
  • Step 405 When the CRC check is incorrect, the second MAC PDU is stored.
  • the second MAC PDU is a MAC PDU of a MAC PDU that is checked for a wrong CRC check. That is, if the CRC check error occurs, the MAC PDU verified by the CRC is transmitted incorrectly, and the MAC PCU that transmits the error is the second MAC PDU.
  • An error occurs in the transmission of a bit, such as a bit flip or deletion, which may cause a transmission error in the MAC PDU.
  • the MAC PDU if the MAC PDU has a transmission error, the MAC PDU is discarded, and the RLC PDU carried by the MAC PDU that is transmitted incorrectly is discarded.
  • the method of the embodiment of the present invention caches the incorrectly transmitted MAC PDU for subsequent use. Since the MAC PDU with the transmission error is buffered, the RLC PDU is not extracted therefrom, so that when the PDCP PDU is combined, the RLC PDU on the MAC PDU is in a lost state.
  • the physical layer of the base station transmits the check result, the MAC PDU packet, and its corresponding initial transmission time T to the MAC layer.
  • the MAC layer passes the incorrectly transmitted MAC PDU and its corresponding initial transmission time T to the ErrorBuffer of the RLC layer.
  • the ErrorBuffer is a preset RLC layer buffer, and stores the MAC PDU with a final transmission error.
  • Step 406 Determine whether the PDCP PDU to be combined has a preset RLC PDU. If the PDCP PDU to be combined has a preset RLC PDU, go to step 407.
  • the preset RLC PDU is an RLC PDU belonging to a non-A substream fragment, that is, the voice data of the data domain data of the RLC PDU belongs to an RLC PDU that is not an A substream bit.
  • an RLC PDU belonging to a B substream slice or a C substream slice is an RLC PDU belonging to a non-A substream fragment.
  • the steps 406 to the step are repeated. 410.
  • the RLC_Buffer is updated, the PDCP PDUs to be combined with the lost RLC PDUs in the RLC_Buffer are all attempted to be compensated for the lost RLC PDUs.
  • the update process of the RLC_Buffer is the process of recovering the lost RLC PDU as much as possible by using the ErrorBuffer.
  • MAC PDUs may arrive out of order.
  • RLC_Buffer has a wait mechanism. If the RLC PDU to be waited for does not arrive, the packet window will wait for a while, expecting it to arrive later. Only the trigger timeout can be determined that the RLC PDU to be waited for is indeed lost, so that the PDCP PDU is not prematurely repackaged. among them.
  • QCI is a QoS (Quality of Service) level identifier, and QCI 1 represents a voice service.
  • step 406 can be implemented as follows:
  • determining whether the PDCP PDU to be combined has a lost RLC PDU includes the following steps:
  • Step A1 Determine a starting RLC PDU and a terminating RLC PDU of the PDCP PDU to be combined according to the FI of the RLC PDU in the pre-stored RLC group packet range.
  • the RLC group packet range includes the correctly received RLC PDU, or the RLC group packet range includes the starting RLC PDU of the next PDCP PDU of the PDCP PDU to be combined and the correctly received RLC PDU.
  • the target RLC PDU in the RLC group packet range includes a starting RLC PDU and a terminating RLC PDU
  • the starting RLC PDU of the next PDCP PDU may not be used
  • the target RLC PDU in the RLC group packet range includes the starting RLC PDU.
  • the RLC PDU is terminated as a lost RLC PDU, and the starting RLC PDU of the next PDCP PDU of the PDCP PDU to be combined is used to determine the terminating RLC PDU of the PDCP PDU to be combined.
  • the FI flag in the RLC header is parsed to determine the start and end RLC fragments of the PDCP.
  • the PDCP PDU can be used as a unit to determine whether each PDCP PDU in the base station needs compensation.
  • the two RLC fragments including the serial PDCP PDU and the non-contiguous PDCP PDU can be parsed from the RLC header whether the RLC fragment is concatenated or not concatenated.
  • the FI bit is generally "11", representing both the end of the previous PDCP PDU and the start of the next PDCP PDU; for the non-contiguous RLC, the starting RLC flag of the PDCP PDU is FI. ", the end of the PDCP PDU, the RLC flag bit FI is "10"
  • Step A2 It is judged whether each RLC PDU ranging from the RLCSN of the starting RLC PDU to the RLCSN of the terminating RLC PDU is not included in the range of the RLC PDU group packet.
  • the PDCP PDU to be combined has a lost RLC PDU.
  • the RLCSN from the RLCSN of the starting RLC PDU to the RLCSN of the terminating RLC PDU is the RLCSN of all RLC PDUs of the PDCP PDU to be combined, all RLC PDUs include correctly received RLC PDUs, ie correctly received RLC PDUs, also including lost The RLC PDU, so that if the RLCSN does not have a corresponding RLC PDU, the RLC PDU is a lost RLC PDU.
  • determining whether the lost RLC PDU belongs to the non-A substream fragment includes the following steps:
  • Step A3 If each RLC PDU from the RLCSN of the starting RLC PDU to the RLCSN of the terminating RLC PDU is not included in the RLC PDU group packet, it is determined that the correctly received RLC PDU received from the starting RLC PDU is received. Whether the total length of the voice data is greater than the total length of the preset A substream.
  • the total length of the A substream is the length of the A substream bits in the voice data of the PDCP PDU to be combined. As shown in the voice frame format shown in FIG. 2 above, the A substream bit is first, and then the B substream bit and the C substream bit, respectively, if the voice data of the correctly received RLC PDU is connected from the starting RLC PDU.
  • the total length of the A substream is greater than the total length of the A substream, it indicates that the A substream bits of the PDCP PDU have been correctly received, that is, the voice data of the correctly received RLC PDU includes all the A substream bits, so that the voice data bits of the lost RLC PDU belong to the non- A substream, in the AMR-WB (Wide Brand; broadband) scenario, as shown in Table 1, the length of the C substream is 0, so that the speech frame only contains the A substream and the B substream, and the non-A substream is B substream.
  • AMR-WB Wide Brand; broadband
  • the lost RLC PDU belonging to the A substream fragment is compensated.
  • the speech frame will be seriously damaged, which not only has a significant drop in the subjective experience of the speech, but also may cause a problem that the receiving side cannot correctly decode. Therefore, in order to improve the execution efficiency of the method of the embodiment of the present invention, the RLC PDU belonging to the A substream fragment may not be compensated.
  • the method for obtaining the total length of the preset A substream may be that before the step A3, the base station can know the current coding rate and the RoHC switch information by parsing the PDCP PDU of the successful group packet, so as to know the length of the A substream,
  • the data processing method of the embodiment of the invention is prepared.
  • Step A4 If the total length of the voice data of the correctly received RLC PDU received from the start RLC PDU connection is greater than the preset total length of the A substream, the following step of acquiring the data domain data of the RLC PDU is performed.
  • the data processing method of the embodiment of the present invention is discarded.
  • the method of the embodiment of the present invention can only determine whether the lost RLC PDU belongs to the A substream fragment or the B substream fragment, and in some embodiments, the RLC PDU that is correctly received can also be determined. That is, whether the correctly received RLC PDU belongs to the A substream or the B substream. In this way, according to whether the total length of the data of the continuously correctly received RLC PDU exceeds the total length of the A substream, it can be determined whether each received and lost RLC PDU belongs to the A substream fragment or the B substream fragment.
  • the A substream is an important bit in a voice PDCP PDU.
  • the A substream includes a high layer header (including a PDCP header, a RoHC compression header, an ETP payload header of a BE or OA mode), and an A substream bit in a protocol-defined voice payload.
  • the A substream contains only the unpredictable parts of the high layer header and the A substream bits in the voice payload.
  • the B substream is a relatively unimportant bit in a voice PDCP PDU.
  • the B substream is the B substream bit in the voice payload specified by the protocol.
  • the lost RLC PDU belongs to the B substream fragment.
  • the lost RLC PDU may not be discriminated, and the lost RLC PDU may be determined to execute the present.
  • the data processing method provided by the embodiment of the present invention is performed when it is determined that the PDCP PDU to be combined has a preset RLC PDU, and the preset RLC PDU is the voice data of the RLC PDU.
  • the RLC PDU whose voice quality parameter is smaller than the preset quality threshold that is, the voice data of the preset RLC PDU is not important in the PDCP PDU to be combined, and after the loss, the compensation for the error-containing RLC PDU has little effect. It also has a positive gain for the overall voice quality; if it is an important RLC PDU, it compensates for it, because the compensation is the wrong RLC PDU, which may have a greater subjective experience impact on the user, and compensate the meaning of the lost RLC PDU. If the PDCP PDU to be combined is discarded directly, the execution efficiency can be improved.
  • the RLC PDU belonging to the non-A substream is a specific example of the RLC PDU whose voice quality parameter of the voice data is smaller than the preset quality threshold.
  • Step 407 Acquire data domain data.
  • the data domain data is data domain data of the RLC PDU type.
  • the voice data of the RLC PDU is acquired to compensate the data domain data of the lost RLC PDU.
  • the embodiment of the present invention describes the data domain data of the RLC PDU as voice data.
  • Step B1 Acquire a target MAC PDU.
  • the target MAC PDU is a CRC check MAC PDU that is checked incorrectly, and the RLC PDU included in the target MAC PDU is an RLC PDU that is to be combined with the PDCP PDU to be combined. That is, the second MAC PDU in step 405, except that the RLC PDU carried by the second MAC PDU is the RLC PDU lost by the PDCP PDU to be combined.
  • the target MAC PDU is obtained, and the specific implementation manner is: determining the receiving times T1 and T2 according to the target correspondence.
  • T1 is the initial reception time of the correctly received RLC PDU before the lost RLC PDU in the PDCP PDU to be combined
  • T2 is the initial of the correctly received RLC PDU after the lost RLC PDU in the PDCP PDU to be combined.
  • Receive time is the target correspondence.
  • the target MAC PDU whose reception time is between T1 and T2 is selected from the second MAC PDU. That is, the target MAC PDU is selected from the stored second MAC PDUs by the correspondence between the initial reception time and the MAC PDU obtained in the above steps.
  • the MAC PDU whose initial reception time is between T1 and T2 is selected from the ErrorBuffer. If the corresponding MAC PDU can be found, the fragment is compensated.
  • the ErrorBuffer is a preset RLC layer buffer that stores the MAC PDUs that are ultimately transmitted incorrectly.
  • the initial transmission reception time selected from the ErrorBuffer is between T1 and T2 If the MAC PDU is not unique, then the compensation for the lost RLC PDU is abandoned. At this time, these non-unique MAC PDUs may be mixed with RTCP packets (RTP Control Packets) and other non-voice bearer packets. The compensation is abandoned in order not to treat the content of the non-voice packet as a voice packet.
  • RTCP Control Packets RTP Control Packets
  • one MAC PDU may be selected to perform the method of the embodiment of the present invention.
  • Step B2 Acquire a voice payload from the target MAC PDU.
  • the data domain data of the RLC PDU can be obtained according to the target MAC PDU. Because the RLC PDU in the target MAC PDU is the lost RLC PDU, the BER in the compensated RLC PDU is small, and the impact on the sound quality is minimized after decoding. In this way, it is achieved that the data in the compensated RLC PDU is the voice payload obtained from the erroneously received MAC PDU.
  • the method for acquiring the data domain data based on the error-accepting MAC PDU of the embodiment of the present invention is also abandoned, thereby giving up Execution of the data processing method of the embodiment of the present invention.
  • This is to make the bit position of receiving the correct RLC fragment good, and the lost RLC PDU is generally no more than 2 B substream fragments.
  • this may not be specifically limited in some embodiments.
  • the specific method for obtaining the voice payload from the target MAC PDU may be different according to different specific locations of the lost RLC PDU in the PDCP PDU to be combined, as follows:
  • Example 1 The number of lost RLC PDUs is one.
  • Step C1 When the lost RLC PDU is located at the non-end position of the PDCP PDU to be combined and the number is one, the voice of the lost RLC PDU is calculated according to the voice data length of the correctly received RLC PDU and the preset voice frame length. The length of the load.
  • the RLCSN of the lost RLC PDU may be determined according to the specific location of the PDCP PDU to be combined.
  • the length of the voice frame can be preset, or the base station can know the current coding rate and RoHC switch information by parsing the PDCP PDU of the successful group packet, thereby knowing the voice frame length.
  • the voice payload length Lm of the lost RLC PDU is calculated based on the correctly received RLC PDU and the voice frame length. For example, using the preset voice frame length minus the total voice data length of the correctly received RLC PDU, the voice payload length of the lost RLC PDU can be obtained.
  • Step C2 For the target MAC PDU, the voice payload whose length is the voice payload length is intercepted from the back.
  • the voice payload length of Lm is intercepted from the back, and the voice payload length is the length of the RLC PDU carried in the target MAC PDU. Since the length of the (MAC+RLC) header of the MAC PDU is not fixed, it is difficult to determine the starting position of the data part, but the voice payload in the target MAC PDU can be obtained by taking the data from the back to the front in the embodiment of the present invention.
  • Example 2 The number of lost RLC PDUs is multiple
  • Obtain the voice payload from the target MAC PDU including:
  • Step D1 Determine the MAC header length of the target MAC PDU when the lost RLC PDU is located at a non-end position of the PDCP PDU to be combined.
  • the base station first determines the head length of the target MAC PDU to prepare for removing the head length. For example, according to whether the periodicBSR_Timer times out, it is determined whether there is a short BSR (Buffer Status Report) in the MAC header. If the MAC PDU to which the RLC PDU in the middle location belongs contains the periodic BSR, the type of the periodic BSR is a short period BSR.
  • a short BSR Buffer Status Report
  • the periodBSR_Timer is a periodic BSR timer. The timer is cleared when the BSR is reported. When the timer expires, the BSR is triggered. The current BSR reporting period is set to 10ms.
  • Step D2 Remove the MAC header length of the target MAC PDU and the preset RLC header length to obtain a voice payload.
  • the RLC header length is predictable. After determining the MAC header length, the MAC header length and the preset RLC header length can be removed from the target MAC PDU, and the obtained data is the voice payload in the target MAC PDU.
  • the MAC header length is 3 bytes, otherwise the MAC header length is 1 Byte.
  • the MAC header length and the RLC header length are removed from the MAC PDU, that is, the voice payload is taken out.
  • the RLCSN length is 10 bits
  • the RLC header length is 2 bytes.
  • the acquisition method of the second embodiment is particularly applicable to a scenario where the number of lost RLC PDUs is multiple.
  • the method of the first embodiment may be used, or the method of the second embodiment may be used.
  • a MAC PDU containing a short period BSR of the last destination MAC PDU is determined from the first MAC PDU. Then, it is judged whether the reception time of the target MAC PDU is longer than the preset period BSR_Timer from the reception time of the MAC PDU containing the short period BSR.
  • the periodicBSR_Timer times out, and the MAC header has a short period BSR.
  • the MAC header length of the target MAC PDU is determined to be 3 bytes. Determining that the MAC header length of the target MAC PDU is 1 Byte if the receiving time of the target MAC PDU is less than the preset periodic BSR_Timer of the MAC PDU with the short BSR of the previous period;
  • the voice payload of most of the lost RLC PDUs can be obtained by the method of Example 2 until one RLC PDU remains, and then Example 1 is executed. Methods.
  • Described above 1.1 is a voice payload compensation method in which the lost RLC PDU fragment belongs to the B substream and is located at the non-end position of the PDCP PDU to be combined.
  • the voice payload compensation method described in 1.2 below is directed to a lost RLC PDU fragment belonging to the B substream and located at the end position of the PDCP PDU to be combined.
  • the voice payload is obtained from the target MAC PDU, including:
  • Step E1 When the lost RLC PDU is located at the end position of the PDCP PDU to be combined, the end position and the start position of the voice payload of the target MAC PDU are determined.
  • the target MAC PDU includes the MAC header and the RLC.
  • the voice payload on the header and data fields it may also include padding bits at the end of the destination MAC PDU, so that the voice payload can be accurately extracted after the start and end positions of the target MAC PDU are determined.
  • Step E2 The voice payload is intercepted from the target MAC PDU according to the start position and the end position.
  • the data between the start position and the end position can be extracted from the target MAC PDU, and the obtained data is the voice payload.
  • Step D1 Calculate the voice payload length of the lost RLC PDU according to the voice data length of the non-end RLC PDU of the PDCP PDU to be combined and the preset voice frame length.
  • Step D2 determining, from the first MAC PDU, a MAC PDU of the last short BSR of the target MAC PDU;
  • Step D3 determining whether the receiving moment of the target MAC PDU is longer than the preset periodic BSR_Timer of the MAC PDU with the short period BSR; if the receiving moment of the target MAC PDU is shorter than the receiving moment of the last short BSR with the short BSR If the preset time of the target MAC PDU is longer than the preset periodic BSR_Timer, the step D5 is performed.
  • step D2 and step D3 are specific implementation methods for determining whether there is a periodic BSR in the MAC header according to whether the periodicBSR_Timer times out.
  • Step D4 determining that the MAC header length of the target MAC PDU is 1 Byte
  • Step D5 Determine, according to the acquired Tb size and the voice payload length of the lost RLC PDU, a period BSR type and a padding length Padding length included in the MAC header length of the target MAC PDU.
  • the periodic BSR type includes a long BSR and a short BSR.
  • the BSR type affects the MAC header length, which in turn affects whether there are padding bits.
  • Step D6 Determine a MAC header length of the target MAC PDU according to the BSR type.
  • Step D7 determining an end position of a voice payload of the target MAC PDU according to the padding bit length
  • Lremain TB Size-7*8-Lm
  • MAC header length is 5Byte
  • padding bit length is Lremain
  • Lremain 16
  • Lremain 16
  • MAC header length is 5Byte
  • no padding bit if Lremian>16, it is a long BSR, the MAC header length is 7Byte, and the padding bit is Lremain-16. Determining the length of the padding bits also determines the end of the voice payload. 7*8 indicates that the MAC header length is 7 bytes and 56 bits when "short BSR, padding" or "long BSR, no padding".
  • Step D8 determining a start position of a voice payload of the target MAC PDU according to the MAC header length and the preset RLC header length;
  • the voice payload can be intercepted from the target MAC PDU according to the start location and the end location.
  • the foregoing method is to acquire data domain data based on the MAC PDU that accepts the error, that is, to reserve the RLC PDU in the MAC PDU with the CRC check error, even if the RLC PDU has an error in the data during the transmission, but retains the RLC PDU of the received error.
  • Voice payload data thereby constructing a compensated RLC PDU, which can retain the corresponding PDCP PDU.
  • Obtaining the data domain data of the RLC PDU can be obtained as follows:
  • the voice payload length of the lost RLC PDU is calculated according to the voice data length of the non-end RLC PDU of the PDCP PDU to be combined and the preset voice frame length, for example, The preset voice frame length of the current mode minus the voice payload length of the RLC PDU that has been received. Then, when the voice payload length is less than the preset length threshold, an all-zero sequence whose length conforms to the voice payload length is acquired.
  • the non-end RLC PDUs may all be correctly received RLC PDUs, or may include RLC PDUs that receive correct correct reception, and also include RLC PDUs obtained through the compensation method of 1.1 above, that is, when the lost RLC PDUs are located.
  • the RLC PDU located at the end of the PDCP PDU is less important voice data, and even if the data is erroneous, the user constructs the voice data constructed with the error-containing data.
  • the subjective experience is not affected much, so the voice data of the lost RLC PDU can be directly compensated by a preset bit sequence, which may be all 1 bit in addition to the all-zero sequence already described.
  • the sequence, or other 01-bit sequence, is not specifically limited in this embodiment of the present invention.
  • the length of the preset bit sequence for compensation may be equal to the length of the voice payload as long as the voice payload length in the end RLC PDU is calculated based on the correctly received or compensated RLC PDU. In this way, the efficiency of generating compensated RLC PDUs can be improved.
  • the embodiment of the present invention further adds a step of determining, that is, if the voice payload length of the lost RLC PDU is not greater than the security length threshold, the preset bit sequence is executed. The method of data domain data.
  • the voice payload length of the lost RLC PDU and the preset security length threshold may be used for determining.
  • the method of 1.2 is performed; if the voice payload length of the lost RLC PDU is not greater than the security length threshold, the method of the above 2 is performed to Compensate for the RLC PDU at the end.
  • the foregoing method for obtaining the data domain data of the RLC PDU is based on receiving the wrong MAC PDU, and some using the preset bit sequence as the data domain data for the RLC PDU located at the end. Another method of obtaining data domain data of an RLC PDU will be presented below.
  • obtaining data domain data of the RLC PDU including:
  • Step E1 Calculate the voice payload length of the lost RLC PDU according to the voice data length of the correctly received RLC PDU and the preset voice frame length.
  • the base station calculates the voice payload length of the lost RLC PDU according to the difference between the voice data length of the received RLC PDU and the preset voice frame length of the PDCP PDU to be combined.
  • the lost RLC PDU may be multiple or one. When there are multiple, the calculated voice payload length is the voice of all lost RLC PDUs. A collection of lengths.
  • Step E2 Acquire a random bit sequence whose length conforms to the voice payload length.
  • the base station After calculating the voice payload length of the lost RLC PDU, the base station acquires a random bit sequence of the same length as the voice payload length of the lost RLC PDU, and uses the random bit sequence as the data domain data of the compensated RLC PDU.
  • the foregoing method of using the random bit sequence as the data domain data can be used as an alternative to the method of acquiring the data domain data of the RLC PDU based on the received error MAC PDU.
  • the method of 3 reference may be made to the description of the embodiment shown in FIG. 5 below.
  • Step 408 Acquire an RLC header.
  • the base station When the PDCP PDU to be combined has a preset RLC PDU lost, the base station also acquires an RLC header to obtain a compensated RLC PDU using the RLC header.
  • the RLC header includes RLCSN and FI.
  • step 408 can be implemented as follows:
  • the RLCSN of the RLC header is determined based on the RLCSN of the correctly received RLC PDU located before and/or after the lost RLC PDU.
  • the FI of the RLC header is 10
  • the FI of the RLC header is 11.
  • the RLCSN of the lost fragment can be accurately estimated.
  • the FI is determined based on whether the lost RLC PDU is the last slice of the PDCP PDU to which it belongs.
  • Step 409 Generate a compensated RLC PDU using the data domain data and the RLC header.
  • the compensated RLC PDU can be generated according to the two data, and the compensated RLC PDU can be used to replace the lost RLC PDU of the PDCP PDU to be combined, so that the base station The PDCP layer can still obtain all the RLC PDUs of the PDCP PDU to be combined if the RLC PDU is lost.
  • the base station combines the inferred RLCSN and the FI flag to form a legal RLC PDU and writes it into the RLC_Buffer of the to-be-packet based on the acquired voice payload.
  • the RLC_Buffer is a buffer of the RLC layer, and stores RLC PDUs that are correctly received and to be grouped.
  • Step 410 Generate a PDCP PDU according to the correctly received RLC PDU and the compensated RLC PDU.
  • the base station includes the correct received RLC PDU of the PDCP PDU to be combined, and the compensated RLC PDU for compensating the lost RLC PDU, so that the PDCP PDU to be combined with the originally lost RLC PDU has no RLC PDU lost, and the base station can A PDCP PDU is generated based on the correctly received RLC PDU and the compensated RLC PDU.
  • the RLC_Buffer in the base station is updated, and the RLC_Buffer includes the correctly received RLC PDU, and also includes the compensated RLC PDU, so that the PDCP PDU can be formed according to the normal group packet flow.
  • At least one correctly received RLC PDU of the PDCP PDU to be combined is obtained, wherein the correctly received RLC PDU is the correct RLC PDU of the CRC of the associated MAC PDU. Then, it is determined whether the PDCP PDU to be combined has a lost RLC PDU, and if the PDCP PDU to be combined has a lost RLC PDU, the data domain data of the RLC PDU is acquired, and the RLC header is obtained. Thus, after the data domain data and the RLC header are used to generate the compensated RLC PDU, the PDCP PDU is generated according to the correctly received RLC PDU and the compensated RLC PDU.
  • the RLC PDU is generated to compensate the lost RLC PDU, and the PDCP PDU can be combined according to the compensated RLC PDU and the correctly received RLC PDU, thereby avoiding discarding the PDCP PDU with the lost RLC PDU.
  • the PDCP packet is retained, thereby reducing data loss and improving the integrity of the frame data.
  • the subjective experience of the user can be improved.
  • the embodiment shown in FIG. 4 mainly provides a method for compensating for a lost RLC PDU based on a received MAC PDU. Another method for compensating for a lost RLC PDU using a random bit sequence will be described below.
  • the MAC PDU needs to be stored. For example, in the RLC layer, a buffer area ErrorBuffer is set to buffer the MAC PDU that receives the error. Therefore, the lost RLC PDU is compensated by the received error MAC PDU in the ErrorBuffer in the subsequent group packet.
  • the embodiment shown in Figure 5 eliminates the need to utilize the buffer area ErrorBuffer.
  • the data field data of the lost RLC PDU is directly compensated by using a random bit sequence.
  • AMR has better robustness to B substreams. Even if the BER of the B substream bits reaches 50%, the MOS score still has a positive gain, which provides a basis for the random compensation scheme.
  • FIG. 5 it is possible to compensate for the loss of any RLC PDU scene belonging to the B substream fragment.
  • FIG. 5 is a flowchart of a method for processing a data according to an embodiment of the present invention. Referring to FIG. 5, the method of the embodiment of the present invention includes:
  • Step 501 Acquire N correctly received RLC PDUs of the PDCP PDU.
  • the correctly received RLC PDU is the correct RLC PDU of the CRC of the associated MAC PDU, and N is a positive integer.
  • Step 501 can be implemented by step 401 and step 404 of the foregoing embodiment.
  • Step 501 can be implemented by step 401 and step 404 of the foregoing embodiment.
  • step 401 and step 404 For details, refer to the detailed description of step 401 and step 404.
  • Step 502 Determine whether the PDCP PDU to be combined has a missing preset RLC PDU. If the PDCP PDU to be combined has a lost RLC PDU, step 503 is performed.
  • the RLC PDU in which the preset RLC PDU is the voice payload is the RLC PDU of the B substream.
  • the missing preset RLC PDUs may be one or more.
  • Step 502 can refer to the detailed description of step 406.
  • the starting RLC PDU and the terminating RLC PDU of the PDCP PDU to be combined are found, and the A and B substream flags of each RLC PDU are determined. Determining whether the PDCP PDU to be combined only loses the RLC PDU belonging to the B substream, and if yes, executing step 503, otherwise determining whether another PDCP PDU to be combined has lost the RLC PDU belonging to the B substream within the RLC group packet range, Similar to the loop.
  • step 502 may be to determine whether a PDCP PDU to be combined has a lost RLC PDU, or the preset RLC PDU belongs to other definitions.
  • Step 503 Calculate a voice payload length of the lost RLC PDU according to the voice data length of the correctly received RLC PDU and the preset voice frame length.
  • Step 504 Acquire a random bit sequence whose length is consistent with the voice payload length.
  • the base station After the base station determines the voice payload length of the lost RLC PDU, the base station only needs to obtain the same length as the voice payload length.
  • the random bit sequence is sufficient, and the acquired random bit sequence can be used as the voice payload of the lost RLC PDU.
  • Step 503 and step 504 are specific methods for obtaining data domain data of the RLC PDU.
  • the number of RLC PDUs belonging to the B substream that are compensated by the random bit sequence is not limited, as long as the compensation is obtained, the RLC PDU obtained according to the compensation and the correctly received RLC are guaranteed.
  • the PDCP PDU obtained by the PDU is the legal voice frame length.
  • the random bit sequence whose length matches the voice payload length is directly used as the voice payload of the lost RLC PDU; when the number of lost RLC PDUs is multiple, the acquisition length is in accordance with the voice net.
  • the specific implementation method of the random bit sequence of the length is to obtain a plurality of random bit sequences of preset length or random length, and the total length of the multiple random bit sequences is equal to the voice payload length. For example, when three RLC PDUs are lost, after calculating the total voice payload length of all the lost RLC PDUs, the voice payload length is divided into three equal parts, and then three segments are randomly selected with each equal length. The bit sequence is OK. Or acquiring a random sequence of 3 random lengths, as long as the total length of the 3-segment random bit sequence is equal to the total voice payload length of the lost RLC PDU.
  • the loss of RLC PDUs can be as follows:
  • the PDCP PDU to be combined only loses one RLC PDU belonging to the B substream fragment, as shown in Fig. 6a. After determining the voice payload of the lost RLC PDU, it can be filled with a random 0-1 bit sequence. At this time, according to the correctly received RLC fragment, in the PDCP PDU to be combined, the missing location of the voice payload can be determined.
  • the PDCP PDU to be combined loses a plurality of consecutive RLC PDUs belonging to the B substream fragment, as shown in FIG. 6b. After determining the voice payload of the lost RLC PDU, the data field data of the RLC PDU may be filled with a random 01 bit sequence. At this time, similarly to Case I, in the PDCP PDU to be combined, the missing location of the voice payload can also be determined.
  • the PDCP PDU to be combined loses a plurality of non-contiguous RLC PDUs belonging to the B substream fragments, as shown in FIG. 6c.
  • the B substream corresponding to each lost RLC PDU is compensated by a random 0-1 bit sequence, and the length of the B substream corresponding to each fragment can be flexibly selected, as long as the total length of the missing B substream is guaranteed to be equal to the total of the lost RLC PDU.
  • the voice payload length can be.
  • Step 505 Acquire an RLC header.
  • Step 505 can refer to the detailed description of step 408.
  • Step 506 Generate a compensated RLC PDU by using the random bit sequence and the RLC header.
  • Step 506 can refer to the detailed description of step 409.
  • the random bit sequence is the data domain data of the RLC PDU.
  • a correct RLC header is added to the voice payload of the randomly compensated RLC PDU to form a legal compensated RLC PDU, which is written into the RLC_Buffer of the packet to be grouped.
  • Step 507 Generate a PDCP PDU according to the correctly received RLC PDU and the compensated RLC PDU.
  • the PDCP PDU includes M RLC PDUs, where M is a positive integer and N ⁇ M. That is, N correctly received RLC PDUs and (M-N) compensated RLC PDUs are combined to obtain PDCP PDUs.
  • Step 507 can refer to the detailed description of step 410.
  • step 502 is performed; otherwise, for the updated RLC_Buffer, the PDCP may be formed according to the normal group packet process.
  • the data processing method of the embodiment of the present invention acquires at least one correctly received RLC PDU of the PDCP PDU to be combined, wherein the correctly received RLC PDU is the correct RLC PDU of the CRC of the associated MAC PDU. Then, it is determined whether the PDCP PDU to be combined has a lost RLC PDU, and if the PDCP PDU to be combined has a lost RLC PDU, the data domain data of the RLC PDU is acquired, and the RLC header is obtained. Thus, after the data domain data and the RLC header are used to generate the compensated RLC PDU, the PDCP PDU is generated according to the correctly received RLC PDU and the compensated RLC PDU.
  • the RLC PDU when the RLC PDU is lost in the PDCP PDU, the RLC PDU is generated to compensate the lost RLC PDU, and the PDCP PDU can be combined according to the compensated RLC PDU and the correctly received RLC PDU, thereby avoiding the loss of the RLC.
  • the PDCP PDU of the PDU is discarded, and the PDCP packet is retained, thereby reducing data loss and improving the integrity of the frame data.
  • the subjective experience of the user can be improved.
  • the RLC layer group packet is allowed to contain a certain BER, and the entire PDCP packet is discarded, thereby retaining more effective voice bits, thereby improving the subjective experience of the voice. Thereby achieving unequal protection of data.
  • the data processing method of the embodiment of the present invention is applicable to the QCI1 and QCI2 services; it is not only suitable for AMR voice, but also can be used for other voice codec systems such as EVS, as well as video telephony services, wireless video service high-level, and low-level packets. Any scenario that prioritizes and importances during the sending process.
  • a scheme to terminate retransmission is also provided.
  • the method for terminating the retransmission may be combined with the foregoing method for compensating for the lost RLC PDU, for example, performing the amplification of the termination retransmission of the embodiment shown in FIG. 7 before step 406; the method shown in FIG. 7 may also be used independently, the embodiment of the present invention This is not specifically limited.
  • the method for performing the method for terminating the retransmission in the embodiment shown in FIG. 7 is to: obtain the MAC PDU corresponding to the non-significant bit in the PDCP PDU; and then, if the number of retransmissions of the MAC PDU corresponding to the non-significant bit reaches the number of retransmissions, The sender sends an ACK message, where the number of retransmissions is less than the maximum number of retransmissions of the MAC PDU.
  • FIG. 7 a flowchart of a method for terminating retransmission is provided.
  • the method of the embodiment of the present invention includes:
  • Step 701 Acquire a transport block including a MAC PDU and a CRC for verifying the MAC PDU through the Harq process.
  • the base station acquires the Harq process through the physical layer, thereby acquiring a transport block including a MAC PDU and a CRC for verifying the MAC PDU.
  • the MAC PDU may include a MAC PDU corresponding to the non-significant bit and a MAC PDU corresponding to the important bit.
  • the MAC PDU corresponding to the non-significant bit is a MAC PDU whose quality parameter of the data bit is smaller than a preset quality threshold, for example, the voice data bit belongs to the MAC PDU of the B substream bit; correspondingly, the MAC PDU corresponding to the important bit is the quality of the data bit.
  • a MAC PDU whose parameter is greater than a preset quality threshold such as a MAC PDU whose voice data bits belong to the A substream bit.
  • Step 701 can be a specific implementation of step 401.
  • Step 702 Determine the type of the MAC PDU.
  • the type of the MAC PDU includes a first type and a second type.
  • the MAC PDU of the first type includes the important bits in the PDCP PDU, that is, the MAC PDU corresponding to the important bit.
  • the second type of MAC PDU includes the non-important in the PDCP PDU.
  • Bit that is, a MAC PDU corresponding to a non-significant bit, the maximum number of retransmissions corresponding to the first type is greater than the number The number of termination retransmissions corresponding to the second type.
  • the number of termination retransmissions is less than the maximum number of retransmissions of the MAC PDU.
  • the maximum number of retransmissions of the MAC PDU is the maximum number of retransmissions set by the system.
  • the maximum number of retransmissions of the MAC PDU and the maximum number of retransmissions corresponding to the first type may be the same value.
  • the voice data of the first type is a MAC PDU belongs to the type of the A substream bit
  • the voice data of the second type of the MAC PDU belongs to the type of the B substream bit
  • the maximum number of retransmissions corresponding to the first type is greater than the second
  • the number of termination retransmissions corresponding to the type is a MAC PDU belongs to the type of the A substream bit
  • the second type is the type of the B substream bit of the voice data of the MAC PDU
  • the maximum number of retransmissions corresponding to the first type is greater than the second type.
  • the maximum number of retransmissions corresponding to the first type may be the maximum number of retransmissions set by the communication system
  • the number of termination retransmissions corresponding to the second type is a preset value.
  • the maximum number of retransmissions or the number of retransmissions corresponding to the type of the MAC PDU can be obtained.
  • the MAC PDU belongs to the second type it is detected whether the number of retransmissions of the MAC PDU belonging to the second type reaches the number of termination retransmissions of the second type.
  • the determining the type of the MAC PDU includes the following steps:
  • Step F1 Acquire an SR (Scheduling Request) message of a PUCCH (Physical Uplink Control Channel).
  • Step F2 Determine, according to the SR message, a starting RLC PDU of the PDCP PDU to be combined from the MAC PDU.
  • the starting fragment of each PDCP PDU can be known.
  • Step F3 Determine the voice data length of the RLC included in the transport block according to the TB Size of the transport block, the average MAC header length, and the RLC header length.
  • Step F4 determining, in the voice data length of the RLC PDU that is continuously received from the starting RLC PDU, that the MAC PDU to which the RLC PDU located in the total length of the preset A substream belongs belongs to the first type, and is located in the preset A sub
  • Step 703 If the number of retransmissions of the second type of MAC PDU reaches the number of termination retransmissions, send an ACK message to the sending end.
  • the base station determines that the MAC PDU transmission error belongs to the second type based on the CRC, and the transmitting end retransmits the MAC PDU.
  • the base station sends an ACK to the sending device. Acknowledgement; Acknowledgement message, the ACK message is used to terminate the retransmission of the second type of MAC PDU by the transmitting device to the transmission error.
  • the maximum number of transmissions corresponding to the A substream type may be a set value, that is, the maximum number of retransmissions set by the system.
  • the number of termination retransmissions of the B substream type is smaller than the maximum number of transmissions of the A substream type, and the number of termination retransmissions of the B substream type is less than the maximum number of retransmissions set by the system, such as a value set by a staff member.
  • the number of termination retransmissions of the B substream type uses a preset value that is less than the maximum number of transmissions of the A substream; or, by comparing the BLER of the currently received B substream with the preset BLER (block error) Rate; block error rate) threshold, adaptively determining the maximum number of retransmissions of the B substream. That is, if the BLER corresponding to the retransmission of the B hop type MAC PDU (the base station will count the BLER of each retransmission in the latest period) is less than the preset BLER threshold, the retransmission can be stopped immediately.
  • the MAC PDUs belonging to the B substream type may be retransmitted 3 times to satisfy the BLER threshold; if the channel conditions are poor, the MAC PDUs belonging to the B substream type may be It is necessary to retransmit 5 times to meet the BLER threshold.
  • the maximum number of retransmissions of the B substream can be adaptively adjusted according to channel conditions.
  • the base station When the number of retransmissions of the B-stream type MAC PDU reaches the number of termination retransmissions corresponding to the B-substream type, the base station sends an ACK message to the UE, so that the UE stops retransmitting the B-stream type MAC PDU, without The retransmission is stopped when the number of retransmissions of the MAC PDU reaches the maximum number of retransmissions set by the system, thereby realizing the early termination of retransmission of the MAC PDU of the B substream type.
  • the HARQ transmission mechanism causes the uplink transmission delay to increase. This phenomenon is particularly serious under the TTI Bundling, and the high latency causes the PDCP layer to actively drop packets. A large number of high-level packet loss will also have a serious impact on the main experience of voice.
  • the existing MAC processing mechanism gives equal transmission opportunities for data packets transmitted by the physical layer.
  • Voice bits of different importance should receive different transmission opportunities (resources). Important sub-flows (such as A sub-streams) should obtain more transmission opportunities than other sub-streams, and appropriate reduction of transmission opportunities for non-significant sub-streams will reduce the probability of high-level packet loss, thereby increasing the overall voice.
  • the throughput rate is particularly serious under the TTI Bundling, and the high latency causes the PDCP layer to actively drop packets. A large number of high-level packet loss will also have a serious impact on the main experience of voice.
  • the existing MAC processing mechanism gives equal transmission opportunities for data packets transmitted by the physical layer.
  • the method of the embodiment shown in FIG. 7 can reduce the number of transmissions of non-significant substreams under a certain BER condition, that is, introduce the characteristic of "early termination" of transmission of non-essential substreams.
  • the MAC PDU of the A substream type is obtained with more transmission opportunities, the probability of packet loss of the high layer PDCP is reduced, the throughput of the voice is improved, and the subjective experience of the voice is finally improved.
  • the method of the embodiment shown in Figure 7 is particularly applicable to TTIB scenarios.
  • Figure 8a and Figure 8b respectively simulate the APR-WB speech rate of 12.65, 23.85k, when the high-level packet loss rate (FER) is 1%, 0.2%, when the B substream contains certain error bits (BER) Not equal to 0), the impact of MOS (Mean Opinion Score subjective evaluation score) points.
  • MOS Mean Opinion Score subjective evaluation score
  • the MOS in the drawing is an average opinion score
  • Es/N0 is the energy power spectral density per symbol energy ratio.
  • Figure 8a and Figure 8b show that, given the high-level packet loss rate, when the B-substream BER is not zero, if the BER is controlled within a certain range, the impact on the subjective experience of speech is almost negligible.
  • Voice service AMR WB service, the rate is set to 12.65k and 23.85k respectively;
  • Evaluation method observe the change of MOS score in the case of channel quality change
  • Abscissa EsN0 normalized value, which represents the average value of the current channel quality
  • FIG. 9 is a gain diagram of a data processing method according to an embodiment of the present invention.
  • the data processing method is specifically a data processing method for acquiring data domain data based on an error-accepting MAC PDU in the method shown in FIG. 4 .
  • the curve shown by Recovery in the figure is the curve generated by the data processing method of the embodiment of the present invention.
  • the MOS score has a significant improvement.
  • Figure 9b compares the performance of the data processing method (i.e., ErrorBufferRecover error buffer compensation) and the random compensation data processing method (i.e., RandomRecover random compensation, the method of the embodiment shown in Fig. 5) based on the error domain MAC PDU acquisition data field data, It can be seen that no matter which compensation scheme is adopted, there is a relatively obvious MOS gain.
  • the compensation scheme based on the error-obtained MAC PDU to obtain the data domain data has the best performance, but the complexity is high; the random compensation scheme has low complexity and the performance is not obvious, which is a good compromise solution.
  • FIG. 10a is a schematic structural diagram of a data processing device according to an embodiment of the present invention
  • FIG. 10b is a partial structural schematic diagram of the data processing device shown in FIG. 10a.
  • the data processing device can be used to perform the data processing method provided by the foregoing embodiments.
  • the data processing device provided by the embodiment of the present invention includes:
  • the obtaining unit 1001 is configured to acquire N correctly received radio link control RLC PDUs of the packet data convergence protocol PDCP protocol data unit PDU;
  • the generating unit 1002 is configured to generate a PDCP PDU according to the N correctly received RLC PDUs, where the PDCP PDU includes M RLC PDUs, where N and M are positive integers, N ⁇ M.
  • the generating unit 1002 includes:
  • the generating module 1004 is configured to generate a PDCP PDU according to the N correctly received RLC PDUs and the compensated RLC PDUs.
  • the data in the compensated RLC PDU is the voice payload obtained from the erroneously received MAC PDU.
  • the obtaining module 1003 includes:
  • the first determining sub-module 1004 is configured to determine, when the PDCP PDU loses the RLC PDU, the receiving times T1 and T2, where T1 is the receiving moment of the correctly received RLC PDU before the lost RLC PDU in the PDCP PDU, and T2 is in the PDCP The receiving moment of the correctly received RLC PDU after the missing RLC PDU in the PDU;
  • the second determining sub-module 1005 is configured to determine, from the MAC PDU that receives the error, a target medium access control MAC PDU whose receiving time is between T1 and T2, where the receiving moment of the RLC PDU is a receiving moment of the MAC PDU to which the RLC PDU belongs The receiving moment of the MAC PDU is the time recorded when the MAC PDU is received;
  • the generating submodule 1007 is configured to generate a compensated RLC PDU according to the voice payload, and compensate the RLC PDU for replacing the lost RLC PDU.
  • the data in the compensated RLC PDU is an all zero sequence or a random bit sequence.
  • the data in the N correctly received RLC PDUs includes the important bits in the PDCP PDU.
  • the device of the embodiment of the present invention further includes a sending unit 1008,
  • the obtaining unit 1001 is further configured to acquire a MAC PDU corresponding to a non-significant bit in the PDCP PDU.
  • the sending unit 1008 is configured to send an ACK message to the sending end, if the number of retransmissions of the MAC PDU corresponding to the non-significant bit reaches the number of retransmissions, where the number of retransmissions is less than the maximum number of retransmissions of the MAC PDU.
  • the obtaining unit 1001 acquires N correctly received RLC PDUs of the PDCP PDU, and then the generating unit 1002 generates PDCP PDUs according to the N correctly received RLC PDUs, where the PDCP PDU includes M RLC PDUs, and N and M are Positive integer, N ⁇ M.
  • the PDCP PDU can be generated by receiving the correct RLC PDU according to the part of the PDCP PDU, and it is not necessary to use all the RLC PDUs to combine and obtain one PDCP PDU, so that the RLC PDU that is not needed to be used is not received or lost, and does not affect. Generation of the PDCP PDU.
  • FIG. 11 is a schematic diagram of a hardware structure of a data processing device according to an embodiment of the present invention.
  • the data processing device 1100 may generate a large difference due to different configurations or performances, and may include one or more central processing units (central processing units). , CPU) 1122 (eg, one or more processors) and memory 1132, one or more storage media 1130 storing application 1142 or data 1144 (eg, one or one storage device in Shanghai).
  • the memory 1132 and the storage medium 1130 may be short-term storage or persistent storage.
  • the program stored on storage medium 1130 may include one or more modules (not shown), each of which may include a series of instruction operations on the data processing device.
  • central processor 1122 can be arranged to communicate with storage medium 1130 to perform a series of instruction operations in storage medium 1130 on data processing device 1100.
  • Data processing device 1100 may also include one or more power sources 1126, one or more wired or wireless network interfaces 1150, one or more input and output interfaces 1158, and/or one or more operating systems 1141, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM and more.
  • one or more power sources 1126 one or more wired or wireless network interfaces 1150, one or more input and output interfaces 1158, and/or one or more operating systems 1141, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM and more.
  • the steps performed by the data processing apparatus in the above embodiments may be based on the data processing apparatus structure shown in FIG. 11, that is, the data processing apparatus shown in FIG. 11 may be used to execute the data processing method of each of the above embodiments.
  • the functional blocks of the data processing device shown in Figures 10a and 10b can be integrated on the data processing device shown in Figure 11.
  • the processor 1122 is configured to perform the following steps by calling an operation instruction stored in the memory 1132:
  • a PDCP PDU is generated according to N correctly received RLC PDUs, the PDCP PDU includes M RLC PDUs, N and M are positive integers, N ⁇ M.
  • the processor 1122 is configured to perform the following steps by calling an operation instruction stored in the memory 1132:
  • a PDCP PDU is generated based on the N correctly received RLC PDUs and the compensated RLC PDUs.
  • the data in the compensated RLC PDU is the voice obtained from the erroneously received Media Access Control MAC PDU. Lotus.
  • the processor 1122 is configured to perform the following steps by calling an operation instruction stored in the memory 1132:
  • T1 is the reception time of the correctly received RLC PDU before the lost RLC PDU in the PDCP PDU
  • T2 is after the missing RLC PDU in the PDCP PDU. The receiving time of the correctly received RLC PDU;
  • the target medium access control MAC PDU whose reception time is between T1 and T2 is determined from the MAC PDU receiving the error.
  • the receiving moment of the RLC PDU is the receiving moment of the MAC PDU to which the RLC PDU belongs, and the receiving moment of the MAC PDU is receiving.
  • the compensated RLC PDU is generated based on the voice payload, and the compensated RLC PDU is used to replace the lost RLC PDU.
  • the data in the compensated RLC PDU is an all zero sequence or a random bit sequence.
  • the data in the N correctly received RLC PDUs includes the important bits in the PDCP PDU.
  • the processor 1122 is configured to perform the following steps by calling an operation instruction stored in the memory 1132:
  • the ACK message is sent to the sender; wherein the number of retransmissions is less than the maximum number of retransmissions of the MAC PDU.
  • the device acquires N correctly received RLC PDUs of the PDCP PDU, and then generates PDCP PDUs according to the N correctly received RLC PDUs, where the PDCP PDU includes M RLC PDUs, and N and M are positive integers, N ⁇ M.
  • the PDCP PDU can be generated by receiving the correct RLC PDU according to the part of the PDCP PDU, and it is not necessary to use all the RLC PDUs to combine and obtain one PDCP PDU, so that the RLC PDU that is not needed to be used is not received or lost, and does not affect. Generation of the PDCP PDU.
  • the computer program product includes one or more computer instructions.
  • the computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable device.
  • the computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be from a website site, computer, server or data center Transfer to another website site, computer, server, or data center by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL), or wireless (eg, infrared, wireless, microwave, etc.).
  • wire eg, coaxial cable, fiber optic, digital subscriber line (DSL), or wireless (eg, infrared, wireless, microwave, etc.).
  • the computer readable storage medium can be any available media that can be stored by a computer or a data storage device such as a server, data center, or the like that includes one or more available media.
  • the usable medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Detection And Prevention Of Errors In Transmission (AREA)
  • Communication Control (AREA)

Abstract

本发明实施例提供了一种数据处理方法和设备，用于根据PDCP PDU的部分正确接收的RLC PDU得到PDCP PDU。获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；根据所述N个正确接收的RLC PDU生成PDCP PDU，所述PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。这样，即可根据PDCP PDU的部分接收正确的RLC PDU即可生成该PDCP PDU，无需使用所有的RLC PDU来组合得到一个PDCP PDU，这样无需使用到的RLC PDU即使错误接收或者丢失，也不影响该PDCP PDU的生成。

Description

数据处理方法和设备 技术领域
本发明涉及通信领域，尤其涉及一种数据处理方法和设备。
背景技术
长期演进(long term evolution，LTE)用户面协议栈，包括分组数据汇聚协议(PDCP)、无线链路协议(RLC)和媒体介入协议(MAC)。其中，RLC层位于PDCP层和MAC层之间，它通过业务接入点(SAP)与PDCP层通信，通过逻辑信道与MAC层通信。在接收侧，一个PDCP PDU由多个RLC PDU组成，若组成PDCP PDU的RLC PDU有至少一个丢失时，MAC层将进行重传，如果重传不能恢复丢失的RLC PDU，则所有与此PDCP PDU相关的RLC PDU分片都会被丢弃，影响用户的主观体验。
发明内容
本发明实施例提供了一种数据处理方法和设备，用于根据PDCP PDU的部分正确接收的RLC PDU得到完整的PDCP PDU，改善用户的主观体验。
第一方面，本发明实施例提供了一种数据处理方法，该方法包括：获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；根据N个正确接收的RLC PDU生成PDCP PDU，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。其中，正确接收的RLC PDU为从正确接收的MAC PDU得到的RLC PDU，正确接收的MAC PDU为对应的CRC校验正确的MAC PDU。
这样，即可根据PDCP PDU的部分接收正确的RLC PDU即可生成该PDCP PDU，无需使用所有的RLC PDU来组合得到一个PDCP PDU，这样无需使用到的RLC PDU即使错误接收或者丢失，也不影响该PDCP PDU的生成。
结合第一方面，在第一种可能的实现方式中，根据N个正确接收的RLC PDU生成PDCP PDU，包括：获取至少一个补偿RLC PDU，根据N个正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。正确接收的RLC PDU为部分，从而还需要获取补偿RLC PDU以与正确接收的RLC PDU组合得到PDCP PDU。例如，PDCP PDU丢失RLC PDU后，获取补偿RLC PDU以替代丢失的RLC PDU，从而以实现组合PDCP PDU。当然，也可以不管有没丢失RLC PDU，设备只用部分正确接收的RLC PDU和补偿RLC PDU即可组合得到PDCP PDU。
结合第一方面第一种可能的实现方式，在第二种可能的实现方式中，根据N个目标RLC PDU生成PDCP PDU，包括：判断PDCP PDU是否有丢失RLC PDU；若待组合PDCP PDU有丢失RLC PDU，则获取至少一个补偿RLC PDU，补偿RLC PDU用于补偿丢失的RLC PDU；根据N个目标RLC PDU和补偿RLC PDU，生成PDCP PDU。即当设备有丢失RLC PDU时，使用补偿RLC PDU作为对丢失的RLC PDU的补偿。
结合第一方面的第一种或第二种可能的实现方式，在第三种可能的实现方式中，获取 至少一个补偿RLC PDU，包括：获取数据域数据，数据域数据为RLC PDU类型的数据域数据，获取RLC头。即分别获取RLC PDU的部分，从而根据数据域数据和RLC头，生成补偿RLC PDU。
结合第一方面的第一种或第二种可能的实现方式，在第四种可能的实现方式中，因错误接收的MAC PDU导致RLC PDU丢失，从而从错误接收的MAC PDU中获取语音数据作为补偿RLC PDU的数据，即补偿RLC PDU中的数据是从错误接收的MAC PDU中获取到的语音净荷。其中，错误接收的MAC PDU为对应的CRC校验错误的MAC PDU。该错误接收的MAC PDU包括的RLC PDU和该PDCP PDU丢失的RLC PDU可为同一RLC PDU。
虽然该错误接收的MAC PDU的数据是含错的，但是可根据这些含错的数据得到补偿RLC PDU，从而根据补偿RLC PDU保留了原来要丢弃的PDCP PDU，减少了语音数据的损失，提高了用户的语音主观体验。
结合第一方面的第三种可能的实现方式，在第五种可能的实现方式中，获取数据域数据，包括：获取目标MAC PDU，目标MAC PDU为被校验错误的CRC校验的MAC PDU，目标MAC PDU包括的RLC PDU为待组合PDCP PDU丢失的RLC PDU。从目标MAC PDU中获取语音净荷。这样的操作，即可实现补偿RLC PDU中的数据是从错误接收的MAC PDU中获取到的语音净荷。
结合第一方面的第五种可能的实现方式，在第六种可能的实现方式中，从目标MAC PDU中获取语音净荷，包括：当丢失的RLC PDU位于PDCP PDU的非末端位置且数量为一个时，根据目标RLC PDU的语音数据长度和预设的语音帧长度，计算得到丢失的RLC PDU的语音净荷长度；对目标MAC PDU，从后往前截取长度为语音净荷长度的语音净荷。在丢失的RLC PDU位于PDCP PDU的非末端位置且数量为一个时，与该丢失的RLC PDU对应的目标MAC PDU的语音数据一直到该目标MAC PDU的末端，从而在得到丢失RLC PDU的语音数据后，可从目标MAC PDU的末端开始获取符合长度的语音数据。
结合第一方面的第五种可能的实现方式，在第七种可能的实现方式中，从目标MAC PDU中获取语音净荷，包括：当丢失的RLC PDU位于待组合PDCP PDU的非末端位置时，确定目标MAC PDU的MAC头长；去除目标MAC PDU的MAC头长和预设的RLC头长，得到语音净荷。这样，将目标MAC PDU的头长去除，即可得到其语音净荷，这种方法尤其适合丢失多个RLC PDU的场景，因为此时不好确定每一丢失的RLC PDU的语音净荷。
结合第一方面的第五种可能的实现方式，在第八种可能的实现方式中，从目标MAC PDU中获取语音净荷，包括：当丢失的RLC PDU位于待组合PDCP的末端位置时，确定目标MAC PDU的语音净荷的结束位置和开始位置；根据开始位置和结束位置，从目标MAC PDU中截取语音净荷。因与末尾的RLC PDU对应的额MAC PDU包括MAC头和RLC头，以及末端还包括填充比特，从而要去除一头一尾的这两类数据，才能得到目标MAC PDU的语音数据。
结合第一方面的第五种可能的实现方式，在第九种可能的实现方式中，该获取至少一个补偿RLC PDU，包括：当PDCP PDU丢失RLC PDU时，确定接收时刻T1和T2，T1 为在PDCP PDU中位于丢失的RLC PDU前的正确接收的RLC PDU的接收时刻，T2为在PDCP PDU中位于丢失的RLC PDU后的正确接收的RLC PDU的接收时刻。然后，从接收错误的MAC PDU中确定接收时刻介于T1和T2之间的目标媒体接入控制MAC PDU，RLC PDU的接收时刻为RLC PDU所属的MAC PDU的接收时刻，MAC PDU的接收时刻为在接收MAC PDU时记录的时刻，即接收端接收MAC PDU时记录的时刻。从目标MAC PDU中获取语音净荷，从而根据语音净荷生成补偿RLC PDU，补偿RLC PDU用于替代丢失的RLC PDU。例如，获取媒体接入控制MAC PDU和对MAC PDU进行校验的循环冗余校验CRC；记录MAC PDU的接收时刻；获取MAC PDU和接收时刻的目标对应关系，MAC PDU包括RLC PDU；当CRC校验正确时，由第一MAC PDU获取正确接收的RLC PDU，第一MAC PDU为MAC PDU中被校验正确的CRC校验的MAC PDU；当CRC校验错误时，存储第二MAC PDU，第二MAC PDU为MAC PDU中被校验错误的CRC校验的MAC PDU；根据目标对应关系，确定接收时刻T1和T2，T1为在待组合PDCP PDU中位于丢失的RLC PDU前的正确接收的RLC PDU的接收时刻，T2为在待组合PDCP PDU中位于丢失的RLC PDU后的正确接收的RLC PDU的接收时刻；根据目标对应关系，从第二MAC PDU中选出接收时刻介于T1和T2之间的目标MAC PDU；从目标MAC PDU中获取语音净荷；根据语音净荷生成补偿RLC PDU。
这样，即可实现从获取的多个接收错误的MAC PDU中选出与丢失的RLL PDU对应的目标MAC PDU，即该目标MAC PDU的接收错误是导致该RLC PDU丢失的原因。
结合第一方面的第一种可能的实现方式，在第十种可能的实现方式中，补偿RLC PDU中的数据为全零序列或者随机比特序列。这样，也能对丢失的RLC PDU进行数据补偿。
结合第一方面，在第十一种可能的实现方式中，N个正确接收的RLC PDU中的数据包括PDCP PDU中的重要比特。正确接收的RLC PDU包括PDCP PDU的重要比特，从而即使对丢失的RLC PDU进行含错补偿，对PDPC PDU的整体数据也影响不大。其中，重要比特可以为语音质量参数大于预设语音质量阈值的比特，例如属于A子流的比特。
结合第一方面的第三种可能的实现方式，在第十二种可能的实现方式中，获取数据域数据，包括：根据N个目标RLC PDU的语音数据长度和预设的语音帧长度，计算丢失的RLC PDU的语音净荷长度；获取长度符合语音净荷长度的随机比特序列。这样，即可实现补偿RLC PDU中的数据为随机比特序列。
结合第一方面的第三种可能的实现方式，在第十三种可能的实现方式中，RLC头包括RLCSN和FI，获取RLC头，包括：根据位于丢失的RLC PDU的前面和/或后面的目标RLC PDU的RLCSN，确定RLC头的RLCSN。当丢失的RLC PDU在待组合PDCP PDU中的末端位置时，确定RLC头的FI为10，当丢失的RLC PDU不在待组合PDCP PDU中的末端位置时，确定RLC头的FI为11。这样，即可实现获取RLC头。
结合第一方面的第二种可能的实现方式，在第十四种可能的实现方式中，判断PDCP是否有丢失RLC PDU，包括：根据预存的RLC组包范围内的RLC PDU的FI确定待组合PDCP PDU的起始RLC PDU和终止RLC PDU，其中，RLC组包范围包括目标RLC PDU，或者RLC组包范围包括待组合PDCP PDU的下一PDCP的起始RLC PDU和目标RLC PDU。
判断在RLC PDU组包范围内是否未包括从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的每一RLC PDU，若在RLC PDU组包范围内未包括从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的每一RLC PDU，则待组合PDCP PDU有丢失RLC PDU。从而，可通过RLCSN可判断PDCP PDU的RLC PDU是否有丢失。
结合第一方面，在第十五种可能的实现方式中，本实现方式的方法还包括：
获取PDCP PDU中非重要比特对应的MAC PDU。若非重要比特对应的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息；其中，终止重传次数小于MAC PDU的最大重传次数。其中，非重要比特对应的MAC PDU例如可以为语音数据属于B子流比特的MAC PDU。该MAC PDU的最大重传次数可以为系统设定的数值。获取MAC PDU的接收端在发送ACK消息后，即可终止发送端对该非重要比特对应的MAC PDU的重传，从而实现了对该MAC PDU的提前终止，无需等到最大重传次数时才终止，避免了高层传输时延的增大，从而减少了高层丢包率，而提前终止重传的是非重要比特对应的MAC PDU，该类型的比特即使含错对用户的主观体验较小。
第二方面，本申请实施例中提供一种数据处理设备，该数据处理设备具有上述方法中数据处理设备的功能。该功能可以通过硬件实现，也可能通过硬件执行相应的软件实现。该硬件或软件包括一个或多个与上述功能相对应的模块。
一种可能的实现方式中，该数据处理设备包括：
获取单元，用于获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；
生成单元，用于根据所述N个正确接收的RLC PDU生成PDCP PDU，所述PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
另一种可能的实现方式中，该数据处理设备包括：
收发器，处理器；
该收发器执行如下动作：获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；
该处理器执行如下动作：根据N个正确接收的RLC PDU生成PDCP PDU，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
本申请的又一方面提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述各方面所述的方法。
本申请的又一方面提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述各方面所述的方法。
本发明实施例提供的技术方案中，设备获取PDCP PDU的N个正确接收的RLC PDU，然后根据N个正确接收的RLC PDU生成PDCP PDU，其中，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。这样，即可根据PDCP PDU的部分接收正确的RLC PDU即可生成该PDCP PDU，无需使用所有的RLC PDU来组合得到一个PDCP PDU，这样无需使用到的RLC PDU即使错误接收或者丢失，也不影响该PDCP PDU的生成。
附图说明
图1a为现有的LTE用户平面协议框图；
图1b为现有的VoLTE下一个语音帧的组包格式示意图；
图1c为现有的RLC SDU分段与串接示意图；
图2为本发明实施例提供的一种数据处理方法的方法流程图；
图3a为本发明实施例提供的一种数据处理方法的方法流程图；
图3b为本发明实施例提供的现有技术和本发明实施例的数据处理方法的对比图；
图4为本发明实施例提供的一种数据处理方法的方法流程图；
图5为本发明实施例提供的一种数据处理方法的方法流程图；
图6a为图5所示实施例涉及的仅丢失1个B子流分片的示意图；
图6b为图5所示实施例涉及的仅丢失丢失多个连续的B子流分片的示意图；
图6c为图5所示实施例涉及的仅丢失多个非连续的B子流分片的示意图；
图7为本发明实施例提供的一种终止重传的方法的方法流程图；
图8a为本发明实施例提供的AMR-WB为12.65k时B子流含错误比特MOS分示意图；
图8b为本发明实施例提供的AMR-WB 23.85k时B子流含错误比特MOS分示意图；
图8c为本发明实施例提供的MOS-vs-BER曲线示意图；
图9a为本发明实施例提供的数据处理方法的增益图；
图9b为本发明实施例提供的基于接受错误的MAC PDU获取数据域数据的数据处理方法和随机补偿的数据处理方法的性能对比示意图；
图10a为本发明实施例提供的一种数据处理设备的结构示意图；
图10b为图10a所示数据处理设备的局部结构示意图；
图11是本发明实施例提供的一种数据处理设备的硬件结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
图1a为LTE(Long Term Evolution；长期演进)用户平面协议框图，表示LTE传输过程中用户平面所涉及到的相关协议层的示意图。其中，RLC层的功能大致如下：
LTE中RLC的功能可以归纳为分段或重分段(segmentation or resegmentation)、串接(concatenation)以及重排(Reordering)、重组(Regroup)。
其中，分段只在UM/AM实体的发送端执行。如图1c所示，当由MAC(Medium Access Control；媒体接入控制)层指示的RLC(Radio Link Control；无线链路控制)PDU(Protocol Data Unit；协议数据单元)大小小于RLC SDU(Service Data Unit；服务数据单元)时，RLC实体就会对RLC SDU执行分段操作，让生成的RLC PDU能适配进去重分段只在AM实体的发送端执行。RLC AM实体支持ARQ重传，当重传的RLC PDU无法适配到由MAC层指示的总的RLC PDU大小时，RLC层就会根据此情况对重传的RLC PDU执行分段操作，让生成的RLC PDU分段能适配进去。串接只在UM/AM的发送端执行。当由MAC层指示的总的RLC PDU大小大于RLC SDU时，RLC层就会根据此情况对RLC SDU执行串接操作，让一个RLC SDU和其他RLC SDU或者其他RLC SDU的分段串联起来，让生成的RLC PDU能适配进去。分段、重分段、串接由发送端完成(图1a中的UE(User Equipment；用户设备)侧)，相应的重排、重组功能则发生在接收端(图1a中的eNodeB(Evolved Node B；演进型Node B)侧)。
重排：HARQ(Hybrid Automatic Repeat reQuest；混合自动重传请求)重传会引起发送端依RLC SN号发送的RLC SDU最终乱序到达接收侧，此时eNodeB检查RCLSN号，任何重复接收的RLC SDU首先被丢弃，剩余的RLC SDU按照RLCSN号完成排序。
重排之后的RLC SDU按照RLC头中的SN、FI位的信息，重新组成一个完整的PDCP SDU,并被发送到上层。但是，如果至少有一个RLC PDU分片丢失了，则与之相关的整个PDCP中的所有其它RLC PDU分片都会被丢弃。
对此，为了进行直观的说明，先介绍下AMR(Adaptive Multi-Rate；自适应多码率)场景。
传统的通信系统中，使用的语音编码系统大部分是基于窄带语音的，频带一般限制在200-3400Hz以内。这一内在的带宽局限性从人类实现电话通信时开始，就已经限制了语音质量的进一步提高。为了实现高品质的语音通信，带宽的限制必须打破。此外，数字通信网络技术的不断突破，如第三代移动通信的出现，给宽带语音编码的应用提供了更广阔的空间。AMR-WB(Adaptive Multi-Rate Wideband；自适应多码率-宽带)语音编码即自适应多码率宽带语音编码，是一种最初设计用于GSM(Global System for Mobile Communication；全球移动通信系统)和第三代移动通信系统WCDMA(Wideband Code Division Multiple Access；宽带码分多址)中的宽带语音编码方式，之后又被ITU(International Telecommunication Union；国际电信联盟)采纳，作为一种高品质数字宽带语音编码新标准,当前被广泛使用在VoLTE(Voice over LTE)中。因此它是通信史上第一种可以同时用于有线与无线业务的语音编码系统。AMR-WB不同于以往的固定码率语音编码方式，它的码率不是固定不变，而是随不同的传输环境而变化的，固有自适应之称。AMR-WB语音带宽为500-7000Hz，采样率为16kHz，相对于传统的200-3400Hz带宽，拓宽的50-200Hz部分增强了语音的自然度和舒适度，高频带扩充的3400-7000Hz部分增强了摩擦音的区分度，从而提高了语音的可理解性，因此AMR-WB是一种高品质的数字宽带语音编码系 统。
AMR语音编码器输出的比特会按照其主观重要性进行重新排序。所谓主观重要性，指当该比特在传输中发生错误(比特翻转或删除)时对语音质量产生的影响，从人的主观体验角度观察，会因比特位置的而在或显著或微弱的差异，因此不同位置的比特对语音质量在主观体验上的重要性是不同的。
为了更好的区分这种因比特位置不同而引起的主观体验的差异，AMR编码器将编码生成的比特分为了三个部分，称为A、B、C子流。A子流包含的比特是所有编码比特中对错误最敏感的比特，当此类编码比特发生错误时，语音帧会产生严重的损坏，不仅在语音的主观体验上有大幅下降，同时也可能会引起在接收侧无法正确译码的问题。因此协议中(3GPP TS 26.201)，A子流有额外的CRC对其进行校验。B、C子流中的语音比特产生错误时，对语音的主观体验按照比特的排列顺序逐渐递减，与A子流不同的是，当B、C子流中含错时，接收侧译码器能够正常完成译码。A、B、C子流对错误比特的敏感度依次为：
A子流>B子流>C子流
同时，A、B、C子流各自内部比特的重要性也是按照排序递减的。协议3GPP TS26.201中列出了各个速率下A、B、C子流的划分，以AMR-WB为例，如表1所示：
表1：
Figure PCTCN2017079327-appb-000001
VoLTE下，语音传输的详细格式如图1b所示。
图1b描述了VoLTE下，一个语音帧的典型组包格式，其中图1b上半部分中的RTP Payload表示语音帧数据，其具体结构如图1b下半部分所示，分为A、B、C三个子流，每个子流的具体长度值遵循3GPP TS 26.201中的要求。以AMR-WB为例，如表1所示，其中C子流的长度为0，语音帧仅仅包含A子流和B子流。
RLC分段、重组机制决定了当属于同一PDCP(Packet Data Convergence Protocol；分组数据汇聚协议)SDU内至少一个分片丢失时，所有与此PDCP相关的RLC分片都会被丢弃。VoLTE场景下，如图1a所示，假设整个PDCP PDU分为[0,…,n-1]个RLC SDU，此 时如果在传输过程中如果第n-1个RLC SDU丢失，则按照RLC层处理机制，全部n个RLC SDU会被丢弃。n为正整数。
由AMR编码方式可知，此时第n-1个RLC SDU处于语音帧的B子流或C子流部分，而处于该部分的编码比特，对于语音的主观体验影响相对于A子流较小。
将所有相关RLC SDU丢失，最终丢失一个PDCP语音包与保留该错误RLC SDU，组成一个含错的PDCP，将其递交的PDCP层，这两种方案相比，直接丢弃所有的RLC SDU在主观体验上引起的损失更大。例如，以AMR-WB 23.85k为例，语音帧全长476比特，当前RLC SDU平均TBSize为100比特，在此场景下，当100比特中发生错误时，丢弃整个476比特，在语音的主观体验上要远差于保留这476比特。
可以理解，上述场景是以AMR为例进行说明，在视频帧传输等领域也可以以此类推。
为此，为了避免在一PDCP PDU有RLC PDU丢失时，要丢弃该PDCP PDU的所有RLC PDU，从而带来较大的数据损失。本发明实施例提供了一种数据处理方法，用于对丢失的RLC PDU进行补偿，以避免丢弃对应的PDCP的所有RLC PDU，从而减少帧数据的损失，提高用户的主观体验。
图2为本发明实施例提供的一种数据处理方法的方法流程图，该方法包括：
步骤201：获取PDCP PDU的N个正确接收的RLC PDU。
数据处理设备只要在生成PDCP PDU前获取到该PDCP PDU中的部分正确接收的RLC PDU。
其中，正确接收的RLC PDU为所属的MAC PDU的CRC(Cyclic Redundancy Check；循环冗余校验)校验正确的RLC PDU，即由基于CRC校验正确的MAC PDU提取得到的RLC PDU，该MAC PDU为传输接收正确的MAC PDU。
其中本发明实施例的数据处理设备可以是基站或UE等设备，本发明实施例对此不作具体限定。
步骤202：根据该N个正确接收的RLC PDU生成PDCP PDU。
其中，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
数据处理设备只要获取到一个正确接收的RLC PDU，即表示获取到该RLC PDU所属的PDCP的有效数据，即使该PDCP含有错误的数据，也可以对其进行保留，或者该PDCP没有获取完整，也可以基于获取到的至少一个正确接收的RLC PDU生成PDCP，也或者，数据处理设备只要基于获取到的正确接收的RLC PDU中的部分RLC PDU，也可以得到PDCP，
本发明实施例的步骤202包括多种实现方式，例如在获取到部分正确接收的RLC PDU后即可使用这些正确接收的RLC PDU生成PDCP。
在本发明有的实施例中，根据N个正确接收的RLC PDU生成PDCP PDU，包括：获取至少一个补偿RLC PDU；然后，根据N个正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
数据数据补偿设备在获取到PDCP PDU的部分正确接收的RLC PDU时，为了生成 PDCP PDU，需要获取满足该PDCP对RLC PDU的数量要求，除了正确接收的RLC PDU外，还要预设数量的的补偿RLC PDU，该补偿RLC PDU和正确接收的RLC PDU可作为PDCP PDU包括的所有RLC PDU，从而根据N个正确接收的RLC PDU和补偿RLC PDU，可该生成PDCP PDU。
其中，补偿RLC PDU的数据域数据有多种来源或类型，例如，补偿RLC PDU中的数据是从错误接收的MAC PDU中获取到的语音净荷，或者，补偿RLC PDU中的数据为全零序列或者随机比特序列。
在本发明有的实施例中，该N个正确接收的RLC PDU中的数据包括PDCP PDU中的重要比特，从而只要获取到重要比特，即可生成该PDCP PDU，剩下的非重要比特可有其它方式填补，例如使用预设比特序列，或者，从接受错误的MAC PDU中获取等。
在本发明的一些实施例中，本发明实施例的待组合PDCP PDU有丢失RLC PDU，从而，获取部分正确接收的RLC PDU后，获取的补偿RLC PDU用于对丢失的RLC PDU进行补偿。这时，本发明实施例的数据处理方法的执行过程如下所述。
图3a为本发明实施例提供的一种数据处理方法的方法流程图。参阅图3a，本发明实施例的数据处理方法，包括：
步骤301：获取待组合PDCP PDU的至少一个正确接收的RLC PDU。
其中，正确接收的RLC PDU为所属的MAC PDU的CRC校验正确的RLC PDU，即由基于CRC(Cyclic Redundancy Check；循环冗余校验)校验正确的MAC PDU提取得到的RLC PDU，该MAC PDU为传输接收正确的MAC PDU。
步骤302：判断待组合PDCP PDU是否有丢失RLC PDU，若待组合PDCP PDU有丢失RLC PDU，则执行步骤303。
数据补偿设备获取到待组合PDCP PDU的至少一个正确接收的RLC PDU后，数据补偿设备判断该待组合PDCP PDU是否有丢失RLC PDU，若有，则尝试补偿该丢失的RLC PDU，具体的补偿方法可参考步骤303至步骤305的详细描述。若待组合PDCP PDU没有丢失RLC PDU，则根据这些正确接收的RLC PDU组合得到PDCP。
其中，具体的判断方法有多种，本发明实施例对此不作具体限定。例如：
1)确定出该待组合PDCP PDU的起始RLC PDU的RLCSN和终止RLC PDU的RLCSN后，当在数据补偿设备的RLC层上未包括从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的每一RLC PDU时，该没包括的RLC PDU即为丢失的RLC PDU，即表示待组合PDCP PDU有丢失RLC PDU，否则，该PDCP PDU没丢失RLC PDU。
2)数据补偿设备在获取了PDCP包的部分RLC PDU后，停止获取剩下的RLC PDU，此时确定该PDCP PDU有丢失RLC PDU，停止的原因可以是设备主动停止获取，或者设备因故障而停止获取等等。
可以理解，该丢失的RLC PDU可以为一个或多个，本发明实施例对此不作具体限定，对该丢失的RLC PDU的类型也不作具体限定。
在本发明的一些实施例中，步骤302具体来说是判断待组合PDCP PDU是否有丢失预设RLC PDU，预设RLC PDU为RLC PDU的语音数据的语音质量参数小于预设质量阀值 的RLC PDU，或者，预设RLC PDU为数据域数据的语音数据属于A子流比特的RLC PDU。即在有特定的RLC PDU丢失时，数据补偿设备才对该丢失的RLC PDU做补偿，否则不对其做补偿，这样，可在不方便补偿的RLC PDU丢失时，停止对该丢失的RLC PDU做补偿，直接丢弃该待组合PDCP PDU包，实现了有选择地对PDCP PDU的丢失RLC PDU做补偿，提高了本发明实施例的方法的灵活性，从而提高执行效率。
步骤303：获取RLC PDU类型的数据域数据。
当该待组合PDCP PDU有丢失RLC PDU时，获取RLC PDU的数据域数据，以该数据域数据对丢失的RLC PDU的数据域数据进行补偿。
其中，具体的获取RLC PDU的数据域数据的方法有多种，本发明实施例对此不作具体限定。例如：
1)使用随机比特序列作为RLC PDU的数据域数据，此时只要该随机比特序列的长度符合丢失的RLC PDU的数据域长度即可。
2)使用预设比特序列作为RLC PDU的数据域数据，此时只要该预设比特序列的长度符合丢失的RLC PDU的数据域长度即可。其中，预设比特序列可以是全零序列，或者根据统计分析得到的对语音或视频等数据有有益补偿效果的比特序列。
3)从接受错误的MAC PDU中截取数据净荷作为RLC PDU的数据域数据，其中该接受错误的MAC PDU为CRC校验错误的MAC PDU，其被设备丢弃，从而设备丢失该接受错误的MAC PDU包括的RLC PDU。但是本发明实施例的方法，使用一预设的缓存区域对该接受错误的MAC PDU进行缓存，在判断出有RLC PDU丢失后，从缓存区域中获取丢失的RLC PDU所属的MAC PDU，从中截取数据净荷，例如语音净荷，即使该数据净荷是错误的，但是仍可使用，以组成一含错的PDCP PDU，从而在设备中保留了该PDCP PDU。
步骤304：获取RLC头。
当该待组合PDCP PDU有丢失RLC PDU时，还要获取RLC头，以使用该RLC头作为补偿RLC PDU的RLC头。其中，该RLC头包括RLCSN和FI。
其中，RLCSN为RLC序列号，当采用10bit的长序列号时，RLCSN从0～1023循环取值。FI为记录RLC分段情况的2bit标志位。
获取RLC头的具体方法有多种，例如：
1)获取到待组合PDCP PDU的所有RLC PDU的RLCSN后，可确定出丢失的RLC PDU的RLCSN，即待组合PDCP PDU的所有RLCSN中，若有RLCSN没丢应的正确接收的RLC PDU，则该RLCSN为丢失的RLC PDU的RLCSN。或者，根据该丢失的RLC PDU前后正确接收的正确接收的RLC PDU的RLCSN可确定出该丢失的RLC PDU的RLCSN，这在丢失的RLC PDU为多个时，尤其适用。然后后根据丢失的RLC PDU在PDCP PDU中的位置确定该RLC PDU的FI，当丢失的RLC PDU在待组合PDCP PDU中的末端位置时，确定RLC头的FI为10，否则FI为11。
2)获取到丢失RLC PDU对应的接收错误的MAC PDU后，从该MAC PDU中提取出RLC头，直接以该RLC头作为丢失的RLC PDU的RLC头。
步骤305：使用数据域数据和RLC头生成补偿RLC PDU；
数据补偿设备获取到RLC头和数据域数据后，即可根据该数据域数据和RLC头生成补偿RLC PDU，该补偿RLC PDU包括该数据域数据和RLC头。该补偿RLC PDU用于对该待组合PDCP PDU丢失的RLC PDU做补偿，即替代该丢失的RLC PDU。从而该待组合PDCP PDU即包括了所有的RLC PDU。
这样，步骤303至步骤305即可获取到补偿RLC PDU。
步骤306：根据正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
数据补偿设备获取了接收正确的正确接收的RLC PDU，以及通过上述的补偿方法对丢失的RLC PDU使用生成的补偿RLC PDU做补偿，这样该数据补偿设备即包括了该待组合PDCP PDU的所有RLC PDU，从而根据这些RLC PDU可组成得到PDCP PDU。
例如，如图3b所示，其示出了本发明实施例的数据处理方法的其中一种具体实现和现有技术的对比图。现有技术中若有属于B子流分片的RLC PDU丢失，则PDCP PDU生成失败，但是本发明实施例通过对丢失的RLC PDU进行补偿，根据正确接收的RLC PDU和补偿得到的RLC PDU即可生成PDCP PDU。
综上所述，本发明实施例提供的技术方案中，获取待组合PDCP PDU的至少一个正确接收的RLC PDU，其中，正确接收的RLC PDU为所属的MAC PDU的CRC校验正确的RLC PDU。然后，判断待组合PDCP PDU是否有丢失RLC PDU，若待组合PDCP PDU有丢失RLC PDU，则获取RLC PDU的数据域数据，以及获取RLC头。从而使用数据域数据和RLC头生成补偿RLC PDU后，根据正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。这样，当PDCP PDU中有RLC PDU丢失时，生成RLC PDU以对该丢失的RLC PDU进行补偿，根据该补偿的RLC PDU和正确接收的RLC PDU即可组合得到PDCP PDU，避免了对有丢失RLC PDU的PDCP PDU的丢弃，保留了该PDCP PDU，从而减少了数据的损失，提高了帧数据的完整性，当用户使用该PDCP PDU相关的数据时，可以提高用户的主观体验。即允许RLC层组包时含有一定的BER，避免丢弃整个PDCP包，从而保留更多的有效的语音比特，从而提升语音的主观体验。
为了更加直观地对本发明实施例的数据处理方法进行说明，下面即以上文有介绍的VoLTE和AMR-WB场景下在基站侧执行本发明实施例的方法为例进行说明。可以理解，本发明实施例的方法除了可以用于AMR-WB的场景外，还可以用于其它场景，例如视频传输场景等等。下述的语音净荷也可以为其它类型的数据。
图4为本发明实施例提供的一种数据处理方法的方法流程图。参阅图4，本发明实施例的数据处理方法包括：
步骤401：获取MAC PDU和对MAC PDU进行校验的CRC。
基站获取MAC PDU和对该MAC PDU进行校验的CRC。对该CRC进行校验，若该CRC校验正确则该MAC PDU传输正确，即基站接收的是正确的MAC PDU；若该CRC校验错误，则该MAC PDU接收错误。
例如，基站的物理层通过Harq进程获取传输块TB和CRC，该传输块上携带有MAC PDU，该CRC用于对该MAC PDU进行校验。其中，MAC PDU为Turbo码编码前的传输块TB，MAC PDU和CRC通过编码、调制，最终承载在物理信道上发送。
步骤402：记录MAC PDU的接收时刻。
基站在获取到携带该MAC PDU的TB(Transport Block,传输块)时，记录接收时刻，该接收时刻用于记录基站获取到该MAC PDU时的时刻。其中，在Harq进程场景中，该接收时刻为初传接收时刻，初传接收时刻用于记录基站初次获取该MAC PDU时的时刻，若该MAC PDU对应的CRC校验错误，即该MAC PDU传输错误时，通过基站的重传机制对该MAC PDU进行重传，以获取传输正确的MAC PDU，重传的MAC PDU的接收时刻可不用记录。为了描述统一，本发明实施例的具体场景的接收时刻为初传接收时刻T。即T为每个Harq进程的初传数据的接收时刻，初传接收时刻T对应MAC PDU，或者说对应TB。
步骤403：获取MAC PDU和接收时刻的目标对应关系。
基站记录接收时刻后，即获取到MAC PDU和初传接收时刻的目标对应关系，该目标对应关系为MAC PDU和初传接收时刻的对应关系。因该MAC PDU包括RLC PDU，从而该目标对应关系也包括RLC PDU和初传接收时刻T的对应关系。
步骤404：当CRC校验正确时，由第一MAC PDU获取正确接收的RLC PDU。
其中，第一MAC PDU为MAC PDU中被该校验正确的CRC校验的MAC PDU。正确接收的RLC PDU为所属的MAC PDU的CRC校验正确的RLC PDU。
基站获取到相互对应的CRC和MAC PDU后，若该CRC校验争取，则对应的MAC PDU为传输正确的MAC PDU，即第一MAC PDU。从该传输正确的MAC PDU中提取出的RLC PDU为传输正确的RLC PDU，该传输正确的RLC PDU即为正确接收的RLC PDU。
步骤404即为基站获取待组合PDCP PDU的至少一个正确接收的RLC PDU的具体方法，基站获取到一待组合PDCP PDU的正确接收的RLC PDU，该正确接收的RLC PDU的数量可以为一个也可以为多个。
例如，基站获取到携带MAC PDU的TB和CRC后，当CRC＝0时，基站的物理层按照正常流程将数据包传递给MAC层，且同时将与其对应初传接收时刻T传递到MAC层。当CRC校验正确时，MAC层将传输正确的MAC PDU及其对应的初传接收时刻T按照正常的处理流程传递到RLC层。其中，CRC为传输块的循环冗余校验值，“0”代表传输正确，“1”代表传输有误。
步骤405：当CRC校验错误时，存储第二MAC PDU。
其中，第二MAC PDU为MAC PDU中被校验错误的CRC校验的MAC PDU。即CRC校验错误，则被该CRC验证的MAC PDU传输错误，该传输错误的MAC PCU即为第二MAC PDU。比特在传输中发生错误，例如比特翻转或删除时，都可能导致MAC PDU发生传输错误。现有的技术中，若该MAC PDU发生传输错误，则丢弃该MAC PDU，从而导致该传输错误的MAC PDU携带的RLC PDU被丢弃。但是本发明实施例的方法将该传输错误的MAC PDU缓存起来，以供后续使用。因该传输错误的MAC PDU被缓存，不从中提取RLC PDU，从而在组合PDCP PDU时，该MAC PDU上的RLC PDU为丢失状态。
例如，当CRC校验错误且该CRC校验的MAC PDU达到最大传输次数时，基站的物理层将校验结果、MAC PDU数据包及其对应的初传接收时刻T传递到MAC层。当CRC 校验错误时，MAC层将传输错误的MAC PDU及其对应的初传接收时刻T传递到RLC层的ErrorBuffer(错误缓存单元)中。其中，ErrorBuffer为预设的RLC层缓存，存放终传错误的MAC PDU。
步骤406：判断待组合PDCP PDU是否有丢失预设RLC PDU，若待组合PDCP PDU有丢失预设RLC PDU，则执行步骤407。
其中该预设RLC PDU为属于非A子流分片的RLC PDU，即RLC PDU的数据域数据的语音数据属于非A子流比特的RLC PDU。例如，属于B子流分片或C子流分片的RLC PDU。
例如，在预设用于缓存接收错误的MAC PDU的ErrorBuffer、而正确接收的RLC PDU存储在RLC_Buffer(RLC缓存单元)中的场景，当触发QCI1的RLC超时组包时，重复步骤步骤406至步骤410，直至RLC_Buffer更新完毕，即使得RLC_Buffer中有丢失RLC PDU的待组合PDCP PDU都尝试被补偿丢失的RLC PDU。在基于接收错误的MAC PDU对丢失的RLC PDU进行补偿的方案中，RLC_Buffer的更新过程即为利用ErrorBuffer尽可能恢复丢失的RLC PDU的过程。由于重传机制，MAC PDU可能是乱序到达的。RLC_Buffer有一个等待机制，如果要等的RLC PDU没到达，组包窗口会等待一段时间，期望它后续会到达。只有触发超时，可确定要等的RLC PDU确实丢失，这样不至于过早地对PDCP PDU进行补包。其中。QCI为QoS(Quality of Service；服务质量)等级标识符，QCI 1代表语音业务。
以AMR-WB场景为例，步骤406可通过如下方式实现：
首先，判断待组合PDCP PDU是否有丢失RLC PDU，包括如下步骤：
步骤A1：根据预存的RLC组包范围内的RLC PDU的FI确定待组合PDCP PDU的起始RLC PDU和终止RLC PDU。
其中，RLC组包范围包括正确接收的RLC PDU，或者RLC组包范围包括待组合PDCP PDU的下一PDCP PDU的起始RLC PDU和正确接收的RLC PDU。
其中，若RLC组包范围中的目标RLC PDU包括起始RLC PDU和终止RLC PDU，可以不使用下一PDCP PDU的起始RLC PDU，若RLC组包范围中的目标RLC PDU包括起始RLC PDU，而终止RLC PDU为丢失的RLC PDU，则使用待组合PDCP PDU的下一PDCP PDU的起始RLC PDU来确定待组合PDCP PDU的终止RLC PDU。
对于正确接收的RLC PDU，解析RLC报头中的FI标志位，可以确定PDCP的起始和末尾RLC分片。通过确定待组合PDCP PDU的起始RLC PDU和终止RLC PDU，即可以PDCP PDU为单位，判断基站中的各个PDCP PDU是否需要补偿。
例如，包括串接PDCP PDU和不串接PDCP PDU的两种RLC分片，从RLC头中可以解析出该RLC分片是串接的还是非串接的。对于串接RLC，FI位一般是“11”，既代表前一个PDCP PDU的末尾，又代表下一个PDCP PDU的起始；对于非串接RLC，PDCP PDU的起始RLC标识位FI为“01”，PDCP PDU的末尾RLC标识位FI为“10”
步骤A2：判断在RLC PDU组包范围内是否未包括从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的每一RLC PDU。
若在RLC PDU组包范围内未包括从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的每一RLC PDU，则待组合PDCP PDU有丢失RLC PDU。从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的RLCSN为该待组合PDCP PDU所有RLC PDU的RLCSN，所有RLC PDU包括正确接收的RLC PDU，即正确接收的RLC PDU，也包括丢失的RLC PDU，从而若有RLCSN没有对应的RLC PDU，则该RLC PDU为丢失的RLC PDU。
然后，判断丢失的RLC PDU是否属于非A子流分片，包括如下步骤：
步骤A3：若在RLC PDU组包范围内未包括从起始RLC PDU的RLCSN到终止RLC PDU的RLCSN范围内的每一RLC PDU，则判断从起始RLC PDU开始连接接收的正确接收的RLC PDU的语音数据总长是否大于预设的A子流总长度。
A子流总长度为待组合PDCP PDU的语音数据中属于A子流比特的长度。如上述图2所示的语音帧格式所示，A子流比特在前，然后依次是B子流比特和C子流比特，若从起始RLC PDU开始连接正确接收的RLC PDU的语音数据总长大于A子流总长度，则表示该PDCP PDU的A子流比特都已经正确接收，即正确接收的RLC PDU的语音数据包括所有的A子流比特，从而丢失的RLC PDU的语音数据比特属于非A子流，在AMR-WB(Wide Brand；宽带)场景中，如表1所示，C子流的长度为0，从而语音帧仅仅包含A子流和B子流，非A子流即为B子流。
基站若不能获取到起始RLC PDU、或者从起始RLC PDU开始连接正确接收的RLC PDU的语音数据小于A子流总长度，则丢失的RLC PDU属于A子流，本发明实施例的方法不对丢失的属于A子流分片的RLC PDU进行补偿。因为A子流比特发生错误时，语音帧会产生严重的损坏，不仅在语音的主观体验上有大幅下降，同时也可能会引起在接收侧无法正确译码的问题。所以为了提高本发明实施例的方法的执行效率，可以不对属于A子流分片的RLC PDU进行补偿。
其中，预设的A子流总长度的获取方式可以为在步骤A3前，基站通过解析成功组包的PDCP PDU，可以知道当前的编码速率和RoHC开关信息，从而知道A子流长度，为本发明实施例的数据处理方法做好准备。
步骤A4：若从起始RLC PDU开始连接接收的正确接收的RLC PDU的语音数据总长大于预设的A子流总长度，则执行下述获取RLC PDU的数据域数据的步骤。
如上所述，若从起始RLC PDU开始连接接收的正确接收的RLC PDU的语音数据总长大于预设的A子流总长度，则表示丢失的RLC PDU属于B子流分片，从而执行下述的补偿该丢失的RLC PDU的方法。在本发明的一些实施例中，若待组合PDCP PDU丢失属于A子流分片的RLC PDU，则放弃执行本发明实施例的数据处理方法。
可以理解，通过上述的方法，本发明实施例的方法可以只判断丢失的RLC PDU属于A子流分片还是属于B子流分片，在有的实施例中也可以还判断正确接收的RLC PDU，即正确接收的RLC PDU属于A子流还是B子流。这样，根据连续正确接收的RLC PDU的数据总长是否超过A子流总长度，可以确定每个收到的和丢失的RLC PDU属于A子流分片还是B子流分片。
其中，A子流为一个语音PDCP PDU中的重要比特。例如，对于RoHC打开场景，A子流包含高层头(包括PDCP头、RoHC压缩头、BE或OA模式的RTP载荷头)以及协议规定的语音净荷中的A子流比特。对于RoHC关闭场景，A子流仅包含高层头中难以预测的部分以及语音净荷中的A子流比特。B子流为一个语音PDCP PDU中相对不重要的比特。例如，B子流即为协议规定的语音净荷中的B子流比特。
这样，即可实现在RLC组包范围内，找到一PDCP PDU的起始RLC PDU和终止RLC PDU，并确定每个RLC PDU属于A子流分片还是B子流分片。
可以理解，上述方法，是以丢失的RLC PDU属于B子流分片进行描述的，在本发明的一些实施例中，可以不对丢失的RLC PDU进行判别，判断出有丢失RLC PDU即可执行本发明实施例提供的数据处理方法。或者，在本发明的一些实施例中，在判断出待组合PDCP PDU有丢失预设RLC PDU时才执行本发明实施例提供的数据处理方法，而该预设RLC PDU为RLC PDU的语音数据的语音质量参数小于预设质量阀值的RLC PDU，即该预设RLC PDU的语音数据在该待组合PDCP PDU中不算重要，丢失后，对其补偿出含错的RLC PDU不但影响不大，还对整体语音质量有正增益；而若是重要的RLC PDU，就其进行补偿，因补偿的是含错的RLC PDU，可能对用户带来较大的主观体验影响，补偿丢失的RLC PDU的意义不大，此时直接丢弃该待组合PDCP PDU反而能提高执行效率。其中，属于非A子流的RLC PDU即为语音数据的语音质量参数小于预设质量阀值的RLC PDU的具体示例。
步骤407：获取数据域数据。
其中，该数据域数据为RLC PDU类型的数据域数据。当待组合PDCP PDU有丢失预设RLC PDU，则获取RLC PDU的语音数据，以对该丢失的RLC PDU的数据域数据进行补偿。本发明实施例以RLC PDU的数据域数据为语音数据进行说明。
其中，获取RLC PDU的数据域数据的具体方法有多种，现举出其中几个示例：
1、基于接受错误的MAC PDU获取数据域数据。
步骤B1：获取目标MAC PDU。
其中，目标MAC PDU为被校验错误的CRC校验的MAC PDU，该目标MAC PDU包括的RLC PDU为待组合PDCP PDU丢失的RLC PDU。即步骤405中的第二MAC PDU，只是该第二MAC PDU携带的RLC PDU为该待组合PDCP PDU丢失的RLC PDU。
可选地，获取目标MAC PDU，具体的实现方式为：根据目标对应关系，确定接收时刻T1和T2。其中，T1为在待组合PDCP PDU中位于丢失的RLC PDU前的正确接收的RLC PDU的初始接收时刻，该T2为在待组合PDCP PDU中位于丢失的RLC PDU后的正确接收的RLC PDU的初始接收时刻。然后，根据目标对应关系，从第二MAC PDU中选出接收时刻介于T1和T2之间的目标MAC PDU。即通过上述步骤中得到的初始接收时刻和MAC PDU的对应关系从存储的第二MAC PDU中选出目标MAC PDU。
例如，根据丢失RLC PDU前后正确接收的RLC PDU的初传接收时刻T1和T2，从ErrorBuffer中选出初传接收时刻介于T1和T2之间的MAC PDU。若能找到对应的MAC PDU，则补偿该分片。ErrorBuffer为预设的RLC层缓存，存放终传错误的MAC PDU。
在本发明的一些实施例中，若从ErrorBuffer中选出的初传接收时刻介于T1和T2之间 的MAC PDU不唯一，则放弃补偿该丢失的RLC PDU。此时，这些不唯一的MAC PDU有可能会夹杂RTCP包(RTP控制包)和其他非语音承载包。放弃补偿是为了不将非语音包的内容当成语音包。当然，在有的实施例中在获取到多个介于T1和T2之间的MAC PDU后，可从中选出一个MAC PDU执行本发明实施例的方法。
步骤B2：从目标MAC PDU中获取语音净荷。
获取到该目标MAC PDU后，即可根据该目标MAC PDU获取RLC PDU的数据域数据。因目标MAC PDU中的RLC PDU即为丢失的RLC PDU，从而补偿得到的RLC PDU中BER较小，解码后对音质影响最小。这样，即可实现补偿RLC PDU中的数据是从错误接收的MAC PDU中获取到的语音净荷。
在本发明的一些实施例中，若丢失的RLC PDU属于B子流分片且数量大于两个时，也放弃执行本发明实施例的基于接受错误的MAC PDU获取数据域数据的方法，从而放弃本发明实施例的数据处理方法的执行。这是为了使得接收正确的RLC分片的比特位置不错位，丢失的RLC PDU一般为不超过2个的B子流分片。当然，在有的实施例中可以不对此做具体限定。
其中，具体的从目标MAC PDU中获取语音净荷的方法可依据丢失的RLC PDU在待组合PDCP PDU中的不同的具体位置而不同，如下所示：
1.1)当丢失的RLC PDU位于待组合PDCP PDU中的非末端位置时
例1：该丢失的RLC PDU的数量为一个
步骤C1：当丢失的RLC PDU位于待组合PDCP PDU的非末端位置且数量为一个时，根据正确接收的RLC PDU的语音数据长度和预设的语音帧长度，计算得到丢失的RLC PDU的语音净荷长度。
其中，在获取待组合PDCP PDU的所有RLC PDU的RLCSN后，可根据丢失的RLC PDU的RLCSN确定丢失的RLC PDU位于待组合PDCP PDU的具体位置。语音帧长度可预先设定，也可以为基站通过解析成功组包的PDCP PDU，可以知道当前的编码速率和RoHC开关信息，从而知道语音帧长度。
根据正确接收的RLC PDU和语音帧长度，计算丢失RLC PDU的语音净荷长度Lm。例如使用预设的语音帧长度减去正确接收的RLC PDU的总语音数据长度，即可得到丢失的RLC PDU的语音净荷长度。
步骤C2：对目标MAC PDU，从后往前截取长度为语音净荷长度的语音净荷。
对于选出的目标MAC PDU，从后往前截取长为Lm的语音净荷长度，该语音净荷长度即为该目标MAC PDU中携带的RLC PDU的长度。由于MAC PDU的(MAC+RLC)报头长度不定，难以确定数据部分的起始位置，但是通过本发明实施例的从后往前取数据的方式即可获取到目标MAC PDU中的语音净荷。
例2：该丢失的RLC PDU的数量为多个
从目标MAC PDU中获取语音净荷，包括：
步骤D1：当丢失的RLC PDU位于待组合PDCP PDU的非末端位置时，确定目标MAC PDU的MAC头长。
基站先确定目标MAC PDU的头长，以为去除该头长做准备。例如根据periodicBSR_Timer是否超时，确定MAC头中是否有周期短BSR(Buffer Status Report；缓冲状态报告)。因据观察现网抓包数据，处于中间位置的RLC PDU所属的MAC PDU若包含周期BSR，则该周期BSR的类型为周期短BSR。
其中，periodicBSR_Timer为周期BSR定时器。每当上报BSR时定时器清零，当定时器超过预设周期时触发BSR上报。当前BSR上报周期设为10ms。
步骤D2：去除目标MAC PDU的MAC头长和预设的RLC头长，得到语音净荷。
RLC头长可预知，确定了MAC头长后，即可从该目标MAC PDU中去除MAC头长和预设的RLC头长，得到的数据即为该目标MAC PDU中的语音净荷。
例如，如果目标MAC PDU包含周期短BSR，则MAC头长为3Byte，否则MAC头长为1Byte。从MAC PDU中去除MAC头长和RLC头长，即取出了语音净荷。RLCSN长度为10bit时，RLC头长为2Byte。
可以理解，例2的获取方法尤其适用于丢失的RLC PDU为多个时的场景，在丢失的RLC PDU为一个时，可以使用例1的方法，也可以使用例2的方法。
其中，根据periodicBSR_Timer是否超时，确定MAC头中是否有周期短BSR的具体方法如下：
从第一MAC PDU中确定目标MAC PDU的上一含周期短BSR的MAC PDU。然后，判断目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻是否超过预设的periodicBSR_Timer。
若目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻超过预设的periodicBSR_Timer，则periodicBSR_Timer超时，MAC头中有周期短BSR。
即若目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻超过预设的periodicBSR_Timer，则确定目标MAC PDU的MAC头长为3Byte。若目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻不超过预设的periodicBSR_Timer，则确定目标MAC PDU的MAC头长为1Byte；
在本发明有的实施例中，在丢失的RLC PDU为多个时，可以通过例2的方法获取到大多数的丢失RLC PDU的语音净荷，直至剩下一个RLC PDU时，再执行例1的方法。
上面1.1描述的是丢失的RLC PDU分片属于B子流且其位于待组合PDCP PDU的非末端位置的语音净荷补偿方法。下面1.2描述的语音净荷补偿方法针对的是丢失的RLC PDU分片属于B子流且其位于待组合PDCP PDU的末端位置。
1.2当丢失的RLC PDU位于待组合PDCP PDU中的末端位置时
当丢失的RLC PDU位于待组合PDCP PDU中的末端位置时，从目标MAC PDU中获取语音净荷，包括：
步骤E1：当丢失的RLC PDU位于待组合PDCP PDU的末端位置时，确定目标MAC PDU的语音净荷的结束位置和开始位置。
因目标MAC PDU的RLC PDU为待组合PDCP PDU的丢失的RLC PDU，该丢失的RLC PDU位于待组合PDCP PDU的末端位置，则目标MAC PDU除了包括MAC头、RLC 头和数据域上的语音净荷外，可能还包括位于目标MAC PDU末尾的填充比特，从而，确定出语音净荷在目标MAC PDU上的开始位置和结束位置后，才能准确地提取出语音净荷。
步骤E2：根据开始位置和结束位置，从目标MAC PDU中截取语音净荷。
如上所述，确定出语音净荷在PDCP PDU上的起始位置和结束位置后，就可以将起始位置到结束位置间的数据从目标MAC PDU上提取出来，得到的数据即为语音净荷。
其中，确定目标MAC PDU的语音净荷的结束位置和开始位置的方法有多种，例如：
步骤D1：根据待组合PDCP PDU的非末尾RLC PDU的语音数据长度和预设的语音帧长度，计算丢失的RLC PDU的语音净荷长度。
步骤D2：从第一MAC PDU中确定目标MAC PDU的上一含周期短BSR的MAC PDU；
步骤D3：判断目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻是否超过预设的periodicBSR_Timer；若目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻不超过预设的periodicBSR_Timer，则执行步骤D4；若目标MAC PDU的接收时刻距离上一含周期短BSR的MAC PDU的接收时刻超过预设的periodicBSR_Timer，则执行步骤D5。
上述的步骤D2和步骤D3即为根据periodicBSR_Timer是否超时，确定MAC头中是否有周期BSR的具体实现方法。
步骤D4：确定目标MAC PDU的MAC头长为1Byte；
步骤D5：根据获取的Tb size和丢失的RLC PDU的语音净荷长度确定目标MAC PDU的MAC头长包括的周期BSR类型和填充比特Padding长度。
其中，周期BSR类型包括长BSR和短BSR。BSR类型影响MAC头长，进而影响是否有填充比特。
步骤D6：根据BSR类型确定目标MAC PDU的MAC头长；
步骤D7：根据填充比特长度确定目标MAC PDU的语音净荷的结束位置；
例如，根据获取的TB Size，MAC头长中的确定部分、RLC头长和净荷长度Lm，推断填充比特的长度和BSR类型。即依据“各MAC控制元素的头长+数据净荷长度+填充长度＝TB Size”这一关系推出各部分长度。譬如，令Lremain＝TB Size-7*8-Lm，若Lremain<16，则为短BSR，MAC头长为5Byte，填充比特长为Lremain；若Lremain＝16，则为长BSR，MAC头长为5Byte，无填充比特；若Lremian>16，则为长BSR，MAC头长为7Byte，填充比特为长为Lremain-16。确定填充比特的长度也就确定了语音净荷的结束位置。7*8表示“短BSR、有填充”或“长BSR、无填充”时的MAC头长为7Byte，56bit。
步骤D8：根据MAC头长和预设的RLC头长确定目标MAC PDU的语音净荷的开始位置；
确定出语音净荷在目标MAC PDU中的开始位置和结束位置后，即可根据开始位置和结束位置，从目标MAC PDU中截取语音净荷。
上述方法为基于接受错误的MAC PDU获取数据域数据，即将CRC校验错误的MAC PDU中的RLC PDU进行保留，即使该RLC PDU在传输过程中数据发生错误，但是保留该接收错误的RLC PDU的语音净荷数据，从而构造出补偿的RLC PDU，可以保留对应的 PDCP PDU。下面将介绍另一种获取RLC PDU的数据域数据的方法。
2、以预设比特序列为数据域数据
获取RLC PDU的数据域数据，可通过如下方式获取：
当丢失的RLC PDU位于待组合PDCP PDU的末端位置时，根据待组合PDCP PDU的非末端RLC PDU的语音数据长度和预设的语音帧长度，计算丢失的RLC PDU的语音净荷长度，例如为当前模式的预设的语音帧长度减去已经接收到RLC PDU的语音净荷长度。然后，当语音净荷长度小于预设长度门限时，获取长度符合语音净荷长度的全零序列。
其中，非末端RLC PDU可以都是正确接收的RLC PDU，也可以是即包括接收正确的正确接收的RLC PDU，也包括通过上述1.1的补偿方式获取到的RLC PDU，即当丢失的RLC PDU位于待组合PDCP PDU中的非末端位置时根据上述方法获取的RLC PDU的数据域数据生成的RLC PDU。
由上文对语音帧的格式的相关内容的描述可知，位于PDCP PDU末尾的RLC PDU为较不重要的语音数据，这些数据即使是含错的，用户对以这些含错数据构造出的语音数据的主观体验受到的影响也不大，所以可以直接以预设比特序列对丢失的RLC PDU的语音数据进行补偿，该预设比特序列除了是已述的全零序列外，还可以是全1比特序列，或其它的01比特序列，本发明实施例对此不作具体限定。只要根据正确接收或已补偿的RLC PDU计算末端RLC PDU中的语音净荷长度后，该用于补偿的预设比特序列的长度与该语音净荷长度相等即可。这样，可提高生成补偿RLC PDU的效率。为了进一步减少含错数据对语音帧的影响，本发明实施例还增加了判断的步骤，即若丢失的RLC PDU的语音净荷长度不大于安全长度门限时，才执行该以预设比特序列为数据域数据的方法。
在本发明的一些实施例中，为了在丢失的RLC PDU位于待组合PDCP PDU的末端位置时，自适应地选择即高效执行又减少错误比特的补偿RLC PDU的方法，在根据待组合PDCP PDU的非末尾RLC PDU的语音数据长度和预设的语音帧长度，计算丢失的RLC PDU的语音净荷长度后，可使用该丢失的RLC PDU的语音净荷长度和预设的安全长度门限进行判断，当丢失的RLC PDU的语音净荷长度大于安全长度门限时，则执行上述的1.2的方法；若丢失的RLC PDU的语音净荷长度不大于安全长度门限时，则执行上述的2的方法，以对位于末端的RLC PDU进行补偿。
上述的获取RLC PDU的数据域数据方法有的是基于接收错误的MAC PDU，有的是对位于末端的RLC PDU使用预设比特序列作为数据域数据。下面将提出另一种获取RLC PDU的数据域数据的方法。
3、以随机比特序列为数据域数据
即获取RLC PDU的数据域数据，包括：
步骤E1：根据正确接收的RLC PDU的语音数据长度和预设的语音帧长度，计算丢失的RLC PDU的语音净荷长度。
基站根据待组合PDCP PDU的接收正确的RLC PDU的语音数据长度和预设的语音帧长度求差，可计算得到丢失的RLC PDU的语音净荷长度。该丢失的RLC PDU可以为多个，也可以为一个，在为多个时，该计算得到的语音净荷长度是所有丢失的RLC PDU的语音净 荷长度的集合。
步骤E2：获取长度符合语音净荷长度的随机比特序列。
基站计算出该丢失的RLC PDU的语音净荷长度后，获取与该丢失的RLC PDU的语音净荷长度等长的随机比特序列，以这些随机比特序列作为补偿RLC PDU的数据域数据。
其中，上述以随机比特序列为数据域数据的方法可作为对基于接收错误的MAC PDU获取RLC PDU的数据域数据的方法的替代。对3的方法的详细描述可参考下述图5所示的实施例的描述。
步骤408：获取RLC头；
当待组合PDCP PDU有丢失预设RLC PDU时，基站还要获取RLC头，以使用该RLC头得到补偿RLC PDU。其中，RLC头包括RLCSN和FI。
可选地，为了准确地获取RLC头，在本发明有的实施例中，步骤408可通过如下方式实现：
根据位于丢失的RLC PDU的前面和/或后面的正确接收的RLC PDU的RLCSN，确定RLC头的RLCSN。当丢失的RLC PDU在待组合PDCP PDU中的末端位置时，确定RLC头的FI为10，当丢失的RLC PDU不在待组合PDCP PDU中的末端位置时，确定RLC头的FI为11。
从而根据丢失的RLC PDU前后正确接收的RLC PDU，可准确推测该丢失分片的RLCSN。根据丢失的RLC PDU是否是所属PDCP PDU的最后一个分片来确定FI。
步骤409：使用数据域数据和RLC头生成补偿RLC PDU。
在通过上述方法获取到RLC PDU的数据域数据和RLC头后，即可根据该两个数据，生成补偿RLC PDU，该补偿RLC PDU可用于替代该待组合PDCP PDU的丢失的RLC PDU，从而基站的PDCP层在丢失RLC PDU的情况下，依然可以获取到待组合PDCP PDU的所有RLC PDU。
例如，基站在获取的语音净荷的基础上，再结合推断的RLCSN和FI标志，组成合法的RLC PDU，写入待组包的RLC_Buffer中。其中RLC_Buffer为RLC层的缓存，存放正确接收、待组包的RLC PDU。
步骤410：根据正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
这样，基站包括待组合PDCP PDU的接收正确的正确接收的RLC PDU，也包括了用于补偿丢失RLC PDU的补偿RLC PDU，从而原本丢失RLC PDU的待组合PDCP PDU已无RLC PDU丢失，基站可根据该正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
例如，通过上述方法的执行，基站中的RLC_Buffer更新完毕，RLC_Buffer包括正确接收的RLC PDU，也包括补偿RLC PDU，从而按正常组包流程即可组成PDCP PDU。
综上所述，获取待组合PDCP PDU的至少一个正确接收的RLC PDU，其中，正确接收的RLC PDU为所属的MAC PDU的CRC校验正确的RLC PDU。然后，判断待组合PDCP PDU是否有丢失RLC PDU，若待组合PDCP PDU有丢失RLC PDU，则获取RLC PDU的数据域数据，以及获取RLC头。从而使用数据域数据和RLC头生成补偿RLC PDU后，根据正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。这样，当PDCP PDU中有 RLC PDU丢失时，生成RLC PDU以对该丢失的RLC PDU进行补偿，根据该补偿的RLC PDU和正确接收的RLC PDU即可组合得到PDCP PDU，避免了对有丢失RLC PDU的PDCP PDU的丢弃，保留了该PDCP包，从而减少了数据的损失，提高了帧数据的完整性，当用户使用该PDCP PDU相关的数据时，可以提高用户的主观体验。
图4所示的实施例主要是提供了一种基于接收错误的MAC PDU对丢失的RLC PDU进行补偿的方法，下面将介绍另一种使用随机比特序列对丢失的RLC PDU进行补偿的方法。在图4所示的方法中在根据CRC校验出接收错误的MAC PDU后需对该MAC PDU进行存储，例如，在RLC层中，设置一缓存区域ErrorBuffer，用于缓存出接收错误的MAC PDU，从而后续组包时利用该ErrorBuffer中的接收错误的MAC PDU对丢失的RLC PDU进行补偿。图5所示的实施例即无需利用该缓存区域ErrorBuffer。而是直接利用随机比特序列对丢失的RLC PDU的数据域数据进行补偿。例如，AMR场景中，AMR对B子流比特有较好的容错鲁棒性，即使B子流比特的BER达到50％，MOS分仍有正增益，这为随机补偿方案提供了依据。在图5所示的实施例中，可以补偿丢失任意个属于B子流分片的RLC PDU场景
图5为本发明实施例提供的一种数据处理方法的方法流程图。参阅图5，本发明实施例的方法包括：
步骤501：获取PDCP PDU的N个正确接收的RLC PDU。
其中，正确接收的RLC PDU为所属的MAC PDU的CRC校验正确的RLC PDU，N为正整数
步骤501可通过上述实施例的步骤401和步骤404实现，具体可参考步骤401和步骤404的详细描述。
步骤502：判断待组合PDCP PDU是否有丢失预设RLC PDU。若待组合PDCP PDU有丢失RLC PDU，则执行步骤503。
其中预设RLC PDU为语音净荷的比特为B子流的RLC PDU。其中丢失的预设RLC PDU可以是一个也可以是多个。
步骤502可参考步骤406的详细描述。
例如，当触发QCI1的RLC超时组包时，在RLC组包范围内，找到待组合PDCP PDU的起始RLC PDU和终止RLC PDU，确定每个RLC PDU的A、B子流标志。判断该待组合PDCP PDU是否仅丢失属于B子流的RLC PDU，若是则执行步骤503，否则在RLC组包范围内，判断另一待组合PDCP PDU是否有丢失属于B子流的RLC PDU，如此类似循环。
在本发明有的实施例中，步骤502可以是判断待组合PDCP PDU是否有丢失RLC PDU，或者该预设RLC PDU属于其它的定义。
步骤503：根据正确接收的RLC PDU的语音数据长度和预设的语音帧长度，计算丢失的RLC PDU的语音净荷长度；
步骤504：获取长度符合语音净荷长度的随机比特序列。
基站确定出丢失的RLC PDU的语音净荷长度后，基站只要获取与该语音净荷长度等长 的随机比特序列即可，该获取的随机比特序列即可作为丢失的RLC PDU的语音净荷。
步骤503和步骤504即为获取RLC PDU的数据域数据的具体方法。
其中，该丢失的RLC PDU可以是一个也可以是多个，即用随机比特序列补偿的属于B子流的RLC PDU数目不限，只要保证补偿后，根据补偿得到的RLC PDU和正确接收的RLC PDU得到的PDCP PDU为合法语音帧长即可。
当丢失的RLC PDU为一个时，直接使用长度符合语音净荷长度的随机比特序列作为丢失的RLC PDU的语音净荷即可；当丢失的RLC PDU的数量为多个时，获取长度符合语音净荷长度的随机比特序列的具体实现方法是：获取多个预设长度或者随机长度的随机比特序列，该多个随机比特序列的总长度等于语音净荷长度。例如，丢失3个RLC PDU时，计算出所有丢失的RLC PDU的总的语音净荷长度后，将该语音净荷长度分为3等分，然后获取3段与每一等分等长的随机比特序列即可。或者获取3段随机长度的随机比特序列，只要该3段随机比特序列的总长度等于丢失的RLC PDU的总的语音净荷长度即可。
例如，RLC PDU丢失可为以下三类情况：
I)待组合PDCP PDU仅丢失1个属于B子流分片的RLC PDU，如图6a所示。确定出丢失的RLC PDU的语音净荷后，用随机0-1比特序列填充即可。此时根据已正确接收的RLC分片，在待组合PDCP PDU中，语音净荷的缺失位置可以确定。
II)待组合PDCP PDU丢失多个连续的属于B子流分片的RLC PDU，如图6b所示。确定出丢失的RLC PDU的语音净荷后，用随机01比特序列填充RLC PDU的数据域数据即可。此时，类似情形I，在待组合PDCP PDU中，语音净荷的缺失位置同样可以确定。
III)待组合PDCP PDU丢失多个非连续的属于B子流分片的RLC PDU，如图6c所示。该情形下用随机0-1比特序列补偿各丢失RLC PDU对应的B子流，各分片对应的B子流长度可以灵活选取，只要保证补偿的缺失B子流总长等于丢失的RLC PDU的总的语音净荷长度即可。在待组合PDCP PDU中，尽管这样会引起非连续丢失的B子流分片中间的正确RLC PDU的语音数据的移位，但仿真表明这样处理影响很小且不会引入负增益，且该情形出现的比例很小。
步骤505：获取RLC头。
步骤505可参考步骤408的详细描述。
步骤506：使用该随机比特序列和RLC头生成补偿RLC PDU。
步骤506可参考步骤409的详细描述。该随机比特序列即为RLC PDU的数据域数据，
例如，为随机补偿的RLC PDU的语音净荷添加正确的RLC头，组成合法的补偿RLC PDU，写入待组包的RLC_Buffer中。
步骤507：根据正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
其中PDCP PDU包括M个RLC PDU，M为正整数，N<M。即N个正确接收的RLC PDU和(M—N)个补偿RLC PDU组合得到PDCP PDU。
步骤507可参考步骤410的详细描述。
例如，若当前PDCP PDU不是RLC_Buffer的RLC组包范围内最后一个PDCP，则执行步骤502；否则对于更新完毕的RLC_Buffer，按正常组包流程组成PDCP即可。
综上所述，本发明实施例的数据处理方法，获取待组合PDCP PDU的至少一个正确接收的RLC PDU，其中，正确接收的RLC PDU为所属的MAC PDU的CRC校验正确的RLC PDU。然后，判断待组合PDCP PDU是否有丢失RLC PDU，若待组合PDCP PDU有丢失RLC PDU，则获取RLC PDU的数据域数据，以及获取RLC头。从而使用数据域数据和RLC头生成补偿RLC PDU后，根据正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。这样，当PDCP PDU中有RLC PDU丢失时，生成RLC PDU以对该丢失的RLC PDU进行补偿，根据该补偿的RLC PDU和正确接收的RLC PDU即可组合得到PDCP PDU，避免了对有丢失RLC PDU的PDCP PDU的丢弃，保留了该PDCP包，从而减少了数据的损失，提高了帧数据的完整性，当用户使用该PDCP PDU相关的数据时，可以提高用户的主观体验。
通过上述的数据处理方法的执行，允许RLC层组包时含有一定的BER，避免丢弃整个PDCP包，从而保留更多的有效的语音比特，从而提升语音的主观体验。从而实现了数据的不等保护。
可以理解，本发明实施例的数据处理方法适用于QCI1和QCI2业务；不仅适合用于AMR语音，同样可用于EVS等着其它语音编解码系统，以及视频电话业务、无线下视频业务高层、底层包发送过程中任何可区分优先级、重要性的场景。
在本发明的一些实施例中，还提供了终止重传的方案。如图7所示的实施例。该终止重传的方法可配合上述补偿丢失RLC PDU的方法，例如在步骤406之前执行图7所示实施例的终止重传的放大；图7所示的方法也可以独立使用，本发明实施例对此不作具体限定。
图7所示实施例的终止重传的方法主要执行过程是：获取PDCP PDU中非重要比特对应的MAC PDU；然后，若非重要比特对应的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息；其中，终止重传次数小于MAC PDU的最大重传次数。
具体来说，如图7所示，其提供了一种终止重传的方法的方法流程图，参阅图7，本发明实施例的方法包括：
步骤701：通过Harq进程获取包括MAC PDU的传输块和对MAC PDU进行校验的CRC。
基站通过物理层获取Harq进程，从而获取包括MAC PDU的传输块和对MAC PDU进行校验的CRC。该MAC PDU可包括非重要比特对应的MAC PDU和重要比特对应的MAC PDU。其中非重要比特对应的MAC PDU为数据比特的质量参数小于预设质量阈值的MAC PDU，例如语音数据比特属于B子流比特的MAC PDU；相应的，重要比特对应的MAC PDU为数据比特的质量参数大于预设质量阈值的MAC PDU，例如语音数据比特属于A子流比特的MAC PDU。
步骤701可为步骤401的具体实现。
步骤702：确定MAC PDU的类型。
其中，MAC PDU的类型包括第一类型和第二类型，第一类型的MAC PDU包括PDCP PDU中的重要比特，即重要比特对应的MAC PDU；第二类型的MAC PDU包括PDCP PDU中的非重要比特，即非重要比特对应的MAC PDU，第一类型对应的最大重传次数大于第 二类型对应的终止重传次数。该终止重传次数小于MAC PDU的最大重传次数。MAC PDU的最大重传次数即系统设定的最大重传次数，MAC PDU的最大重传次数和第一类型对应的最大重传次数可以为相同的数值。
例如，第一类型为MAC PDU的语音数据属于A子流比特的类型，第二类型为MAC PDU的语音数据属于B子流比特的类型，其中，第一类型对应的最大重传次数大于第二类型对应的终止重传次数。例如，第一类型为MAC PDU的语音数据属于A子流比特的类型，第二类型为MAC PDU的语音数据属于B子流比特的类型时，第一类型对应的最大重传次数大于第二类型对应的终止重传次数。其中第一类型对应的最大重传次数可以为通信系统设定的最大重传次数，第二类型对应的终止重传次数为预设的数值。
确定MAC PDU的类型后，即可获取与MAC PDU的类型对应的最大重传次数或者终止重传次数。当MAC PDU属于第二类型时，检测该属于第二类型的MAC PDU的重传次数是否达到第二类型的终止重传次数。其中，确定MAC PDU的类型，包括如下步骤：
步骤F1：获取PUCCH(Physical Uplink Control Channel；物理上行链路控制信道)的SR(Scheduling Request)消息。
步骤F2：根据SR消息，从MAC PDU中确定待组合PDCP PDU的起始RLC PDU。
根据PUCCH的SR消息，可以知道每个PDCP PDU的起始分片。
步骤F3：根据传输块的TB Size和平均MAC头长、和RLC头长，确定传输块包括的RLC的语音数据长度。
步骤F4：在从起始RLC PDU开始连续接收的RLC PDU的语音数据长度中，确定位于预设的A子流总长度内的RLC PDU所属的MAC PDU属于第一类型，位于预设的A子流总长度外的RLC PDU所属的MAC PDU属于第二类型，A子流总长度为待组合PDCP PDU的语音数据中属于A子流比特的长度。
步骤703：若第二类型的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息。
基站基于CRC确定属于第二类型的MAC PDU传输错误，则发送端对该MAC PDU进行重传，当第二类型的MAC PDU的重传次数达到终止重传次数时，基站向发送设备发送ACK(Acknowledgement；确认)消息，ACK消息用于终止发送设备对传输错误的属于第二类型的MAC PDU的重传。
例如，MAC PDU的类型分为A子流类型和B子流类型时A子流类型对应的最大传输次数可为设置值，即系统设定的最大重传次数。B子流类型的终止重传次数小于A子流类型的最大传输次数，B子流类型的终止重传次数为小于系统设定的最大重传次数，例如工作人员设定的数值。对B子流类型的终止重传次数有两种确定方式：采用小于A子流最大传输次数的预设值；或者，通过比较当前收到的B子流的BLER和预设的BLER(block error rate；误块率)门限，自适应地确定B子流的最大重传次数。即如果为B子流类型的MAC PDU的某次重传对应的BLER(基站会统计最近一段时期内各次重传的BLER)小于预设的BLER门限，则可以立即停止重传。例如，如果信道条件好，属于B子流类型的MAC PDU可能重传3次就能满足BLER门限；如果信道条件差，属于B子流类型的MAC PDU可能 要重传5次才能满足BLER门限。B子流的最大重传次数可以根据信道条件自适应地调整。
当B子流类型的MAC PDU的重传次数达到B子流类型对应的终止重传次数时，基站向UE发送ACK消息，从而使得UE停止对该B子流类型的MAC PDU的重传，无需等到该MAC PDU的重传次数达到系统设定的最大重传次数时才停止重传，从而实现了提前终止对B子流类型的MAC PDU的重传。
在LTE场景下，HARQ传输机制会引起高层传输时延增大，这一现像在TTI Bundling下尤为严重，而高时延会引起PDCP层主动丢包。大量的高层丢包，同样会对语音的主体体验带来严重的影响。现有的MAC处理机制对于物理层传输的数据包都是给予同等的传输机会，然而在VoLTE这个场景下，语音帧本所包含的比特本身的重要性是不同的，从最佳传输的角度分析，不同重要性的语音比特，应该获得不同的传输机会(资源)。重要的子流(如A子流)应该比其它的子流获得更多的传输机会，而适当的减少非重要子流的传输机会，会降低高层丢包的的概率，从而从整体上增加语音的吞吐率。
综上所述，图7所示的实施例的方法即可在一定BER的条件下，减少非重要子流的传输次数，即引入对非重要子流的传输“提前终止”的特性。使A子流类型的MAC PDU获得更多的传输机会，降低高层PDCP丢包的概率，提升语音的吞吐量，最终提升语音的主观体验。图7所示的实施例的方法尤其适用于TTIB场景。
为了对上述的方法的效果进行验证，以在AMR场景执行本发明实施例提供的数据处理方法后，对执行效果进行试验验证。
一、AMR不同子流容错能力仿真
因当AMR的A子流发生错误时，不仅语音的主观体验会受到极大影响，同时也会引起译码器的异常，因此认为A子流是十分重要的，BER(Bit Error Ratio；误比特率)应该严格控制在为0，所以此处不需要任何仿真。
图8a和图8b分别仿真了AMR-WB下语音速率为12.65,23.85k时，当高层的丢包率(FER)分别为1％，0.2％时，B子流含有一定的错误比特时(BER不等于0)，对MOS(Mean Opinion Score主观评测分数)分的影响。其中，在附图中的MOS为平均意见得分；Es/N0为每符号能量比上噪声功率谱密度。
图8a和图8b分别表明，给定高层丢包率的前提下，当B子流BER不为0时，如将BER控制在一定的范围内，对语音的主观体验的影响几乎可以忽略不计。
图8c展示了当FER＝10％，A子流错误率＝0，30％，50％，70％时，MOS分随B子流BER变化，基线是丢弃整个PDCP的MOS分。该结果表明，AMR对B子流有较好的容错鲁棒性，即使B子流的BER达到50％，MOS分仍有正增益，这为本发明实施例提供的数据处理方法提供了依据。
二、数据处理方法的仿真验证
信道类型：ETU70信道；
语音业务：AMR WB业务，速率分别设置12.65k和23.85k；
评估方式：在信道质量变化的情况下，观察MOS分的变化；
横坐标：EsN0归一化值,表征当前信道质量的平均值；
纵坐标：MOS分，使用P863-Ploqa直接测量所得。
图9a为本发明实施例的数据处理方法的增益图，该数据处理方法具体为上述图4所示的方法中的基于接受错误的MAC PDU获取数据域数据的数据处理方法。其中，图中Recovery(补偿)所示的曲线即为本发明实施例的数据处理方法产生的曲线。
从上图可知：
相同EsN0/相同信道质量的条件下，使用本发明实施例提供的数据处理方法后，MOS分值有明显的提升。
图9b对比了基于接受错误的MAC PDU获取数据域数据的数据处理方法(即ErrorBufferRecover错误缓存补偿)和随机补偿数据处理方法(即RandomRecover随机补偿，图5所示的实施例的方法)的性能，可见无论采用哪种补偿方案，都有比较明显的MOS增益。基于接受错误的MAC PDU获取数据域数据的补偿方案性能最好，但是复杂度高；随机补偿方案复杂度低，性能降低不明显，不失为一种较好的折中方案。
图10a为本发明实施例提供的一种数据处理设备的结构示意图，图10b为图10a所示数据处理设备的局部结构示意图。该数据处理设备可用于执行上述各实施例提供的数据处理方法，参阅图10a，本发明实施例提供的数据处理设备包括：
获取单元1001，用于获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；
生成单元1002，用于根据N个正确接收的RLC PDU生成PDCP PDU，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
可选地，
生成单元1002，包括：
获取模块1003，用于获取至少一个补偿RLC PDU；
生成模块1004，用于根据N个正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
可选地，
补偿RLC PDU中的数据是从错误接收的MAC PDU中获取到的语音净荷。
可选地，参阅图10b，
获取模块1003，包括：
第一确定子模块1004，用于当PDCP PDU丢失RLC PDU时，确定接收时刻T1和T2，T1为在PDCP PDU中位于丢失的RLC PDU前的正确接收的RLC PDU的接收时刻，T2为在PDCP PDU中位于丢失的RLC PDU后的正确接收的RLC PDU的接收时刻；
第二确定子模块1005，用于从接收错误的MAC PDU中确定接收时刻介于T1和T2之间的目标媒体接入控制MAC PDU，RLC PDU的接收时刻为RLC PDU所属的MAC PDU的接收时刻，MAC PDU的接收时刻为在接收MAC PDU时记录的时刻；
获取子模块1006，用于从目标MAC PDU中获取语音净荷；
生成子模块1007，用于根据语音净荷生成补偿RLC PDU，补偿RLC PDU用于替代丢失的RLC PDU。
可选地，
补偿RLC PDU中的数据为全零序列或者随机比特序列。
可选地，
N个正确接收的RLC PDU中的数据包括PDCP PDU中的重要比特。
可选地，
本发明实施例的设备还包括发送单元1008，
获取单元1001，还用于获取PDCP PDU中非重要比特对应的MAC PDU；
发送单元1008，用于若非重要比特对应的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息；其中，终止重传次数小于MAC PDU的最大重传次数。
综上所述，获取单元1001获取PDCP PDU的N个正确接收的RLC PDU，然后生成单元1002根据N个正确接收的RLC PDU生成PDCP PDU，其中，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。这样，即可根据PDCP PDU的部分接收正确的RLC PDU即可生成该PDCP PDU，无需使用所有的RLC PDU来组合得到一个PDCP PDU，这样无需使用到的RLC PDU即使错误接收或者丢失，也不影响该PDCP PDU的生成。
图11是本发明实施例提供的一种数据处理设备的硬件结构示意图，该数据处理设备1100可因配置或性能不同而产生比较大的差异，可以包括一个或一个以上中央处理器(central processing units，CPU)1122(例如，一个或一个以上处理器)和存储器1132，一个或一个以上存储应用程序1142或数据1144的存储介质1130(例如一个或一个以上海量存储设备)。其中，存储器1132和存储介质1130可以是短暂存储或持久存储。存储在存储介质1130的程序可以包括一个或一个以上模块(图示没标出)，每个模块可以包括对数据处理设备中的一系列指令操作。更进一步地，中央处理器1122可以设置为与存储介质1130通信，在数据处理设备1100上执行存储介质1130中的一系列指令操作。
数据处理设备1100还可以包括一个或一个以上电源1126，一个或一个以上有线或无线网络接口1150，一个或一个以上输入输出接口1158，和/或，一个或一个以上操作系统1141，例如Windows ServerTM，Mac OS XTM，UnixTM,LinuxTM，FreeBSDTM等等。
上述各实施例中由数据处理设备所执行的步骤可以基于该图11所示的数据处理设备结构，即图11所示的数据处理设备可用于执行上述各实施例的数据处理方法。图10a和图10b所示的数据处理设备的各功能模块可集成在图11所示的数据处理设备上。
具体来说，通过调用存储器1132存储的操作指令，处理器1122，用于执行如下步骤：
获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；
根据N个正确接收的RLC PDU生成PDCP PDU，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
可选地，通过调用存储器1132存储的操作指令，处理器1122，用于执行如下步骤：
获取至少一个补偿RLC PDU；
根据N个正确接收的RLC PDU和补偿RLC PDU，生成PDCP PDU。
可选地，
补偿RLC PDU中的数据是从错误接收的媒体接入控制MAC PDU中获取到的语音净 荷。
可选地，通过调用存储器1132存储的操作指令，处理器1122，用于执行如下步骤：
当PDCP PDU丢失RLC PDU时，确定接收时刻T1和T2，T1为在PDCP PDU中位于丢失的RLC PDU前的正确接收的RLC PDU的接收时刻，T2为在PDCP PDU中位于丢失的RLC PDU后的正确接收的RLC PDU的接收时刻；
从接收错误的MAC PDU中确定接收时刻介于T1和T2之间的目标媒体接入控制MAC PDU，RLC PDU的接收时刻为RLC PDU所属的MAC PDU的接收时刻，MAC PDU的接收时刻为在接收MAC PDU时记录的时刻；
从目标MAC PDU中获取语音净荷；
根据语音净荷生成补偿RLC PDU，补偿RLC PDU用于替代丢失的RLC PDU。
可选地，
补偿RLC PDU中的数据为全零序列或者随机比特序列。
可选地，
N个正确接收的RLC PDU中的数据包括PDCP PDU中的重要比特。
可选地，通过调用存储器1132存储的操作指令，处理器1122，用于执行如下步骤：
获取PDCP PDU中非重要比特对应的MAC PDU；
若非重要比特对应的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息；其中，终止重传次数小于MAC PDU的最大重传次数。
综上所述，设备获取PDCP PDU的N个正确接收的RLC PDU，然后根据N个正确接收的RLC PDU生成PDCP PDU，其中，PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。这样，即可根据PDCP PDU的部分接收正确的RLC PDU即可生成该PDCP PDU，无需使用所有的RLC PDU来组合得到一个PDCP PDU，这样无需使用到的RLC PDU即使错误接收或者丢失，也不影响该PDCP PDU的生成。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。
所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本发明实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存储的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘Solid State Disk(SSD))等。

Claims (16)

  1. 一种数据处理方法，其特征在于，所述方法包括：
    获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；
    根据所述N个正确接收的RLC PDU生成PDCP PDU，所述PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
  2. 根据权利要求1所述的方法，其特征在于，
    所述根据所述N个正确接收的RLC PDU生成PDCP PDU，包括：
    获取至少一个补偿RLC PDU；
    根据所述N个正确接收的RLC PDU和所述补偿RLC PDU，生成PDCP PDU。
  3. 根据权利要求2所述的方法，其特征在于，
    所述补偿RLC PDU中的数据是从错误接收的媒体接入控制MAC PDU中获取到的语音净荷。
  4. 根据权利要求3所述的方法，其特征在于，
    所述获取至少一个补偿RLC PDU，包括：
    当所述PDCP PDU丢失RLC PDU时，确定接收时刻T1和T2，所述T1为在所述PDCP PDU中位于所述丢失的RLC PDU前的正确接收的RLC PDU的接收时刻，所述T2为在所述PDCP PDU中位于所述丢失的RLC PDU后的正确接收的RLC PDU的接收时刻；
    从所述接收错误的MAC PDU中确定接收时刻介于T1和T2之间的目标媒体接入控制MAC PDU，RLC PDU的接收时刻为所述RLC PDU所属的MAC PDU的接收时刻，所述MAC PDU的接收时刻为在接收所述MAC PDU时记录的时刻；
    从所述目标MAC PDU中获取语音净荷；
    根据所述语音净荷生成补偿RLC PDU，所述补偿RLC PDU用于替代所述丢失的RLC PDU。
  5. 根据权利要求2所述的方法，其特征在于，
    所述补偿RLC PDU中的数据为全零序列或者随机比特序列。
  6. 根据权利要求1至5任一项所述的方法，其特征在于，
    所述N个正确接收的RLC PDU中的数据包括所述PDCP PDU中的重要比特。
  7. 根据权利要求1至6任一项所述的方法，其特征在于，
    获取所述PDCP PDU中非重要比特对应的MAC PDU；
    若所述非重要比特对应的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息；其中，所述终止重传次数小于所述MAC PDU的最大重传次数。
  8. 一种数据处理设备，其特征在于，所述设备包括：
    获取单元，用于获取分组数据汇聚协议PDCP协议数据单元PDU的N个正确接收的无线链路控制RLC PDU；
    生成单元，用于根据所述N个正确接收的RLC PDU生成PDCP PDU，所述PDCP PDU包括M个RLC PDU，N和M为正整数，N<M。
  9. 根据权利要求8所述的设备，其特征在于，
    所述生成单元，包括：
    获取模块，用于获取至少一个补偿RLC PDU；
    生成模块，用于根据所述N个正确接收的RLC PDU和所述补偿RLC PDU，生成PDCP PDU。
  10. 根据权利要求9所述的设备，其特征在于，
    所述补偿RLC PDU中的数据是从错误接收的MAC PDU中获取到的语音净荷。
  11. 根据权利要求10所述的设备，其特征在于，
    所述获取模块，包括：
    第一确定子模块，用于当所述PDCP PDU丢失RLC PDU时，确定接收时刻T1和T2，所述T1为在所述PDCP PDU中位于所述丢失的RLC PDU前的正确接收的RLC PDU的接收时刻，所述T2为在所述PDCP PDU中位于所述丢失的RLC PDU后的正确接收的RLC PDU的接收时刻；
    第二确定子模块，用于从所述接收错误的MAC PDU中确定接收时刻介于T1和T2之间的目标媒体接入控制MAC PDU，RLC PDU的接收时刻为所述RLC PDU所属的MAC PDU的接收时刻，所述MAC PDU的接收时刻为在接收所述MAC PDU时记录的时刻；
    获取子模块，用于从所述目标MAC PDU中获取语音净荷；
    生成子模块，用于根据所述语音净荷生成补偿RLC PDU，所述补偿RLC PDU用于替代所述丢失的RLC PDU。
  12. 根据权利要求9所述的设备，其特征在于，
    所述补偿RLC PDU中的数据为全零序列或者随机比特序列。
  13. 根据权利要求8至12任一项所述的设备，其特征在于，
    所述N个正确接收的RLC PDU中的数据包括所述PDCP PDU中的重要比特。
  14. 根据权利要求8至13任一项所述的设备，其特征在于，
    所述设备还包括发送单元，
    所述获取单元，还用于获取所述PDCP PDU中非重要比特对应的MAC PDU；
    所述发送单元，用于若所述非重要比特对应的MAC PDU的重传次数达到终止重传次数，则向发送端发送ACK消息；其中，所述终止重传次数小于所述MAC PDU的最大重传次数。
  15. 一种计算机可读存储介质，包括指令，当其在计算机上运行时，使得计算机执行如权利要求1-7任意一项所述的方法。
  16. 一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行如权利要求1-7任意一项所述的方法。
PCT/CN2017/079327 2017-04-01 2017-04-01 数据处理方法和设备 WO2018176480A1 (zh)

Priority Applications (7)

Application Number Priority Date Filing Date Title
JP2019553840A JP7101702B2 (ja) 2017-04-01 2017-04-01 データ処理方法及び装置
CN201780087344.2A CN110326357B (zh) 2017-04-01 2017-04-01 数据处理方法和设备
BR112019020346A BR112019020346A2 (pt) 2017-04-01 2017-04-01 método de processamento de dados, dispositivo, meio de armazenamento legível por computador e produto de programa de computador
EP17903926.8A EP3592101B1 (en) 2017-04-01 2017-04-01 Data processing method and device
KR1020197030561A KR102255051B1 (ko) 2017-04-01 2017-04-01 데이터 처리 방법 및 장치
PCT/CN2017/079327 WO2018176480A1 (zh) 2017-04-01 2017-04-01 数据处理方法和设备
US16/587,910 US11290214B2 (en) 2017-04-01 2019-09-30 Data processing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/079327 WO2018176480A1 (zh) 2017-04-01 2017-04-01 数据处理方法和设备

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/587,910 Continuation US11290214B2 (en) 2017-04-01 2019-09-30 Data processing method and device

Publications (1)

Publication Number Publication Date
WO2018176480A1 true WO2018176480A1 (zh) 2018-10-04

Family

ID=63674051

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/079327 WO2018176480A1 (zh) 2017-04-01 2017-04-01 数据处理方法和设备

Country Status (7)

Country Link
US (1) US11290214B2 (zh)
EP (1) EP3592101B1 (zh)
JP (1) JP7101702B2 (zh)
KR (1) KR102255051B1 (zh)
CN (1) CN110326357B (zh)
BR (1) BR112019020346A2 (zh)
WO (1) WO2018176480A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11564096B2 (en) * 2018-09-13 2023-01-24 Lenovo (Singapore) Pte. Ltd. Integrity protection for a packet data unit
WO2024017054A1 (zh) * 2022-07-22 2024-01-25 大唐移动通信设备有限公司 数据处理方法、装置及存储介质
WO2024066898A1 (zh) * 2022-09-26 2024-04-04 荣耀终端有限公司 一种数据包丢弃的方法及相关设备

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA3233016A1 (en) * 2021-12-23 2023-06-29 Congchi ZHANG Methods and apparatuses for supporting a packet discarding operation in a pdcp layer due to a packet loss
WO2023197312A1 (zh) * 2022-04-15 2023-10-19 Oppo广东移动通信有限公司 无线通信的方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080240011A1 (en) * 2005-10-05 2008-10-02 Electronics And Telecommunications Research Institute Method and Apparatus for Error Correction in Mbms Receipt System
WO2010080916A1 (en) * 2009-01-07 2010-07-15 Qualcomm Incorporated Unbundling packets received in wireless communications
CN102265700A (zh) * 2008-12-26 2011-11-30 Lg电子株式会社 在无线通信系统中释放无线承载的方法和接收机
WO2017024581A1 (zh) * 2015-08-13 2017-02-16 华为技术有限公司 数据传输方法、基站及用户设备

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1255368A1 (en) 2001-04-30 2002-11-06 Siemens Information and Communication Networks S.p.A. Method to perform link adaptation in enhanced cellular communication systems with several modulation and coding schemes
US7904055B2 (en) * 2005-08-23 2011-03-08 Lg Electronics Inc. Communicating message in mobile communication system
EP1764980B8 (en) * 2005-09-20 2009-01-07 Panasonic Corporation Method and apparatus for packet segmentation and concatenation signaling in a communication system
US8503423B2 (en) * 2007-02-02 2013-08-06 Interdigital Technology Corporation Method and apparatus for versatile MAC multiplexing in evolved HSPA
US8315243B2 (en) * 2007-08-14 2012-11-20 Qualcomm Incorporated Transport of PDCP control PDUs within MAC frames
US20090207739A1 (en) * 2008-02-01 2009-08-20 Sung-Duck Chun Mobile communication system and method for transmitting pdcp status report thereof
CN102104535B (zh) * 2009-12-18 2013-12-18 华为技术有限公司 一种pdcp数据发送方法、装置及系统
WO2011077039A1 (fr) 2009-12-23 2011-06-30 France Telecom Procede de communication vocale par paquets de données avec differents niveaux de protection
CN102469511A (zh) * 2010-11-15 2012-05-23 中兴通讯股份有限公司 一种分组数据汇聚协议层处理数据的方法及系统
CN102547848B (zh) * 2011-01-04 2015-08-05 华为技术有限公司 一种处理业务数据流的方法和装置
KR20120130419A (ko) * 2011-05-23 2012-12-03 삼성전자주식회사 이동통신 시스템에서 오류 보정 암호화를 위한 방법 및 장치
CN106304127B (zh) * 2015-05-13 2019-11-01 苏州简约纳电子有限公司 一种lte数据面下行检错纠错方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080240011A1 (en) * 2005-10-05 2008-10-02 Electronics And Telecommunications Research Institute Method and Apparatus for Error Correction in Mbms Receipt System
CN102265700A (zh) * 2008-12-26 2011-11-30 Lg电子株式会社 在无线通信系统中释放无线承载的方法和接收机
WO2010080916A1 (en) * 2009-01-07 2010-07-15 Qualcomm Incorporated Unbundling packets received in wireless communications
WO2017024581A1 (zh) * 2015-08-13 2017-02-16 华为技术有限公司 数据传输方法、基站及用户设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3592101A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11564096B2 (en) * 2018-09-13 2023-01-24 Lenovo (Singapore) Pte. Ltd. Integrity protection for a packet data unit
WO2024017054A1 (zh) * 2022-07-22 2024-01-25 大唐移动通信设备有限公司 数据处理方法、装置及存储介质
WO2024066898A1 (zh) * 2022-09-26 2024-04-04 荣耀终端有限公司 一种数据包丢弃的方法及相关设备

Also Published As

Publication number Publication date
KR20190127862A (ko) 2019-11-13
JP2020516178A (ja) 2020-05-28
US20200028626A1 (en) 2020-01-23
US11290214B2 (en) 2022-03-29
EP3592101A1 (en) 2020-01-08
JP7101702B2 (ja) 2022-07-15
BR112019020346A2 (pt) 2020-04-28
CN110326357A (zh) 2019-10-11
EP3592101B1 (en) 2022-07-20
CN110326357B (zh) 2021-06-22
EP3592101A4 (en) 2020-06-03
KR102255051B1 (ko) 2021-05-24

Similar Documents

Publication Publication Date Title
WO2018176480A1 (zh) 数据处理方法和设备
US8958375B2 (en) Framing for an improved radio link protocol including FEC
JP4243023B2 (ja) パケット−ベースの信頼性レベル設定を伴うａｒｑプロトコル
KR100989529B1 (ko) 무선 통신 시스템에서 향상된 업링크로 오버헤드 감소를 위한 방법 및 장치
KR100978306B1 (ko) 셀룰러 무선 통신 네트워크를 통해 ＶｏＩＰ 서비스를지원하는 방법 및 장치
US9049017B2 (en) Efficient TCP ACK prioritization in wireless networks
US20090319850A1 (en) Local drop control for a transmit buffer in a repeat transmission protocol device
US20120208580A1 (en) Forward error correction scheduling for an improved radio link protocol
US20090074088A1 (en) Adaptive Fragmentation for HARQ in Wireless OFDMA Networks
MXPA06014835A (es) Metodo y aparato para control de enlace en comunicaciones inalambricas.
KR20080012274A (ko) 인터넷 프로토콜 어플리케이션들에 대한 상이한 에러 방지및 상이한 에러 검출을 제공하기 위한 방법
KR20150129048A (ko) 고속 다운링크 패킷 액세스를 통해 패킷을 전송하고 수신하는 방법 및 장치
KR20040023568A (ko) 패킷 기반 통신 시스템을 위한 순방향 에러 보정 시스템및 방법
EP2673925A1 (en) Method for improved robust header compression with low signal energy
WO2014205814A1 (zh) 一种数据传输方法、装置、基站及用户设备
US7924710B2 (en) Method for transmitting data including an error control mechanism designed for unreliable networks and error resilience applications
WO2017143538A1 (zh) 语音数据传输方法以及装置
WO2020238689A1 (zh) 一种无线局域网中的通信方法及设备
TWI538459B (zh) 蜂巢式無線通信網路上支援網路語音服務方法及裝置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17903926

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019553840

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017903926

Country of ref document: EP

Effective date: 20191001

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112019020346

Country of ref document: BR

ENP Entry into the national phase

Ref document number: 20197030561

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 112019020346

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20190927