WO2012163099A1 - 语音信号处理方法、装置和接入网系统 - Google Patents

语音信号处理方法、装置和接入网系统 Download PDF

Info

Publication number
WO2012163099A1
WO2012163099A1 PCT/CN2012/070658 CN2012070658W WO2012163099A1 WO 2012163099 A1 WO2012163099 A1 WO 2012163099A1 CN 2012070658 W CN2012070658 W CN 2012070658W WO 2012163099 A1 WO2012163099 A1 WO 2012163099A1
Authority
WO
WIPO (PCT)
Prior art keywords
substream
decoding
result
crc
base station
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2012/070658
Other languages
English (en)
French (fr)
Inventor
魏岳军
唐欣
朱佥
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to KR1020137034285A priority Critical patent/KR101611018B1/ko
Priority to EP12793985.8A priority patent/EP2709280A4/en
Priority to RU2013158335/08A priority patent/RU2543944C1/ru
Priority to BR112013030548-7A priority patent/BR112013030548A2/pt
Priority to CN201280025927.XA priority patent/CN103782518B/zh
Priority to JP2014511711A priority patent/JP5827743B2/ja
Publication of WO2012163099A1 publication Critical patent/WO2012163099A1/zh
Priority to US14/092,319 priority patent/US9171540B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03MCODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/03Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
    • H03M13/05Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits
    • H03M13/09Error detection only, e.g. using cyclic redundancy check [CRC] codes or single parity bit
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/08Speech classification or search
    • G10L15/14Speech classification or search using statistical models, e.g. Hidden Markov Models [HMMs]
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/08Error detection or correction by redundancy in data representation, e.g. by using checking codes
    • G06F11/10Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
    • G06F11/1004Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's to protect a block of data words, e.g. CRC or checksum
    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03MCODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/03Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
    • H03M13/23Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using convolutional codes, e.g. unit memory codes
    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03MCODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/37Decoding methods or techniques, not specific to the particular type of coding provided for in groups H03M13/03 - H03M13/35
    • H03M13/39Sequence estimation, i.e. using statistical methods for the reconstruction of the original codes
    • H03M13/41Sequence estimation, i.e. using statistical methods for the reconstruction of the original codes using the Viterbi algorithm or Viterbi processors
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/14Relay systems
    • H04B7/15Active relay systems
    • H04B7/155Ground-based stations
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/004Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0045Arrangements at the receiver end
    • H04L1/0054Maximum-likelihood or sequential decoding, e.g. Viterbi, Fano, ZJ algorithms
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/004Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056Systems characterized by the type of code used
    • H04L1/0059Convolutional codes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/004Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056Systems characterized by the type of code used
    • H04L1/0061Error detection codes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/004Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056Systems characterized by the type of code used
    • H04L1/0064Concatenated codes
    • H04L1/0065Serial concatenated codes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/004Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056Systems characterized by the type of code used
    • H04L1/007Unequal error protection

Definitions

  • the present invention relates to the field of communications, and in particular, to a voice signal processing method, apparatus, and access network system. Background technique
  • FIG. 1 is a schematic diagram of a system architecture of a voice coding process in the prior art.
  • AMR Speech Encoder (hereinafter referred to as AMR Speech Encoder) in the User Equipment (hereinafter referred to as UE) is encoded by a convolutional code encoder (Convolutional Code Encoder, CC Encoder).
  • the AMR voice signal encoded by the CC Encoder is sent to the base station through an air interface (hereinafter referred to as: NodeB), and the CC decoder in the NodeB (hereinafter referred to as CC Decoder) can decode the AMR voice signal, and the CC Decoder includes two
  • the output of the channel is sent to the radio network controller (Radio Network Controller, RNC for short) through the Iub port, and the RNC is sent to the core network (Core Network, hereinafter referred to as CN) through the Iu port.
  • RNC Radio Network Controller
  • AMR speech decoder (hereinafter referred to as: AMR Speech Decoder), another way through the Iub port will be a circular convolution check (Cyclic Redundancy Check, hereinafter referred to as: CRC), the CRC indicator (CRC Indicator, hereinafter referred to as CRCI) is sent to the RNC, and the RNC can send a bad frame indication to the AMR Speech Decoder in the CN through the Iu interface according to the CRCI.
  • Frame Indicator hereinafter referred to as: BFI
  • CC Decoder will also The CRCI is sent to the outer loop power control (hereinafter referred to as Outer-Loop Power Control) in the RNC.
  • the AMR Speech Decoder After receiving the decoded bit stream and BFI, the AMR Speech Decoder can perform decoding processing.
  • the Outer-Loop Power Control can adjust the target block error rate (BLER) according to the CRCI, and send the inner ring power control (hereinafter referred to as Inner-Loop Power Control) in the NodeB according to the adjusted BLER.
  • BLER target block error rate
  • Target SINR Target signal-to-noise ratio
  • the inner loop power control sends power commands to the UE's power transmitter (hereinafter referred to as Power Transmitter) according to the measured signal-to-noise ratio (hereinafter referred to as Measured SINR) and Target SINR (hereinafter referred to as: Power Transmitter) Abbreviation: Power Commander) to adjust the UE's transmit power.
  • Power Transmitter the UE's power transmitter
  • Measured SINR the measured signal-to-noise ratio
  • Power Transmitter hereinafter referred to as: Power Transmitter
  • Power Commander Abbreviation: Power Commander
  • AMR voice signals can be divided into three substreams A, B, and C, that is, Class A, Class B and Class C, where A substream has the greatest impact on voice quality, and most importantly, its data block is followed by a 12-bit CRC check.
  • the importance of B and C substreams is relatively low, and there is no data block.
  • the CC Decoder in the NodeB uses the Viterbi Algorithm (hereinafter referred to as VA) decoder, and in the decoding result of the VA decoder, only the A substream has the CRCI.
  • VA Viterbi Algorithm
  • Embodiments of the present invention provide a voice signal processing method, apparatus, and access network system to improve decoding performance of a convolutional code for an A substream.
  • the embodiment of the invention provides a voice signal processing method, including:
  • the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a cyclic convolution check CRC; Decoding processing the first substream, the second substream, and the third substream, where the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision; and sent to the base station controller a decoding result of the first substream, the second substream, and the third substream, where the decoding result of the first substream includes a decoded bitstream and a CRC check result.
  • An embodiment of the present invention provides another voice signal processing method, including:
  • the decoding result of the first substream is translated by using a decoding algorithm based on a cyclic convolution check CRC auxiliary decision a decoding result obtained after the code processing, and the decoding result includes a decoded bit stream and a CRC check result;
  • An embodiment of the present invention provides a base station, including:
  • a first receiving module configured to receive a coded voice signal sent by the user equipment UE, where the coded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a cyclic convolutional school CRC
  • a decoding processing module configured to perform decoding processing on the first substream, the second substream, and the third substream by using a decoding algorithm, where a decoding algorithm based on a CRC auxiliary decision is used to the first sub Streaming for decoding processing;
  • a first sending module configured to send, to the base station controller, a decoding result of the first substream, the second substream, and the third substream, where the decoding result of the first substream includes a decoded bit stream and CRC check result.
  • An embodiment of the present invention provides a base station controller, including:
  • a second receiving module configured to receive a decoding result of the first substream, the second substream, and the third substream sent by the base station, where the decoding result of the first substream is a CRC auxiliary based on a cyclic convolution check a decoding result obtained by the decoding algorithm of the decoding process, and the decoding result includes a decoding bit stream and a CRC check result; and the decoded bit stream and the CRC check result of the first substream and The decoding results of the second substream and the third substream are sent to the core network.
  • An embodiment of the present invention provides an access network system, including: a base station and a base station controller; wherein, the base station uses the foregoing base station, and the base station controller uses the foregoing base station controller.
  • the embodiment of the invention further provides another voice signal processing method, including:
  • an encoded voice signal where the encoded voice signal includes a first substream, a second substream and a third substream, wherein the first substream includes a cyclic convolution check CRC; and the first substream, the second substream, and the third substream are decoded and processed by using a decoding algorithm to obtain a decoding result of the first substream, the second substream, and the third substream, wherein the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision, the first substream
  • the decoding result includes a decoded bit stream and a CRC check result;
  • Decoding processing of the AM R speech signal is performed according to the decoded bit stream of the first substream and the CRC check result and the decoding results of the second substream and the third substream.
  • the embodiment of the invention further provides a user equipment, including:
  • a receiving module configured to receive a coded voice signal sent by the base station, where the coded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a cyclic convolution check CRC;
  • a first decoding processing module configured to perform decoding processing on the first substream, the second substream, and the third substream by using a decoding algorithm, to obtain the first substream, the second substream, and the third a decoding result of the substream, wherein the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision, where the decoding result of the first substream includes a decoded bit stream and a CRC check a second decoding processing module, configured to perform translation of the AMR voice signal according to the decoded bit stream and the CRC check result of the first substream and the decoding result of the second substream and the third substream Code processing.
  • the base station may perform decoding processing on the first substream by using a decoding algorithm based on a CRC auxiliary decision, which is compared with the prior art using a common VA decoding algorithm for decoding processing.
  • the decoding performance of the first substream can be improved, and the importance of the first substream is higher in the voice quality. Therefore, the embodiment of the present invention can improve the decoding performance of the first substream and improve the voice quality. Meet the higher requirements of users for voice quality.
  • the UE may perform decoding processing on the first substream by using a decoding algorithm based on a CRC auxiliary decision, compared to the prior art using a common VA decoding algorithm for decoding processing.
  • the decoding performance of the first substream can be improved, and the first substream is highly important in the voice quality. Therefore, the embodiment of the present invention can improve the decoding performance of the first substream and improve the voice quality. , to meet the higher requirements of users for voice quality.
  • DRAWINGS The drawings used in the embodiments or the description of the prior art are briefly described. It is obvious that the drawings in the following description are some embodiments of the present invention, and are not creative to those skilled in the art. Other drawings can also be obtained from these drawings on the premise of labor.
  • FIG. 1 is a schematic diagram of a system architecture of a speech coding process in the prior art
  • FIG. 2 is a schematic structural diagram of processing three substreams in the system architecture shown in FIG. 1.
  • FIG. 3 is a flowchart of Embodiment 1 of a voice signal processing method according to the present invention.
  • FIG. 4 is a schematic structural diagram of a PLVA used in an embodiment of a method for processing a voice signal according to the present invention
  • FIG. 5 is a schematic structural diagram of another PLVA decoder used in an embodiment of a method for processing a voice signal according to the present invention
  • Embodiment 6 is a flowchart of Embodiment 2 of a method for processing a voice signal according to the present invention
  • Embodiment 7 is a flowchart of Embodiment 3 of a method for processing a voice signal according to the present invention.
  • Embodiment 8 is a flowchart of Embodiment 4 of a method for processing a voice signal according to the present invention.
  • FIG. 9 is a schematic structural diagram of processing three substreams in Embodiment 4 of the method shown in FIG. 8;
  • FIG. 10 is a flowchart of Embodiment 5 of a voice signal processing method according to the present invention.
  • FIG. 1 is a flowchart of Embodiment 6 of a method for processing a voice signal according to the present invention
  • Embodiment 1 of a base station is a schematic structural diagram of Embodiment 1 of a base station according to the present invention.
  • Embodiment 13 is a schematic structural diagram of Embodiment 2 of a base station according to the present invention.
  • Embodiment 3 of a base station according to the present invention is a schematic structural diagram of Embodiment 3 of a base station according to the present invention.
  • Embodiment 4 of a base station according to the present invention is a schematic structural diagram of Embodiment 4 of a base station according to the present invention.
  • Embodiment 16 is a schematic structural diagram of Embodiment 1 of a base station controller according to the present invention.
  • Embodiment 17 is a schematic structural diagram of Embodiment 2 of a base station controller according to the present invention.
  • Embodiment 3 of a base station controller is a schematic structural diagram of Embodiment 3 of a base station controller according to the present invention.
  • Embodiment 4 of a base station controller is a schematic structural diagram of Embodiment 4 of a base station controller according to the present invention.
  • FIG. 20 is a schematic structural diagram of an embodiment of an access network system according to the present invention.
  • FIG. 21 is a flowchart of Embodiment 7 of a method for processing a voice signal according to the present invention.
  • Embodiment 8 is a flowchart of Embodiment 8 of a method for processing a voice signal according to the present invention.
  • FIG. 23 is a schematic structural diagram of a UE in the method embodiment shown in FIG. 22; 24 is a schematic structural diagram of Embodiment 1 of a user equipment according to the present invention;
  • Embodiment 25 is a schematic structural diagram of Embodiment 2 of a user equipment according to the present invention.
  • FIG. 26 is a schematic structural diagram of Embodiment 3 of a user equipment according to the present invention.
  • FIG. 27 is a schematic structural diagram of Embodiment 4 of a user equipment according to the present invention.
  • the technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention.
  • the embodiments are a part of the embodiments of the invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.
  • GSM Global System for Mobile Communications
  • CDMA Code Division Multiple Access 2000 system.
  • WCDMA Wideband Code Division Multiple Access
  • LTE Long Term Evolution
  • the base station may be a base station (Base Transceiver Station, hereinafter referred to as BTS) in GSM or CDMA2000, or may be a base station NodeB in WCDMA, or may be an evolved Node B in LTE (hereinafter referred to as eNB or eNodeB).
  • BTS Base Transceiver Station
  • eNB evolved Node B in LTE
  • the present invention is not limited, but for convenience of description, the following embodiments are described by taking a NodeB as an example.
  • the base station controller may be a base station controller (hereinafter referred to as BSC) in GSM or CDMA2000, or may be an RNC in WCDMA.
  • BSC base station controller
  • RNC radio network controller
  • FIG. 3 is a flowchart of Embodiment 1 of a method for processing a voice signal according to the present invention. As shown in FIG. 3, the method in this embodiment is an improvement on a method performed by the CC Decoder of the NodeB in FIG. 1. The method in this embodiment may include :
  • Step 301 Receive an encoded voice signal sent by the UE, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a cyclic convolution check CRC.
  • the NodeB can receive the encoded voice signal sent by the UE, and the encoded voice
  • the signal can be the AMR speech signal processed by the CC Encoder in FIG.
  • the AMR voice signal may include three substreams A, B, and C in FIG. 2, corresponding to the first substream, the second substream, and the third substream, respectively.
  • the first substream that is, the A substream, contains a CRC.
  • Step 302 Perform decoding processing on the first substream, the second substream, and the third substream by using a decoding algorithm, where the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision deal with.
  • the NodeB may be a CC Decoder in the NodeB, and the first substream, the second substream, and the third substream may be separately decoded by using a decoding algorithm.
  • the decoding algorithm based on the CRC auxiliary decision is used to decode the first substream, that is, the A substream. deal with.
  • the decoding algorithm based on the CRC auxiliary decision requires the CRC auxiliary decision, and the second substream and the third substream do not contain the CRC, the second substream and the third substream, that is, the B substream and the C substream
  • the stream can be decoded using a VA decoder in the prior art.
  • the decoding algorithm based on CRC-assisted decision can effectively improve the decoding performance of convolutional codes.
  • the basic principle is: output the global optimal multiple candidate paths through the Viterbi algorithm, and pass the CRC.
  • the CRC check is performed on the decoding results corresponding to the paths, and the correct decoding result of the CRC check is selected as the final result. If the decoding results corresponding to all the paths fail to pass the CRC check, the decoding of the best path is output. The result is the final result. Since this decoding algorithm can choose between multiple paths including the best path, performance is better than the normal Viterbi algorithm that only selects the best path.
  • the CRC-assisted decision-based decoding algorithm is a parallel enumeration Viterbi decoding algorithm for four candidate paths (Parallel List Viterbi Algoriyhm-4, hereinafter referred to as PLVA-4)
  • PLVA-4 Parallel List Viterbi Algoriyhm-4
  • the decoding performance is approximately 0.2 to 0.8 dB higher than the VA decoding performance.
  • Step 303 Send a decoding result of the first substream, the second substream, and the third substream to the base station controller, where the decoding result of the first substream includes a decoding bit stream and a CRC check result.
  • the NodeB After completing the decoding process performed in step 302, the NodeB, specifically the CC Decoder in the NodeB, can send the decoding result to the base station controller, such as the RNC, so that the RNC can adopt the manner shown in FIG.
  • the decoding result is sent to the AMR Speech Decoder in the CN, and the CRC check result included in the decoding result of the first substream can be sent to the Outer-Loop Power Control in the RNC.
  • the subsequent implementation process is the same as the prior art. I will not repeat them here.
  • the base station may perform decoding processing on the first substream by using a CRC-assisted decision-based decoding algorithm, which is compared to the prior art using a common VA decoding algorithm for decoding processing.
  • the decoding performance of the first substream is improved, and the importance of the first substream is higher in the voice quality. Therefore, the performance of the first substream can be improved by improving the decoding performance of the first substream. Higher requirements for voice quality.
  • the CRC-assisted decision-based decoding algorithm used in the above embodiments may be a List Viterbi Algorithm (LVA), or a bit-reversed decoding algorithm.
  • the above embodiment may preferably be PL VA, or serial LVA (Serial LVA, hereinafter referred to as SLVA).
  • the PLVA decoder includes a VA decoder and a CRC Check&Choose module.
  • the VA decoder includes K candidate paths, namely Path1 ⁇ PathK:.
  • the A substream is input to the VA decoder.
  • the VA decoder can output K global optimal candidate paths Path1 ⁇ PathK by using the Viterbi algorithm.
  • the CRC Check&Choose module can correspond to the Path1 ⁇ PathK corresponding to the CRC included in the A substream.
  • the decoding result is respectively subjected to CRC check, and the decoding result with the correct CRC check result is selected as the final decoding result.
  • the decoding result corresponding to Path2 is selected as the final result, and the decoding results corresponding to Path1 ⁇ PathK cannot pass.
  • the CRC check that is, all the decoding results are erroneous, and the decoding result of the best path is output as the final decoding result, and the optimal path may be, for example, a preset Path1, and the optimal path is VA.
  • the CRC Check&Choose module can output the PLVA CRC Indicator (PLVA CRCI) and the PLVA decoded bit stream to the RNC.
  • PLVA CRC Indicator PLVA CRCI
  • the above embodiment may preferably employ PLVA-4.
  • the use of PLVA-4 is a compromise between current performance gain and complexity tradeoff. When the number of candidate paths K>4, the performance gain does not increase much, and the larger the ⁇ , the higher the probability of CRC miss detection.
  • PLVA-2, PLVA-6, PLVA-8, PLVA-12 or PLVA-16 can also be utilized in the above embodiments.
  • the decoding algorithm based on the CRC-assisted decision can also adopt other algorithms, such as SLVA, a bit-reversed decoding algorithm, etc., and the implementation principle is similar, and details are not described herein again.
  • the direct replacement of a conventional VA decoder in a base station with a CRC-based decision-based decoder can indeed improve
  • the first substream that is, the decoding performance of the A substream, but reduces the Mean Opinion Score (hereinafter referred to as MOS) of the speech.
  • MOS Mean Opinion Score
  • Target BLER a same target BLER (hereinafter referred to as: Target BLER) is preset for the A substream, the B substream, and the C substream, and once the decoding performance of the substream is improved, Then the BLER of the A substream is lower than the Target BLER set by the Outer-Loop Power Control. Therefore, the Outer-Loop Power Control needs to reduce the AMR power. However, once the AMR power is reduced, the final result is that the BLER of the A substream remains unchanged, but the BLER of the B and C substreams will increase, resulting in a decrease in the voice MOS score. Experiments show that the AMR power is reduced by 0.3 dB, MOS. The score is reduced by about 0.1 points.
  • the embodiment of the present invention provides three solutions, which are described in detail below.
  • Solution 1 Reduce the target block error rate of the outer loop power control.
  • Outer-Loop Power Control reduces the Target SINR of the A substream, while the BLER of the B and C substreams can remain unchanged. Therefore, the scheme does not require Outer-Loop Power Control to reduce the AMR power, so the MOS score of the voice is not reduced. .
  • the channel resources occupied by the first substream are reduced, and the channel resources occupied by the second substream and the third substream are increased.
  • the method may also be used by a person skilled in the art to reconfigure channel resources between the three substreams, as long as the channel resources occupied by the first substream are reduced.
  • the purpose of the channel resources occupied by the second substream and the third substream may be.
  • the scheme can reconfigure the rate matching parameters of the A, B, and C substreams, so that the rate matching parameter of the A substream is reduced, and the rate matching parameters of the B substream and the C substream are increased, thereby A part of the transmission resources of the substream are transferred to the B and C substreams, so that the three substreams reach a new balance under the PLVA, thereby avoiding the reduction of the MOS score.
  • the base station may use a decoding algorithm based on the CRC auxiliary decision to obtain decoding results on multiple candidate paths, and apply CRC included in the first substream to perform CRC on decoding results on multiple candidate paths. Verifying, obtaining the decoding result on the correct path of the CRC check result, and the CRC check result of the correct path and the CRC check result of the best path; then, the base station may send the decoding result on the correct path to the base station controller The CRC check result of the correct path and the check result of the best path, so that the base station controller sends the check result of the best path to the outer loop power control module, and sends the decoding result on the correct path to the core network. And sending a bad frame indication to the core network according to the CRC check result of the correct path.
  • FIG. 5 is a schematic structural diagram of another PLVA decoder used in the embodiment of the voice signal processing method of the present invention. As shown in FIG. 5, the PLVA decoder is shown. Compared with the PLVA decoder shown in FIG. 4, three outputs may be included, and the three outputs include VA CRCI, PLVA CRCI, and PLVA decoded bit stream, and the VA CRCI is the CRC check result of the optimal path.
  • the best path is a preset path, such as Pathl, and the VA CRCI is equivalent to the CRCI output by the VA decoder.
  • the PLVA CRCI is the CRC check result of the correct path, and the correct path may be optimal.
  • the path is the same.
  • the correct path and the best path are Pathl.
  • the PLVA CRCI is the same as the VA CRCI.
  • the correct path may be different from the best path. For example, if the correct path is Path2, the PLVA CRCI is the result of the Path2 check.
  • the PLVA decoding bit stream is the decoding result of the correct path.
  • the CRC Check&Choose module of the PLVA decoder also outputs the CRC check result of the best path, that is, the three outputs of the PLVA decoder are: As a result, the CRC check result of the best path and the CRC check result of the best path, at this time the PLVA decoder is equivalent to the VA decoder.
  • Outer-Loop Power Control still uses VA CRCI, while PLVA CRCI gives AMR Speech Decoder to indicate if a voice frame is available.
  • the result of the VA decoding is the same as the result of the PLVA decoding, and both CRC check results are positive.
  • the VA CRC check result is wrong
  • the PLVA CRC check result is correct
  • the PLVA outputs the correct decoding result if all candidate paths
  • the corresponding decoding results are all wrong, and the PLVA outputs the decoding result corresponding to the optimal path. Therefore, when the VA CRC check result is correct, the CRC check result of the PLVA is necessarily correct; on the contrary, the PLVA CRC check result is correct, but the VA CRC check result is not necessarily correct.
  • the NodeB can pass the VA CRCI and the PLVA CRCI to the RNC.
  • the RNC can use the VA CRCI for outer loop power control.
  • the BFI can be sent to the CN to indicate whether the corresponding voice frame is correct.
  • the NodeB can also use the PL sub-decoded A substream and the VA decoded B and C.
  • the substream is framing and sent to the AMR Speech Decoder.
  • the AMR Speech Decoder can perform speech decoding based on receiving an AMR speech signal including three substreams and a corresponding BFI indication.
  • RNC's Outer-Loop Power Control module can be used for power control based on the VA CRCI output from the PLVA. It can be implemented by existing technology and will not be described here.
  • FIG. 6 is a flowchart of the second embodiment of the voice signal processing method of the present invention. As shown in FIG. 6, the method in this embodiment is used to implement the foregoing solution 1. The method in this embodiment may include:
  • Step 601 Receive an encoded voice signal sent by the UE, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a CRC.
  • Step 602 Perform decoding processing on the first substream, the second substream, and the third substream by using a decoding algorithm, where the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision deal with.
  • Step 603 Send a decoding result of the first substream, the second substream, and the third substream to a base station controller, where the decoding result of the first substream includes a decoded bitstream and a CRC check result. .
  • steps 601 to 603 are similar to the implementations of steps 301 to 303 in the method embodiment shown in FIG. 3, and details are not described herein again.
  • Step 604 Receive a reduced target signal to noise ratio sent by the base station controller.
  • Step 605 Perform inner loop power control according to the target signal to noise ratio.
  • Step 604 and step 605 may specifically be Inner-Loop Power in NodeB.
  • the Control module executes.
  • the Outer-Loop Power Control can reduce the Target SINR of the first substream, and the BLER of the second and third substreams can remain unchanged.
  • the solution does not require Outer-Loop Power Control to reduce the AMR power, so it does not reduce the MOS score of the voice.
  • this embodiment does not need to modify the product code, and only needs to modify the Target BLER of the Outer-Loop Power Control, which is easy to implement.
  • FIG. 7 is a flowchart of Embodiment 3 of a method for processing a voice signal according to the present invention. As shown in FIG. 7, the method in this embodiment may include:
  • Step 701 Receive an encoded voice signal sent by the UE, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a CRC.
  • Step 702 Perform decoding processing on the first substream, the second substream, and the third substream by using a decoding algorithm, where the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision deal with.
  • Step 703 Send a decoding result of the first substream, the second substream, and the third substream to a base station controller, where the decoding result of the first substream includes a decoded bitstream and a CRC check result. .
  • steps 701 to 703 are similar to the implementations of steps 301 to 303 in the method embodiment shown in FIG. 3, and details are not described herein again.
  • Step 704 Decrease a rate matching parameter of the first substream, and increase a rate matching parameter of the second substream and the third substream.
  • the rate matching parameter of the first substream is decreased, and the rate matching parameters of the second substream and the third substream are increased, thereby A part of the transmission resources are transferred to the second and third substreams, so that the three substreams reach a new balance under the PLVA, thereby avoiding the reduction of the MOS score.
  • the technical solution has less change to the product and is simple to implement.
  • FIG. 8 is a flowchart of Embodiment 4 of a method for processing a voice signal according to the present invention
  • FIG. 9 is a schematic structural diagram of processing three substreams in Embodiment 4 of the method shown in FIG. 8.
  • the method in this embodiment may be Includes:
  • Step 801 Receive an AMR voice signal sent by the UE.
  • the AMR voice signal includes three substreams A, B, and C, namely Class A, Class B, and Class C, where the data block of Class A is followed by a CRC check, and the data block of the B and C substreams has no CRC check.
  • Step 802 Decode the A substream by using an LVA decoder, and decode the B substream and the C substream by using a VA decoder.
  • Step 803 the outer loop power control of the LVA decoder to the RNC (Outer-Loop Power)
  • Step 804 The LVA decoder sends the LVA CRCI to the AMR Speech Decoder through the RNC.
  • Step 805 The decoding result of the A substream decoded by the LVA decoder and the decoding result of the ⁇ C substream decoded by the two VA decoders are sent to the AMR speech decoder through the RNC.
  • PLVA-4 has a performance gain of about 0.3 dB with respect to VA.
  • a voice processing system with power control such as a WCDMA system, does not need to modify the Target BLER of the Outer-Loop Power Control, and does not need to modify the A, B, and C substreams.
  • the rate matching parameter but directly converts the performance gain brought by LVA into the gain of the speech MOS. Since the A substream is of the highest importance in AMR speech, this method maximizes speech performance while minimizing the impact on existing systems.
  • FIG. 10 is a flowchart of Embodiment 5 of a method for processing a voice signal according to the present invention. As shown in FIG. 10, the method in this embodiment may include:
  • Step 101 The decoding result of the first substream, the second substream, and the third substream sent by the base station is received, where the decoding result of the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision a decoding result obtained afterwards and the decoding result includes a decoding bit stream and a CRC check result;
  • Step 102 Send the verification result of the optimal path to the outer loop power control module, and send the decoding result and the CRC check result on the correct path to the core network.
  • This embodiment is a technical solution executed by the base station controller corresponding to the technical solution executed by the base station shown in FIG. 3, and the implementation principle thereof is described in detail in the description of the foregoing technical solution, and details are not described herein again.
  • the base station controller of this embodiment may be an RNC or a BSC.
  • the decoding algorithm based on the CRC-assisted decision in this embodiment may include PL VA, SLVA, etc., and details are not described herein again.
  • the base station controller may receive the decoded bit stream and the CRC check result after the base station decodes the first substream by using a decoding algorithm based on the CRC auxiliary decision, which is compared with the prior art.
  • the VA decoding algorithm performs the decoding process.
  • the decoding performance of the first substream can be improved, and since the first substream is of high importance in the voice quality, the embodiment can pass Improve the decoding performance of the first substream, improve the voice quality, and meet the higher requirements of the user for voice quality.
  • the method further includes: instructing the base station to reduce channel resources occupied by the first substream, and adding the second substream And the channel resources occupied by the third substream.
  • the method of this embodiment corresponds to the method described in the foregoing method, and the implementation principle and the technical effect are similar, and details are not described herein again.
  • the method further includes: reducing a target block error rate of the outer loop power control module, so that the outer loop power control module is The base station transmits the reduced target signal to noise ratio.
  • the method of this embodiment corresponds to the method described in the foregoing solution 2. The implementation principle and the technical effect are similar, and details are not described herein again.
  • FIG. 1 is a flowchart of Embodiment 6 of a method for processing a voice signal according to the present invention. As shown in FIG. 11, the method in this embodiment may include:
  • Step 201 The decoding result of the first substream, the second substream, and the third substream sent by the base station is received, where the decoding result of the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision
  • the decoded result obtained later includes the decoded bit stream and the CRC check result.
  • the CRC check result may include a CRC check result of the correct path and a CRC check result of the best path, and the decoded bit stream is a decoded bit stream on the correct path.
  • Step 202 Send a CRC check result of the best path to the outer loop power control module.
  • Step 203 Send the decoded bit stream on the correct path and the CRC check result of the correct path. Send it to the core network.
  • Step 204 Send the decoding result of the second substream and the third substream to the core network.
  • steps 202 to 204 There may be no sequential execution order between steps 202 to 204.
  • a voice processing system with power control such as a WCDMA system, does not need to modify the Target BLER of the Outer-Loop Power Control, and does not need to modify the rate matching parameter between the three substreams. Instead, the performance gain brought by the decoding algorithm based on the CRC auxiliary decision is directly converted into the gain of the speech MOS. Since the first substream is of the highest importance in AMR speech, this embodiment can maximize speech performance while minimizing the impact on existing systems.
  • the wideband AMR voice and the partial narrowband AMR voice have a bit number of 0 in the C substream.
  • the foregoing embodiment is only described by taking the narrowband AMR voice whose bit number of the C substream is not 0 as an example, those skilled in the art can understand that the technical solution of the embodiment of the present invention is also applicable to the broadband AMR voice and the C subframe.
  • the narrow-band AMR voice with a bit number of 0 is similar to the foregoing embodiment, and details are not described herein again.
  • FIG. 12 is a schematic structural diagram of Embodiment 1 of a base station according to the present invention.
  • the base station in this embodiment may include: a first receiving module 11, a decoding processing module 12, and a first sending module 13, where the first receiving The module 11 is configured to receive the encoded voice signal sent by the UE, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a CRC.
  • the first sending module 13 is configured to send, to the base station controller, a decoding result of the first substream, the second substream, and the third substream, where the decoding result of the first substream includes a decoded bit stream and CRC check result.
  • the base station in this embodiment may be used to perform the method in the method embodiment shown in FIG. 3, and the implementation principle and the technical effect are similar.
  • the base station in this embodiment may be a BTS, a NodeB, or an eNB.
  • FIG. 13 is a schematic structural diagram of Embodiment 2 of a base station according to the present invention. As shown in FIG. 13, the base station in this embodiment is based on the base station shown in FIG. 12, and further, the first receiving module 11 may include: a first receiving unit 111.
  • the decoding processing module 12 may include: a first decoding processing unit 121, a second decoding processing unit 122, and a third decoding processing unit 123
  • the first decoding processing unit 121 is configured to perform decoding processing on the first substream by using a parallel enumeration Viterbi decoding algorithm, obtain decoding results on multiple candidate paths, and apply the CRC to multiple strips.
  • the decoding result on the candidate path is subjected to CRC check, and the decoding result on the correct path of the CRC check result and the CRC check result of the correct path and the CRC check result of the best path are obtained, and the best path is adopted.
  • the maximum likelihood path determined by the Viterbi decoding algorithm; the second decoding processing unit 122 is configured to perform decoding processing on the second substream by using a Viterbi decoding algorithm to obtain a decoding result;
  • the first sending module 13 may include: a first sending unit 131, a second sending unit 132, and a third sending list, by using a Viterbi decoding algorithm to perform decoding processing on the third substream to obtain a decoding result. 133.
  • the first sending unit 131 is configured to send the decoding result on the correct path, the CRC check result of the correct path, and the check result of the best path obtained by the first decoding processing unit. Giving the base station controller, so that the base station controller sends the verification result of the best path to the outer loop power control module, and sends the decoding result and the CRC check result on the correct path to the core network;
  • the second sending unit 132 is configured to send the decoding result acquired by the second decoding processing unit to the base station controller, and the third sending unit 133 is configured to use the decoding result obtained by the third decoding processing unit. Sended to the base station controller.
  • the base station in this embodiment may be used to perform the technical solution described in the foregoing solution 3.
  • the method may be specifically implemented in the method embodiment shown in FIG. 8. The implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 14 is a schematic structural diagram of Embodiment 3 of a base station according to the present invention.
  • the base station of the present embodiment further includes: a channel resource control module 14 for reducing a The channel resources occupied by the first substream are increased, and the channel resources occupied by the second substream and the third substream are increased.
  • the base station in this embodiment may be used to perform the technical solution described in the foregoing solution 1.
  • the method in the embodiment of the method shown in FIG. 6 may be specifically implemented. The implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 15 is a schematic structural diagram of Embodiment 4 of a base station according to the present invention.
  • the base station of this embodiment further includes: an inner loop power control module based on the base station shown in FIG. Block 15 is configured to receive the reduced target signal to noise ratio sent by the base station controller, and perform inner loop power control according to the target signal to noise ratio.
  • the base station of this embodiment may be used to perform the technical solution described in the foregoing solution 2.
  • the method of the method embodiment shown in FIG. 7 may be specifically implemented, and the implementation principle and the technical effect are similar, and details are not described herein again.
  • FIG. 16 is a schematic structural diagram of Embodiment 1 of a base station controller according to the present invention.
  • the base station controller of this embodiment may include: a second receiving module 21 and a second sending module 22, where the second receiving module 21 is used.
  • the second sending module 22 is configured to send the CRC check result to the outer loop power control module, where The decoded bit stream and CRC check result of the first substream and the decoded result of the second substream and the third substream are sent to the core network.
  • the base station controller of this embodiment may be used to perform the technical solution of the method embodiment shown in FIG. 10, and the implementation principle and technical effects are similar, and details are not described herein again.
  • the CRC check result includes a CRC check result of a correct path and a CRC check result of an optimal path, and the decoded bit stream is a correct path.
  • the decoded bit stream, the optimal path is a maximum likelihood path determined by using a Viterbi decoding algorithm; the base station controller of this embodiment is based on the base station controller shown in FIG.
  • the second sending module 22 includes: a fourth sending unit 221 and a fifth sending unit 222, where the fourth sending unit 221 is configured to send a CRC check result of the best path to the outer loop power control module;
  • the fifth sending unit 222 is configured to send the decoded bit stream on the correct path and the CRC check result of the correct path to the core network, and send the decoded result of the second sub-stream and the third sub-stream to The core network.
  • the base station controller of this embodiment may be used to perform the technical solution described in the foregoing solution 3.
  • the technical solution shown in FIG. 11 may be specifically implemented, and the implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 18 is a schematic structural diagram of Embodiment 3 of a base station controller according to the present invention.
  • the base station controller of this embodiment further includes: an indication module 23, based on the base station controller shown in FIG. Instructing the base station to reduce channel resources occupied by the first substream, and The channel resources occupied by the second substream and the third substream are increased.
  • the base station controller of this embodiment may be used to perform the technical solution described in the foregoing solution 1.
  • the implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 19 is a schematic structural diagram of Embodiment 4 of a base station controller according to the present invention.
  • the base station controller of this embodiment further includes: a parameter control module 24, based on the base station controller shown in FIG. And a method for reducing a target block error rate of the outer loop power control module, so that the outer loop power control module sends the reduced target signal to noise ratio to the base station.
  • the base station controller of this embodiment may be used to perform the technical solution described in the foregoing solution 2.
  • the implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 20 is a schematic structural diagram of an embodiment of an access network system according to the present invention.
  • the access network system of this embodiment may include a base station 1 and a base station controller 2, where the base station 1 may adopt FIG. 12 to FIG.
  • the foregoing embodiment describes the process of processing the uplink AMR voice signal sent by the UE by the network side.
  • the process of processing the downlink AMR voice signal sent by the base station by the UE is described in detail below.
  • FIG. 21 is a flowchart of Embodiment 7 of a method for processing a voice signal according to the present invention. As shown in FIG. 21, the method in this embodiment may include:
  • Step 211 Receive an encoded voice signal sent by the base station, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a CRC.
  • Step 212 Decode the first substream, the second substream, and the third substream by using a decoding algorithm, and obtain decoding results of the first substream, the second substream, and the third substream, where
  • the decoding algorithm based on the CRC auxiliary decision decodes the first substream, and the decoding result of the first substream includes a decoded bit stream and a CRC check result.
  • Step 213 Perform decoding processing of the AM R speech signal according to the decoded bit stream of the first substream and the CRC check result and the decoding results of the second substream and the third substream.
  • the UE may receive the encoded voice signal sent by the NodeB, and the encoded voice signal may be an AMR voice signal.
  • the AMR voice signal can include three sub-streams A, B, and C. Corresponding to the first substream, the second substream, and the third substream, respectively.
  • the first substream that is, the A substream, contains a CRC.
  • the CC decoder in the UE may perform decoding processing on the first substream, the second substream, and the third substream, respectively, by using a decoding algorithm.
  • the decoding algorithm based on CRC-assisted decision can effectively improve the decoding performance of convolutional codes.
  • the basic principle is: output the global optimal multiple candidate paths through the Viterbi algorithm, and pass the CRC.
  • the CRC check is performed on the decoding results corresponding to the paths, and the correct decoding result of the CRC check is selected as the final result. If the decoding results corresponding to all the paths fail to pass the CRC check, the decoding of the best path is output. The result is the final result. Since this decoding algorithm can choose between multiple paths including the best path, performance is better than the normal Viterbi algorithm that only selects the best path.
  • the decoding performance is about 0.2 ⁇ 0.8dB higher than the VA decoding performance. .
  • the decoding algorithm based on the CRC auxiliary decision is used to decode the first substream, that is, the A substream. Since the decoding algorithm based on the CRC auxiliary decision requires the CRC auxiliary decision, and the second substream and the third substream do not contain the CRC, the second substream and the third substream, that is, the B substream and the C substream The stream can be decoded using a VA decoder in the prior art.
  • the UE may perform decoding processing on the first substream by using a decoding algorithm based on a CRC auxiliary decision, which is compared to the prior art using a common VA decoding algorithm for decoding processing.
  • the decoding performance of the first substream is improved, and the importance of the first substream is higher in the voice quality. Therefore, the performance of the first substream can be improved by improving the decoding performance of the first substream. Higher requirements for voice quality.
  • the CRC-assisted decision-based decoding algorithm used in the above embodiment may be an LVA, or a bit-reversed decoding algorithm or the like.
  • the above embodiment may preferably be PLVA, or SLVA.
  • PL VA can adopt PLVA including 2, 4, 6, 8, 12 or 16 candidate paths.
  • the inventors have found that the direct replacement of the ordinary VA decoder in the UE into the decoder based on the CRC-assisted decision, for example, directly replacing the PLVA decoder, can indeed improve The first substream, that is, the decoding performance of the A substream, but will drop Low voice MOS points.
  • the first substream that is, the decoding performance of the A substream, but will drop Low voice MOS points.
  • a same target BLER is preset for the A substream, the B substream, and the C substream, and if the decoding performance of the A substream becomes better, the Bstream of the A substream is low.
  • the Target BLER is set for the outer loop power control. Therefore, the outer loop power control needs to reduce the AMR power.
  • the embodiment of the present invention provides three solutions, which are described in detail below.
  • Solution 1 Reduce the target of outer loop power control BLER.
  • the outer loop power control can reduce the target signal to noise ratio of the A substream, and the BLER of the C substream can remain unchanged. Therefore, the scheme does not require outer loop power control to reduce the AMR power. Therefore, the MOS score of the voice will not be lowered.
  • the base station is instructed to reduce channel resources occupied by the first substream, and to increase channel resources occupied by the second substream and the third substream.
  • the base station may be instructed to reduce the rate matching parameter of the A substream, increase the rate matching parameter of the B substream and the C substream, reduce the channel resource occupied by the first substream, and increase the second substream and the
  • the channel resources occupied by the three substreams the person skilled in the art may also instruct the base station to use other means to reconfigure the channel resources between the three substreams, as long as the channel resources occupied by the first substream can be reduced, and the second subroutine is added.
  • the purpose of the channel resources occupied by the stream and the third substream is sufficient.
  • the solution may indicate that the base station reconfigures the channel resources occupied by the A, B, and C substreams, for example, reconfigures the rate matching parameter, so that the rate matching parameter of the A substream is reduced, and the rate matching parameters of the B substream and the C substream are reduced. Increase, thereby transferring part of the transmission resources of the A substream to the B and C substreams, so that the three substreams reach a new balance under the PLVA, thereby avoiding the reduction of the MOS score.
  • the UE may use a decoding algorithm based on a CRC auxiliary decision to obtain
  • the decoding result on the multiple candidate paths is used to perform CRC check on the decoding result on the multiple candidate paths by using the CRC included in the first substream, and obtain the decoding result on the correct path of the CRC check result and the correct path.
  • the CRC check result and the CRC check result of the best path wherein the best path is the maximum likelihood path determined by the Viterbi decoding algorithm; then, the UE can use the best path CRC check result for the outer ring
  • the power control performs decoding processing of the AMR voice signal according to the decoding result and the CRC check result on the correct path and the decoding result of the second substream and the third substream.
  • FIG. 22 is a flowchart of Embodiment 8 of a method for processing a voice signal according to the present invention
  • FIG. 23 is a schematic diagram of a structure of a UE in the method of the method shown in FIG. 22, and as shown in FIG. 22 and FIG. 23, the method in this embodiment may include:
  • Step 251 Receive an encoded voice signal sent by the base station, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a CRC;
  • Step 252 Perform a decoding process based on the CRC auxiliary decision, decode the first substream, obtain decoding results on the multiple candidate paths, and apply CRC to perform CRC check on the decoding results on the multiple candidate paths. Obtaining a decoding result on a correct path of the CRC check result, a CRC check result of the correct path, and a CRC check result of the best path, where the best path is a maximum likelihood path determined by a Viterbi decoding algorithm;
  • Step 253 Perform outer loop power control according to the CRC check result of the best path.
  • Step 254 Perform AM according to the decoding result and the CRC check result on the correct path and the decoding result of the second substream and the third substream. Decoding processing of R speech signals.
  • the PLVA decoder may include three outputs including VA CRCI, PLVA CRCI, and PLVA decoded bit stream, the VA CRCI. That is, the CRC check result of the best path, the optimal path is a preset path, such as Pathl, and the VA CRCI is equivalent to the CRCI output by the VA decoder, and the PLVA CRCI is the CRC of the correct path.
  • the correct path may be the same as the best path. For example, the correct path and the best path are both Pathl.
  • the PLVA CRCI is the same as the VA CRCI, and the correct path may be different from the best path.
  • the correct path is Path2.
  • the PLVA CRCI is the result of the check of Path2, and the PLVA decoded bit stream is positive.
  • the decoding result of the path is confirmed. If all the paths in the candidate path are not correct, the CRC Check&Choose module of the PLVA decoder also outputs the CRC check result of the best path, that is, the three outputs of the PLVA decoder are:
  • the CRC check result of the best path and the CRC check result of the best path at this time, the PLVA decoder is equivalent to the VA decoding cry.
  • the result of the decoding of the Bay' J VA is the same as the result of the PLVA decoding, and both CRC check results are correct; if the best path corresponds If the decoding result is incorrect, the decoding result corresponding to the other candidate paths is correct, the VA CRC check result is incorrect, the PLVA CRC check result is correct, and the PLVA outputs the correct decoding result; if the decoding results corresponding to all candidate paths are incorrect , PLVA outputs the decoding result corresponding to the best path. Therefore, when the VA CRC check result is correct, the PLVA CRC check result is necessarily correct; otherwise, the PLVA CRC check result is correct, but the VA CRC check result is not necessarily correct.
  • the receiving module of the UE may receive a downlink AMR voice signal that is sent by the base station and includes the A substream, the B substream, and the C substream, where the A substream includes a CRC.
  • the UE includes two types of decoders, one is a PLVA decoder, and the other is a VA decoder, wherein the PLVA decoder can decode the A substream, and the VA decoder can The B substream and the C substream are decoded. For the process of decoding the B substream and the C substream by using the VA decoder, the process can be implemented by using the prior art. The description will not be repeated here.
  • the decoded bit stream can be sent to the AMR voice in the UE.
  • the decoder performs decoding.
  • the process of decoding the A substream by the PLVA decoder will be described in detail below.
  • the A substream using the PLVA decoder of Figure 5, three decoding results, VA CRCI, PLVA CRCI, and the decoded bit stream can be output.
  • the PLVA CRCI and the decoded bit stream can be sent to the AMR speech decoder, wherein the PLVA CRCI can be used to indicate whether the speech frame is available, and the VA CRCI can be sent to the outer loop power control module in the UE for outer loop power control. .
  • the UE may framing the PL sub-decoded A substream and the VA decoded B and C substreams to the AMR speech decoder, and the AMR speech decoder may be included according to the received The AMR speech signals of the three substreams, and the corresponding BFI indications, are speech decoded.
  • a voice processing system with power control such as a WCDMA system
  • WCDMA power control
  • FIG. 24 is a schematic structural diagram of Embodiment 1 of a user equipment according to the present invention.
  • the UE in this embodiment may include: a receiving module 26, a first decoding processing module 27, and a second decoding processing module 28, where The receiving module 26 is configured to receive the encoded voice signal sent by the base station, where the encoded voice signal includes a first substream, a second substream, and a third substream, where the first substream includes a cyclic convolution check CRC;
  • the first decoding processing module 27 is configured to perform decoding processing on the first substream, the second substream, and the third substream by using a decoding algorithm, to obtain the first substream, the second substream, and the first a decoding result of the three substreams, wherein the first substream is decoded by using a decoding algorithm based on a CRC auxiliary decision, and the decoding result of the first substream includes a decoded bit stream and a CRC calibration
  • the second decoding processing module 28 is configured to perform
  • the receiving module 26 is a module for receiving a downlink signal sent by the base station in the UE
  • the first decoding processing module 27 is a module for performing convolutional code decoding processing in the UE
  • the second decoding processing module is 28, which is the AMR speech decoder in the UE.
  • the module in this embodiment may adopt a hardware implementation manner, for example, multiplexing corresponding modules in an existing UE structure, or may be implemented in software, for example, loading corresponding program code in a memory, or implementing a combination of software and hardware.
  • the first decoding processing module 27 and the second decoding processing module 28 can all be implemented in software
  • the receiving module 26 can be implemented in hardware.
  • the UE in this embodiment is used to perform the technical solution of the method embodiment shown in FIG. 21, and the principle and the technical effect are similar, and details are not described herein again.
  • Embodiment 2 of a user equipment according to the present invention. As shown in FIG. 25, this embodiment is based on the UE structure shown in FIG. 24, and further includes: an outer loop power control module. Block 29, where:
  • the receiving module 26 includes: a first receiving unit 261, configured to receive the first substream; a second receiving unit 262, configured to receive the second substream; and a third receiving unit 263, configured to receive the third Substream
  • the first decoding processing module 27 includes: a first decoding processing unit 271, configured to adopt
  • a decoding algorithm of the CRC auxiliary decision performing decoding processing on the first substream, acquiring decoding results on multiple candidate paths, and applying the CRC to perform CRC check on decoding results on multiple candidate paths, Obtaining a decoding result on a correct path of the CRC check result, and a CRC check result of the correct path and a CRC check result of the best path, where the best path is a maximum likelihood path determined by a Viterbi decoding algorithm; a second decoding processing unit 272, configured to perform decoding processing on the second substream by using a Viterbi decoding algorithm to obtain a decoding result, and a third decoding processing unit 273, configured to use a Viterbi decoding algorithm Decoding the third substream to obtain a decoding result;
  • the outer loop power control module 29 is configured to perform outer loop power control according to the CRC check result of the best path;
  • the second decoding processing module 28 is specifically configured to: according to the decoding result and the CRC check result on the correct path acquired by the first decoding processing unit 271, and the second acquired by the second decoding processing unit 272
  • the decoding result of the third substream acquired by the two substreams and the third decoding processing unit 273 performs decoding processing of the AMR speech signal.
  • three receiving units may be physically combined, and the second decoding processing unit 272 and the third decoding processing unit 273 may be implemented by using a VA decoder.
  • the UE of this embodiment is used to perform the technical solution of the method embodiment shown in FIG. 22.
  • the specific logical architecture is shown in the architecture shown in FIG. 23, and the implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 26 is a schematic structural diagram of Embodiment 3 of a user equipment according to the present invention. As shown in FIG. 26, this embodiment is based on the UE structure shown in FIG. 24, and further includes: an outer loop power control module 29 and an inner loop power. Control module 30, wherein:
  • the outer loop power control module 29 is configured to reduce a target block error rate of the outer loop power control, and send the reduced target signal to the inner loop power control module according to the reduced target block error rate. Noise ratio
  • the inner loop power control module 30 is configured to perform inner loop power control according to the reduced target signal to noise ratio and the measured signal to noise ratio.
  • the UE in this embodiment may be used to perform the foregoing solution 1.
  • the implementation principle and technical effects are similar, and details are not described herein again.
  • Figure 27 is a schematic structural diagram of Embodiment 4 of the user equipment of the present invention. As shown in Figure 27, this embodiment is based on the UE structure shown in Figure 24, and further includes: an indication module 31, configured to indicate the base station Reducing channel resources occupied by the first substream, and increasing channel resources occupied by the second substream and the third substream.
  • an indication module 31 configured to indicate the base station Reducing channel resources occupied by the first substream, and increasing channel resources occupied by the second substream and the third substream.
  • the UE in this embodiment may be used to perform the foregoing solution 2.
  • the implementation principle and technical effects are similar, and details are not described herein again.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Computer Security & Cryptography (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Error Detection And Correction (AREA)
  • Detection And Prevention Of Errors In Transmission (AREA)

Abstract

一种语音信号处理方法、装置和接入网系统,其中该方法包括接收用户设备UE发送的编码语音信号,所述编码语音信号包括第一子流、第二子流和第三子流,所述第一子流中包含循环卷积校验CRC(301),采用译码算法对所述第一子流、第二子流和第三子流进行译码处理,其中采用基于CRC辅助判决的译码算法对所述第一子流进行译码处理(302),向基站控制器发送所述第一子流、第二子流和第三子流的译码结果,所述第一子流的译码结果中包括译码比特流和CRC校验结果(303),通过提高第一子流的译码性能,提高语音质量,满足用户对语音质量的更高要求。

Description

语音信号处理方法、 装置和接入网系统
本申请要求于 2011 年 5 月 27 日 提交的 国 际申请号为 PCT/CN2011/074801、发明名称为"语音信号处理方法、装置和接入网系统" 的 PCT国际专利申请的优先权, 其全部内容通过引用结合在本申请中。
技术领域 本发明实施例涉及通信领域, 尤其涉及一种语音信号处理方法、 装置 和接人网系统。 背景技术
在现有通信系统, 例如通用移动通讯系统 ( Universal Mobile Telecommunications System, 以下简称: UMTS ) 中, 语音编码大量采用 卷积码作为信道编码, 并利用功率控制机制保障其语音质量。 图 1为现有 技术中语音编码过程的系统架构示意图, 如图 1所示, 以 UMTS网络举例 来说, 上行速率自适应(Adaptive Muti-Rate, 以下简称: AMR )语音信号 处理过程为, 用户设备( User Equipment, 以下简称: UE ) 中的 AMR语 音编码器(以下简称: AMR Speech Encoder ) 的语音编码经过卷积码编码 器 ( Convolutional Code Encoder, 以下简称: CC Encoder ) 采用卷积码进 行编码处理, CC Encoder编码后的 AMR语音信号通过空口发送给基站(以 下简称: NodeB ) , NodeB中的 CC译码器 (以下简称: CC Decoder ) 可 以对 AMR语音信号进行译码,该 CC Decoder包括两路输出,一路通过 Iub 口将译码后的比特流发送给无线网络控制器 (Radio Network Controller, 以下简称: RNC ) , RNC再通过 Iu口发送到核心网 ( Core Network, 以 下简称: CN )中的 AMR语音译码器(以下简称: AMR Speech Decoder ) , 另一路通过 Iub口将循环卷积校验 ( Cyclic Redundancy Check, 以下简称: CRC )校验结果, 即 CRC指示 (CRC Indicator, 以下简称: CRCI )发送 给 RNC, RNC即可根据该 CRCI通过 Iu口向 CN中的 AMR Speech Decoder 发送坏帧指示( Bad Frame Indicator, 以下简称: BFI ) , CC Decoder还将 该 CRCI发送给 RNC 中的外环功率控制 (以下简称: Outer-Loop Power Control ) 。 AMR Speech Decoder在接收到译码后的比特流以及 BFI之后, 即可进行译码处理。而 Outer-Loop Power Control可以根据 CRCI调整目标 块误码率 ( Block Error Ratio, 以下简称: BLER ) , 并根据调整的 BLER 向 NodeB中的内环功率控制 (以下简称: Inner-Loop Power Control )发送 目标信噪比(以下简称: Target SINR ) , 内环功率控制根据测量的信噪比 (以下简称: Measured SINR )和 Target SINR向 UE的功率发射器(以下 简称: Power Transmitter )发送功率指令(以下简称: Power Commander ) , 以调整 UE的发射功率。 图 2为在图 1所示系统架构中处理三个子流的结 构示意图, 如图 2所示, 在现有技术中, AMR语音信号可以分为 A、 B、 C三个子流, 即 Class A、 Class B和 Class C, 其中, A子流对语音质量影 响最大, 也最重要, 其数据块后附有 12比特的 CRC校验, B、 C子流的 重要性相对较低, 数据块之后没有 CRC校验。 NodeB中的 CC Decoder均 采用维特比算法 ( Viterbi Algorithm, 以下简称: VA )译码器, 而 VA译 码器的译码结果中, 只有 A子流存在 CRCI。
但是, 在实现本发明过程中, 发明人发现现有技术中 NodeB 对上行 AMR语音信号中 A子流的卷积码译码性能较低,或者 UE对下行 AMR语 音信号中的 A子流的卷积码译码性能较低, 对语音质量的影响较大, 无法 满足用户对语音质量的更高要求。 发明内容 本发明实施例提供一种语音信号处理方法、 装置和接入网系统, 以提 高对 A子流的卷积码译码性能。
本发明实施例提供一种语音信号处理方法, 包括:
接收用户设备 UE发送的编码语音信号, 所述编码语音信号包括第一 子流、 第二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC; 采用译码算法对所述第一子流、 第二子流和第三子流进行译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理; 向基站控制器发送所述第一子流、 第二子流和第三子流的译码结果, 所述第一子流的译码结果中包括译码比特流和 CRC校验结果。 本发明实施例提供另一种语音信号处理方法, 包括:
接收基站发送的第一子流、 第二子流和第三子流的译码结果, 其中, 第一子流的译码结果为采用基于循环卷积校验 CRC辅助判决的译码算法 进行译码处理后获取的译码结果且该译码结果中包括译码比特流和 CRC 校验结果;
将所述 CRC校验结果发送给外环功率控制模块, 将所述第一子流的 译码比特流和 CRC校验结果以及所述第二子流和第三子流的译码结果发 送给核心网。
本发明实施例提供一种基站, 包括:
第一接收模块, 用于接收用户设备 UE发送的编码语音信号, 所述编 码语音信号包括第一子流、 第二子流和第三子流, 所述第一子流中包含循 环卷积校验 CRC;
译码处理模块, 用于采用译码算法对所述第一子流、 第二子流和第三 子流进行译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第 一子流进行译码处理;
第一发送模块, 用于向基站控制器发送所述第一子流、 第二子流和第 三子流的译码结果, 所述第一子流的译码结果中包括译码比特流和 CRC 校验结果。
本发明实施例提供一种基站控制器, 包括:
第二接收模块, 用于接收基站发送的第一子流、 第二子流和第三子流 的译码结果, 其中, 第一子流的译码结果为采用基于循环卷积校验 CRC 辅助判决的译码算法进行译码处理后获取的译码结果且该译码结果中包 括译码比特流和 CRC校验结果; 将所述第一子流的译码比特流和 CRC校验结果以及所述第二子流和第三 子流的译码结果发送给核心网。
本发明实施例提供一种接入网系统, 包括: 基站和基站控制器; 其中, 基站采用上述的基站, 基站控制器采用上述的基站控制器。
本发明实施例还提供另一种语音信号处理方法, 包括:
接收基站发送的编码语音信号, 所述编码语音信号包括第一子流、 第 二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC; 采用译码算法对所述第一子流、 第二子流和第三子流进行译码处理, 获取所述第一子流、 第二子流和第三子流的译码结果, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理,所述第一子流的 译码结果中包括译码比特流和 CRC校验结果;
根据所述第一子流的译码比特流和 CRC校验结果以及所述第二子流 和第三子流的译码结果进行 AM R语音信号的译码处理。
本发明实施例还提供一种用户设备, 包括:
接收模块, 用于接收基站发送的编码语音信号, 所述编码语音信号包 括第一子流、 第二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC;
第一译码处理模块, 用于采用译码算法对所述第一子流、 第二子流和 第三子流进行译码处理, 获取所述第一子流、 第二子流和第三子流的译码 结果, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进行译 码处理, 所述第一子流的译码结果中包括译码比特流和 CRC校验结果; 第二译码处理模块, 用于根据所述第一子流的译码比特流和 CRC校 验结果以及所述第二子流和第三子流的译码结果进行 AMR语音信号的译 码处理。
本发明实施例中, 基站可以采用基于 CRC辅助判决的译码算法对所 述第一子流进行译码处理, 相比于现有技术中采用普通的 VA译码算法进 行译码处理来说, 可以提高对第一子流的译码性能, 而由于第一子流在语 音质量中的重要性较高, 因此, 本发明实施例可以通过提高第一子流的译 码性能, 提高语音质量, 满足用户对语音质量的更高要求。
本发明上述实施例中, UE可以采用基于 CRC辅助判决的译码算法对 所述第一子流进行译码处理, 相比于现有技术中采用普通的 VA译码算法 进行译码处理来说, 可以提高对第一子流的译码性能, 而由于第一子流在 语音质量中的重要性较高, 因此, 本发明实施例可以通过提高第一子流的 译码性能, 提高语音质量, 满足用户对语音质量的更高要求。 附图说明 实施例或现有技术描述中所需要使用的附图作一简单地介绍, 显而易见 地, 下面描述中的附图是本发明的一些实施例, 对于本领域普通技术人员 来讲, 在不付出创造性劳动性的前提下, 还可以根据这些附图获得其他的 附图。
图 1为现有技术中语音编码过程的系统架构示意图;
图 2为在图 1所示系统架构中处理三个子流的结构示意图; 图 3为本发明语音信号处理方法实施例一的流程图;
图 4为本发明语音信号处理方法实施例中所使用的 PLVA的一种结构 示意图;
图 5为本发明语音信号处理方法实施例所使用的另一种 PLVA译码器 的结构示意图;
图 6为本发明语音信号处理方法实施例二的流程图;
图 7为本发明语音信号处理方法实施例三的流程图;
图 8为本发明语音信号处理方法实施例四的流程图;
图 9为图 8所示方法实施例四中处理三个子流的结构示意图; 图 10为本发明语音信号处理方法实施例五的流程图;
图 1 1为本发明语音信号处理方法实施例六的流程图;
图 12为本发明基站实施例一的结构示意图;
图 13为本发明基站实施例二的结构示意图;
图 14为本发明基站实施例三的结构示意图;
图 15为本发明基站实施例四的结构示意图;
图 16为本发明基站控制器实施例一的结构示意图;
图 17为本发明基站控制器实施例二的结构示意图;
图 18为本发明基站控制器实施例三的结构示意图;
图 19为本发明基站控制器实施例四的结构示意图;
图 20为本发明接入网系统实施例的结构示意图;
图 21为本发明语音信号处理方法实施例七的流程图;
图 22为本发明语音信号处理方法实施例八的流程图;
图 23为图 22所示方法实施例中 UE的架构示意图; 图 24为本发明用户设备实施例一的结构示意图;
图 25为本发明用户设备实施例二的结构示意图;
图 26为本发明用户设备实施例三的结构示意图;
图 27为本发明用户设备实施例四的结构示意图。 具体实施方式 为使本发明实施例的目的、 技术方案和优点更加清楚, 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, 本领域普通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。
本发明的技术方案, 可以应用于各种通信系统, 例如: 全球移动通信 系统 ( Global System for Mobile Communications, 以下简称: GSM ) , 码 分多址 ( Code Division Multiple Access , 以下简称: CDMA ) 2000系统, 宽带码分多址 (Wideband Code Division Multiple Access , 以下简称: WCDMA ) 系统, 长期演进( Long Term Evolution, 以下简称: LTE ) 系 统等。 但为描述方便, 下述实施例以 WCDMA为例进行说明。
基站, 可以是 GSM 或者 CDMA2000 中的基站 (Base Transceiver Station, 以下简称: BTS ) , 也可以是 WCDMA中的基站 NodeB , 还可以 是 LTE中的演进型基站( Evolutional Node B ,以下简称: eNB或 eNodeB ) , 本发明并不限定, 但为描述方便, 下述实施例以 NodeB为例进行说明。
基站控制器, 可以是 GSM或者 CDMA2000 中的基站控制器 (Base Station Controller, 以下简称: BSC ) , 也可以是 WCDMA中的 RNC, 本 发明并不限定, 但为描述方便, 下述实施例以 RNC为例进行说明。
图 3为本发明语音信号处理方法实施例一的流程图, 如图 3所示, 本 实施例的方法是对图 1中 NodeB的 CC Decoder所执行的方法的改进, 本 实施例的方法可以包括:
步骤 301、 接收 UE发送的编码语音信号, 所述编码语音信号包括第 一子流、 第二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC。
举例来说, NodeB, 可以接收 UE发送的编码语音信号, 该编码语音 信号即可为图 1中经过 CC Encoder编码处理后的 AMR语音信号。该 AMR 语音信号即可包括图 2中的 A、 B、 C三个子流, 分别对应第一子流、 第 二子流和第三子流。 其中第一子流, 即 A子流中包含 CRC。
步骤 302、 采用译码算法对所述第一子流、 第二子流和第三子流进行 译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进 行译码处理。
NodeB, 具体地可以是 NodeB中的 CC Decoder, 可以采用译码算法 分别对所述第一子流、 第二子流和第三子流进行译码处理。 为了提升现有 技术中 NodeB对第一子流, 即 A子流的卷积码译码性能, 本实施例采用 基于 CRC辅助判决的译码算法对第一子流, 即 A子流进行译码处理。 由 于基于 CRC辅助判决的译码算法需要 CRC的辅助判决, 而第二子流和第 三子流中均不包含 CRC, 因此, 第二子流和第三子流, 即 B子流和 C子 流可以采用现有技术中的 VA译码器进行译码处理。
具体来说, 发明人经过研究发现, 基于 CRC辅助判决的译码算法可 以有效提高卷积码的译码性能, 其基本原理是: 通过 Viterbi 算法输出全 局最优的多条候选的路径, 通过 CRC对这些路径对应的译码结果分别进 行 CRC校验, 选择 CRC校验正确的译码结果作为最终结果, 如果所有路 径对应的译码结果都无法通过 CRC校验, 则输出最佳路径的译码结果作 为最终结果。 由于这种译码算法可以在包括最佳路径在内的多条路径间做 选择, 因此性能比只选择最佳路径的普通 Viterbi 算法性能更好。 通过研 究和仿真, 在 1%的 BLER条件下, 以基于 CRC辅助判决的译码算法为 4 条候选路径的并行列举维特比译码算法( Parallel List Viterbi Algoriyhm-4 , 以下简称: PLVA-4 )为例来说,译码性能比 VA译码性能大致高 0.2〜0.8dB。
步骤 303、 向基站控制器发送第一子流、 第二子流和第三子流的译码 结果, 该第一子流的译码结果中包括译码比特流和 CRC校验结果。
在完成步骤 302所执行的译码处理后, NodeB, 具体地可以是 NodeB 中的 CC Decoder即可将译码结果发送给基站控制器, 例如 RNC, 从而使 得 RNC可以采用图 2所示的方式将译码结果发送给 CN中的 AMR Speech Decoder, 而第一子流的译码结果中包含的 CRC 校验结果则可以发送到 RNC中的 Outer-Loop Power Control 后续的实现过程与现有技术相同, 此处不再赘述。
本实施例中, 基站可以采用基于 CRC辅助判决的译码算法对所述第 一子流进行译码处理, 相比于现有技术中采用普通的 VA译码算法进行译 码处理来说, 可以提高对第一子流的译码性能, 而由于第一子流在语音质 量中的重要性较高, 因此, 本实施例可以通过提高第一子流的译码性能, 提高语音质量, 满足用户对语音质量的更高要求。
进一步的, 上述实施例中所使用的基于 CRC辅助判决的译码算法可 以为列举维特比译码算法( List Viterbi Algorithm, 以下简称: LVA ) , 或 者比特反转译码算法等。 进一步地, 上述实施例可以优选 PL VA, 或者, 串行 LVA ( Serial LVA, 以下简称: SLVA ) 。
图 4为本发明语音信号处理方法实施例中所使用的 PLVA的一种结构 示意图, 如图 4所示, 该 PLVA译码器包括 VA译码器和 CRC校验和选 择(CRC Check&Choose )模块, 其中 VA译码器包括 K个候选路径, 即 Pathl〜PathK:。 A子流输入到 VA译码器中, VA译码器采用 Viterbi算法可 以输出 K条全局最优的候选路径 Pathl〜PathK, CRC Check&Choose模块 可以通过 A子流中包含的 CRC对 Pathl〜PathK对应的译码结果分别进行 CRC校验, 并选择 CRC校验结果正确的译码结果作为最终译码结果, 例 如选择 Path2对应的译码结果作为最终结果, 如果 Pathl〜PathK对应的译 码结果都不能通过 CRC校验, 也即所有译码结果都是错误的, 则输出最 佳路径的译码结果作为最终译码结果, 该最佳路径例如可以为预先设定的 Pathl , 最佳路径即为采用 VA 算法确定的最大似然路径。 最后, CRC Check&Choose模块可以将 PLVA CRC指示 ( PLVA CRC Indicator, 以下 简称: PLVA CRCI ) 和 PLVA译码比特流输出给 RNC。
更进一步地, 上述实施例可以优选采用包括 PLVA-4。 采用 PLVA-4 是当前性能增益和复杂度权衡的一个折中, 当候选路径的条数 K>4以后, 性能增益增加不多, 而 Κ越大, 意味着 CRC漏检概率增加。 本领域技术 人员可以理解的是, PLVA-2, PLVA-6, PLVA- 8, PLVA-12或者 PLVA-16 也可以利用到上述实施例中。 另外, 本领域技术人员可以理解的是, 基于 CRC辅助判决的译码算法还可以采用其它算法, 例如 SLVA, 比特反转译 码算法等, 其实现原理类似, 此处不再赘述。 在上述图 3 所示实施例的基础上, 发明人发现, 在将基站中普通的 VA译码器直接替换成基于 CRC 辅助判决的译码器, 例如直接替换成 PLVA译码器, 确实能够提高第一子流, 即 A子流的译码性能, 但是会降 低语音的平均主观分(Mean Opinion Score, 以下简称: MOS ) 。 具体来 说, Outer-Loop Power Control中针对 A子流、 B子流和 C子流均预先设 定一相同的目标 BLER (以下简称: Target BLER ) , 一旦 Α子流的译码 性能变好,则 A子流的 BLER低于 Outer-Loop Power Control设定的 Target BLER, 因此, Outer-Loop Power Control需要将 AMR功率降下去。 但是, 一旦 AMR功率降低, 其最终结果是 A子流的 BLER保持不变, 但 B、 C 子流的 BLER则会升高, 从而导致语音 MOS分下降, 实验表明, AMR功 率降低 0.3dB, MOS分约降低 0.1分。
为了在采用图 3所示实施例的技术方案时, 避免降低 MOS分, 本发 明实施例提供了三种解决方案, 下面对这三种方案进行详细说明。
方案一、 降低外环功率控制的目标块误码率。
通过降低 Outer-Loop Power Control 的 Target BLER , 可以使得
Outer-Loop Power Control降低 A子流的 Target SINR,而 B、 C子流的 BLER 则可维持不变, 因此, 该方案无需 Outer-Loop Power Control降低 AMR功 率, 因此, 不会降低语音的 MOS分。
本方案的优点在于:不用修改产品代码,只需要修改 Outer-Loop Power Control的 Target BLER。
方案二、 减少第一子流所占用的信道资源, 增加第二子流和第三子流 所占用的信道资源。
在具体实现时, 可以通过减小 A子流的速率匹配参数, 增大 B子流 和 C子流的速率匹配参数来减少所述第一子流所占用的信道资源,增加所 述第二子流和第三子流所占用的信道资源, 本领域技术人员也可以采用其 它手段来重新配置三个子流之间的信道资源, 只要能够达到减少所述第一 子流所占用的信道资源, 增加所述第二子流和第三子流所占用的信道资源 的目的即可。
本方案可以通过重新配置 A、 B、 C子流的速率匹配参数, 使得 A子 流的速率匹配参数减小, B子流和 C子流的速率匹配参数增大, 从而将 A 子流的一部分传输资源转移到 B、 C子流, 从而使得三个子流在 PLVA下 达到新的平衡, 从而避免降低 MOS分。
本方案的优点在于: 对产品改动小。
方案三: 采用双 CRC上报的技术方案。
在该技术方案中, 基站可以采用基于 CRC辅助判决的译码算法, 获 取多条候选路径上的译码结果, 应用第一子流中包含的 CRC对多条候选 路径上的译码结果进行 CRC校验, 获取 CRC校验结果正确路径上的译码 结果以及该正确路径的 CRC校验结果和最佳路径的 CRC校验结果;然后, 基站可以向基站控制器发送正确路径上的译码结果、 正确路径的 CRC校 验结果以及最佳路径的校验结果, 以使基站控制器将最佳路径的校验结果 发送给外环功率控制模块、 将正确路径上的译码结果发送给核心网并根据 正确路径的 CRC校验结果向核心网发送坏帧指示。
以基于 CRC辅助判决的译码算法为 PLVA举例来说, 图 5为本发明 语音信号处理方法实施例所使用的另一种 PLVA译码器的结构示意图,如 图 5所示, PLVA译码器相对于图 4所示的 PLVA译码器来说, 可以包括 三路输出, 这三路输出包括 VA CRCI、 PLVA CRCI以及 PLVA译码比特 流, 该 VA CRCI即为最佳路径的 CRC校验结果, 该最佳路径为预先设定 的路径, 例如 Pathl , 而该 VA CRCI与 VA译码器输出的 CRCI是等同的, 该 PLVA CRCI即为正确路径的 CRC校验结果,正确路径可能与最佳路径 相同,例如正确路径和最佳路径均为 Pathl ,此时 PLVA CRCI与 VA CRCI 相同, 正确路径也可能与最佳路径不同, 例如, 正确路径为 Path2, 则该 PLVA CRCI为 Path2的校验结果, PLVA译码比特流则为正确路径的译码 结果。 如果候选路径中所有路径都不正确, 则 PLVA 译码器的 CRC Check&Choose模块也输出最佳路径的 CRC校验结果, 也即, PLVA译码 器的三路输出分别为: 最佳路径的译码结果, 最佳路径的 CRC校验结果 以及最佳路径的 CRC校验结果, 此时 PLVA译码器等价于 VA译码器。 Outer-Loop Power Control仍然使用 VA CRCI,而 PLVA CRCI则送给 AMR Speech Decoder, 用于指示语音帧是否可用。
由上述 PLVA的原理描述可知, 如果最佳路径对应的译码结果正确, 则 VA译码的结果和 PLVA译码的结果相同, 两个 CRC校验结果均为正 确; 如果最佳路径对应的译码结果错误, 其他候选路径对应的译码结果正 确, 则 VA CRC校验结果错误, PLVA CRC校验结果正确, PLVA输出正 确的译码结果; 如果所有候选路径对应的译码结果均错误, PLVA输出最 佳路径对应的译码结果。 因此, VA CRC校验结果正确的时候, PLVA的 CRC校验结果必然正确; 反之, PLVA CRC校验结果正确, 但 VA CRC 校验结果不一定正确。
然后, 通过 NodeB和 RNC之间的 Iub接口, NodeB可以将 VA CRCI 和 PLVA CRCI传给 RNC。 RNC可以利用 VA CRCI进行外环功率控制, 根据 PLVA CRCI, 则可以向 CN发送 BFI指示对应的语音帧是否正确, NodeB还可以将 PLVA译码后的 A子流和 VA译码得到的 B、 C子流进行 组帧, 送给 AMR Speech Decoder。
AMR Speech Decoder可以根据收到包括三个子流的 AMR语音信号、 以及对应的 BFI指示进行语音解码。 而 RNC的 Outer-Loop Power Control 模块则可以根据 PLVA输出的 VA CRCI进行功率控制, 此处可采用现有 技术实现, 不再赘述。
下面采用三个具体实施例, 对上述三个技术方案进行详细说明。
图 6为本发明语音信号处理方法实施例二的流程图, 如图 6所示, 本 实施例的方法用于实现上述方案一, 本实施例的方法可以包括:
步骤 601、 接收 UE发送的编码语音信号, 所述编码语音信号包括第 一子流、 第二子流和第三子流, 所述第一子流中包含 CRC。
步骤 602、 采用译码算法对所述第一子流、 第二子流和第三子流进行 译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进 行译码处理。
步骤 603、 向基站控制器发送所述第一子流、 第二子流和第三子流的 译码结果, 所述第一子流的译码结果中包括译码比特流和 CRC校验结果。
上述步骤 601〜步骤 603与图 3所示方法实施例中的步骤 301〜步骤 303 的实现原理类似, 此处不再赘述。
步骤 604、 接收基站控制器发送的降低后的目标信噪比。
步骤 605、 根据所述目标信噪比, 进行内环功率控制。
步骤 604和步骤 605具体来说可以为 NodeB 中的 Inner-Loop Power Control模块执行的。
本实施例, 通过降低 Outer-Loop Power Control的 Target BLER, 可以 使得 Outer-Loop Power Control降低第一子流的 Target SINR, 而第二、 第 三子流的 BLER则可维持不变,因此,该方案无需 Outer-Loop Power Control 降低 AMR功率, 因此, 不会降低语音的 MOS分。 而且, 本实施例不用 修改产品代码, 只需要修改 Outer-Loop Power Control的 Target BLER, 实 现容易。
图 7为本发明语音信号处理方法实施例三的流程图, 如图 7所示, 本 实施例的方法可以包括:
步骤 701、 接收 UE发送的编码语音信号, 所述编码语音信号包括第 一子流、 第二子流和第三子流, 所述第一子流中包含 CRC。
步骤 702、 采用译码算法对所述第一子流、 第二子流和第三子流进行 译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进 行译码处理。
步骤 703、 向基站控制器发送所述第一子流、 第二子流和第三子流的 译码结果, 所述第一子流的译码结果中包括译码比特流和 CRC校验结果。
上述步骤 701〜步骤 703与图 3所示方法实施例中的步骤 301〜步骤 303 的实现原理类似, 此处不再赘述。
步骤 704、 减小第一子流的速率匹配参数, 增大第二子流和第三子流 的速率匹配参数。
本实施例可以通过重新配置三个子流的速率匹配参数, 使得第一子流 的速率匹配参数减小, 第二子流和第三子流的速率匹配参数增大, 从而将 第一子流的一部分传输资源转移到第二、 第三子流, 从而使得三个子流在 PLVA下达到新的平衡, 从而避免降低 MOS分。 该技术方案对产品的改 动较小, 实现简单。
图 8为本发明语音信号处理方法实施例四的流程图, 图 9为图 8所示 方法实施例四中处理三个子流的结构示意图, 如图 8和 9所示, 本实施例 的方法可以包括:
步骤 801、 接收 UE发送的 AMR语音信号。
该 AMR语音信号包括 A、 B、 C三个子流, 即 Class A、 Class B和 Class C, 其中 Class A的数据块后附有 CRC校验, B、 C子流的数据块之 后没有 CRC校验。
步骤 802、 采用 LVA译码器对 A子流进行译码处理, 采用 VA译码 器对 B子流和 C子流进行译码处理。
步骤 803、 LVA译码器向 RNC中的外环功率控制( Outer-Loop Power
Control )发送 VA CRCI。
步骤 804、 LVA译码器通过 RNC向 AMR语音译码器 ( AMR Speech Decoder )发送 LVA CRCI。
步骤 805、 LVA译码器译码后的 A子流的译码结果和两个 VA译码器 译码后的 ^ C子流的译码结果通过 RNC发送给 AMR语音译码器。
需要说明的是, 步骤 803〜步骤 804之间可以没有先后顺序。
发明人采用上述技术方案了系统仿真, 从仿真结果可知, 对于 AMR 12.2k业务的 A子流, PLVA-4相对于 VA, 大约有 0.3dB的性能增益。 通 过双 CRC上报的方案, 在 BLER=1%的时候, 可以得到 0.1分的 MOS分 增益, 在 BLER=10%的时候, 可以获得 0.35分的 MOS分增益。 当系统中 BLER越大, 由 PLVA带来的 MOS分增益也越大。
本实施例, 通过采用双 CRC 的方案, 使得带有功率控制的语音处理 系统, 例如 WCDMA系统等, 不需要修改 Outer-Loop Power Control的 Target BLER, 也不需要修改 A、 B、 C子流间的速率匹配参数, 而是直接 将 LVA带来的性能增益转化为语音 MOS分的增益。 由于 A子流在 AMR 语音中重要性最高, 该方法能在最大程度上提升语音性能, 同时对现有系 统影响最小。
图 10为本发明语音信号处理方法实施例五的流程图, 如图 10所示, 本实施例的方法可以包括:
步骤 101、 接收基站发送的第一子流、 第二子流和第三子流的译码结 果, 其中, 第一子流的译码结果为采用基于 CRC辅助判决的译码算法进 行译码处理后获取的译码结果且该译码结果中包括译码比特流和 CRC校 验结果;
步骤 102、 将所述最佳路径的校验结果发送给外环功率控制模块, 将 所述正确路径上的译码结果和 CRC校验结果发送给核心网。 本实施例是图 3所示基站执行的技术方案相应的基站控制器执行的技 术方案,其实现原理已在上述技术方案的描述中详细说明,此处不再赘述。 本实施例的基站控制器可以为 RNC或者 BSC。本实施例中的基于 CRC辅 助判决的译码算法可以包括 PL VA、 SLVA等, 此处不再赘述。
本实施例中, 基站控制器可以接收基站采用基于 CRC辅助判决的译 码算法对第一子流进行译码处理后的译码比特流和 CRC校验结果, 相比 于现有技术中采用普通的 VA译码算法进行译码处理来说, 本实施例可以 提高对第一子流的译码性能, 而由于第一子流在语音质量中的重要性较 高, 因此, 本实施例可以通过提高第一子流的译码性能, 提高语音质量, 满足用户对语音质量的更高要求。
在本发明另一个实施例中, 在图 10所示方法实施例的步骤 102之后 还可以包括: 指示所述基站减少所述第一子流所占用的信道资源, 并增加 所述第二子流和第三子流所占用的信道资源。 该实施例的方法对应上述方 案一所述的方法, 其实现原理和技术效果类似, 此处不再赘述。
在本发明再一个实施例中, 在图 10所示方法实施例的步骤 102之后 还可以包括: 降低所述外环功率控制模块的目标块误码率, 以使所述外环 功率控制模块向所述基站发送降低后的目标信噪比。 该实施例的方法对应 上述方案二所述的方法, 其实现原理和技术效果类似, 此处不再赘述。
下面对上述方案三的具体实现过程进行详细说明。
图 1 1为本发明语音信号处理方法实施例六的流程图, 如图 1 1所示, 本实施例的方法可以包括:
步骤 201、 接收基站发送的第一子流、 第二子流和第三子流的译码结 果, 其中, 第一子流的译码结果为采用基于 CRC辅助判决的译码算法进 行译码处理后获取的译码结果且该译码结果中包括译码比特流和 CRC校 验结果。
其中, CRC校验结果可以包括正确路径的 CRC校验结果和最佳路径 的 CRC校验结果, 所述译码比特流为正确路径上的译码比特流。
步骤 202、 将最佳路径的 CRC校验结果发送给所述外环功率控制模 块。
步骤 203、将正确路径上的译码比特流和正确路径的 CRC校验结果发 送给核心网。
步骤 204、 将第二子流和第三子流的译码结果发送给核心网。
步骤 202〜步骤 204之间可以没有先后的执行顺序。
本实施例, 通过采用双 CRC 的方案, 使得带有功率控制的语音处理 系统, 例如 WCDMA系统等, 不需要修改 Outer-Loop Power Control的 Target BLER, 也不需要修改三个子流间的速率匹配参数, 而是直接将基 于 CRC辅助判决的译码算法带来的性能增益转化为语音 MOS分的增益。 由于第一子流在 AMR语音中重要性最高, 本实施例可以在最大程度上提 升语音性能, 同时对现有系统影响最小。
需要说明的是, 宽带 AMR语音与部分窄带 AMR语音, 其 C子流的 比特数为 0。 尽管上述实施例仅以 C子流的比特数不为 0的窄带 AMR语 音为例进行说明, 本领域技术人员可以理解的是, 本发明实施例的技术方 案也同样适用于宽带 AMR语音以及 C子流的比特数为 0的窄带 AMR语 音, 其实现原理与上述实施例类似, 此处不再赘述。
图 12为本发明基站实施例一的结构示意图, 如图 12所示, 本实施例 的基站可以包括:第一接收模块 11、译码处理模块 12和第一发送模块 13 , 其中, 第一接收模块 11 , 用于接收 UE发送的编码语音信号, 所述编码语 音信号包括第一子流、 第二子流和第三子流, 所述第一子流中包含 CRC; 译码处理模块 12用于采用译码算法对所述第一子流、 第二子流和第三子 流进行译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第一 子流进行译码处理; 第一发送模块 13 用于向基站控制器发送所述第一子 流、 第二子流和第三子流的译码结果, 所述第一子流的译码结果中包括译 码比特流和 CRC校验结果。
本实施例的基站可以用于执行图 3所示方法实施例的方法, 其实现原 理和技术效果类似,此处不再赘述,本实施例中的基站可以为 BTS、 NodeB 或者 eNB。
图 13为本发明基站实施例二的结构示意图, 如图 13所示, 本实施例 的基站在图 12所示基站的基础上,进一步地,第一接收模块 11可以包括: 第一接收单元 111、 第二接收单元 112和第三接收单元 113 , 其中, 第一 接收单元 111用于接收所述第一子流; 第二接收单元 112用于接收所述第 二子流; 第三接收单元 1 13用于接收所述第三子流; 译码处理模块 12可 以包括: 第一译码处理单元 121、 第二译码处理单元 122和第三译码处理 单元 123 , 其中, 第一译码处理单元 121用于采用并行列举维特比译码算 法对所述第一子流进行译码处理, 获取多条候选路径上的译码结果, 应用 所述 CRC对多条候选路径上的译码结果进行 CRC校验, 获取 CRC校验 结果正确路径上的译码结果以及该正确路径的 CRC校验结果和最佳路径 的 CRC校验结果, 所述最佳路径为采用维特比译码算法确定的最大似然 路径; 第二译码处理单元 122用于采用维特比译码算法对所述第二子流进 行译码处理, 获取译码结果; 第三译码处理单元 123用于采用维特比译码 算法对所述第三子流进行译码处理, 获取译码结果; 第一发送模块 13 可 以包括: 第一发送单元 131、 第二发送单元 132和第三发送单元 133 , 其 中, 第一发送单元 131用于将所述第一译码处理单元获取的所述正确路径 上的译码结果、 所述正确路径的 CRC校验结果以及最佳路径的校验结果 发送给基站控制器, 以使所述基站控制器将所述最佳路径的校验结果发送 给外环功率控制模块、 将所述正确路径上的译码结果和 CRC校验结果发 送给核心网; 第二发送单元 132用于将所述第二译码处理单元获取的译码 结果发送给所述基站控制器; 第三发送单元 133用于将所述第三译码处理 单元获取的译码结果发送给所述基站控制器。
本实施例的基站可以用于执行上述方案三所描述的技术方案, 其具体 可以执行图 8所示方法实施例的方法, 其实现原理和技术效果类似, 此处 不再赘述。
图 14为本发明基站实施例三的结构示意图, 如图 14所示, 本实施例 的基站在图 12所示基站的基础上, 进一步地, 还包括: 信道资源控制模 块 14, 用于减少所述第一子流所占用的信道资源, 增加所述第二子流和第 三子流所占用的信道资源。
本实施例的基站可以用于执行上述方案一所描述的技术方案, 其具体 可以执行图 6所示方法实施例的方法, 其实现原理和技术效果类似, 此处 不再赘述。
图 15为本发明基站实施例四的结构示意图, 如图 15所示, 本实施例 的基站在图 12所示基站的基础上, 进一步地, 还包括: 内环功率控制模 块 15 , 用于接收所述基站控制器发送的降低后的目标信噪比, 并根据所述 目标信噪比, 进行内环功率控制。
本实施例的基站可以用于执行上述方案二所描述的技术方案, 其具体 可以执行图 7所示方法实施例的方法, 其实现原理和技术效果类似, 此处 不再赘述。
图 16为本发明基站控制器实施例一的结构示意图, 如图 16所示, 本 实施例的基站控制器可以包括: 第二接收模块 21和第二发送模块 22, 第 二接收模块 21 用于接收基站发送的第一子流、 第二子流和第三子流的译 码结果, 其中, 第一子流的译码结果为采用基于循环卷积校验 CRC辅助 判决的译码算法进行译码处理后获取的译码结果且该译码结果中包括译 码比特流和 CRC校验结果; 第二发送模块 22用于将所述 CRC校验结果 发送给外环功率控制模块, 将所述第一子流的译码比特流和 CRC校验结 果以及所述第二子流和第三子流的译码结果发送给核心网。
本实施例的基站控制器可以用于执行上述图 10所示方法实施例的技 术方案, 其实现原理和技术效果类似, 此处不再赘述。
图 17为本发明基站控制器实施例二的结构示意图,如图 17所示, CRC 校验结果包括正确路径的 CRC校验结果和最佳路径的 CRC校验结果, 译 码比特流为正确路径上的译码比特流, 所述最佳路径为采用维特比译码算 法确定的最大似然路径; 本实施例的基站控制器在图 16所示的基站控制 器的基础上, 进一步, 所述第二发送模块 22包括: 第四发送单元 221和 第五发送单元 222, 其中, 第四发送单元 221用于将所述最佳路径的 CRC 校验结果发送给所述外环功率控制模块; 第五发送单元 222用于将所述正 确路径上的译码比特流和所述正确路径的 CRC校验结果发送给核心网, 将所述第二子流和第三子流的译码结果发送给所述核心网。
本实施例的基站控制器可以用于执行上述方案三所述的技术方案, 其 具体可以执行图 11 所示的技术方案, 其实现原理和技术效果类似, 此处 不再赘述。
图 18为本发明基站控制器实施例三的结构示意图, 如图 18所示, 本 实施例的基站控制器在图 16所示的基站控制器的基础上, 进一步地包括: 指示模块 23 , 用于指示所述基站减少所述第一子流所占用的信道资源, 并 增加所述第二子流和第三子流所占用的信道资源。
本实施例的基站控制器可以用于执行上述方案一所述的技术方案, 其 实现原理和技术效果类似, 此处不再赘述。
图 19为本发明基站控制器实施例四的结构示意图, 如图 19所示, 本 实施例的基站控制器在图 16所示的基站控制器的基础上, 进一步地包括: 参数控制模块 24, 用于降低所述外环功率控制模块的目标块误码率, 以使 所述外环功率控制模块向所述基站发送降低后的目标信噪比。
本实施例的基站控制器可以用于执行上述方案二所述的技术方案, 其 实现原理和技术效果类似, 此处不再赘述。
图 20为本发明接入网系统实施例的结构示意图, 如图 20所示, 本实 施例的接入网系统可以包括基站 1和基站控制器 2, 其中, 基站 1可以采 用图 12〜图 15任一所示的基站的结构, 其对应地可以执行图 3、 图 6〜8中 任一实施例所述的技术方案,基站控制器 2可以采用图 16〜图 19任一所示 的基站的结构, 其可以执行图 10或图 11所示的技术方案, 其实现原理和 技术效果类似, 此处不再赘述。
上述实施例描述了网络侧对 UE发送的上行 AMR语音信号进行处理 的过程, 下面详细描述 UE对基站发送的下行 AMR语音信号进行处理的 过程。
图 21为本发明语音信号处理方法实施例七的流程图, 如图 21所示, 本实施例的方法可以包括:
步骤 211、 接收基站发送的编码语音信号, 编码语音信号包括第一子 流、 第二子流和第三子流, 该第一子流中包含 CRC。
步骤 212、 采用译码算法对第一子流、 第二子流和第三子流进行译码 处理, 获取第一子流、 第二子流和第三子流的译码结果, 其中, 采用基于 CRC辅助判决的译码算法对第一子流进行译码处理,该第一子流的译码结 果中包括译码比特流和 CRC校验结果。
步骤 213、根据第一子流的译码比特流和 CRC校验结果以及第二子流 和第三子流的译码结果进行 AM R语音信号的译码处理。
具体来说, UE可以接收 NodeB发送的编码语音信号, 该编码语音信 号即可为 AMR语音信号。该 AMR语音信号即可包括 A、 B、 C三个子流, 分别对应第一子流、 第二子流和第三子流。 其中第一子流, 即 A子流中包 含 CRC。 UE中的 CC译码器, 可以采用译码算法分别对所述第一子流、 第二子流和第三子流进行译码处理。
具体来说, 发明人经过研究发现, 基于 CRC辅助判决的译码算法可 以有效提高卷积码的译码性能, 其基本原理是: 通过 Viterbi 算法输出全 局最优的多条候选的路径, 通过 CRC对这些路径对应的译码结果分别进 行 CRC校验, 选择 CRC校验正确的译码结果作为最终结果, 如果所有路 径对应的译码结果都无法通过 CRC校验, 则输出最佳路径的译码结果作 为最终结果。 由于这种译码算法可以在包括最佳路径在内的多条路径间做 选择, 因此性能比只选择最佳路径的普通 Viterbi 算法性能更好。 通过研 究和仿真, 在 1%的 BLER条件下, 以基于 CRC辅助判决的译码算法为 4 条候选路径的 PLVA-4 为例来说, 译码性能比 VA 译码性能大致高 0.2〜0.8dB。
因此, 为了提升 UE对第一子流, 即 A子流的卷积码译码性能, 本实 施例采用基于 CRC辅助判决的译码算法对第一子流, 即 A子流进行译码 处理。 由于基于 CRC辅助判决的译码算法需要 CRC的辅助判决, 而第二 子流和第三子流中均不包含 CRC, 因此, 第二子流和第三子流, 即 B子 流和 C子流可以采用现有技术中的 VA译码器进行译码处理。
本实施例中, UE可以采用基于 CRC辅助判决的译码算法对所述第一 子流进行译码处理, 相比于现有技术中采用普通的 VA译码算法进行译码 处理来说, 可以提高对第一子流的译码性能, 而由于第一子流在语音质量 中的重要性较高, 因此, 本实施例可以通过提高第一子流的译码性能, 提 高语音质量, 满足用户对语音质量的更高要求。
进一步的, 上述实施例中所使用的基于 CRC辅助判决的译码算法可 以为 LVA, 或者比特反转译码算法等。 进一步地, 上述实施例可以优选 PLVA, 或者 SLVA。 其中, PL VA可以采用包括 2条、 4条、 6条、 8条、 12条或者 16条候选路径的 PLVA。
在上述图 21 所示实施例的基础上, 发明人发现, 在将 UE 中普通的 VA译码器直接替换成基于 CRC 辅助判决的译码器, 例如直接替换成 PLVA译码器, 确实能够提高第一子流, 即 A子流的译码性能, 但是会降 低语音的 MOS分。 具体来说, 外环功率控制中针对 A子流、 B子流和 C 子流均预先设定一相同的目标 BLER, —旦 A子流的译码性能变好, 则 A 子流的 BLER低于外环功率控制设定的 Target BLER, 因此, 外环功率控 制需要将 AMR功率降下去。但是, 一旦 AMR功率降低, 其最终结果是 A 子流的 BLER保持不变, 但 C子流的 BLER则会升高, 从而导致语音 MOS分下降, 实验表明, AMR功率降低 0.3dB, MOS分约降低 0.1分。
为了在采用图 21所示实施例的技术方案时, 避免降低 MOS分, 本发 明实施例提供了三种解决方案, 下面对这三种方案进行详细说明。
方案一、 降低外环功率控制的目标 BLER。
通过降低外环功率控制的目标 BLER, 可以使得外环功率控制降低 A 子流的目标信噪比, 而 C子流的 BLER则可维持不变, 因此, 该方案 无需外环功率控制降低 AMR功率, 因此, 不会降低语音的 MOS分。
本方案的优点在于: 不用修改产品代码, 只需要修改外环功率控制的 目标 BLER。
方案二、 指示基站减少第一子流所占用的信道资源, 增加第二子流和 第三子流所占用的信道资源。
在具体实现时, 可以指示基站减小 A 子流的速率匹配参数, 增大 B 子流和 C子流的速率匹配参数来减少第一子流所占用的信道资源,增加第 二子流和第三子流所占用的信道资源, 本领域技术人员也可以指示基站采 用其它手段来重新配置三个子流之间的信道资源, 只要能够达到减少第一 子流所占用的信道资源, 增加第二子流和第三子流所占用的信道资源的目 的即可。
本方案可以通过指示基站重新配置 A、 B、 C子流所占用的信道资源, 例如重新配置速率匹配参数, 使得 A子流的速率匹配参数减小, B子流和 C子流的速率匹配参数增大, 从而将 A子流的一部分传输资源转移到 B、 C子流,从而使得三个子流在 PLVA下达到新的平衡,从而避免降低 MOS 分。
本方案的优点在于: 对产品改动小。
方案三: 采用双 CRC上报的技术方案。
在该技术方案中, UE可以采用基于 CRC辅助判决的译码算法, 获取 多条候选路径上的译码结果, 应用第一子流中包含的 CRC对多条候选路 径上的译码结果进行 CRC校验, 获取 CRC校验结果正确路径上的译码结 果以及该正确路径的 CRC校验结果和最佳路径的 CRC校验结果, 其中, 最佳路径为采用维特比译码算法确定的最大似然路径; 然后, UE 可以采 用最佳路径的 CRC校验结果进行外环功率控制, 根据所述正确路径上的 译码结果和 CRC校验结果以及所述第二子流和第三子流的译码结果进行 AMR语音信号的译码处理。
下面采用一个具体的实施例, 对上述方案三进行详细说明。
图 22为本发明语音信号处理方法实施例八的流程图, 图 23 为图 22 所示方法实施例中 UE的架构示意图, 如图 22和 23所示, 本实施例的方 法可以包括:
步骤 251、 接收基站发送的编码语音信号, 该编码语音信号包括第一 子流、 第二子流和第三子流, 所述第一子流中包含 CRC;
步骤 252、 采用基于 CRC辅助判决的译码算法, 对第一子流进行译码 处理, 获取多条候选路径上的译码结果, 应用 CRC对多条候选路径上的 译码结果进行 CRC校验, 获取 CRC校验结果正确路径上的译码结果以及 该正确路径的 CRC校验结果和最佳路径的 CRC校验结果, 最佳路径为采 用维特比译码算法确定的最大似然路径;
步骤 253、 根据最佳路径的 CRC校验结果进行外环功率控制; 步骤 254、根据正确路径上的译码结果和 CRC校验结果以及第二子流 和第三子流的译码结果进行 AM R语音信号的译码处理。
需要说明的是, 步骤 253和步骤 254之间并不限定执行的先后顺序。 具体地, 以图 5所示 PLVA译码器的结构为例来说, 该 PLVA译码器 可以包括三路输出, 这三路输出包括 VA CRCI、 PLVA CRCI以及 PLVA 译码比特流, 该 VA CRCI即为最佳路径的 CRC校验结果, 该最佳路径为 预先设定的路径, 例如 Pathl , 而该 VA CRCI与 VA译码器输出的 CRCI 是等同的, 该 PLVA CRCI即为正确路径的 CRC校验结果, 正确路径可能 与最佳路径相同, 例如正确路径和最佳路径均为 Pathl , 此时 PLVA CRCI 与 VA CRCI相同, 正确路径也可能与最佳路径不同, 例如, 正确路径为 Path2 , 则该 PLVA CRCI为 Path2的校验结果, PLVA译码比特流则为正 确路径的译码结果。 如果候选路径中所有路径都不正确, 则 PLVA译码器 的 CRC Check&Choose模块也输出最佳路径的 CRC校验结果,也即, PLVA 译码器的三路输出分别为: 最佳路径的译码结果, 最佳路径的 CRC校验 结果以及最佳路径的 CRC校验结果, 此时 PLVA译码器等价于 VA译码 哭口 。
由上述 PLVA的原理描述可知, 如果最佳路径对应的译码结果正确, 贝' J VA译码的结果和 PLVA译码的结果相同, 两个 CRC校验结果均为正 确; 如果最佳路径对应的译码结果错误, 其他候选路径对应的译码结果正 确, 则 VA CRC校验结果错误, PLVA CRC校验结果正确, PLVA输出正 确的译码结果; 如果所有候选路径对应的译码结果均错误, PLVA输出最 佳路径对应的译码结果。 因此, VA CRC校验结果正确的时候, PLVA的 CRC校验结果必然正确; 反之, PLVA CRC校验结果正确, 但 VA CRC 校验结果不一定正确。
在具体实现时, UE的接收模块可以接收基站发送的包含 A子流、 B 子流以及 C子流的下行 AMR语音信号, 其中 A子流包含 CRC。 该 UE中 包括两种译码器,一种是 PLVA译码器,另一种是 VA译码器,其中, PLVA 译码器可以针对 A子流进行译码, 而 VA译码器则可以针对 B子流和 C 子流进行译码。对于采用 VA译码器对 B子流和 C子流进行译码的过程来 说, 其可以采用现有技术实现, 此处不再赘述, 译码后的比特流可以发送 给 UE中的 AMR语音译码器进行译码。
下面对 PLVA译码器对 A子流进行译码的过程进行详细说明。对于 A 子流来说,采用图 5的 PLVA译码器,可以输出三个译码结果, VA CRCI、 PLVA CRCI以及译码后的比特流。 PLVA CRCI以及译码后的比特流可以 发送给 AMR语音译码器, 其中 PLVA CRCI可以用于指示语音帧是否可 用,而 VA CRCI则可以发送给 UE中的外环功率控制模块进行外环功率控 制。 在具体实现时, UE可以将 PLVA译码后的 A子流和 VA译码得到的 B、 C子流进行组帧, 送给 AMR语音译码器, AMR语音译码器可以根据 收到的包括三个子流的 AMR语音信号、 以及对应的 BFI指示进行语音译 码。
发明人采用上述技术方案进行了系统仿真, 从仿真结果可知, 对于 AMR 12.2k业务的 A子流, PLVA-4相对于 VA, 大约有 0.3dB的性能增 益。 通过双 CRC上报的方案, 在 BLER=1%的时候, 可以得到 0.1 分的 MOS分增益, 在 BLER=10%的时候, 可以获得 0.35分的 MOS分增益。 当系统中 BLER越大, 由 PLVA带来的 MOS分增益也越大。
本实施例, 通过采用双 CRC 的方案, 使得带有功率控制的语音处理 系统, 例如 WCDMA 系统等, 可以将 LVA 带来的性能增益转化为语音 MOS分的增益。 由于 A子流在 AMR语音中重要性最高, 该方法能在最 大程度上提升语音性能, 同时对现有系统影响最小。
图 24为本发明用户设备实施例一的结构示意图, 如图 24所示, 本实 施例的 UE可以包括: 接收模块 26、 第一译码处理模块 27和第二译码处 理模块 28, 其中, 接收模块 26, 用于接收基站发送的编码语音信号, 所 述编码语音信号包括第一子流、 第二子流和第三子流, 所述第一子流中包 含循环卷积校验 CRC; 第一译码处理模块 27, 用于采用译码算法对所述 第一子流、 第二子流和第三子流进行译码处理, 获取所述第一子流、 第二 子流和第三子流的译码结果, 其中, 采用基于 CRC辅助判决的译码算法 对所述第一子流进行译码处理, 所述第一子流的译码结果中包括译码比特 流和 CRC校验结果; 第二译码处理模块 28, 用于根据所述第一子流的译 码比特流和 CRC校验结果以及所述第二子流和第三子流的译码结果进行 AMR语音信号的译码处理。
本实施例中,接收模块 26即为 UE中用于接收基站发送的下行信号的 模块, 第一译码处理模块 27即为 UE中进行卷积码译码处理的模块, 第二 译码处理模块 28, 即为 UE中的 AMR语音译码器。 本实施例中的模块可 以采用硬件实现方式, 例如复用现有 UE结构中的相应模块, 也可以采用 软件形式实现, 例如在存储器中装载相应的程序代码, 又或者软硬结合的 方式实现, 例如第一译码处理模块 27和第二译码处理模块 28均可以采用 软件形式实现, 而接收模块 26可以采用硬件形式实现。
本实施例的 UE, 其用于执行图 21所示方法实施例的技术方案, 其实 现原理和技术效果类似, 此处不再赘述。
图 25为本发明用户设备实施例二的结构示意图, 如图 25所示, 本实 施例在图 24所示 UE结构的基础上, 进一步地, 还包括: 外环功率控制模 块 29, 其中:
接收模块 26包括: 第一接收单元 261 , 用于接收所述第一子流; 第二 接收单元 262, 用于接收所述第二子流; 第三接收单元 263 , 用于接收所 述第三子流;
第一译码处理模块 27 包括: 第一译码处理单元 271 , 用于采用基于
CRC辅助判决的译码算法, 对所述第一子流进行译码处理, 获取多条候选 路径上的译码结果,应用所述 CRC对多条候选路径上的译码结果进行 CRC 校验, 获取 CRC校验结果正确路径上的译码结果以及该正确路径的 CRC 校验结果和最佳路径的 CRC校验结果, 所述最佳路径为采用维特比译码 算法确定的最大似然路径; 第二译码处理单元 272, 用于采用维特比译码 算法对所述第二子流进行译码处理, 获取译码结果; 第三译码处理单元 273 , 用于采用维特比译码算法对所述第三子流进行译码处理, 获取译码 结果;
外环功率控制模块 29, 用于根据所述最佳路径的 CRC校验结果进行 外环功率控制;
第二译码处理模块 28 ,具体用于根据所述第一译码处理单元 271获取 的所述正确路径上的译码结果和 CRC 校验结果以及第二译码处理单元 272获取的所述第二子流和第三译码处理单元 273获取的所述第三子流的 译码结果进行 AMR语音信号的译码处理。
需要说明的是, 本实施例中, 三个接收单元可以在物理上合为一个、 第二译码处理单元 272和第三译码处理单元 273可以采用一个 VA译码器 实现。
本实施例的 UE, 其用于执行图 22所示方法实施例的技术方案, 其具 体的逻辑架构参见图 23 所示的架构, 其实现原理和技术效果类似, 此处 不再赘述。
图 26为本发明用户设备实施例三的结构示意图, 如图 26所示, 本实 施例在图 24所示 UE结构的基础上, 进一步地, 还包括: 外环功率控制模 块 29和内环功率控制模块 30, 其中:
外环功率控制模块 29 , 用于降低外环功率控制的目标块误码率, 并根 据降低后的目标块误码率向所述内环功率控制模块发送降低后的目标信 噪比;
内环功率控制模块 30,用于根据所述降低后的目标信噪比和测量的信 噪比进行内环功率控制。
本实施例的 UE, 其可以用于执行前述方案一, 其实现原理和技术效 果类似, 此处不再赘述。
图 27为本发明用户设备实施例四的结构示意图, 如图 27所示, 本实 施例在图 24所示 UE结构的基础上, 进一步地, 还包括: 指示模块 31 , 用于指示所述基站减少所述第一子流所占用的信道资源, 增加所述第二子 流和第三子流所占用的信道资源。
本实施例的 UE, 其可以用于执行前述方案二, 其实现原理和技术效 果类似, 此处不再赘述。
本领域普通技术人员可以理解: 实现上述方法实施例的全部或部分步 骤可以通过程序指令相关的硬件来完成, 前述的程序可以存储于一计算机 可读取存储介质中, 该程序在执行时, 执行包括上述方法实施例的步骤; 而前述的存储介质包括: ROM、 RAM, 磁碟或者光盘等各种可以存储程 序代码的介质。 最后应说明的是: 以上实施例仅用以说明本发明的技术方案, 而非对 其限制; 尽管参照前述实施例对本发明进行了详细的说明, 本领域的普通 技术人员应当理解: 其依然可以对前述各实施例所记载的技术方案进行修 改, 或者对其中部分技术特征进行等同替换; 而这些修改或者替换, 并不 使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims

权 利 要 求 书
1、 一种语音信号处理方法, 其特征在于, 包括:
接收用户设备 UE发送的编码语音信号, 所述编码语音信号包括第一 子流、 第二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC; 采用译码算法对所述第一子流、 第二子流和第三子流进行译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理; 向基站控制器发送所述第一子流、 第二子流和第三子流的译码结果, 所述第一子流的译码结果中包括译码比特流和 CRC校验结果。
2、 根据权利要求 1所述的方法, 其特征在于, 所述基于 CRC辅助判 决的译码算法为列举维特比译码算法或者比特反转译码算法。
3、 根据权利要求 2所述的方法, 其特征在于, 所述列举维特比译码 算法为并行列举维特比译码算法或者串行列举维特比译码算法。
4、 根据权利要求 3 所述的方法, 其特征在于, 所述并行列举维特比 译码算法为包括 2条、 4条、 6条、 8条、 12条或者 16条候选路径的并行 列举维特比译码算法。
5、 根据权利要求 1〜4 中任一权利要求所述的方法, 其特征在于, 所 述采用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理, 包 括:
采用基于 CRC辅助判决的译码算法, 获取多条候选路径上的译码结 果, 应用所述 CRC对多条候选路径上的译码结果进行 CRC校验, 获取 CRC校验结果正确路径上的译码结果以及该正确路径的 CRC校验结果和 最佳路径的 CRC校验结果, 所述最佳路径为采用维特比译码算法确定的 最大似然路径;
向基站控制器发送所述第一子流的译码结果, 包括:
向所述基站控制器发送所述正确路径上的译码结果、 所述正确路径的
CRC校验结果以及最佳路径的校验结果,以使所述基站控制器将所述最佳 路径的校验结果发送给外环功率控制模块、 将所述正确路径上的译码结果 和 CRC校验结果发送给核心网。
6、 根据权利要求 1〜4 中任一权利要求所述的方法, 其特征在于, 还 包括: 减少所述第一子流所占用的信道资源, 增加所述第二子流和第三子流 所占用的信道资源。
7、 根据权利要求 6所述的方法, 其特征在于, 所述减少所述第一子 流所占用的信道资源, 增加所述第二子流和第三子流所占用的信道资源, 包括:
减小所述第一子流的速率匹配参数, 增大所述第二子流和第三子流的 速率匹配参数。
8、 根据权利要求 1〜4 中任一权利要求所述的方法, 其特征在于, 还 包括:
接收所述基站控制器发送的降低后的目标信噪比, 并根据所述目标信 噪比, 进行内环功率控制。
9、 一种语音信号处理方法, 其特征在于, 包括:
接收基站发送的第一子流、 第二子流和第三子流的译码结果, 其中, 第一 子流的译码结果为采用基于循环卷积校验 CRC辅助判决的译码算法进行译码 处理后获取的译码结果且该译码结果中包括译码比特流和 CRC校验结果; 将所述 CRC校验结果发送给外环功率控制模块,将所述第一子流的译码比 特流和 CRC校验结果以及所述第二子流和第三子流的译码结果发送给核心网。
10、 根据权利要求 9所述的方法, 其特征在于, 所述 CRC校验结果 包括正确路径的 CRC校验结果和最佳路径的 CRC校验结果, 所述译码比 特流为正确路径上的译码比特流, 所述最佳路径为采用维特比译码算法确 定的最大似然路径;
将所述 CRC校验结果发送给外环功率控制模块, 包括:
将所述最佳路径的 CRC校验结果发送给所述外环功率控制模块; 将所述第一子流的译码比特流和 CRC校验结果发送给核心网, 包括: 将所述正确路径上的译码比特流和所述正确路径的 CRC校验结果发 送给核心网。
11、 根据权利要求 9所述的方法, 其特征在于, 还包括:
指示所述基站减少所述第一子流所占用的信道资源, 并增加所述第二 子流和第三子流所占用的信道资源。
12、 根据权利要求 9所述的方法, 其特征在于, 还包括: 降低所述外环功率控制模块的目标块误码率, 以使所述外环功率控制 模块向所述基站发送降低后的目标信噪比。
13、 一种基站, 其特征在于, 包括:
第一接收模块, 用于接收用户设备 UE发送的编码语音信号, 所述编 码语音信号包括第一子流、 第二子流和第三子流, 所述第一子流中包含循 环卷积校验 CRC;
译码处理模块, 用于采用译码算法对所述第一子流、 第二子流和第三 子流进行译码处理, 其中, 采用基于 CRC辅助判决的译码算法对所述第 一子流进行译码处理;
第一发送模块, 用于向基站控制器发送所述第一子流、 第二子流和第 三子流的译码结果, 所述第一子流的译码结果中包括译码比特流和 CRC 校验结果。
14、 根据权利要求 13所述的基站, 其特征在于, 所述第一接收模块, 包括:
第一接收单元, 用于接收所述第一子流;
第二接收单元, 用于接收所述第二子流;
第三接收单元, 用于接收所述第三子流;
所述译码处理模块, 包括:
第一译码处理单元, 用于采用并行列举维特比译码算法对所述第一子 流进行译码处理, 获取多条候选路径上的译码结果, 应用所述 CRC对多 条候选路径上的译码结果进行 CRC校验, 获取 CRC校验结果正确路径上 的译码结果以及该正确路径的 CRC校验结果和最佳路径的 CRC校验结 果, 所述最佳路径为采用维特比译码算法确定的最大似然路径;
第二译码处理单元, 用于采用维特比译码算法对所述第二子流进行译 码处理, 获取译码结果;
第三译码处理单元, 用于采用维特比译码算法对所述第三子流进行译 码处理, 获取译码结果;
所述第一发送模块, 包括:
第一发送单元, 用于将所述第一译码处理单元获取的所述正确路径上 的译码结果、 所述正确路径的 CRC校验结果以及最佳路径的校验结果发 送给基站控制器, 以使所述基站控制器将所述最佳路径的校验结果发送给 外环功率控制模块、 将所述正确路径上的译码结果和 CRC校验结果发送 给核心网;
第二发送单元, 用于将所述第二译码处理单元获取的译码结果发送给 所述基站控制器;
第三发送单元, 用于将所述第三译码处理单元获取的译码结果发送给 所述基站控制器。
15、 根据权利要求 13所述的基站, 其特征在于, 还包括:
信道资源控制模块, 用于减少所述第一子流所占用的信道资源, 增加 所述第二子流和第三子流所占用的信道资源。
16、 根据权利要求 13所述的基站, 其特征在于, 还包括:
内环功率控制模块, 用于接收所述基站控制器发送的降低后的目标信 噪比, 并根据所述目标信噪比, 进行内环功率控制。
17、 一种基站控制器, 其特征在于, 包括:
第二接收模块, 用于接收基站发送的第一子流、 第二子流和第三子流 的译码结果, 其中, 第一子流的译码结果为采用基于循环卷积校验 CRC 辅助判决的译码算法进行译码处理后获取的译码结果且该译码结果中包 括译码比特流和 CRC校验结果; 将所述第一子流的译码比特流和 CRC校验结果以及所述第二子流和第三 子流的译码结果发送给核心网。
18、 根据权利要求 17所述的基站控制器, 其特征在于, 所述 CRC校 验结果包括正确路径的 CRC校验结果和最佳路径的 CRC校验结果, 所述 译码比特流为正确路径上的译码比特流, 所述最佳路径为采用维特比译码 算法确定的最大似然路径; 所述第二发送模块, 包括:
第四发送单元, 用于将所述最佳路径的 CRC校验结果发送给所述外 环功率控制模块;
第五发送单元, 用于将所述正确路径上的译码比特流和所述正确路径 的 CRC校验结果发送给核心网, 将所述第二子流和第三子流的译码结果 发送给所述核心网。
19、 根据权利要求 17所述的基站控制器, 其特征在于, 还包括: 指示模块, 用于指示所述基站减少所述第一子流所占用的信道资源, 并增加所述第二子流和第三子流所占用的信道资源。
20、 根据权利要求 17所述的基站控制器, 其特征在于, 还包括: 参数控制模块, 用于降低所述外环功率控制模块的目标块误码率, 以 使所述外环功率控制模块向所述基站发送降低后的目标信噪比。
21、 一种接入网系统, 其特征在于, 包括: 基站和基站控制器, 所述 基站采用权利要求 13〜16中任一项所述的基站, 所述基站控制器采用权利 要求 17〜20中任一项所述的基站控制器。
22、 一种语音信号处理方法, 其特征在于, 包括:
接收基站发送的编码语音信号, 所述编码语音信号包括第一子流、 第 二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC;
采用译码算法对所述第一子流、 第二子流和第三子流进行译码处理, 获取所述第一子流、 第二子流和第三子流的译码结果, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理,所述第一子流的 译码结果中包括译码比特流和 CRC校验结果;
根据所述第一子流的译码比特流和 CRC校验结果以及所述第二子流 和第三子流的译码结果进行 AM R语音信号的译码处理。
23、 根据权利要求 22所述的方法, 其特征在于, 所述基于 CRC辅助 判决的译码算法为列举维特比译码算法或者比特反转译码算法。
24、 根据权利要求 23 所述的方法, 其特征在于, 所述列举维特比译 码算法为并行列举维特比译码算法或者串行列举维特比译码算法。
25、 根据权利要求 24所述的方法, 其特征在于, 所述并行列举维特 比译码算法为包括 2条、 4条、 6条、 8条、 12条或者 16条候选路径的并 行列举维特比译码算法。
26、 根据权利要求 22〜25中任一项所述的方法, 其特征在于, 所述采 用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理, 包括: 采用基于 CRC辅助判决的译码算法, 对所述第一子流进行译码处理, 获取多条候选路径上的译码结果, 应用所述 CRC对多条候选路径上的译 码结果进行 CRC校验, 获取 CRC校验结果正确路径上的译码结果以及该 正确路径的 CRC校验结果和最佳路径的 CRC校验结果, 所述最佳路径为 采用维特比译码算法确定的最大似然路径;
所述采用基于 CRC辅助判决的译码算法对所述第一子流进行译码处 理之后, 还包括:
根据所述最佳路径的 CRC校验结果进行外环功率控制;
所述根据所述第一子流的译码比特流和 CRC校验结果以及所述第二 子流和第三子流的译码结果进行 AMR语音信号的译码处理, 包括:
根据所述正确路径上的译码结果和 CRC校验结果以及所述第二子流 和第三子流的译码结果进行 AM R语音信号的译码处理。
27、 根据权利要求 22〜25中任一项所述的方法, 其特征在于, 所述采 用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理之后, 还 包括:
降低外环功率控制的目标块误码率, 并采用根据降低后的目标块误码 率获得的降低后的目标信噪比和测量的信噪比进行内环功率控制。
28、 根据权利要求 22〜25中任一项所述的方法, 其特征在于, 所述采 用基于 CRC辅助判决的译码算法对所述第一子流进行译码处理之后, 还 包括:
指示所述基站减少所述第一子流所占用的信道资源, 增加所述第二子 流和第三子流所占用的信道资源。
29、 一种用户设备, 其特征在于, 包括:
接收模块, 用于接收基站发送的编码语音信号, 所述编码语音信号包 括第一子流、 第二子流和第三子流, 所述第一子流中包含循环卷积校验 CRC;
第一译码处理模块, 用于采用译码算法对所述第一子流、 第二子流和 第三子流进行译码处理, 获取所述第一子流、 第二子流和第三子流的译码 结果, 其中, 采用基于 CRC辅助判决的译码算法对所述第一子流进行译 码处理, 所述第一子流的译码结果中包括译码比特流和 CRC校验结果; 第二译码处理模块, 用于根据所述第一子流的译码比特流和 CRC校 验结果以及所述第二子流和第三子流的译码结果进行 AMR语音信号的译 码处理。
30、 根据权利要求 29所述的用户设备, 其特征在于, 所述基于 CRC 辅助判决的译码算法为列举维特比译码算法或者比特反转译码算法。
31、 根据权利要求 29所述的用户设备, 其特征在于, 所述列举维特 比译码算法为并行列举维特比译码算法或者串行列举维特比译码算法。
32、 根据权利要求 29所述的用户设备, 其特征在于, 所述并行列举 维特比译码算法为包括 2条、 4条、 6条、 8条、 12条或者 16条候选路径 的并行列举维特比译码算法。
33、 根据权利要求 29〜32中任一项所述的用户设备, 其特征在于, 还 包括: 外环功率控制模块;
所述接收模块, 包括:
第一接收单元, 用于接收所述第一子流;
第二接收单元, 用于接收所述第二子流;
第三接收单元, 用于接收所述第三子流;
所述第一译码处理模块, 包括:
第一译码处理单元, 用于采用基于 CRC辅助判决的译码算法, 对所 述第一子流进行译码处理, 获取多条候选路径上的译码结果, 应用所述 CRC对多条候选路径上的译码结果进行 CRC校验,获取 CRC校验结果正 确路径上的译码结果以及该正确路径的 CRC校验结果和最佳路径的 CRC 校验结果, 所述最佳路径为采用维特比译码算法确定的最大似然路径; 第二译码处理单元, 用于采用维特比译码算法对所述第二子流进行译 码处理, 获取译码结果;
第三译码处理单元, 用于采用维特比译码算法对所述第三子流进行译 码处理, 获取译码结果;
所述外环功率控制模块, 具体用于根据所述最佳路径的 CRC校验结 果进行外环功率控制;
所述第二译码处理模块, 具体用于根据所述第一译码处理单元获取的 所述正确路径上的译码结果和 CRC校验结果以及第二译码处理单元获取 的所述第二子流和第三译码处理单元获取的所述第三子流的译码结果进 行 AMR语音信号的译码处理。
34、 根据权利要求 29〜32中任一项所述的用户设备, 其特征在于, 还 包括: 外环功率控制模块和内环功率控制模块;
所述外环功率控制模块, 用于降低外环功率控制的目标块误码率, 并 根据降低后的目标块误码率向所述内环功率控制模块发送降低后的目标 信噪比;
所述内环功率控制模块, 用于根据所述降低后的目标信噪比和测量的 信噪比进行内环功率控制。
35、 根据权利要求 29〜32中任一项所述的用户设备, 其特征在于, 还 包括:
指示模块, 用于指示所述基站减少所述第一子流所占用的信道资源, 增加所述第二子流和第三子流所占用的信道资源。
PCT/CN2012/070658 2011-05-27 2012-01-20 语音信号处理方法、装置和接入网系统 Ceased WO2012163099A1 (zh)

Priority Applications (7)

Application Number Priority Date Filing Date Title
KR1020137034285A KR101611018B1 (ko) 2011-05-27 2012-01-20 음성 신호 처리 방법과 장치, 및 액세스 네트워크 시스템
EP12793985.8A EP2709280A4 (en) 2011-05-27 2012-01-20 METHOD FOR LANGUAGE SIGNAL PROCESSING AND DEVICE AND ACCESS NETWORK SYSTEM
RU2013158335/08A RU2543944C1 (ru) 2011-05-27 2012-01-20 Способ, устройство и система сети доступа для обработки речевых сигналов
BR112013030548-7A BR112013030548A2 (pt) 2011-05-27 2012-01-20 método, aparelho e sistema de rede de acesso para processamento de sinal de voz.
CN201280025927.XA CN103782518B (zh) 2011-05-27 2012-01-20 语音信号处理方法、装置和接入网系统
JP2014511711A JP5827743B2 (ja) 2011-05-27 2012-01-20 音声信号処理のための方法、装置、及びアクセス・ネットワーク・システム
US14/092,319 US9171540B2 (en) 2011-05-27 2013-11-27 Method, apparatus, and access network system for speech signal processing

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
PCT/CN2011/074801 WO2011144112A2 (zh) 2011-05-27 2011-05-27 语音信号处理方法、装置和接入网系统
CNPCT/CN2011/074801 2011-05-27

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/092,319 Continuation US9171540B2 (en) 2011-05-27 2013-11-27 Method, apparatus, and access network system for speech signal processing

Publications (1)

Publication Number Publication Date
WO2012163099A1 true WO2012163099A1 (zh) 2012-12-06

Family

ID=44992121

Family Applications (2)

Application Number Title Priority Date Filing Date
PCT/CN2011/074801 Ceased WO2011144112A2 (zh) 2011-05-27 2011-05-27 语音信号处理方法、装置和接入网系统
PCT/CN2012/070658 Ceased WO2012163099A1 (zh) 2011-05-27 2012-01-20 语音信号处理方法、装置和接入网系统

Family Applications Before (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/074801 Ceased WO2011144112A2 (zh) 2011-05-27 2011-05-27 语音信号处理方法、装置和接入网系统

Country Status (8)

Country Link
US (2) US9177548B2 (zh)
EP (2) EP2706709B1 (zh)
JP (2) JP5881818B2 (zh)
KR (2) KR101568921B1 (zh)
CN (3) CN102907030A (zh)
BR (2) BR112013030371A2 (zh)
RU (2) RU2546321C1 (zh)
WO (2) WO2011144112A2 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015120575A1 (zh) * 2014-02-11 2015-08-20 华为技术有限公司 信道译码的方法、装置和分布式译码器

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011144112A2 (zh) 2011-05-27 2011-11-24 华为技术有限公司 语音信号处理方法、装置和接入网系统
WO2014011622A2 (en) 2012-07-09 2014-01-16 Vid Scale, Inc. Power aware video decoding and streaming
CN106452680B (zh) * 2013-09-23 2019-10-18 华为技术有限公司 信号处理方法及设备
CN104796160B (zh) * 2014-01-22 2019-04-12 华为技术有限公司 译码方法和装置
CA2967718A1 (en) * 2014-11-19 2016-05-26 Lantiq Beteiligungs-GmbH & Co.KG Link loss detection
US20160204908A1 (en) * 2015-01-14 2016-07-14 Qualcomm Incorporated Adaptive multi-rate partial decode
CN107005254B (zh) * 2015-06-12 2020-03-20 华为技术有限公司 信道译码的方法、装置和系统
US20170118707A1 (en) * 2015-10-21 2017-04-27 Qualcomm Incorporated Techniques for achieving optimal power and mean opinion scores for internet protocol multimedia subsystem-based dual connectivity calling
JP6807956B2 (ja) * 2016-05-20 2021-01-06 華為技術有限公司Huawei Technologies Co.,Ltd. パケットドメイン内の音声サービスをスケジューリングするための方法および装置
US10992729B2 (en) * 2017-04-18 2021-04-27 Microsoft Technology Licensing, Llc Endpoint configuration for a communication session
CN111294060B (zh) * 2019-02-28 2023-04-21 展讯通信(上海)有限公司 一种语音信息译码方法及装置
CN117091462B (zh) * 2022-11-11 2025-12-23 上海芯跳科技有限公司 高可靠通信的电子雷管电路、应用方法及系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1221169A (zh) * 1997-10-17 1999-06-30 索尼公司 编码方法和装置以及解码方法和装置
EP0930738A2 (en) * 1998-01-13 1999-07-21 Lucent Technologies Inc. Optimal complementary punctured convolutional codes for use in digital audio broadcasting and other applications
US6161210A (en) * 1998-04-03 2000-12-12 Lucent Technologies Inc. List Viterbi algorithms for tailbiting convolutional codes
CN1360796A (zh) * 1999-07-09 2002-07-24 艾利森电话股份有限公司 面向分组的信息网络中具有实时需求的压缩信息的传输
CN1483257A (zh) * 2000-01-10 2004-03-17 �����ɷ� 在cdma通信系统中支持自适应多速率(amr)数据的方法和设备
US20100058152A1 (en) * 2008-09-04 2010-03-04 Kabushiki Kaisha Toshiba Decoding apparatus and method
WO2011144112A2 (zh) * 2011-05-27 2011-11-24 华为技术有限公司 语音信号处理方法、装置和接入网系统

Family Cites Families (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ZA947317B (en) * 1993-09-24 1995-05-10 Qualcomm Inc Multirate serial viterbi decoder for code division multiple access system applications
JP3239795B2 (ja) * 1997-04-23 2001-12-17 三菱電機株式会社 誤り訂正復号装置および誤り訂正復号方式
JP3567067B2 (ja) 1997-11-04 2004-09-15 株式会社日立グローバルストレージテクノロジーズ ディジタル磁気記録再生装置
US6141388A (en) * 1998-03-11 2000-10-31 Ericsson Inc. Received signal quality determination method and systems for convolutionally encoded communication channels
US6105158A (en) * 1998-04-03 2000-08-15 Lucent Technologies, Inc. Screening for undetected errors in data transmission systems
US6108386A (en) * 1998-04-03 2000-08-22 Lucent Technologies Inc. List Viterbi algorithms for continuous data transmission
US6272660B1 (en) * 1998-04-03 2001-08-07 Agere Systems Guardian Corp. Screening for errors in data transmission systems
JP3653391B2 (ja) * 1998-06-11 2005-05-25 株式会社日立グローバルストレージテクノロジーズ ビタビ検出器及びこれを用いたディジタル磁気記録再生装置
WO2001003448A2 (en) * 1999-07-05 2001-01-11 Nokia Corporation Method for coding mode selection
RU2168858C1 (ru) * 1999-12-30 2001-06-10 Закрытое акционерное общество "Кодофон" Способ декодирования данных в прямом канале umts-2000 для второго типа прерывистой передачи
US7746832B2 (en) * 2001-01-05 2010-06-29 Qualcomm Incorporated Method and apparatus for supporting adaptive multi-rate (AMR) data in a CDMA communication system
WO2003021905A1 (fr) * 2001-08-31 2003-03-13 Fujitsu Limited Recepteur et procede de reception pour systeme de communication amdc
CN1284319C (zh) * 2002-04-22 2006-11-08 西安大唐电信有限公司 一种多通道amr声码器的实现方法和设备
US7023880B2 (en) * 2002-10-28 2006-04-04 Qualcomm Incorporated Re-formatting variable-rate vocoder frames for inter-system transmissions
JP4701343B2 (ja) * 2002-10-30 2011-06-15 エスティー‐エリクソン、ソシエテ、アノニム トレリスに基づく受信器
US7643993B2 (en) * 2006-01-05 2010-01-05 Broadcom Corporation Method and system for decoding WCDMA AMR speech data using redundancy
JP4305341B2 (ja) * 2004-09-09 2009-07-29 富士通株式会社 無線通信装置、無線基地局、無線基地局制御装置、送信電力制御方法
US20080139235A1 (en) 2004-12-22 2008-06-12 Claes Tidestav Method and Arrangement for Improved Outer Loop Power Control
US8019615B2 (en) 2005-07-26 2011-09-13 Broadcom Corporation Method and system for decoding GSM speech data using redundancy
CN101336450B (zh) * 2006-02-06 2012-03-14 艾利森电话股份有限公司 在无线通信系统中用于语音编码的方法和装置
US7831895B2 (en) * 2006-07-25 2010-11-09 Communications Coding Corporation Universal error control coding system for digital communication and data storage systems
CN101155141A (zh) * 2006-09-29 2008-04-02 华为技术有限公司 分组语音数据包传输方法及其系统
JP4326561B2 (ja) * 2006-12-18 2009-09-09 株式会社エヌ・ティ・ティ・ドコモ 移動通信端末及び送信電力制御方法
JP4888245B2 (ja) * 2007-06-25 2012-02-29 富士通株式会社 受信品質測定方法及び送信電力制御方法ならびにそれらの装置
US20090276221A1 (en) 2008-05-05 2009-11-05 Arie Heiman Method and System for Processing Channel B Data for AMR and/or WAMR
CN101321035B (zh) * 2008-07-09 2012-03-21 上海华为技术有限公司 差值上限获取方法、定点方法及装置
JP4785899B2 (ja) * 2008-09-05 2011-10-05 富士通株式会社 送信電力制御方法
CN101841340B (zh) * 2010-05-28 2013-04-24 华为技术有限公司 差值上限获取方法及装置、维特比译码方法及装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1221169A (zh) * 1997-10-17 1999-06-30 索尼公司 编码方法和装置以及解码方法和装置
EP0930738A2 (en) * 1998-01-13 1999-07-21 Lucent Technologies Inc. Optimal complementary punctured convolutional codes for use in digital audio broadcasting and other applications
US6161210A (en) * 1998-04-03 2000-12-12 Lucent Technologies Inc. List Viterbi algorithms for tailbiting convolutional codes
CN1360796A (zh) * 1999-07-09 2002-07-24 艾利森电话股份有限公司 面向分组的信息网络中具有实时需求的压缩信息的传输
CN1483257A (zh) * 2000-01-10 2004-03-17 �����ɷ� 在cdma通信系统中支持自适应多速率(amr)数据的方法和设备
US20100058152A1 (en) * 2008-09-04 2010-03-04 Kabushiki Kaisha Toshiba Decoding apparatus and method
WO2011144112A2 (zh) * 2011-05-27 2011-11-24 华为技术有限公司 语音信号处理方法、装置和接入网系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2709280A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015120575A1 (zh) * 2014-02-11 2015-08-20 华为技术有限公司 信道译码的方法、装置和分布式译码器
US10285086B2 (en) 2014-02-11 2019-05-07 Huawei Technologies Co., Ltd. Channel decoding method and apparatus, and distributed decoder

Also Published As

Publication number Publication date
CN103782518A (zh) 2014-05-07
US20140088963A1 (en) 2014-03-27
EP2706709A4 (en) 2014-04-09
JP2014519743A (ja) 2014-08-14
WO2011144112A2 (zh) 2011-11-24
BR112013030548A2 (pt) 2018-04-24
JP5881818B2 (ja) 2016-03-09
US20140089770A1 (en) 2014-03-27
WO2011144112A3 (zh) 2012-03-08
JP5827743B2 (ja) 2015-12-02
BR112013030371A2 (pt) 2016-12-13
KR20140015577A (ko) 2014-02-06
EP2706709B1 (en) 2017-07-05
CN103782518B (zh) 2017-02-22
CN102907030A (zh) 2013-01-30
JP2014519742A (ja) 2014-08-14
EP2706709A2 (en) 2014-03-12
KR101568921B1 (ko) 2015-11-12
US9171540B2 (en) 2015-10-27
EP2709280A1 (en) 2014-03-19
RU2546321C1 (ru) 2015-04-10
KR20140016394A (ko) 2014-02-07
EP2709280A4 (en) 2014-04-09
RU2543944C1 (ru) 2015-03-10
CN106788904A (zh) 2017-05-31
US9177548B2 (en) 2015-11-03
KR101611018B1 (ko) 2016-04-08

Similar Documents

Publication Publication Date Title
WO2012163099A1 (zh) 语音信号处理方法、装置和接入网系统
US10476998B2 (en) Reinforced list decoding
WO2017185377A1 (zh) 极化Polar码的编译码方法及装置
EP3533147A1 (en) Iterative decoding of polar code with bit-flipping of unreliable bits
US10715272B2 (en) Signal processing method and device
US11128313B2 (en) Apparatus and method for decoding signal in wireless communication system
CN110166166B (zh) 网络数据处理装置及方法
CN103532660B (zh) 信源信道的解码方法、装置及终端设备

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: 12793985

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2014511711

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2012793985

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20137034285

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2013158335

Country of ref document: RU

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112013030548

Country of ref document: BR

ENP Entry into the national phase

Ref document number: 112013030548

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20131127