WO2003061214A1

WO2003061214A1 - A hybird arq scheme for packet data transmission 0ver wireless channel

Info

Publication number: WO2003061214A1
Application number: PCT/CN2002/000018
Authority: WO
Inventors: Chundi Xiu; Yongzhong Zou; Yongsheng Zhang
Original assignee: Linkair Communications,Inc.
Priority date: 2002-01-15
Filing date: 2002-01-15
Publication date: 2003-07-24
Also published as: AU2002226259A1; CN1494791A

Abstract

The present invention is a hybird ARQ scheme for packet dat a transmission over wireless channel. For the first transmission, only the single turbo product code is transmitted. The transmitter responds to the retransmission request by sending the punctured code derived from the output of turbo-ARQ coding struture. The receiver emloys inner iterative decoding algorithm for turbo product code and outer iterative decoding algorithm corresponding to turbo-ARQ decoding structure for retransmitted packets. Duplex turbo iterative dexoding used in this scheme can improve decoding performance and further increase system throughput and reliability. The proposed scheme jointly used turbo product code and turbo -ARQ structure thus to provide a better performance with tolerable complexity.

Description

Hybrid ARQ method for wireless channel packet data transmission TECHNICAL FIELD

The invention belongs to the transmission field of electric communication technology, and relates to an error control method in a digital mobile communication network. The invention uses a Turbo Product Code (TPC- Turbo Product Code) and Turbo- ARQ (Turbo- Automat ic Repeat reQues). t) A hybrid ARQ method that implements reliable transmission of packet data over a wireless channel, specifically a hybrid ARQ method for packet data transmission over a wireless channel. Background technique

After experiencing the first generation of analog systems and the second generation of digital systems (represented by GSM and narrowband CDMA), modern mobile communications are coordinated by the International Telecommunication Union (ITU) to meet the requirements of market development. Since 1996, the standardization process of third-generation (3G) broadband digital systems has begun. The urgent demand for higher bit rate data services and better spectrum utilization is the main driving force for the development of 3G systems. As a result, efficient and reliable data transmission is becoming one of the focus topics in the communications field.

Mobile communication systems use error control technology to provide certain error protection for data transmission, and the choice of error control scheme usually depends on the quality of service and wireless channel required by the system. Quality of Service (QoS — Quality of Service), that is, the various services provided by the system to users should meet certain performance criteria, including BER-Bit Error Rate or FER — Frame Error Rate ), The maximum allowed delay and the communication channel capacity available to the user (usually expressed in terms of system throughput). Existing 3G packet data traffic system in claim error rate ^10-6 or better, and in the high data rate and high-speed mobile environment, the channel condition is often due to multipath, shadowing, fading, multiple access interference suffered serious damage. Therefore, an effective error control mechanism must be adopted to achieve high quality in a wireless channel. Data transfer.

Channel coding (or error control coding) is one of the most effective methods to improve the reliability of data transmission. The way to use coding technology to implement error control in communication systems is called forward error correction.

(FEC— Forward Error Correction). The FEC method selects a code with a certain error correction capability. The system can provide effective throughput by adjusting the encoding rate, but the limitation of the error protection capability of the error correction code itself limits the reliability of the system. Common error correction codes include block codes, convolutional codes, and concatenated forms of the two. In 1993, a turbo code (C. Berrou, A. Glavieux, and P. Thi t imajshima, "Near Shannon l imi t error-correct ing coding and decoding: Turbo-codes" that can provide coding gains similar to Shannon bounds was proposed. , "in Proc. ICC'93, pp. 1064- 1070, May 1993.), which broadened the way for the development of channel coding theory and technology. In the general sense, Turbo codes use a parallel convolutional convolutional code (PCCC — Parallel Concatenated Convo lutional Code) scheme, whose member codes are convoluted by two or more code rates. RSC—Recursive Systematic Convolution Code). The Turbo interleaver is used to connect each member code. The good performance of the Turbo code is particularly suitable for the transmission of packet data, because the data service can tolerate a large delay caused by the turbo iterative decoding and retransmission process. However, when the channel conditions are deteriorated (especially in various complicated channel environments such as deep fading or multiple access interference), even the 1/3 bit rate Turbo code will fail. Therefore, the pure FEC method can no longer be adapted to the data service requirements of the 3G system, and hybrid ARQ technology combining the FEC method and the ARQ method is receiving more and more attention.

Hybrid ARQ technology refers to any combination of FEC and ARQ methods. Among them, the ARQ method improves the reliability of the system by retransmitting the detected error data frames, and the FEC method is used to correct common errors in the channel to reduce the number of retransmissions and increase the system throughput. Most of the existing 3G systems use the PCCC-based Turbo code as the error correction code, and use the rate-matching-based Turbo (RCPT — Rate Compa ti bl e Punctured) Turbo) Chase combination method and incremental redundancy (IR-Incremental l

Redundancy) method to implement error control at the physical layer (3G TR 25. 848 V0. 6. (2000-05); Motoro la, Performance Compar i son of Hybr id-ARQ Schemes-Addi t iona l Resul ts, TSGR1 # 18 (01) 0044). In the Chase combination method, the transmitting end retransmits the RCPT code with a certain initial bit rate, and the receiving end performs Chase combining and decoding on the received repeated codewords. This method has simple decoding, small buffering requirements, and low system implementation complexity, but the system throughput is low at low signal-to-noise ratio (SNR—Signal Noi se Ratio). The IR method responds to a retransmission request that fails to decode with increased retransmission. The receiver combines the retransmitted data with the initial codeword to form a RCPT code with a lower code rate, and obtains a larger coding gain. In this method, the incremental redundancy determines whether to retransmit according to the change in channel conditions. The worse the channel conditions, the greater the number of retransmissions. At the same time, the PCCC scheme is required to cascade more member codes to provide sufficient redundancy, which results in an increase in the complexity of the corresponding decoding algorithms and decoding equipment. In addition, the buffer requirements at both ends of the system will also increase as the number of retransmissions increases. It can be seen that although this method improves the system performance, it increases the complexity and cost of system implementation.

The Turbo code used in the existing 3G system adopts the PCCC scheme. Its soft-in, soft-out iterative MAP decoding algorithm or simplified algorithm is a two-way algorithm with a large amount of calculation. The corresponding decoder has a complicated structure, high cost, and processing speed. slow. Therefore, the performance improvement of hybrid ARQ systems based on RCPT codes comes at the cost of increasing the complexity of the system. Summary of the Invention

An object of the present invention is to provide a hybrid ARQ method for wireless channel packet data transmission. This method uses a TPC coding scheme with strong error correction capability for hybrid ARQ systems to improve system frame error rate performance and improve system throughput. At the same time, a Turbo-ARQ structure (. R. Narayanan, and GL is used in the hybrid ARQ mechanism). Stuber, A Novel ARQ Technique us ing the Turbo Coding Pr inc iple, IEEE Commun i cations Le t ter, vo l. 1, No. 2, Mar. 1997.), to improve the performance of communication systems to a greater extent.

In the hybrid ARQ method proposed by the present invention, the selection of the TPC coding scheme enables the system to obtain a larger coding gain, and the application of the Turbo-ARQ structure enables the system to make full use of the useful information of each transmission data, thereby to a greater extent Improve communication system performance.

The technical solution of the present invention is:

The present invention provides a hybrid ARQ method for wireless channel packet data transmission, which includes: the transmitting end sends only a single Turbo product codeword during the first transmission process, and the transmitting end sends The truncated codeword output by the structure responds to the retransmission request that is feedbacked by the receiving end due to decoding failure. The receiving end uses internal iterative turbo decoding for decoding the turbo product codeword, and the receiving end uses retransmission data decoding Turbo-ARQ decoding structure with external iterative Turbo decoding.

The Turbo-ARQ coding structure is formed by parallel concatenation of two or more member codes, where: the first member code used for the parallel concatenation is a Turbo product code;

The Turbo-ARQ encoding structure further includes a truncation circuit, wherein: the transmitting end obtains a required codeword by Turbo-ARQ encoding of an information frame to be transmitted, and inputs the codeword into the truncation circuit to obtain a different truncated codeword, It is stored in the originating buffer for retransmission.

The Turbo-ARQ coding structure may be formed by parallel concatenation of two Turbo product codes and adding a truncation circuit, wherein: the transmitting end obtains a required codeword by Turbo-ARQ encoding of an information frame to be transmitted, and converts the codeword The input truncation circuit obtains different truncated codewords and stores them in the originating buffer for retransmission.

The Turbo-ARQ coding structure can be formed by parallel concatenation of a Turbo product code and a convolutional code and adding a truncation circuit, wherein: the transmitting end obtains the required codeword by Turbo-ARQ coding, This codeword is input to the truncation circuit to obtain different truncated codewords, and stored in the originating buffer for retransmission.

The Turbo-ARQ coding structure may be a coding structure including interleaving. The turbo-ARQ decoding structure is composed of turbo product code decoding and other member code decoding, wherein: the receiving end uses the turbo product code to receive the received turbo product code word and the prior information retained in the previous transmission process. Decoding performs internal iterative Turbo decoding;

The external information output by the internal iterative Turbo decoding and the prior information retained in the previous transmission process and the corresponding codeword are sent to other member codes for decoding for further decoding; the externalities of the further decoding output The information can be used as part of the a priori information of the turbo product code decoding to form an outer reiterated turbo decoding.

The Turbo-ARQ decoding structure can be composed of two Turbo product code decodings, where: the receiving end decodes the received Turbo product codeword and the a priori information retained in the previous transmission process by the first Turbo product code The code performs internal iterative Turbo decoding;

The external information output by the internal iterative Turbo decoding, the prior information retained in the previous transmission process, and the corresponding codeword are sent to a second Turbo product code decoding for further decoding; the further decoding The output external information can be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

The Turbo-ARQ decoding structure may be composed of one turbo product code decoding and one convolution code decoding, wherein: the receiving end transmits the received turbo product codeword and the a priori information retained in the previous transmission process through the turbo The product code decoding performs internal iterative turbo decoding; the external information output by the internal iterative turbo decoding, the a priori information retained in the previous transmission process, and the corresponding codeword are sent to the convolutional code decoding for further Decoding; the external information output by the further decoding may be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

The Turbo-ARQ decoding structure may be a decoding structure including interleaving and de-interleaving, wherein: the receiving end decodes the received turbo product codeword and the prior information retained in the previous transmission process by the turbo product code. Perform internal iterative Turbo decoding;

The external information outputted by the internal iterative Turbo decoding is interleaved with the first face information retained in the previous transmission process, and then sent to the corresponding codeword for further decoding. Decoding; the external information output by further decoding can be used as the

Part of the a priori information decoded by Turbo product codes forms an outer reiterated Turbo decoding.

The Turbo-ARQ encoding structure further includes a truncation circuit, wherein: the transmitting end obtains a required codeword by Turbo-ARQ encoding of an information frame to be transmitted, and inputs the codeword into the truncation circuit to obtain a different truncated codeword, And store it in the originating buffer for retransmission;

The turbo-ARQ decoding structure is composed of turbo product code decoding and other member code decoding, wherein: the receiving end uses the turbo product code to receive the received turbo product code word and the prior information retained in the previous transmission process. Decoding performs internal iterative Turbo decoding;

The external information outputted by the internal iterative Turbo decoding and the prior information retained in the previous transmission process and the corresponding codeword are sent to the member code decoding for further decoding; the external information outputted by the further decoding It can also be used as part of the prior information of the turbo product code decoding to form an outer-reiterating iterative turbo decoding.

The Turbo-ARQ coding structure can be formed by parallel concatenation of two Turbo product codes, where: the transmitting end obtains the required codeword by Turbo-ARQ encoding of the information frame to be transmitted, and enters the codeword into a truncation circuit to obtain a different codeword. Truncate the codeword and store it in the originating buffer for retransmission;

The turbo-ARQ decoding structure is composed of two turbo product code decodings, where: the receiving end decodes the received turbo product codeword and the risk information retained by the previous transmission process through the first turbo product code The code performs internal iterative Turbo decoding;

The external information output by the internal iterative Turbo decoding, the prior information retained in the previous transmission process, and the corresponding codeword are sent to a second Turbo product code decoding for further decoding; the further decoding The output external information can be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding. The Turbo-ARQ coding structure can be formed by concatenating a Turbo product code and a convolution code in parallel, where: · The sender obtains the required codeword by Turbo-ARQ encoding of the information frame to be sent, and inputs the codeword The truncation circuit obtains different truncated codewords and stores them in the sending buffer for retransmission;

The turbo-ARQ decoding structure is composed of a turbo product code decoding and a convolution code decoding, wherein: the receiving end transmits the received turbo product codeword and the a priori information retained in the previous transmission process through the turbo Product code decoding performs internal overlapping K Turbo decoding; the external information output by the internal iterative Turbo decoding, the a priori information retained in the previous transmission process, and the corresponding codeword are sent to the convolutional code decoding to do Further decoding; the external information output by the further decoding may be used as part of the prior information decoded by the Turbo product code to perform external iterative turbo decoding.

The Turbo-ARQ coding structure may be an interleaving coding structure;

The external information outputted by the internal iterative Turbo decoding is interleaved with the first face information retained in the previous transmission process, and then is sent to the member code decoding for further decoding with the corresponding codeword; the further decoding After de-interleaving, the output external information can be used as part of the prior information of the turbo product code decoding to form an outer re-iterating turbo decoding.

The member code may be a block code.

The member code may be a convolutional code.

The member codes may be parallel or serial concatenated convolutional codes.

The member code may be a concatenated block code, including a Turbo product code.

The member code may be a concatenation of a block code and a convolutional code.

The steps of the hybrid ARQ method include:

The transmitting end sends the Turbo product code code for the first time; The receiving end decodes the corresponding received codeword;

If the CRC check indicates that the decoding is correct, the frame data is accepted, and an ACK signal is fed back to notify the sender to send the next frame data;

According to the CRC check, if the decoding fails, the corresponding codeword and the external information corresponding to the decoded output are stored in the receiving buffer and fed back to the sender with a NAK signal requesting retransmission; the sender receives the first NAK signal Then, retransmit the backup codeword stored in the originating buffer;

The receiving end combines the received retransmitted data with the data in the receiving end buffer to obtain a new codeword;

Turbo-ARQ decoder for error correction of new codewords;

The external information corresponding to the decoding output of the previous transmission process is also sent to the Turbo-ARQ decoder as the prior information of this decoding attempt;

The data frame is accepted after the CRC check that the decoding is successful;

After the CRC check, the decoding is considered to be failed. The combined new codeword and the external information corresponding to the decoded output will be stored in the cache to replace the original data. At the same time, the K signal will be fed back to the sender to request the data to be retransmitted;

This process continues until the data frame is received correctly.

The beneficial effects of the present invention are: The new hybrid ARQ method proposed by the present invention is an effective hybrid ARQ method based on the Turbo product code and the Turbo-ARQ structure. It uses a combination of TPC coding and Turbo-ARQ structure. Can provide better system performance, where:

In the present invention, a simple TPC codeword is used as the first transmission data. Product codes are a class of good codes with strong error correction capabilities and simple code construction, and are particularly suitable for use in complex interference channel environments. The product code using the Turbo iterative decoding scheme is TPC. The system can obtain a more flexible code rate by selecting subcodes reasonably and truncating them appropriately. J. Hagenauer in the literature (J. Hanenauer, Itera tive Decoding of Binary Block and Convolut ion Codes, IEEE Trans. On Information Theory, vol. 42, No. 2, Mar. 1996.) pointed out that when the code rate is greater than 2/3, the performance of the TPC scheme is better than the PCCC scheme. In addition, TPC is more suitable for short frame structures.

The invention also selects a Turbo-ARQ structure. This is a parallel cascading scheme that comprehensively utilizes Turbo encoding and iterative decoding. Its member codes can be block codes, convolutional codes, and cascaded forms of the two-any one suitable for soft-in and soft-out decoding algorithms. Code. Moreover, this scheme does not add much burden to the complexity of the system. In the Turbo-ARQ structure proposed by the present invention, the first member code is TPC, and the remaining member codes may be block codes, convolution codes, parallel or serial concatenated convolution codes, concatenated block codes, and block codes and The concatenation of convolutional codes. The Turbo interleaver between member codes is optional.

The decoding scheme of the present invention uses dual Turbo iterative decoding. The receiving end adopts iterative decoding for the first received TPC codeword. The present invention may also select a TPC iterative decoding algorithm based on subcode adjoint decoding. The inventors of the method are: Li Zongwang, Xu Youyun, and the invention name is: Iterative cascade block code based on subcode adjoint decoding The decoding method is disclosed in the invention patent application PCT / CNOl / 01289. The advantage is that it can obtain better decoding performance without increasing the complexity of the algorithm, and can support more types of subcodes. On the other hand, the Turbo-ARQ decoder combines the current retransmitted data with the codewords that were previously stored in the cache and failed to decode them in a certain way (including possible Chase combinations), and then performs iterative outer decoding. At the same time, external information corresponding to the decoding output of the previous transmission process is used as prior information for this decoding attempt.

The hybrid ARQ method proposed by the present invention adopts a technology combining a TPC coding and a Turbo-ARQ structure, and its effect is as follows: Since the hybrid ARQ method proposed by the present invention uses the TPC coding, the probability of successful first transmission is increased. At the same time, its simple code structure and decoding algorithm simplify the encoding and decoding equipment, and speed up the decoding processing speed. The method of the present invention uses a Turbo-ARQ structure with TPC as the first member code, which not only makes full use of the advantages of TPC, but also comprehensively utilizes the useful information of each transmitted data, thereby improving the system performance to a greater extent.

The selection of dual Turbo iterative decoding in the method of the invention improves the decoding performance of the system and increases the reliability and effectiveness of the system.

The method of the present invention comprehensively utilizes the advantages of the TPC coding and the Turbo-ARQ structure, so that the system can achieve better performance without adding a large complexity burden. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows a f-diagram of a turbo-ARQ coding structure according to the present invention;

FIG. 2 is a block diagram of a turbo-ARQ decoding structure according to the present invention;

FIG. 3 shows a working flowchart of the hybrid ARQ method of the present invention;

Figure 4 shows a block diagram of a CDMA system using the hybrid ARQ method of the present invention. detailed description

Example 1

As shown in FIG. 1 and FIG. 2, the present invention provides a hybrid ARQ method for wireless channel packet data transmission, which includes: the transmitting end sends only a single Turbo product code word during the first transmission, and the transmitting end is retransmitting In the process, the truncated codeword output through the Turbo-ARQ coding structure is sent to respond to the retransmission request fed back due to the decoding failure; the receiving end uses internal re-iteration Turbo decoding to decode the turbo product codeword. The end-to-end retransmission data decoding adopts the external re-iteration Turbo decoding through the Turbo-ARQ decoding structure.

The Turbo-ARQ coding structure is formed by parallel concatenation of two or more member codes, where: the first member code used for the parallel concatenation is a Turbo product code; The Turbo-ARQ encoding structure further includes a truncation circuit, wherein: the transmitting end obtains a required codeword by Turbo-ARQ encoding of an information frame to be transmitted, and inputs the codeword into the truncation circuit to obtain a different truncated codeword, It is stored in the originating buffer for retransmission.

The Turbo-ARQ coding structure may be a coding structure including interleaving.

The external information output by the internal iterative Turbo decoding, the prior information retained in the previous transmission process, and the corresponding codeword are sent to a second Turbo product code decoding for further decoding; the further decoding The output external information can be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding. The Turbo-ARQ decoding structure may be composed of one turbo product code decoding and one convolution code decoding, wherein: the receiving end transmits the received turbo product codeword and the a priori information retained in the previous transmission process through the turbo The product code decoding performs internal iterative turbo decoding; the external information output by the internal iterative turbo decoding, the a priori information retained in the previous transmission process, and the corresponding codeword are sent to the convolutional code decoding for further Decoding; the external information output by the further decoding may be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

The external information outputted by the internal iterative Turbo decoding is interleaved with the a priori information retained in the previous transmission process, and then is sent to the member code decoding for further decoding with the corresponding codeword; the further decoding After de-interleaving, the output external information can be used as part of the prior information of the turbo product code decoding to form an outer re-iterating turbo decoding.

The external information outputted by the internal iterative Turbo decoding and the prior information retained in the previous transmission process and the corresponding codeword are sent to the member code decoding for further decoding; the external information outputted by the further decoding It can also be used as part of the prior information of the turbo product code decoding to form an outer-reiterating iterative turbo decoding. The Turbo-ARQ coding structure can be formed by parallel concatenation of two Turbo product codes, where: the transmitting end obtains the required codeword by Turbo-ARQ encoding of the information frame to be transmitted, and enters the codeword into a truncation circuit to obtain a different Truncate the codeword and store it in the originating buffer for retransmission;

The turbo-ARQ decoding structure is composed of two turbo product code decodings, where: the receiving end decodes the received turbo product codeword and the a priori information retained in the previous transmission process by the first turbo product code The code performs internal iterative Turbo decoding;

The Turbo-ARQ coding structure can be formed by parallel concatenation of a Turbo product code and a convolutional code, where: the sender obtains the required codeword by Turbo-ARQ encoding of the information frame to be transmitted, and truncates the codeword input. The circuit obtains different truncated codewords and stores them in the originating buffer for retransmission;

The turbo-ARQ decoding structure is composed of a turbo product code decoding and a convolution code decoding, wherein: the receiving end transmits the received turbo product codeword and the a priori information retained in the previous transmission process through the turbo The product code decoding performs internal iterative turbo decoding; the external information output by the internal iterative turbo decoding, the a priori information retained in the previous transmission process, and the corresponding codeword are sent to the convolutional code decoding for further Decoding; the external information output by the further decoding may be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

The Turbo-ARQ coding structure may be a coding structure including interleaving; the Turbo-ARQ decoding structure may be a decoding structure including interleaving and deinterleaving, wherein: the turbo product codeword to be received by the receiver The a priori information retained from the previous transmission process is subjected to internal iterative turbo decoding via Turbo product code decoding; The external information outputted by the internal iterative Turbo decoding is interleaved with the a priori information retained in the previous transmission process, and then is sent to the member code decoding for further decoding with the corresponding codeword; the further decoding The output external information can be used as the 0 after de-interleaving.

The member code may be a block code.

The member code may be a convolutional code.

The member codes may be parallel or serial concatenated convolutional codes.

As shown in FIG. 3, the working process of the present invention is: The sending end sends the TPC codeword C for the first time.

(Step 501), the receiving end decodes the corresponding received codeword C (step 50 ² ), and if the CRC check (step 503) determines that the decoding is correct, the frame data is accepted, and an ACK signal is fed back Notify the sender to send the next frame of data (step 510); otherwise, if the decoding fails, store C. and its corresponding decoded external information in the receiving buffer, and feed it back to the sender with a NAK signal requesting retransmission. (Step 504). After receiving the first NAK signal, the transmitting end retransmits the codeword Cj (step 505). At this time, the user 1 and the receiving end combine the received C / and G in the receiving end buffer to obtain ( (Step 506), and then use the Turbo-ARQ decoder to perform error correction (step 507). Corresponding to G, the external information output from the decoding is also sent to Turbo at the same time as the prior information of this decoding attempt. -ARQ decoder, if the CRC check (step 508) considers that the decoding is successful, the data frame is accepted (step 510); otherwise, if the decoding fails, the combined (and its corresponding decoded output external The information will be stored in the cache to replace the original data, and the K signal will be fed back The sender requests the + 1th retransmission of the codeword (step 509). This process continues until the data frame is received correctly.

The novel hybrid ARQ method proposed by the present invention is an effective hybrid ARQ method based on the Turbo product code and the Turbo-ARQ structure. It uses a combination of TPC coding and Turbo-ARQ structure, which can provide a better system. Performance, where: In the present invention, a simple TPC codeword is used as the first transmission data. Product codes are a class of good codes with strong error correction capabilities and simple code construction, and are particularly suitable for use in complex interference channel environments. The product code using the Tu r bo iterative decoding scheme is TPC. The system can obtain a more flexible code rate by selecting subcodes reasonably and truncating them appropriately. J. Hagenauer in the literature (J. Hanenauer, Iterative Decoding of Binary Block and Convolutiona l Codes, IEEE Trans. On Information Theory, vo l. 42, No. 2, Mar. 1996.) states that when When the bit rate is greater than 2/3, the performance of the TPC scheme is better than the PCCC scheme. In addition, TPC is more suitable for short frame structures.

The invention also selects a Turbo-ARQ structure. This is a parallel concatenation scheme that uses Turbo encoding and iterative decoding. Its member codes can be block codes, convolutional codes, and concatenated forms of the two. They are suitable for soft-in and soft-out decoding algorithms. code. Moreover, this scheme does not add much burden to the complexity of the system. In the Turbo-ARQ structure proposed by the present invention, the first member code is TPC, and the remaining member codes may be block codes, convolution codes, parallel or serial concatenated convolution codes, concatenated block codes, and block codes and The concatenation of convolutional codes. The Turbo interleaver between member codes is optional.

The decoding scheme of the present invention uses dual Turbo iterative decoding. The receiving end adopts iterative decoding for the first received TPC codeword. The present invention may also select a TPC iterative decoding algorithm based on subcode accompanying decoding. The advantage is that it can obtain better decoding performance without increasing the complexity of the algorithm, and can support more types of subcodes. On the other hand, the Turbo-ARQ decoder combines the current retransmitted data with the codewords that were previously stored in the cache and failed to decode them in a certain way (including possible Chase combinations), and then performs iterative outer decoding. At the same time, the external information corresponding to the decoding output of the previous transmission process is used as a priori information for this decoding attempt.

The hybrid ARQ method proposed by the present invention adopts a technology combining TPC coding and Turbo-ARQ structure, and the effect is: Since the hybrid ARQ method proposed by the present invention uses TPC coding Code, which increases the probability of successful first transmission. At the same time, its simple code structure and decoding algorithm simplify the encoding and decoding equipment and speed up the decoding processing speed.

The method of the present invention adopts the Turbo-ARQ structure with TPC as the first code, which not only makes full use of the advantages of TPC, but also comprehensively utilizes the useful information of each transmission data, thereby improving the system performance to a greater extent.

The method of the present invention comprehensively utilizes the advantages of the TPC coding and the Turbo-ARQ structure, so that the system can achieve better performance without adding a large complexity burden.

Example 2

As shown in FIG. 4, a block diagram of a CDMA system using the hybrid ARQ method of the present invention is given. The system transmitter consists of a CRC encoder 10, a turbo encoder 11, a modulator 12, and a spread-spectrum device 13. The sender first frames the information bits to be sent with a fixed length, and then uses the CRC encoder 10 to add check bits for error detection (the system should design the CRC check bits long enough so that the probability of undetectable errors in the system design The allowable range can be ignored), and then sent to the Turbo-ARQ encoder 11 to encode at the code rate required by the design. The sender first sends a single TPC codeword to the modulator 12, and at the same time, the output of the Turbo-ARQ encoder obtains different truncated codewords according to a certain truncated matrix. The modulated symbols are spread in the spreading device 13 and finally reach the receiver of the CDMA system via the wireless channel 14.

The CDMA receiver includes a despreading device 15, a demodulator 16, a combiner 17, a Turbo-ARQ decoder 18, and a CRC decoder 19. At the receiving end, the despreading device 15 and the demodulator 16 first perform the despreading and demodulation function on the received codeword, and then use the combiner 17 to appropriately combine the current received codeword with the data previously retained in the buffer, and then Turbo -The ARQ decoder 18 codes the combined data, and then uses the CRC decoder 19 to detect errors. If the decoding is correct, it receives the data and feeds back an ACK (Acknowledge) signal to notify the originator; otherwise, such as decoding Error, the receiver will fail to decode the codeword and its external information It is stored in the receiving buffer, and a NAK (Nega Tive Acknowl edge) signal is fed back through the feedback channel to request the retransmission of the data at the transmitting end.

The effect of the method of the present invention in a CDMA system is as follows: Since the hybrid ARQ method proposed by the present invention uses TPC coding, the probability of successful first transmission is increased. At the same time, its simple code structure and decoding algorithm simplify the encoding and decoding equipment, and speed up the decoding processing speed.

The method of the present invention uses a Turbo-ARQ structure with TPC as the first member code. Not only does it take full advantage of TPC, but it also comprehensively utilizes the useful information of each transmitted data, thereby improving the system performance to a greater degree. .

The selection of silent turbo iterative decoding in the method of the invention improves the decoding performance of the system and increases the reliability and effectiveness of the system.

The above embodiments are only used to illustrate the present invention, but not intended to limit the present invention.

Claims

Rights request

1. A hybrid ARQ method for wireless channel packet data transmission, comprising: the transmitting end sends only a single Turbo product codeword during the first transmission, and the transmitting end outputs the result through the Turbo-ARQ encoding structure during retransmission The truncated codeword is used to respond to the retransmission request fed back by the receiving end due to decoding failure. The receiving end uses internal iterative Turbo decoding for decoding the turbo product codeword, and the receiving end uses Turbo- External De-Iteration Turbo Decoding of ARQ Decoding Structure.

2. The method according to claim 1, characterized in that: the Turbo-ARQ coding structure is formed by parallel concatenation of two or more member codes, wherein: the first member code used for parallel concatenation is Turbo product code;

3. The method according to claim 1, characterized in that: the Turbo-ARQ coding structure can be formed by concatenating two Turbo product codes in parallel and adding a truncation circuit, wherein: the transmitting end sends the information frame to be transmitted through Turbo-ARQ encoding obtains the required codeword, enters the codeword into the truncation circuit to obtain a different truncated codeword, and stores it in the originating buffer for retransmission.

4. The method according to claim 1, characterized in that: the Turbo-ARQ coding structure can be formed by concatenating a Turbo product code and a convolution code in parallel and adding a truncation circuit, wherein: the transmitting end is to be waited for. The transmitted information frame is encoded by Turbo-ARQ to obtain the required codeword, and the codeword is input to the truncation circuit to obtain a different truncated codeword, which is stored in the transmitting buffer for retransmission.

5. The method according to claim 2 or 3 or 4, wherein: the Turbo-ARQ coding structure is a coding structure including interleaving.

6. The method according to claim 1, wherein: the Turbo-ARQ decoding structure is composed of turbo product code decoding and other member code decoding, wherein: the receiving end will receive the turbo product code code. The word and the prior information retained in the previous transmission process are decoded by Turbo product code to perform internal iterative Turbo decoding;

7. The method according to claim 1, wherein: the Turbo-ARQ decoding structure is composed of two Turbo product code decodings, wherein: the receiving end receives the received Turbo product codeword from the previous transmission The a priori information retained in the process is decoded by the first Turbo product code to perform iterative T _ur bo decoding;

The external information output by the internal iterative Turbo decoding and the a priori information retained in the previous transmission process and the corresponding codeword are sent to a second Turbo product code decoder for further decoding; the further decoding The output external information can be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

8. The method according to claim 1, wherein: the Turbo-ARQ decoding structure comprises one Turbo product code decoding and one Convolution code decoding, wherein: the Turbo product to be received by the receiving end The code codeword and the prior information retained in the previous transmission process are decoded by Turbo product code to perform internal iterative Turbo decoding;

The external information output by the internal iterative Turbo decoding and the a priori information retained in the previous transmission process and the corresponding codeword are sent to the convolutional code decoding for further decoding; the externalities of the further decoding output The information can be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

9. The method according to claim 6 or 7 or 8, wherein:

The Turbo-ARQ decoding structure may be a decoding structure including interleaving and deinterleaving, where: the receiving end The turbo product code word received and the prior information retained in the previous transmission process are decoded by the turbo product code to perform internal iterative turbo decoding;

10. The method according to claim 1, wherein: the Turbo-ARQ coding structure is formed by parallel concatenation of two or more member codes, wherein: the first member code used for parallel concatenation is Turbo product code;

The Turbo-ARQ coding structure further includes a truncation circuit, where the transmitting end obtains a required codeword by Turbo-ARQ encoding of the information frame to be transmitted, and inputs the codeword into the truncation circuit to obtain a different truncated codeword, And store it in the originating buffer for retransmission;

11. The method according to claim 1, wherein: the Turbo-

The ARQ coding structure can be formed by concatenating two Turbo product codes in parallel, where: the transmitting end obtains the required codeword by Turbo-ARQ encoding of the information frame to be transmitted, and inputs the codeword into a truncation circuit to obtain different truncated codewords, And store it in the originating buffer for retransmission;

The turbo-ARQ decoding structure is composed of two turbo product code decodings, where: the receiving end decodes the received turbo product codeword and the a priori information retained in the previous transmission process by the first turbo product code The code performs internal iterative Turbo decoding; The external information output by the internal iterative Turbo decoding, the prior information retained in the previous transmission process, and the corresponding codeword are sent to a second Turbo product code decoding for further decoding; the further decoding The output external information can be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

12. The method according to claim 1, wherein: the Turbo-

The AQ coding structure can be formed by parallel concatenation of a Turbo product code and a convolutional code, where: The transmitting end obtains the required codeword by Turbo-ARQ encoding of the information frame to be transmitted, and enters the codeword into a truncation circuit to obtain different truncations. Short codeword and store it in the originating buffer for retransmission;

The turbo-ARQ decoding structure is composed of a turbo product code decoding and a convolution code decoding, wherein: the receiving end transmits the received turbo product codeword and the a priori information retained in the previous transmission process through the turbo The product code decoding performs internal iterative turbo decoding; the external information output by the internal iterative turbo decoding and the a priori information retained in the previous transmission process and the corresponding codeword are sent to the convolutional code decoding for further Decoding; the external information output by the further decoding may be used as part of the prior information of the turbo product code decoding to perform external iterative turbo decoding.

13. The method according to claim 10 or 11 or 12, wherein: the Turbo-ARQ coding structure is a coding structure including interleaving;

14. The method according to claim 2 or 6 or 10, wherein: the member code is a block code.

15. The method according to claim 2 or 6 or 10, wherein the member code is a convolutional code.

16. The method according to claim 2 or 6 or 10, wherein: the member codes are parallel or serial concatenated convolutional codes.

17. The method according to claim 2 or 6 or 10, characterized in that: the member code is a concatenated block code, including a turbo product code.

18. The method according to claim 2 or 6 or 10, wherein: the member code is a concatenation of a block code and a convolutional code.

19. The method according to claim 1, wherein the steps of the hybrid ARQ method include:

The transmitting end sends the Turbo product code code for the first time;

The receiving end decodes the corresponding received codeword;

Turbo-ARQ decoder for error correction of new codewords;

The CRC check indicates that the decoding is successful, then the data frame is accepted; After the CRC check, it is considered that the decoding has failed. The combined new codeword and the external information corresponding to the decoded output will be stored in the cache to replace the original data. At the same time, the NAK signal is fed back to the sender to request the data to be retransmitted;

This process continues until the data frame is received correctly.