WO2003030424A1

WO2003030424A1 - Error control method for mobile communication system

Info

Publication number: WO2003030424A1
Application number: PCT/CN2001/001466
Authority: WO
Inventors: Chundi Xiu; Youghui Li
Original assignee: Linkair Communications,Inc.
Priority date: 2001-09-29
Filing date: 2001-09-29
Publication date: 2003-04-10
Also published as: CN1476692A

Abstract

Present invention discloses a new RCPT/II type of HARQ method, that is to say, RCPT/CPC HARQ method, for the packet data service in CDMA communication system. The method uses the RCPT coding scheme of the code rate matching Turbo code, the alternate re-transmitting scheme of code word check, and the form for combining the Chase combining with increment redundancy IR method, that is, responding to NAK signal of feedback due to decoding lose at the receiving end, in alternately re-transmitting higher code rate RCPT child and check code block; when the receiving end decode, it making Chase combining for received repetitive high code rate child code, while the combination of the high code rate child with the redundancy increased in check code block are made into low code rate parent code, thereby increasing error correction capability for FEC system. Unsing the method of present invention, the decoding complexity and the decoding delay are reduced, system design is simplified, and full IR performance is achieved.

Description

Method for controlling error in mobile communication system

The present invention relates to an error control method in a mobile communication system, and in particular, to an error control method suitable for a high-speed packet data service of a Code Division Multiple Access CDMA (Code Division Multiple Access) communication system. technical background

The third generation mobile communication (3G) system is required to support multiple services such as voice, image, and data, especially to be able to support multimedia and high-bit-rate packet data services. In order to achieve high data rate transmission, an efficient and reliable communication mechanism must be used. For this reason, a high-dimensional modulation method is applied to a 3G system, thereby increasing the peak rate of the system. However, the problem caused by this is that the reliability of the system is seriously challenged, because on the wireless channel, multipath, shadow, and Doppler frequency shifts will seriously degrade the performance of the high-dimensional system, causing random errors and bursts. Coexist errors. If an effective error control mechanism is not adopted, it is difficult to achieve reliable data transmission. It is bound to fail to meet the requirements of data communication with a bit error rate less than 10 " ^6. In 3G systems, it is recommended to use hybrid error correction HARQ (Hybrid Automat ic Repeat). reques t) is used as a system link-level error control technology to meet the requirements for Quality of Service (QoS).

There are three basic error control methods in a data communication system: forward error control (FEC), error detection and retransmission ARQ (Automatic Repeat Reques t), and: combined error correction HARQ. The HARQ method refers to any combination of the FEC method and the ARQ method, that is, the sender first performs FEC encoding on the information bits to be sent, and then sends the encoded codeword to the receiver. If the receiver finds that there is a When the error is corrected, a veto letter is fed back through the feedback channel. Retransmission gradually increases the redundancy (or check code) in response to each NAK signal at the receiving end. ^ According to how much IR is redundant in the retransmission data, it can be divided into two ways: Partial IR and Full IR. Partial IR means that each retransmitted data is composed of information bits and newly added redundancy, and the receiving end combines each received data to form a codeword with a lower bit rate; thereby improving the system's error correction capability. Since each retransmitted codeword in this method can be decoded separately, it is also called multiple redundant type III HARQ. Full IR is a type II HARQ scheme. The sender responds to the signal that the receiver failed to decode with only the increased redundancy. The receiver combines the previously received data with the redundancy of the retransmission to form a more ^ [ Codeword code rate, to improve the performance of the system to a greater extent.

The above-mentioned HARQ method in the prior art has the following disadvantages: Although the Chase combining method is simple and easy to implement, its performance is slightly worse; the full IR method has the best performance, but with the increase of the number of retransmissions, the buffer requirements at the sending and receiving ends The amount of error is increased, and the enhancement of the error correction capability of FEC codes is also at the cost of more complicated decoding algorithms and decoding equipment. Although part of the IR is between the first two in terms of performance and complexity, this method can The system performance provided is also not optimal. Object of the invention

In order to solve the above-mentioned shortcomings and deficiencies in the prior art, the present invention proposes a new RCPT / II type HARQ method. It can be called RCPT / CPC (alternate "Code retransmission" and "Parity retransmission" & Chase combing) HARQ method. This method combines the code rate matching truncated Turbo code RCPT encoding with the codeword check alternate retransmission scheme, and combines the advantages of the Chase combining and IR methods, which can not only improve the performance of the HARQ system, but also reduce the system. Complexity.

The present invention is implemented using the following technical solutions:

The RCPT / CPC HARQ method of the present invention is characterized by: adopting RCPT coding and coding The word resurgence alternate retransmission scheme and the technology of combining Chase combining and IR methods, that is, the transmitting end retransmits the higher code rate RCPT subcode and check code group in response to the MK signal returned by the receiving end due to decoding failure. ; Chase comb ining the received repeated high code rate subcodes at the receiving end decoding, and combine the increased redundancy with the high code rate subcodes to form a low code rate mother code to improve the error correction capability of the FEC system , So as to optimize the entire system in terms of performance and complexity.

The RCPT / CPC HARQ method of the present invention further includes the following steps:

1) The sender performs RCPT error correction coding on the bits to be sent to obtain the RCPT subcode A and the check code group B;

2) The sending end sends the subcode A, and the receiving end decodes it accordingly and performs a CRC check;

3) Corresponding to the first NAK signal, the transmitting end retransmits the check code group B, and when decoding is performed at the receiving end, the received additional check code group is combined with the subcode received in step 2 and then decoded. ;

4) Corresponding to the second NAK signal, the transmitting end retransmits the subcode A again; when the receiving end decodes, all received repeated subcodes are first subjected to Chase combining, and then the check code received in step 3 is used. Combine and decode again;

5) Corresponding to the ACK signal, the sender sends the next frame, and the receiver accepts the decoded data frame and sends it to the user.

By adopting the RCPT / CPC HARQ method of the present invention, the system uses the codeword check alternate retransmission scheme, so that the system no longer needs the mother code with too low code rate in the full IR method, thereby reducing decoding complexity and The decoding delay reduces the overhead of the receiving buffer, and because the present invention combines the Chase combing technology, the system obtains performance comparable to that of the full IR. jg illustration

The following describes the present invention in detail with reference to the accompanying drawings and preferred embodiments of the present invention. In the drawings: FIG. 1 shows a block diagram of a DS-CDMA system using a DSSS method of the present invention; FIG. 2 is a structural diagram of an RCPT code used by the method of the present invention;

FIG. 3 is a schematic diagram of a coding and decoding scheme corresponding to the three retransmission processes in the RCPT / CPC HARQ method of the present invention;

FIG. 4 shows a working flowchart for implementing the method of the present invention.

The RCPT / CPC HARQ method of the present invention can be applied to any wireless system that needs to provide high-speed packet data services. The DS-CDMA system, more specifically, the physical layer of the system is taken as an example to describe the present invention for high-speed packet data services. However, the present invention is not limited to this. Preferred embodiment

In the physical layer of the existing mobile communication system, the choice of the error control method depends on the quality of service QoS and wireless channel required by the system. For packet data services, the bit error rate of the system output must be 1 (Γ ^δ or better. Therefore, under harsh channel conditions, the system requires FEC codes with strong error correction capabilities to ensure reliable communication of the system. However, when When channel conditions continue to deteriorate (especially in various complex channel environments such as deep fading or multiple access interference), even the use of a 1/3 bit rate Turbo code as the error correction code will fail. Therefore, the HARQ technology due to the combination Because of the high reliability of the ARQ method and the high throughput efficiency of the FEC method, they are widely used in error control systems at the physical layer.

The new error control method proposed by the present invention is a new HARQ method suitable for TDD system based on RCPT coding, alternate retransmission of codeword check and combination of Chase combining method and IR method, which is called RCPT The / CPC HARQ method, which uses a combination of a codeword check alternate retransmission scheme and a Chase combining technology, reduces the complexity of the full IR method, and can achieve the optimal performance that the full IR method can provide. Compared with the traditional HARQ technology, the RCPT / CPC HARQ method mainly has the following main advantages. 1. FEC system uses RCPT code for error correction. RCPT codes are a family of codes generated by using a low bit rate Turbo code as the mother code, following the bit rate matching principle and truncating in a certain way. The system achieves reliable data transmission by properly selecting the check bits required for each retransmission to form a lower code rate RCPT code.

2. The ARQ system uses a codeword check alternate retransmission scheme. The sender alternately retransmits the check code group and the RCPT subcode, and the receiver alternately stores the received RCPT subcode and check code group accordingly.

3. Based on 1, 2, the RCPT / CPC HARQ method proposed by the present invention uses the technology of combining the Chase combining method and the IR method, which is also the core idea of the present invention.

The first advantage is that the RCPT / CPC HARQ method uses the RCPT code to make the system obtain the best reliability and effectiveness, because the Turbo code is the best error correction code in today's coding theory, and the code rate matching principle In addition, the code rate design of the error correction code is adapted to the channel conditions, which improves the effectiveness of the system.

The second advantage explains that the RCPT / CPC HARQ method does not require a turbo mother code with a low code rate similar to the full IR method, which can reduce the number of stages of the turbo decoder, thereby reducing decoding complexity and decoding delay.

The third advantage shows that the RCPT / CPC HARQ method combines the advantages of the RCPT encoding and codeword check alternate retransmission scheme, and uses the combination of Chase comb ining and IR methods to reduce the complexity of the system, while achieving the same The best performance that the full IR method can achieve.

Referring to FIG. 1, a block diagram of a DS-CDMA system using the method of the present invention is shown in FIG. The system transmitter consists of a CRC encoder 101, an RCPT encoder 102, a spread spectrum modulator 103, and a symbol interleaver 104. The sender first frames the information bits to be sent with a fixed length, and then uses the CRC encoder 101 to add check bits for error detection. (The system should design the CRC check bits long enough so that the probability of undetectable errors is in the system design. The allowable range can be ignored), and then sent to the RCPT encoder 102 for encoding according to the mother code rate required by the design, and all codewords are stored in the originating buffer. The device prepares for retransmission. The encoded codewords are spread-spectrum-modulated in the spread-spectrum modulator 103, channel-interleaved in the symbol interleaver 104, and finally transmitted to the receiver of the C-plane A via the wireless channel 105.

The CDMA receiver includes a deinterleaver 106, a despread demodulator 107, an RCPT decoder 108, and a CRC decoder 109. At the receiving end, the deinterleaver 106 and the despread demodulator 107 first complete the deinterleaving and despread demodulation functions of the received codeword, and then use the RCPT decoder 108 to correct the received signal and use CRC decoding. The detector 109 detects an error. If the decoding is correct, the frame information is received. Otherwise, when there is an uncorrectable error code in the received codeword, a NAK signal is fed back through the feedback channel to request retransmission of the data at the transmitting end.

In FIG. 2, a structural diagram of an RCPT code applied in the method of the present invention is shown. In the preferred embodiment of the present invention, a turbo mother code with a code rate of k / 2n is used to perform FEC encoding on a k-bit length information frame, and then the check bits are truncated in a certain truncation manner to obtain a code rate. Is a subcode A of k / n, and the truncated n face correction bits form a face correction code group B for retransmission. 2, the subcode information A by the length of the block length k I and the check block is configured of k N-, group B efficacy code length n from the remaining parity block P ₂ configuration, the A and B binding The mother code C constituting the code rate is k / 2n.

FIG. 3 is a schematic diagram of a coding and decoding scheme corresponding to the three retransmission processes in the preferred embodiment of the RCPT / CPC HARQ method of the present invention.

According to a preferred embodiment of the present invention, the HARQ system selects a maximum number of retransmissions of three. At the transmitting end, the RCPT encoder is first used to obtain the sub code A and the school insurance code group B according to the RCPT code structure shown in FIG. 2. The transmitting end sends the RCPT subcode A for the first time with a code rate of k / n, and the receiving end RCPT decoder decodes the corresponding codeword A *; if there is an uncorrectable error, the transmitting end retransmits the check code group B responds to the first NAK signal at the receiving end, and the receiving end decoder corrects the received valid code group B * and A * retained in the buffer into a mother code C * with a code rate of k / 2n. ; If there is still an uncorrectable error, the transmitting end retransmits the RCPT subcode A again in response to the second NAK signal of the receiving end, and the receiving end will newly receive The data of 01 01466 and A * in the buffer are used for Chase combining to obtain A **, and A ** and the effective code group B * in the buffer are combined to form C ** and sent to the decoder.

Referring to FIG. 4, in the RCPT / CPC HARQ scheme proposed by the present invention, the sender first sorts and serializes information of a certain bit length (step 401), uses a CRC encoder to perform error detection coding (step 402), and then performs The RCPT error correction coding obtains an RCPT subcode A and a check code group B with a required code rate of k / n (step 403).

After the RCPT code is determined, the retransmission process of the RCPT / CPC HARQ method of the present invention includes the following steps: The sender first sends subcode A (step 404), and stores the subcode A and the check code group B in the sender buffer ( Step 405). The receiving end decodes the received subcode A * (step 407), if the CRC check (step 408) is performed correctly (step 409), the data frame is accepted and sent to the user (step 410), and Feedback to the sender with an ACK signal; otherwise, if an uncorrectable error is found in A * (step 409), store it temporarily in the receiving buffer (step 411), and feed back to the sender with a NAI (signal request for retransmission) After the sender receives the first NAK signal, it retransmits the check code group B in the sender's buffer (step 411), and the receiver combines the received B * with the A * retained in the buffer to form a low bit rate. The mother code O and decode C * (step 413), if the CRC detection (step 414) is decoded correctly (step 415), accept the data frame (step 410); otherwise, if the decoding fails (step 415), B * is stored in the buffer (step 416), and the NA signal is fed back to the originator to request retransmission. The originator retransmits the subcode A in response to the second NAK signal (step 416), and the receiver will newly receive Chase combining the obtained subcode A * and A * in the buffer to obtain A ** (step 418), that is, the received subcode A * The complex subcode is weighted and combined according to the SNR conditions of the corresponding transmission channel, and then A ** and B * in the buffer are combined to form a new C ** and sent to the decoder (step 419) for decoding. If the CRC check (step 420) decodes correctly (step 421), the data frame is accepted (step Step 410); otherwise, if the decoding still fails (Steps 420, 421), the packet is discarded (Step 422), and the next packet is processed.

Although the present invention has been described in detail above with reference to the accompanying drawings, the present invention is not limited to this. According to the method of the present invention, those skilled in the art can easily implement the present invention or improve it by other methods. It is known that any improvement without departing from the idea of the present invention is within the scope of the claims of the present invention.

Claims

Claim

1. An error control method, characterized in that the transmitting end responds to a rejection signal returned by the receiving end due to a decoding failure by alternately retransmitting a higher bit rate bit rate to match the truncated Turbo code RCPT subcode and a check code group. NAK signal; when decoding at the receiving end, Chase combining is performed on the received repeated high code rate sub-codes, and the added check code group and high code rate sub-codes are combined to form a mother code of 4's code rate Decoding.

2. The error control method according to claim 1, further comprising the following steps:

1) The sender performs RCPT error correction coding on the bits to be sent to obtain the RCPT subcode A and the face correction group B;

3) Corresponding to the first NAK signal, the transmitting end retransmits the check code group B, and when the receiving end decodes, the received additional correction code group is combined with the subcode received in step 2 and then decoded. ;

3. The error control method according to claim 1, wherein the Chase combining is based on the signal-to-noise ratio conditions of the corresponding transmission channel for all received subcodes.

4. The error control method according to claim 1, 2 or 3, characterized in that, by a feedback signal Channel feedback NAK signal, ACK signal.

5. The error control method according to claim 2, wherein the maximum number of retransmissions is

3 times.

6. The error control method according to claim 1, further comprising the step of storing a check code.

II