CN1144429C - Packet check data transmission method - Google Patents

Abstract

The present invention relates to a packet check data transmission method which comprises: packet checking and error correcting codes are introduced into transmitted data frames, so that the positions of error packets can be determined, and further the bits of the errors can be corrected. The present invention greatly reduce the frame error rate and the retransmission rate of communication systems, increase system efficiency, and greatly save frequency resources particularly for wireless communication systems; the present invention can be widely applied to the communication field and the computer network field.

Description

Packet check data transmission method

Technical field: The present invention relates to a communication and computer system data transmission method, especially a packet verification data transmission method.

Background technology: In the field of communication and computer network, data transmission mainly adopts data frame transmission mode. Generally speaking, each frame consists of three parts: frame header, data and check code (Cyclic Redandency Check or CRC). The frame header is used to sort the received data packets, independent of the data and CRC part, and the check part is used to check whether the received useful data part is correct. When the received data cannot pass the CRC check, it means that the frame has an error during transmission, even if only 1 bit in the frame is wrong, the data packet must be retransmitted. This is because the function of the CRC check is only to detect whether the data block is received correctly, and it has no ability to correct errors. Retransmission in a communication system will reduce system transmission efficiency and bring system reception delay at the same time. A block diagram of an existing frame transmission communication system is shown in FIG. 1 .

Invention content:

The purpose of the present invention is to provide a method for reducing the frame error rate of the communication system and reducing the receiving time delay.

The packet verification data transmission method of the present invention, its steps are:

1 Convert the data source into a P-ary data flow sequence, where P≥2, which is an integer;

2. Divide the data stream sequence into data blocks, and add frame check code and other system information to each data block to form an L-bit long transmission end data frame sequence, where L is an integer greater than 1;

3 M data frames are arranged in parallel as M×L data matrix, and the matrix data or the matrix data without system information is divided into two or more data groups according to the column direction, and a group check code is added to each group to form a message to be transmitted , where M is an integer greater than or equal to 1;

Data sets can be of equal dimensions or overlap each other;

4 After the information to be transmitted is modulated, the transmitting end sends it to the transmission channel, and transmits it in the channel;

5. Demodulate the received signal at the receiving end to obtain receiving information including M data frames at the receiving end;

6 For each data frame at the receiving end, perform frame verification according to the frame verification code, and identify the frame that passes the frame verification as a correct received frame; for the data frame that does not pass the frame verification, then

6-1 Check each group according to the group check code to find out the error data group;

6-2 Perform a digital replacement of the possibly erroneous data of the frame in the erroneous data group to generate a new frame, and perform a frame check on the new frame;

When the number of data frames that may fail the frame check is greater than the predetermined limit

, the data frame can also be directly identified as an error frame and retransmission is required.

6-3 If the new frame passes the frame check, it is determined that the new frame data is a correctly received frame;

If the new frame does not pass the frame verification, perform the next replacement and verification;

  If all other combinations are replaced, or the number of replacements is higher than the established upper limit, the

The frame was received in error and requires retransmission.

In this method, the forward error correction coding technology can also be used to perform error correction coding on the data frame of the sending end at the sending end, and then perform error correction decoding on the data frame of the receiving end and the new frame at the receiving end; Error correction encoding is performed on the transmitted information, and then error correction decoding is performed on the received information at the receiving end.

The present invention can determine the position of the error group by introducing a packet check error correction code into the transmitted data frame, and then perform bit correction on the error, greatly reducing the frame error rate and retransmission rate of the communication system, and improving the system efficiency, especially For wireless communication systems, it can greatly save frequency resources

Description of drawings:

Figure 1 Block diagram of the existing frame transmission communication system (taking P=2 as an example, other P-ary data are the same)

Fig. 2 frame transmission communication system block diagram of the present invention

Figure 3 Generation of parity frame

Figure 4 Determination of "possibly erroneous bits"

The frame error rate of the transmission system of the present invention and the comparison of the frame error rate of the existing transmission system in Fig. 5

The circles represent the frame error rate using the traditional method,

The solid line represents the theoretical result given by Equation 3,

The dotted line represents the frame error rate under the condition that the maximum allowable error bit number is not limited (called the ideal case).

implementation plan:

The existing frame traditional communication system can be decomposed into eight parts, as shown in Figure 1, 1 - the data source generates a binary data stream; 2 - the data stream is divided into data blocks, and each data block is generated for each data block The check code for checking whether the received data is correct or not, the original data, the check part and other system information are arranged to form a data frame; 3--the data frame is transmitted to the channel after modulation and other operations; 4--the channel is generally non- The ideal channel may cause some data in the data frame to be wrong during transmission; 5 --- demodulate the received signal to obtain an error or correct data frame; 6 --- generate a correction according to module 2 7---If the received data frame passes the verification, it is judged that the received data frame is correct; 8---If the received data frame does not pass the verification, it is judged that the received data frame is wrong And ask the transmitting end to retransmit.

The frame transmission communication system of the present invention has been improved on the basis of the existing frame transmission communication system, and the packet verification part has been increased, as shown in Figure 2, wherein 1 ^* -- the data in the data frame is divided into equal length or different Several groups of equal length, the data between groups can overlap. Each group adds a check code for verifying the correctness of the data received by this group (the method can be the same as that of module 2); 2 ^* -- in the case that the received frame does not pass the correctness check, according to the rule of 1 ^* Check each group to find out the wrong data group. Since the exact position of the error bit in the error group cannot be determined, all the bits in the error group are considered "possibly erroneous bits"; 3 ^* -- one or several bits in the error group in the original received frame are 0/ 1 Replacement, and generate a new frame; 4 ^* -- Perform frame verification on the newly generated frame, the method is the same as that of module 6. If the new frame passes the frame check, its data is considered to be correct; 5 ^* --If the new frame fails the frame check, replace the other bits of the error group with 0/1. Returns 3 ^* if all possible 0/1 permutations have not been exhausted. If all possible 0/1 permutations have been exhausted or the number of permutations exceeds the upper limit and the frame check still fails, then the received frame is determined to be an error received frame.

Assuming that the sum of the data before channel coding and the CRC part is N bits, for the packet check error correction method, the system needs to process M frames in parallel. That is, the N bits of M frames are formed into a matrix with M rows and N columns at the transmission end. The main point of block check is to generate a parity check frame in the M+1th row of the matrix, where each bit is the modulo 2 addition of all bits in each column, as shown in Figure 3. In order to reduce the bit error rate of the system, the transmitting end performs channel coding on M+1 frames respectively and then transmits them to the channel.

At the receiving end, channel decoding is first performed on M+1 frames respectively, and a traditional CRC check is performed on the first M frames. If a row passes the CRC check, it can be considered that each bit of the row is correct, as shown in row 1 and row 4 in FIG. 4 . For a frame that does not pass the CRC check, the traditional frame transmission method requires that the frame must be retransmitted. As for the block check correction method, since the parity line is introduced, the system can check the received data column by column. In this way, the position of the column where the error bit is located in the error frame can also be roughly determined. As shown in the example in FIG. 2 , the error bit positions of the received data are the 3rd bit in the 2nd row and the 7th bit in the 3rd row, denoted as (2, 3), (3, 7). The results of the parity check give the error codes in the third column and the seventh column. Considering that neither the 2nd row nor the 3rd row has passed the CRC check, the receiver thinks that the possible error bit positions are (2, 3), (2, 7) (3, 3), (3, 7), such as The four bits framed by dashed lines in the figure are shown. In this way, the approximate position of the error bit in each frame is determined.

Since the exact position of the erroneous bit cannot be clearly given by the column parity method, the erroneous bit cannot be corrected yet. Taking row 2 as an example, the error bit may be bit 3 or bit 7. In order to find the correct value of these two bits, 4 bit combinations need to be considered: 00, 01, 10, 11. Here, it is recommended to adopt the method of replacing one by one, that is, substitute these four combinations into the corresponding positions of the second frame one by one, and perform a CRC check for each substitution. When a certain combination passes the CRC check, it is considered that the error in this frame has been corrected. If the number of erroneous bits in the M frame is n, at most 2 ⁿ CRC checks are required to correct the errors.

The frame error rate of the system is greatly reduced by using the group verification method. Under ideal conditions, the frame error rate is only about one tenth of that of the traditional system. A comparison of frame error rate results using this method with traditional methods is given here. Both systems have the same spectral efficiency. In order to shorten the processing time of the receiver, it is assumed that the system limits that when the number of bit errors in the received M frames is greater than ErrMax, the receiver directly requires frame retransmission without error bit correction. Here ErrMax corresponds to the content involved in the above description (2), which means that when the number of detected bit errors is greater than a certain upper limit ErrMax, the system does not perform operations from 2 ^* to 5 ^* and directly requires retransmission. But this method makes the system frame error rate larger than ideal. The dotted and dotted lines in Fig. 5 show the results of the frame error rate when ErrMax is equal to 6 and 9. It can be seen that the frame error rate of the system has been significantly reduced in all cases.

Claims (5)

1. A packet verification data transmission method, the steps of which are 1-1 Convert the data source into a P-ary data flow sequence, where P≥2, which is an integer; 1-2 Divide the data stream sequence into data blocks, and add a frame check code and other system information to each data block to form an L-bit long transmission end data frame sequence, where L>1 is an integer; 1-3 Parallel M data frames into an M×L data matrix, and divide the matrix data or the matrix data without system information into two or more data groups in the column direction, add a group check code for each group, and form a waiting list Send information, where M≥1, which is an integer; 1-4 Modulate the information to be transmitted and send it to the transmission channel from the transmission end, and transmit it in the channel; 1-5 Demodulate the received signal at the receiving end to obtain receiving information including M data frames at the receiving end; 1-6 For each data frame at the receiving end, perform frame verification according to the frame verification code, and identify the frame that passes the frame verification as the correct received frame; for the data frame that does not pass the frame verification, then 1-6-1 Check each group according to the group check code to find out the error data group; 1-6-2 Replace the possibly erroneous data of the frame in the error data group with other data in the P base system, generate a new frame, and perform frame verification on the new frame; 1-6-3 If the new frame passes the frame check, it is determined that the new frame data is a correctly received frame; If the new frame does not pass the frame check, perform the next replacement and check; If all other combinations are permuted, or the number of permutations is higher than the established upper limit, the frame is considered to be an error received frame and retransmission is required. 2. A packet verification data transmission method as claimed in claim 1, characterized in that the data group is divided into equal-dimensional data groups. 3. A packet verification data transmission method as claimed in claim 1, characterized in that the group data can overlap. 4. A packet verification data transmission method as claimed in claim 1, characterized in that forward error correction coding technology is adopted, at the transmission end, the transmission end data frame is carried out error correction encoding, and then at the receiving end, the receiving end data is frames and new frames are decoded for error correction. 5. A packet verification data transmission method as claimed in claim 1, characterized in that forward error correction coding technology is used to carry out error correction coding on the information to be transmitted at the transmission end, and then the received information is corrected at the receiving end. Wrong decoding.

Images