JP4025709B2 - Error correction encoding apparatus, error correction decoding apparatus, error correction encoding program, error correction decoding program, error correction encoding program, or computer readable recording medium on which error correction decoding program is recorded - Google Patents

Description

  The present invention relates to a technique for performing end-to-end reliable communication in a packet communication method using IP (Internet Protocol) or the like.

  In the Internet, IP, which is a best-effort protocol, is adopted. However, when performing point-to-point communication, TCP (which improves the reliability of communication by performing retransmission control as an upper protocol of this IP) Generally, Transmission Control Protocol is employed. On the other hand, when performing multicast for transmitting the same data to a plurality of nodes, UDP (User Datagram Protocol) that does not ensure quality is generally adopted as a higher IP protocol.

  However, in recent years, with increasing communication needs using multicast, so-called Reliable Multicast (error-free multicast) is being developed in order to increase communication reliability even when multicast is performed. Specifically, the development of the reliable multicast is advanced in a scheme for performing retransmission control and a scheme for adding an error correction code (FEC: Forward Error Correction).

  As a method for performing this retransmission control, there are two methods, an Ack method that replies that it has been received and a Nack method that replies that it has not been received.

  A method for adding an error correction code is defined in RFC 2733 (An RTP Payload Format for Generic Forward Error Correction), Patent Document 1, and the like.

  This Patent Document 1 discloses a communication method for adding error correction cells using Reed-Solomon codes (hereinafter referred to as RS codes) to a plurality of ATM cells as an example of the FEC method in packet communication. . As shown in FIG. 8, this method divides cells (packets) into data cells (packets) and code cells (packets), and error correction is performed for a certain number of cell discards (packet discards). This is possible. This method can be applied to IP communication as it is if a cell is replaced with a packet. That is, since the IP packet has a variable length, the packet length may be different. In this case, padding processing such as adding 0 to other short packets to match the longest packet among the FEC target packets. And matching the lengths of all the packets can be applied to IP communication.

  By the way, it is desirable to change the redundancy and the encoding method even for the same stream due to the difference in network quality. In addition, even if the network quality is the same, it is desirable to change the redundancy and the encoding method in consideration of the delay time in the case of different streams having different stream speeds.

  Therefore, conventionally, a method for changing the redundancy has been proposed (see Patent Document 2).

In the method of Patent Document 2, a syndrome value is calculated by substituting the root of a generator polynomial for information data on the FEC encoding side, and the value is simultaneously transmitted as information “syndrome value on transmission side” together with the information data. It is a method. For example, in the case of an RS code, calculation for substituting the root of the generator polynomial G (X) into information data is performed. The change in redundancy can be achieved by adjusting the number of calculations (that is, the number of roots of the generator polynomial to be substituted).
JP-A-8-186570 (Error control method in ATM network) JP-A-11-136220 (Code Transmission Method, Transmitting Device, Receiving Device, and Communication System)

  However, in the method described in Patent Document 2, in order to change the redundancy and the encoding method, different encoding calculation processes must be performed.

  Therefore, in the method described in Patent Document 2, for example, RS code requires division by different generator polynomials, and hardware requires a circuit for each setting with different redundancy. Also in software, a separate program must be created for each.

  Therefore, the method described in Patent Document 2 has a problem that the circuit scale and the software scale must be increased in order to change the redundancy and the encoding method in consideration of the delay time.

  Therefore, in view of such circumstances, the present invention provides an error correction coding apparatus and error correction capable of controlling redundancy and delay time without increasing the circuit scale and software code amount when the stream speed and network quality are different. It is an object of the present invention to provide a computer-readable recording medium in which a decoding device, an error correction encoding program, an error correction decoding program, an error correction encoding program, or an error correction decoding program is recorded.

  According to the present invention, the above problem is solved by the following means.

In a first aspect of the present invention, in packet communication, in an error correction coding apparatus that performs error correction coding using a block code, an information packet to be subjected to an error correction coding operation is divided into information packet elements that are parts of the information packet. Means for generating a matrix having the information packet element as a row element by a predetermined number of rows, and performing error correction coding operation in the column direction of the generated matrix for the number of columns, Means for generating a check packet element, means for generating at least one information packet using the information packet element such that the information packet element and the check packet element are separate, and the information packet element wherein as and test packets element is separately and less by using a test packet element the generated Means for generating one test packet is also error correction coding apparatus characterized by comprising a means for separately transmitting said information packet and the test packet.

A second invention is an error correction decoding apparatus capable of performing error correction decoding of a data string subjected to error correction coding by a block code in packet communication. From the error correction coding apparatus according to the first invention, the transmitted packets, and means for separating the inspection packets generated by performing error correction coding operation on the information packet with the information packet to be subjected to error correction encoding operation, the separated information packets Means for dividing the information packet element which is a part of the information packet, dividing the separated inspection packet into inspection packet elements which are part of the inspection packet; and the divided information packet element and inspection packet element are predetermined. Means for generating a matrix having as many rows as the number of rows, and error correction decoding in the column direction of the generated matrix Means for repeatedly performing the operation for the number of columns and generating a row element composed of packet elements that are part of a packet that could not be received; and means for generating one or more packets using the row elements of the matrix An error correction decoding apparatus comprising:

  According to a third aspect of the present invention, in packet communication, in an error correction coding apparatus that performs error correction coding using a block code, the number of rows is constant and an information packet that is a target of error correction coding calculation is less than a predetermined number of rows Means for generating a matrix having only row elements, means for padding a line in which the information packet is not a row element in the generated matrix, and columning error correction coding operations in the column direction of the generated matrix. Repeating several minutes, generating a predetermined number of inspection packets, generating at least one information packet using the information packet, and generating at least one inspection packet using the generated inspection packet And means for transmitting the information packet and the inspection packet separately. An error correction coding apparatus that.

  A fourth invention is an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding by a block code in packet communication, wherein the received packet is subjected to error correction encoding calculation. Means for separating the information packet into a check packet generated by performing error correction coding operation on the information packet, and a matrix having the separated information packet and check packet as a row element by a predetermined number of rows Generating means, padding rows in which the information packet and check packet are not row elements in the generated matrix, and repeating error correction decoding operations in the column direction of the generated matrix for the number of columns Means for generating a row element composed of packets that could not be received, and the matrix Means for generating one or more packets using the row element is an error correction decoding apparatus comprising the.

According to a fifth aspect of the present invention, in packet communication, in an error correction coding apparatus that performs error correction coding using a block code, an information packet that is a target of an error correction coding operation is divided into information packet elements that are part of the information packet. Means for generating a matrix having a fixed number of rows and having the information packet element as a row element by a number less than a predetermined number of rows, and a row in which the information packet element is not a row element in the generated matrix. Means for repeating error correction coding operations in the column direction of the generated matrix for the number of columns, and generating a predetermined number of check packet elements that are part of the check packet, and the information packet element wherein as and test packets element is separate, at least one information using the information packet element and Means for generating a packet, such as the information packet element and the test packet element is separately and means for generating at least one test packet using the test packets element said generated the information packet and the And an error correction coding apparatus comprising: means for separately transmitting a check packet.

A sixth invention is an error correction decoding apparatus capable of performing error correction decoding of a data string subjected to error correction coding by a block code in packet communication. From the error correction coding apparatus according to the fifth invention, the transmitted packets, and means for separating the inspection packets generated by performing error correction coding operation on the information packet with the information packet to be subjected to error correction encoding operation, the separated information packets Means for dividing the information packet element which is a part of the information packet, dividing the separated inspection packet into inspection packet elements which are part of the inspection packet; and the divided information packet element and inspection packet element are predetermined. Means for generating a matrix having as many row elements as the number of rows, and the information packet in the generated matrix A packet that is a part of a packet that could not be received by means of padding a row in which the elementary and check packet elements are not row elements, and by repeating error correction decoding operation for the number of columns in the column direction of the generated matrix An error correction decoding apparatus comprising: means for generating row elements composed of elements; and means for generating one or more packets using the row elements of the matrix.

  According to a seventh aspect of the present invention, in packet communication, in an error correction coding apparatus that performs error correction coding using a block code, an information packet to be subjected to an error correction coding operation is divided into information packet elements that are part of the information packet. Means for generating a plurality of matrixes having the information packet element as a row element by a predetermined number of rows, and performing an error correction coding operation in the column direction of the plurality of generated matrices, respectively, for the number of columns, Means for generating a plurality of inspection packet elements that are part of the inspection packet; means for generating at least one information packet using the information packet element; and at least one inspection packet using the generated inspection packet element. Generating means, and information packets and inspection packets based on each of the plurality of matrices The door, an error correction coding apparatus characterized by comprising means for transmitting alternately for each matrix, a.

  An eighth invention provides an error correction decoding apparatus capable of performing error correction decoding of a data string subjected to error correction encoding with a block code in packet communication, wherein received packets for a plurality of matrices are converted into error correction codes. Means for separating the information packet to be subjected to the merging operation and the inspection packet generated by performing an error correction coding operation on the information packet, and the information packet that is a part of the information packet. A means for dividing the packet into elements, and dividing the packet into check packet elements which are parts of the separated check packet, and generating a plurality of matrixes having the divided information packet elements and check packet elements as row elements by a predetermined number of rows. And an error correction decoding operation for each of the plurality of generated matrixes in the column direction. Means for generating, in each matrix, row elements composed of packet elements that are part of a packet that has not been received, and means for generating one or more packets using the row elements of each matrix; An error correction decoding apparatus characterized by comprising:

  According to a ninth aspect of the present invention, in a packet communication, in an error correction encoding apparatus that performs error correction encoding using a block code, the number of rows is constant, and an information packet that is an object of error correction encoding calculation is less than a predetermined number Means for generating a plurality of matrixes having only row elements, means for padding each row in which the information packet is not a row element in the plurality of generated matrices, and a column direction of each of the plurality of generated matrices. Means for repeatedly performing an error correction coding operation for the number of columns and generating a predetermined number of check packets for each matrix; means for generating at least one information packet using the information packet; and the generated check packet Means for generating at least one inspection packet using a plurality of matrixes; Information packets and the test packets based on, respectively, an error correction coding apparatus characterized by comprising means for transmitting alternately for each matrix, a.

  A tenth aspect of the present invention is an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding with a block code in packet communication. Means for separating an information packet to be operated into an inspection packet generated by performing error correction encoding operation on the information packet; and a row element of the separated information packet and inspection packet by a predetermined number of rows Means for generating a plurality of matrices, a means for padding each row in which the information packet and the inspection packet are not row elements in the plurality of generated matrices, and a column direction of each of the generated matrices Repeat the error correction decoding operation for the number of columns, and Means for generating a row element configured, an error correction decoding apparatus characterized by comprising: means for generating one or more packets, the using the line elements of each matrix.

  An eleventh aspect of the invention relates to an error correction coding apparatus that performs error correction coding using a block code in packet communication, and divides an information packet to be subjected to an error correction coding operation into information packet elements that are part of the information packet. Means for generating a plurality of matrices in which the number of rows is constant and the information packet element is a row element less than a predetermined number of rows, and the information packet element is a row element in the generated plurality of matrices. Means for padding each non-existing row, and means for generating a predetermined number of check packet elements which are part of a check packet by repeatedly performing error correction coding operations for each number of columns in the column direction of the plurality of generated matrices. Means for generating at least one information packet using the information packet element; and Means for generating at least one inspection packet using the prepared inspection packet element, and means for alternately transmitting information packets and inspection packets based on each of the plurality of matrices for each matrix. It is an error correction encoding device.

  A twelfth aspect of the present invention is an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding with a block code in packet communication. Means for separating an information packet to be operated into an inspection packet generated by performing an error correction coding operation on the information packet, and an information packet element that is a part of the information packet Means for dividing the packet into check packet elements that are parts of the separated check packets, and means for generating a plurality of matrices having the divided information packet elements and check packet elements as row elements by a predetermined number of rows And the information packet element and the inspection packet element in the plurality of generated matrices. It is composed of a means for padding each non-primitive row, and a packet element that is a part of a packet that cannot be received by repeatedly performing error correction decoding operations for each column in the column direction of the generated matrix. An error correction decoding apparatus comprising: means for generating row elements; and means for generating one or more packets using the row elements of each matrix.

According to a thirteenth aspect of the present invention, in packet communication, an error correction coding apparatus that performs error correction coding using a block code is divided into information packet elements that are part of the information packet. A step of generating a matrix having the information packet element as a row element by a predetermined number of rows, repeating an error correction encoding operation for the number of columns in the column direction of the generated matrix, Generating an inspection packet element, generating at least one information packet using the information packet element such that the information packet element and the inspection packet element are separate, and the information packet element wherein as and test packets element is separately and the test packets element the generated Generating at least one test packet have a error-correcting encoding program for causing to execute a step of separately transmitting the information packet and the test packet.

A fourteenth aspect of the present invention is an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding by a block code in packet communication. The error correction encoding apparatus according to the thirteenth aspect of the present invention the transmitted packets, and separating the inspection packets generated by performing error correction coding operation on the information packet with the information packet to be subjected to error correction encoding operation, the separated information packets Dividing the information packet element that is part of the information packet, dividing the separated inspection packet into inspection packet elements that are part of the inspection packet; and dividing the divided information packet element and inspection packet element Generating a matrix having row elements equal to the number of rows, and an error in the column direction of the generated matrix. Repeat the correction decoding operation for the number of columns to generate a row element composed of packet elements that are part of a packet that could not be received, and generate one or more packets using the row element of the matrix And an error correction decoding program.

  In a fifteenth aspect of the present invention, in packet communication, an error correction encoding apparatus that performs error correction encoding with a block code has a fixed number of rows, and information packets subject to error correction encoding operations are less than a predetermined number of rows. A step of generating a matrix having only row elements, a step of padding rows in which the information packet is not a row element in the generated matrix, and a column of error correction coding operations in the column direction of the generated matrix Repeating several minutes, generating a predetermined number of inspection packets, generating at least one information packet using the information packet, and generating at least one inspection packet using the generated inspection packet And transmitting the information packet and the inspection packet separately; An error correction encoding program for causing the execution.

  According to a sixteenth aspect of the present invention, in packet communication, an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding using a block code is configured to subject the received packet to an error correction encoding operation. Separating the information packet into a check packet generated by performing an error correction coding operation on the information packet, and a matrix having a predetermined number of rows as the separated information packet and check packet A step of padding a row in which the information packet and the check packet are not row elements in the generated matrix, and repeating an error correction decoding operation in the column direction of the generated matrix for the number of columns. Generating a row element composed of packets that could not be received and before An error correction decoding program characterized by executing the steps of generating one or more packets using the line elements of the matrix, a.

According to a seventeenth aspect of the present invention, in packet communication, an error correction encoding apparatus that performs error correction encoding using a block code is divided into information packet elements that are part of the information packet. Generating a matrix in which the number of rows is constant and having the information packet element as a row element by a number less than a predetermined number of rows, and a row in which the information packet element is not a row element in the generated matrix A step of repeatedly performing an error correction coding operation in the column direction of the generated matrix for the number of columns to generate a predetermined number of check packet elements that are part of the check packet, and the information packet element wherein as and test packets element is separately and reduced by using the information packet element and Generating one packet of information, the so information packet element and the test packet element is separate, and generating at least one test packet using the test packets element said generated the information And a step of separately transmitting the packet and the check packet.

Eighteenth aspect of the present invention, in the packet communication, the error correction decoding apparatus which can error-correcting decoding of the data string error correction coding is performed by the block code, the error correction encoding apparatus according to the invention of a 17 the transmitted packets, and separating the inspection packets generated by performing error correction coding operation on the information packet with the information packet to be subjected to error correction encoding operation, the separated information packets Dividing the information packet element that is part of the information packet, dividing the separated inspection packet into inspection packet elements that are part of the inspection packet; and dividing the divided information packet element and inspection packet element Generating a matrix having as many row elements as the number of rows, and the information in the generated matrix. It is a part of a packet that could not be received by padding a row in which the packet element and the check packet element are not row elements, and by repeating error correction decoding operation for the number of columns in the column direction of the generated matrix. An error correction decoding program characterized by executing a step of generating a row element composed of packet elements and a step of generating one or a plurality of packets using the row element of the matrix.

  According to a nineteenth aspect of the present invention, in packet communication, an error correction encoding device that performs error correction encoding using a block code is divided into information packet elements that are parts of the information packet. A step of generating a plurality of matrixes having the information packet element as a row element by a predetermined number of rows, and performing error correction encoding operations in the column direction of the generated plurality of matrices, respectively, for the number of columns, Generating a plurality of inspection packet elements that are parts of the inspection packet; generating at least one information packet using the information packet element; and at least one inspection packet using the generated inspection packet element. Generating and an information packet based on each of the plurality of matrices. And a test packet, an error correction encoding program characterized by executing the steps of alternately transmitted for each matrix, a.

  According to a twentieth aspect of the present invention, in a packet communication, an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding using a block code, Separating the information packet to be subjected to the merging operation and the inspection packet generated by performing error correction coding operation on the information packet; and the information packet that is a part of the information packet. Dividing into elements, dividing into inspection packet elements that are parts of the separated inspection packets, and generating a plurality of matrices having the divided information packet elements and inspection packet elements as row elements by a predetermined number of rows Error correction decoding operations in the column direction of the generated plurality of matrices Generating a row element composed of packet elements that are part of a packet that could not be received in each matrix, and generating one or more packets using the row elements of each matrix And an error correction decoding program characterized in that

  In a twenty-first aspect, an error correction encoding apparatus that performs error correction encoding with a block code in packet communication has a fixed number of rows, and an information packet to be subjected to error correction encoding calculation is less than a predetermined number of rows. Generating a plurality of matrixes having only row elements, padding each row in which the information packet is not a row element in the plurality of generated matrices, and a column direction of each of the plurality of generated matrices. Repeating the error correction coding operation for the number of columns, generating a predetermined number of check packets for each matrix, generating at least one information packet using the information packet, and the generated check packet Generating at least one inspection packet using Information packets and the test packets based on each trix, an error correction encoding program characterized by executing the steps of alternately transmitted for each matrix, a.

  According to a twenty-second aspect of the present invention, an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding using a block code in packet communication. A step of separating an information packet to be operated into an inspection packet generated by performing error correction encoding operation on the information packet; and a row element of the separated information packet and inspection packet by a predetermined number of rows A step of generating a plurality of matrices, a step of padding each row in which the information packet and the inspection packet are not row elements in the generated plurality of matrices, and a column direction of each of the generated matrices The error correction decoding operation is repeated for the number of columns, and the packet that could not be received. Generating a row elements constituted by said an error correction decoding program characterized by executing the steps of generating one or more packets using the line elements of each matrix, a.

  According to a twenty-third aspect of the present invention, in packet communication, an error correction encoding apparatus that performs error correction encoding using a block code is divided into information packet elements that are part of the information packet. Generating a plurality of matrices having a fixed number of rows and having the information packet element as a row element by a number less than a predetermined number of rows, and the information packet element is not a row element in the generated plurality of matrices. A step of padding each non-existing row, and a step of repeatedly performing an error correction encoding operation for each column in the column direction of the plurality of generated matrices to generate a predetermined number of check packet elements which are parts of the check packet And a step of generating at least one information packet using the information packet element. Generating at least one check packet using the generated check packet element; and alternately transmitting information packets and check packets based on each of the plurality of matrices for each matrix. An error correction coding program characterized by being executed.

  According to a twenty-fourth aspect of the present invention, an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding using a block code in packet communication. A step of separating an information packet to be operated into an inspection packet generated by performing an error correction coding operation on the information packet, and an information packet element that is a part of the information packet And dividing into divided inspection packet elements that are parts of the separated inspection packet, and generating a plurality of matrices in which the divided information packet elements and inspection packet elements are row elements by a predetermined number of rows. And the information packet element and the inspection packet element in the generated plurality of matrices. Each padding each row that is not a row element, and repeating the error correction decoding operation for each column in the column direction of the generated matrix to form a packet element that is a part of the packet that could not be received. An error correction decoding program characterized by causing a row element to be generated and a step of generating one or a plurality of packets using the row element of each matrix to be executed.

  A twenty-fifth aspect of the invention is the error correction coding program according to the thirteenth aspect, fifteenth aspect, seventeenth aspect, nineteenth aspect, twenty-first aspect, or twenty-third aspect, or the fourteenth aspect. , An error correction decoding program according to the sixteenth, eighteenth, twentieth, twenty-second, or twenty-fourth invention.

  As described above, according to the present invention, when the stream speed and the network quality are different, the redundancy and the delay time can be controlled without increasing the circuit scale and the software code amount.

  The best mode for carrying out the present invention will be described below with reference to the accompanying drawings.

  The embodiment of the present invention is a method that can be applied to block codes of error correction codes. In this block code, a data string that is a target for performing an error correction coding operation (hereinafter referred to as FEC operation) or the like. A bit string and a data string and a bit string generated by performing an FEC operation on the data string and the bit string are separately generated. Here, in this specification, a data string or a bit string that is a target for the FEC calculation is referred to as information data, and a data string or a bit string generated by the FEC calculation is referred to as inspection data. A data string or bit string obtained by combining the information data and the inspection data is referred to as encoded data. The redundancy is a value obtained by dividing the check data by the encoded data. Examples of cyclic codes are shown below.

Information data: 100
Inspection data: 1011
Encoded data: 100 1011
Redundancy: about 0.57
Further, in this specification, information data packetized is called an information packet, and inspection data packetized is called an inspection packet. The information packet portion is called an information packet element, and the check packet portion is called a check packet element.

[First Embodiment]
FIG. 1 is a diagram for explaining an error correction coding apparatus according to the first embodiment of the present invention.

First, the error correction coding apparatus according to the first embodiment divides an information packet into a plurality of information packet elements (packet division).

  Next, the error correction coding apparatus according to the first embodiment generates a matrix having the plurality of information packets as row elements (matrix generation). Here, the row element refers to data constituting one row in the matrix.

  Then, the error correction coding apparatus according to the first embodiment repeats this packet division and matrix generation until a matrix having a preset number of rows is completed.

  For example, when the preset number of rows is 8, the error correction coding apparatus according to the first embodiment first converts the information packet A into four information packet elements (1, 2, 3, 4). Information packet element 1 is the first line element, information packet element 2 is the second line element, information packet element 3 is the third line element, and information packet element 4 is the fourth line. Element. Then, the error correction coding apparatus according to the first embodiment divides the information packet B into four information packet elements (5, 6, 7, 8), and the information packet element 5 is the fifth row element. The information packet element 6 is a row element of the sixth row, the information packet element 7 is a row element of the seventh row, and the information packet element 8 is a row element of the eighth row. Since the matrix has 8 matrix elements and 8 matrix elements are completed, the error correction coding apparatus according to the first embodiment ends packet division and matrix generation.

  Next, the error correction coding apparatus according to the first embodiment performs FEC calculation for the number of columns in the column direction for the column elements of the completed matrix (FEC calculation), and generates check packet elements (encoding). Data complete). Here, the column element is data constituting one column in the matrix.

  Next, the error correction coding apparatus according to the first embodiment generates one or a plurality of transmission packets using the information packet element and the check packet element (transmission packet generation). Note that the error correction coding apparatus according to the first embodiment generates a transmission packet so that the information packet element and the check packet element constitute separate transmission packets. For example, in the above example, the error correction coding apparatus according to the first embodiment includes the first to eighth row elements as information packet elements and the ninth and 10th check packet elements. Transmission packet generation is performed so that the row element in the row does not constitute the same packet.

  Then, the error correction coding apparatus according to the first embodiment transmits a transmission packet.

  According to the error correction coding apparatus according to the first embodiment described above, since packet division is performed, for example, even when the network quality and the stream speed are different, the change in redundancy and the change in delay time are controlled. it can.

  The packet division may be performed by equally dividing the packet, by dividing the packet unequally, or by dividing in accordance with a predetermined rule. An example in the case of unequal division of packets is shown in FIG.

  The error correction decoding apparatus according to the first embodiment that has received the transmission packet transmitted from the error correction decoding apparatus according to the first embodiment performs error correction decoding as follows.

  First, the error correction decoding apparatus according to the first embodiment uses a received packet to perform an error correction encoding operation on an information packet and an information packet that are subject to error correction encoding operation. Separated into packets.

  Next, the error correction decoding apparatus according to the first embodiment divides the separated information packet into information packet elements that are part of the information packet, and the separated check packet is the part of the check packet. Divide into inspection packet elements.

  Next, the error correction decoding apparatus according to the first embodiment generates a matrix having the divided information packet elements and check packet elements as row elements by a predetermined number of rows.

  Next, the error correction decoding apparatus according to the first embodiment repeats the error correction decoding operation for the number of columns in the column direction of the generated matrix, and uses the packet element that is the part of the packet that could not be received. Generates a composed row element.

  The error correction decoding apparatus according to the first embodiment generates one or a plurality of packets using the matrix row elements.

[Second Embodiment]
FIG. 3 is a diagram for explaining an error correction coding apparatus according to the second embodiment of the present invention.

  In the second embodiment, the number of columns of the matrix is predetermined.

  First, the error correction coding apparatus according to the second embodiment generates a matrix having a plurality of information packets as row elements (matrix generation). Here, the number of rows in which the information packet is a row element is less than the total number of rows in the matrix. For example, if there are 10 information packets, the number of information packets used for matrix generation is 9 or less.

  Next, the error correction coding apparatus according to the second embodiment pads the remaining row elements that are not filled with information packets (padding processing).

  Next, the error correction coding apparatus according to the second embodiment performs FEC operation for the number of columns in the column direction for the column elements of the matrix (FEC operation), and generates a check packet (completed code data).

  Next, the error correction coding apparatus according to the second embodiment generates one or a plurality of transmission packets using the information packet and the check packet (transmission packet generation). However, the padding line is not used for generating a transmission packet. The point that the transmission packet is generated so that the information packet and the inspection packet constitute separate transmission packets is the same as in the case of the first embodiment.

  Then, the error correction coding apparatus according to the second embodiment transmits a transmission packet.

  According to the error correction coding apparatus according to the second embodiment described above, since the padding process is performed, the network quality and the stream speed are different even if the information data and the inspection data are fixed. It is possible to control changes in redundancy and delay time.

  The error correction decoding apparatus according to the second embodiment that has received the transmission packet transmitted from the error correction encoding apparatus according to the second embodiment performs error correction decoding as follows. .

  First, the error correction decoding apparatus according to the second embodiment uses a received packet to perform an error correction encoding operation on an information packet and an information packet that are subject to error correction encoding operation. Separated into packets.

  Next, the error correction decoding apparatus according to the second embodiment generates a matrix having separated information packets and check packets as row elements by a predetermined number of rows.

  Next, the error correction decoding apparatus according to the second embodiment pads the rows in which the information packet and the check packet are not row elements in the generated matrix.

  Next, the error correction decoding apparatus according to the second embodiment repeats the error correction decoding operation for the number of columns in the column direction of the generated matrix, and obtains row elements composed of packets that could not be received. Generate.

  Then, the error correction decoding apparatus according to the second embodiment generates one or a plurality of packets using the matrix row elements.

[Third Embodiment]
Next, an error correction coding apparatus according to the third embodiment will be described. The third embodiment is a combination of the first embodiment and the second embodiment.

  First, the error correction coding apparatus according to the third embodiment divides an information packet into a plurality of information packet elements (packet division).

  Next, the error correction coding apparatus according to the third embodiment generates a matrix having a plurality of information packet elements as row elements (matrix generation). Here, the number of rows in which the information packet element is a row element is less than the total number of rows in the matrix, as in the second embodiment.

  Next, the error correction coding apparatus according to the third embodiment pads the remaining rows that are not filled with information packet elements (padding processing).

  Next, the error correction coding apparatus according to the third embodiment performs FEC operation on the column elements of the matrix in the column direction by the number of columns (FEC operation), and generates check packet elements (code data completion).

  Next, the error correction coding apparatus according to the third embodiment generates one or a plurality of transmission packets using the information packet element and the check packet element (transmission packet generation). However, the padding line is not used for generating a transmission packet. The point that the transmission packet is generated so that the information packet element and the inspection packet element form separate transmission packets is the same as in the case of the first embodiment and the second embodiment.

  Then, the error correction coding apparatus according to the third embodiment transmits a transmission packet.

  According to the error correction coding apparatus according to the third embodiment described above, since packet division and padding are performed, for example, a change in redundancy and a change in delay time due to different network quality and stream speed Can be controlled.

  The error correction decoding apparatus according to the third embodiment that has received the transmission packet transmitted from the error correction encoding apparatus according to the third embodiment performs error correction decoding as follows. .

  First, the error correction decoding apparatus according to the third embodiment performs a check on a received packet by performing an error correction coding operation on an information packet to be subjected to an error correction coding operation and the information packet. Separated into packets.

  Next, the error correction decoding apparatus according to the third embodiment divides the separated information packet into information packet elements that are part of the information packet, and the separated check packet is the check that is the part of the check packet. Split into packet elements.

  Next, the error correction decoding apparatus according to the third embodiment generates a matrix having the divided information packet elements and check packet elements as row elements by a predetermined number of rows.

  Next, the error correction decoding apparatus according to the third embodiment pads the rows in which the information packet element and the check packet element are not row elements in the generated matrix.

  Next, the error correction decoding apparatus according to the third embodiment repeats the error correction decoding operation for the number of columns in the column direction of the generated matrix, and uses the packet element that is the part of the packet that could not be received. Generates a composed row element.

  Then, the error correction decoding apparatus according to the third embodiment generates one or a plurality of packets using the matrix row elements.

[Fourth Embodiment]
FIG. 4 is a diagram for explaining an error correction coding apparatus according to the fourth embodiment of the present invention.

  In the fourth embodiment, interleaving processing is added to the first embodiment, the second embodiment, or the third embodiment.

  The error correction coding apparatus according to the fourth embodiment alternately transmits transmission packets based on a plurality of matrices for each matrix.

  According to the fourth embodiment, it is possible to improve the burst packet loss tolerance. For example, if packets are transmitted alternately from two matrices, the burst packet loss tolerance is doubled.

  The error correction decoding apparatus according to the fourth embodiment, which has received transmission packets for a plurality of matrices from the error correction encoding apparatus according to the fourth embodiment based on the first embodiment, is as follows. In this manner, error correction decoding is performed.

  First, an error correction decoding apparatus according to the fourth embodiment performs an error correction coding operation on an information packet to be subjected to an error correction coding operation and the information packet for received packets for a plurality of matrices. To the inspection packet generated by the above.

  Next, the error correction decoding apparatus according to the fourth embodiment divides the separated information packet into information packet elements that are part of the information packet, and converts the information packet element into a check packet element that is a part of the separated check packet. To divide.

  Next, the error correction decoding apparatus according to the fourth embodiment generates a plurality of matrices having the divided information packet elements and check packet elements as row elements by a predetermined number of rows.

  Next, the error correction decoding apparatus according to the fourth embodiment repeats error correction decoding operations for the number of columns in the column direction of a plurality of generated matrices, and is a part of a packet that cannot be received. Row elements composed of packet elements are generated in each matrix.

  Then, the error correction decoding apparatus according to the fourth embodiment generates one or a plurality of packets using the row elements of each matrix.

  The error correction decoding apparatus according to the fourth embodiment, which has received transmission packets for a plurality of matrices from the error correction encoding apparatus according to the fourth embodiment based on the second embodiment, In this manner, error correction decoding is performed.

  First, an error correction decoding apparatus according to the fourth embodiment performs an error correction coding operation on a plurality of matrix received packets on an information packet to be subjected to an error correction coding operation and the information packet. Separated into generated inspection packets.

  Next, the error correction decoding apparatus according to the fourth embodiment generates a plurality of matrices having separated information packets and check packets as row elements by a predetermined number of rows.

  Next, the error correction decoding apparatus according to the fourth embodiment pads each row in which the information packet and the check packet are not row elements in the generated plurality of matrices.

  Next, the error correction decoding apparatus according to the fourth embodiment is configured with packets that could not be received by repeatedly performing error correction decoding operations for each of the generated matrix in the column direction. Generate row elements.

  Then, the error correction decoding apparatus according to the fourth embodiment generates one or a plurality of packets using the row elements of each matrix.

  The error correction decoding apparatus according to the fourth embodiment, which has received transmission packets for a plurality of matrices from the error correction encoding apparatus according to the fourth embodiment based on the third embodiment, In this manner, error correction decoding is performed.

  First, an error correction decoding apparatus according to the fourth embodiment performs an error correction coding operation on a plurality of matrix received packets on an information packet to be subjected to an error correction coding operation and the information packet. Separated into generated inspection packets.

  Next, the error correction decoding apparatus according to the fourth embodiment divides the separated information packet into information packet elements that are part of the information packet, and converts the information packet element into a check packet element that is a part of the separated check packet. To divide.

  Next, the error correction decoding apparatus according to the fourth embodiment generates a plurality of matrices having the divided information packet elements and check packet elements as row elements by a predetermined number of rows.

  Next, the error correction decoding apparatus according to the fourth embodiment pads each row in which the information packet element and the check packet element are not row elements in the generated plurality of matrices.

Next, the error correction decoding apparatus according to the fourth embodiment repeatedly performs error correction decoding operations in the column direction of the generated matrix for the number of columns, and packet elements that are portions of packets that could not be received. A row element constituted by is generated.

  Then, the error correction decoding apparatus according to the fourth embodiment generates one or a plurality of packets using the row elements of each matrix.

  Next, an error correction coding apparatus according to Embodiment 1 of the present invention will be described. The error correction coding apparatus according to Example 1 is an example of the error correction coding apparatus according to the third embodiment.

  FIG. 5 is a diagram illustrating a matrix used in the error correction coding apparatus according to the first embodiment.

  In the error correction coding apparatus according to the first embodiment, the coding method is shortened Reed-Solomon (14, 10) on the Galois extension field GF (2 ^ 8), and the generator polynomial is G (X). . That is, the information data is 10 bytes, the inspection data is 4 bytes, the encoded data is 14 bytes, the number of rows that can be subjected to FEC computation is 10 rows, the number of rows that can contain computation results is 4, and the total number of rows in the matrix The number is 14 lines.

  In addition, the information packet length is 100 bytes, and the information packet is divided into two equal parts. Therefore, the matrix column is 50 bytes.

  Also, the number of lines to be filled with information packet elements is eight. That is, the number of padding lines is two.

  FIG. 6 is a diagram for explaining the error correction coding apparatus according to the first embodiment.

  First, the error correction coding apparatus according to the first embodiment equally divides the data portion of four information packets into two (step 1).

  Next, the error correction coding apparatus according to the first embodiment stores information packet elements in a matrix in order, with each information packet element as one row (step 2). That is, information packet element 1-1, information packet element 1-2, information packet element 2-1, ..., information packet element 4-2 are stored in the first to eighth rows of the matrix.

  Step 1 and step 2 may be performed simultaneously. That is, the information packet element may be stored in the matrix every time one information packet is divided.

  Next, the error correction coding apparatus according to the first embodiment performs a padding process on the 9th and 10th rows (step 3). Note that step 3 can be performed out of order with the processing of step 1 and step 2.

  Next, the error correction coding apparatus according to the first embodiment performs an FEC operation on the column elements (step 4). Specifically, the first 1 byte of the line is divided by the generator polynomial G (X).

  Next, the error correction coding apparatus according to the first embodiment stores the check packet element, which is the result of the FEC calculation, in the first 1 byte of the 11th to 14th rows of the matrix.

  The error correction coding apparatus according to the first embodiment repeats this FEC calculation 50 times (for 50 bytes) in the column direction.

  Then, the error correction coding apparatus according to the first embodiment constructs two rows of the completed matrix as one transmission packet. That is, two information packet elements are set as one transmission packet.

  Here, the characteristics of the encoded data generated by the error correction encoding apparatus according to the first embodiment will be described.

  First, since the number of information packets is 4 and the number of inspection packets is 2, the redundancy is about 0.33. Further, it is possible to recover up to one packet loss among a total of six packets (four information packets and two inspection packets).

  Here, the case where the parameters are changed in the error correction coding apparatus according to the first embodiment will be described.

  First, consider the case where the number of padding lines is changed from 2 to 0. As a result, the number of packets used for the calculation is five. That is, the redundancy can be changed from about 0.33 to about 0.29. In other words, in the above description, recovery was possible from a total of 6 packets to 1 packet loss. However, after the parameter change, the recovery changes from a total of 7 packets to 1 packet loss. Also, the delay time is 1.25 times since it takes 4 packets to wait for 5 packets.

  Furthermore, in the error correction coding apparatus according to the first embodiment, a case where parameters are changed so as to be different from the above will be described.

  Here, a case where a packet that has been divided into two is not divided will be considered. As a result, the number of packets used for the calculation is eight. The redundancy is about 0.33, but in the above example, it was possible to recover from a total of 6 packets to 1 packet loss, but after the parameter change, 2 packets lost from a total of 12 packets. Recovery is possible. That is, it is possible to handle up to two burst packet losses. In addition, the delay time is doubled since the time for waiting for 8 packets is required from the time for waiting for 4 packets.

  As described above, according to the error correction coding apparatus according to the first embodiment, it is possible to change the redundancy and the delay time with one algorithm such as the shortened Reed-Solomon (14, 10). It becomes.

  In the first embodiment, the error correction coding method is the shortened Reed-Solomon (14, 10) on the Galois extension field GF (2 ^ 8), but other methods are used in the present invention. However, the same effect can be obtained.

  In the first embodiment, the information packet length is set to 100 bytes. However, it is obvious that the present invention does not limit the information packet length to 100 bytes.

  In the first embodiment, the information packet is divided into two equal parts, but it is clear that the present invention does not limit the information packet division to two equal parts.

  In the present invention, when the packet length cannot be divided equally, padding processing can be performed. For example, when the information packet length is 100 bytes and the three-division processing is performed, the first two rows may be 34 bytes, and the third row may be padded with 2 bytes to the data 32 bytes and the matrix column byte may be aligned with 34 bytes. The padding process can be realized by a process of filling a row with 0, but the effect of the present invention can be obtained by padding by another method.

  Next, an error correction coding apparatus according to the second embodiment will be described. This second example is an example of the fourth embodiment.

  In the error correction coding apparatus according to the second embodiment, the coding method is shortened Reed-Solomon (14, 10) on the Galois extension field GF (2 ^ 8), and the generator polynomial is G (X). . That is, the information data is 10 bytes, the inspection data is 4 bytes, the encoded data is 14 bytes, the number of rows in which the information packet can be put is 10, the number of rows in which the check packet can be put is 4, the total number of rows in the matrix is 14 rows It becomes. The information packet length is 50 bytes. That is, the matrix column is 50 bytes. In addition, the number of lines for filling the information packet is 8 lines. That is, the number of padding lines is two. The matrix used by the error correction coding apparatus according to the second embodiment is the same as that of the first embodiment and is shown in FIG. Note that the number of matrices for performing packet interleaving performed by the error correction coding apparatus according to the second embodiment is two.

  FIG. 7 is a diagram for explaining an error correction coding apparatus according to the second embodiment.

  First, the error correction coding apparatus according to the second embodiment sequentially stores information packets in a matrix 1 in one row (step 1). That is, information packet 1, information packet 2,..., Information packet 8 are stored in the first to eighth rows of matrix 1.

  Next, the error correction coding apparatus according to the second embodiment performs padding processing on the 9th and 10th rows (step 2). This step 2 can be performed out of order with the processing of step 1.

  Next, the error correction coding apparatus according to the second embodiment performs an FEC operation on the column elements of the matrix (step 3). Specifically, the first 1 byte of the line is divided by the generator polynomial G (X).

  Next, the error correction coding apparatus according to the second embodiment stores the check packet, which is the result of the FEC calculation, in the first 1 byte of the 11th to 14th rows of the matrix.

  The error correction coding apparatus according to the second embodiment repeats this FEC operation 50 times (for 50 bytes) in the row direction.

  Next, the error correction coding apparatus according to the second embodiment sequentially stores the information packets in the matrix 2 as one row (step 4). That is, the information packet 11, the information packet 12,..., The information packet 18 are stored in the first to eighth rows of the matrix 2.

  Next, the error correction coding apparatus according to the second embodiment performs padding processing on the 9th and 10th rows (step 5). This step 5 can be performed out of order with the processing of step 4.

  Next, the error correction coding apparatus according to the second embodiment performs an FEC operation on the column elements of the matrix (step 6). Specifically, the first 1 byte of the line is divided by the generator polynomial G (X).

Next, the error correction coding apparatus according to the second embodiment stores the check packet, which is the result of the FEC calculation, in the first 1 byte of the 11th to 14th rows of the matrix.

  The error correction coding apparatus according to the second embodiment repeats this FEC operation 50 times (for 50 bytes) in the row direction.

  Then, the error correction coding apparatus according to the second embodiment alternately transmits packets from the two completed matrices 1 and 2 (step 7).

  Here, characteristics of encoded data generated by the error correction encoding apparatus according to the second embodiment will be described.

  First, the number of information packets is 8, the number of inspection packets is 4, and the redundancy is about 0.33. In addition, the total of 8 information packets and 4 check packets is 12, and packet interleaving is performed between two matrices. As a result, up to 4 packet losses can be recovered from 24 packets. is there. This can cope with burst packet loss of 4 packets.

  Here, the case where the parameters are changed in the error correction coding apparatus according to the second embodiment will be described.

  First, consider a case where the number of padding lines is changed from 2 to 4. As a result, the number of packets used for the calculation is six. That is, the redundancy can be changed from about 0.33 to 0.4. In the above description, recovery was possible from a total of 24 packets to 4 packet losses, but after a parameter change, recovery from a total of 20 packets to 4 packet losses is possible. In addition, the delay time is 0.8 times since it is a time for waiting for 20 packets from a time for waiting for 24 packets.

  Although the number of matrices for packet interleaving is two, if the number of matrices is three, it is possible to recover from 24 packets to 4 packet losses. To 6 packet losses are changed so as to be recoverable. However, the delay time is 1.5 times because the time for waiting for 36 packets is required from the time for waiting for 24 packets.

  In the second embodiment, the error correction coding method is the shortened Reed-Solomon (14, 10) on the Galois extension field GF (2 ^ 8). However, the effect of the present invention is to use another method. However, it can be obtained similarly.

  In the second embodiment, the information packet length is 50 bytes, but it is clear that the present invention does not limit the information packet length to 50 bytes.

  The padding process can be realized by a process of filling a row with 0, but the effect of the present invention can be obtained by padding by another method.

  As described above, the embodiment of the present invention basically uses the algorithm of the existing coding method as it is (that is, when considering using the RS coding method, for example, normal RS coding). As with the method, division processing is performed with a generator polynomial on the FEC encoding side.) When generating a matrix for performing FEC calculation, packet division and row padding are performed. Therefore, according to the embodiment of the present invention, it is possible to change the redundancy in various ways and to suppress the change in the delay time.

  Note that the above-described error correction encoding apparatus and error correction decoding apparatus according to the embodiment of the present invention can be realized not only in hardware but also in software. When implemented in software, an error correction encoding program and an error correction decoding program are installed in the computer. These error correction encoding program and error correction decoding program are programs that cause a CPU, a memory, or the like to execute processing in the error correction encoding device or processing in the error correction decoding device. These error correction encoding programs and error correction decoding programs can be obtained by purchasing CD-ROMs or DVD-ROMs on which these programs are recorded, or downloading them from a server on the Internet. Can do.

It is a figure explaining the error correction coding apparatus which concerns on the 1st Embodiment of this invention. It is a figure explaining the example in the case of dividing | segmenting a packet equally in the 1st Embodiment of this invention. It is a figure explaining the error correction coding apparatus which concerns on the 2nd Embodiment of this invention. It is a figure explaining the error correction coding apparatus which concerns on the 4th Embodiment of this invention. It is a figure which shows the matrix used with the error correction coding apparatus which concerns on Example 1,2. 1 is a diagram illustrating an error correction coding apparatus according to Embodiment 1. FIG. It is a figure explaining the error correction coding apparatus which concerns on Example 2. FIG. It is a figure which shows the FEC calculation and error correction decoding calculation of a prior art.

Claims (5)

In an error correction coding apparatus that performs error correction coding using a block code in packet communication,
Means for dividing an information packet to be subjected to an error correction coding operation into information packet elements that are parts of the information packet;
A number of rows is constant, means for generating a matrix to said information packet element predetermined by less than the number of rows row element,
Means for padding each row for which the information packet is not a row element in the generated matrix;
Means for repeatedly performing an error correction coding operation for the number of columns in the column direction of the generated matrix, and generating a predetermined number of check packets;
Means for generating at least one information packet using the information packet such that the information packet element and the inspection packet element are separate;
Means for generating at least one inspection packet using the generated inspection packet such that the information packet element and the inspection packet element are separate;
Means for separately transmitting the information packet and the inspection packet;
An error correction coding apparatus comprising: In packet communication, in an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding by a block code,
The packet transmitted from the error correction coding apparatus according to claim 1 is converted into an information packet that is an object of an error correction coding operation and a check packet that is generated by performing an error correction coding operation on the information packet. Means for separating;
Means for generating a matrix having the separated information packet and check packet as a row element by a predetermined number of rows, and padding the row in which the information packet and check packet are not row elements in the generated matrix Means to
Means for repeatedly performing error correction decoding operation in the column direction of the generated matrix for the number of columns, and generating row elements constituted by packets that could not be received;
Means for generating one or more packets using the row elements of the matrix;
An error correction decoding apparatus comprising: In packet communication, an error correction encoding device that performs error correction encoding using a block code,
Dividing an information packet to be subjected to an error correction coding operation into information packet elements that are parts of the information packet;
A number of rows is constant, and generating a matrix of said predetermined by less than the number of rows row elements of said information packet element,
A step of padding the row in which the information packet is not the row element in the matrix which is the product for each row,
Repeating the error correction encoding operation for the number of columns in the column direction of the generated matrix to generate a predetermined number of check packets;
Generating at least one information packet using the information packet such that the information packet element and the inspection packet element are separate;
Generating at least one inspection packet using the generated inspection packet such that the information packet element and the inspection packet element are separate;
Transmitting the information packet and the inspection packet separately;
An error correction coding program characterized in that In packet communication, an error correction decoding apparatus capable of performing error correction decoding of a data string that has been subjected to error correction encoding by a block code,
A packet transmitted from the error correction coding apparatus according to claim 3 is converted into an information packet to be subjected to error correction coding calculation and a check packet generated by performing error correction coding calculation on the information packet. Separating, and
Generating a matrix having a predetermined number of rows as the separated information packets and inspection packets;
A step of padding the row in which the information packets and the test packets are not the row element in the matrix which is the product for each row,
Repeating the error correction decoding operation for the number of columns in the column direction of the generated matrix to generate row elements composed of packets that could not be received;
Generating one or more packets using row elements of the matrix;
An error correction decoding program characterized in that Error correction encoding program according to claim 3 or claim 4 also error correction decoding program, a computer-readable recording medium according to.