WO2011062444A2

WO2011062444A2 - Method and apparatus for transmitting and receiving broadcasting signal

Info

Publication number: WO2011062444A2
Application number: PCT/KR2010/008203
Authority: WO
Inventors: 백서영; 고우석; 문상철
Original assignee: 엘지전자 주식회사
Priority date: 2009-11-20
Filing date: 2010-11-19
Publication date: 2011-05-26
Also published as: WO2011062444A3

Abstract

The present invention relates to a method and an apparatus for transmitting and receiving a broadcasting signal, and more particularly to a method and an apparatus for transmitting and receiving a broadcasting signal by performing Low-Density Parity-Check (LDPC) coding to enable a receiving part to obtain original information without loss of information. The method for transmitting the broadcasting signal according to the present invention comprises the steps of: generating information; inserting zero padding information into the generated information; performing Bose-Chaudhuri-Hocquenghem (BCH) encoding for the information into which the zero padding information has been inserted; performing LDPC encoding for the BCH-encoded information, the LDPC-encoded information including an information area and a parity area; removing a zero value included in the information area of the LDPC-encoded information; and puncturing the parity area of the LDPC-encoded information.

Description

Broadcast signal transmission and reception method and apparatus

The present invention relates to a method and apparatus for transmitting and receiving broadcast signals, and more particularly, to a method for transmitting and receiving broadcast signals by performing low-density parity-check (LDPC) coding so that a receiver can decode original information without loss of information. Relates to a device.

As the point of stopping the transmission of the analog broadcast signal is approaching, various technologies for transmitting and receiving digital broadcast signals have been developed. The digital broadcast signal may transmit a larger amount of video / audio data than the analog broadcast signal, and may include various additional data in addition to the video / audio data.

The digital broadcasting system can provide HD (High Definition) level video, multi-channel sound, and various additional services. However, data transmission efficiency for high-capacity data transmission, robustness of the transmission / reception network, and flexibility of the network considering mobile reception equipment still need to be improved.

The present invention relates to a method and apparatus for transmitting and receiving information using LDPC coding, and does not cause loss of information according to performing LDPC coding, and is capable of maintaining a constant robustness regardless of the size of information to be transmitted. And an apparatus.

Technical solution of the present invention,

An information generation unit for generating information, a zero padding unit for inserting zero padding information into the generated information, and a BCH for performing Bose-Chaudhuri-Hocquenghem (BCH) encoding on the information into which the zero padding information is inserted An LDPC encoding unit for performing low-density parity-check (LDPC) encoding on the BCH-encoded information, and the LDPC-encoded information includes an information region and a parity region, and an information region of the LDPC encoded information. It may include a shortening and puncturing unit to remove a '0' (zero) included in, and to puncture a parity region of the LDPC encoded information.

Therefore, according to the present invention, since the influence of puncturing is distributed and distributed without focusing a specific portion of the LDPC information area, transmission can be performed without loss of information, and the size change of the information to be transmitted by adjusting the size of the punctured LDPC parity is related. It is possible to maintain a certain robustness (robustness) for the broadcast signal transmission without.

1 is a block diagram showing a system for transmitting and receiving information according to the present invention.

2 is an embodiment of an LDPC block according to the present invention.

3 is another embodiment of an LDPC block according to the present invention.

4 is an embodiment of a puncturing process according to the present invention.

5 is an embodiment of a tanner graph according to an embodiment of the puncturing process of FIG. 4 according to the present invention.

6 is an embodiment of an update process of an LDPC bit node according to the present invention.

7 is a block diagram illustrating a relationship between a puncturing group and an information group according to the present invention.

8 is an embodiment of a Tanner graph associated with the block diagram of FIG. 7 in accordance with the present invention.

9 is another embodiment of a Tanner graph associated with the block diagram of FIG. 7 in accordance with the present invention.

10 is a table showing values for determining a puncturing pattern according to the present invention.

11 is a block diagram illustrating an embodiment of a signaling information transmitter according to the present invention.

12 is a block diagram illustrating signaling information according to the present invention.

13 is a graph showing performance of Signal to Noise Ratio (SNR) versus Bit Error Rate (BER) according to the present invention.

14 is a block diagram illustrating a relationship between variables for obtaining a compensation value according to the present invention.

15 is a graph illustrating a change in code rate according to a change in a compensation value according to the present invention.

16 is a table illustrating SNR size according to a modulation order of signaling information and data according to the present invention.

FIG. 17 is a table illustrating a result of comparing an SNR value and an SNR value according to a change in the size of signaling information according to the present invention with an SNR value of the data of FIG. 16.

18 is a flowchart illustrating an embodiment of a broadcast signal transmission method according to the present invention.

19 is a flowchart illustrating an embodiment of a broadcast signal receiving method according to the present invention.

With the development of digital broadcasting technology, digital broadcasting technology can be used to transmit video and audio data with a larger capacity than existing analog broadcasting signals, and can also include various additional data. As the spread of digital broadcasting spreads, demands for better services such as video and audio are increasing, and the size of data desired by the user and the number of broadcasting channels are gradually increasing. In the case of transmitting data as described above, the transmitting side processes and transmits the data so that the receiving unit can restore the original data in spite of an error caused by the loss of the signal. It will go through a series of processes to recover. Therefore, the transmitting side should transmit the information necessary to recover the data as completely as possible at the receiving side. Such information may be referred to as signaling information.

The signaling information includes information for data recovery, that is, coding information, data scheduling information, and the like, and may include not only data currently transmitted but also information necessary for recovering data to be transmitted in the future.

Although signaling information has a smaller amount of data than information such as video and audio data related to a service, a technique such as error correction coding (ECC) may be used to transmit the signaling information without loss because it includes information for data recovery. have.

Low Density Parity Check (LDPC) coding is a linear error correcting code (linear error correcting code) as one of error correction coding methods for transmitting information with a minimum probability of information loss.

LDPC coding may include an H-matrix determined according to an LDPC block size for performing LDPC coding. The LDPC block may be represented by parameters represented by N and K, where N represents a block length (# bits) and K represents a number of encoded information bits included in one LDPC block. In the present invention, an area including encoded information bits may be referred to as an LDPC information part or an information area, and K may be used as a parameter indicating the size of the LDPC information area.

The LDPC encoder can add a fixed number of parity bits to each block containing K information bits and form an N bit encoded block having a code rate (R = K / N). . In the present invention, an area including parity bits added during LDPC encoding may be referred to as an LDPC parity part or a parity area. The amount of data that one LDPC block can transmit may be determined according to the size and code rate of the LDPC parity region.

However, as described above, the signaling information may include a relatively smaller amount of information than the data related to the service. Therefore, when performing LDPC coding on signaling information, LDPC coding is inevitably performed on a smaller amount of information than the amount of data that can be transmitted through a preset LDPC block.

In other words, since a smaller amount of signaling information must be transmitted through a block having the same size as a general LDPC block, some kind of data processing for insufficient information amount is required, which is called shortening and puncturing.

As an example, shortening may be performed on the LDPC information region, and puncturing may be performed on the LDPC parity region. Although puncturing is performed on the LDPC parity region, the puncturing may also affect the information region associated with the parity region. In this case, if the influence of puncturing on the specific area is concentrated, the specific part of the information area including the information for data recovery may be intensively damaged and loss of signaling information may occur.

Accordingly, there is a need to develop a method of puncturing, that is, a puncturing pattern, so that the effect of puncturing on the information area can be distributed and distributed without being concentrated on a specific part of the LDPC information area.

In the present invention, a puncturing pattern for performing puncturing in the LDPC parity region and performing puncturing so that the influence of the puncturing on the LDPC information region is distributed. In addition, to determine the size of the LDPC parity region to be punctured, a method of determining the code rate that can perform the most efficient LDPC coding.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

According to an embodiment of the present invention, the OFDM transmission / reception system may be a part included in a bit interleaver and coded modulation (BICM) module.

According to an embodiment of the present invention, a system for transmitting and receiving information includes a transmitter 101100 for transmitting information, a channel 101200 for transmitting information, and a receiver 101300 for receiving information. The information transmitting unit 101100 includes an information generating unit 101110, a zero padding insertion unit 101120, a Bose-Chaudhuri-Hocquenghem encoding unit 101130, and an LDPC encoding unit. unit 101140 and a shortening and puncturing unit 101150.

The information generator 101110 generates information to be transmitted. According to an embodiment of the present invention, the generated information may be signaling information. In this case, the number of bits of signaling information is variable. In addition, the signaling information may include information necessary for recovering data related to a service currently being transmitted, and may include information necessary for recovering data to be subsequently transmitted.

The information generated by the information generator 101110 is input to the zero padding inserter 101120. The zero padding inserter 101120 inserts a zero padding value to enable BCH encoding on the information. In this case, the position of the zero padding to be inserted is determined according to a shortening pattern to be performed in the shortening step.

The BCH encoder 101130 performs BCH encoding on information in which a zero padding value is inserted. The BCH encoder 101130 adds BCH parity check bits to the information and transmits the BCH parity check bits to the LDPC encoder 101140.

The LDPC encoder 101140 may generate an LDPC block by adding LDPC parities to the received BCH encoding information, and perform LDPC encoding on the basis of the generated LDPC block. The encoded LDPC block may include an LDPC information part including information and an LDPC parity part including parity used for LDPC encoding.

The size of the LDPC parity region may be determined according to the size of the information region, and a constant value may be used in the equation for determining the relationship between the size of the LDPC parity region and the size of the information region. The constant value used at this time may be referred to as a compensation value. The compensation value can be used to change the code rate according to the size of the information to be transmitted. Detailed description thereof will be described later.

The encoded LDPC block is input to the shortening and puncturing unit 101150.

The shortening and puncturing unit 101150 may remove the value '0' inserted into the information area of the LDPC block according to the shortening pattern, and this process may be referred to as shortening. In addition, the shortening and puncturing unit 101150 may perform puncturing on the parity region of the LDPC block according to the puncturing pattern. have.

Although not shown in the figure, the information which has undergone the shortening and puncturing steps may be input to the OFDM modulator and modulated into an OFDM signal to be output. The output signal may be transmitted to the receiver through the channel 101200.

The receiver 101300 includes an OFDM demodulation 101310, an unshortening and de-puncturing unit 101320, an LDPC decoding unit 101330, and a BCH decoding unit ( A BCH decoding unit 101340 and an L1 signaling extraction unit 101350 may be included.

The OFDM demodulator 101310 may perform OFDM demodulation on the signal received through the channel 101200. The unshortening and depuncturing unit 101320 performs unshortening by inserting a maximum value according to a log-likelihood ratio (LLR) operation into an information region according to a shortening pattern defined by the transmitting side with respect to OFDM demodulated information. According to the puncturing pattern previously defined by the side, de-punching may be performed by assigning a value ('0'; zero) corresponding to an unknown to a parity region where puncturing is performed.

The LDPC decoding unit 101330 may perform LDPC decoding on unshortened and depunctured information. The BCH decoding unit 101340 may perform BCH decoding on the LDPC decoded information, and the information extraction unit 101350 may extract information to be received in the remaining portions except for the zero-padded position with respect to the BCH decoded information. Can be.

2 is an embodiment of an LDPC block according to the present invention.

As shown in FIG. 2, the LDPC block is formed in the form of an H-matrix, and includes an LDPC information region (or information region) 102100 and an LDPC parity region (or parity region) 102200 including an LDPC parity. can do. The LDPC information area 102100 may include a plurality of information groups, and the LDPC parity area 102200 may include a plurality of parity groups (or puncturing groups).

N shown in FIG. 2 is a parameter indicating an LDPC block size, and K is a parameter indicating the size of an information region of an LDPC block or the number of encoded information bits included in one LDPC block. M is a parameter indicating the size of one information group included in the LDPC information area 102100. In the present invention, M may be 360. In addition, Q is a parameter indicating the size of one parity group (or puncturing group) included in the LDPC parity region 102200.

According to an embodiment of the present invention, a column of an LDPC H-matrix is called a variable node and a row is called a check node.

The information groups included in the LDPC information region 102100 or the parity groups included in the LDPC parity region 102300 may include a matrix including a check node and a versatile node, and may include one information group or one parity group. The check node and the versatile node included in may be matched with each other. As such, when the check node and the versatile node match each other, '1' may be displayed as shown in FIG. 2. '1' may be distributed intermittently in the matrix of the LDPC information area according to the matching state of the check node and the versatile node in each information group or parity group.

According to an embodiment of the present invention, the position of '1' in the information area 102100 is determined as follows.

First, the position of '1' included in the first column included in each information group is determined first. Thereafter, the position of '1' included in another column included in the information group may be determined using the position of '1' in a periodic manner. According to an embodiment of the present invention, the position of '1' in the first column of the information group and the position of '1' in the second column are determined to be different by the size of Q.

As shown in Fig. 2, the right side of the information area A parity region 102200 in a dual diaginal matrix form is located.

Puncture may be performed on the versatile nodes included in the parity region 102200. In an embodiment, the parity region 102200 may be divided by the size of Q and the columns included in the region divided by the Q size may be bundled to form a parity group (or puncturing group). In addition, puncturing may be performed for each column included in the parity region. The receiver 101300 may perform LDPC decoding by using a matching portion of '1', that is, a matching relationship between a check node and a versatile node.

However, the LDPC information region 102100 may be affected by the puncturing when the LDPC parity region 102200 is punctured.

As shown in FIG. 2, a case in which the variable node 102300 included in the LDPC parity region 102200 becomes punctured is described. Since two '1's are displayed in the displayed variable node 102300, it can be seen that there are two check nodes matched with the punctured variable node 102300.

Looking at the LDPC information area 102100 on the left, it can be seen that the information groups include check nodes matched with the punctured variable node 102300. Also, when the dotted line connected to the indication '1' of the parity region is examined, it can be seen that there are two

information groups

102310 and 102320 including the check node matched with the punctured versatile node 102300. In the two

information groups

102310 and 102320, when the barrier node matched with the check node affected by the punctured barrier node 102300 is included in the two information groups 10210 and 102420 ) May be present. Therefore, when the versatile node 102300 is punctured, the same result as that of the

portions

102410 and 102420 marked with two '1's is removed to become' 0 '(zero). That is, the puncturing disappears the matching relationship between the check node and the versatile node of the information area, and as a result, the information to be used for LDPC decoding is reduced, thereby reducing the LDPC decoding performance of the receiver 101300.

3 is another embodiment of an LDPC block according to the present invention. The LDPC block may include a K size information region 103100 and a parity region 103200 having N-K sizes. N is a parameter representing the total size of the LDPC block. The information area 103100 may include a BCH information block and a BCH Forward Error Correction (BCH FEC) block.

In addition, the parity region 103200 may be divided for each Q size, and puncturing may be performed by grouping columns included in the parity region divided into Q units into one parity group or a puncturing group. In the present invention, a first parity group (or a first puncturing group) is formed by collecting the first columns included in the parity regions divided by Q units, and the second columns included in the parity regions divided by Q units, respectively. In one embodiment, gathering may form a second parity group (second puncturing group). As described above, puncturing is performed only on the parity region 103200, and the punctured bits may not be transmitted to the receiver 101300.

4 is an embodiment of a puncturing process according to the present invention.

4 illustrates an LDPC block in detail, and the LDPC block may include an information region 104100 and a parity region 104200. In the present invention, the variable node is represented by vn and the check node is represented by cn according to an embodiment.

The variable nodes v21, v26, and v31 indicated in FIG. 4 are punctured barrierable nodes, and the check nodes c1, c5, c6, c10, and c11 indicated are punctured by check nodes matching the punctured barrier nodes. Affected by

FIG. 4 illustrates that when the Q value is 5, the parity region 104200 is divided into intervals of Q, that is, a unit including five columns, and a parity group including columns of parity regions divided by five intervals is formed. An embodiment of performing the processing is shown. In this case, the information group having the size of M included in the information area 104100 of the LDPC block may include check nodes c1, c6, and c11 affected by puncturing.

The check node c1 matches the versatile node v1 among the versatile nodes v1, v2, and v3 included in the information group having the size of M, the check node c6 matches the versatile node v2, and the check node c11 is the versatile node. You can see that it matches v3.

The Tanner graph is a graph showing the relationship between the versatile node and the check node based on the position of '1' included in the H-matrix of the LDPC.

FIG. 5 illustrates a Tanner graph for the variable node v2 of FIG. 4. In a Tanner graph, a rectangle is a check node. In one embodiment, a circle represents a variable node. Also, the circle indicated indicates a variable node on which puncturing is performed.

In order to decode the variable node v2, the value of the check node c6 matched with the variable node v2 must be updated. As shown in FIG. 4, in order to update the value of the check node c6, the variable nodes matching the check node and c6, that is, the values of the variable nodes v26 and v27 included in the parity region 104200 should be updated.

However, when only the variable node v26 is punctured as shown in FIG. 5, the value of the variable node v26 will be removed by puncturing. Therefore, in order to know the punctured variable node v26 value, the value of the other check node c5 matching the variable node v26 needs to be updated. Also, in order for the check node c5 to be updated, the value of the variable node v25 matched with the check node c5 must be updated. As a result, the punctured variable node v26 may have no information until it is updated by the variable node v25 connected to the bottom. Therefore, when the versatile node v26 is punctured, the value of the versatile node v26 is unknown, so the check node c6 is affected by the puncturing.

If the check node c6 is updated, the value of the updated check node c6 may be used to update the versatile node v2 again.

Referring to an embodiment illustrated in FIG. 4, when the versatile node v26 is punctured, the check nodes c5 and c6 may be affected, and when the versatile node v31 is punctured, the check nodes c10 and c11 are affected. It can be seen.

In FIG. 6, m denotes an information value to be updated, and m1 to m5 denote information values to be used for the update, respectively. When the receiver 101300 performs decoding on the punctured LDPC block, the amount of information required for decoding may be reduced. Therefore, if puncturing and decoding are performed repeatedly, the amount of information to be used for decoding will be reduced, which will inevitably reduce the decoding performance.

FIG. 6A illustrates an embodiment of updating a bit node when puncturing is not performed. As shown in FIG. 6A, when updating the information value m without performing puncturing, the information values including m1 to m5 are not affected by the puncturing and thus may be used for updating. Therefore, since various information values are used to perform the update, the accuracy of the update can be improved and the update rate can also be increased.

6B illustrates an embodiment of updating a bit node after performing puncturing. When puncturing is performed as shown in FIG. 6B, the information values m1, m3, and m5 are removed by puncturing, and only the information values m2 and m4 may be used for updating. Therefore, since the amount of information used for updating is small as compared with FIG. 6 (a), the update rate may be reduced and performance may be degraded.

Therefore, in the present invention, while puncturing the parity region of the LDPC to improve the transmission efficiency while maintaining a constant code rate, puncturing is performed so that the influence of the information region due to the puncturing is not concentrated in a specific portion and distributedly arranged. In other words, we propose a puncturing pattern.

FIG. 7 is a block diagram illustrating the relationship between the puncturing group and the information group in the LDPC block corresponding to the size of the puncturing group, that is, the Q value is 36 and the block size of 16,200 and the 1/4 code rate.

In the present invention, the information group may be represented as an information group (IG), and in one embodiment, the LDPC information area 102100 includes nine information groups from IG_0 to IG_8. Also, in the present invention, the puncturing group may be expressed as G (Group), and when the Q value is 36, that is, when the columns spaced by the interval of 36 are grouped into one puncturing group, each of 36 punctures from G0 to G35 One example may include including cheering groups.

As shown in FIG. 7, the left rows represent respective groups of information, and the top columns represent respective puncturing groups. In addition, the rectangles displayed in the inner region indicate that each group of information is affected by the puncturing when puncturing is performed for a specific puncturing group.

For example, if puncturing group G0 is punctured, if information groups IG_0, IG_3, and IG_5 are affected by puncturing, puncturing group G0 column 107100 is related to information groups IG_0, IG_3, and IG_5. It may include three rectangles (107110, 107120, 107130) representing. Since the rectangle 107120 representing the relationship with the information group IG_3 is more affected by the puncturing than the other information groups IG_0 and IG_5, the rectangle 107120 shows the relationship with the information groups IG_0 and IG_5 as shown in FIG. The

rectangles

107110 and 107130 may be displayed differently.

When puncturing group G11 is punctured, information groups IG_0, IG_1, IG_3 and IG_8 may be affected by puncturing. In this case, the puncturing group G11 column 107200 may include four

rectangles

107210, 107220, 107230, and 107240 indicating a relationship with the information groups IG_0, IG_1, IG_3, and IG_8. In this case, the display of the four

rectangles

107210, 107220, 107230, and 107240 is the same because the information groups IG_0, IG_1, IG_3, and IG_8 are affected by the puncturing.

As described above, the information group and the puncturing group may include a versatile node corresponding to a column and a check node corresponding to a row. The Tanner graph of FIG. 8 illustrates the variable node and the check node included in each information group and the puncturing group in a figure, and more specifically illustrates the block diagram of FIG. 7 by expressing the relationship between the variable node and the check node as a line. will be.

A circle shown in FIG. 8 means a variable node, and a rectangle means a check node. Marked rectangles 108110.108310,108320 denote check nodes affected by puncturing, and

marked circles

108120,108315,108325 denote punctured variable nodes.

FIG. 8 shows a check node affected by puncturing and any verifiable nodes affected by puncturing and punctured if any puncturing group G0 is punctured 108100, 108200 Tanner graphs are shown for each information group IG_0, IG_1, and IG_3.

(A) of FIG. 8 is a Tanner graph of the information group IG_0, and any variable node 108100 included in the information group IG_0 may be connected with the check node 108110 updated by the punctured variable node 108120. have.

8 (b) is a Tanner graph of the information group IG_1, and any variable node 108200 included in the information group IG_1 is not connected to the check node updated by the punctured variable node.

8 (c) is a Tanner graph of information group IG_3, in which any variable node 108300 included in information group IG_3 is updated by two check nodes punctured by punctured

variable nodes

108315 and 108325. And the

respective ones

108310 and 108320, respectively.

As shown in FIG. 8, any variable node 108200 included in the information group IG_1 is not connected to the check node affected by the puncturing, so the update rate may be the fastest. On the other hand, any of the variable nodes 108300 included in the information group IG_3 are affected by two

check nodes

108310 and 108320 which are updated by puncturing, so the update speed may be slow.

As in FIG. 8, the circle illustrated in FIG. 9 means a variable node, and the rectangle means a check node.

Marked rectangles

109110, 109210, 109310, 109320, 109330 and 109340 represent check nodes affected by puncturing, and

marked circles

109120, 109220, 109315, 108325, 109335 and 109345 are punctured barriers Means nodes.

FIG. 9 shows check nodes affected by puncturing, check nodes affected by puncturing, and any connected variable nodes affected by puncturing, when puncturing groups G0 and G5 are punctured. Tanner graphs showing the relationship with (109100, 109200, 109300) are shown for each information group IG_0, IG_1, and IG_3.

(A) of FIG. 9 is a Tanner graph of the information group IG_0, and any variable node 109100 included in the information group IG_0 may be connected to the check node 109110 updated by the punctured variable node 109120. have.

9 (b) is a Tanner graph of the information group IG_1, and any variable node 109200 included in the information group IG_1 may be connected to the check node 109210 updated by the punctured variable node 109220. have.

9 (c) is a Tanner graph of the information group IG_3, in which any variable node 109300 included in the information group IG_3 is updated by the punctured punctureable

variable nodes

109315, 109325, 109335, and 109345. Four

check nodes

109310, 109320, 109330, and 109340 may be connected, respectively.

As shown in FIG. 9, any

variable nodes

109100 and 109200 included in the information group IG_0 and the information group IG_1 are connected to one check node affected by puncturing. On the other hand, any of the variable nodes 109300 included in the information group IG_3 are affected by four

check nodes

109310, 109320, 109330, and 109340 updated by puncturing, so that the update speed may be slow.

When puncturing is performed on the parity region 102200 of the LDPC H-matrix, a particular versatile node among the versatile nodes included in the information region 102100 is intensively matched with check nodes affected by the puncturing. Can be. In this case, as described above, the decoding performance of the receiver may be degraded. Therefore, in the present invention, the order in which puncturing is performed so that matching between a specific versatile node and a check node updated by puncturing among the versatile nodes included in the information area 102100 is not centralized, that is, a puncturing pattern We will present a way to determine this.

As shown in FIG. 10, the upper end of the table represents an information group and the left side represents a puncturing group. In the present invention, the number, average, and variance of the check nodes updated by puncturing to obtain a puncturing pattern such that matching between the versatile node included in the information area 102100 and the check node updated by the puncturing is distributed. May be calculated for each puncturing group and information group, and the puncturing group having the smallest variance value for all information groups may be selected.

In addition, in the present invention, the puncturing group having the least influence on the information groups is selected as the group to perform the first puncturing, and the rest of the puncturing group except the group to perform the first puncturing to determine the second puncturing group. In an embodiment, the puncturing group having the smallest dispersion value may be determined as the second puncturing group by calculating the variance of the processing groups, and the process may be repeated to determine the puncturing pattern.

Specifically, it is as follows.

As shown in FIG. 7, the information groups IG_0 to IG_3 may include up to 24 check nodes updated by puncturing, but the information groups IG4 to IG8 include up to six check nodes updated by puncturing. can do. Therefore, when the puncturing group is punctured, the influence of each puncturing group on the information groups IG_0 to IG_3 may be at least 1/24, and the influence on the information groups IG_4 to IG_8 may be 1/6.

Accordingly, as shown in FIG. 10, the puncturing groups having the least influence on the information groups by puncturing are G13 and G26. This is because when the puncturing group G13 is punctured, it only affects the information groups IG_1 and IG_2 by 1/24, and when the puncturing group G26 is punctured, it affects the information groups IG_0 and IG_1 by 1/24, respectively. . Therefore, the puncturing group G13 may be determined as the group to perform the first puncturing.

When the puncturing group G13 is first punctured, the information groups IG_1 and IG_2 are each affected by 1/24 by the check nodes updated by the puncturing. Subsequently, when the puncturing groups other than the puncturing group G13 are punctured, it may be analyzed to affect how each information group is affected, and among them, the puncturing group having the smallest dispersion value of the influence on the information group may be selected.

For example, when the puncturing group G0 is punctured, the puncturing group G0 is affected by 1/24 in the information group IG_1 and IG_3, 1/24 in the information group IG_2 and 1/6 in the information group IG_5. Can be. If the puncturing group G5 is punctured, it may affect the information groups IG_0, IG_1 and IG_2 by 1/24 and the information group IG_3 by 1/12.

As shown in the table of FIG. 10, since the variance values of the puncturing groups G5 and G6 are 8.145 * 10 ^ -4, it can be seen that the variance value is smaller than that of other puncturing groups. Therefore, one of the puncturing groups G5 and G6 may be selected as the puncturing group to perform the second puncturing. By repeating this process, the puncturing pattern can be determined by selecting the puncturing group to be punctured last.

According to an embodiment of the present invention, the present invention may be included in a signaling information transmission part of a bit interleaver and coded modulation (BICM) module of an OFDM transmission system.

In the present invention, the signaling information transmitter includes a signaling generator 111100, a first FEC encoding unit 111110, a first mapping unit 111120, and a second FEC encoding. One embodiment includes a second FEC encoding unit 111210, a bit interleaver 111220, a demux unit 111230, and a second mapping unit 111240. Yes.

The signaling generator 111100 may generate signaling information. The signaling information may include pre-signaling information (L1-pre signaling) and post-signaling information (L1-post signaling) according to the information content. The pre-signaling information may include information necessary for decoding the post-signaling information, and the post-signaling information may include information necessary for decoding the data to be transmitted by the transmitter. The pre-signaling information may be included in the foremost part of the transmission frame and transmitted, and the post-signaling information and data may be included after the pre-signaling information and transmitted.

The first FEC encoder 111110 may perform FEC coding on the pre-signaling information, and the first mapping unit 111120 may symbol-map the FEC coded pre-signaling information to constellations. In order to properly decode the signaling information and data, it is necessary to decode the pre-signaling information accurately and quickly. Therefore, in the present invention, a bit interleaver may not be used for fast decoding of pre-signaling information, and a binary phase shift keying (BPSK) modulation scheme and code rate 1 having good recovery performance in order to secure high robustness are obtained. Transmission using / 4 may be an example.

The second FEC encoder 111210 may perform FEC coding on post-signaling information, and the bit interleaver 111220 may perform interleaving on a bit basis on the FEC-coded post-signaling information. The demux unit 111230 demuxes the interleaved post-signaling information in units of cells, and the second mapping unit 111240 symbol-maps the demuxed post-signaling information in units of cells to constellations. You can.

The post-signaling information amount is variable and has a large amount of information compared to the pre-signaling information. Since the pre-signaling information includes information related to the post-signaling information, the pre-signaling information can be decoded first and then the decoding of the post-signaling information can be performed.

In the present invention, one of BPSK, QPSK, 16QAM, and 64QAM may be selected according to a case to transmit post-signaling information, and 1/2 may be used as a code rate. However, since post-signaling information includes information for decoding data related to a service, it should be transmitted so as to be more robust than data. In the present invention, when transmitting data with 16 QAM, post-signaling information may be transmitted using BPSK or QPSK having better performance than 16 QAM in the same channel.

The signaling information may include pre-signaling information and post-signaling information. The post-signaling information may include a configurable block, a dynamic block, an extension block, and a CRC block. ) And a padding block. The size of the pre-signaling information is fixed and according to the embodiment of the present invention is 200 bits. The size of the post-signaling information may vary depending on the number of data blocks and the like, and may be a size between 398 and 18408 bits.

The configurable block may include information that may be equally applied over one transmission frame, and the dynamic block may include characteristic information corresponding to the transmission frame currently being transmitted. The extension block may be used when the post-signaling information is extended, and the CRC block may include information used for error correction of the post-signaling information and may have a 32-bit size. In addition, when the post-signaling information is divided into several LDPC blocks and transmitted, the padding block may be used to equally size the information included in each LDPC block, and the size thereof is variable. By adding a variable padding block, post-signaling information values divided into several LDPC blocks can maintain constant performance.

Post-signaling information may be included in one LDPC block and transmitted, or may be divided into several LDPC blocks and transmitted. Therefore, in case of transmitting the post-signaling information to several LDPC blocks, the transmitter 101100 must determine how many LDPC blocks to divide and transmit to each LDPC block. If the size of the post-signaling information to be transmitted (K_post_ex-pad) is known and the size (K_bch) of the information area 102100 included in the LDPC block transmitted through one LDPC block is known, the post-signaling information is included and transmitted. The number N_post_FEC_Block of the LDPC blocks may be determined through Equation 1 below.

[Revision 22.02.2011 under Rule 26]
Equation 1

The size K_L1_PADDING of the padding block added so that post-signaling information may be divided into a plurality of LDPC blocks by the same size may be determined through Equation 2 below.

[Revision 22.02.2011 under Rule 26]
Equation 2

The total size of post-signaling information to be transmitted (K_post) is a value obtained by adding the size of the padding block to the size of the original post-signaling information, which can be expressed as Equation 3 below.

[Revision 22.02.2011 under Rule 26]
Equation 3

In addition, the size (K_sig) of the post-signaling information transmitted in one LDPC block is divided by the number of blocks (N_post_FEC_Block) required to transmit the size (K_post) of the entire post-signaling information. It can be expressed as 4.

[Revision 22.02.2011 under Rule 26]
Equation 4

The above equations may be used to determine the size (Ksig) of post-signaling information to be transmitted in one LDPC block. The above-mentioned shortening and puncturing may be performed to efficiently transmit post-signaling information having a determined size to be transmitted.

The number of bits to be shortened is the size of the post-signaling information to be transmitted through one LDPC block (K_sig) at the size of the information area of the determined LDPC block (K_bch: 16K, K_bch = 7032 when the effective code rate is 4/9). Can be expressed as a subtracted value. The number of punctured bits (N_punc_temp) is determined by multiplying the number of bits to be shortened by a constant value of "6/5", which can be expressed by Equation 5 below.

[Revision 22.02.2011 under Rule 26]
Equation 5

The reason why the number of bits to be shortened is multiplied by a constant of “6/5” is because the LDPC code rate may change according to the size (K_sig) of post-signaling information to be transmitted through one LDPC block.

As described above, the code rate may be determined by a ratio between the size of the entire LDPC block and the size of information (the size of the post-signaling information (K_sig)). The code rate may change, that is, the code rate may change according to the size of the post-signaling information, resulting in a difference in performance of the post-signaling information transmission.

However, since the post-signaling information includes information necessary for the recovery of the data, when it is transmitted, more robustness is required than data transmitted together, but even in this case, robustness must be maintained to a certain degree.

Therefore, when the size of post-signaling information (K_sig) to be transmitted in one LDPC block is small, the code rate is lowered by puncturing for a relatively small number of bits, but the decoding performance decrease due to puncturing is reduced. Regardless of the size of the signaling information (K_sig) it needs to be adjusted to be constant. At this time, the value used to adjust is “6/5”.

The size of post-signaling information to be transmitted (N_post_temp) after shortening and puncturing is performed is the size of post-signaling information (K_sig), BCH parity size (K_bch_parity), and LDPC parity size (N_ldpc). A value obtained by subtracting the number of bits to be punctured (N_punc_temp) from the value of * (1-R_eff_16K_LDPC_1_2)) may be expressed as Equation 6 below.

[Revision 22.02.2011 under Rule 26]
Equation 6

[Revision 22.02.2011 under Rule 26]
In the equation

Is a modulation order and N_P2 is the number of symbols according to a given fast fourier transform (FFT) size. The size N_post of the post-signaling information to be finally transmitted may be determined according to the number of symbols according to the modulation order and the FFT size.

Post-signaling information to be transmitted after the shortening and puncturing is performed after passing through the bit interleaver 111220. If the modulation order is BPSK or QPSK, the bit interleaver may not be used. In the case where the modulation order is 16 QAM, the size (N_post_temp) of post-signaling information to be transmitted after shortening and puncturing is performed may be an integer multiple of 8 and a multiple of 12 in 64 QAM. Therefore, when N_post_temp = 1, the size of N_post_temp can be adjusted by using 2 so that the number of columns of the bit interleaver may be an integer multiple.

The graph shown in FIG. 13 maintains a code rate of 1/4 at 16K, and when transmitted in BPSK modulation, SNR (Signal) according to a change in the size (K_sig) of post-signaling information transmitted in one LDPC block to Noise Ratio) versus BER (Bit Error Rate).

As shown in the graph of FIG. 13, the case where the size (K_sig) of the post-signaling information transmitted in one LDPC block is the smallest 90 bits and the largest 3072 bits will be compared. If the BER is 10 ^ -05, the SNR is -3.8 for 90 bits and the SNR is -6.7 for 3072 bits, which is about 2.8dB.

Therefore, in order to maintain constant performance regardless of the change in the size of post-signaling information transmitted in one LDPC block (K_sig), regardless of the change in the size of the post-signaling information transmitted in one LDPC block (K_sig) according to BER The difference in SNR should be kept to a minimum.

In order to maintain a minimum performance difference, when the size of the post-signaling information transmitted in one LDPC block (K_sig) is small, the number of bits to be punctured may be reduced to reduce the code rate. As shown in Equation 5, the number of bits to be punctured is the number of bits to be shortened, that is, the size of the information area of the predetermined LDPC block minus the size (K_sig) of post-signaling information to be transmitted through one LDPC block. Can be determined by multiplying by a constant value. Eventually, the number of bits to be punctured can be adjusted by adjusting a constant value irrespective of a change in the size K_sig of post-signaling information to be transmitted through one LDPC block. This constant value is called a compensation value.

14 is a block diagram showing a relationship between variables for obtaining a compensation value according to the present invention.

The block diagram shown in FIG. 14 illustrates the relationship between the size (K_sig), the BCH parity size (K_bch_parity), and the size of the LDPC block (N_LDPC) of post-signaling information transmitted in one LDPC block.

Equation 7 below illustrates a process of obtaining a compensation value when using code rate 1/4 (effective code rate 1/5). The number of bits to be punctured (N_punc_temp) may be a value obtained by multiplying the number of bits to be shortened by a compensation value (X).

[Revision 22.02.2011 by Rule 91]
Equation 7

When the compensation value (X) is used, the final code rate (R_eff_post) is the sum of the size of the post-signaling information (K_sig) and the BCH parity size (K_bch_parity) transmitted in one LDPC block. It can be divided by the size (N_post).

Finally, the size of post-signaling information (N_post) to be transmitted is obtained by subtracting the number of shortening bits (K_bch-K_sig) and the number of puncturing bits (X * (K_bch-K_sig)) from the size of the LDPC block (N_LDPC). This can be In this case, when the compensation value X is 4, R_eff_post may be 1/5, which is always 1/5, which is an effective code rate value regardless of a change in the value of K_sig. Therefore, it is understood that the compensation value (X) should have a value less than 4 when reducing and transmitting the performance difference caused by the change of the size (K_sig) of the post-signaling information transmitted in one LDPC block having a code rate of 1/4. Can be.

Equation 8 below shows a process of obtaining a size (N_post) of post-signaling information to be finally transmitted when signaling information having a variable code rate having an effective code rate is 1/5.

[Revision 22.02.2011 by Rule 91]
Equation 8

By substituting the determined compensation value X into Equation 8, the number of bits to be shortened (K_bch-K_sig) and the number of bits to be punctured (N_punc_temp) can be calculated and then the calculated number of bits to be punctured (N_punc_temp) is used. The size N_post of the post-signaling information to be finally transmitted may be expressed as in Equation 9 below.

[Revision 22.02.2011 by Rule 91]
Equation 9

In equation (9)

Is a modulation order and N_P2 is the number of symbols according to a given fast fourier transform (FFT) size.

Although not shown in the drawing, the receiver 101300 of the present invention may include a BICM decoder (Bit Interleaved Coded Modulation decoder). The BICM decoder is for decoding the pre-signaling information and the post-signaling information. The BICM decoder can decode the pre-signaling information before the post-signaling information.

When the pre-signaling information is decoded, the receiver 101300 may extract information for decoding the post-signaling information from the decoded pre-signaling information. Information for decoding the post-signaling information includes an L1_POST_INFO_SIZE field indicating the size of the information area except for the CRC block and the padding block in the post-signaling information, an L1_MOD field indicating the signaling format of the post-signaling information, and a code rate of the post-signaling information. It may include an L1_COD field.

Since the size of the post-signaling information (K_post_ex-pad) is a 32-bit CRC field added to L1_POST_INFOSIZE, the receiver 101300 uses Equation 2 in the number of paddings added by the transmitter 101100 and one LDPC. The size (K_sig) of post-signaling information to be transmitted through the block can be calculated.

In addition, if the size (K_sig) of the post-signaling information to be transmitted through one LDPC block is known, the number of bits (N_punc_temp) to be punctured may be calculated using Equation (8). The compensation value X used at this time is a value preset by the transmitter 101100. When the size of post-signaling information (K_sig) to be transmitted through one LDPC block and the number of bits (N_punc_temp) to be punctured are calculated, the post-signaling information to be finally transmitted using the L1_MOD field and the N_P2 value according to the FFT size The size of N_post can be calculated.

The unshortening and depuncturing unit 101320 knows the number of bits to be shortened (K_bch-K_sig) and the number of bits to be punctured (N_LDPC * (1-R_eff_16K_LDPC_1_4) + (K_sig + K_bch_parity) -N_post) using equations. Can be. In addition, the unshortening and depuncturing unit 101320 may perform depuncturing using the puncturing pattern proposed by the present invention.

The graph shown in FIG. 15 shows a code rate (R_eff_post) according to the size (K_sig) of post-signaling information to be transmitted through one LDPC block when the compensation value (X) is changed by 0.1 to a value between 3.5 and 4 or less. ) Is indicated. The X axis of the graph represents the size (K_sig) of post-signaling information to be transmitted through one LDPC block, and the Y axis represents a code rate (R_eff_post).

When the compensation value (X) is 4, the effective code rate (R_eff_post) is constant at 1/5 (0.2), but when the compensation value (X) is less than 4, post-signaling information to be transmitted through one LDPC block As the size K_sig is smaller, the value of the effective code rate may be smaller.

As shown in FIG. 15, when the compensation value X is 3.5, when the size K_sig of post-signaling information to be transmitted through one LDPC block is 500, R_eff_post drops to less than 0.15. Smaller code rate values increase overhead, resulting in lower transmission efficiency but higher transmission performance. Therefore, in case of signaling information having a small amount of information, a compensation value may be changed to compensate for a decrease in transmission performance due to shortening and puncturing.

According to the present invention, the code rate is 1/4 (effective code rate 1/5), the code length is 16200, and the frame error ratio (FER) is 10 ^ -04.

FIG. 16A illustrates a compensation value X of 3.5 or more when the size K_sig of post-signaling information to be transmitted through one LDPC block has a value between 312 and 3072 bits. When changing to the following values, it is a table showing the size of the SNR gap (Signal to Noise Ratio Gap) for each modulation order according to the compensation value.

As the size of the SNR gap is smaller, it means that the robustness is constant regardless of the change in the size of the signaling information. Therefore, an X value having the smallest SNR value for each modulation order used when transmitting the signaling information can be selected and used as a compensation value. have.

As shown in (a) of FIG. 16, when the modulation order is BPSK and QPSK, it may have the smallest SNR gap value when the compensation value X is between 3.5 and 3.7, and when the modulation order is 16 QAM. It can be seen that (X) has the smallest SNR gap value when the value is between 3.6 and 3.8, and the modulation order has the smallest SNR gap size when the compensation value (X) is between 3.8 and 4.0 when the modulation order is 64QAM. It can also be seen that the compensation value (X) that satisfies to have a small SNR gap size in all modulation orders can be 3.8.

FIG. 16B is a table showing the size of the SNR for each modulation order when transmitting data related to a service.

As described above, when transmitting the signaling information, it is necessary to secure greater robustness than when transmitting data, and thus it may be transmitted using a modulation order lower than the modulation order used when transmitting data. In the present invention, when the QPSK is used for data transmission, the BSPK may be used for signaling transmission as an embodiment.

FIG. 17 is a table illustrating a result of comparing an SNR value according to a change in the size of signaling information and an SNR value according to a change in the size of signaling information with an SNR value of the data of FIG. 16.

FIG. 17 (a) shows the SNR of the case where the performance is the worst at each compensation value when a given compensation value is changed to a value between 3.5 and 4 according to a change in the size of signaling information K_sig according to the present invention. One embodiment shows the values.

FIG. 17B is a diagram illustrating a difference between an SNR value of signaling information shown in FIG. 17A and an SNR value of data shown in FIG. 16B according to the present invention. As described above, transmission of signaling information requires greater robustness than transmission of data.

Therefore, when data is transmitted in QPSK, signaling information may be transmitted in BPSK having a modulation order lower than that of QPSK. In the case of data transmitted in 16QAM, signaling information may be transmitted in QPSK. In this case, the greater the difference between the SNR value of the data and the SNR value of the signaling information, the greater the robustness required for transmitting the signaling information.

As shown in (b) of FIG. 17, when the signaling information is transmitted to the BPSK, when the compensation value is 3.5, it can be seen that the difference between the SNR value of the data and the SNR value of the signaling information is greatest and transmitted to the QPSK. When the compensation value is 3.5, it can be seen that the difference between the SNR value of the data and the SNR value of the signaling information is greatest. In addition, when the signaling information is transmitted in 16QAM, it can be seen that the difference between the SNR value of the data and the SNR value of the signaling information is greatest when the compensation values are 3.5 and 3.6. When the signaling information is transmitted by 64QAM, the compensation values are 3.5 to 3.7. In this case, it can be seen that the difference between the SNR value of the data and the SNR value of the signaling information is greatest.

16 (a) and (b) of FIG. 16 show that the SNR gap according to the size of the signaling information is the smallest and the difference between the SNR of the data and the SNR of the signaling information is the largest regardless of the modulation order. It can be seen that it can be an optimal compensation value.

The information generator 101110 generates information to be transmitted (S118100). According to an embodiment, information to be transmitted may be generated in an OFDM broadcasting system, and the information may include signaling information including data related to a service or information necessary for decoding data as described above.

The zero padding unit 101120 inserts zero padding information to enable BCH encoding on the generated information (S118200). In this case, the position of the zero padding to be inserted may be determined according to a shortening pattern to be performed in the shortening step.

The BCH encoding unit 101130 performs BCH encoding on information in which zero padding information is inserted (S118300). The BCH encoder 101130 may add BCH parity check bits to the LDPC encoder 101140 by adding BCH parity check bits to information in which zero padding information is inserted.

The LDPC encoder 101140 may generate an LDPC block by adding LDPC parities to the received BCH encoding information, and perform LDPC encoding on the basis of the generated LDPC block (S118400). In this case, the LDPC block may include an information region 102100 including information and a parity region 102200 including LDPC parities.

The shortening and puncturing unit 101150 may perform shortening and puncturing on the encoded LDPC block. The shortening is performed on the information area 102100 of the LDPC block, and in this case, the shortening is performed according to a specific shortening pattern (S118500). In addition, puncturing is performed according to the puncturing pattern for the parity region 102200 of the LDPC block (S118600). In this case, as illustrated in FIG. 9, the puncturing pattern may be a puncturing pattern proposed by the present invention.

The OFDM demodulator 101310 OFDM demodulates the received information (S119100). The information to be OFDM demodulated may include signaling information including data related to a service or information necessary for decoding data as described above.

The unshortening and depuncturing unit 101320 performs unshortening by inserting a maximum value of a log-likelihood ratio (LLR) operation into the information area according to a shortening pattern defined by the transmitter 101100 with respect to the OFDM demodulated information. It may be (S119200). In addition, according to the puncturing pattern previously defined by the transmission unit 101100, de-punching may be performed by inserting a value ('0'; zero) corresponding to an unknown in the parity region where puncturing is performed (S119300). .

The LDPC decoding unit 101330 may perform LDPC decoding on the unshortened and depunctured information (S119400).

The BCH decoding unit 101340 may perform BCH decoding on the LDPC decoded information (S119500), and the information extraction unit 101350 may receive information to be received in the remaining portions except for the zero-padded position with respect to the BCH decoded information. It may be extracted (S119600).

As described above, in the best mode for carrying out the invention, related matters have been described.

As described above, the present invention may be applied in whole or in part to a digital broadcasting system.

Claims

Generating information;

Inserting zero padding information into the generated information;

Performing BCH (Bose-Chaudhuri-Hocquenghem) encoding on the information into which the zero padding information is inserted;

Performing low-density parity-check (LDPC) encoding on the BCH-encoded information, wherein the LDPC-encoded information includes an information region and a parity region;

Removing a zero value included in the information area of the LDPC encoded information; And

And performing puncturing on the parity region of the LDPC encoded information.
The method of claim 1,

The parity region of the LDPC encoded information includes a plurality of puncturing groups.
The method of claim 2,

The information area of the LDPC encoded information includes a plurality of information groups.
The method of claim 3, wherein

Performing puncturing on the parity region of the LDPC encoded information may include: information affected by the puncturing when the plurality of information groups include information affected by the puncturing; Determining a puncturing order of the plurality of puncturing groups so that they are distributedly arranged; And

Puncturing the plurality of puncturing groups according to the determined puncturing order.
The method of claim 1,

And performing LDPC encoding on the BCH encoded information comprises determining a code rate for performing the LDPC encoding.
The method of claim 5,

Determining a compensation value using the code rate, and determining a size of a parity region for performing the puncturing using the determined compensation value;

And varying the code rate using the determined compensation value.
An information generator for generating information;

A zero padding unit inserting zero padding information into the generated information;

A BCH encoding unit for performing BCH (Bose-Chaudhuri-Hocquenghem) encoding on the information into which the zero padding information is inserted;

An LDPC encoding unit performing low-density parity-check (LDPC) encoding on the BCH-encoded information, wherein the LDPC-encoded information includes an information region and a parity region;

Remove '0' included in the information region of the LDPC encoded information,

And a shortening and puncturing unit for performing puncturing on the parity region of the LDPC encoded information.
The method of claim 7, wherein

The parity region of the LDPC encoded information includes a plurality of puncturing groups.
The method of claim 8,

And the information region of the LDPC encoded information includes a plurality of information groups.
The method of claim 9,

The shortening and puncturing unit, in the case where the plurality of information groups include information affected by the puncturing, puncturing order of the plurality of puncturing groups so that the information affected by the puncturing is distributedly arranged. To determine,

And puncturing the plurality of puncturing groups according to the determined puncturing order.
The method of claim 7, wherein

And the LDPC encoder determines a code rate for performing LDPC encoding.
The method of claim 11,

The LDPC encoder determines a compensation value using the code rate, determines a size of a parity region for performing the puncturing using the determined compensation value, and changes the code rate using the determined compensation value. Broadcasting signal transmission apparatus.
OFDM demodulating information including an information area and a parity area;

Performing unshortening on the information area included in the information, and performing depuncturing on the parity area included in the information;

Performing LDPC decoding on the unshortened and depunctured information;

Performing BCH decoding on the LDPC decoded information; And

And extracting information included in the BCH decoded information.
An OFDM demodulator for OFDM demodulating information including an information area and a parity area;

An unshortening and depuncturing unit for performing unshortening on the information area included in the information and performing depuncturing on the parity area included in the information;

An LDPC decoding unit for performing LDPC decoding on the unshortened and depunctured information;

A BCH decoding unit for performing BCH decoding on the LDPC decoded information; And

And an information extraction unit for extracting information included in the BCH decoded information.