Apparatus and method for transmitting and receiving a signal

The present invention relates to a method of transmitting and receiving a signal and apparatus for transmitting and receiving a signal, and more precisely, a method of transmitting and receiving a signal and an apparatus for transmitting and receiving a signal that is capable of improving data transmission efficiency.

Description of the state of the art [0002] Due to the development of digital broadcasting technology, users can choose a high definition (HD) video image. With the continuous development of the compression algorithm and high performance of the equipment, users will have better conditions in the future. A digital television system (DTV) can receive a digital broadcast signal and provide users with various additional services, as well as a video signal and an audio signal.

[0003] The Digital Video Broadcasting (DVB) -C2 standard is the third specification that has joined the second generation DVB family of transmission systems. Developed in 1994, DVB-C is currently used in more than 50 million cable receivers around the world. In line with other second-generation DVB systems, DVB-C2 uses a combination of Low-density parity-check (LDPC) and BCH (Bose-ChaudhuriHocquengham) codes. This strong prediction error correction (FEC) improves the carrier-to-noise ratio by approximately 5 dB relative to DVB-C. Suitable bit interleaving schemes optimize the overall robustness of the FEC system. These frames, along with their header, are referred to as Physical Layer Pipes (PLP). One or more PLPs are multiplexed to form data slices. For each segment, two-dimensional interleaving (in terms of time and frequency) is used, which allows the receiver to eliminate the impact of pulse weakening and selective frequency interference, such as single-frequency input.

[0004] With the development of these digital broadcasting technologies, the requirements for services associated with, for example, video signals and audio signals, as well as the amount of data desired by users or the number of broadcasting channels have been gradually increasing. The DVB project "Frame structure channel and modulation for a second generation terrestrial television broadcasting system (DVB-T2)", Digital Video Broadcasting [online], June 2008 (2008-06-01), reveals techniques for coding a channel frame structure for systems broadcast transmission.

SUMMARY OF THE INVENTION [0005] Each occurrence of the expression "embodiment" herein should be considered as an "aspect of the invention", the invention being defined in the appended independent claims. Accordingly, the present invention is directed to a method of transmitting and receiving a signal and an apparatus for transmitting and receiving a signal that substantially eliminates one or more problems resulting from limitations and disadvantages of methods and devices known in the art.

VP / 4263 / AGR

[0006] The object of the present invention is to provide a method for transmitting a broadcasting signal to a receiver, according to claim 1.

[0007] In a further aspect, the present invention provides a method of receiving a broadcasting signal according to claim 9.

[0008] Yet another aspect of the present invention provides a transmitter for transmitting a broadcasting signal to a receiver according to claim 5.

[0009] Yet another aspect of the present invention provides a receiver for receiving a broadcasting signal according to claim 10.

BRIEF DESCRIPTION OF THE DRAWINGS [0010] The accompanying drawings, which are included in the present description to enable further understanding of the invention and are included in it and form a part of this application, illustrate embodiment (s) of the invention and together with the description serve to explain the essence of the invention. . In the figures:

Fig. 1 illustrates an example of 64-quadrature amplitude modulation (QAM) used in European DVB-T.

Fig. 2 illustrates a method for creating Gray binary reflection codes (BRGC 'Binary Reflected Gray Code).

Fig. 3 illustrates a Gaussian-like output by modifying the 64-QAM modulation used in DVB-T.

Figure 4 illustrates Hamming distance between reflected pairs in BRGC.

Fig. 5 illustrates the characteristics of the QAM modulation where there is a reflected pair for each axis I and the Q axis.

Fig. 6 is a method of modifying QAM using Reflected Pair of BRGC.

Fig. 7 is an example of a modified 64/256/1024/4096-QAM modulation.

<td>Figs. 8-9 illustrate an example reflected BRGC.</td><td>modified</td><td>64-QAM modulation</td><td>with</td><td>application</td><td>steam</td>

<td>Fig. 10-11 is an example reflected BRGC.</td><td>modified</td><td>256-QAM modulation</td><td>with</td><td>application</td><td>steam</td>

<td>Fig. 12-13 is an example reflected BRGC (0 ~ 511).</td><td>modified</td><td>1024-QAM modulation</td><td>with</td><td>application</td><td>steam</td>

<td>Figs. 14-15 are an example reflected BRGC (512 ~ 1023).</td><td>modified</td><td>1024-QAM modulation</td><td>with</td><td>application</td><td>steam</td>

<td>Figs. 16-17 illustrate an example</td><td>modified</td><td>4096-QAM modulation</td><td>with</td><td>application</td><td>steam</td>

reflected BRGC (0 ~ 511).

VP / 4263 / AGR

EP 2 323 331 B1

<td>Fig. 18-19 illustrates an example reflected BRGC (512 ~ 1023).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

<td>Figs. 20-21 illustrate an example reflected BRGC (1024 ~ 1535).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

<td>Fig. 22-23 is an example reflected BRGC (1536 ~ 2047).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

<td>Fig. 24-25 illustrates an example reflected BRGC (2048 ~ 2559).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

<td>Figs. 26-27 illustrate an example reflected BRGC (2560 ~ 3071).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

<td>Figs. 28-29 are an example reflected BRGC (3072 ~ 3583).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

<td>Figs. 30-31 illustrate an example reflected BRGC (3584 ~ 4095).</td><td>modified</td><td>modulation</td><td>4096-QAM</td><td>with</td><td>application</td><td>steam</td>

Fig. 32 is an example of bit mapping of modified QAM, where 256-QAM is modified using BRGC.

Fig. 33 is an example of transformation of MQAM into a non-uniform constellation.

Fig. 34 is an example of a digital broadcasting system.

Fig. 35 shows an example of an input processor.

Fig. 36 is an illustration that can be included in the base band (BB).

Fig. 37 is an example of BICM.

Fig. 38 is an example of a shortened / punctured encoder.

Fig. 39 is an example of using various constellations.

Fig. 40 is another example of cases where compatibility between conventional systems is considered.

Fig. 41 shows a frame structure which comprises preamble for L1 signaling and data symbol for PLP data.

Fig. 42 is an example of a frame builder.

Fig. 43 is an example of pilot insert (404) shown in Fig. 4.

Fig. 44 is a structure of SP.

Fig. 45 is a new SP structure or pilot signal (PP) 5 'pattern.

Fig. 46 is a suggested PP5 'construction.

Fig. 47 is a relationship between data symbol and preamble.

Fig. 48 is another relationship between data symbol and preamble.

VP / 4263 / AGR

EP 2 323 331 B1

Fig. 49 is an example of a cable channel delay profile.

Fig. 50 is a scattered pilot structure that uses z = 56 and z = 112.

Fig. 51 is an example of OFDM based modulator.

Fig. 52 is an example of preamble structure.

Fig. 53 is an example of preamble decoding.

Fig. 54 is a process for designing a more optimized preamble.

Fig. 55 is another example of a preamble structure

Fig. 56 is another example of Preamble decoding.

Fig. 57 is an example of preamble structure.

Fig. 58 is an example of L1 decoding.

Fig. 59 is an example of an analog processor.

Fig. 60 is an example of a digital receiver system.

Fig. 61 is an example of an analog processor used in a receiver.

Fig. 62 is an example of a demodulator.

Fig. 63 is an example of a frame parser.

Fig. 64 is an example of a BICM demodulator.

Fig. 65 is an example of LDPC decoding using shortening / puncturing.

Fig. 66 is an example of an output processor.

Fig. 67 is an example of L1 block repetition rate of 8 MHz.

Fig. 68 is an example of L1 block repetition rate of 8 MHz.

Fig. 69 is a new L1 block repetition rate of 7.61 MHz.

Fig. 70 is an example of L1 signaling block that is transmitted in a frame header.

Fig. 71 is a simulation of the preamble structure and the L1 block.

Fig. 72 is an example of symbol interleaver.

Fig. 73 is an example of transmitting L1 block.

Fig. 74 is another example of L1 signaling block that is transmitted within a frame header.

Fig. 75 is an example of frequency or time interleaving / deinterleaving.

DESCRIPTION OF THE PREFERRED EMBODIMENTS [0011] Detailed information will now be provided about preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Anywhere

VP / 4263 / AGR

As it is possible, the same reference numerals will be used in all figures of the drawing with reference to the same or similar parts.

[0012] In the following description, the term "service" as used indicates the transmitted content that can be transmitted / received by the signal transmission / reception apparatus.

[0013] Quadrature Amplitude Modulation (QAM) using Gray binary reflection code (BRGC) is used for modulation in a broadcast environment when conventional coded modulation is applied with bit interlaced (BICM). Bit Interleaved Coded Modulation. Fig. 1 is an example of 64-QAM modulation used in European DVB-T.

[0014] The BRGC may be created by the method shown in Fig. 2. The N-bit BRGC may be formed by adding the reverse code (nl) -bit BRGC (i.e., the reflected code) at the back (nl) -bit, by adding zeros in front of the original (nl) -bit BRGC, and by adding the ones in front of the reflected code. The BRGC code created in this way has a Hamming distance between adjacent codes of one (1). In addition, when the BRGC is used in QAM, the Hamming distance between the point and the four points nearest to this point is one (1) and the Hamming interval between the point and the other four points that are the closest to this point is two (2).

[0015] In order to immunize the system against Gaussian additive white noise (AWGN), it is possible to cause the distribution of signals transmitted from the transmitter to be close to the Gaussian distribution. To make this possible, the location of points in the constellation can be modified. Fig. 3 illustrates a Gaussian-like output by modifying the 64-QAM modulation used in DVB-T. Such a constellation may be referred to as non-uniform QAM (NU-QAM) (Non-Uniform QAM)).

[0016] In order to create a constellation of non-uniform QAM, a Gaussian Cumulative Distribution Function (CDF) can be used. In the case of modulation 64, 256, or 1024 QAM, i.e., 2^ΛΝ AM, QAM can be divided into two independent N-PAM modulations (Pulse-Amplitude Modulation). By dividing the Gaussian distributor of CDF into N parts with identical probabilities and allowing the signal point in each part to represent this part, a constellation having a Gaussian distribution can be obtained. In other words, the xj coordinate of the newly defined non-uniform NPAM can be defined as follows:

<img file="PL2323331T3_D0001.tif" />

e 2 dx =

pj '

3 2N - 1) ^{Pj 6 {}2N'2N '' '2N ^} (Equation 1) [0017] Fig. 3 is an example of transformation of 64QAM modulation used in DVB-T into NU-64QAM modulation using the above methods. Figure 3 shows the result of modifying the coordinates of each I axis and Q axis using the above methods and mapping

VP / 4263 / AGR

The previous constellation points have been redefined. In the case of 32, 128 or 512 QAM modulation, i.e., cross-QAM modulation which is not modulation 2^ΛΝ QAM, as a result of proper modification of the Pj value, a new coordinate can be found.

[0018] In one embodiment of the present invention, QAM modulation using BRGC may be modified by using BRGC characteristics. As illustrated in Fig. 4, the Hamming distance between reflected pairs in the BRGC is one, because it differs only in one bit that is added to the front of each code. Fig. 5 illustrates the characteristics of the QAM modulation where there is a reflected pair for each axis I and the Q axis. In this figure there is a reflected pair on each side of the black dotted line.

[0019] By using reflection pairs existing in QAM, the average power of the QAM constellation can be lowered while maintaining the Gray mapping rule in QAM. In other words, in a constellation in which the average power is normalized to 1, the minimum Euclidean distance in the constellation can be increased. When such modified QAM is used in broadcast or communication systems, it is possible to obtain a more resilient system using the same energy as a conventional system or a system with the same performance as the conventional system, but which uses less energy.

[0020] Fig. 6 is a method of modifying QAM using Reflected Pair of BRGC. Fig. 6a illustrates a constellation and Fig. 6b is a flowchart for modifying QAM using Reflected Pair of BRGC. At the beginning you need to find the destination point that has the most power from the constellation points. Candidate points are points at which you can move the destination point and which are in the immediate vicinity of the points from the pair of the reflected destination. Then, from among the candidate points, you need to select an empty point (ie, a point that is not yet occupied by other points) having the lowest power, after which the strength of the target point and the power of the candidate point are compared. If the strength of the candidate point is smaller, the destination point moves to the candidate point.

[0021] Fig. 7 is an example of modified 64/256/1024/4096-QAM. The values ​​mapped according to the Gray principle correspond to the values ​​shown in Figs. 8 ~ 31 respectively. In addition to these examples, other types of modified QAM can be implemented that allow identical power optimization. This is due to the fact that the destination can go to many candidate points. The suggested modified QAM modulation can be used not only for 64/256/1024/4096-QAM modulation, but also for QAM cross-distribution, QAM larger size modulation, or for modulations using different BRGC other than QAM.

[0022] Fig. 32 is an example of bit mapping according to a modified QAM when the 256-QAM modulation is modified using the BRGC. Fig. 32a and Fig. 32b show the mapping of the Most Significant Bits (MSB). Points marked as filled circles represent one-way mapping, while points marked as empty circles represent zero mappings. In the same way, each bit is mapped as

VP / 4263 / AGR

FIG. 2 is shown in FIGS. (A) to (h) in FIG. 32, until the least significant bits (LSB) are mapped to Least Significant Bits. As illustrated in Fig. 32, a modified QAM may allow the decision to receive a bit using only I or Q axis as in conventional QAM, except for the bit that is closest to the MSB (Fig. 32c and Fig. 32d).

By using these characteristics, a simple receiver can be obtained by partially modifying the receiver for QAM. An efficient receiver can be obtained by checking both I and Q values ​​only when determining a bit adjacent to the MSB and by calculating only I or Q for the remaining bits. This method can be used in an approximate LLR (Approximate LLR) (LLR; Log-Likelihood Ratio)), in exact

LLR (called Exact LLR), or in a hard decision.

[0023] By using modified QAM (Modified-QAM) or MQAM, which uses the characteristics of the above BRGC, a non-uniform or NU-MQAM constellation can be obtained. In the above equation, in which the CDF distributor of the Gaussian distribution is used, the value Pj can be modified to be adapted to MQAM. As in QAM, in MQAM, two PAMs having an I axis and a Q axis can be considered. However, unlike QAM, where the number of points corresponding to the values ​​of each PAM axis is identical, the number of points changes in MQAM. If the number of points corresponding to the j-th PAM value is defined as nj in MQAM, where there are a total of M constellation points, then the Pj value can be defined as follows:

^cj 1 ii.

, - == e 2 dx = p, ·,

-rn / Ζπ

Pj =

Σ '=^{] -1}η · + ^ 7-ti = o + 7 'A = 0 (equation 2) [0024] By applying the newly defined value Pj, the MQAM can be transformed into a non-uniform constellation. The value Pj can be defined as follows, e.g. for 256-MQAM.

(2,5 10 22 36 51 67 84 102 119.5 136.5 154 172 189 205 220 234 246 253.5) ^{Pi 6 {}256^.256^.256^.256^.256^.256^.256^.256^. 256 ^. 256 ^.256^.256^.256^.256^.256^.256^.256^. 256 ^} [0025] Fig. 33 is an example of transformation of MQAM into non-uniform constellation. Constellation

5 NU-MQAM created using these methods can retain receiver characteristics

MQAM with modified coordinates for each PAM. In this way, an effective receiver can be obtained. In addition, a more noise-resistant system than the earlier NUQAM can be obtained. In order to obtain a more efficient transmission system of the broadcast signal, it is possible to hybridize MQAM and NU-MQAM. In other words, the system is more resistant to interference

0 can be obtained by using MQAM for an environment in which an error correction code with a high code rate is applied and in addition through the use of NU-MQAM. In this case, the transmitter may provide the receiver with information regarding the code rate of the error correction code currently used and the type of modulation currently used, so that the receiver can conduct demodulation according to the currently used modulation.

[0026] Fig. 34 is an example of a digital transmission system. The inputs can include multiple MPEG-TS streams or GSE streams (generic stream encapsulation).

VP / 4263 / AGR

EP 2 323 331 B1

GeneralStream Encapsulation). The input processor module 101 may add transmission parameters to the input stream and perform scheduling for the BICM 102. The BICM 102 module may add redundancy and interleaving data to correct the transmission channel error. The frame builder 103 can build frames by adding information signaling the physical layer and pilot signals. The modulator 104 can modulate the input symbols with effective methods. The analog processor 105 may carry out various processes to convert the input digital signals into analog output signals.

[0027] Fig. 35 shows an example of an input processor. The MPEG-TS or GSE input stream can be transformed by the input preprocessor in total of the streams that will be processed independently. Each of these streams may be in the form of a complete TS frame that includes components for multiple services or in the form of a minimal TS frame that includes a service component (i.e., video or audio). In addition, each of these streams may be a GSE stream that broadcasts multiple services or a single service.

[0028] The input interface module 202-1 may allocate a number of input bits equal to the maximum capacity of the baseband data area (BB). A padding can be inserted to fill the LDPC / BCH code block capacity. The input stream synchronizing module 203-1 may provide a regeneration mechanism, in the receiver, of a transport stream clock (or packetized generic stream) to guarantee a constant bit rate and delay between the ends of the path.

[0029] In order to allow the Transport Stream recombining without requiring additional memory in the receiver, the input Transport Streams are delayed by delay compensators 204-1 ~ n considering interleaving parameters of the data PLPs in the group and the corresponding common PLP. Null packet deleting modules 205-1 ~ n can increase transmission efficiency by removing the inserted null packet for a case of VBR (Variable Bit Rate) service. Cyclic Redundancy Check (RCR) 206-1 ~ n coding modules can add CRC parity to increase the reliability of BB frame transmissions. BB header insert 207-1 n can add a BB frame header in the start portion of the BB frame. information,

[0030] A Merger / slicer module 208 can perform BB frame slicing from each PLP, merging BB frames from multiple PLPs, and scheduling each BB frame within a transmission frame. In this way, the combining / slicing module 208 can output L1 signaling information that relates to the PLP allocation in the frame. Finally, the BB 209 mixing module can randomize the bit input streams to minimize the correlation between the bits within the bit streams. The modules that are shaded in Fig. 35 mean modules used when the transmitting system uses a single PLP, while the remaining modules shown in Fig. 35 are modules used when the transmitting device uses a plurality of PLPs.

[0031] Fig. 37 is an example of BICM module. Fig. 37a shows the data path and Fig. 37b shows the L1 path of the BICM module. The external encoder 301 module and the internal encoder module 303 may add redundancy to the input bitstreams to correct the error. Module

VP / 4263 / AGR

The external interleaver 302 and the interleaving 304 can interleave the bits to prevent pulse errors. External interleaver module 302 may be omitted if BICM modulation is intended for DVB-C2. The bit demultiplexer module 305 may control the reliability of the output of each bit from the interleaving module 304. The symbol mapper 306 may convert the bit input streams into symbol streams. At this time, it is possible to use any of the conventional QAM modulations, MQAM modulation, which uses the above mentioned BRGC to improve performance, NU-QAM modulation that uses non-uniform modulation, or NU-MQAM modulation that uses non-uniform modulation using BRGC in to improve performance. To construct the system, which is more resistant to noise, you can consider combinations of modulations using MQAM and / or NUMQAM depending on the code rate of the error correction code and the constellation capacity. At this time, the symbol mapper module 306 may use the corresponding constellation according to the coding and constellation capacity of the constellation. Fig. 39 is an example of such combinations.

[0032] Case 1 illustrates an example of using only NU-MQAM with low coding efficiency to implement a simplified system. Case 2 illustrates an example of using an optimized constellation for each coding efficiency. The transmitter can send to the receiver information regarding the efficiency of coding the error correction code and constellation capacity, so that the receiver can use the appropriate constellation. Fig. 40 is another example of cases in which compatibility between conventional systems is considered. In addition to these examples, further combinations are possible to optimize the system.

The ModCod 30 inserter module shown in Fig. 37 can receive feedback regarding Adaptive Coding and Modulation (VCM; Variable Coding and Modulation) and add information about parameter used in coding and modulation to the FEC block in the form of a header. The header containing the modulation type / coding efficiency (ModCod) may contain the following information:

* FEC type (1 bit) - long or short LDPC * Coding efficiency (3 bits) * Modulation (3 bits) - up to 64K QAM * PLP identifier (8 bits) [0034] Symbol interleaver 308 may perform interleaving in the symbol art in for additional interlacing effects. Processes similar to those performed on the data path can be performed on the L1 signal path, but using possibly different parameters (301-1 ~ 308-1). At this point, the shortened / punctured coding module (303-1) can be used for internal coding.

[0035] Fig. 38 shows an example of LDPC encoding using shortening / puncturing. The shortening process can be performed on input blocks that contain fewer bits than the required number of bits for LDPC encoding, because many zero bits can be entered

VP / 4263 / AGR

EP 2 323 331 B1 required for LDPC encoding (301c). Input streams of bits with entered zeros may contain parity bits throughout the LDPC encoding process (302c). At this time, in order to obtain bitstreams that correspond to the original bitstreams, zeros (303c) can be removed and bit-perforation (304c) can be performed to obtain the evenness of the bitstreams according to coding efficiency. These processed information bitstreams and even bit streams can be multiplexed to primary sequences and outputted (305c).

[0036] Fig. 41 shows a frame structure which comprises preamble for L1 signaling and data symbol for PLP data. It can be seen that preamble and data symbols are generated cyclically, using one frame as a unit. The data symbols include PLP type 0 which is transmitted using fixed modulation / fixed coding, and PLP type 1 which is transmitted using variable modulation / variable coding. In the case of the PLP type 0, information such as modulation, type FEC, and FEC coding efficiency are transmitted in the preamble (see FIG. 42, inserting the frame header 401). In the case of a type 1 PLP, the corresponding information may be transmitted in the header of the FEC block of the data symbol (see Fig. 37, inserting the ModCod 307 header). As a result of separating the types of PLPs, The ModCod overhead can be reduced by 3 ~ 4% relative to the total transmission rate for a type PLP PLP that is transmitted at a fixed data rate. At the receiver, in the case of a fixed modulation / fixed PLP coding of the PLP type 0, the frame header removal element r401 shown in Fig. 63 may extract modulation information and FEC coding efficiency and provide the information obtained to the BICM decoding module. In the PLP type 1 modulation / variable PLP coding variable, the ModCod extraction modules, r307 and r307-1 shown in Fig. 64, can extract and provide parameters necessary for BICM decoding. in the case of a fixed modulation / fixed PLP encoding of the PLP type 0, the frame header removal element r401 shown in Fig. 63 may extract modulation information and FEC coding efficiency and provide the information obtained to the BICM decoding module. In the PLP type 1 modulation / variable PLP coding variable, the ModCod extraction modules, r307 and r307-1 shown in Fig. 64, can extract and provide parameters necessary for BICM decoding. in the case of a fixed modulation / fixed PLP encoding of the PLP type 0, the frame header removal element r401 shown in Fig. 63 may extract modulation information and FEC coding efficiency and provide the information obtained to the BICM decoding module. In the PLP type 1 modulation / variable PLP coding variable, the ModCod extraction modules, r307 and r307-1 shown in Fig. 64, can extract and provide parameters necessary for BICM decoding.

[0037] Fig. 42 is an example of a frame builder. The frame header inserter 401 can form a frame from the input symbol streams and can add a frame header at the front of each transmitted frame. The frame header can contain the following information:

* Number of connected channels (4 bits) * Protective period (2 bits) * PAPR (2 bits) * Pilot pattern (2 bits) * Digital System identification (16 bits) * Frame identification (16 bits) * Frame length (16 bits) ) - number of Orthogonal Frequency Division Multiplexing (OFDM) symbols per frame * Super frame length (16 bits) - number of frames per super frame * number of PLP frames (8 bits) * for each PLP

PLP identification (8 bits)

VP / 4263 / AGR

Channel bonding (4 bits)

Start of PLP (9 bits)

PLP type (2 bits) - common PLP or other type PLP data block (5 bits)

MC type (1 bit) - fixed / variable modulation and coding

When type MC == fixed modulation and coding

Type FEC (1 bit) - long or short LDPC Efficiency of coding (3 bits)

Modulation (3 bits) - up to 64K QAM end of the loop when;

Number of channel notches (2 bits) for each cutout

Start of cut (9 bits)

Cutout width (9 bits)

End of the loop for;

PLP width (9 bits) - maximum number of FEC blocks in the PLP

PLP time interleaving type (2 bits) end of the loop for;

* CRC-32 (32 bits) [0038] It is assumed that the channel bonding environment for L1 information transmitted in the frame header and data that correspond to each data slice is defined as a PLP. Therefore, information such as PLP identifier, channel bond identifier, and starting PLP address are required for each channel used in the join. In one embodiment of the present invention, it is suggested to transfer a ModCod data field in the FEC frame header if the PLP type supports variable modulation / variable coding and ModCod data field transfer in the frame header if the PLP type supports fixed modulation / fixed coding to reduce the overhead signal. Additionally, if there is a Notch band for each PLP,

[0039] Fig. 43 is an example of pilot pattern 5 (PP5) used in a channel bonding environment. As shown, in the case where the SP positions are coincidental with the pilot signal preamble positions, an irregular pilot structure may occur.

[0040] Fig. 43a shows an example of pilot inserter 404 as shown in Fig. 42. As shown in Fig. 43, when a single frequency band (e.g., 8 MHz) is used, the available bandwidth is 7 , 61 MHz, but if they are combined

VP / 4263 / AGR

In this case, the bandwidths can be removed, which can significantly increase the frequency efficiency. Fig. 43b is an example of the preamble inserting module 504 shown in Fig. 51 that is transmitted at the front of the frame and even when combining channels, the preamble has a repetition rate of 7.61 MHz, which is the L1 block bandwidth. It is a construction that takes into account the frequency bandwidth of the tuner, which performs initial channel scanning.

[0041] Pilot pattern patterns exist for both preamble and data symbols. For scatter symbols, scattered pilot (SP) patterns can be used. The pilot pattern 5 (PP5) and pilot pattern 7 (PP7) T2 can be good candidates for frequency-only interpolation. PP5 has x = 12, y = 4, z = 48 for GI = 1/64 and PP7 has x = 24, y = 4, z = 96 for GI = 1/128. Additional time interpolation is also available for better channel estimation. Pilot patterns for the preamble can cover all possible pilot positions to detect the initial channel. In addition, pilot positions for the preamble should be coincidental with SP positions and a single pilot pattern is required for both preamble and SP. Pilot signals for the preamble may also be used for time interpolation and each preamble may have an identical pilot pattern. These requirements are important for C2 detection during the search and necessary to estimate the frequency shift using the mixing sequence correlation. In a channel bonding environment, the coincidence in pilot positions should also be maintained during channel bonding because an irregular pilot structure can reduce interpolation performance.

In particular, if the distance z between the scattered pilot signals (SP) in the OFDM symbol is 48, and if the distance y between the scattered pilot signals SP corresponding to a specific carrier SP along the time axis is 4, the usable distance x after time interpolation will be 12 This is the case if the Guardian Interval (Guardian Interval) is 1/64. If the GI fraction is 1/128, then the parameters x = 24, y = 4, and z = 96 may be used. In the case where channel bonding is used, the positions SP may be set to coincidence with pilot positions for the preamble by generating discontinuous points in the dispersed pilot structure.

[0043] At this time, the pilot positions for the preamble may be coincident with all SP data symbol positions. When channel bonding is used, the data segment in which the service is transmitted may be determined regardless of the 8 MHz granularity of the bandwidth. However, in order to reduce the overhead for the address data slice, a transmission starting from the position SP and ending at the position SP can be selected.

[0044] When the receiver receives such scattered pilot signals SP, if necessary, the channel estimating module r501 shown in Fig. 62 may perform time interpolation to obtain pilot signals represented by dashed lines in Fig. 43 and may perform frequency interpolation. At this time, for discontinuous points whose compartments are indicated as 32 in Fig. 43, interpolations on the left and right can be performed separately or interpolations can be performed only on one side, and interpolation on the other side can be performed by using interpolated ones pilot position,

VP / 4263 / AGR

Whose interval is 12 as a reference point. At this time, the width of the data slice can vary within the 7.61 MHz range, and thus the receiver can minimize power consumption by performing channel estimation and decoding only the necessary subcarriers.

[0045] Fig. 44 is another example of a PP5 pattern used in a channel bonding environment or a SP construct to maintain a usable x-distance at level 12 to avoid the irregular SP structure shown in Fig. 43 when channel bonding is used. Fig. 44a illustrates a SP structure for data symbols, and Fig. 44b illustrates a SP structure for preamble symbols.

[0046] As shown, if the SP distance is consistently maintained in the case of channel bonding, then there will be no problems during frequency interpolation, but the pilot positions between data symbol and preamble may not be coincidental. In other words, this construction does not require additional channel estimation for the irregular SP structure, however, the SP positions used in channel bonding and pilot positions for the preamble become different for each channel.

[0047] Fig. 45 shows a new structure SP or PP5 ', which is intended to provide a solution to two problems mentioned above in a channel bonding environment. Specifically, the pilot distance of x = 16 can solve these problems. In order to protect the pilot signal density or maintain the same overhead, the PP5 'pattern may have x = 16, y = 3, z = 48 for GI = 1/64 and the PP7' pattern may have x = 16, y = 6, z = 96 for GI = 1/128. The ability to interpolate the same frequency can still be maintained. The pilot positions are shown in Fig. 45 for comparison with the PP5 construction.

[0048] Fig. 46 is an example of a new SP pattern or a PP5 'structure in a channel bonding environment. As shown in Fig. 46, regardless of whether a single channel or channel bonding is used, a useful pilot distance x = 16 can be provided. In addition, since it is possible to make the SPs coincidental with the pilot positions for the preamble, the deterioration of the channel estimation caused by the SP irregularity or the lack of the coincidence of the SP position can be avoided. In other words, there is no irregular SP position for the frequency interpolator and a coincidence between the preamble and the SP positions is provided.

[0049] Consequently, the proposed new SP standards may be advantageous in that a single SP pattern can be used for both a single and a combined channel; it is possible to cause the lack of an irregular pilot structure, thanks to which a good channel estimation is possible; both preamble positions and SP pilot positions may be maintained coincidental; the pilot signal density can be kept the same as for PP5 and PP7 standards, respectively; and the ability to interpolate the same frequency can also be maintained.

[0050] In addition, the preamble structure can meet requirements such as pilot positions for the preamble, which should cover all possible SP positions to detect the initial channel; the maximum number of carriers should be 3409 (7.61 MHz) for the initial search; exactly the same pilot pattern and mixing sequence should be

VP / 4263 / AGR

Used for detection of C2; and a detection-specific preamble such as P1 in T2 is not required.

[0051] In terms of the relationship with the frame structure, the granularity of the data slice position can be modified to 16 carriers rather than 12, so that a smaller overhead associated with addressing the position may occur and no other problems regarding the state of the data slice, zero gap state can be expected, e.t.c.

[0052] Therefore, in the estimator module r501 shown in Fig. 62, pilot signals in each preamble may be used when SP interpolation of data symbols is performed. Thus, channel detection and channel estimation at frame boundaries may be improved.

[0053] Now, regarding requirements related to the preamble and the pilot structure, there is consensus in that positions of preamble pilots and SPs should coincide regardless of channel bonding; the number of all carriers in L1 block should be divided by the distance of the pilot signal in order to avoid an irregular structure at the edge of the band; L1 blocks should be repeated in the frequency domain; and L1 blocks should always be capable of decoding in any position of the tuner window. Additional requirements may be such that pilot positions and standards should be repeated at a period of 8 MHz; the correct carrier frequency shift should be estimated without knowledge about channel bonding; and L1 decoding (re-ordering) is impossible before the frequency offset is equalized.

[0054] Fig. 47 is a relationship between data symbol and preamble when the preamble structures depicted in Fig. 52 and in Fig. 53 are used. L1 block can be repeated with a period of 6 MHz. In order to decode L1, both the frequency shift and the preamble shift pattern should be found. L1 decoding is not possible in any tuner position without channel bonding information and the receiver can not distinguish between the preamble shift value and the frequency offset.

[0055] Therefore, the receiver must obtain a channel bonding structure, in particular for the frame header removal element r401, shown in Fig. 63, to be able to decode the L1 signal. Because the expected preamble shift amount in the two vertically shadowed regions in Fig. 47 is known, the time / freq synchronizing module r505, shown in Fig. 62, can estimate the carrier frequency offset. Based on this estimation, the signal path L1 (r308-1 ~ r301-1) shown in Fig. 64 can decode L1.

[0056] Fig. 48 is a relationship between data symbol and preamble when the preamble structure as shown in Fig. 55 is used. L1 block can be repeated with a period of 8 MHz. In order to decode L1, you only need to find the frequency shift and the knowledge about channel bonding may not be required. The frequency offset can be easily estimated by using a known binary pseudo random sequence (PRBS). As illustrated in Fig. 48, the preamble and data symbols are aligned, and thus an additional search for synchronization may become unnecessary. Therefore, for the receiver, in particular for the frame header remover module r401 shown in Fig. 63,

VP / 4263 / AGR

Decoding the L1 signal. The time / freq synchronizing module r505, shown in Fig. 62, can estimate the carrier frequency offset relative to a peak position.

[0057] Fig. 49 is an example of a cable channel delay profile.

[0058] From the point of view of the pilot structure, the commonly used GI already with excess secures the propagation of the cable channel delay. In the worst case scenario, rebuilding the channel model can be an option. In order to repeat the pattern exactly every 8 MHz, the distance of the pilot signal should be a carrier divider of 3584 (z = 32 or 56). A pilot density of z = 32 may increase the pilot overhead, and therefore a density of z = 56 may be selected. A slightly smaller area of ​​delay coverage may not be relevant for the cable channel. For example, it can be 8 μs for PP5 'and 4 μs for PP7', compared to 9.3 μs (PP5) and 4.7 μs (PP7). Significant delays can be covered by both pilot patterns, even in the worst case scenario. For pilot signal preamble items,

[0059] If the delay path -40 dB can be ignored, the actual delay spread may start to be 2.5 μs, 1/64 GI = 7 μs, or 1/128 GI = 3.5 μs. This illustrates that the distance parameter of the pilot signal, z = 56 can have a good enough value. In addition, z = 56 may be a convenient value for constructing a pilot pattern, which allows the preamble structure shown in Fig. 48.

[0060] Fig. 50 is scattered pilot structure using z = 56 and z = 112, which is constructed at pilot inserter 404 shown in Fig. 42. Proposes are: PP5 '(x = 14, y = 4, z = 56) and PP7 '(x = 28, y = 4, z = 112). Edge carriers can be inserted for the final edge.

[0061] As illustrated in Fig. 50, pilots are aligned at 8 MHz from each edge of the band, each pilot position and pilot structure can be repeated every 8 MHz. In this way, this structure can support the preamble structure shown in Fig. 48. In addition, a common pilot structure between preamble and data symbols can be used. Therefore, the channel estimating module r501, shown in Fig. 62, can perform channel estimation using interpolation on preamble and data symbols, because no irregular pilot pattern can appear, regardless of the position of the window determined by the position data segment. . At this time, using only frequency interpolation, it is possible to sufficiently compensate the channel deformation resulting from the expansion of the delay. If time interpolation is performed additionally, a more accurate channel estimation can be performed.

[0062] Consequently, in the new proposed pilot pattern, pilot position and pattern can be repeated based on a period of 8 MHz. A single pilot pattern can be used for both preamble and data symbols. L1 decoding can always be possible without knowledge about channel bonding. In addition, the proposed pilot pattern may not affect the commonality with T2, because the same pilot strategy of the pilot scatter pattern may be used; T2 already uses 8 different pilot patterns; and modified pilot pattern can not

VP / 4263 / AGR

This may result in a significant increase in the complexity of the receiver. For a pilot mixing sequence, the PRBS can be 2047 (m-sequence); generating PRBS whose period is 3584 can be reset every 8 MHz; the pilot repeating frequency 56 may also be jointly first with period 2047; and you can expect that there will be no problems with PAPR.

[0063] Fig. 51 is an example of OFDM-based modulator. The input symbol streams can be converted to the time domain by the IFFT 501 module. If necessary, the Peak-to-Average Power Ratio (PAPR) can be reduced in the PAPR 502 decreasing module. In the PAPR methods, use the method of expanding an active constellation (ACE) or the tone reservation method. The GI inserting module 503 may copy the last part of the useful OFDM symbol to complete the guard period in the form of a cyclic prefix.

[0064] Preamble inserting module 504 can insert preamble at the front of each transmitted frame such that a receiver can detect a digital signal, a frame, and acquire time / freq offset acquisition. At this time, the preamble signal can perform signaling at the physical layer of information such as the FFT size (3 bits) and the size of the guard period (3 bits). The preamble inserting module 504 can be omitted if the modulator is intended for DVB-C2.

[0065] Fig. 52 is an example of a preamble channel bond structure generated in preamble inserting module 504 shown in Fig. 51. One complete L1 block should be &amp; quot; always decodable &amp; quot; at any arbitrary position of the 7.61 MHz tuning window and not L1 signal loss should occur irrespective of the position of the tuner window. As shown, the L1 blocks can be repeated in the frequency domain with a period of 6 MHz. The data symbol can be a combined channel for every 8 MHz. If, for L1 decoding, the receiver uses a tuner, such as the r603 tuner shown in Fig. 61, which uses a 7.61 MHz bandwidth, the frame header removal element r401, shown in Fig. 63, needs a rearrangement of the received cyclic shifted L1 block (Figure 53) to its original form. Such rearrangement is possible because the L1 block is repeated for each 6 MHz block. Fig. 53a can be rearranged to Fig. 53b.

[0066] Fig. 54 is a process for designing a more optimized preamble. The preamble structure shown in Fig. 52 uses only 6 MHz with a total tuner bandwidth of 7.61 MHz for L1 decoding. In terms of spectral efficiency, the tuner frequency bandwidth of 7.61 MHz is not fully utilized. Therefore, further optimization regarding spectral efficiency can be performed.

[0067] Fig. 55 is another example of a preamble structure or preamble symbol structure for achieving full spectrum efficiency generated in the frame header inserter 401 shown in Fig. 42. Like the data symbol, L1 blocks can be repeated in the frequency domain. with a period of 8 MHz. One complete L1 block is still "always decodable" in any arbitrary position of the 7.61 MHz tuning window. After performing the tuning, the given 7.61 MHz frequencies can be considered as virtually punctured code. With exactly the same frequency bandwidth for both preamble and data symbols and exactly the same pilot structure for both preamble and data symbols, spectral efficiency can be maximized. Other characteristics

VP / 4263 / AGR

In the absence of a data slice, they can be maintained unchanged. In other words, the frequency bandwidth of the preamble symbols may be identical to the data bandwidth of the data symbols or, as illustrated in Fig. 57, the bandwidth of the preamble symbols may be the bandwidth of the tuner (in this case 7.61 MHz). The frequency bandwidth of the tuner can be defined as the frequency bandwidth that corresponds to the number of all active carriers when a single channel is used. That is, the frequency bandwidth of the preamble symbol may correspond to the number of all active carriers (in this case 7.61 MHz).

[0068] Fig. 56 shows a virtually punctured code. Data 7.61 MHz between 8 MHz L1 block can be considered to be coded as a punctured code. When the r603 tuner shown in Fig. 61 uses a bandwidth of 7.61 MHz for L1 decoding, the frame header removal element r401, shown in Fig. 63, needs to rearrange the received cyclic shifted L1 block to its original form, like illustrated in Fig. 56. At this time, L1 decoding is performed using the entire tuner bandwidth. After the L1 block is rearranged, the spectrum of the rearranged L1 block may include a hollow region within the spectrum, as shown in Fig. 56 in the upper right side, since the L1 block primary size has 8 MHz bandwidth.

[0069] After zero-filled null space, or after performing deinterleaving in the symbol domain by frequency deinterleaver r403, shown in Fig. 63, or by de-interleaver in symbol domain r308-1, shown in Fig. 64, or after performing bit-deinterleaving in the symbol demapper r3061, the bit multiplexer r305-1, and the internal de-interleaver r304-1 shown in Fig. 64, the block may be in a form that appears to be pierced as shown in Figure 56 in the bottom right part.

[0070] This L1 block can be decoded in the punctured / shortened decode module r303-1, shown in Fig. 64. By using these preamble structure, the entire tuner bandwidth can be used, and thus an increase in spectral efficiency and efficiency can be obtained. coding. In addition, identical frequency bandwidth and pilot structure can be used for preamble and data symbols.

[0071] Furthermore, if the preamble bandwidth or the frequency band of the preamble symbols is set like the tuner bandwidth as illustrated in Fig. 58, (in this example it is 7.61 MHz), after regrouping, complete L1 block does not even use piercing. In other words, for a frame containing preamble symbols, in which the preamble symbols comprise at least one block layer 1 (L1), it can be said that the L1 block comprises 3408 active subcarriers and that 3408 active subcarriers correspond to 7.61 MHz from the radio frequency band ( RF; Radio Frequency) of 8 MHz.

[0072] In this way, the spectral efficiency and L1 decoding performance can be maximized. In other words, at the receiver, the decoding can be performed in the punctured / shortened decode module r303-1, shown in Fig. 64, after only de-interleaving in the symbol domain.

VP / 4263 / AGR

[0073] Consequently, the proposed new preamble structure can be advantageous in that it is fully compatible with previously used preamble except that the bandwidth is different; L1 blocks are repeated with a period of 8 MHz; the L1 block can always be suitable for decoding, regardless of the position of the tuner window; the full tuner frequency bandwidth can be used for L1 decoding; maximum spectral efficiency can guarantee higher coding efficiency; an incomplete L1 block can be considered to be coded in a punctured manner; a simple and the same pilot structure can be used for both preamble and data; and the same frequency bandwidth can be used for both preamble and data.

[0074] Fig. 59 is an example of an analog processor. The DAC 601 module can convert an input digital signal into an analog signal. After up conversion of the transmit frequency bandwidth (602) and analogue filtering (603), the signal may be transmitted.

[0075] Fig. 60 is an example of a digital reception system. The received signal is converted into a digital signal in the analog processing module r105. The demodulator r104 can convert a signal to data in the frequency domain. The r103 frame parser may remove pilot signals and headers and allow selection of service information that must be decoded. The BICM demodulator r102 can correct errors in the transmission channel. The r101 output processor can restore the originally transmitted service stream and timing information.

[0076] Fig. 61 is an example of an analog processor used at a receiver. The tuner / AGC module r603 can select the desired bandwidth from the received signal. The down-change frequency module r602 can restore the basic band. The r601 ADC module can convert an analog signal into a digital signal.

[0077] Fig. 62 is an example of demodulator. The frame detection module r506 can detect the preamble, check for a corresponding digital signal, and detect the beginning of the frame. The time / frequency synchronization module r505 can perform synchronization in time and frequency domains. At this time, the correlation of the guard interval may be used to synchronize in the time domain. For synchronization in the frequency domain, a correlation may be used, or the offset may be estimated based on the subcarrier information that is transmitted in the frequency domain. The preamble remover module r504 can remove the preamble from the front portion of the detected frame. The GI r503 removal module can remove the protection period. The FFT module r501 can convert the time domain signal into a frequency domain signal. The channel estimation / equalization module r501 can compensate for errors by estimating the deformation in the transmission channel using the pilot signal symbol. The preamble remover r504 may be omitted if the demodulator is adapted for DVB-C2.

[0078] Fig. 63 is an example of a frame parser. The r404 pilot removal module can remove the pilot symbol. Frequency deinterleaving module r403 can perform frequency unraveling in the frequency domain. The element combining the OFDM symbol r402 can restore the data frame from symbol streams transferred in the form of OFDM symbols. Removal module

VP / 4263 / AGR

The frame header r401 may extract physical layer signals from the header of each transmitted frame and delete the header. The information extracted can be used as parameters for subsequent processes carried out in the receiver.

[0079] Fig. 64 is an example of a BICM demodulator. Fig. 64a illustrates the data path, and Fig. 64b shows the L1 signaling path. The de-interleaver in the field of symbols r308 may perform deinterleaving in the symbol domain. The ModCod r307 extractor can extract ModCod parameters from the front of each BB frame and cause that these parameters become available for subsequent adaptive / variable demodulation processes and decoding processes. The symbol demapper r306 may mirror the input symbol streams inversely into bitstreams with a log-likelihood ratio (Log-Likelyhood Ratio, LLR). The output streams of the LLRs can be calculated using the constellation used in the transmitter mapper 306 as a reference point. In this point, when the above-mentioned MQAM or NU-MQAM is used, by calculating both the I axis and the Q axis when calculating the bit closest to the MSB and by calculating the axis I or the Q axis, the effective symbol demapper can be obtained when calculating the remaining bits. This method can be used, for example, in an approximate LLR, in an accurate LLR, or in a Hard decision.

When an optimized constellation according to the constellation capacity and the code rate of the error correction code encoding in the transmitter mapper 306 is used, the symbol demapper mice receiver r306 can receive the constellation using the coding efficiency and constellation capacity information transmitted from the transmitter. The receiver multiplexer r305 may perform a reverse function than the transmitter bit demoder 305. The internal de-interleaver r304 and the external de-interleaver r302 of the receiver can perform inverse functions than the inner interleaver 304 and the external interleaver 302, respectively, to obtain a bit stream in its original sequence. The external de-interleaver r302 may be omitted in the case of

[0081] The inner decoder r303 and the outer decoder r301 of the receiver may perform corresponding decoding processes to the inner coder module 303 and outer transmitter encoder 301, respectively, to correct errors in the transmission channel. Similar processes to be performed on a data path can be performed on the L1 signal path, but with different parameters (r308-1 ~ r301-1). At this point, as explained in the introductory part, the shortened / punctured coding module r303-1 can be used to decode the L1 signal.

[0082] Fig. 65 is an example of LDPC decoding using shortening / puncturing. The r301a demultiplexer can separately output the information part and the parity part of the systematic code from the input bitstreams. For the information part, padding (r302a) can be performed according to the number of input bit streams of the LDPC decoder, while for the parity portion, the input bit streams for (r303a) LDPC decoder can be generated by piercing the inverse punctured part. decoding

VP / 4263 / AGR

EP 2 323 331 B1

LDPC (r304a) can be performed on the generated bit streams, zeros in the information part can be deleted and outputted (r305a).

[0083] Fig. 66 is an example of output processor. The BB descrambler r209 can restore encrypted (209) bit streams at the transmitter. The r208 splitter can restore the BB frames corresponding to the plurality of PLPs that are multiplexed and transmitted from the transmitter according to the PLP path. For each PLP path, the BB header erasing element r207-1 n can delete the header that is transmitted in front of the BB frame. The CRC decoder r206-1? N can perform CRC decoding and cause the reliable BB frames to become available for selection. Null packet inserting modules r205-1-n can restore null packets that have been removed for higher transmission efficiency to their original location. The delay recovery modules r204-1~n can restore the delay that exists between each PLP path.

[0084] The r203-1-n clocks recovering timer may restore the primary timing of the service stream from timing information transmitted from the input stream synchronizing modules 203-1 ~ n. The r202-1 ~ n output interface modules can restore data in the TS / GS packet from the input bitstreams that are segmented in the BB frame. The r201-1 ~ n output postprocess modules can restore multiple TS / GS streams into a complete TS / GS stream, if necessary. The shaded blocks shown in Fig. 66 represent modules that can be used when a single PLP is processed at one time, while the remaining blocks represent modules that can be used when multiple PLP frames are processed at the same time.

[0085] Pilot pattern patterns for the preamble have been carefully constructed to avoid increasing PAPR, and therefore, it is necessary to consider whether the L1 block repetition rate can increase PAPR. The number of information bits of the L1 block varies dynamically according to channel joining, number of PLPs, etc. In particular, it is necessary to consider the following issues: fixed L1 block size, which may introduce unnecessary overhead; the L1 signal block should be more protected than the data symbols; and time interleaving of the L1 block, which can improve the channel weakness resistance such as the shortage of impulsive interference.

[0086] For the L1 block repetition rate of 8 MHz, as shown in Fig. 67, the total spectral efficiency (26.8% increase in BW) is demonstrated using virtual piercing, but the PAPR ratio can be increased because the L1 block bandwidth is is the same as the frequency band of the data symbols. For a repetition rate of 8 MHz, 4K-FFT DVB-T2 frequency interleaving may be used for commonality and the same pattern may be repeated in the 8 MHz period after interleaving.

[0087] For a L1 block repetition rate of 6 MHz, as shown in Fig. 68, reduced spectral efficiency may be exhibited without using virtual punching. A similar problem regarding the PAPR ratio as in the case of 8 MHz can occur because the bandwidth of the L1 block and the data symbol are shared by LCM = 24 MHz. For a repetition rate of 6 MHz, 4K-FFT DVB-T2 frequency interleaving may be used for commonality and the same pattern may be repeated with a period of 24 MHz after interleaving.

VP / 4263 / AGR

[0088] Fig. 69 is a new L1 block repetition rate of 7.61 MHz or total tuner bandwidth. Total spectral efficiency (26.8% increase in BW) can be obtained without using virtual piercing. In this case, there can be no problem with PAPR, because the bandwidth of L1 block and data symbol are shared by LCM = 1704 MHz. For a repetition rate of 7.61 MHz, 4K-FFT DVB-T2 frequency interleaving can be used for commonality and the same pattern can be repeated with a period of about 1704 MHz after interleaving.

[0089] Fig. 70 is an example of L1 signaling block that is transmitted in the frame header. Any information in the L1 signal block can be transmitted to the receiver and can be used as a decoding parameter. Specifically, the information may be used in the L1 signaling path shown in Fig. 64 and the PLPs may be transmitted in each data slice. In this way, an increased resistance for each PLP can be obtained.

[0090] Fig. 72 is an example of symbol interleaver 308-1 as shown in L1 signal path in Fig. 37, which may also serve as an example of a corresponding symbol deinterleaver r308-1, as shown in the L1 signaling path in FIG. 64. Blocks with sloping lines represent L1 blocks, and solid blocks represent data carriers. L1 blocks may be transmitted not only within a single preamble, but may also be transmitted within multiple OFDM blocks. Depending on the size of the L1 block, the size of the interleaving block may vary. In other words, the num_L1_sym and L1_span parameters may be different from each other. In order to minimize unnecessary overhead, data may be transmitted within the rest of the OFDM symbol carriers in which the L1 block is transmitted. In this point, total spectral efficiency can be guaranteed because the repetitive L1 block cycle is still the total tuner bandwidth. In Fig. 72, numbers in blocks with slanted lines represent the order of bits within a single LDPC block.

[0091] Consequently, when the bits are written in the interleaved memory in the row direction according to the symbol index as shown in Fig. 72 and are read in the column direction according to the carrier index, a block interleaving effect can be obtained. In other words, one LDPC block can be flown in the time domain and in the frequency domain, and can then be transmitted. The num_L1_sym parameter can have a predetermined value, e.g., a number between 2-4 can be set as the number of OFDM symbols. At this point, to increase the granularity of the L1 block size, a punctured / shortened LDPC code having a minimum length of the code word can be used to protect the L1 block.

[0092] Fig. 73 is an example of transmitting L1 blocks. Fig. 73 is Fig. 72 in the field of the frame. As shown in Fig. 73a, the L1 blocks may occupy the total tuner bandwidth or as shown in Fig. 73b, the L1 blocks may occupy a part, and the rest of the carriers may be used as a data carrier. In any case, it can be seen that the repetition rate of the L1 block can be identical to the total tuner bandwidth. Additional property amenitiesthat, for OFDM symbols that use the L1 signal block including the preamble, only symbol interleaving can be performed that does not allow data to be transmitted in these OFDM symbols. Consequently, in the case of the OFDM symbol used for the L1 signaling block, the receiver can L1 decode by performing deinterleaving without decoding the data. At this point, block L1

VP / 4263 / AGR

It may transmit the L1 signaling block of the current frame or the L1 signaling block of the next frame. On the receiver side, decoded L1 parameters from the L1 signaling decoding path shown in Fig. 64 may be applied to the decoding process for the data path from the frame parser of the next frame.

[0093] In brief, at a transmitter, interleaving blocks of L1 can be performed by storing blocks in a memory in the row direction and reading the stored blocks from the memory in the direction determined by the columns. At the receiver, deinterleaving of blocks from the L1 area can be performed by writing the blocks in the memory in the direction determined by the columns and reading the stored blocks from the memory in the direction determined by the rows. The directions of reading and writing in the transmitter and receiver can be swapped.

[0094] After performing the simulation, using assumptions such as CR = 1/2 to protect the L1 block and to enable the T2 mutualization; symbol mapping using 16-QAM modulation; pilot density of 6 in the preamble; and the short LDPC number, which implies the required amount of piercing / shortening, results or conclusions can be obtained, such as that for transmitting the L1 block the preamble itself may not be sufficient; the number of OFDM symbols depends on the L1 block size; the shortest codeword LDPC (e.g., 192-bit information) from shortened / punctured codes can be used to obtain flexibility and fine granularity; and filling may be added, if required, with a slight overhead. This result is summarized in Fig. 71.

[0095] Consequently, for the L1 block repetition rate, using the full tuner bandwidth without using virtual punching can be a good solution and still no PAPR problem can arise with full spectral efficiency. For the L1 signal block, the effective signal structure can allow the maximum configuration in the 8-channel connection environment, 32 notches, 256 data segments, and 256 PLPs. For the structure of the L1 block, the L1 signal flexible block can be implemented according to the size of the L1 block. Time interleaving can be performed to obtain better resistance to T2 commonality. A lower overhead can allow data to be transmitted in the preamble.

[0096] Block interleaving of L1 block can be performed for better robustness. The interleaving can be performed using a pre-defined number of L1 symbols (num_L1_sym) and the number of carriers joined by L1 as parameter (L1_span). The same technique is used for interleaving the preamble P2 in DVB-T2.

[0097] A L1 block of variable size may also be used. The magnitude can be adjusted to the number of bits of the L1 signaling block, resulting in a reduction of the overhead. The total spectral efficiency can be obtained without any problems with PAPR. Repetition lower than 7.61 MHz may mean that more redundancy can be transmitted, but this is unused. In this case, there can be no problem with PAPR due to 7.61 MHz repetition rate for L1 block.

[0098] Fig. 74 is another example of L1 signaling transmitted within a frame header. This Fig. 74 is different from Fig. 70 in that the L1_span field having 12 bits has been divided into two fields. In other words, the L1_span field has been divided into an L1_column field having 9 bits and an L1_row field having 3 bits. The L1_column field represents the subscript of the carrier that includes block L1. Because the data segment starts and ends every 12 media, which is the pilot density, 12 bits of overhead can be reduced by 3 bits to obtain 9 bits.

VP / 4263 / AGR

[0099] The L1_row field represents the number of OFDM symbols that comprise the L1 block when using time interleaving. Consequently, time interleaving can be performed inside the L1_columns fields multiplied by the L1_rows fields. Alternatively, L1 blocks with a total size may be transmitted so that the L1_span field shown in Fig. 70 can be used when time interleaving is not performed. In this case, the L1 block size is 11.776 x 2 bits in this example, and therefore 15 bits is sufficient. Consequently, the L1_span field can be created by 15 bits.

[0100] Fig. 75 is an example of frequency or time interleaving / deinterleaving. Fig. 75 shows a part of the entire transmitted frame. Fig. 75 also shows the combining of a plurality of 8 MHz bandwidths. The frame may consist of a preamble that transmits L1 blocks and a data symbol that transmits data. Different types of data symbols represent data segments for different services. As shown in Fig. 75, the preamble transmits L1 blocks for every 7.61 MHz.

[0101] In the case of a preamble, interleaving in the frequency or time domain is performed within L1 blocks and is not performed between the L1 blocks. That is, for the preamble, it can be said that interleaving is carried out at L1 block level. This enables L1 blocks to be decoded by sending L1 blocks within the tuner window bandwidth, even when the tuner window has moved to a random location within the channel bonding system.

[0102] For decoding a data symbol at a random tuner window bandwidth, interleaving between data slices should not occur. This means that it can be said that for data segments, interleaving is performed at the data slice level. Consequently, frequency interleaving and time interleaving should be performed within the data slice. Therefore, the symbol interleaver 308 in the data path of the BICM of the transmitter, as shown in Fig. 37, can perform symbol interleaving for each data slice. The symbol interleaver 308-1 in the L1 signaling path can perform symbol interleaving for each L1 block.

[0103] Frequency interleaver 403, illustrated in Fig. 42, must perform interleaving on preamble and data symbols separately. Specifically, in the case of a preamble, frequency interleaving can be performed for each L1 block, and for a data symbol, frequency interleaving can be performed for each data slice. At this point, time interleaving in the data path or L1 signal path may not be performed considering the low latency mode.

[0104] Using the proposed methods and devices, among others advantages it is possible to implement an efficient digital transmitter, receiver and structure of physical layer signaling.

[0105] By transmitting ModCod information in each BB frame header that is necessary for ACM / VCM and transmitting the rest of the physical layer signaling in the frame header, signaling overhead can be minimized.

VP / 4263 / AGR

[0106] Modified QAM may be implemented for a more energy efficient transmission or a more noise-robust digital broadcasting system. The system may include a transmitter and a receiver for each example disclosed and combinations thereof.

[0107] An improved uneven QAM may be implemented for a more energy-efficient broadcast or a more noise-robust digital broadcasting system. Also described is the use of the error correction code with high encoding efficiency in NU-MQAM and MQAM. For each disclosed example, the system may include a transmitter and a receiver and combinations thereof. [0108] The proposed L1 signaling method can reduce overhead by 3 ~ 4% by minimizing signaling overhead during channel bonding.

[0109] It will be apparent to those skilled in the art that the present invention may be subjected to various modifications and variations, without departing from the scope of the invention.

VP / 4263 / AGR

EP 2 323 331 B1