KR20230002144A - Apparatus for time de-interleaving and method using the same - Google Patents

Abstract

Disclosed are a time interleaving device corresponding to a hybrid time interleaving mode and a time interleaving method thereof. According to an embodiment of the present invention, the time interleaving device comprises: a twisted block interleaver that performs intra-subframe interleaving corresponding to a time interleaving block; and a convolutional delay line for performing inter-subframe interleaving by using an output of the twisted block interleaver.

Description

Time deinterleaving device and time deinterleaving method using the same

The present invention relates to broadcast signal transmission/reception technology used in a broadcasting system, and more particularly to a broadcast signal transmission/reception system using time interleaving.

BICM (Bit-Interleaved Coded Modulation) is a bandwidth-efficient transmission technology that includes an error-correction coder, a bit-by-bit interleaver, and a high-order modulator. It is a combined form.

BICM can provide excellent performance with a simple structure by using a low-density parity check (LDPC) coder or a turbo coder as an error correction coder. In addition, BICM provides a high level of flexibility because it can select a modulation order, error correction code length and code rate in various ways. Because of these advantages, BICM is not only used in broadcasting standards such as DVB-T2 or DVB-NGH, but is highly likely to be used in other next-generation broadcasting systems as well.

When one service is transmitted using multiple physical layer pipes, a decoding process may vary greatly depending on the time interleaving mode corresponding to the physical layer pipe.

Therefore, it is very important to determine the time interleaving mode of the physical layer pipe and parameters related to time interleaving in consideration of decoding complexity and the like.

When inter-subframe interleaving is performed, decoding complexity greatly increases. That is, when inter-subframe interleaving is performed, it is very important to reduce the increase in decoding complexity.

An object of the present invention is to reduce the decoding complexity of a broadcast system by appropriately setting the position of a virtual cell when inter-subframe interleaving is performed.

Also, an object of the present invention is to operate a convolutional delay line for hybrid-time interleaving using a new virtual cell regardless of a virtual cell included in a twisted block interleaver, so that a broadcast system can operate efficiently.

In addition, an object of the present invention is to allow a broadcasting system to operate efficiently without increasing decoding complexity by operating by appropriately utilizing an initial value of a FIFO register included in a convolutional delay line.

To achieve the above object, a time interleaving apparatus according to the present invention includes a twisted block interleaver performing intra-subframe interleaving corresponding to time interleaving blocks; and a convolutional delay line for performing inter-subframe interleaving using an output of the twisted block interleaver.

In this case, the twisted block interleaver may perform the intra-subframe inleaving by performing a column-wise write operation and a diagonal-wise read operation.

At this time, the convolutional delay line can read only data cells excluding virtual cells from the twisted block interleaver.

In this case, the convolutional delay line may store new virtual cells after each row of the data cells is written from the twisted block interleaver and before the switches move to the next branch.

At this time, the new virtual cells are calculated by subtracting the number of FEC blocks (N _{FEC_TI} ) in the TI block of the interleaving frame from the maximum value (N _{FEC_TI_MAX} ) of the number of FEC blocks in the TI block of the interleaving frame in each branch. can match

In this case, new virtual cells may not be output as an output of the time interleaving apparatus.

In this case, the convolutional delay line may include branches corresponding to the time interleaving unit (N _IU ), and the remaining branches excluding the first branch among the branches may include one or more FIFO registers.

At this time, the convolutional delay line may output only some of the initial values stored in the FIFO register.

In this case, some of the initial values may correspond to one initialization cell for each of the remaining branches.

In this case, the TI apparatus may further include a cell interleaver interleaving cells within the TI block.

Also, a time interleaving method according to an embodiment of the present invention includes performing twisted block interleaving corresponding to intra-subframe interleaving; and performing inter-subframe interleaving using an output of the twisted block interleaving.

In this case, the twisted block interleaving may be performed using a column-wise write operation and a diagonal-wise read operation.

In this case, the step of performing inter-subframe interleaving may be performed using a convolutional delay line, and the convolutional delay line reads only data cells excluding virtual cells corresponding to the twisted block interleaving. can

In this case, the convolutional delay line may store new virtual cells after each row of the data cells is written from the output of the twisted block interleaving and before the switches move to the next branch.

At this time, the new virtual cells are calculated by subtracting the number of FEC blocks (N _{FEC_TI} ) in the TI block of the interleaving frame from the maximum value (N _{FEC_TI_MAX} ) of the number of FEC blocks in the TI block of the interleaving frame in each branch. can match

In this case, new virtual cells may not be output as an output of the inter-subframe interleaving.

In this case, the convolutional delay line may include branches corresponding to the time interleaving unit (N _IU ), and the remaining branches excluding the first branch among the branches may include one or more FIFO registers.

At this time, the convolutional delay line may output only some of the initial values stored in the FIFO register.

In this case, some of the initial values may correspond to one initialization cell for each of the remaining branches.

In this case, the TI method may further include performing cell interleaving corresponding to cells within the TI block.

According to the present invention, when inter-subframe interleaving is performed, decoding complexity of a broadcast system can be reduced by appropriately setting the position of a virtual cell.

In addition, since the convolutional delay line for hybrid-time interleaving operates using a new virtual cell regardless of the virtual cell included in the twisted block interleaver, the broadcasting system can operate more efficiently.

In addition, the present invention operates by appropriately utilizing the initial value of the FIFO levisor included in the convolutional delay line, so that the broadcasting system can operate efficiently without increasing decoding complexity.

1 is a block diagram showing a system for transmitting/receiving broadcast signals according to an embodiment of the present invention.
FIG. 2 is an operational flowchart illustrating a method for transmitting/receiving broadcast signals according to an embodiment of the present invention. FIG. 1 is a block diagram illustrating a system for transmitting/receiving broadcast signals according to an embodiment of the present invention.
2 is an operational flowchart illustrating a broadcast signal transmission/reception method according to an embodiment of the present invention.
FIG. 3 is a block diagram illustrating an example of an apparatus for generating a broadcast signal frame shown in FIG. 1 .
4 is a diagram illustrating an example of a broadcast signal frame structure.
FIG. 5 is a diagram illustrating an example of a process in which the broadcast signal frame shown in FIG. 4 is received.
FIG. 6 is a diagram illustrating another example of a process in which the broadcast signal frame shown in FIG. 4 is received.
FIG. 7 is a block diagram illustrating another example of the apparatus for generating a broadcast signal frame shown in FIG. 1 .
FIG. 8 is a block diagram illustrating an example of the signal demultiplexing device shown in FIG. 1 .
FIG. 9 is a block diagram illustrating an example of a core layer BICM decoder and an enhanced layer symbol extractor shown in FIG. 8 .
FIG. 10 is a block diagram illustrating another example of a core layer BICM decoder and an enhanced layer symbol extractor shown in FIG. 8 .
FIG. 11 is a block diagram illustrating another example of a core layer BICM decoder and an enhanced layer symbol extractor shown in FIG. 8 .
FIG. 12 is a block diagram illustrating another example of the signal demultiplexing device shown in FIG. 1 .
13 is a diagram illustrating power increase due to a combination of a core layer signal and an enhanced layer signal.
14 is an operational flowchart illustrating a method for generating a broadcast signal frame according to an embodiment of the present invention.
15 is a diagram showing a superframe structure including a broadcast signal frame according to an embodiment of the present invention.
16 is a diagram illustrating an example of an LDM frame to which LDM using two layers and a multiple-physical layer pipe are applied.
17 is a diagram illustrating another example of an LDM frame to which an LDM using two layers and a multiple-physical layer pipe are applied.
18 is a diagram showing an example of utilization of an LDM frame to which LDM using two layers and a multiple-physical layer pipe are applied.
19 is a diagram illustrating another utilization example of an LDM frame to which LDM using two layers and a multiple-physical layer pipe are applied.
20 is a diagram illustrating an example of a case in which a convolutional time interleaver is used.
21 is a diagram illustrating another example of a case in which a convolutional time interleaver is used.
22 is a diagram illustrating an example of a case where a hybrid time interleaver is used.
FIG. 23 is a diagram illustrating a time interleaver group in the example shown in FIG. 22 .
24 to 26 are diagrams illustrating a process of calculating the size of an incomplete FEC block in the example shown in FIG. 23 .
27 is a diagram for explaining the number of bits required for L1D_plp_fec_block_start when L1D_plp_TI_mode = “00”.
28 and 29 are diagrams for explaining the number of bits required for L1D_plp_CTI_fec_block_start when L1D_plp_TI_mode = “01”.
30 is a diagram illustrating insertion of enhanced layer dummy values when an HTI mode is used together with layered division multiplexing.
31 is a diagram illustrating an example of a shift register used to generate dummy values according to an embodiment of the present invention.
32 is a diagram illustrating types of time interleaving modes.
33 is a diagram illustrating a case in which intra-subframe interleaving and inter-subframe interleaving are simultaneously used.
34 is a diagram illustrating subframes when intra-subframe interleaving and inter-subframe interleaving are simultaneously used.
35 is a diagram illustrating a case in which different TI units are simultaneously used.
36 is a diagram illustrating subframes when the same TI unit is simultaneously used.
37 is a diagram illustrating a case where one complete transport product is composed of a plurality of physical layer pipes.
FIG. 38 is a block diagram illustrating an example of the time interleaver shown in FIG. 3 or FIG. 7 .
FIG. 39 is a diagram illustrating a write operation of the twisted block interleaver shown in FIG. 38;
40 is a diagram illustrating a read operation of the twisted block interleaver shown in FIG. 38;
FIG. 41 is a block diagram illustrating an example of the convolutional delay line shown in FIG. 38 .
42 is a diagram illustrating an example of an operation of the twisted block interleaver shown in FIG. 38;
43 is a diagram illustrating an example of an operation of the convolutional delay line shown in FIG. 38;
FIG. 44 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 43 .
FIG. 45 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 42 .
46 is a diagram illustrating another example of an operation of the twisted block interleaver shown in FIG. 38;
47 is a diagram illustrating another example of an operation of the convolutional delay line shown in FIG. 38;
FIG. 48 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 47 .
FIG. 49 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 46 .
50 is a diagram showing initial values of FIFO registers included in a convolutional delay line.
51 is a block diagram illustrating an example of the time deinterleaver shown in FIG. 8 or 12;
52 is an operational flowchart illustrating a time interleaving method according to an embodiment of the present invention.

The present invention will be described in detail with reference to the accompanying drawings. Here, repeated descriptions, well-known functions that may unnecessarily obscure the subject matter of the present invention, and detailed descriptions of configurations are omitted. Embodiments of the present invention are provided to more completely explain the present invention to those skilled in the art. Accordingly, the shapes and sizes of elements in the drawings may be exaggerated for clarity.

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

1 is a block diagram showing a system for transmitting/receiving broadcast signals according to an embodiment of the present invention.

Referring to FIG. 1 , a broadcast signal transmission/reception system according to an embodiment of the present invention includes a broadcast signal transmission device 110, a radio channel 120, and a broadcast signal reception device 130.

The broadcast signal transmission apparatus 110 includes a broadcast signal frame generator 111 and an OFDM transmitter 113 generating broadcast signal frames by multiplexing core layer data and enhanced layer data.

The apparatus 111 for generating a broadcast signal frame combines a core layer signal corresponding to core layer data and an enhanced layer signal corresponding to enhanced layer data, and converts power of the combined signal to power corresponding to the core layer signal. After power normalization is performed, time interleaving is performed to generate a time interleaved signal. In this case, the core layer signal and the enhanced layer signal may be combined at different power levels. In this case, time interleaving may be applied to both the core layer signal and the enhanced layer signal. In this case, the broadcasting signal frame generating device 111 may generate a broadcasting signal frame including a bootstrap and a preamble using the time-interleaved signal. In this case, the broadcast signal frame may be an ATSC 3.0 frame.

According to an embodiment, the broadcast signal frame generating apparatus 111 may time-interleave one layer signal (BICM output signal) and generate a broadcast signal frame without combining two layer signals.

In this case, the preamble may signal a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes.

In this case, the physical layer pipes may include one enhanced layer physical layer pipe and a plurality of core layer physical layer pipes layered division multiplexed to the one enhanced layer physical layer pipe.

In this case, the time interleaving mode corresponding to the enhanced layer physical layer pipe may be the same as the time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipe is layered division multiplexed.

In this case, time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed may all be no time interleaving modes or all hybrid time interleaving modes.

In this case, when the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are all hybrid time interleaving modes, the core layer physical layer pipes all use the intra-subframe interleaving mode. can be used

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are no time interleaving modes, each of the core layer physical layer pipes is in each subframe. It may consist of an integer number of FEC blocks.

At this time, the subframe may be created by first filling all available data cells of the subframe with dummy modulation values, and then overwriting the actual physical layer pipe data.

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

In this case, the physical layer pipes include a plurality of core layer physical layer pipes corresponding to one complete delivered product, and the core layer physical layer pipes may not be layered division multiplexed.

In this case, each of the core layer and physical layer pipes may use either a time interleaving mode of a no time interleaving mode or a hybrid time interleaving mode, and may not use a convolutional time interleaving mode.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, all of the core layer physical layer pipes use intra-subframe interleaving mode or all use inter-subframe interleaving mode. can

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, the core layer physical layer pipes may all have the same L1D_plp_HTI_inter_subframe value.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes and all of the core layer physical layer pipes use the inter-subframe interleaving mode, the core layer physical layer pipes have the same time interleaving mode. An interleaving unit (N _IU ) may be used.

In this case, when at least one of the time interleaving modes corresponding to the core layer physical layer pipes is the no time interleaving mode, all core layer pipes configured in the hybrid time interleaving mode among the core layer pipes are in the intra-subframe interleaving mode. can be used.

At this time, one complete transmission product corresponds to one or more subframes, in which all available data cells of the subframe are first filled with dummy modulation values, and then the real physical layer pipe data is overwritten. can be created

In this case, time interleaving uses one of time interleaver groups, and a boundary between the time interleaver groups is between physical layer pipes (PLPs) of the core layer corresponding to the core layer signal. may be the boundary of That is, one of the boundaries between physical layer pipes of the core layer may be a boundary between time interleaver groups.

In this case, enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

At this time, dummy values are applied after the actual data cells of the last enhanced PLP in the PLP group so that the total number of enhanced layer cells in the PLP group is equal to the total number of core layer cells in the PLP group. of the last Enhanced PLP) can be inserted.

In this case, dummy values may not be inserted into core layer data.

In this case, dummy values may be inserted after completing the core layer BICM and the enhanced layer BICM and before combining the core layer signal and the enhanced layer signal.

In this case, the dummy values may correspond to a preset scrambling sequence.

At this time, the scrambling sequence may be modulated using the constellation mapping used for the last enhanced PLP.

In this case, dummy values may have the same power as the last enhanced PLP.

In this case, the scrambling sequence may be generated using a 16-bit shift register corresponding to a preset generator polynomial.

In this case, the scrambling sequence may be generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ .

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), and the sixth bit of the shift register initialized to 0xF180 (1111 0001 1000 0000). Using 8 bits generated using output (x ¹¹ ), 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x) can be created by

The OFDM transmitter 113 transmits the generated broadcast signal frame through the antenna 117 using the OFDM communication method, and the transmitted OFDM signal is sent to the antenna 137 of the broadcast signal receiving device 130 through the radio channel 120. to be received through

The broadcast signal receiving device 130 includes an OFDM receiver 133 and a signal demultiplexing device 131. When a signal transmitted through the radio channel 120 is received through the antenna 137, the OFDM receiver 133 receives the OFDM signal through synchronization, channel estimation, and equalization processes. do.

At this time, the OFDM receiver 133 detects and demodulates a bootstrap from the OFDM signal, demodulates a preamble using information included in the bootstrap, and generates a super-imposed payload using information included in the preamble. You can also demodulate.

The signal demultiplexing device 131 first restores core layer data from the signal (super-imposed payload) received through the OFDM receiver 133, and performs cancellation corresponding to the restored core layer data. Restore enhanced layer data. At this time, the signal demultiplexing device 131 first generates a broadcast signal frame, restores a bootstrap from the broadcast signal frame, restores a preamble using information included in the bootstrap, and then signaling information data included in the preamble. It can be used for signal recovery. In this case, the signaling information may be L1 signaling information and may include injection level information, normalizing factor information, and the like.

In this case, the preamble may include a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes.

At this time, the preamble includes PLP identification information for identifying Physical Layer Pipes (PLPs); and layer identification information for identifying layers corresponding to the hierarchical division.

In this case, PLP identification information and layer identification information may be included in the preamble as separate fields.

In this case, time interleaver information may be included in the preamble based on the core layer.

In this case, the preamble may selectively include injection level information corresponding to the injection level controller according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes.

In this case, the preamble may include type information, start position information, and size information of physical layer pipes.

In this case, the type information may be for identifying one of a first type corresponding to a non-dispersed physical layer pipe and a second type corresponding to a dispersed physical layer pipe.

In this case, non-distributed physical layer pipes are allocated for contiguous data cell indices, and the distributed physical layer pipes may include two or more subslices.

In this case, type information may be selectively signaled according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes.

In this case, type information may be signaled only for the core layer.

In this case, the start position information may be set equal to an index corresponding to the first data cell of the physical layer pipe.

In this case, the start position information may indicate a start position of the physical layer pipe using a cell addressing scheme.

In this case, start position information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, size information may be set based on the number of data cells allocated to the physical layer pipe.

In this case, size information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, time interleaver information may be signaled based on the core layer.

In this case, the time interleaver may correspond to a hybrid time interleaver. In this case, physical layer pipes (PLPs) of the core layer and the enhanced layer may include only complete FEC blocks.

In this case, the preamble is information for identifying a part of the FEC block of the enhanced layer corresponding to the boundary of the time interleaver groups when the boundary of the time interleaver groups does not correspond to the boundary of the FEC blocks of the enhanced layer. can be signaled.

In this case, the information for identifying a part of the FEC block includes start position information of the physical layer pipe of the core layer, start position information of the physical layer pipe of the enhanced layer, modulation information corresponding to the enhanced layer, and the enhancement It may include any one or more of FEC type information corresponding to the hard layer.

In this case, the start position information of the physical layer pipe may correspond to the index of the first data cell of the physical layer pipe.

In this case, the modulation information may be signaled only when the FEC type information satisfies a preset condition.

In this case, the enhanced layer signal may correspond to enhanced layer data that is restored based on cancellation corresponding to restoration of core layer data corresponding to the core layer signal.

In this case, the time interleaver corresponds to a convolutional time interleaver, and the time interleaver groups include a physical layer pipe (PLP) including an incomplete FEC block. and the preamble may signal start position information of a first complete FEC block in the physical layer pipe.

In this case, the time interleaver may perform the interleaving in one of a plurality of operation modes.

In this case, the operating modes may include a first mode in which time interleaving is omitted, a second mode in which convolutional time interleaving is performed, and a third mode in which hybrid time interleaving is performed. can

In this case, the preamble includes a field indicating a start position of a first complete FEC block corresponding to a current physical layer pipe for the first mode and the second mode, and For mode 3, the field indicating the start position of the first FEC block may not be included. In this case, the field indicating the start position may indicate the start position of the first FEC block starting within the current physical layer pipe during the current subframe.

In this case, the field indicating the start position of the first FEC block is either a first field used in the first mode or a second field used in the second mode, and the first field and the second field are lengths. may be different.

In this case, the length of the second field may be longer than that of the first field.

In this case, the length of the first field is determined based on the length and modulation order of the LDPC codeword, and the length of the second field is determined based on the length and modulation order of the LDPC codeword as well as the depth of the convolutional time interleaver ) can be determined by further considering.

In this case, the length of the first field may be 15 bits, and the length of the second bit may be 22 bits.

In this case, the first field and the second field may be separately signaled for each of the core layer corresponding to the core layer signal and the enhanced layer corresponding to the enhanced layer signal.

As will be described later, the apparatus 111 for generating a broadcast signal frame shown in FIG. 1 includes a combiner generating a multiplexed signal by combining a core layer signal and an enhanced layer signal; a power normalizer performing power normalization to lower the power of the multiplexed signal to a power corresponding to the core layer signal; a time interleaver to generate a time interleaved signal by performing time interleaving after the power normalization has been performed; and a frame builder generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes. In this case, the time interleaver uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values. At this time, the broadcast signal transmission apparatus 110 shown in FIG. 1 includes a combiner generating a multiplexed signal by combining a core layer signal and an enhanced layer signal; a power normalizer performing power normalization to lower the power of the multiplexed signal to a power corresponding to the core layer signal; a time interleaver to generate a time interleaved signal by performing time interleaving after the power normalization has been performed; a frame builder generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes; and an OFDM transmitter for transmitting the broadcast signal frame through an antenna using an OFDM communication scheme. In this case, the time interleaver uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

According to an embodiment, the apparatus 111 for generating a broadcast signal frame shown in FIG. 1 includes a time interleaver for generating a time interleaved signal by performing time interleaving on a BICM output signal in the case of a single layer; and a frame builder generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes. In this case, the BICM output signal may be an output signal of a BICM device to be described later. At this time, the broadcast signal transmission apparatus 110 shown in FIG. 1 includes a time interleaver that performs time interleaving on a BICM output signal to generate a time interleaved signal; a frame builder generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes; and an OFDM transmitter for transmitting the broadcast signal frame through an antenna using an OFDM communication scheme.

As will be described later, the signal demultiplexing apparatus shown in FIG. 1 includes a time deinterleaver for generating a time deinterleaving signal by applying time deinterleaving to a received signal corresponding to a broadcast signal frame; a de-normalizer that increases the power of the received signal or the time deinterleaved signal by the power reduction by the power normalizer of the transmitter; a core layer BICM decoder restoring core layer data from the signal power-adjusted by the de-normalizer; Enhancement for extracting an enhanced layer signal by performing cancellation corresponding to the core layer data on the signal power-adjusted by the de-normalizer using an output signal of the core layer FEC decoder of the core layer BICM decoder de-layer symbol extractor; a de-injection level controller that increases the power of the enhanced layer signal by the power reduction of the injection level controller of the transmitter; and an enhanced layer BICM decoder for restoring enhanced layer data using an output signal of the de-injection level controller. At this time, the broadcast signal receiving apparatus 130 shown in FIG. 1 includes an OFDM receiver generating a received signal by performing one or more of synchronization, channel estimation, and equalization on a transmitted signal corresponding to a broadcast signal frame; a time deinterleaver generating a time deinterleaving signal by applying time deinterleaving to the received signal; a de-normalizer that increases the power of the received signal or the time deinterleaved signal by the power reduction by the power normalizer of the transmitter; a core layer BICM decoder restoring core layer data from the signal power-adjusted by the de-normalizer; Enhancement for extracting an enhanced layer signal by performing cancellation corresponding to the core layer data on the signal power-adjusted by the de-normalizer using an output signal of the core layer FEC decoder of the core layer BICM decoder de-layer symbol extractor; a de-injection level controller that increases the power of the enhanced layer signal by the power reduction of the injection level controller of the transmitter; and an enhanced layer BICM decoder for restoring enhanced layer data using an output signal of the de-injection level controller.

In this case, the time deinterleaver uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

In this case, the time deinterleaver may correspond to the time interleaving mode.

In this case, the preamble may include a time interleaving mode corresponding to the time interleaver for each of all physical layer pipes.

Although not explicitly shown in FIG. 1, the broadcast signal transmission/reception system according to an embodiment of the present invention may multiplex/demultiplex one or more enhancement layer data in addition to core layer data and enhanced layer data. In this case, the enhancement layer data may be multiplexed at a lower power level than the core layer data and the enhanced layer data. Furthermore, when two or more extension layers are included, the injection power level of the second extension layer is lower than the injection power level of the first extension layer, and the injection power level of the third extension layer is lower than the injection power level of the second extension layer. can

2 is an operational flowchart illustrating a broadcast signal transmission/reception method according to an embodiment of the present invention.

Referring to FIG. 2 , in the broadcast signal transmission/reception method according to an embodiment of the present invention, a broadcast signal frame is generated by combining and multiplexing a core layer signal and an enhanced layer signal (S210). In this case, the core layer signal and the enhanced layer signal may be combined at different power levels. In this case, the broadcast signal frame may be generated by performing time interleaving, and the preamble may include a time interleaving mode corresponding to time interleaving for each of all physical layer pipes.

According to an embodiment, in step S210, a time interleaved signal is generated by performing time interleaving on the BICM output signal, and a preamble for signaling a time interleaving mode corresponding to the time interleaving for each of all physical layer pipes is provided. It is also possible to generate a broadcast signal frame including

In this case, the physical layer pipes include a plurality of core layer physical layer pipes corresponding to one complete delivered product, and the core layer physical layer pipes may not be layered division multiplexed.

In this case, each of the core layer and physical layer pipes may use either a time interleaving mode of a no time interleaving mode or a hybrid time interleaving mode, and may not use a convolutional time interleaving mode.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, all of the core layer physical layer pipes use intra-subframe interleaving mode or all use inter-subframe interleaving mode. can

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, the core layer physical layer pipes may all have the same L1D_plp_HTI_inter_subframe value.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes and all of the core layer physical layer pipes use the inter-subframe interleaving mode, the core layer physical layer pipes have the same time interleaving mode. An interleaving unit (N _IU ) may be used.

In this case, when at least one of the time interleaving modes corresponding to the core layer physical layer pipes is the no time interleaving mode, all core layer pipes configured in the hybrid time interleaving mode among the core layer pipes are in the intra-subframe interleaving mode. can be used.

At this time, one complete transmission product corresponds to one or more subframes, in which all available data cells of the subframe are first filled with dummy modulation values, and then the real physical layer pipe data is overwritten. can be created

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

In this case, the physical layer pipes may include one enhanced layer physical layer pipe and a plurality of core layer physical layer pipes layered division multiplexed to the one enhanced layer physical layer pipe.

In this case, the time interleaving mode corresponding to the enhanced layer physical layer pipe may be the same as the time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipe is layered division multiplexed.

In this case, time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed may all be no time interleaving modes or all hybrid time interleaving modes.

In this case, when the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are all hybrid time interleaving modes, the core layer physical layer pipes all use the intra-subframe interleaving mode. can be used

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are no time interleaving modes, each of the core layer physical layer pipes is in each subframe. It may consist of an integer number of FEC blocks.

At this time, the subframe may be created by first filling all available data cells of the subframe with dummy modulation values, and then overwriting the actual physical layer pipe data.

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

At this time, the broadcast signal frame generated in step S210 may include a bootstrap, a preamble, and a super-imposed payload. At this time, any one or more of the bootstrap and preamble may include L1 signaling information. In this case, the L1 signaling information may include injection level information and normalizing factor information.

At this time, the preamble includes PLP identification information for identifying Physical Layer Pipes (PLPs); and layer identification information for identifying layers corresponding to the hierarchical division.

In this case, PLP identification information and layer identification information may be included in the preamble as separate fields.

In this case, time interleaver information may be included in the preamble based on the core layer.

In this case, the preamble may selectively include injection level information corresponding to the injection level controller according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes.

In this case, the preamble may include type information, start position information, and size information of physical layer pipes.

In this case, the type information may be for identifying one of a first type corresponding to a non-dispersed physical layer pipe and a second type corresponding to a dispersed physical layer pipe.

In this case, non-distributed physical layer pipes are allocated for contiguous data cell indices, and the distributed physical layer pipes may include two or more subslices.

In this case, type information may be selectively signaled according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes.

In this case, type information may be signaled only for the core layer.

In this case, the start position information may be set equal to an index corresponding to the first data cell of the physical layer pipe.

In this case, the start position information may indicate a start position of the physical layer pipe using a cell addressing scheme.

In this case, start position information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, size information may be set based on the number of data cells allocated to the physical layer pipe.

In this case, size information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, time interleaver information may be signaled based on the core layer.

In this case, the step of generating the time-interleaved signal may perform the interleaving using a hybrid time interleaver. In this case, physical layer pipes (PLPs) of the core layer and the enhanced layer may include only complete FEC blocks.

In this case, the preamble is information for identifying a part of the FEC block of the enhanced layer corresponding to the boundary of the time interleaver groups when the boundary of the time interleaver groups does not correspond to the boundary of the FEC blocks of the enhanced layer. can be signaled.

In this case, the information for identifying a part of the FEC block includes start position information of the physical layer pipe of the core layer, start position information of the physical layer pipe of the enhanced layer, modulation information corresponding to the enhanced layer, and the enhancement It may include any one or more of FEC type information corresponding to the hard layer.

In this case, the start position information of the physical layer pipe may correspond to the index of the first data cell of the physical layer pipe.

In this case, the modulation information may be signaled only when the FEC type information satisfies a preset condition.

In this case, the enhanced layer signal may correspond to enhanced layer data that is restored based on cancellation corresponding to restoration of core layer data corresponding to the core layer signal.

In this case, in the step of generating the time interleaved signal, the interleaving is performed using a convolutional time interleaver, and the time interleaver groups are physical including incomplete FEC blocks. A layer pipe (Physical Layer Pipe; PLP) is included, and the preamble may signal start position information of a first complete FEC block in the physical layer pipe.

In this case, the interleaving uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

At this time, dummy values are applied after the actual data cells of the last enhanced PLP in the PLP group so that the total number of enhanced layer cells in the PLP group is equal to the total number of core layer cells in the PLP group. of the last Enhanced PLP) can be inserted.

In this case, dummy values may not be inserted into core layer data.

In this case, dummy values may be inserted after completing the core layer BICM and the enhanced layer BICM and before combining the core layer signal and the enhanced layer signal.

In this case, the dummy values may correspond to a preset scrambling sequence.

At this time, the scrambling sequence may be modulated using the constellation mapping used for the last enhanced PLP.

In this case, dummy values may have the same power as the last enhanced PLP.

In this case, the scrambling sequence may be generated using a 16-bit shift register corresponding to a preset generator polynomial.

In this case, the scrambling sequence may be generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ .

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

In this case, the interleaving may be performed using one of a plurality of operation modes.

In this case, the operating modes may include a first mode in which time interleaving is omitted, a second mode in which convolutional time interleaving is performed, and a third mode in which hybrid time interleaving is performed. there is.

In this case, the preamble includes a field indicating a start position of a first complete FEC block corresponding to a current physical layer pipe for the first mode and the second mode, and For mode 3, the field indicating the start position of the first FEC block may not be included.

In this case, the field indicating the start position of the first FEC block is either a first field used in the first mode or a second field used in the second mode, and the first field and the second field are lengths. may be different.

In this case, the length of the second field may be longer than that of the first field.

In this case, the length of the first field is determined based on the length and modulation order of the LDPC codeword, and the length of the second field is determined based on the length and modulation order of the LDPC codeword as well as the depth of the convolutional time interleaver ) can be determined by further considering.

In this case, the length of the first field may be 15 bits, and the length of the second bit may be 22 bits.

In this case, the first field and the second field may be separately signaled for each of the core layer corresponding to the core layer signal and the enhanced layer corresponding to the enhanced layer signal.

In addition, in the method for transmitting/receiving broadcast signals according to an embodiment of the present invention, OFDM transmission of broadcast signal frames is performed (S220).

In addition, in the broadcast signal transmission/reception method according to an embodiment of the present invention, the transmitted signal is OFDM received (S230).

At this time, in step S230, synchronization, channel estimation, and equalization may be performed.

At this time, in step S230, the bootstrap may be restored, the preamble may be restored using a signal included in the restored bootstrap, and a data signal may be restored using signaling information included in the preamble.

In addition, in the broadcast signal transmission/reception method according to an embodiment of the present invention, core layer data is restored from the received signal (S240).

In addition, in the broadcast signal transmission/reception method according to an embodiment of the present invention, enhanced layer data is restored through core layer signal cancellation (S250).

In particular, steps S240 and S250 shown in FIG. 2 may correspond to demultiplexing operations corresponding to step S210.

As will be described later, step S210 shown in FIG. 2 includes generating a time interleaved signal by performing time interleaving on a BICM output signal in the case of a single layer; and generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaving for each of all physical layer pipes. In this case, the broadcast signal transmission method of steps S210 and S220 includes performing time interleaving on a BICM output signal in the case of a single layer to generate a time interleaved signal; generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaving for each of all physical layer pipes; and transmitting the broadcast signal frame through an antenna using an OFDM communication scheme.

As will be described later, step S210 shown in FIG. 2 includes generating a multiplexed signal by combining a core layer signal and an enhanced layer signal; performing power normalization to lower the power of the multiplexed signal to a power corresponding to the core layer signal; generating a time interleaved signal by performing time interleaving after the power normalization is performed; and generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaving for each of all physical layer pipes. In this case, the time interleaving uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values. At this time, the broadcast signal transmission method of steps S210 and S220 includes generating a multiplexed signal by combining a core layer signal and an enhanced layer signal; performing power normalization to lower the power of the multiplexed signal to a power corresponding to the core layer signal; generating a time interleaved signal by performing time interleaving after the power normalization is performed; generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaving for each of all physical layer pipes; and transmitting the broadcast signal frame through an antenna using an OFDM communication scheme. In this case, the time interleaving uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

As will be described later, steps S240 and S250 shown in FIG. 2 include generating a time deinterleaving signal by applying time deinterleaving to a received signal corresponding to a broadcast signal frame; increasing the power of the received signal or the time deinterleaving signal by a power reduction by a power normalizer of a transmitter; restoring core layer data from the power-adjusted signal; extracting an enhanced layer signal by performing cancellation corresponding to the core layer data on the power-adjusted signal; increasing the power of the enhanced layer signal by a power reduction of an injection level controller of a transmitter; and restoring enhanced layer data using the power-adjusted enhanced layer signal. At this time, a broadcast signal reception method according to an embodiment of the present invention includes generating a received signal by performing at least one of synchronization, channel estimation, and equalization on a transmitted signal corresponding to a broadcast signal frame; generating a time deinterleaving signal by applying time deinterleaving to the received signal; increasing the power of the received signal or the time deinterleaving signal by a power reduction by a power normalizer of a transmitter; restoring core layer data from the power-adjusted signal; extracting an enhanced layer signal by performing cancellation corresponding to the core layer data on the power-adjusted signal; increasing the power of the enhanced layer signal by a power reduction of an injection level controller of a transmitter; and restoring enhanced layer data using the power-adjusted enhanced layer signal.

In this case, time deinterleaving may correspond to a time interleaving mode.

In this case, the preamble may include a time interleaving mode corresponding to time interleaving for each of all physical layer pipes.

In this case, time deinterleaving may be performed in one of a plurality of operation modes.

In this case, time deinterleaving uses one of the time interleaver groups, and enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

FIG. 3 is a block diagram illustrating an example of an apparatus for generating a broadcast signal frame shown in FIG. 1 .

Referring to FIG. 3, an apparatus for generating broadcast signal frames according to an embodiment of the present invention includes a core layer BICM unit 310, an enhanced layer BICM unit 320, an injection level controller 330, a combiner 340, a power A normalizer 345, a time interleaver 350, a signaling generator 360, and a frame builder 370 may be included.

In general, a BICM (Bit-Interleaved Coded Modulation) device is composed of an error correction encoder, a bit interleaver, and a symbol mapper, and the core layer BICM unit 310 and the enhanced layer BICM unit 320 shown in FIG. 3 also have errors, respectively. It may include a correction coder, a bit interleaver and a symbol mapper. In particular, the error correction encoders (CORE LAYER FEC ENCODER and ENHANCED LAYER FEC ENCODER) shown in FIG. 3 may be serially coupled of a BCH encoder and an LDPC encoder, respectively. At this time, the input of the error correction encoder may be input to the BCH encoder, the output of the BCH encoder may be input to the LDPC encoder, and the output of the LDPC encoder may be the output of the error correction encoder.

As shown in FIG. 3 , core layer data and enhanced layer data are combined through a combiner 340 after passing through different BICM units. That is, in the present invention, Layered Division Multiplexing (LDM) may mean combining and transmitting multiple layers into one using a power difference.

That is, core layer data passes through the core layer BICM unit 310, and enhanced layer data passes through the enhanced layer BICM unit 320, passes through the injection level controller 330, and is combined in the combiner 340. At this time, the enhanced layer BICM unit 320 may perform BICM encoding different from that of the core layer BICM unit 310 . That is, the enhanced layer BICM unit 320 may perform error correction encoding or symbol mapping corresponding to a higher bit rate than the core layer BICM unit 310 . In addition, the enhanced layer BICM unit 320 may perform error correction encoding or symbol mapping that is less robust than the core layer BICM unit 310 .

For example, the core layer error correction coder may have a lower bit rate than the enhanced layer error correction coder. In this case, the enhanced layer symbol mapper may be less robust than the core layer symbol mapper.

The combiner 340 may be viewed as combining a core layer signal and an enhanced layer signal. In this case, the combiner 340 may combine the core layer signal and the enhanced layer signal at different power levels. Depending on embodiments, power level control may be performed on a core layer signal rather than an enhanced layer signal. In this case, the power of the core layer signal may be adjusted to be greater than the power of the enhanced layer signal.

Core layer data uses low code rate forward error correction (FEC) codes for robust reception, whereas enhanced layer data uses high code rate FEC codes for high data rates. can

That is, compared to enhanced layer data, core layer data may have a wider coverage in the same reception environment.

The gain (or power) of the enhanced layer data passing through the enhanced layer BICM unit 320 is adjusted through the injection level controller 330 and combined with the core layer data by the combiner 340 .

That is, the injection level controller 330 generates a power reduced enhanced layer signal by reducing the power of the enhanced layer signal. At this time, the level of the signal controlled by the injection level controller 330 may be determined according to the injection level. At this time, the injection level when signal B is inserted into signal A may be defined as in Equation 1 below.

[Equation 1]

For example, assuming that the injection level is 3 dB when inserting the enhanced layer signal into the core layer signal, the enhanced layer signal has a power corresponding to half that of the core layer signal.

In this case, the injection level controller 330 may adjust the power level of the enhanced layer signal from 0 dB to 25.0 dB at 0.5 dB or 1 dB intervals.

In general, transmit power allocated to the core layer is allocated higher than transmit power allocated to the enhanced layer, and through this, a receiver can preferentially decode the core layer.

In this case, the combiner 340 may be regarded as generating a multiplexed signal by combining the core layer signal and the power reduced enhanced layer signal.

The signal combined by the combiner 340 is provided to the power normalizer 345 to lower the power by the power increase caused by the combination of the core layer signal and the enhanced layer signal, and power control is performed. That is, the power normalizer 345 lowers the power of the signal multiplexed by the combiner 340 to a power level corresponding to the core layer signal. Since the level of the combined signal is higher than that of one layer signal, power normalization of the power normalizer 345 is required to prevent amplitude clipping in the rest of the broadcast signal transmission/reception system.

At this time, the power normalizer 345 may multiply the magnitude of the combined signal by the normalizing factor of Equation 2 below to adjust the magnitude of the appropriate signal. Injection level information for calculating Equation 2 below may be delivered to the power normalizer 345 through a signaling flow.

[Equation 2]

와 같이 표현될 수 있다.Assuming that the power levels of the core layer signal and the enhanced layer signal are normalized to 1 when the enhanced layer signal S _E is injected into the core layer signal S _C by a predetermined injection level, the combined signal is

can be expressed as

At this time, α represents a scaling factor corresponding to various injection levels. That is, the injection level controller 330 may correspond to the scaling factor.

와 같이 표현될 수 있다.For example, if the injection level of the enhanced layer is 3 dB, the combined signal is

can be expressed as

Since the power of the combined signal (multiplexed signal) is increased compared to the core layer signal, the power normalizer 345 must mitigate this power increase.

와 같이 표현될 수 있다.The output of the power normalizer 345 is

can be expressed as

In this case, β represents a normalizing factor according to various injection levels of the enhanced layer.

와 같이 표현될 수 있다.When the injection level of the enhanced layer is 3dB, the power increase of the combined signal compared to the core layer signal is 50%. Thus, the output of power normalizer 345 is

can be expressed as

Table 1 below shows scaling factors α and normalizing factors β according to various injection levels (CL: Core Layer, EL: Enhanced Layer). The relationship between the injection level, the scaling factor α, and the normalization factor β can be defined as follows.

[Equation 3]

EL Injection level relative to CL

Scaling factor

Normalizing factor

3.0 dB 0.7079458 0.8161736 3.5 dB 0.6683439 0.8314061 4.0 dB 0.6309573 0.8457262 4.5 dB 0.5956621 0.8591327 5.0 dB 0.5623413 0.8716346 5.5 dB 0.5308844 0.8832495 6.0 dB 0.5011872 0.8940022 6.5 dB 0.4731513 0.9039241 7.0 dB 0.4466836 0.9130512 7.5 dB 0.4216965 0.9214231 8.0 dB 0.3981072 0.9290819 8.5 dB 0.3758374 0.9360712 9.0 dB 0.3548134 0.9424353 9.5 dB 0.3349654 0.9482180 10.0 dB 0.3162278 0.9534626

Depending on the embodiment, the injection level may be a value ranging from 0 dB to 25 dB. When the injection level is 0 dB, the core layer signal and the enhanced layer signal may be combined with the same power. In this case, the scaling factor may be 1 and the normalizing factor may be 0.7071068. That is, the power normalizer 345 corresponds to the normalizing factor, and the power of the multiplexed signal is increased by the combiner 340. In this case, the normalizing factor and the scaling factor may be rational numbers greater than 0 and less than 1, respectively.

In this case, the scaling factor may decrease as the power reduction corresponding to the injection level controller 330 increases, and the normalizing factor may increase as the power reduction corresponding to the injection level controller 330 increases.

The power-normalized signal passes through a time interleaver 350 for distributing burst errors generated in the channel.

In this case, the time interleaver 350 can be regarded as performing interleaving applied to both the core layer signal and the enhanced layer signal. That is, since the core layer and the enhanced layer share a time interleaver, unnecessary memory usage can be prevented and latency in the receiver can be reduced.

As will be described later, the enhanced layer signal may correspond to enhanced layer data that is reconstructed based on cancellation corresponding to restoration of the core layer data corresponding to the core layer signal, and the combiner 340 may correspond to the core layer data. One or more extension layer signals having a power level lower than that of the signal and the enhanced layer signal may be combined with the core layer signal and the enhanced layer signal.

Meanwhile, the L1 signaling information including the injection level information is encoded in the signaling generator 360 including BICM dedicated to signaling. At this time, the signaling generating unit 360 may receive injection level information IL INFO from the injection level controller 330 and generate an L1 signaling signal.

In L1 signaling, L1 represents Layer-1, which is the lowest layer of the ISO 7 layer model. At this time, L1 signaling may be included in the preamble.

In general, L1 signaling may include FFT size, guard interval size, etc., which are main parameters of an OFDM transmitter, and channel code rate, modulation information, etc., which are main BICM parameters. The L1 signaling signal is combined with the data signal to form a broadcast signal frame.

The frame builder 370 combines the L1 signaling signal and the data signal to generate a broadcast signal frame. At this time, the frame builder 370 uses the time-interleaved signal to signal time interleaver information shared by the core layer signal and the enhanced layer signal and size information of Physical Layer Pipes (PLPs). It is possible to generate a broadcast signal frame including a preamble for In this case, the broadcast signal frame may further include a bootstrap.

At this time, the frame builder 370 can be regarded as generating a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaver 350.

In this case, the time interleaving mode may be signaled for each of all physical layer pipes.

In this case, the physical layer pipes may include one enhanced layer physical layer pipe and a plurality of core layer physical layer pipes layered division multiplexed to the one enhanced layer physical layer pipe.

In this case, the time interleaving mode corresponding to the enhanced layer physical layer pipe may be the same as the time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipe is layered division multiplexed.

In this case, time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed may all be no time interleaving modes or all hybrid time interleaving modes.

In this case, when the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are all hybrid time interleaving modes, the core layer physical layer pipes all use the intra-subframe interleaving mode. can be used

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are no time interleaving modes, each of the core layer physical layer pipes is in each subframe. It may consist of an integer number of FEC blocks.

At this time, the subframe may be created by first filling all available data cells of the subframe with dummy modulation values, and then overwriting the actual physical layer pipe data.

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

In the case of a single layer, the enhanced layer BICM unit 320, injection level controller 330, combiner 340, and power normalizer 345 may be omitted. In this case, the time interleaver 350 may perform time interleaving on the BICM output signal of the core layer BICM unit 310 to generate a time interleaved signal. Also, the frame builder 370 may generate a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaver 350 for each of all physical layer pipes.

In this case, the physical layer pipes include a plurality of core layer physical layer pipes corresponding to one complete delivered product, and the core layer physical layer pipes may not be layered division multiplexed.

In this case, each of the core layer and physical layer pipes may use either a time interleaving mode of a no time interleaving mode or a hybrid time interleaving mode, and may not use a convolutional time interleaving mode.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, all of the core layer physical layer pipes use intra-subframe interleaving mode or all use inter-subframe interleaving mode. can

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, the core layer physical layer pipes may all have the same L1D_plp_HTI_inter_subframe value.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes and all of the core layer physical layer pipes use the inter-subframe interleaving mode, the core layer physical layer pipes have the same time interleaving mode. An interleaving unit (N _IU ) may be used.

In this case, when at least one of the time interleaving modes corresponding to the core layer physical layer pipes is the no time interleaving mode, all core layer pipes configured in the hybrid time interleaving mode among the core layer pipes are in the intra-subframe interleaving mode. can be used.

At this time, one complete transmission product corresponds to one or more subframes, in which all available data cells of the subframe are first filled with dummy modulation values, and then the real physical layer pipe data is overwritten. can be created

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

In this case, the time interleaver 350 uses one of the time interleaver groups, and a boundary between the time interleaver groups includes physical layer pipes of a core layer corresponding to the core layer signal; It may be a boundary between PLPs). That is, one of the boundaries between physical layer pipes of the core layer may become a boundary between time interleaver groups.

In this case, enhanced layer data corresponding to one of the time interleaver groups may include dummy values.

At this time, dummy values are applied after the actual data cells of the last enhanced PLP in the PLP group so that the total number of enhanced layer cells in the PLP group is equal to the total number of core layer cells in the PLP group. of the last Enhanced PLP) can be inserted.

In this case, dummy values may not be inserted into core layer data.

In this case, dummy values may be inserted after completing the core layer BICM and the enhanced layer BICM and before combining the core layer signal and the enhanced layer signal.

In this case, the dummy values may correspond to a preset scrambling sequence.

At this time, the scrambling sequence may be modulated using the constellation mapping used for the last enhanced PLP.

In this case, dummy values may have the same power as the last enhanced PLP.

In this case, the scrambling sequence may be generated using a 16-bit shift register corresponding to a preset generator polynomial.

In this case, the scrambling sequence may be generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ .

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

In this case, time interleaver information may be signaled based on the core layer.

According to embodiments, some of the time interleaver information may be signaled based on a core layer, and other parts of the time interleaver information may be signaled regardless of the layer.

That is, time interleaver information may be signaled based on layer identification information corresponding to the core layer.

In this case, the time interleaver 350 may correspond to a hybrid time interleaver. In this case, the physical layer pipes of the core layer and the enhanced layer may include only complete FEC blocks.

In this case, the preamble is information for identifying a part of the FEC block of the enhanced layer corresponding to the boundary of the time interleaver groups when the boundary of the time interleaver groups does not correspond to the boundary of the FEC blocks of the enhanced layer. can be signaled.

In this case, the information for identifying a part of the FEC block includes start position information of the physical layer pipe of the core layer, start position information of the physical layer pipe of the enhanced layer, modulation information corresponding to the enhanced layer, and the enhancement It may include any one or more of FEC type information corresponding to the hard layer.

In this case, the start position information of the physical layer pipe may correspond to the index of the first data cell of the physical layer pipe.

In this case, the modulation information may be signaled only when the FEC type information satisfies a preset condition.

In this case, the enhanced layer signal may correspond to enhanced layer data that is restored based on cancellation corresponding to restoration of core layer data corresponding to the core layer signal.

In this case, the time interleaver 350 corresponds to a convolutional time interleaver, and the time interleaver groups include physical layer pipes (PLPs) including incomplete FEC blocks. ), and the preamble may signal start position information of a first complete FEC block in the physical layer pipe.

In this case, the time interleaver 350 may perform the interleaving in one of a plurality of operation modes.

At this time, the operation modes include a first mode (L1D_plp_TI_mode = 00) that omits time interleaving, a second mode (L1D_plp_TI_mode = 01) that performs convolutional time interleaving, and hybrid time interleaving (Hybrid time interleaving). ) may include a third mode (L1D_plp_TI_mode=10) for performing.

In this case, the preamble includes a field indicating a start position of a first complete FEC block corresponding to a current physical layer pipe for the first mode and the second mode, For the third mode, the field indicating the start position of the first FEC block may not be included.

In this case, the field indicating the start position of the first FEC block is a first field (L1D_plp_fec_block_start) used in the first mode (L1D_plp_TI_mode = 00) and a second field (L1D_plp_TI_mode = 01) used in the second mode ( L1D_plp_CTI_fec_block_start), and the first field and the second field may have different lengths. At this time, the first field (L1D_plp_fec_block_start) indicates the start position of the first FEC block starting within the current physical layer pipe during the current subframe, and the second field (L1D_plp_CTI_fec_block_start) indicates the current or subsequent subframe. may indicate the start position of the first complete FEC block of the current physical layer pipe leaving the convolutional time interleaver in . In this case, both the first field (L1D_plp_fec_block_start) and the second field (L1D_plp_CTI_fec_block_start) may be signaled after interleaving. In particular, in the case of the second field (L1D_plp_CTI_fec_block_start), the number of bits required for signaling can be increased if signaling is performed based on the period after interleaving.

In this case, the length of the second field may be longer than that of the first field.

At this time, the length of the first field is determined based on the length of the LDPC codeword and the modulation order, and the length of the second field is the depth of the convolutional time interleaver as well as the length and modulation order of the LDPC codeword It can be determined by further considering.

In this case, the length of the first field may be 15 bits, and the length of the second field may be 22 bits.

In this case, the first field and the second field may be separately signaled for each of the core layer corresponding to the core layer signal and the enhanced layer corresponding to the enhanced layer signal.

At this time, the frame builder 370 includes a bootstrap generating unit generating the bootstrap; a preamble generating unit generating the preamble; and a super-imposed payload generating unit generating a super-imposed payload corresponding to the time-interleaved signal.

In this case, the bootstrap may be shorter than the preamble and may have a fixed length.

At this time, the bootstrap includes a symbol representing the structure of the preamble,

The symbol may correspond to a fixed-length bit string representing a combination of a modulation method/code rate, an FFT size, a guard interval length, and a pilot pattern of the preamble.

At this time, when the modulation method/code rate is the same, the preamble structure corresponding to the second FFT size smaller than the first FFT size is preferentially allocated to the symbol over the preamble structure corresponding to the first FFT size, and the modulation When the method/code rate and the FFT size are the same, a preamble structure corresponding to a second guard interval length greater than the first guard interval length is preferentially assigned to a lookup table over a preamble structure corresponding to the first guard interval length may be commensurate.

A broadcast signal frame is transmitted through an OFDM transmitter that is robust to multi-path and Doppler. At this time, the OFDM transmitter can be regarded as being in charge of generating a transmission signal for a next-generation broadcasting system.

At this time, the preamble includes PLP identification information for identifying Physical Layer Pipes (PLPs); and layer identification information for identifying layers corresponding to the hierarchical division.

In this case, PLP identification information and layer identification information may be included in the preamble as separate fields.

In this case, time interleaver information may be included in the preamble based on the core layer.

At this time, the preamble may selectively include injection level information corresponding to the injection level controller according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes (IF(j>0)). can

In this case, the preamble may include type information, start position information, and size information of physical layer pipes.

In this case, the type information may be for identifying one of a first type corresponding to a non-dispersed physical layer pipe and a second type corresponding to a dispersed physical layer pipe.

In this case, non-distributed physical layer pipes are allocated for contiguous data cell indices, and the distributed physical layer pipes may include two or more subslices.

In this case, type information may be selectively signaled according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes.

In this case, type information may be signaled only for the core layer.

In this case, the start position information may be set equal to an index corresponding to the first data cell of the physical layer pipe.

In this case, the start position information may indicate a start position of the physical layer pipe using a cell addressing scheme.

In this case, start position information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, size information may be set based on the number of data cells allocated to the physical layer pipe.

In this case, size information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

4 is a diagram illustrating an example of a broadcast signal frame structure.

Referring to FIG. 4 , a broadcast signal frame includes a bootstrap 410 , a preamble 420 and a super-imposed payload 430 .

The frame shown in FIG. 4 may be included in a super-frame.

In this case, the broadcast signal frame may be composed of one or more OFDM symbols. A broadcast signal frame may include a reference symbol or a pilot symbol.

A frame structure to which Layered Division Multiplexing (LDM) is applied includes a bootstrap 410, a preamble 420, and a super-imposed payload 430 as shown in FIG.

At this time, the bootstrap 410 and the preamble 420 can be regarded as two preambles hierarchical.

In this case, the bootstrap 410 may have a shorter length than the preamble 420 for fast acquisition and detection. At this time, the bootstrap 410 may have a fixed length. At this time, the bootstrap 410 may include symbols of a fixed length. For example, the bootstrap 410 may consist of 4 OFDM symbols each 0.5 ms long and have a total fixed time length of 2 ms.

In this case, the bootstrap 410 has a fixed bandwidth, and the preamble 420 and the super-imposed payload 430 may have a wider and more variable bandwidth than the bootstrap 410 .

The preamble 420 may transmit detailed signaling information using a robust LDPC code. In this case, the length of the preamble 420 may vary according to signaling information.

At this time, both the bootstrap 410 and the payload 430 can be regarded as corresponding to common signals shared by several layers.

The super-imposed payload 430 may correspond to a signal obtained by multiplexing two or more layer signals. In this case, the super-imposed payload 430 may be a combination of a core layer payload and an enhanced layer payload at different power levels. In this case, the core layer payload may include an in-band signaling section. In this case, the in-band signaling unit may include signaling information for enhanced layer service.

At this time, the bootstrap 410 may include a symbol representing a preamble structure.

At this time, symbols included in the bootstrap to indicate the structure of the preamble may be set as shown in Table 2 below.

preamble_structure L1-Basic Mode FFT Size GI Length (samples) Pilot Pattern (D _X ) 0 L1-Basic Mode 1 8192 2048 3 One L1-Basic Mode 1 8192 1536 4 2 L1-Basic Mode 1 8192 1024 3 3 L1-Basic Mode 1 8192 768 4 4 L1-Basic Mode 1 16384 4096 3 5 L1-Basic Mode 1 16384 3648 4 6 L1-Basic Mode 1 16384 2432 3 7 L1-Basic Mode 1 16384 1536 4 8 L1-Basic Mode 1 16384 1024 6 9 L1-Basic Mode 1 16384 768 8 10 L1-Basic Mode 1 32768 4864 3 11 L1-Basic Mode 1 32768 3648 3 12 L1-Basic Mode 1 32768 3648 8 13 L1-Basic Mode 1 32768 2432 6 14 L1-Basic Mode 1 32768 1536 8 15 L1-Basic Mode 1 32768 1024 12 16 L1-Basic Mode 1 32768 768 16 17 L1-Basic Mode 2 8192 2048 3 18 L1-Basic Mode 2 8192 1536 4 19 L1-Basic Mode 2 8192 1024 3 20 L1-Basic Mode 2 8192 768 4 21 L1-Basic Mode 2 16384 4096 3 22 L1-Basic Mode 2 16384 3648 4 23 L1-Basic Mode 2 16384 2432 3 24 L1-Basic Mode 2 16384 1536 4 25 L1-Basic Mode 2 16384 1024 6 26 L1-Basic Mode 2 16384 768 8 27 L1-Basic Mode 2 32768 4864 3 28 L1-Basic Mode 2 32768 3648 3 29 L1-Basic Mode 2 32768 3648 8 30 L1-Basic Mode 2 32768 2432 6 31 L1-Basic Mode 2 32768 1536 8 32 L1-Basic Mode 2 32768 1024 12 33 L1-Basic Mode 2 32768 768 16 34 L1-Basic Mode 3 8192 2048 3 35 L1-Basic Mode 3 8192 1536 4 36 L1-Basic Mode 3 8192 1024 3 37 L1-Basic Mode 3 8192 768 4 38 L1-Basic Mode 3 16384 4096 3 39 L1-Basic Mode 3 16384 3648 4 40 L1-Basic Mode 3 16384 2432 3 41 L1-Basic Mode 3 16384 1536 4 42 L1-Basic Mode 3 16384 1024 6 43 L1-Basic Mode 3 16384 768 8 44 L1-Basic Mode 3 32768 4864 3 45 L1-Basic Mode 3 32768 3648 3 46 L1-Basic Mode 3 32768 3648 8 47 L1-Basic Mode 3 32768 2432 6 48 L1-Basic Mode 3 32768 1536 8 49 L1-Basic Mode 3 32768 1024 12 50 L1-Basic Mode 3 32768 768 16 51 L1-Basic Mode 4 8192 2048 3 52 L1-Basic Mode 4 8192 1536 4 53 L1-Basic Mode 4 8192 1024 3 54 L1-Basic Mode 4 8192 768 4 55 L1-Basic Mode 4 16384 4096 3 56 L1-Basic Mode 4 16384 3648 4 57 L1-Basic Mode 4 16384 2432 3 58 L1-Basic Mode 4 16384 1536 4 59 L1-Basic Mode 4 16384 1024 6 60 L1-Basic Mode 4 16384 768 8 61 L1-Basic Mode 4 32768 4864 3 62 L1-Basic Mode 4 32768 3648 3 63 L1-Basic Mode 4 32768 3648 8 64 L1-Basic Mode 4 32768 2432 6 65 L1-Basic Mode 4 32768 1536 8 66 L1-Basic Mode 4 32768 1024 12 67 L1-Basic Mode 4 32768 768 16 68 L1-Basic Mode 5 8192 2048 3 69 L1-Basic Mode 5 8192 1536 4 70 L1-Basic Mode 5 8192 1024 3 71 L1-Basic Mode 5 8192 768 4 72 L1-Basic Mode 5 16384 4096 3 73 L1-Basic Mode 5 16384 3648 4 74 L1-Basic Mode 5 16384 2432 3 75 L1-Basic Mode 5 16384 1536 4 76 L1-Basic Mode 5 16384 1024 6 77 L1-Basic Mode 5 16384 768 8 78 L1-Basic Mode 5 32768 4864 3 79 L1-Basic Mode 5 32768 3648 3 80 L1-Basic Mode 5 32768 3648 8 81 L1-Basic Mode 5 32768 2432 6 82 L1-Basic Mode 5 32768 1536 8 83 L1-Basic Mode 5 32768 1024 12 84 L1-Basic Mode 5 32768 768 16 85 L1-Basic Mode 6 8192 2048 3 86 L1-Basic Mode 6 8192 1536 4 87 L1-Basic Mode 6 8192 1024 3 88 L1-Basic Mode 6 8192 768 4 89 L1-Basic Mode 6 16384 4096 3 90 L1-Basic Mode 6 16384 3648 4 91 L1-Basic Mode 6 16384 2432 3 92 L1-Basic Mode 6 16384 1536 4 93 L1-Basic Mode 6 16384 1024 6 94 L1-Basic Mode 6 16384 768 8 95 L1-Basic Mode 6 32768 4864 3 96 L1-Basic Mode 6 32768 3648 3 97 L1-Basic Mode 6 32768 3648 8 98 L1-Basic Mode 6 32768 2432 6 99 L1-Basic Mode 6 32768 1536 8 100 L1-Basic Mode 6 32768 1024 12 101 L1-Basic Mode 6 32768 768 16 102 L1-Basic Mode 7 8192 2048 3 103 L1-Basic Mode 7 8192 1536 4 104 L1-Basic Mode 7 8192 1024 3 105 L1-Basic Mode 7 8192 768 4 106 L1-Basic Mode 7 16384 4096 3 107 L1-Basic Mode 7 16384 3648 4 108 L1-Basic Mode 7 16384 2432 3 109 L1-Basic Mode 7 16384 1536 4 110 L1-Basic Mode 7 16384 1024 6 111 L1-Basic Mode 7 16384 768 8 112 L1-Basic Mode 7 32768 4864 3 113 L1-Basic Mode 7 32768 3648 3 114 L1-Basic Mode 7 32768 3648 8 115 L1-Basic Mode 7 32768 2432 6 116 L1-Basic Mode 7 32768 1536 8 117 L1-Basic Mode 7 32768 1024 12 118 L1-Basic Mode 7 32768 768 16 119 Reserved Reserved Reserved Reserved 120 Reserved Reserved Reserved Reserved 121 Reserved Reserved Reserved Reserved 122 Reserved Reserved Reserved Reserved 123 Reserved Reserved Reserved Reserved 124 Reserved Reserved Reserved Reserved 125 Reserved Reserved Reserved Reserved 126 Reserved Reserved Reserved Reserved 127 Reserved Reserved Reserved Reserved

For example, to represent the preamble structure shown in Table 2, a fixed symbol of 7 bits may be allocated. L1-Basic Mode 1, L1-Basic Mode 2 and L1-Basic Mode 3 described in Table 2 above may correspond to QPSK and 3/15 LDPC. The L1-Basic Mode 4 described in Table 2 may correspond to 16-NUC (Non Uniform Constellation) and 3/15 LDPC.

L1-Basic Mode 5 described in Table 2 may correspond to 64-NUC (Non Uniform Constellation) and 3/15 LDPC.

L1-Basic Mode 6 and L1-Basic Mode 7 described in Table 2 may correspond to 256-NUC (Non Uniform Constellation) and 3/15 LDPC. The modulation method/code rate described below represents a combination of modulation method and code rate, such as QPSK and 3/15 LDPC.

The FFT size described in Table 2 may indicate a Fast Fourier Transform size.

The GI length described in Table 2 indicates a guard interval length, and may indicate the length of a guard interval rather than data in the time domain. At this time, the longer the guard interval length, the more robust the system.

The Pilot Pattern described in Table 2 above may indicate Dx of the pilot pattern. Although not explicitly described in Table 2, Dy may all be 1 in the examples described in Table 2. For example, Dx = 3 may mean that one of three pilots for channel estimation is included in the x-axis direction. For example, Dy = 1 may mean that a pilot is included every time in the y-axis direction.

As can be seen from the example of Table 2, the preamble structure corresponding to the second modulation method / code rate, which is more robust than the first modulation method / code rate, is looked up with priority over the preamble structure corresponding to the first modulation method / code rate can be assigned to a table.

In this case, being preferentially assigned may mean being stored corresponding to a smaller number of indexes in the lookup table.

Also, in the case of the same modulation method/code rate, a preamble structure corresponding to a second FFT size smaller than the first FFT size may be preferentially allocated to the lookup table than a preamble structure corresponding to the first FFT size.

Also, in the case of the same modulation method/code rate and FFT size, a preamble structure corresponding to a second guard interval larger than the first guard interval may be assigned to the lookup table with priority over a preamble structure corresponding to the first guard interval.

As described in Table 2, by setting the order in which the preamble structures are allocated in the lookup table, identification of the preamble structure using bootstrap can be performed more efficiently.

FIG. 5 is a diagram illustrating an example of a process in which the broadcast signal frame shown in FIG. 4 is received.

Referring to FIG. 5, a bootstrap 510 is detected and demodulated, and a preamble 520 is demodulated using the demodulated information to restore signaling information.

The core layer data 530 is demodulated using the signaling information, and the enhanced layer signal is demodulated through a cancellation process corresponding to the core layer data. In this case, cancellation corresponding to the core layer data will be described in more detail later.

FIG. 6 is a diagram illustrating another example of a process in which the broadcast signal frame shown in FIG. 4 is received.

Referring to FIG. 6, a bootstrap 610 is detected and demodulated, and a preamble 620 is demodulated using the demodulated information to restore signaling information.

Core layer data 630 is demodulated using signaling information. In this case, the in-band signaling unit 650 is included in the core layer data 630. The in-band signaling unit 650 includes signaling information for enhanced layer service. Through the in-band signaling unit 650, more efficient bandwidth utilization is possible. In this case, the in-band signaling unit 650 is preferably included in a core layer that is stronger than the enhanced layer.

In the example shown in FIG. 6, basic signaling information and information for a core layer service may be delivered through the preamble 620, and signaling information for an enhanced layer service may be delivered through the in-band signaling unit 650. there is.

The enhanced layer signal is demodulated through a cancellation process corresponding to core layer data.

In this case, the signaling information may be L1 (Layer-1) signaling information. The L1 signaling information may include information necessary for configuring physical layer parameters.

Referring to FIG. 4, a broadcast signal frame includes an L1 signaling signal and a data signal. For example, the broadcast signal frame may be an ATSC 3.0 frame.

FIG. 7 is a block diagram illustrating another example of the apparatus for generating a broadcast signal frame shown in FIG. 1 .

Referring to FIG. 7, it can be seen that the apparatus for generating broadcast signal frames multiplexes data corresponding to N (N is a natural number of 1 or more) extension layers in addition to core layer data and enhanced layer data together. .

That is, the apparatus for generating broadcast signal frames shown in FIG. 7 includes a core layer BICM unit 310, an enhanced layer BICM unit 320, an injection level controller 330, a combiner 340, a power normalizer 345, a time In addition to the interleaver 350, the signaling generator 360, and the frame builder 370, N enhancement layer BICM units 410, ..., 430 and injection level controllers 440, ..., 460 are included. .

The core layer BICM unit 310, the enhanced layer BICM unit 320, the injection level controller 330, the combiner 340, the power normalizer 345, the time interleaver 350, and the signaling generator shown in FIG. 7 360 and the frame builder 370 have already been described in detail with reference to FIG. 3 .

The N enhancement layer BICM units 410, ..., 430 independently perform BICM encoding, and the injection level controllers 440, ..., 460 perform power reduction corresponding to each enhancement layer. The power-reduced enhancement layer signal is combined with other layer signals through the combiner 340.

At this time, the error correction encoder of each of the enhancement layer BICM units 410, ..., 430 may be a serial connection of a BCH encoder and an LDPC encoder.

In particular, the power reduction corresponding to each of the injection level controllers 440, ..., 460 is preferably greater than the power reduction of the injection level controller 330. That is, the injection level controllers 330, 440, ..., 460 shown in FIG. 7 may correspond to a large power reduction as they go down.

The injection level information provided from the injection level controllers 330, 440, and 460 shown in FIG. 7 is included in a broadcast signal frame of the frame builder 370 via the signaling generator 360 and transmitted to the receiver. That is, the injection level of each layer is contained in L1 signaling information and delivered to the receiver.

In the present invention, power control may increase or decrease the power of an input signal or increase or decrease the gain of an input signal.

The power normalizer 345 mitigates a power increase caused by combining the plurality of layer signals by the combiner 340 .

In the example shown in FIG. 7 , the power normalizer 345 multiplies the normalizing factor by the magnitude of the signal combined with the signals of each layer using Equation 4 below to adjust the signal power to an appropriate signal magnitude. .

[Equation 4]

Normalizing Factor =

The time interleaver 350 interleaves the signals combined by the combiner 340, thereby interleaving the signals of the layers.

FIG. 8 is a block diagram illustrating an example of the signal demultiplexing device shown in FIG. 1 .

Referring to FIG. 8, a signal demultiplexing apparatus according to an embodiment of the present invention includes a time deinterleaver 510, a de-normalizer 1010, a core layer BICM decoder 520, and an enhanced layer symbol extractor 530. , a de-injection level controller 1020 and an enhanced layer BICM decoder 540.

In this case, the signal demultiplexing device shown in FIG. 8 may correspond to the broadcasting signal frame generating device shown in FIG. 3 .

The time deinterleaver 510 receives a received signal from an OFDM receiver that performs operations such as time/frequency synchronization, channel estimation, and equalization, and detects burst errors generated in the channel. Perform operations related to dispersion. At this time, the L1 signaling information may be decoded first in the OFDM receiver and used for data decoding. In particular, among the L1 signaling information, injection level information may be delivered to the de-normalizer 1010 and the de-injection level controller 1020. At this time, the OFDM receiver may decode the received signal in the form of a broadcast signal frame (eg, ATSC 3.0 frame), extract a data symbol portion of the frame, and provide the extracted data symbol portion to the time deinterleaver 510. That is, the time deinterleaver 510 disperses clustering errors generated in the channel by performing an inverse interleaving process while passing data symbols.

At this time, the time deinterleaver 510 may perform an operation corresponding to the time interleaver. In this case, the time deinterleaver 510 may perform deinterleaving in one of a plurality of operation modes, and may perform deinterleaving using time interleaver information signaled in relation to the operation of the time interleaver.

The de-normalizer 1010 corresponds to the power normalizer of the transmitter and increases the power by the amount reduced by the power normalizer. That is, the de-normalizer 1010 divides the received signal by the normalizing factor of Equation 2 above.

In the example shown in FIG. 8 , the de-normalizer 1010 is shown as adjusting the power of the output signal of the time interleaver 510, but according to the embodiment, the de-normalizer 1010 is the time interleaver 510 It may be located in front of , so that power control is performed before interleaving.

That is, the de-normalizer 1010 may be positioned before or after the time interleaver 510 to amplify the signal amplitude for LLR calculation of the core layer symbol demapper.

An output of the time deinterleaver 510 (or an output of the denormalizer 1010) is provided to the core layer BICM decoder 520, and the core layer BICM decoder 520 restores core layer data.

In this case, the core layer BICM decoder 520 includes a core layer symbol demapper, a core layer bit deinterleaver, and a core layer error correction decoder. The core layer symbol demapper calculates LLR (Log-Likelihood Ratio) values related to symbols, the core layer bit deinterleaver strongly mixes the calculated LLR values with clustering errors, and the core layer error correction decoder calculates errors generated in the channel. Correct.

In this case, the core layer symbol demapper may calculate an LLR value for each bit using a predetermined constellation. In this case, constellations used in the core layer symbol mapper may be different according to a combination of a code rate and a modulation order used in the transmitter.

In this case, the core layer bit deinterleaver may perform inverse interleaving on the calculated LLR values in units of LDPC codewords.

In particular, the core layer error correction decoder may output only information bits or may output all bits in which information bits and parity bits are combined. In this case, the core layer error correction decoder may output only information bits as core layer data, and output all bits obtained by combining the information bits with parity bits to the enhanced layer symbol extractor 530.

The core layer error correction decoder may have a form in which a core layer LDPC decoder and a core layer BCH decoder are serially connected. That is, the input of the core layer error correction decoder is input to the core layer LDPC decoder, the output of the core layer LDPC decoder is input to the core layer BCH decoder, and the output of the core layer BCH decoder is input to the core layer error correction decoder. output can be At this time, the LDPC decoder performs LDPC decoding, and the BCH decoder performs BCH decoding.

Furthermore, the enhanced layer error correction decoder may also have an enhanced layer LDPC decoder and an enhanced layer BCH decoder connected in series. That is, the input of the enhanced layer error correction decoder is input to the enhanced layer LDPC decoder, the output of the enhanced layer LDPC decoder is input to the enhanced layer BCH decoder, and the output of the enhanced layer BCH decoder is input to the enhanced layer BCH decoder. It can be the output of the layer error correction decoder.

The enhanced layer symbol extractor 530 receives all bits from the core layer error correction decoder of the core layer BICM decoder 520 and uses the output signal of the time deinterleaver 510 or the de-normalizer 1010 to enhance the layer. symbols can be extracted. According to an embodiment, the enhanced layer symbol extractor 530 does not receive all bits from the error correction decoder of the core layer BICM decoder 520, but receives information bits of LDPC or BCH information bits. can be provided.

In this case, the enhanced layer symbol extractor 530 includes a buffer, a subtracter, a core layer symbol mapper, and a core layer bit interleaver. The buffer stores an output signal of the time deinterleaver 510 or the de-normalizer 1010. The core layer bit interleaver receives all bits (information bits + parity bits) of the core layer BICM decoder and performs the same core layer bit interleaving as the transmitter. The core layer symbol mapper generates the same core layer symbol as the transmitter from the interleaved signal. The subtractor obtains an enhanced layer symbol by subtracting the output signal of the core layer symbol mapper from the signal stored in the buffer and transfers it to the de-injection level controller 1020 . In particular, when receiving LDPC information bits, the enhanced layer symbol extractor 530 may further include a core layer LDPC encoder. In addition, when receiving BCH information bits, the enhanced layer symbol extractor 530 may further include a core layer BCH encoder as well as a core layer LDPC encoder.

At this time, the core layer LDPC encoder, core layer BCH encoder, core layer bit interleaver, and core layer symbol mapper included in the enhanced layer symbol extractor 530 are the LDPC encoder, BCH encoder, and bit interleaver of the core layer described with reference to FIG. 3 and symbol mapper.

The de-injection level controller 1020 receives the enhanced layer symbol and increases the power by the amount of power dropped by the injection level controller of the transmitter. That is, the de-injection level controller 1020 amplifies the input signal and provides it to the enhanced layer BICM decoder 540 . For example, if the transmitter combines the power of the enhanced layer signal to be 3 dB smaller than the power of the core layer signal, the de-injection level controller 1020 serves to increase the power of the input signal by 3 dB.

In this case, the de-injection level controller 1020 may receive the injection level information from the OFDM receiver and multiply the extracted enhanced layer signal by the enhanced layer gain of Equation 5 below.

[Equation 5]

Enhanced Layer Gain =

The enhanced layer BICM decoder 540 receives the enhanced layer symbol whose power is increased by the de-injection level controller 1020 and restores enhanced layer data.

In this case, the enhanced layer BICM decoder 540 may include an enhanced layer symbol demapper, an enhanced layer bit deinterleaver, and an enhanced layer error correction decoder. The enhanced layer symbol demapper calculates LLR (Log-Likelihood Ratio) values related to the enhanced layer symbol, and the enhanced layer bit deinterleaver strongly mixes the calculated LLR values with clustering errors and performs enhanced layer error correction decoding. The machine corrects errors that have occurred in the channel.

The enhanced layer BICM decoder 540 performs tasks similar to those of the core layer BICM decoder 520, but generally, the enhanced layer LDPC decoder performs LDPC decoding for a code rate of 6/15 or higher.

For example, the core layer may use an LDPC code having a code rate of 5/15 or less, and the enhanced layer may use an LDPC code having a code rate of 6/15 or more. At this time, in a receiving environment in which decoding of enhanced layer data is possible, core layer data can be decoded with only a small number of LDPC decoding iterations. By using this property, the receiver hardware can share one LDPC decoder with the core layer and the enhanced layer, thereby reducing hardware implementation costs. At this time, the core layer LDPC decoder uses only some time resources (LDPC decoding iterations), and the enhanced layer LDPC decoder can use most of the time resources.

The signal demultiplexing apparatus shown in FIG. 8 first restores core layer data, cancels core layer symbols from received signal symbols to leave only enhanced layer symbols, and then increases power of enhanced layer symbols to perform enhancement. Restore layer data. As already described with reference to FIGS. 3 and 5 , since the signals corresponding to each layer are combined at different power levels, data restoration with the least error is possible only when the signal combined with the strongest power is restored.

Eventually, in the example shown in FIG. 8, the signal demultiplexing apparatus includes a time deinterleaver 510 for generating a time deinterleaving signal by applying time deinterleaving to a received signal; a de-normalizer (1010) which increases the power of the received signal or the time deinterleaved signal by the power reduction by the power normalizer of the transmitter; a core layer BICM decoder 520 for restoring core layer data from the signal power-adjusted by the de-normalizer 1010; By using the output signal of the core layer FEC decoder of the core layer BICM decoder 520, cancellation corresponding to the core layer data for the signal power-adjusted by the de-normalizer 1010 is performed, thereby enhancing an enhanced layer symbol extractor 530 for extracting layer signals; a de-injection level controller (1020) that increases the power of the enhanced layer signal by the power reduction of the injection level controller of the transmitter; and an enhanced layer BICM decoder 540 for restoring enhanced layer data using an output signal of the de-injection level controller 1020 .

In this case, the enhanced layer symbol extractor may receive all codewords from the core layer LDPC decoder of the core layer BICM decoder and directly perform bit interleaving of the entire codewords.

In this case, the enhanced layer symbol extractor may receive information bits from the core layer LDPC decoder of the core layer BICM decoder, perform bit interleaving after performing core layer LDPC encoding on the information bits.

In this case, the enhanced layer symbol extractor may receive information bits from the core layer BCH decoder of the core layer BICM decoder, perform bit interleaving after core layer BCH encoding and core layer LDPC encoding of the information bits.

At this time, the de-normalizer and the de-injection level controller may receive injection level information (IL INFO) provided based on L1 signaling and perform power control based on the injection level information.

In this case, the core layer BICM decoder may have a lower bit rate than the enhanced layer BICM decoder and be more robust than the enhanced layer BICM decoder.

In this case, the de-normalizer may correspond to the reciprocal of the normalizing factor.

In this case, the de-injection level controller may correspond to the reciprocal of the scaling factor.

In this case, the enhanced layer data may be restored based on cancellation corresponding to restoration of the core layer data corresponding to the core layer signal.

At this time, the signal demultiplexing apparatus includes at least one enhancement layer symbol extractor extracting an enhancement layer signal by performing cancellation corresponding to previous layer data; One or more de-injection level controllers that increase the power of the enhancement layer signal by a power reduction of the injection level controller of the transmitter, and one or more extensions that restore one or more enhancement layer data using output signals of the one or more de-injection level controllers. A layer BICM decoder may be further included.

A signal demultiplexing method according to an embodiment of the present invention through the configuration shown in FIG. 8 includes generating a time deinterleaving signal by applying time deinterleaving to a received signal; increasing the power of the received signal or the time deinterleaving signal by a power reduction by a power normalizer of a transmitter; restoring core layer data from the power-adjusted signal; extracting an enhanced layer signal by performing cancellation corresponding to the core layer data on the power-adjusted signal; increasing the power of the enhanced layer signal by a power reduction of an injection level controller of a transmitter; and restoring enhanced layer data using the power-adjusted enhanced layer signal.

In this case, in the step of extracting the enhanced layer signal, entire codewords may be received from the core layer LDPC decoder of the core layer BICM decoder, and the entire codeword may be directly bit-interleaved.

In this case, the step of extracting the enhanced layer signal may receive information bits from the core layer LDPC decoder of the core layer BICM decoder, perform core layer LDPC encoding on the information bits, and then perform bit interleaving.

At this time, the step of extracting the enhanced layer signal receives information bits from the core layer BCH decoder of the core layer BICM decoder, performs core layer BCH encoding and core layer LDPC encoding on the information bits, and then performs bit interleaving. .

FIG. 9 is a block diagram illustrating an example of the core layer BICM decoder 520 and the enhanced layer symbol extractor 530 shown in FIG. 8 .

Referring to FIG. 9 , the core layer BICM decoder 520 includes a core layer symbol demapper, a core layer bit deinterleaver, a core layer LDPC decoder, and a core layer BCH decoder.

That is, in the example shown in FIG. 9, the core layer error correction decoder includes a core layer LDPC decoder and a core layer BCH decoder.

In addition, in the example shown in FIG. 9 , the core layer LDPC decoder provides a whole codeword including parity bits to the enhanced layer symbol extractor 530 . That is, generally, an LDPC decoder outputs only information bits among all LDPC codewords, but it is also possible to output all codewords.

In this case, the enhanced layer symbol extractor 530 does not need to separately have a core layer LDPC encoder or a core layer BCH encoder, so implementation is simple, but there is a possibility that residual errors remain in the LDPC code parity part.

FIG. 10 is a block diagram illustrating another example of the core layer BICM decoder 520 and the enhanced layer symbol extractor 530 shown in FIG. 8 .

Referring to FIG. 10 , the core layer BICM decoder 520 includes a core layer symbol demapper, a core layer bit deinterleaver, a core layer LDPC decoder, and a core layer BCH decoder.

That is, in the example shown in FIG. 10, the core layer error correction decoder includes a core layer LDPC decoder and a core layer BCH decoder.

In addition, in the example shown in FIG. 10, the core layer LDPC decoder provides information bits that do not include parity bits to the enhanced layer symbol extractor 530.

In this case, the enhanced layer symbol extractor 530 does not need to include a core layer BCH encoder, but must include a core layer LDPC encoder.

The example shown in FIG. 10 can remove residual errors that may remain in the LDPC code parity part compared to the example shown in FIG. 9 .

FIG. 11 is a block diagram showing another example of the core layer BICM decoder 520 and the enhanced layer symbol extractor 530 shown in FIG. 8 .

Referring to FIG. 11 , the core layer BICM decoder 520 includes a core layer symbol demapper, a core layer bit deinterleaver, a core layer LDPC decoder, and a core layer BCH decoder.

That is, in the example shown in FIG. 11, the core layer error correction decoder includes a core layer LDPC decoder and a core layer BCH decoder.

In the example shown in FIG. 11, the output of the core layer BCH decoder corresponding to core layer data is provided to the enhanced layer symbol extractor 530.

In this case, the enhanced layer symbol extractor 530 has high complexity because it must include both the core layer LDPC encoder and the core layer BCH encoder, but guarantees the highest performance compared to the examples of FIGS. 9 and 10 .

FIG. 12 is a block diagram illustrating another example of the signal demultiplexing device shown in FIG. 1 .

Referring to FIG. 12, a signal demultiplexing apparatus according to an embodiment of the present invention includes a time deinterleaver 510, a de-normalizer 1010, a core layer BICM decoder 520, and an enhanced layer symbol extractor 530. , an enhanced layer BICM decoder 540, one or more enhancement layer symbol extractors 650, 670, one or more enhancement layer BICM decoders 660, 680 and de-injection level controllers 1020, 1150, 1170. include

In this case, the signal demultiplexing device shown in FIG. 12 may correspond to the broadcasting signal frame generating device shown in FIG. 7 .

The time deinterleaver 510 receives a received signal from an OFDM receiver that performs operations such as synchronization, channel estimation, and equalization, and relates to dispersion of burst errors generated in the channel. perform the action At this time, the L1 signaling information may be decoded first in the OFDM receiver and used for data decoding. In particular, among the L1 signaling information, injection level information may be transmitted to the de-normalizer 1010 and the de-injection level controllers 1020, 1150, and 1170.

At this time, the de-normalizer 1010 obtains injection level information of all layers, obtains a de-normalizing factor using Equation 6 below, and then multiplies the input signal.

[Equation 6]

De-Normalizing factor = (Normalizing factor) ^-1 =

That is, the de-normalizing factor is the reciprocal of the normalizing factor expressed by Equation 4 above.

According to an embodiment, when N1 signaling includes injection level information as well as normalizing factor information, the de-normalizer 1010 simply takes the reciprocal of the normalizing factor and simply de-normalizes it without the need to calculate the de-normalizing factor using the injection level. A normalizing factor can be obtained.

The de-normalizer 1010 corresponds to the power normalizer of the transmitter and increases the power by the amount reduced by the power normalizer.

In the example shown in FIG. 12, the de-normalizer 1010 is shown as adjusting the power of the output signal of the time interleaver 510, but according to the embodiment, the de-normalizer 1010 is the time interleaver 510 It may be located in front of , so that power control is performed before interleaving.

That is, the de-normalizer 1010 may be positioned before or after the time interleaver 510 to amplify the signal amplitude for LLR calculation of the core layer symbol demapper.

An output of the time deinterleaver 510 (or an output of the denormalizer 1010) is provided to the core layer BICM decoder 520, and the core layer BICM decoder 520 restores core layer data.

In this case, the core layer BICM decoder 520 includes a core layer symbol demapper, a core layer bit deinterleaver, and a core layer error correction decoder. The core layer symbol demapper calculates LLR (Log-Likelihood Ratio) values related to symbols, the core layer bit deinterleaver strongly mixes the calculated LLR values with clustering errors, and the core layer error correction decoder calculates errors generated in the channel. Correct.

In particular, the core layer error correction decoder may output only information bits or may output all bits in which information bits and parity bits are combined. In this case, the core layer error correction decoder may output only information bits as core layer data, and output all bits obtained by combining the information bits with parity bits to the enhanced layer symbol extractor 530.

The core layer error correction decoder may have a form in which a core layer LDPC decoder and a core layer BCH decoder are serially connected. That is, the input of the core layer error correction decoder is input to the core layer LDPC decoder, the output of the core layer LDPC decoder is input to the core layer BCH decoder, and the output of the core layer BCH decoder is input to the core layer error correction decoder. output can be At this time, the LDPC decoder performs LDPC decoding, and the BCH decoder performs BCH decoding.

An enhanced layer error correction decoder may also have a form in which an enhanced layer LDPC decoder and an enhanced layer BCH decoder are serially connected. That is, the input of the enhanced layer error correction decoder is input to the enhanced layer LDPC decoder, the output of the enhanced layer LDPC decoder is input to the enhanced layer BCH decoder, and the output of the enhanced layer BCH decoder is input to the enhanced layer BCH decoder. It can be the output of the layer error correction decoder.

Furthermore, the enhancement layer error correction decoder may also have an enhancement layer LDPC decoder and an enhancement layer BCH decoder connected in series. That is, the input of the enhancement layer error correction decoder is input to the enhancement layer LDPC decoder, the output of the enhancement layer LDPC decoder is input to the enhancement layer BCH decoder, and the output of the enhancement layer BCH decoder is input to the enhancement layer error correction decoder. output can be

In particular, the trade-off between implementation complexity and performance according to which of the outputs of the error correction decoder described with reference to FIGS. 9, 10, and 11 is to be used is the core layer BICM decoder 520 of FIG. and the enhanced layer symbol extractor 530, as well as the enhancement layer symbol extractors 650 and 670 and the enhancement layer BICM decoders 660 and 680.

The enhanced layer symbol extractor 530 receives all bits from the core layer error correction decoder of the core layer BICM decoder 520 and uses the output signal of the time deinterleaver 510 or the de-normalizer 1010 to enhance the layer. symbols can be extracted. According to an embodiment, the enhanced layer symbol extractor 530 does not receive all bits from the error correction decoder of the core layer BICM decoder 520, but receives information bits of LDPC or BCH information bits. can be provided.

In this case, the enhanced layer symbol extractor 530 includes a buffer, a subtracter, a core layer symbol mapper, and a core layer bit interleaver. The buffer stores an output signal of the time deinterleaver 510 or the de-normalizer 1010. The core layer bit interleaver receives all bits (information bits + parity bits) of the core layer BICM decoder and performs the same core layer bit interleaving as the transmitter. The core layer symbol mapper generates the same core layer symbol as the transmitter from the interleaved signal. The subtractor obtains an enhanced layer symbol by subtracting the output signal of the core layer symbol mapper from the signal stored in the buffer and transfers it to the de-injection level controller 1020 .

In this case, the core layer bit interleaver and the core layer symbol mapper included in the enhanced layer symbol extractor 530 may be the same as the core layer bit interleaver and symbol mapper shown in FIG. 7 .

The de-injection level controller 1020 receives the enhanced layer symbol and increases the power by the amount of power dropped by the injection level controller of the transmitter. That is, the de-injection level controller 1020 amplifies the input signal and provides it to the enhanced layer BICM decoder 540 .

The enhanced layer BICM decoder 540 receives the enhanced layer symbol whose power is increased by the de-injection level controller 1020 and restores enhanced layer data.

In this case, the enhanced layer BICM decoder 540 may include an enhanced layer symbol demapper, an enhanced layer bit deinterleaver, and an enhanced layer error correction decoder. The enhanced layer symbol demapper calculates LLR (Log-Likelihood Ratio) values related to the enhanced layer symbol, and the enhanced layer bit deinterleaver strongly mixes the calculated LLR values with clustering errors and performs enhanced layer error correction decoding. The machine corrects errors that have occurred in the channel.

In particular, the enhanced layer error correction decoder may output only information bits or may output all bits in which information bits and parity bits are combined. In this case, the enhanced layer error correction decoder may output only information bits as enhanced layer data and output all bits obtained by combining parity bits with information bits to the enhancement layer symbol extractor 650.

The extension layer symbol extractor 650 receives all bits from the enhanced layer error correction decoder of the enhanced layer BICM decoder 540 and extracts extension layer symbols from the output signal of the de-injection level controller 1020. do.

In this case, the de-injection level controller 1020 may amplify the power of the output signal of the subtractor of the enhanced layer symbol extractor 530.

At this time, the enhancement layer symbol extractor 650 includes a buffer, a subtracter, an enhanced layer symbol mapper, and an enhanced layer bit interleaver. The buffer stores the output signal of the de-injection level controller 1020. The enhanced layer bit interleaver receives all bits (information bits + parity bits) of the enhanced layer BICM decoder and performs the same enhanced layer bit interleaving as the transmitter. The enhanced layer symbol mapper generates the same enhanced layer symbol as the transmitter from the interleaved signal. The subtractor obtains an enhancement layer symbol by subtracting the output signal of the enhanced layer symbol mapper from the signal stored in the buffer and transfers it to the de-injection level controller 1150.

In this case, the enhanced layer bit interleaver and the enhanced layer symbol mapper included in the enhancement layer symbol extractor 650 may be the same as the enhanced layer bit interleaver and symbol mapper shown in FIG. 7 .

The de-injection level controller 1150 increases power by the amount reduced by the injection level controller of the corresponding layer in the transmitter.

At this time, the de-injection level controller can be regarded as performing an operation of multiplying the enhancement layer gain of Equation 7 below. At this time, the 0th injection level may be regarded as 0dB.

[Equation 7]

n-th Extension Layer Gain =

The enhancement layer BICM decoder 660 receives the enhancement layer symbol whose power is increased by the de-injection level controller 1150 and restores enhancement layer data.

In this case, the enhancement layer BICM decoder 660 may include an enhancement layer symbol demapper, an enhancement layer bit deinterleaver, and an enhancement layer error correction decoder. The extension layer symbol demapper calculates log-likelihood ratio (LLR) values related to the extension layer symbols, the extension layer bit deinterleaver strongly mixes the calculated LLR values with clustering errors, and the extension layer error correction decoder Correct errors.

In particular, two or more enhancement layer symbol extractors and two or more enhancement layer BICM decoders may be provided when there are two or more enhancement layers.

That is, in the example shown in FIG. 12, the enhancement layer error correction decoder of the enhancement layer BICM decoder 660 may output only information bits or output all bits in which information bits and parity bits are combined. may be At this time, the enhancement layer error correction decoder may output only information bits as enhancement layer data, and output all bits obtained by combining information bits with parity bits to the next enhancement layer symbol extractor 670.

The structure and operation of the enhancement layer symbol extractor 670, the enhancement layer BICM decoder 680, and the de-injection level controller 1170 are the same as the enhancement layer symbol extractor 650, the enhancement layer BICM decoder 660, and the de-injection level controller 1170. It can be readily seen from the structure and operation of the level controller 1150.

The de-injection level controllers 1020, 1150, and 1170 shown in FIG. 12 may correspond to a greater power increase as they go down. That is, the de-injection level controller 1150 increases the power more than the de-injection level controller 1020, and the de-injection level controller 1170 increases the power more than the de-injection level controller 1150. can make it

The signal demultiplexing apparatus shown in FIG. 12 first restores core layer data, restores enhanced layer data using cancellation of core layer symbols, and restores enhancement layer data using cancellation of enhanced layer symbols. know to restore. Two or more enhancement layers may be provided, and in this case, an enhancement layer combined with a higher power level is restored.

13 is a diagram illustrating power increase due to a combination of a core layer signal and an enhanced layer signal.

Referring to FIG. 13 , when a multiplexed signal is generated by combining an enhanced layer signal whose power is reduced by an injection level with a core layer signal, the power level of the multiplexed signal is different from that of the core layer signal or the enhanced layer signal. It can be seen that it is higher than the power level.

At this time, the injection level controlled by the injection level controller shown in FIGS. 3 and 7 may be adjusted from 0 dB to 25.0 dB at 0.5 dB or 1 dB intervals. When the injection level is 3.0 dB, the power of the enhanced layer signal is lower than the power of the core layer signal by 3 dB. When the injection level is 10.0 dB, the power of the enhanced layer signal is lower than the power of the core layer signal by 10 dB. This relationship may be applied not only between the core layer signal and the enhanced layer signal, but also between the enhanced layer signal and the enhancement layer signal or enhancement layer signals.

The power normalizer shown in FIGS. 3 and 7 can solve problems such as signal distortion that may be caused by an increase in power due to coupling by adjusting the power level after coupling.

14 is an operational flowchart illustrating a method for generating a broadcast signal frame according to an embodiment of the present invention.

Referring to FIG. 14, in the broadcast signal frame generation method according to an embodiment of the present invention, BICM is applied to core layer data (S1210).

Also, in the broadcast signal frame generating method according to an embodiment of the present invention, BICM is applied to enhanced layer data (S1220).

The BICM applied in step S1220 and the BICM applied in step S1210 may be different. At this time, the BICM applied in step S1220 may be less robust than the BICM applied in step S1210. At this time, the bit rate of BICM applied in step S1220 may be greater than the bit rate applied in step S1210.

In this case, the enhanced layer signal may correspond to enhanced layer data that is restored based on cancellation corresponding to restoration of core layer data corresponding to the core layer signal.

Also, in the broadcast signal frame generating method according to an embodiment of the present invention, a power reduced enhanced layer signal is generated by reducing the power of the enhanced layer signal (S1230).

At this time, step S1230 may change the injection level between 0dB and 25.0dB at 0.5dB or 1dB intervals.

In addition, in the broadcast signal frame generation method according to an embodiment of the present invention, a multiplexed signal is generated by combining a core layer signal and a power reduced enhanced layer signal (S1240).

In this case, in step S1240, the core layer signal and the enhanced layer signal may be combined at different power levels, but the power level of the enhanced layer signal may be lower than the power level of the core layer signal.

In this case, in step S1240, one or more extension layer signals having a lower power level than the core layer signal and the enhanced layer signal may be combined with the core layer signal and the enhanced layer signal.

In addition, in the method for generating broadcast signal frames according to an embodiment of the present invention, power normalization is performed to lower the power of the multiplexed signal in step S1250 (S1250).

At this time, in step S1250, the power of the multiplexed signal may be lowered by the power of the core layer signal. At this time, in step S1250, the power of the multiplexed signal may be lowered by the amount increased by step S1240.

Also, in the broadcast signal frame generation method according to an embodiment of the present invention, time interleaving is performed to generate a time interleaved signal (S1260).

Depending on the embodiment, in the case of a single layer, in step S1260, a time interleaved signal may be generated by performing time interleaving on the BICM output signal.

At this time, step S1260 uses one of the time interleaver groups, and the boundary between the time interleaver groups is Physical Layer Pipes (PLPs) of the core layer corresponding to the core layer signal. ) may be the boundary between

At this time, in step S1260, the interleaving may be performed using a hybrid time interleaver. In this case, the physical layer pipes of the core layer and the enhanced layer may include only complete FEC blocks.

At this time, in step S1260, the interleaving is performed using a convolutional time interleaver, and the time interleaver groups include physical layer pipes including incomplete FEC blocks. Layer Pipe (PLP), and the preamble may signal start position information of a first complete FEC block in the physical layer pipe.

At this time, step S1260 may be performed using one of a plurality of operation modes.

In this case, the operating modes may include a first mode in which time interleaving is omitted, a second mode in which convolutional time interleaving is performed, and a third mode in which hybrid time interleaving is performed. can

In this case, the operation mode may correspond to a time interleaving mode. In this case, a time interleaving mode corresponding to time interleaving may be signaled for each of all physical layer pipes. In this case, the time interleaving mode may be included in the preamble.

In addition, the method for generating a broadcast signal frame according to an embodiment of the present invention generates a broadcast signal frame including a preamble for signaling a time interleaving mode corresponding to the time interleaving for each of all physical layer pipes (S1270). .

In this case, the physical layer pipes include a plurality of core layer physical layer pipes corresponding to one complete delivered product, and the core layer physical layer pipes may not be layered division multiplexed.

In this case, each of the core layer and physical layer pipes may use either a time interleaving mode of a no time interleaving mode or a hybrid time interleaving mode, and may not use a convolutional time interleaving mode.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, all of the core layer physical layer pipes use intra-subframe interleaving mode or all use inter-subframe interleaving mode. can

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes, the core layer physical layer pipes may all have the same L1D_plp_HTI_inter_subframe value.

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes are hybrid time interleaving modes and all of the core layer physical layer pipes use the inter-subframe interleaving mode, the core layer physical layer pipes have the same time interleaving mode. An interleaving unit (N _IU ) may be used.

In this case, when at least one of the time interleaving modes corresponding to the core layer physical layer pipes is the no time interleaving mode, all core layer pipes configured in the hybrid time interleaving mode among the core layer pipes are in the intra-subframe interleaving mode. can be used.

At this time, one complete transmission product corresponds to one or more subframes, in which all available data cells of the subframe are first filled with dummy modulation values, and then the real physical layer pipe data is overwritten. can be created

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

In this case, the physical layer pipes may include one enhanced layer physical layer pipe and a plurality of core layer physical layer pipes layered division multiplexed to the one enhanced layer physical layer pipe.

In this case, the time interleaving mode corresponding to the enhanced layer physical layer pipe may be the same as the time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipe is layered division multiplexed.

In this case, time interleaving modes corresponding to core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed may all be no time interleaving modes or all hybrid time interleaving modes.

In this case, when the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are all hybrid time interleaving modes, the core layer physical layer pipes all use the intra-subframe interleaving mode. can be used

In this case, when all of the time interleaving modes corresponding to the core layer physical layer pipes in which the enhanced layer physical layer pipes are layered division multiplexed are no time interleaving modes, each of the core layer physical layer pipes is in each subframe. It may consist of an integer number of FEC blocks.

At this time, the subframe may be created by first filling all available data cells of the subframe with dummy modulation values, and then overwriting the actual physical layer pipe data.

In this case, the dummy modulation values are generated using a scrambling sequence generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can

In this case, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases having a 180 degree phase difference.

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

In this case, time interleaver information may be signaled based on the core layer.

In this case, the preamble is information for identifying a part of the FEC block of the enhanced layer corresponding to the boundary of the time interleaver groups when the boundary of the time interleaver groups does not correspond to the boundary of the FEC blocks of the enhanced layer. can be signaled.

In this case, the information for identifying a part of the FEC block includes start position information of the physical layer pipe of the core layer, start position information of the physical layer pipe of the enhanced layer, modulation information corresponding to the enhanced layer, and the enhancement It may include any one or more of FEC type information corresponding to the hard layer.

In this case, the start position information of the physical layer pipe may correspond to the index of the first data cell of the physical layer pipe.

In this case, the modulation information may be signaled only when the FEC type information satisfies a preset condition.

In this case, the enhanced layer signal may correspond to enhanced layer data that is restored based on cancellation corresponding to restoration of core layer data corresponding to the core layer signal.

At this time, step (S1270) is the step of generating the bootstrap; generating the preamble; and generating a super-imposed payload corresponding to the time-interleaved signal.

At this time, the preamble includes PLP identification information for identifying Physical Layer Pipes (PLPs); and layer identification information for identifying layers corresponding to the hierarchical division.

In this case, PLP identification information and layer identification information may be included in the preamble as separate fields.

In this case, time interleaver information may be selectively included in the preamble according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes (IF(j>0)).

At this time, the preamble may selectively include injection level information corresponding to the injection level controller according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes (IF(j>0)). can

In this case, the bootstrap may be shorter than the preamble and have a fixed length.

In this case, the bootstrap includes a symbol representing the structure of the preamble, and the symbol may correspond to a fixed bit stream representing a combination of a modulation method/code rate, an FFT size, a guard interval length, and a pilot pattern of the preamble. .

At this time, when the modulation method/code rate is the same, the preamble structure corresponding to the second FFT size smaller than the first FFT size is preferentially allocated to the symbol over the preamble structure corresponding to the first FFT size, and the modulation When the method/code rate and the FFT size are the same, a preamble structure corresponding to a second guard interval length greater than the first guard interval length corresponds to a lookup table in which a preamble structure corresponding to the first guard interval length is preferentially allocated it may be

In this case, the broadcast signal frame may be an ATSC 3.0 frame.

In this case, the L1 signaling information may include injection level information and/or normalizing factor information.

In this case, the preamble may include type information, start position information, and size information of physical layer pipes.

In this case, the type information may be for identifying one of a first type corresponding to a non-dispersed physical layer pipe and a second type corresponding to a dispersed physical layer pipe.

In this case, non-distributed physical layer pipes are allocated for contiguous data cell indices, and the distributed physical layer pipes may include two or more subslices.

In this case, type information may be selectively signaled according to a result of comparing the layer identification information with a preset value for each of the physical layer pipes.

In this case, type information may be signaled only for the core layer.

In this case, the start position information may be set equal to an index corresponding to the first data cell of the physical layer pipe.

In this case, the start position information may indicate a start position of the physical layer pipe using a cell addressing scheme.

In this case, start position information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, size information may be set based on the number of data cells allocated to the physical layer pipe.

In this case, size information may be included in the preamble for each of the physical layer pipes without determining a condition of a conditional statement corresponding to the layer identification information.

In this case, the preamble includes a field indicating a start position of a first complete FEC block corresponding to a current physical layer pipe for the first mode and the second mode, and For mode 3, the field indicating the start position of the first FEC block may not be included.

In this case, the field indicating the start position of the first FEC block is either a first field used in the first mode or a second field used in the second mode, and the first field and the second field are lengths. may be different.

In this case, the length of the second field may be longer than that of the first field.

In this case, the length of the first field is determined based on the length and modulation order of the LDPC codeword, and the length of the second field is determined based on the length and modulation order of the LDPC codeword as well as the depth of the convolutional time interleaver ( depth) may be further considered.

In this case, the length of the first field may be 15 bits, and the length of the second field may be 22 bits.

In this case, the first field and the second field may be separately signaled for each of the core layer corresponding to the core layer signal and the enhanced layer corresponding to the enhanced layer signal.

Although not explicitly shown in FIG. 14, the method of generating a broadcast signal frame may further include generating signaling information including injection level information corresponding to step S1230. In this case, the signaling information may be L1 signaling information.

The broadcast signal frame generating method shown in FIG. 14 may correspond to step S210 shown in FIG. 2 .

Although not explicitly shown in FIG. 14 , the broadcast signal frame generating method may further include inserting dummy values into enhanced layer data between steps S1220 and S1230.

At this time, dummy values are applied after the actual data cells of the last enhanced PLP in the PLP group so that the total number of enhanced layer cells in the PLP group is equal to the total number of core layer cells in the PLP group. of the last Enhanced PLP) can be inserted.

In this case, dummy values may not be inserted into core layer data.

In this case, dummy values may be inserted after completing the core layer BICM and the enhanced layer BICM and before combining the core layer signal and the enhanced layer signal.

In this case, the dummy values may correspond to a preset scrambling sequence.

At this time, the scrambling sequence may be modulated using the constellation mapping used for the last enhanced PLP.

In this case, dummy values may have the same power as the last enhanced PLP.

In this case, the scrambling sequence may be generated using a 16-bit shift register corresponding to a preset generator polynomial.

In this case, the scrambling sequence may be generated using a generator polynomial corresponding to 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ .

At this time, the scrambling sequence is the third bit output (x ¹⁴ ), the fourth bit output (x ¹³ ), the fifth bit output (x ¹² ), the sixth bit output (x ¹¹ ), It can be generated using 8 bits generated using the 10th bit output (x ⁷ ), 13th bit output (x ⁴ ), 14th bit output (x ³ ) and 16th bit output (x).

15 is a diagram showing a super-frame structure including broadcast signal frames according to an embodiment of the present invention.

Referring to FIG. 15 , it can be seen that a superframe based on Layered Division Multiplexing (LDM) is composed of one or more frames, and one frame is composed of one or more OFDM symbols.

At this time, each OFDM symbol may start with one or more preamble symbols. Also, a frame may include a reference symbol or a pilot symbol.

The superframe 1510 shown in FIG. 15 includes an LDM frame 1520, a single-layer frame 1530 without LDM, and a future extension frame for future extensibility ( Including a Future Extension Frame (FEF) 1540 and the like, it may be configured in a Time Division Multiplexing (TDM) method.

The LDM frame 1520 may include an upper layer (UL) 1553 and a lower layer (LL) 1555 when two layers are applied.

In this case, the upper layer 1553 may correspond to the core layer, and the lower layer 1555 may correspond to the enhanced layer.

In this case, the LDM frame 1520 including the upper layer 1553 and the lower layer 1555 may include a bootstrap 1552 and a preamble 1551.

In this case, the data of the upper layer 1553 and the data of the lower layer 1555 may share a time interleaver and use the same frame length and FFT size in order to reduce complexity and memory size.

In addition, the single-layer frame 1530 may also include a bootstrap 1562 and a preamble 1561 .

In this case, the single-layer frame 1530 may use a different FFT size, time interleaver, and frame length than the LDM frame 1520. In this case, the single-layer frame 1530 can be viewed as being multiplexed with the LDM frame 1520 in the superframe 1510 in a TDM manner.

16 is a diagram illustrating an example of an LDM frame to which LDM using two layers and a multiple-physical layer pipe (PLP) are applied.

Referring to FIG. 16, it can be seen that the LDM frame starts with a bootstrap signal including system version information or general signaling information. After bootstrapping, an L1 signaling signal including code rate, modulation information, and number information of physical layer pipes may follow as a preamble.

Following the preamble (L1 SIGNAL), a common Physical Layer Pipe (PLP) in the form of a burst may be transmitted. In this case, the common physical layer pipe may transmit data that can be shared with other physical layer pipes within the frame.

After the common physical layer pipe, a multiple-physical layer pipe for servicing different broadcast signals is transmitted in an LDM scheme of two layers. At this time, services (720p or 1080p HD, etc.) requiring robust reception, such as indoor/mobile, pass through core layer (upper layer) data physical layer pipes, and fixed reception services (4K-UHD) requiring high transmission rates or multiple HD) can be transmitted through enhanced layer (lower layer) data physical layer pipes.

When multiple-physical layer pipes are layered division multiplexed, it can be seen that the total number of multiple-physical layer pipes increases as a result.

In this case, the core layer data physical layer pipe and the enhanced layer data physical layer pipe may share a time interleaver to reduce complexity and memory size. In this case, the core layer data physical layer pipe and the enhanced layer data physical layer pipe may have the same physical layer pipe size (PLP size) or different physical layer pipe sizes.

Depending on the embodiment, physical layer pipes divided into layers may have different PLP sizes, and in this case, information for identifying a PLP start position or PLP size may be signaled.

17 is a diagram illustrating another example of an LDM frame to which LDM using two layers and a multiple-physical layer pipe (PLP) are applied.

Referring to FIG. 17 , it can be seen that the LDM frame may include a common physical layer pipe after bootstrap and preamble (L1 SIGNAL). After the common physical layer pipe, core layer data physical layer pipes and enhanced layer data physical layer pipes can be transmitted in a 2-layer LDM scheme.

In particular, the core layer data physical layer pipes and enhanced layer data physical layer pipes shown in FIG. 17 may have one of type 1 and type 2, and type 1 and type 2 may be defined as follows. there is.

- Type 1 PLP

If a common PLP exists, it is transmitted after the common PLP

Transmitted in one burst form (one slice) within a frame

-Type 2 PLP

If Type 1 PLP exists, transmitted after Type 1 PLP

Distributed and transmitted in the form of two or more sub-slices within a frame

As the number of sub-slices increases, time diversity increases, which has the effect of power consumption.

In this case, the type 1 PLP may correspond to a nun-dispersed PLP, and the type 2 PLP may correspond to a dispersed PLP. In this case, the non-distributed PLPs may be allocated to contiguous data cell indices. In this case, the distributed PLPs may be divided and allocated to two or more subslices.

18 is a diagram showing an example of utilization of an LDM frame to which LDM using two layers and a multiple-physical layer pipe (PLP) are applied.

18, the LDM frame may include a common physical layer pipe (PLP(1,1)) after bootstrap and preamble, and a data physical layer pipe (PLP(2,1) for robust audio service) ) may be included in a time-division method.

In addition, core layer data physical layer pipe (PLP(3,1)) for mobile/indoor service (720p or 1080p HD) and enhanced layer data physical layer for high data rate service (4K-UHD or multiple HD) A pipe (PLP(3,2)) may be transmitted in a 2-layer LDM scheme.

19 is a diagram illustrating another utilization example of an LDM frame to which LDM using two layers and a multiple-physical layer pipe are applied.

Referring to FIG. 19, an LDM frame may include a bootstrap, a preamble, and a common physical layer pipe (PLP(1,1)). At this time, the robust audio service and mobile/indoor service (720p or 1080p HD) are divided into core layer data and physical layer pipes (PLP(2,1), PLP(3,1)) and transmitted, and high data rate A service (4K-UHD or multiple HD) can be transmitted by enhanced layer data physical layer pipes (PLP(2,2), PLP(3,2)).

In this case, the core layer data physical layer pipe and the enhanced layer data physical layer pipe may use the same time interleaver.

In this case, the physical layer pipes (PLP(2,2) and PLP(3,2)) providing the same service may signal that they provide the same service by using PLP_GROUP_ID indicating the same PLP group.

Depending on the embodiment, when physical layer pipes of different sizes are used for each LDM layer, a service may be identified according to the starting position and size of each of the physical layer pipes without PLP_GROUP_ID.

18 and 19 exemplify the case of identifying multiple physical layer pipes and layers corresponding to layered division multiplexing by PLP(i,j), but PLP identification information and layer identification information may be signaled as separate fields. may be

Depending on embodiments, PLPs of different sizes may be used for each layer. In this case, each service may be identified through a PLP identifier.

When PLPs having different sizes are used for each layer, a PLP start position and a PLP length may be signaled for each PLP.

The pseudo code below shows an example of fields included in a preamble according to an embodiment of the present invention. In this case, the pseudo code below may be included in the L1 signaling information of the preamble.

[water supply code]

SUB_SLICES_PER_FRAME (15 bits)

NUM_PLP (8 bits)

NUM_AUX (4 bits)

AUX_CONFIG_RFU (8 bits)

for i=0.. NUM_RF-1 {

RF_IDX (3 bits)

FREQUENCY (32 bits)

}

IF S2=='xxx1' {

FEF_TYPE (4 bits)

FEF_LENGTH (22 bits)

FEF_INTERVAL (8 bits)

}

for i=0 .. NUM_PLP-1 {

NUM_LAYER (2~3 bits)

for j=0 .. NUM_LAYER-1{

/* Signaling for each layer */

PLP_ID (i, j) (8 bits)

PLP_GROUP_ID (8 bits)

PLP_TYPE (3 bits)

PLP_PAYLOAD_TYPE (5 bits)

PLP_COD (4 bits)

PLP_MOD (3 bits)

PLP_SSD (1 bit)

PLP_FEC_TYPE (2 bits)

PLP_NUM_BLOCKS_MAX (10 bits)

IN_BAND_A_FLAG (1 bit)

IN_BAND_B_FLAG (1 bit)

PLP_MODE (2 bits)

STATIC_PADDING_FLAG (1 bit)

IF (j > 0)

LL_INJECTION_LEVEL (3~8 bits)

} /* End of NUM_LAYER loop */

/* Common signaling for all layers */

FF_FLAG (1 bit)

FIRST_RF_IDX (3 bits)

FIRST_FRAME_IDX (8 bits)

FRAME_INTERVAL (8 bits)

TIME_IL_LENGTH (8 bits)

TIME_IL_TYPE (1 bit)

RESERVED_1 (11 bits)

STATIC_FLAG (1 bit)

PLP_START (24 bits)

PLP_SIZE (24 bits)

} /* End of NUM_PLP loop */

FEF_LENGTH_MSB (2 bits)

RESERVED_2 (30 bits)

for i=0 .. NUM_AUX-1 {

AUX_STREAM_TYPE (4 bits)

AUX_PRIVATE_CONF (28 bits)

}

In the above pseudo code, NUM_LAYER may consist of 2 bits or 3 bits. In this case, NUM_LAYER may be a field used to indicate the number of layers within each temporally divided PLP. In this case, NUM_LAYER may have a different number of layers for each temporally divided PLP defined within the NUM_PLP loop.

In the pseudo code, LL_INJECTION_LEVEL may consist of 3 to 8 bits. In this case, LL_INJECTION_LEVEL may be a field for defining an injection level of a lower layer (enhanced layer). At this time, LL_INJECTION_LEVEL may correspond to injection level information.

In this case, LL_INJECTION_LEVEL may be defined from the second layer (j>0) when there are two or more layers.

Fields such as PLP_ID(i,j), PLP_GROUP_ID, PLP_TYPE, PLP_PAYLOAD_TYPE, PLP_COD, PLP_MOD, PLP_SSD, PLP_FEC_TYPE, PLP_NUM_BLOCKS_MAX, IN_BAND_A_FLAG, IN_BAND_B_FLAG, PLP_MODE, and STATIC_PADDING_FLAG are parameters defined for each layer and are defined within the NUM_LAYER loop. can

In this case, PLP_ID (i, j) may correspond to PLP identification information and layer identification information. For example, i of PLP_ID(i,j) may correspond to PLP identification information and j may correspond to layer identification information.

Depending on embodiments, PLP identification information and layer identification information may be included in the preamble as separate fields.

In addition, fields such as FRAME_INTERVAL, FF_FLAG, FIRST_RF_IDX, FIRST_FRAME_IDX, RESERVED_1, and STATIC_FLAG related to time interleaver information such as TIME_IL_LENGTH or TIME_IL_TYPE or PLP size may be defined outside the NUM_LAYER loop or inside the NUM_PLP loop.

In particular, PLP_TYPE indicates type information of the aforementioned physical layer pipes, and may be composed of 1 bit since it only needs to identify two types, the first type and the second type. In the pseudocode, an example in which PLP_TYPE is included in the preamble without determining a conditional sentence corresponding to layer identification information (j) has been described, but PLP_TYPE is a result of comparing layer identification information (j) with a preset value (0) (if (j = 0)) may be selectively signaled (transmitted only for the core layer).

In the above pseudo code, the case where PLP_TYPE is defined within the NUM_LAYER loop is exemplified, but PLP_TYPE may be defined outside the NUM_LAYER loop or within the NUM_PLP loop, depending on the embodiment.

In the pseudo code, PLP_START represents the start position of the corresponding physical layer pipe. At this time, PLP_START may indicate a start position using a cell addressing scheme. In this case, PLP_START may be an index corresponding to the first data cell of the corresponding PLP.

In particular, PLP_START may be signaled for each of all physical layer pipes, and may be used for service identification using multiple-physical layer pipes along with a field signaling the size of a PLP according to an embodiment.

In the pseudo code, PLP_SIZE is size information of physical layer pipes. In this case, PLP_SIZE may be set equal to the number of data cells allocated to the corresponding physical layer pipe.

That is, in the pseudo code, it can be seen that PLP_TYPE is signaled in consideration of layer identification information, and PLP_SIZE and PLP_START are signaled for all physical layer pipes regardless of layer identification information.

The combiner 340 shown in FIGS. 3 and 7 serves to combine the core layer signal and the enhanced layer signal, and since the core layer signal and the enhanced layer signal share one time interleaver, the core layer signal And combining may be performed in units of time interleaver groups shared with enhanced layer signals.

At this time, it is advantageous in terms of memory efficiency or system efficiency that the time interleaver group is set based on the core layer.

However, when the time interleaver group is set based on the core layer, there may be FEC blocks divided at the boundary of the time interleaver group in the enhanced layer. Signaling of fields necessary to identify a part of the corresponding FEC block may be required.

A time interleaver used for layer division multiplexing may be a convolutional time interleaver (CTI) or a hybrid time interleaver (HTI). In this case, the convolutional time interleaver may be used when the core layer has one physical layer pipe, and the hybrid time interleaver may be used when the core layer has two or more physical layer pipes. When a hybrid time interleaver is used, a physical layer pipe may include only complete FEC blocks.

20 is a diagram illustrating an example of a case in which a convolutional time interleaver is used.

Referring to FIG. 20 , it can be seen that a subframe includes two layers: a core layer and an enhanced layer.

In the example shown in FIG. 20, since the subframe includes only one physical layer pipe (PLP #0) in the core layer, the time interleaver corresponding to the subframe is a convolutional time interleaver. When the convolutional time interleaver is used, the physical layer pipe of each layer may include an incomplete FEC block.

Such an incomplete FEC block may be identified using a field such as "L1D_plp_CTI_fec_block_start" located at the edge of the PLP and indicating the position of the first complete FEC block in each PLP.

In the example shown in FIG. 20 , the start position and size of the physical layer pipe (PLP #0) of the core layer and the physical layer pipe (PLP #1 of the enhanced layer) are the same.

In the example shown in FIG. 20 , it can be seen that the time interleaver group (TI Group) corresponds to the physical layer pipe (PLP #0) of the core layer. The time interleaver group is commonly applied to the core layer and the enhanced layer, and it is advantageous in terms of memory or system efficiency to set the time interleaver group corresponding to the core layer.

21 is a diagram illustrating another example of a case in which a convolutional time interleaver is used.

Referring to FIG. 21 , it can be seen that the start positions and sizes of the core layer physical layer pipe (PLP #0) and the enhanced layer physical layer pipe (PLP #1) are different.

In this way, when the starting positions and sizes of the core layer physical layer pipe (PLP #0) and the enhanced layer physical layer pipe (PLP #1) are different, an empty area may be included in the enhanced layer.

As shown in FIG. 21 , when a blank area is included at the rear end of the enhanced layer physical layer pipe (PLP #1), the enhanced layer physical layer pipe (PLP #1) ends with a complete FEC block.

22 is a diagram illustrating an example of a case where a hybrid time interleaver is used.

Referring to FIG. 22 , it can be seen that the core layer includes two physical layer pipes (PLP #0 and PLP #1).

As such, when the core layer is composed of multiple physical layer pipes, a hybrid time interleaver is used.

When the hybrid time interleaver is used, all physical layer pipes of the core layer and the enhanced layer include only complete FEC blocks.

In this case, a part of the enhanced layer may be vacated for alignment with the boundary of the core layer.

FIG. 23 is a diagram illustrating a time interleaver group in the example shown in FIG. 22 .

Referring to FIG. 23 , it can be seen that the time interleaver group boundary is set to correspond to the boundary of the physical layer pipes of the core layer.

In FIG. 23, a case in which a time interleaver group includes only one core layer physical layer pipe is exemplified, but according to an embodiment, a time interleaver group may include two or more core layer physical pipes.

In the example shown in FIG. 23 , in the case of an enhanced layer, one FEC block may be divided by a TI group boundary.

This is because time interleaver group division is performed on a core layer basis. In this case, information capable of identifying an incomplete FEC block of an enhanced layer corresponding to a time interleaver group boundary can be signaled.

24 to 26 are diagrams illustrating a process of calculating the size of an incomplete FEC block in the example shown in FIG. 23 .

Referring to FIG. 24 , the starting position of the physical layer pipe of the core layer (L1D_plp_start (PLP #0)), the size of the physical layer pipe of the core layer (L1D_plp_size (PLP #0)), and the start of the physical layer pipe of the enhanced layer It is possible to calculate the distance (A) between the start position (L1D_plp_start (PLP #2)) of the enhanced layer physical layer pipe and the time interleaver group boundary using the position (L1D_plp_start (PLP #2)).

Referring to FIG. 25 , it can be seen that the distance B between the start position of the divided FEC block and the time interleaver group boundary is calculated using the FEC block size of the enhanced layer.

In this case, the FEC block size may be determined using modulation information (L1D_plp_mod) corresponding to the enhanced layer and FEC type information (L1D_plp_fec_type) corresponding to the enhanced layer.

Referring to FIG. 26 , it can be seen that a part (C) of the FEC block of the enhanced layer corresponding to the boundary of the time interleaver groups is identified.

Table 3 below shows an example of L1-Detail fields of the preamble according to an embodiment of the present invention.

A preamble according to an embodiment of the present invention may include L1-Basic and L1-Detail.

Syntax # of bits L1_Detail_signaling() { L1D_version 4 L1D_num_rf 3 for L1D_rf_id=1 .. L1D_num_rf { L1D_rf_frequency 19 } if ( L1B_time_info_flag != 00 ) { L1D_time_sec 32 L1D_time_msec 10 if ( L1B_time_info_flag != 01 ) { L1D_time_usec 10 if ( L1B_time_info_flag != 10 ) { L1D_time_nsec 10 } } } for i=0 .. L1B_num_subframes { if (i > 0) { L1D_mimo One L1D_miso 2 L1D_fft_size 2 L1D_reduced_carriers 3 L1D_guard_interval 4 L1D_num_ofdm_symbols 11 L1D_scattered_pilot_pattern 5 L1D_scattered_pilot_boost 3 L1D_sbs_first One L1D_sbs_last One } if (L1B_num_subframes>0) { L1D_subframe_multiplex One } L1D_frequency_interleaver One L1D_num_plp 6 for j=0 .. L1D_num_plp { L1D_plp_id 6 L1D_plp_lls_flag One L1D_plp_layer 2 L1D_plp_start 24 L1D_plp_size 24 L1D_plp_scrambler_type 2 L1D_plp_fec_type 4 if (L1D_plp_fec_type ∈ {0,1,2,3,4,5}) { L1D_plp_mod 4 L1D_plp_cod 4 } L1D_plp_TI_mode 2 if ( L1D_plp_TI_mode=00) { L1D_plp_fec_block_start 15 } if ( L1D_plp_TI_mode=01) { L1D_plp_CTI_fec_block_start 22 } if (L1D_num_rf>0) { L1D_plp_num_channel_bonded 3 if (L1D_plp_num_channel_bonded>0) { L1D_plp_channel_bonding_format 2 for k=0 .. L1D_plp_num_channel_bonded{ L1D_plp_bonded_rf_id 3 } } } if (i=0 && L1B_first_sub_mimo=1) || (i >1 && L1D_mimo=1) { L1D_plp_stream_combining One L1D_plp_IQ_interleaving One L1D_plp_PH One } if (L1D_plp_layer=0) { L1D_plp_type One if L1D_plp_type=1 { L1D_plp_num_subslices 14 L1D_plp_subslice_interval 24 } L1D_plp_TI_extended_interleaving One if (L1D_plp_TI_mode=01) { L1D_plp_CTI_depth 3 L1D_plp_CTI_start_row 11 } else if (L1D_plp_TI_mode=10) { L1D_plp_HTI_inter_subframe One L1D_plp_HTI_num_ti_blocks 4 L1D_plp_HTI_num_fec_blocks_max 12 if (L1D_plp_HTI_inter_subframe=0) { L1D_plp_HTI_num_fec_blocks 12 } else { for (k=0.. L1D_plp_HTI_num_ti_blocks) { L1D_plp_HTI_num_fec_blocks 12 } } L1D_plp_HTI_cell_interleaver One } } else { L1D_plp_ldm_injection_level 5 } } } L1D_reserved as needed L1D_crc 32 }

All bits to which bits are allocated in Table 3 may correspond to an unsigned integer most significant bit first (uimsbf) format. Among the fields listed in Table 3, L1D_plp_layer indicates a layer corresponding to each physical layer pipe. can L1D_plp_start corresponds to start position information of the current PLP and may indicate the index of the first data cell of the current physical layer pipe. L1D_plp_size corresponds to size information of the current physical layer pipe, and may indicate the number of data cells allocated to the current physical layer pipe.

L1D_plp_fec_type corresponds to FEC type information of the current physical layer pipe, and may indicate a Forward Error Correction (FEC) method used to encode the current physical layer pipe.

For example, L1D_plp_fec_type="0000" corresponds to BCH and 16200 LDPC, L1D_plp_fec_type="0001" corresponds to BCH and 64800 LDPC, L1D_plp_fec_type="0010" corresponds to CRC and 16200 LDPC, and L1D_plp_fec_type="0011 " may correspond to CRC and 64800 LDPC, L1D_plp_fec_type="0100" may correspond to 16200 LDPC, and L1D_plp_fec_type="0101" may correspond to 64800 LDPC.

L1D_plp_mod may indicate modulation information of a current physical layer pipe. At this time, L1D_plp_mod can be signaled only when L1D_plp_fec_type satisfies a preset condition as shown in Table 3 above.

For example, L1D_plp_mod="0000" corresponds to QPSK, L1D_plp_mod="0001" corresponds to 16QAM-NUC, L1D_plp_mod="0010" corresponds to 64QAM-NUC, and L1D_plp_mod="0011" corresponds to 256QAM-NUC , L1D_plp_mod="0100" may correspond to 1024QAM-NUC, and L1D_plp_mod="0101" may correspond to 4096QAM-NUM. At this time, L1D_plp_mod can be set to "0100" or "0101" only when L1D_plp_fec_type corresponds to 64800 LDPC.

L1D_plp_TI_mode represents a time interleaving mode of PLP.

For example, L1D_plp_TI_mode="00" may indicate a mode not using a time interleaver, L1D_plp_TI_mode="01" may indicate a convolutional time interleaving mode, and L1D_plp_TI_mode="10" may indicate a hybrid time interleaving mode.

L1D_plp_fec_block_start may correspond to start position information of the first complete FEC block in the physical layer pipe. L1D_plp_fec_block_start can be signaled only when L1D_plp_TI_mode="00".

When layered division multiplexing is used, L1D_plp_fec_block_start may be signaled for each layer because the start position of the first FEC block in each layer may be different.

L1D_plp_CTI_fec_block_start may correspond to start position information of the first complete block in the physical layer pipe. L1D_plp_CTI_fec_block_start can be signaled only when L1D_plp_TI_mode=“01”.

In this case, more bits may be allocated to L1D_plp_CTI_fec_block_start than to L1D_plp_fec_block_start.

As described above, in the case of L1D_plp_TI_mode=10, since all PLPs include only complete FEC blocks, there is no need to separately signal the start position of the first FEC block.

L1D_plp_HTI_num_fec_blocks may correspond to the number of FEC blocks included in the current interleaving frame for the physical layer pipe of the core layer.

In this case, the time interleaver information includes fields (L1D_plp_CTI_depth, L1D_plp_CTI_start_row) corresponding to convolutional time interleaving and fields corresponding to hybrid time interleaving according to whether L1D_plp_TI_mode is 01 or 10 when L1D_plp_layer is 0 (core layer). It can be seen that (L1D_plp_HTI_inter_subframe, L1D_plp_HTI_num_ti_blocks, L1D_plp_HTI_num_fec_blocks_max, L1D_plp_HTI_num_fec_blocks, L1D_plp_HTI_cell_interleaver, etc.) are signaled as time interleaver information.

In this case, L1D_plp_CTI_depth may indicate the number of rows used in the convolutional time interleaver, and L1D_plp_CTI_start_row may indicate the position of an interleaver selector at the start of a subframe.

At this time, L1D_plp_HTI_inter_subframe represents a hybrid time interleaving mode, L1D_plp_HTI_num_ti_blocks represents the number of TI blocks per interleaving frame or the number of subframes in which cells from one TI block are transmitted, and L1D_plp_HTI_num_fec_blocks_max represents an FEC block per interleaving frame of the current physical layer pipe. , L1D_plp_HTI_num_fec_blocks indicates the number of FEC blocks included in the current interleaving frame of the current physical layer pipe minus 1, and L1D_plp_HTI_cell_interleaver indicates whether the cell interleaver is used.

In this case, a field such as L1D_plp_TI_mode may be signaled separately from time interleaver information signaled based on the core layer.

27 is a diagram for explaining the number of bits required for L1D_plp_fec_block_start when L1D_plp_TI_mode = “00”.

Referring to FIG. 27, when L1D_plp_TI_mode = “00” (when time interleaving is omitted), a cell address of FEC block start position before time interleaving (C_in) and time interleaving It can be seen that the cell address of the FEC block start position after time interleaving (C_out) after interleaving is the same.

When time interleaving is omitted as in the case of FIG. 27 , it can be considered that convolutional interleaving is performed with a depth of 0.

In this case, since L1D_plp_fec_block_start is defined after time interleaving, C_out may be signaled for each physical layer pipe within a subframe as L1D_plp_fec_block_start.

When the LDPC codeword is 16200 or 64800 and the modulation order is 2, 4, 6, 8, 10, and 12, the longest FEC block may have a length of 64800/2 = 32400.

Since 32400 can be expressed with 15 bits, if 15 bits are allocated to L1D_plp_fec_block_start, the case of L1D_plp_TI_mode=“00” can be covered.

28 and 29 are diagrams for explaining the number of bits required for L1D_plp_CTI_fec_block_start when L1D_plp_TI_mode = “01”.

Referring to FIG. 28, when L1D_plp_TI_mode = 01 (convolutional time interleaving), the cell address of FEC block start position before time interleaving (C_in) and time interleaving It can be seen that the subsequent cell address of FEC block start position after time interleaving (C_out) changes due to interleaving.

In this case, since L1D_plp_CTI_fec_block_start is defined after time interleaving, C_out may be signaled for each physical layer pipe within a subframe as L1D_plp_CTI_fec_block_start.

Referring to FIG. 29 , it can be seen that the convolutional time interleaver having a depth of 4 operates with C_in as an input and C_out as an output.

At this time, in the case of input, 0 corresponds to the 0th row, 1 corresponds to the 1st row, 2 corresponds to the 2nd row, 3 corresponds to the 3rd row, 4 corresponds to the 0th row, , 5 corresponds to row 1, 6 corresponds to row 2, 7 corresponds to row 3, 8 corresponds to row 0, 9 corresponds to row 1, 10 corresponds to row 2 corresponds to the row

First, in the case of 0, 4, 8, etc. corresponding to the 0th row, they are output immediately without delay.

In the case of 1, 5, 9, etc. corresponding to the first row, they are delayed by 4 and output.

In the case of 2, 6, 10, etc. corresponding to the second row, they are delayed by 8 and output.

In the case of 3, 7, etc. corresponding to the 3rd row, they are delayed by 12 and output.

That is, it can be seen that (n x 4) delays occur for the nth row.

In FIG. 29, the case where the depth is 4 (the row of the time interleaver is 4) is exemplified, but if the row of the time interleaver is N_row, the input corresponding to the nth row is delayed by (n x N_row).

At this time, the cell address (L1D_plp_CTI_fec_block_start) of the FEC block start position after time interleaving may be calculated as (C_in + (n x N_row)). In this case, n is a row corresponding to C_in, and may be determined by L1D_CTI_start_row among time interleaving information signaled by L1-Detail. In this case, n may be ((L1D_CTI_start_row + C_in) % N_row). In this case, L1D_CTI_start_row may indicate the position of the interleaver selector at the start of the subframe.

That is, L1D_plp_CTI_fec_block_start may be calculated by adding delay caused by time interleaving to C_in.

To calculate the number of bits required for signaling L1D_plp_CTI_fec_block_start, the maximum value of L1D_plp_CTI_fec_block_start is required. As already seen, the maximum value of C_in is 32400, the maximum value of n is N_row-1, and N_row can be up to 1024 in the case of non-extended interleaving. At this time, the maximum value of L1D_plp_CTI_fec_block_start is (32400 + (1024-1)x1024) = 1079952. 1079952 can be signaled with at least 21 bits.

N_row can be up to 1448 in case of extended interleaving. At this time, the maximum value of L1D_plp_CTI_fec_block_start is (32400 + (1448-1)x1448) = 2127656. 2127656 requires at least 22 bits to signal.

After all, when L1D_plp_TI_mode="00", the maximum value of L1D_plp_fec_block_start is equal to the maximum value of C_in, and when L1D_plp_TI_mode="01", the maximum value of L1D_plp CTI_fec_block_start is the value obtained by adding the delay due to interleaving to the maximum value of C_in, so L1D_plp_fec_block_start Efficient signaling is possible only when the number of bits used for signaling L1D_plp CTI_fec_block_start is greater than the number of bits used for signaling .

In the case of L1D_plp_TI_mode="10", since all physical layer pipes of the core layer and enhanced layer can contain only complete FEC blocks, the starting position of all physical layer pipes becomes the starting position of the first complete FEC block, There is no need to signal fields such as L1D_plp_fec_block_start or L1D_plp CTI_fec_block_start.

30 is a diagram illustrating insertion of enhanced layer dummy values when an HTI mode is used together with layered division multiplexing.

Referring to FIG. 30 , it can be seen that dummy values (Dummy) are inserted into the enhanced layer data (L1D_PLP_layer = 1) of the time interleaver group (TI_Group_1).

A set of all PLPs related to a specific end product transmitted to a receiver within a subframe is referred to as a PLP group.

When layered division multiplexing is used, a PLP group may include at least one core PLP and one or more enhanced PLPs.

When time interleaving in HTI mode is used, where an integer number of FEC blocks are used for the actual PLP data, the total number of cells in the core PLP(s) depends on the ModCod configuration of each PLP within a particular PLP group. ) may be different from the total number of cells in In this case, enhanced layer dummy values may be inserted next to actual data cells of the last enhanced PLP in the PLP group so that the total number of enhanced layer cells in the PLP group is equal to the total number of core layer cells. Since the time interleaver group is set based on the core PLPs, dummy values may not be inserted into the core layer.

Insertion of enhanced layer dummy values may be performed after BICM steps and before core PLP(s) and enhanced PLP(s) are combined. To generate the enhanced layer dummy values, a scrambling sequence is used and this scrambling sequence can be reinitialized for each relevant PLP group. And, this sequence may be modulated using the same constellation mapping as that used for the last enhanced PLP in the current PLP group.

The enhanced layer dummy values may have the same power as the immediately preceding enhanced PLP in the same PLP group so that the same scaling and normalizing factors as those of real data are applied.

31 is a diagram illustrating an example of a shift register used to generate dummy values according to an embodiment of the present invention.

Referring to FIG. 31, a sequence is generated using a 16-bit shift register corresponding to a generator polynomial 1 + X + X ³ + X ⁶ + X ⁷ + X ¹¹ + X ¹² + X ¹³ + X ¹⁶ can be known to be created.

The register shown in FIG. 31 may be initialized with the initial sequence 0xF180 (1111 0001 1000 0000). As described above, reinitialization is performed for each relevant PLP group.

In the example shown in FIG. 31 , the eight shift register outputs (D ⁷ , D ⁶ , ..., D ⁰ ) become the output bits. After the output bits are output, the bits of the shift register are shifted once. After one shift, 0, the exclusive or value of 1 and 1, is stored in register X ¹⁴ , 1, the exclusive or value of 1 and 0, is stored in register X ¹³ , and 1 and 0 are stored in register X ¹² . 1, the exclusive OR value, is stored in register X ¹¹ , the 0 stored in register X ¹⁰ is stored in register X ⁷ , 1, the exclusive OR value of 1 and 0, is stored in register X ⁴ , and 1 and 0 are stored in register X 4. 1, the exclusive or value of , is stored in register X ³ , 0 stored in register X ² is stored, and 1 stored in register X ¹⁶ is stored in register X.

Therefore, in the example shown in FIG. 31, the output sequence (scrambling sequence) becomes 1100 0000 0110 1101 0011 1111 ... (MSB first, or D ⁷ , D ⁶ , ..., D ⁰ , D ⁷ , D ⁶ , ...).

As described above, each physical layer pipe may be configured in any one of a no time interleaving mode, a convolutional time interleaving mode, and a hybrid time interleaving mode.

The time interleaving mode of the PLP can be identified by L1D_plp_TI_mode, and the time interleaving mode of the enhanced PLP must be the same as the time interleaver mode identified for the core PLP in which the enhanced layer is layer division multiplexed.

A complete delivered product consists of only one constant-cell-rate PLP or one constant-cell-rate core PLP and said one core When it consists of one or more constant-cell-rate enhanced PLP(s) multiplexed to a PLP, the PLP(s) constituting the complete transmission product are no time interleaving mode, convolutional time interleaving mode and hybrid It can be configured as one of the time interleaving modes.

When a complete delivered product is composed of PLPs having characteristics different from those described in the previous paragraph, the PLPs constituting the complete delivered product are selected from among the no time interleaving mode and the hybrid time interleaving mode. It can be configured in one time interleaving mode.

At this time, a complete transport product may correspond to one service. That is, a complete transport product may include all PLP data required for one service.

The time interleaving mode(s) for PLPs of a particular complete delivered product is the time interleaving mode(s) for PLP(s) of any other delivered products transmitted within the same RF channel. It can be configured independently of the mode(s). When one particular transmit product includes multiple core PLPs and/or PLPs that are not layered division multiplexed, these PLPs may be configured in the same or different time interleaving modes (e.g., no time interleaving mode and/or hybrid time interleaving mode). ) and/or the same or different time interleaver parameters.

32 is a diagram illustrating types of time interleaving modes.

Referring to FIG. 32, time interleaving modes are largely divided into intra-subframe interleaving and inter-subframe interleaving.

Intra-subframe interleaving corresponds to a case where interleaving is performed within a subframe. In this case, the interleaving frame is mapped to one subframe. That is, when intra-subframe interleaving is performed, the decoder can decode a corresponding physical layer pipe within a subframe.

Inter-subframe interleaving corresponds to a case where interleaving is out of the range of one subframe. In this case, the interleaving frame is mapped to a plurality of subframes. That is, when inter-subframe interleaving is performed, a decoder may require data of a subframe other than one subframe to decode a corresponding physical layer pipe.

As shown in FIG. 32, the no time interleaving mode (NO TI) corresponds to intra-subframe interleaving, and the convolutional time interleaving mode (CTI) corresponds to inter-subframe interleaving. In this case, the no-time interleaving mode (NO TI) may be regarded as an interleaving mode having an interleaving depth of 0.

Hybrid time interleaving mode (HTI) may correspond to intra-subframe interleaving or inter-subframe interleaving. When CDL (Convolutional Delay Line) is turned off, it corresponds to intra-subframe interleaving and CDL When is on, it may correspond to inter-subframe interleaving.

In the hybrid time interleaving mode, intra-subframe interleaving or inter-subframe interleaving can be identified by the L1D_plp_HTI_inter_subframe field. For example, when L1D_plp_HTI_inter_subframe=0, the corresponding time interleaving mode corresponds to intra-subframe interleaving, and when L1D_plp_HTI_inter_subframe=1, the corresponding time interleaving mode corresponds to inter-subframe interleaving.

Since the time interleaving mode or parameters related to time interleaving can be set for each core layer physical layer pipe, if the decoding process is not considered when setting the time interleaving mode or parameters related to time interleaving in the encoding process, one enhanced layer physical layer It may occur that some of the plurality of core layer and physical layer pipes layered on pipes use intra-subframe interleaving and others use inter-subframe interleaving.

33 is a diagram illustrating a case in which intra-subframe interleaving and inter-subframe interleaving are simultaneously used.

Referring to FIG. 33, three core layer physical layer pipes (CORE PLP #0, CORE PLP #1, and CORE PLP #2) are layered division multiplexed on one enhanced layer physical layer pipe (ENHANCED PLP #3). it can be seen that there is

The first core layer physical layer pipe (CORE PLP #0) corresponds to intra-subframe interleaving because it is an HTI mode with CDL turned off, and the second core layer physical layer pipe (CORE PLP #1) is HTI mode with CDL turned on. Therefore, it corresponds to inter-subframe interleaving, and since the third core layer physical layer pipe (CORE PLP #2) is a NO TI mode, it corresponds to intra-subframe interleaving.

Therefore, the first and third core layer physical layer pipes (CORE PLP #0 and CORE PLP #2) can be immediately decoded, but the second core layer physical layer pipe (CORE PLP #1) is a time interleaving unit (N It can be decoded by waiting until the number (N _IU -1) of subframes corresponding to _IU ) is decoded. In this case, the time interleaving unit (N _IU ) may be the number of subframes in which cells from one time interleaving block are transmitted during inter-subframe interleaving.

In the example shown in FIG. 33 , pieces of an enhanced layer physical layer pipe have different decoding timings, which means that additional latency and buffers are required to decode the corresponding enhanced layer physical layer pipe.

34 is a diagram illustrating subframes when intra-subframe interleaving and inter-subframe interleaving are simultaneously used.

Referring to FIG. 34, three core layer physical layer pipes (PLP #0, PLP #1, and PLP #2) are included in one enhanced layer physical layer pipe (PLP #3) over three subframes. It can be seen that layer division multiplexing is performed.

At this time, the pieces of the enhanced layer and physical layer pipes (PLP #3-A, PLP #3-B, and PLP #3-C) have 5, 2, and 4 FEC blocks, respectively, and the core layer and physical layer The time interleaving unit (N _IU ) of the pipe (PLP #1) may be 3. At this time, the enhanced layer physical layer pipe (PLP #3) must wait for subframes corresponding to the time interleaving unit (N _IU ) to be decoded.

In the example shown in FIG. 34 , the output timings of enhanced layer cells are #0, 1, 2, 3, 4, 7, 8, 9, 10, 11, 12, 13, 14, 15, 18, 19, 20 , 21, 22, 23, 24, 25, 26, 5, 6, 29, 39, 31, 32... Accordingly, parts #5 and #6 of the first subframe SUBFRAME #0 are output after waiting for two subframes SUBFRAME #1 and SUBFRAME #2, which becomes a problem of decoding timing.

In order to solve such a problem of decoding timing, when a plurality of core layer physical layer pipes are layered division multiplexed in one enhanced layer physical layer pipe, all core layer physical layer pipes use intra-subframe interleaving. Alternatively, it may be effective to reduce the decoding timing problem and related decoding complexity by making all core layer physical layer pipes use inter-subframe interleaving.

Even when all of the plurality of core layer physical layer pipes layered division multiplexed into one enhanced layer physical layer pipe use inter-subframe interleaving, the time interleaving units (N _IU ) of the core layer physical layer pipes may be different from each other. can

However, decoding complexity may increase even when all core layer physical layer pipes that are layered division multiplexed into one enhanced layer physical layer pipe use inter-subframe interleaving.

35 is a diagram illustrating a case in which different TI units are simultaneously used.

Referring to FIG. 35, three core layer physical layer pipes (CORE PLP #0, CORE PLP #1, and CORE PLP #2) are layered division multiplexed on one enhanced layer physical layer pipe (ENHANCED PLP #3). it can be seen that there is

Since all three core layer physical layer pipes (CORE PLP #0, CORE PLP #1, and CORE PLP #2) are HTI modes in which CDL is turned on, they correspond to inter-subframe interleaving. However, the time interleaving unit (N _IU ) of the first core layer physical layer pipe (CORE PLP #0) is 2, and the time interleaving unit (N _IU ) of the second core layer physical layer pipe (CORE PLP #1) is 4 , and the time interleaving unit (N _IU ) of the third core layer physical layer pipe (CORE PLP #2) is 3.

Therefore, the first core layer physical layer pipe (CORE PLP #0) has to wait for 1 subframe, the second core layer physical layer pipe (CORE PLP #1) has to wait for 3 subframes, and the third core layer physical layer pipe (CORE PLP #1) has to wait for 3 subframes. The pipe (CORE PLP #2) has to wait 2 subframes.

In the example shown in FIG. 35 , pieces of an enhanced layer physical layer pipe have different decoding timings, which means that additional latency and buffers are required to decode the corresponding enhanced layer physical layer pipe.

In this way, all core layer physical layer pipes that are layered division multiplexed into one enhanced layer physical layer pipe use inter-subframe interleaving and all use the same time interleaving unit to reduce the decoding timing problem and related decoding complexity. lowering can be effective.

However, even when a plurality of core layer physical layer pipes layered division multiplexed into one enhanced layer physical layer pipe all use inter-subframe interleaving and all use the same time interleaving unit, decoding problems depend on the subframe structure. may occur.

36 is a diagram illustrating subframes when the same TI unit is simultaneously used.

Referring to FIG. 36, in a first subframe (SUBFRAME 0), two core layer physical layer pipes (CORE PLP #0 and CORE PLP #1) are included in one enhanced layer physical layer pipe (ENHANCED PLP #3). It can be seen that layered division multiplexing is performed.

In addition, in each of the second and third subframes SUBFRAME 1 and 2, one core layer physical layer pipe (CORE PLP #1) is added to one enhanced layer physical layer pipe (ENHANCED PLP #3) for layered division multiplexing. has been

In addition, in the fourth subframe (SUBFRAME 3), one core layer physical layer pipe (CORE PLP #0) is layered division multiplexed on one enhanced physical layer pipe (ENHANCED PLP #3).

At this time, the two core layer physical layer pipes (CORE PLP #0 and CORE PLP #1) of the first subframe both correspond to inter-subframe interleaving and use the same time interleaving unit (N _IU =3). .

However, the core layer physical layer pipe (CORE PLP #1) only needs to wait until the third subframe (SUBFRAME 2), but the core layer physical layer pipe (CORE PLP #0) has to wait until after the fourth subframe (SUBFRAME 3). . This is because structures of subframes after the first subframe (SUBFRAME 0) are different.

In the example shown in FIG. 36, fragments of the enhanced layer physical layer pipe have different decoding timings despite the same time interleaving unit, which means that additional latency and buffers are required to decode the corresponding enhanced layer physical layer pipe. do.

33 to 36, when a plurality of core layer physical layer pipes are layered division multiplexed in one enhanced layer physical layer pipe, the decoding timing of each piece of the enhanced layer physical layer pipe is different, resulting in a decoding problem. Occurs.

When one enhanced layer physical layer pipe is spread over multiple time interleaving groups, all core layer physical layer pipes related to the enhanced layer physical layer pipe may use the same time interleaving mode. In this case, all core layer physical layer pipes related to the enhanced layer physical layer pipe may use a hybrid time interleaving mode or a no time interleaving mode.

That is, in this case, all core layer and physical layer pipes use the same time interleaving mode, but use of the convolutional time interleaving mode may be prohibited.

According to embodiments, when all core layer physical layer pipes related to the enhanced layer physical layer pipe use the convolutional time interleaving mode, interleaving depths (L1D_plp_CTI_depth) of all core layers may be the same.

In this case, when all core layer physical layer pipes related to the enhanced layer physical layer pipe use the hybrid time interleaving mode, each core layer physical layer pipe may use the intra-subframe interleaving mode (L1D_plp_HTI_inter_subframe=0). . That is, when all core layer physical layer pipes related to the enhanced layer physical layer pipe use the hybrid time interleaving mode, inter-subframe interleaving may be prohibited.

According to an embodiment, when core layer physical layer pipes corresponding to the enhanced layer physical layer pipes use a hybrid time interleaving mode corresponding to inter-subframe interleaving, all core layer physical layer pipes use the same time interleaving unit. can be used

In this case, when all core layer physical layer pipes related to the enhanced layer physical layer pipe use the no-time interleaving mode, each core layer physical layer pipe includes an integer number of FEC blocks within a subframe. FEC blocks).

In this case, dummy modulation values may be used to achieve an integer number of FEC blocks per subframe.

Depending on the subframe configuration and the physical layer pipe multiplexing parameters, the available data cells of the subframe may be fully or partially occupied by physical layer pipe data. If all available data cells do not have corresponding physical layer pipe data, it is important that the unoccupied data cells be modulated rather than remaining unmodulated null cells to ensure constant transmit power. This can be achieved by assigning pseudo-random dummy modulation values to unoccupied data cells.

Depending on the physical layer pipe multiplexing parameters, unoccupied data cells can occur anywhere within a subframe. Thus, all available data cells of a subframe are first filled with dummy modulation values, and then the cell multiplexing process may overwrite the dummy modulation values of occupied data cells with real physical layer pipe data. This approach can ensure that all available data cells in a subframe are modulated by a physical layer pipe cell or dummy modulation value.

Let N _cell be the total number of available data cells in a subframe so that data cells are indexed from 0 to N _cell -1, and d _i is dummy modulation for a data cell having an index i (0≤i<N _cell ) value, and b _i (0≤i<N _cell ) represents the i-th value of the scrambling sequence described with reference to FIG. 31 .

In this case, the real value of the dummy modulation value for the i-th data cell (0≤i<N _cell ) may be (1 - 2 * b _i ), and the imaginary value may be 0. That is, dummy modulation values may be generated by mapping the value of the scrambling sequence to one of two phases which are separated by 180 degrees.

Each of the N _cell available data cells in the subframe may have corresponding dummy modulation values before the physical layer pipe data is multiplexed into the subframe. Following insertion of dummy modulation values, physical layer pipe data belonging to the current subframe may be mapped to corresponding data cells assigned to the corresponding physical layer pipe data, and dummy modulation values previously assigned to the data cells may be overwritten. there is.

Decoding timing mismatch may be a problem not only in the case of a plurality of core layer physical layer pipes multiplexed to one enhanced layer physical layer pipe but also in the case of core layer physical layer pipes when layered division multiplexing is not applied.

37 is a diagram illustrating a case where one complete transport product is composed of a plurality of physical layer pipes.

Referring to FIG. 37 , it can be seen that one complete transport product includes three core layer physical layer pipes (PLP #0, PLP #1, and PLP #2). In this case, one complete transport product may correspond to one service (SERVICE A). In this case, the core layer and physical layer pipes PLP #0, PLP #1, and PLP #2 may not be layered division multiplexed.

At this time, the first core layer physical layer pipe (PLP #0) corresponds to video data of one service (SERVICE A), and the second core layer physical layer pipe (PLP #1) corresponds to the service (SERVICE A). Corresponds to the first audio data, and the third core layer physical layer pipe (PLP #2) may correspond to the second audio data of the service (SERVICE A).

Since the characteristics of the corresponding data are different for each core layer physical layer pipe constituting one complete transmission product, the first core layer physical layer pipe (PLP #0) corresponds to a hybrid time interleaving mode that is CDL OFF, The second core layer physical layer pipe (PLP #1) corresponds to the hybrid time interleaving mode with CDL on, and the third core layer physical layer pipe (PLP #2) corresponds to the time interleaving mode with CDL on. can match At this time, as shown in FIG. 37, the time interleaving unit (N _IU ) of the core layer physical layer pipe (PLP #1) is 3, and the time interleaving unit (N _IU ) of the core layer physical layer pipe (PLP #2) may be 4.

If some of the core layer physical layer pipes constituting one complete transport product correspond to intra-subframe interleaving and some correspond to inter-subframe interleaving, the core layer physical layer pipes corresponding to intra-subframe interleaving are Core layer physical layer pipe that is immediately decoded and corresponds to inter-subframe interleaving may have to wait for another subframe.

Furthermore, even when the core layer physical layer pipes use inter-subframe interleaving, decoding timing is changed if the time interleaving units (N _IU ) of the core layer physical layer pipes are different.

In the example shown in FIG. 37, the first physical layer pipe (PLP #0) corresponds to intra-subframe interleaving and therefore is immediately decoded, and the second physical layer pipe (PLP #1) corresponds to inter-subframe interleaving and is time-decoded. Since the interleaving unit (N _IU ) is 3, we have to wait for 2 subframes, and since the third physical layer pipe (PLP #2) corresponds to inter-subframe interleaving and the time interleaving unit (N _IU ) is 4, we have to wait for 3 subframes. do.

In the example shown in FIG. 37, three core layer physical layer pipes have different decoding timing. At this time, the first core layer physical layer pipe (PLP #0) and the second core layer physical layer pipe (PLP #1) are synchronized with the third core layer physical layer pipe (PLP #2) for service (SERVICE A). Since they have to wait for 3 subframes, this causes unnecessary decoding complexity.

To reduce decoding complexity, a particular complete delivered product includes a plurality of core layer physical layer pipes that are not layered-division multiplexed, and all of these core layer physical layer pipes use hybrid time interleaving mode. When used, all of these core layer physical layer pipes may use intra-subframe interleaving mode, or all of these core layer physical layer pipes may use inter-subframe interleaving mode. That is, all of these core layer physical layer pipes may have the same L1D_plp_HTI_inter_subframe value (L1D_plp_HTI_inter_subframe=0 indicates intra-subframe interleaving, and L1D_plp_HTI_inter_subframe=1 indicates inter-subframe interleaving).

If inter-subframe interleaving is used for these core layer physical layer pipes (L1D_plp_HTI_inter_subframe=1), all of these core layer physical layer pipes may use the same time interleaving unit (N _IU ).

If one specific complete physical layer pipe contains a plurality of core layer physical layer pipes that are not layered division multiplexed and at least one of these core layer physical layer pipes uses no time interleaving mode, then these core layer physical layer pipes Among them, those using the hybrid time interleaving mode may all use the intra-subframe interleaving mode (L1D_plp_HTI_inter_subframe=0). That is, if one specific complete physical layer pipe includes a plurality of core layer physical layer pipes that are not layered division multiplexed, and at least one of the core layer physical layer pipes uses the no-time interleaving mode, the core layer physical layer pipe Use of the inter-subframe interleaving mode may be prohibited for all physical layer pipes corresponding to the hybrid time interleaving mode among pipes.

FIG. 38 is a block diagram illustrating an example of the time interleaver shown in FIG. 3 or FIG. 7 .

Referring to FIG. 38, the time interleaver according to an embodiment of the present invention includes a cell interleaver 3810, a twisted block interleaver 3820, and a convolutional delay line 3830.

A cell interleaver 3810 interleaves cells within a time interleaving block.

At this time, the cell interleaver 3810 may arrange the input cells of the FEC blocks into time interleaving blocks. In this case, the TI block may include one or more FEC blocks.

In this case, the time interleaving block may be a basic unit for the operation of the cell interleaver 3810, the twisted block interleaver 3820, and the convolutional delay line 3830.

In this case, time interleaving blocks may include different numbers of FEC blocks.

In this case, the cell interleaver 3810 may interleave cells within each FEC block.

At this time, the cell interleaver 3810 may perform cell interleaving by writing the FEC block to the memory and reading it pseudo-randomly.

Depending on embodiments, the cell interleaver 3810 may be omitted.

The twisted block interleaver 3820 performs intra-subframe interleaving corresponding to time interleaving blocks.

The convolutional delay line 3830 performs inter-subframe interleaving using the output of the twisted block interleaver 3820. That is, the convolutional delay line 3830 spreads block-interleaved time interleaving blocks over multiple subframes.

In this case, the twisted block interleaver 3820 may perform the intra-subframe interleaving by performing a column-wise write operation and a diagonal-wise read operation.

At this time, the convolutional delay line 3830 can read only data cells excluding virtual cells from the twisted block interleaver 3820.

In this case, the convolutional delay line 3830 may store new virtual cells after each row of the data cells is written from the twisted block interleaver 3820 and before the switches move to the next branch. .

At this time, the new virtual cells are calculated by subtracting the number of FEC blocks (N _{FEC_TI} ) in the TI block of the interleaving frame from the maximum value (N _{FEC_TI_MAX} ) of the number of FEC blocks in the TI block of the interleaving frame in each branch. may be commensurate.

In this case, new virtual cells may not be output as an output of the time interleaving apparatus.

In this case, the convolutional delay line 3830 may include branches corresponding to the time interleaving unit (N _IU ), and the remaining branches excluding the first branch among the branches may include one or more FIFO registers.

At this time, the convolutional delay line 3830 may output only some of the initial values stored in the FIFO register.

In this case, some of the initial values may correspond to one initialization cell for each of the remaining branches.

FIG. 39 is a diagram illustrating a write operation of the twisted block interleaver shown in FIG. 38;

Referring to FIG. 39 , it can be seen that the cells of the FEC blocks included in the TI block are written to the memory by a column-wise write operation.

40 is a diagram illustrating a read operation of the twisted block interleaver shown in FIG. 38;

Referring to FIG. 40 , it can be seen that the cells of the FEC blocks included in the TI block are read from the memory by a diagonal-wise read operation.

In the examples shown in FIGS. 39 and 40, N _{FEC_TI_MAX} represents the maximum value of the number of FEC blocks in a time interleaving block of an interleaving frame, and N _r represents the number of cells included in each FEC block.

39 and 40, the twisted block interleaver may perform the intra-subframe interleaving by performing a column-wise write operation and a diagonal-wise read operation.

In this case, the twisted block interleaver may skip virtual FEC cells during the reading process as shown in FIG. 40 .

FIG. 41 is a block diagram illustrating an example of the convolutional delay line shown in FIG. 38 .

Referring to FIG. 41 , a convolutional delay line according to an embodiment of the present invention includes N _IU branches. That is, the convolutional delay line divides (spits) the time interleaving block into N _IU interleaving units, and distributes the interleaving units over N _IU subframes.

Switch S ₀ can connect the twisted block interleaver to the convolutional delay line. Switch S ₁ can connect the convolutional delay line to a framing block such as the frame builder shown in FIG. 3 or FIG. 7 .

At this time, the movements of the switches S ₀ and S ₁ may be synchronized. That is, the switches can always point to the same branches of the convolutional delay line.

The switches can move from the last branch of the convolutional delay line back to the first branch.

When N _{FEC_TI_MAX} cells composed of N _{FEC_TI} data cells and (N _{FEC_TI_MAX} - N _{FEC_TI} ) virtual cells are written as a convolutional delay line, the two switches S ₀ and S ₁ are branch n 0 <= n < N an integer with _IU -1) to the immediately following branch n+1. At this time, N _{FEC_TI_MAX} represents the maximum value of the number of FEC blocks in the time interleaving block of the interleaving frame, N _{FEC_TI} represents the number of FEC blocks corresponding to data in the time interleaving block of the interleaving frame, and N _IU represents the time interleaving unit. can represent In this case, (N _{FEC_TI_MAX} - N _{FEC_TI} ) virtual cells may not be read from the twisted block interleaver, but may be new virtual cells for the convolutional delay line. That is, the new virtual cells may be irrelevant to the twisted block interleaver and may be newly generated in the convolutional delay line.

In this case, the two switches S ₀ and S ₁ may be reset to the first branch (branch 0) of the convolutional delay line at the start of every subframe.

At this time, the virtual cells may not be read from the twisted block interleaver and may not be transferred to the convolutional delay line.

However, after N _{FEC_TI} data cells are written to the convolutional delay line from the twisted block interleaver, before the switches (S ₀ and S ₁ ) move to the next branch of the convolutional delay line, ( N _{FEC_TI_MAX} - N _{FEC_TI} ) New virtual cells may be input to the convolutional delay line.

At this time, the virtual cells may not be written to the output of the time interleaver from either the twisted block interleaver or the convolutional delay line.

42 is a diagram illustrating an example of an operation of the twisted block interleaver shown in FIG. 38;

Referring to FIG. 42 , it can be seen that an example in which the number of cells (N _r ) included in FEC blocks is 8, N _{FEC_TI_MAX} is 5, N _{FEC_TI} is 3, and N _IU is 2 is shown.

In the example shown in FIG. 42 , virtual cells corresponding to two columns are stored in the twisted block interleaver, and twisted block interleaving is performed through column-direction write and diagonal-direction read operations.

In the example shown in FIG. 42, the output memory of the twisted block interleaver includes virtual cells.

43 is a diagram illustrating an example of an operation of the convolutional delay line shown in FIG. 38;

Referring to FIG. 43 , since N _IU is 2, it can be seen that two branches exist in the convolutional delay line, and the FIFO register is included in the second branch.

In the example shown in FIG. 43, virtual cells read from the twisted block interleaver are transferred to the convolutional delay line.

In particular, FIG. 43 shows the first subframe timing when N _IU is 2. At this timing, it can be seen that all data corresponding to the second branch are values stored in the FIFO register.

As already described, the virtual cell may not be included in the transmission signal.

43, the first subframe (subframe #1) is “2, 11, 20, 10, 19, 6, 5, 14, 23, 8, 22, 16" is transmitted, and the second subframe (subframe #2) transmits "7, 1, 15, 0, 9, 18, 17, 4, 13, 3, 12, 21".

In the example shown in FIG. 43, I ₀ , I ₁ , ..., I ₁₉ stored in the FIFO register at a previous timing and then output are the lower part (5) of the memory corresponding to the convolutional delay line. rows to 8), but may be stored in rows 2, 4, 6, and 8 of the memory corresponding to the convolutional delay line according to embodiments.

FIG. 44 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 43 .

Referring to FIG. 44 , a time deinterleaver corresponding to a decoding process may restore memory data (CDL memory state including virtual cells) shown in FIG. 44 from two subframes through a FIFO register. Furthermore, the time deinterleaver may restore data (writing order to TBDI memory) input to the twisted block deinterleaver from memory data.

As shown in FIG. 44, when virtual cells are transferred from the twisted block interleaver to the convolutional delay line, it can be seen that the virtual cells are scattered in the receiver. At this time, the receiver needs to know the writing process of virtual cells.

That is, an inverse convolutional delay line of the receiver requires location information of virtual cells, which increases decoding complexity and memory.

FIG. 45 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 42 .

Referring to FIG. 45, it can be seen that accurate twisted block deinterleaving is possible only when the positions of virtual cells are known. That is, in order to accurately perform twisted block deinterleaving in the time deinterleaver of the receiver, the location of virtual cells or at least the number of virtual cells to be included in each row must be known.

As such, in the embodiments described with reference to FIGS. 42 to 45, since the locations of virtual cells must be known in the time deinterleaver, decoding complexity increases.

46 is a diagram illustrating another example of an operation of the twisted block interleaver shown in FIG. 38;

Referring to FIG. 46 , unlike the example shown in FIG. 42 , in the example shown in FIG. 46 , during a read process from the twisted block interleaver, it can be seen that virtual cells belonging to virtual FEC blocks are skipped.

That is, in the example shown in FIG. 46, the twisted block interleaver may output only data cells corresponding to data and not output virtual cells.

47 is a diagram illustrating another example of an operation of the convolutional delay line shown in FIG. 38;

Referring to FIG. 47 , in the example shown in FIG. 47 , virtual cells are not written to convolutional delay lines from the twisted block interleaver.

That is, the convolutional delay line reads and stores only data cells excluding virtual cells from the twisted block interleaver, and then generates and stores new virtual cells so that the virtual cells are not dispersed.

At this time, when the memory operates in the FIFO (First-In-First-Out) method, the data shown in FIG. 47 can be seen as being read and written first from the left. However, in the example shown in FIG. 47 , virtual cells marked with X may be written to the memory last. That is, in the example shown in FIG. 47, the order of writing to the memory corresponding to the convolutional delay line is 2, 11, 20, X, X, 19, 6, 15, X, X, 5, 14, 23, X , X, ...

In this case, when data cells are written from the twisted block interleaver to the convolutional delay line, the virtual cells may be stored corresponding to the leftmost column (N _{FEC_TI_MAX} - N _{FEC_TI} ) of the convolutional delay line.

47 also shows the first subframe timing when N _IU is 2, as in FIG. 43, it can be seen that at this timing, all data corresponding to the second branch are values stored in the FIFO register.

In this case, the virtual cell may not be included in the transmission signal.

Therefore, assuming that the memory is written and read from the left side of the memory shown in FIG. 47 (excluding virtual cells), the first subframe (subframe #1) is "2, 11, 20, 19, 6, 15, 5, 14, 23 , 13, 22, 16" and the second subframe (subframe #2) transmits "7, 1, 10, 0, 9, 18, 8, 17, 4, 3, 12, 21" can see.

In the example shown in FIG. 47, I ₀ , I ₁ , ..., I ₁₉ stored in the FIFO register at a previous timing and then output are displayed at the bottom (5) of the memory corresponding to the convolutional delay line. rows to 8), but may be stored in rows 2, 4, 6, and 8 of the memory corresponding to the convolutional delay line according to embodiments.

In the examples shown in FIGS. 43 and 47 , the memory corresponding to the convolutional delay line is shown for convenience of description, and the convolutional delay line may not include a separate output memory according to embodiments.

FIG. 48 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 47 .

Referring to FIG. 48, a time deinterleaver corresponding to a decoding process may restore memory data (CDL memory state including virtual cells) shown in FIG. 48 from two subframes through a FIFO register. Furthermore, the time deinterleaver may restore data (writing order to TBDI memory) input to the twisted block deinterleaver from memory data. Unlike the case of FIG. 44, in the example shown in FIG. 48, it can be seen that the positions of virtual cells are known and the virtual cells are not scattered during the decoding process of the inverse convolutional delay line.

Accordingly, decoding complexity is lower in the case of FIG. 48 than in the case of FIG. 44 .

FIG. 49 is a diagram illustrating an example of a decoding process corresponding to the operation shown in FIG. 46 .

Referring to FIG. 49, since reverse processes of the column-wise write operation and the diagonal-wise read operation of twisted block deinterleaving are performed except for virtual cells, it is simpler than the case of FIG. 45. It can be seen that deinterleaving is possible.

50 is a diagram showing initial values of FIFO registers included in a convolutional delay line.

Referring to FIG. 50 , it can be seen that data corresponding to the second branch in the first subframe timing when N _IU is 2 are all values 5010 stored in the FIFO register.

At this time, the convolutional delay line may output only some of the initial values stored in the FIFO register.

In this case, some of the initial values may correspond to one initialization cell for each of the branches other than the first branch among the branches corresponding to the time interleaving unit (N _IU ).

That is, among the values 5010 stored in the FIFO register, only one cell per row may be output and the remaining cells may not be output.

According to embodiments, memories corresponding to the twisted block interleaver (TBI) and the convolutional delay line (CDL) operate in a FIFO manner, and reading and writing to these memories may be performed from right to left. 50, the first row of the twisted block interleaver memory stores 2, 11, and 20 from right to left, the second row stores 7, 1, and 10 from right to left, and the third row stores 7, 1, and 10 from right to left. row stores 19, 6, 15 from right to left, fourth row stores 0, 9, 18 from right to left, fifth row stores 5, 14, 23 from right to left, and six The seventh row could store 8, 17, 4 from right to left, the seventh row could store 13, 22, 16 from right to left, and the eighth row could store 3, 12, 21 from right to left. .

At this time, reading and writing to the FIFO register shown in FIG. 50 may also be performed from right to left. At this time, the first row of the FIFO register is initialized to I ₀ , X, X, X, X from right to left, the second row is initialized to I ₁ , X, X, X, X from right to left, and The fourth row may be initialized to I ₂ , X, X, X, X from right to left, and the fourth row may be initialized to I ₃ , X, X, X, X from right to left.

In the example shown in FIG. 50 , the memory corresponding to the convolutional delay line may be conceptual for explaining the output of the convolutional delay line, and according to the embodiment, the convolutional delay line has a separate output memory. may or may not be included.

At the timing corresponding to the first subframe, the data (20, 11, 2) stored in the first row of the memory corresponding to the twisted block interleaver in the memory (CDL memory) corresponding to the convolutional delay line, and the third The data stored in the row (15, 6, 19), the data stored in the fifth row (23, 14, 5), the data stored in the seventh row (16, 22, 13) and the initialization of the FIFO register values can be stored. At this time, data initialized in the FIFO register may be stored in a memory corresponding to the convolutional delay line. That is, at the timing corresponding to the first subframe, 2, 11, and 20 are stored from right to left in the first row of the memory corresponding to the convolutional delay line, and two new virtual cells X and X can be stored. there is. 19, 6, and 15 may be stored from right to left in the third row of the memory corresponding to the convolutional delay line, and two new virtual cells X and X may be stored. 5, 14, and 23 may be stored from right to left in the fifth row of the memory corresponding to the convolutional delay line, and two new virtual cells X and X may be stored. 13, 22, 16, X, X may be stored from right to left in the seventh row of the memory corresponding to the convolutional delay line.

In this case, one initialization cell (I ₀ ) may be stored in the rightmost row of the memory corresponding to the convolutional delay line, and then four virtual cells may be stored. In this case, one initialization cell (I ₁ ) may be stored at the far right in the fourth row of the memory corresponding to the convolutional delay line, and then four virtual cells may be stored. In this case, one initialization cell (I ₂ ) may be stored in the rightmost row of the memory corresponding to the convolutional delay line, and then four virtual cells may be stored. In this case, one initialization cell (I ₃ ) may be stored at the far right in the eighth row of the memory corresponding to the convolutional delay line, and then four virtual cells may be stored.

That is, at the timing corresponding to the first subframe, the first row of the memory corresponding to the convolutional delay line stores 2, 11, 20, X, X from right to left, and the second row from right to left stores I ₀ , X, X, X, X, the third row stores 19, 6, 15, X, X from right to left, and the fourth row stores I ₁ , X, X from right to left , X, X, the fifth row stores 5, 14, 23, X, X from right to left, the sixth row stores I ₂ , X, X, X, X from right to left, , the seventh row can store 13, 22, 16, X, X from right to left, and the eighth row can store I ₃ , X, X, X, X from right to left. In this case, X may represent a virtual cell. At this time, the first row of the FIFO register shown in FIG. 50 stores 7, 1, 10, X, X from right to left, and the second row stores 0, 9, 18, X, X from right to left. The third row can store 8, 17, 4, X, X from right to left, and the fourth row can store 3, 12, 21, X, X from right to left.

In this case, the virtual cell stored in the memory corresponding to the convolutional delay line may not be output as a time interleaver output. That is, the first subframe transmits "2, 11, 20, I ₀ , 19, 6, 15, I ₁ , 5, 14, 23, I ₂ , 13, 22, 16, I ₃ ", and the second Subframes may transmit "..., 7, 1, 10, ..., 0, 9, 18, ..., 8, 17, 4, ..., 3, 12, 21". In this case, “I ₀ , I ₁ , I ₂ , I ₃ ” may correspond to some of the aforementioned initial values. In this case, "..." may correspond to the output of the next time interleaving block from the twisted block interleaver.

51 is a block diagram illustrating an example of the time deinterleaver shown in FIG. 8 or 12;

Referring to FIG. 51 , the time deinterleaver includes an inverse convolutional delay line 5110, a twisted block deinterleaver 5120, and a cell deinterleaver 5130.

The inverse convolutional delay line 5110 performs the reverse process of the convolutional delay line shown in FIG. 38 .

In this case, the inverse convolutional delay line 5110 may predict a location of a newly added virtual cell in the convolutional delay line of the transmission side, and perform the reverse process based on the predicted location of the virtual cell.

The twisted block deinterleaver 5120 performs the reverse process of the twisted block interleaver shown in FIG. 38 .

In this case, the twisted block deinterleaver 5120 may restore data cells and then newly generate and store virtual cells.

At this time, the twisted block deinterleaver 5120 may predict the position of the newly added virtual cell in the convolutional delay line of the transmitter and perform the reverse process in consideration of the predicted virtual cell position.

In this case, the twisted block deinterleaver 5120 may perform reverse processes of the column-wise write operation and the diagonal-wise read operation of the twisted block deinterleaving, excluding the virtual cells. .

The cell deinterleaver 5130 performs the reverse process of the cell interleaver shown in FIG. 38 .

52 is an operational flowchart illustrating a time interleaving method according to an embodiment of the present invention.

Referring to FIG. 52, the TI method according to an embodiment of the present invention performs cell interleaving corresponding to cells within a TI block (S5210).

Depending on the embodiment, step S5210 may be omitted.

Also, in the time interleaving method according to an embodiment of the present invention, twisted block interleaving corresponding to intra-subframe interleaving is performed (S5220).

In this case, in step S5220, twisted block interleaving may be performed using a column-wise write operation and a diagonal-wise read operation.

Also, in the time interleaving method according to an embodiment of the present invention, inter-subframe interleaving is performed using the output of the twisted block interleaving (S5230).

At this time, step S5230 may be performed using a convolutional delay line.

At this time, the convolutional delay line can read only data cells excluding virtual cells corresponding to the twisted block interleaving.

In this case, the convolutional delay line may store new virtual cells after each row of the data cells is written from the output of the twisted block interleaving and before the switches move to the next branch.

At this time, the new virtual cells are calculated by subtracting the number of FEC blocks (N _{FEC_TI} ) in the TI block of the interleaving frame from the maximum value (N _{FEC_TI_MAX} ) of the number of FEC blocks in the TI block of the interleaving frame in each branch. may be commensurate.

In this case, new virtual cells may not be output as an output of the inter-subframe interleaving.

In this case, the convolutional delay line may include branches corresponding to the time interleaving unit (N _IU ), and the remaining branches excluding the first branch among the branches may include one or more FIFO registers.

At this time, the convolutional delay line may output only some of the initial values stored in the FIFO register.

In this case, some of the initial values may correspond to one initialization cell for each of the remaining branches.

A time deinterleaving method according to an embodiment of the present invention may be provided in a manner corresponding to FIG. 52 .

For example, in the time deinterleaving method according to an embodiment of the present invention, inter-subframe deinterleaving corresponding to the reverse process of step S5230 and twisted interleaving corresponding to the reverse process of step S5220 are performed. It may include performing block deinterleaving. At this time, the time deinterleaving method according to an embodiment of the present invention may further include performing cell deinterleaving corresponding to the reverse process of step S5210.

As described above, in the time interleaving apparatus and method according to the present invention, the configuration and method of the embodiments described above are not limitedly applicable, but the above embodiments are all of the embodiments so that various modifications can be made. Alternatively, some of them may be selectively combined.

3810: cell interleaver
3820: twisted block interleaver
3830: convolutional delay line

Claims (9)

an inverse convolutional delay line that performs a reverse process of inter-subframe interleaving performed by the transmitter's convolutional delay line; and
Twisted block deinterleaver performing the reverse process of intra-subframe interleaving performed by the twisted block interleaver of the transmitter
A time deinterleaving device comprising a. The method of claim 1,
The intra-subframe interleaving is
A time deinterleaving device characterized in that it is performed using a column-wise write operation and a diagonal-wise read operation. The method of claim 1,
The time deinterleaving apparatus of claim 1 , wherein the convolutional delay line reads only data cells excluding virtual cells from the twisted block interleaver. The method of claim 3,
The convolutional delay line is
After each row of data cells is written from the twisted block interleaver, storing new virtual cells before moving switches to the next branch;
The new virtual cells are
Characterized in that each branch corresponds to a number obtained by subtracting the number of FEC blocks (N _{FEC_TI} ) in the TI block of the interleaving frame from the maximum value (N _{FEC_TI_MAX} ) of the number of FEC blocks in the TI block of the interleaving frame in each branch. Deinterleaving device. The method of claim 4,
The new virtual cells are
Time deinterleaving device characterized in that it is not output as an output of the inter-subframe interleaving. The method of claim 5,
The convolutional delay line is
A time deinterleaving device comprising branches corresponding to a time interleaving unit (N _IU ), and branches other than a first branch among the branches including one or more FIFO registers. The method of claim 6,
The convolutional delay line is
and outputting only some of the initial values stored in the FIFO register. The method of claim 7,
Some of these initial values are
and each of the remaining branches corresponds to one initialization cell in a switching unit. performing a reverse process of the inter-subframe interleaving performed by the transmitter's convolutional delay line; and
performing a reverse process of intra-subframe interleaving performed by a twisted block interleaver of the transmitter;
A time deinterleaving method comprising:

Images