JP7741736B2 - Receiving device and receiving method - Google Patents

Description

The present invention relates to a transmission system using Layered Division Multiplexing (LDM) for satellite broadcasting or terrestrial broadcasting, and in particular to an LDM receiving device and receiving method that receives and processes LDM signals in which two modulated signals, one for a higher layer (UL) and one for a lower layer (LL), are multiplexed at different power levels and then orthogonally modulated and transmitted in the same frequency band.

One wireless transmission method aimed at improving frequency utilization efficiency is hierarchical division multiplexing (hereinafter referred to as "LDM"), which multiplexes two modulated signals with different power levels and transmits them over the same frequency band. The use of LDM is also being considered for terrestrial digital broadcasting, and systems that use LDM to multiplex new terrestrial broadcasting formats with current terrestrial broadcasting are being considered (see, for example, non-patent documents 1 and 2).

The LDM method multiplexes two modulated signals in the same frequency band with a power difference between them. The higher power signal is generally referred to as the UL (Upper Layer) and the lower power signal as the LL (Lower Layer). The power ratio between the upper layer (UL) and the lower layer (LL) is defined by the IL (Injection Level), which represents the ratio of the average power of the UL to the average power of the LL.

Figure 11 is a block diagram showing the general configuration of a typical LDM transmitter 10 in the prior art. While Figure 11 illustrates the general configuration of an LDM transmitter 10 using single-carrier transmission, the LDM method can also be applied to multi-carrier OFDM transmission, and the configuration related to LDM signal transmission itself is similar. The LDM transmitter 10 shown in Figure 11 comprises an error correction coding unit 11 and mapping unit 12 for UL, an error correction coding unit 13 and mapping unit 14 for LL, a power adjustment unit 15, a combining unit 16, and an orthogonal modulation unit 17.

The error correction coding unit 11 inputs UL transmission information, which is information to be transmitted via the UL in the LDM method, applies a predetermined error correction coding process (such as an LDPC code) for the UL, and outputs the result to the mapping unit 12.

The mapping unit 12 maps the data obtained from the error correction coding unit 11 after error correction coding processing using a modulation method (such as QPSK) predetermined for UL, generates a modulated signal having UL symbols as an IQ signal that can be represented on the IQ plane with an in-phase component I and a quadrature-phase component Q, and outputs this to the combining unit 16.

The error correction coding unit 13 inputs LL transmission information, which is information to be transmitted on the LL of the LDM method, and applies a predetermined error correction coding process (such as an LDPC code) for the LL, outputting the result to the mapping unit 14. The error correction coding processes for the UL and LL may use the same error correction coding method and coding rate, or different error correction coding methods and coding rates.

The mapping unit 14 maps the error-correction-encoded data obtained from the error-correction encoding unit 13 using a modulation method (such as QPSK) predetermined for LL, generates a modulated signal having LL symbols as an IQ signal, and outputs it to the power adjustment unit 15. The modulation methods for UL and LL may be the same or different, and may be other than QPSK, such as π/2-shift BPSK, 8PSK, or 16APSK.

The power adjustment unit 15 adjusts the amplitude of the LL symbols obtained from the mapping unit 14 using a predetermined IL value (i.e., adjusts the power ratio between layers according to the IL), and outputs the amplitude-adjusted LL symbols to the combining unit 16. Note that while an example configuration is shown here in which the power adjustment unit 15 adjusts the amplitude of the LL symbols, it is sufficient that the power adjustment unit 15 can adjust the power ratio between layers according to the IL, and such a power adjustment unit 15 may adjust the amplitude of either or both of the UL and LL symbols.

The combiner 16 generates an LDM symbol by adding the power of each UL and LL symbol obtained from the mapping unit 12 and power adjustment unit 15, respectively, and outputs it to the orthogonal modulation unit 17.

The quadrature modulation unit 17 generates an LDM signal by quadrature modulating the LDM symbols obtained from the combiner 16 in the same frequency band according to a modulation method predetermined for UL, and transmits the signal through a specified power amplification (not shown) from the transmitting antenna 18 to the receiving device (including via a terrestrial or satellite repeater).

In a receiving device that receives this LDM signal, the reception performance is greatly affected by the accuracy of UL and LL separation during reception processing, so various UL and LL layer separation methods are being considered. The most common layer separation method is SIC (Successive Interference Cancellation), which generates replica signals and performs power subtraction processing (see, for example, Non-Patent Document 1).

Figure 12 is a block diagram showing the general configuration of a typical SIC-type receiving device 20S in the prior art. The SIC-type receiving device 20S shown in Figure 12 includes an orthogonal demodulation unit 22, a UL LLR calculation unit 23 and error correction decoding unit 24, a UL re-modulation replica signal generation unit 25, a subtraction unit 26, and an LL LLR calculation unit 27 and error correction decoding unit 28.

The orthogonal demodulation unit 22 receives an LDM signal via the receiving antenna 21, in which two modulated signals, one for a higher layer (UL) and one for a lower layer (LL), are multiplexed at different power levels and then orthogonally modulated and transmitted in the same frequency band. The received LDM signal is then subjected to power (amplitude) normalization processing using a predetermined gain adjustment (not shown) based on the average power or maximum power, before being input. Symbol synchronization and carrier recovery are then performed, and the LDM symbol is extracted through orthogonal demodulation processing, which is then output to the UL LLR calculation unit 23 and subtraction unit 26.

The LLR calculation unit 23 regards the LDM symbols obtained from the orthogonal demodulation unit 22 as UL symbols, calculates log-likelihood ratios (LLRs) using reference points indicating all possible symbol positions for the LDM symbols (when UL is QPSK and LL is QPSK, the 16-symbol reference points shown in FIG. 4, described below), generates a UL signal indicating the log-likelihood ratios (LLRs), and outputs it to the error correction decoding unit 24. The LLR calculation unit 23 may also regard the LDM symbols obtained from the orthogonal demodulation unit 22 as UL symbols, calculates log-likelihood ratios (LLRs) using reference points indicating symbol positions according to a predetermined modulation method for UL (not shown, but when UL is QPSK, the 4-symbol reference points corresponding to QPSK for UL, regardless of the LL modulation method), and generates a UL signal indicating the log-likelihood ratios (LLRs). In this case, the accuracy of the LLR decreases, especially when the IL is small, but it is the same as the LLR calculation method used in general error correction decoding processing.

The error correction decoding unit 24 performs decoding processing on the UL signal obtained from the LLR calculation unit 23, corresponding to an error correction coding processing (such as an LDPC code) predetermined for the UL on the transmitting side, and generates decoded bits of the UL signal indicating UL transmission information, which are output externally and also to the replica signal generation unit 25.

The replica signal generation unit 25 re-performs error correction coding processing (such as LDPC code) predetermined for the UL on the transmitting side on the decoded bits of the UL signal obtained from the error correction decoding unit 24, and then re-modulates the decoded bits by performing orthogonal modulation processing according to a modulation method predetermined for the UL, thereby generating a replica signal indicating the UL symbol and outputting it to the subtraction unit 26.

The subtraction unit 26 performs delay adjustment for the LDM symbol obtained from the orthogonal demodulation unit 22 up until the replica signal generation unit 25 generates the replica signal, and then subtracts the power of the replica signal indicating the UL symbol from the LDM symbol to obtain the LL symbol, and outputs the LL symbol to the LLR calculation unit 27.

The LLR calculation unit 27 calculates the log-likelihood ratio (LLR) for the LL symbol obtained from the subtraction unit 26 using a reference point indicating the symbol position according to a modulation method predetermined for LL, generates an LL signal indicating the log-likelihood ratio (LLR), and outputs it to the error correction decoding unit 28.

The error correction decoding unit 28 performs a decoding process on the LL signal obtained from the LLR calculation unit 27, corresponding to an error correction coding process (such as an LDPC code) predetermined for the LL on the transmitting side, and generates decoded bits of the LL signal indicating the transmission information for the LL, which are then output externally.

This type of SIC-based layer separation method has high separation accuracy and can achieve good reception performance. However, the reception processing using the above-mentioned replica signal generation and power subtraction process has the problem of significantly higher costs compared to general receiving devices that do not rely on the LDM method.

Therefore, a method has been disclosed in which the LDM symbols obtained by performing orthogonal demodulation processing are decoded separately for the UL and LL (hereinafter referred to as "collective demodulation"), omitting the generation of replica signals and power subtraction processing (see, for example, Patent Document 1).

Figure 13 is a block diagram showing the general configuration of a typical batch demodulation receiving device 20B in the prior art. Note that in Figure 13, components similar to those shown in Figure 12 are assigned the same reference numerals.

The batch demodulation receiving device 20B shown in FIG. 13 includes an orthogonal demodulation unit 22, an LLR calculation unit 23 and an error correction decoding unit 24 for UL, and an LLR calculation unit 27B and an error correction decoding unit 28 for LL.

In the batch demodulation type receiving device 20B shown in Figure 13, the orthogonal demodulation unit 22, LLR calculation unit 23, error correction decoding unit 24, and error correction decoding unit 28 each function in the same way as those shown in Figure 12, and their explanation will be omitted, but the difference is that the LLR calculation unit 27B shown in Figure 13 replaces the LLR calculation unit 27 shown in Figure 12.

That is, the LLR calculation unit 27B shown in Figure 13 regards the LDM symbols obtained from the orthogonal demodulation unit 22 as symbols for LL, calculates the log-likelihood ratio (LLR) using reference points indicating all possible symbol positions for the LDM symbol, generates an LL signal indicating the log-likelihood ratio (LLR), and outputs it to the error correction decoding unit 28.

The error correction decoding unit 28 shown in Figure 13 performs decoding processing on the LL signal obtained from the LLR calculation unit 27B, corresponding to an error correction coding processing (such as an LDPC code) predetermined for the LL on the transmitting side.

However, while this type of layer separation method using batch demodulation can reduce the costs associated with reception processing compared to SIC, it has low noise tolerance and it is difficult to achieve high reception performance in environments with low reception C/N ratios.

Japanese Patent Application Laid-Open No. 2019-87894

Akihiko Sato and 11 others, "A Study on the Application of LDM for Next-Generation Terrestrial Broadcasting," ITE Technical Report, Vol. 41, No. 6, BCT2017-34, pp. 45-48, February 2017 Hiromasa Okada and six others, "A Study on Improving Various Issues When Applying LDM to Terrestrial Digital Broadcasting - Examination of New Broadcasting System Reception Area Expansion Methods and Synchronization Methods," Institute of Image Information and Television Engineers Technical Report, Vol. 42, No. 28, BCT2018-76, pp. 13-16, September 2018.

As mentioned above, two types of layer separation methods, "SIC" and "batch demodulation," are known as conventional techniques for receiving LDM signals.

However, while SIC can achieve high accuracy in separating UL and LL layers and good reception performance, it has the problem of increasing the costs associated with reception processing.

While batch demodulation reduces the cost of reception processing, it has the drawback of making it difficult to achieve high reception performance.

In view of the above-mentioned problems, the object of the present invention is to provide a receiving device and receiving method that reduces the receiving processing of layer division multiplexing (LDM) systems while improving receiving performance with high layer separation accuracy.

The receiving device of the present invention is a hierarchical division multiplexing (LDM) receiving device, which receives an LDM signal in which two modulated signals, a high layer (UL) and a low layer (LL), are multiplexed at different powers and orthogonally modulated in the same frequency band and transmitted, and performs orthogonal demodulation processing to extract an LDM symbol; an UL LLR calculation unit that regards the LDM symbol as a UL symbol and calculates a log-likelihood ratio using all symbol positions that the LDM symbol can take or a reference point indicating a symbol position according to a modulation method predetermined for UL, and generates a UL signal indicating the log-likelihood ratio; an UL error correction decoding unit that performs error correction decoding processing predetermined for UL on the UL signal, and generates decoded bits for the UL signal; and The system is characterized by comprising: a reference point determination unit that determines the reference point used to calculate the log-likelihood ratio for LL decoding according to the decoded bits of the UL signal by narrowing down the reference points that indicate the symbol positions that the LL symbol in the received LDM symbol can take from among the reference points that indicate all symbol positions that the LDM symbol can take, based on the decoded bits of the UL signal; an LL LLR calculation unit that calculates the log-likelihood ratio for the received LDM symbol corresponding to the UL signal based only on the reference points determined by the reference point determination unit and generates an LL signal that indicates the log-likelihood ratio; and an LL error correction decoding unit that performs a predetermined error correction decoding process for LL on the LL signal and generates decoded bits of the LL signal.

Furthermore, in the receiving device of the present invention, the UL error correction decoding unit is characterized in that it attaches identification information that enables the reference point determination unit to identify the LDM symbol corresponding to the UL signal, and feeds back the decoding result indicating the decoded bits of the UL signal to the reference point determination unit.

Furthermore, in the receiving device of the present invention, the LL LLR calculation unit variably adjusts the delay of the received LDM symbol based on the identification information in order to generate an LL signal corresponding to the UL signal, and then calculates a log-likelihood ratio for the received LDM symbol corresponding to the UL signal using only the reference point determined by the reference point determination unit, and generates an LL signal indicating the log-likelihood ratio.

Furthermore, in the receiving device of the present invention, the LL LLR calculation unit performs delay adjustment on the received LDM symbol for a fixed time period required to generate an LL signal corresponding to the UL signal, and then calculates a log-likelihood ratio for the received LDM symbol corresponding to the UL signal using only the reference point determined by the reference point determination unit, and generates an LL signal indicating the log-likelihood ratio.

In addition, in the receiving device of the present invention, the UL LLR calculation unit calculates the log-likelihood ratio using reference points indicating all possible symbol positions of the LDM symbol, and further includes a Euclidean distance accumulation unit that temporarily accumulates, while updating, information on the Euclidean distance to each reference point for each received LDM symbol used in calculating the log-likelihood ratio indicated as the UL signal for a predetermined number of symbols; the reference point determination unit has a Euclidean distance selection unit that selects, for each LDM symbol decoded for the UL signal by the UL error correction decoding unit, the Euclidean distance of a reference point to be used in decoding the LL signal from the Euclidean distances of each reference point accumulated in the Euclidean distance accumulation unit based on the decoded bits of the UL signal; and the LL LLR calculation unit calculates the log-likelihood ratio for the received LDM symbol corresponding to the UL signal using only the Euclidean distance selected for decoding the LL signal by the Euclidean distance selection unit, and generates an LL signal indicating the log-likelihood ratio.

Furthermore, the receiving method of the present invention is a hierarchical division multiplexing (LDM) method, which includes the steps of receiving an LDM signal in which two modulated signals, one for a higher layer (UL) and one for a lower layer (LL), are multiplexed at different power levels and transmitted by orthogonal modulation in the same frequency band, and performing orthogonal demodulation processing to extract an LDM symbol; regarding the LDM symbol as a UL symbol, calculating a log-likelihood ratio using all symbol positions that the LDM symbol can take or a reference point indicating a symbol position according to a modulation method predetermined for UL, and generating a UL signal indicating the log-likelihood ratio; and performing error correction decoding processing predetermined for UL on the UL signal, and generating decoded bits for the UL signal. The method includes the steps of: determining a reference point to be used in calculating a log-likelihood ratio for LL decoding according to the decoded bits of the UL signal by narrowing down reference points indicating possible symbol positions for LL symbols in the received LDM symbols from among all reference points indicating possible symbol positions for LDM symbols based on the decoded bits of the UL signal; calculating a log-likelihood ratio for the received LDM symbol corresponding to the UL signal based only on the reference point determined in this step, and generating an LL signal indicating the log-likelihood ratio; and performing a predetermined error correction decoding process for LL on the LL signal, and generating decoded bits for the LL signal.

The present invention enables hierarchical division multiplexing (LDM) reception processing to achieve excellent reception performance with noise resistance equivalent to or better than SIC, while also significantly reducing the costs associated with reception processing compared to SIC.

1 is a block diagram showing a schematic configuration of an LDM receiving device according to a first embodiment of the present invention; 1 is an IQ plane diagram showing an example of allocation of bits constituting symbols of an LDM signal in an LDM receiving device according to a first embodiment of the present invention; 4 is a flowchart showing a receiving process in the LDM receiving device according to the first embodiment of the present invention. 1 is an IQ plane diagram for explaining the processing of an LLR calculation unit for UL in an LDM receiving device according to a first embodiment of the present invention; 10A and 10B are diagrams for explaining the processing of a reference signal determination unit and an LLR calculation unit for LL in an LDM receiving device according to a first embodiment of the present invention; 1 is a diagram showing C/N vs. BER characteristics under white noise, comparing the LDM receiving device of the first embodiment of the present invention with the prior art; 1 is a diagram showing the required C/N ratio for a comparison between an LDM receiving device according to the first embodiment of the present invention and the prior art; FIG. 10 is a block diagram showing a schematic configuration of an LDM receiving device according to a second embodiment of the present invention. 10 is a flowchart showing a receiving process in an LDM receiving device according to a second embodiment of the present invention. 10A and 10B are diagrams illustrating processing performed by a Euclidean distance storage unit in an LDM receiving device according to a second embodiment of the present invention; 1 is a block diagram showing a schematic configuration of a typical LDM transmitter 10 in the prior art; FIG. 1 is a block diagram showing a schematic configuration of a typical SIC type receiving device 20S in the prior art. FIG. 1 is a block diagram showing a schematic configuration of a typical batch demodulation receiving device 20B in the prior art.

The following describes the LDM receiver 20 and receiving method according to each embodiment of the present invention, with reference to the drawings.

[Receiving device of embodiment 1]
Fig. 1 is a block diagram showing a schematic configuration of an LDM receiving device 20 according to a first embodiment of the present invention. The LDM receiving device 20 according to the first embodiment of the present invention shown in Fig. 1 receives and demodulates an LDM signal transmitted from an LDM transmitting device 10 using single-carrier transmission as shown in Fig. 11, and performs layer separation of the LDM signal into a higher layer (UL) and a lower layer (LL) to decode the signal. The LDM receiving device 20 includes an orthogonal demodulator 22, an LLR calculation unit 23 and an error correction decoding unit 24 for UL, a reference point determination unit 25R, and an LLR calculation unit 27R and an error correction decoding unit 28 for LL. Note that in Fig. 1, components similar to those shown in Fig. 12 are designated by the same reference numerals. Furthermore, while Figure 1 illustrates the general configuration of a receiving device 20 using an LDM system with single-carrier transmission, the device can also be configured in a similar manner when receiving and demodulating an LDM signal transmitted by multi-carrier OFDM transmission, and performing layer separation into a higher layer (UL) and a lower layer (LL) in the LDM signal and decoding the signal.

In other words, the LDM receiving device 20 of Example 1 according to the present invention shown in FIG. 1 is similar to the SIC receiving device 20S shown in FIG. 12 in that it includes an orthogonal demodulation unit 22, an LLR calculation unit 23 for UL, an error correction decoding unit 24, and an error correction decoding unit 28 for LL. However, it differs in that it includes a "reference point determination unit 25R" instead of the SIC-type "UL re-modulation replica signal generation unit 25 and subtraction unit 26," and further includes an "LL LLR calculation unit 27R" according to the present invention that performs LLR calculations using only the reference point determined by the "reference point determination unit 25R" instead of the SIC-type "LL LLR calculation unit 27."

The following describes in more detail the configuration of the LDM receiver 20 of the first embodiment of the present invention shown in Figure 1.

The orthogonal demodulation unit 22 receives an LDM signal via the receiving antenna 21, in which two modulated signals, one for a higher layer (UL) and one for a lower layer (LL), are multiplexed at different power levels and then orthogonally modulated and transmitted in the same frequency band. The received LDM signal undergoes power (amplitude) normalization processing using a predetermined gain adjustment (not shown) based on the average power or maximum power, and is then input. It then performs symbol synchronization and carrier recovery, performs orthogonal demodulation processing to extract the LDM symbol, and outputs it to the UL LLR calculation unit 23 and the LL LLR calculation unit 27R. Here, the LDM signal received via the receiving antenna 21 is a signal transmitted from the LDM-based transmitter 10 shown in FIG. 11.

The LLR calculation unit 23 regards the LDM symbols obtained from the orthogonal demodulation unit 22 as UL symbols, calculates the log-likelihood ratio (LLR) using all possible symbol positions of the LDM symbols or reference points indicating symbol positions according to a predetermined modulation method for UL, generates a UL signal indicating the log-likelihood ratio (LLR), and outputs it to the error correction decoding unit 24.

The error correction decoding unit 24 performs decoding processing on the UL signal obtained from the LLR calculation unit 23, corresponding to an error correction coding processing (such as an LDPC code) predetermined for the UL on the transmitting side, to generate decoded bits of the UL signal indicating UL transmission information, which are output externally and also fed back to the reference point determination unit 25R.

Based on the decoded bits of the UL signal obtained from the error correction decoding unit 24, the reference point determination unit 25R narrows down the reference points indicating the symbol positions that the LL symbols in the received LDM symbols can take from the reference points indicating all the symbol positions that the LDM symbols can take, thereby determining the reference points to be used in calculating the log-likelihood ratio (LLR) for LL decoding according to the decoded bits of the UL signal, and outputs the reference points to the LLR calculation unit 27R.

The LLR calculation unit 27R performs a predetermined delay adjustment on the LDM symbols obtained from the orthogonal demodulation unit 22 to generate an LL signal corresponding to the UL signal, and then, using only the reference point determined by the reference point determination unit 25R, calculates the log-likelihood ratio (LLR) for the received LDM symbol corresponding to the UL signal for which the reference point was determined, generates an LL signal indicating this log-likelihood ratio (LLR), and outputs it to the error correction decoding unit 28.

The error correction decoding unit 28 performs a decoding process on the LL signal obtained from the LLR calculation unit 27R, corresponding to an error correction coding process (such as an LDPC code) predetermined for the LL on the transmitting side, and generates and outputs decoded bits of the LL signal indicating the transmission information for the LL.

As such, in the receiving device 20 shown in Figure 1, the UL decoding process itself is the same as the conventional SIC method, but instead of having the SIC-type "UL re-modulation replica signal generation unit 25 and subtraction unit 26," it is equipped with a "reference point determination unit 25R" and an "LL LLR calculation unit 27R" that performs LL LLR calculations using only the reference point determined by the "reference point determination unit 25R." This makes it possible to significantly reduce the cost associated with LL receiving processing.

[Reception process of embodiment 1]
The reception process in the receiving device 20 shown in FIG. 1 will be described below with more specific examples.

First, in the receiving device 20 shown in Figure 1, the LDM signal received via the receiving antenna 21 is a signal transmitted from the LDM transmitting device 10 shown in Figure 11, and the bits that make up the LDM symbol in this embodiment are configured so that the bits that make up the UL symbol are arranged on the higher order side and the bits that make up the LL symbol are arranged on the lower order side.

For example, Figure 2 shows an IQ plane diagram centered on the origin O when the UL and LL modulation methods are QPSK and the bit allocation to the QPSK symbols is Gray code mapped, as an example of the allocation of bits that make up the symbols of an LDM signal (LDM symbols). In this case, the LDM symbols are composed of a total of 16 symbols, and the bits allocated to these 16 symbols are, as shown in Figure 2 (for convenience of explanation, shown as 16QAM), with the most significant 2 bits being UL and the least significant 2 bits being LL. However, the UL and LL bit allocation in the LDM symbols does not need to be limited to that shown in Figure 2, as long as the UL and LL bits can be distinguished on the LDM symbol.

As a representative example, we will explain an example in which the UL and LL modulation schemes are QPSK and the orthogonal demodulation unit 22 shown in Figure 1 receives and processes LDM symbols using Gray code mapping as shown in Figure 2. Note that the present invention also applies to cases where set partitioning mapping is used instead of Gray code mapping.

Figure 3 is a flowchart showing the reception process in the LDM receiving device 20 of embodiment 1 of the present invention.

First, the receiving device 20 extracts, as a "received symbol," an LDM symbol that has been subjected to orthogonal demodulation processing by the orthogonal demodulation unit 22 from the LDM signal received via the receiving antenna 21 (step S1). Here, the LDM symbol output by the orthogonal demodulation unit 22 is output as a received symbol #n, with a symbol number assigned to make it identifiable for subsequent processing. Furthermore, the receiving device 20 regards the LDM symbol obtained from the orthogonal demodulation unit 22 (i.e., the received symbol) as a UL symbol, and calculates a log-likelihood ratio (LLR) using the LLR calculation unit 23, using all possible symbol positions for the LDM symbol or a reference point indicating a symbol position according to a predetermined modulation method for UL.

Figure 4 is an IQ plane diagram for explaining the processing of the UL LLR calculation unit 23 in the LDM-based receiving device 20 of embodiment 1 of the present invention. As shown in Figure 4, the 16-symbol reference points determined by the QPSK modulation method for the UL and LL and the IL represent all possible symbol positions for the LDM symbol. Therefore, when calculating the log-likelihood ratio (LLR) for received symbol #1 (indicated by the "△" in Figure 4) as the LDM symbol, the UL LLR calculation unit 23 calculates the log-likelihood ratio (LLR) of each bit in the error-correcting code block used by the error-correcting decoder 24 from the Euclidean distance between received symbol #1 and each of the 16-symbol reference points.

Then, the receiving device 20 uses the error correction decoding unit 24 to perform decoding processing on the UL signal indicated by the log-likelihood ratio (LLR) obtained from the LLR calculation unit 23, corresponding to an error correction coding processing (such as an LDPC code) predetermined for the UL on the transmitting side, and generates decoded bits of the UL signal as the decoding result, which are output externally.

The receiving device 20 also causes the error correction decoder 24 to feedback and output the decoded bits of the UL signal generated as the decoding result to the reference point determiner 25R (step S2). Here, we consider received symbol #1 as the LDM symbol, but in actual processing, the error correction decoder 24 sequentially feeds back the decoded results of the UL signal for each received symbol #n to the reference point determiner 25R. Preferably, the error correction decoder 24 sequentially feeds back the decoded results of the UL signal for each received symbol #n to the reference point determiner 25R, along with identification information (e.g., a symbol number) that enables the LDM symbol corresponding to the UL signal to be identified.

Next, the receiving device 20 determines the reference point used to calculate the log-likelihood ratio (LLR) for LL decoding according to the decoded bits of the UL signal by using the reference point determination unit 25R to narrow down the reference points indicating the possible symbol positions of the LL symbols in the received LDM symbols (i.e., received symbols) from among the reference points indicating all possible symbol positions of the LDM symbols (16 symbols in this example) based on the decoded bits of the UL signal (step S3).

For example, in the bit allocation example for the LDM symbol shown in Figure 2, the symbol can be divided into the first to fourth quadrants, R1 to R4, with the most significant two bits being UL and the least significant two bits being LL. Therefore, when the UL decoding result is 00, the reference point determination unit 25R determines the reference point of four symbols in the R1 region of the first quadrant as the reference point to be used in calculating the log-likelihood ratio (LLR) for LL decoding (step S4). When the UL decoding result is 10, the reference point determination unit 25R determines the reference point of four symbols in the R2 region of the second quadrant as the reference point to be used in calculating the log-likelihood ratio (LLR) for LL decoding (step S5). When the UL decoding result is 11, the reference point determination unit 25R determines the reference point of four symbols in the R3 region of the third quadrant as the reference point to be used in calculating the log-likelihood ratio (LLR) for LL decoding (step S6). Furthermore, when the UL decoding result is 01, the reference point determination unit 25R determines the reference points of the four symbols in the fourth quadrant R4 area as the reference points to be used in calculating the log-likelihood ratio (LLR) for LL decoding (step S7). By feeding back the UL decoding result to the reference point determination unit 25R in this way, it is possible to narrow down the reference points to be used for LL decoding.

Next, the receiving device 20 performs a predetermined delay adjustment on the LDM symbol obtained from the orthogonal demodulation unit 22 using the LLR calculation unit 27R to generate an LL signal corresponding to the UL signal, and then calculates the log-likelihood ratio (LLR) for the received LDM symbol corresponding to the UL signal for which the reference point was determined, using only the reference point determined by the reference point determination unit 25R (step S8).

Figure 5 is a diagram illustrating the processing of the reference signal determination unit 25R and the LLR calculation unit 27R for LL in the LDM receiving device 20 of the first embodiment of the present invention.

As shown in FIG. 5, the UL error correction decoder 24 feeds back the UL decoding results to the reference point determiner 25R. Preferably, the error correction decoder 24 sequentially feeds back the UL signal decoding results for each received symbol #n to the reference point determiner 25R, along with identification information (e.g., a symbol number) that identifies the LDM symbol corresponding to that UL signal. The reference point determiner 25R can then narrow down the reference points that indicate the possible symbol positions of the LL symbol in the received LDM symbol (i.e., received symbol #1) from among the reference points that indicate all possible symbol positions of the LDM symbol (16 symbols in this example). In the example shown in FIG. 5, when the UL decoding result is 00, the reference points of four symbols in the R1 region of the first quadrant are determined as the reference points to be used in calculating the log-likelihood ratio (LLR) for LL decoding, and preferably notify the LLR calculator 27R of this together with the identification number (e.g., the symbol number).

Then, the LL LLR calculation unit 27R calculates the Euclidean distance between each of the four symbol reference points (R1 in the example shown in Figure 5) notified by the reference point determination unit 25R and the received symbol #1, and converts the Euclidean distance into a probability to determine the log-likelihood ratio (LLR) of each bit in the error correction code (LDPC, etc.) block in the LL error correction decoding unit 28, generates an LL signal indicating this log-likelihood ratio (LLR), and outputs it to the error correction decoding unit 28.

Here, when the error correction decoder 24 feeds back and notifies the LLR calculator 27R of the decoded result of the UL signal via the reference point determiner 25R, it is preferable to attach identification information (e.g., a symbol number) to the decoded result of the UL signal. In this case, the LLR calculator 27R can variably and accurately adjust the delay of the received LDM symbol based on the identification information in order to generate an LL signal corresponding to the UL signal. After performing this delay adjustment, the LLR calculator 27R calculates a log-likelihood ratio (LLR) for the received LDM symbol corresponding to the UL signal using only the reference point determined by the reference point determiner 25R, and generates an LL signal indicating the log-likelihood ratio (LLR). However, if the LLR calculation process and error correction decoding process for the UL signal by the LLR calculation unit 23 and the error correction decoding unit 24 are configured to be performed in a fixed time, such identification information (e.g., symbol number) can be omitted, and in this case, the LLR calculation unit 27R can perform delay adjustment for a fixed time to generate an LL signal corresponding to the UL signal.

Next, the receiving device 20 uses the error correction decoding unit 28 to perform decoding processing on the UL signal obtained from the LLR calculation unit 27R, corresponding to an error correction coding process (such as an LDPC code) predetermined for the LL on the transmitting side, and generates decoded bits of the LL signal as the decoding result, which are output externally.

In the receiving device 20 according to the present invention, configured as described above, it is possible to narrow down the reference point for calculating the LLR for LL decoding by feeding back the UL decoding results to the reference point determination unit 25R, thereby making it possible to significantly reduce the cost of receiving processing compared to the SIC type illustrated in FIG. 12. Furthermore, by narrowing down the reference point as in the receiving device 20 according to the present invention, it is possible to improve the LL decoding performance, and it is possible to achieve better receiving performance compared to the batch demodulation type illustrated in FIG. 13.

(simulation)
Next, a description will be given of the results of a simulation comparing the LDM type receiving device 20 of the first embodiment of the present invention with the prior art (SIC and collective modulation).

First, as simulation conditions, we assume single-carrier LDM transmission using the transmitter 10 illustrated in Figure 11, and that the information bits constituting the UL and LL transmission information are error-correction coded at coding rates of 3/5 and 4/5, respectively, using a parity check matrix for an LDPC code conforming to ISDB-S3. Furthermore, we assume that the LDPC block (44,880 bits) after error-correction coding is QPSK modulated (22,440 symbols) for both the UL and LL, and that the power of the UL and LL is adjusted to IL = 3 dB. We then assume that the transmitter 10 illustrated in Figure 11 generates an LDM signal that is quadrature-modulated from an LDM symbol combining the UL and LL symbols, and transmits the signal to the LDM receiver 20 of Example 1 of the present invention, simulating a white noise transmission channel.

Figure 6 is a diagram showing the C/N vs. BER characteristics under white noise, comparing the LDM-based receiving device 20 of Example 1 of the present invention with conventional technologies (SIC and collective modulation). Also, Figure 7 is a diagram showing the required C/N, comparing the LDM-based receiving device 20 of Example 1 of the present invention with conventional technologies (SIC and collective modulation). A pseudo-error-free state is defined as a bit error rate (BER) of 1E-11, and the C/N at this point is the required C/N. Since the required C/N is defined as the C/N at which both the UL and LL are pseudo-error-free, the BER of the LL, which has low noise tolerance, was evaluated.

From Figures 6 and 7, it can be seen that the reception performance of the receiving device 20 according to the present invention is almost equivalent to that of the conventional SIC method, and can significantly improve reception performance compared to batch demodulation.

Analyzing the results of Figures 6 and 7, it is clear that with the batch demodulation method, it is not possible to narrow down the reference points, so all 16 symbols shown in Figure 5 are used as reference points, and the Euclidean distance is calculated between received symbol #1 and all reference points to determine the LLR. In this case, the Euclidean distance between received symbol #1 and the reference point in the R2 region shown in Figure 5 (the point corresponding to UL = 10, LL = 01) is the smallest, so with the batch demodulation method, there is a high probability that the LL bit assigned to received symbol #1 is 01, which is thought to be a cause of error correction decoding failure due to incorrect judgment.

Therefore, according to the receiving device 20 of Example 1 of the present invention, by narrowing down the reference point for LL decoding from the UL decoding result as part of the layer division multiplexing (LDM) receiving process, it is possible to achieve good receiving performance with high noise resistance equivalent to or better than SIC, and also to significantly reduce the cost of receiving processing compared to SIC.

[Receiving device of embodiment 2]
8 is a block diagram showing a schematic configuration of an LDM receiving device 20 according to a second embodiment of the present invention. The LDM receiving device 20 according to the second embodiment of the present invention shown in FIG. 8 receives and demodulates an LDM signal transmitted from the LDM transmitting device 10 using single-carrier transmission shown in FIG. 11 , separates the LDM signal into a higher layer (UL) and a lower layer (LL), and decodes the signal. The LDM receiving device 20 includes an orthogonal demodulator 22, an UL LLR calculator 23 and an error correction decoder 24, a Euclidean distance storage unit 25E, a reference point determiner 25R having a Euclidean distance selector 26E, and an LL LLR calculator 27E and an error correction decoder 28. In FIG. 8, components similar to those shown in FIG. 1 are designated by the same reference numerals. Furthermore, while Figure 1 illustrates the general configuration of a receiving device 20 using an LDM system with single-carrier transmission, the device can also be configured in a similar manner when receiving and demodulating an LDM signal transmitted by multi-carrier OFDM transmission, and performing layer separation into a higher layer (UL) and a lower layer (LL) in the LDM signal and decoding the signal.

That is, the LDM receiving device 20 of the second embodiment of the present invention shown in FIG. 8 is similar to the first embodiment shown in FIG. 1 in that it includes an orthogonal demodulation unit 22, an LLR calculation unit 23 for UL, an error correction decoding unit 24, and an error correction decoding unit 28 for LL, but differs in that it includes a "Euclidean distance accumulation unit 25E," that the reference point determination unit 25R shown in FIG. 1 has a "Euclidean distance selection unit 26E," and that, instead of the "LL LLR calculation unit 27R" shown in FIG. 1, it includes an "LL LLR calculation unit 27E" that performs LLR calculations using only the Euclidean distance for the reference point selected by the "Euclidean distance selection unit 26E" from the Euclidean distances for each reference point already calculated at the time of UL decoding.

The following describes in more detail the configuration of the LDM receiver 20 according to the second embodiment of the present invention, as shown in Figure 8.

The quadrature demodulation unit 22 according to the second embodiment, like the first embodiment, receives an LDM signal via the receiving antenna 21, in which two modulated signals are multiplexed at different power levels as a high layer (UL) and a low layer (LL), and then quadrature modulated and transmitted in the same frequency band. The received LDM signal is then subjected to power (amplitude) normalization processing using a predetermined gain adjustment (not shown) based on the average power or maximum power, and input, performs symbol synchronization and carrier recovery, and performs quadrature demodulation processing to extract the LDM symbol, which is then output to the LLR calculation unit 23 for UL.

However, note that the orthogonal demodulation unit 22 according to the second embodiment differs from that of the first embodiment shown in FIG. 1 in that it is not configured to output the received symbol (received LDM symbol) to the LLR calculation unit 27E for LL. Here, the LDM signal received via the receiving antenna 21 is a signal transmitted from the LDM-based transmitting device 10 shown in FIG. 11.

As in Example 1, the LLR calculation unit 23 according to Example 2 regards the received symbols (received LDM symbols) obtained from the orthogonal demodulation unit 22 as symbols for UL, calculates log-likelihood ratios (LLRs) using reference points indicating all possible symbol positions of the LDM symbols, generates a UL signal indicating the log-likelihood ratios (LLRs), and outputs it to the error correction decoding unit 24. Note that Example 2 does not assume a configuration in which the LLR calculation unit 23 may calculate log-likelihood ratios (LLRs) using "reference points indicating symbol positions according to a modulation method predetermined for UL" as in Example 1, but calculates log-likelihood ratios (LLRs) using "reference points indicating all possible symbol positions of the LDM symbols."

However, unlike in Example 1 shown in FIG. 1, the LLR calculation unit 23 in Example 2 outputs, for each received symbol (received LDM symbol), information on the Euclidean distance for each reference point used when calculating the log-likelihood ratio (LLR) for the UL signal to the Euclidean distance accumulation unit 25E.

As in Example 1, the error correction decoding unit 24 according to Example 2 performs a decoding process on the UL signal obtained from the LLR calculation unit 23, corresponding to an error correction coding process (such as an LDPC code) predetermined for the UL on the transmitting side, to generate and externally output decoded bits of the UL signal indicating UL transmission information.

However, unlike in Example 1 shown in FIG. 1, the error correction decoding unit 24 in Example 2 outputs information on the decoded bits (decoding results) of the UL signal to a reference point determination unit 25R having a Euclidean distance selection unit 26E.

The Euclidean distance storage unit 25E inputs information on the Euclidean distance (r1 to r16) for each reference point of the received symbols used in the LLR calculation from the UL LLR calculation unit 23, and temporarily stores, while updating, information on the Euclidean distance (r1 to r16) for each reference point of each received symbol used in the calculation of the log-likelihood ratio (LLR) for UL decoding for a predetermined number of received symbols (received LDM symbols).

The reference point determination unit 25R according to the second embodiment, like that of the first embodiment, narrows down the reference points used to calculate the log-likelihood ratio (LLR) for LL decoding for each received symbol based on the UL decoded bits. However, it differs from the first embodiment in that it includes the functionality of a Euclidean distance selection unit 26E.

For each received symbol decoded by the error correction decoder 24 for the UL signal (received LDM symbol), the Euclidean distance selector 26E selects and reads the Euclidean distance of the reference point to be used for LL decoding from the Euclidean distances of each reference point stored in the Euclidean distance storage unit 25E based on the UL decoded bits, and outputs it to the LL LLR calculator 27E.

As a series of operations following the decoding of the UL signal in the error correction decoding unit 24, the LLR calculation unit 27E calculates, for each received symbol, a log-likelihood ratio (LLR) for the corresponding received symbol (received LDM symbol) using only the Euclidean distance selected for LL decoding by the Euclidean distance selection unit 26E, generates an LL signal indicating this log-likelihood ratio (LLR), and outputs it to the error correction decoding unit 28.

The error correction decoding unit 28 performs a decoding process on the LL signal obtained from the LLR calculation unit 27E that corresponds to an error correction coding process (such as an LDPC code) predetermined for the LL on the transmitting side, and generates and outputs decoded bits of the LL signal that indicate the transmission information for the LL.

As such, in the receiving device 20 shown in Figure 8, the UL decoding process itself is the same as the conventional SIC method, but instead of having a SIC-type "UL re-modulation replica signal generation unit 25 and subtraction unit 26," it is configured to have a "Euclidean distance accumulation unit 25E" and a "reference point determination unit 25R having a Euclidean distance selection unit 26E," and an "LL LLR calculation unit 27E" that calculates the log-likelihood ratio (LLR) for LL using only the Euclidean distance selected by the "Euclidean distance selection unit 26E." This reduces at least the Euclidean distance calculation process, making it possible to significantly reduce the cost associated with LL reception processing.

[Reception process of the second embodiment]
The reception process in the receiving device 20 shown in FIG. 8 will be described below with more specific examples.

First, in the receiving device 20 shown in Figure 8, the LDM signal received via the receiving antenna 21 is a signal transmitted from the LDM transmitting device 10 shown in Figure 11, and the bits constituting the LDM symbol in Example 2 are configured in the same way as in Example 1, with the bits constituting the UL symbol on the higher order side and the bits constituting the LL symbol on the lower order side.

For example, Figure 10(a) shows an IQ plane diagram centered on the origin O when the modulation method for UL and LL is QPSK and the bit allocation to the QPSK symbols is Gray code mapped, as an example of the allocation of bits that make up the symbols of an LDM signal (LDM symbols). In this case, the LDM symbols are composed of a total of 16 symbols, and the bits allocated to these 16 symbols are, as shown in Figure 10(a), with the most significant 2 bits being UL and the least significant 2 bits being LL.

Incidentally, for the sake of convenience in explaining the operation of Example 2, in Figure 10(a), the bits that make up the LDM symbol are represented as [a1 a2 a3 a4], where a1 indicates the first bit (UL), a2 indicates the second bit (UL), a3 indicates the third bit (LL), and a4 indicates the fourth bit (LL), and r1 to r16 indicate the Euclidean distance from a certain received symbol (here, received symbol #1) to each reference point. However, the allocation of UL and LL bits in the LDM symbol does not need to be limited to that shown in Figure 10(a), as long as the UL and LL bits can be identified on the LDM symbol.

As a representative example, we will explain an example in which the UL and LL modulation schemes are QPSK and the orthogonal demodulation unit 22 shown in Figure 8 receives and processes LDM symbols using the Gray code mapping shown in Figure 10(a). Note that the present invention also applies to cases where set partitioning mapping is used instead of Gray code mapping.

Figure 9 is a flowchart showing the reception process in the LDM receiving device 20 of embodiment 2 of the present invention.

First, in the receiving device 20 according to the second embodiment, as in the first embodiment, the orthogonal demodulation unit 22 extracts, as a "received symbol," an LDM symbol obtained by orthogonal demodulation processing of an LDM signal received via the receiving antenna 21 (step S11). Here, the LDM symbol output by the orthogonal demodulation unit 22 is output as a received symbol #n with a symbol number assigned so that it can be identified for subsequent processing. Furthermore, the receiving device 20 regards the LDM symbol obtained from the orthogonal demodulation unit 22 (i.e., the received symbol) as a UL symbol using the LLR calculation unit 23, and calculates a log-likelihood ratio (LLR) using reference points indicating all possible symbol positions for the LDM symbol.

As shown in Figure 10(a), the 16-symbol reference points determined by the IL and the QPSK modulation scheme for the UL and LL represent all possible symbol positions for the LDM symbol. Therefore, when calculating the log-likelihood ratio (LLR) for received symbol #1 (represented by the "△" in Figure 10(a)) as the LDM symbol, the UL LLR calculation unit 23 calculates the log-likelihood ratio (LLR) for each bit in the error correction code block used by the error correction decoding unit 24 from the Euclidean distance (r1 to r16) between received symbol #1 and each of the 16 symbol reference points.

Here, the receiving device 20 according to Example 2 differs from Example 1 in that it includes a Euclidean distance storage unit 25E. The Euclidean distance storage unit 25E receives information on the Euclidean distances (r1 to r16) of received symbols used to calculate the log-likelihood ratios (LLRs) from the UL LLR calculation unit 23, and temporarily stores, while updating, information on the Euclidean distances (r1 to r16) of received symbols used to calculate the log-likelihood ratios (LLRs) for UL decoding for a predetermined number of received symbols (received LDM symbols) (step S12). Figure 10(b) shows an example of storage, in the Euclidean distance storage unit 25E, of information on the Euclidean distances (r1 to r16 corresponding to symbol numbers #1 to #3 indicated by subscripts) of received symbols (e.g., symbol numbers #1 to #3) used to calculate the log-likelihood ratios (LLRs) by the UL LLR calculation unit 23. Although Figure 10(b) only shows three received symbols, symbol numbers #1 to #3, the Euclidean distance storage unit 25E is configured, in terms of its operating principle, to temporarily store information on the Euclidean distances (r1 to r16) for received symbols corresponding to the code length of the error correction code.

The receiving device 20 according to Example 2, as in Example 1, performs a decoding process corresponding to an error correction coding process (such as an LDPC code) predetermined for the UL on the transmitting side on the UL signal indicated by the log-likelihood ratio (LLR) obtained from the LLR calculation unit 23, using the error correction decoding unit 24, and generates decoded bits of the UL signal as the decoding result, which are output to the outside.

In the receiving device 20 according to Example 2, similar to Example 1, the reference point determination unit 25R narrows down the reference points used to calculate the log-likelihood ratio (LLR) for LL decoding for each received symbol based on the UL decoded bits. However, unlike Example 1, a Euclidean distance selection unit 26E is provided as a function of the reference point determination unit 25R. For received symbols (received LDM symbols) decoded for the UL signal by the error correction decoding unit 24, the Euclidean distance selection unit 26E selects and reads the Euclidean distance of the reference point used for LL decoding from the Euclidean distances of each reference point stored in the Euclidean distance storage unit 25E based on the UL decoded bits for each received symbol, and outputs it to the LLR calculation unit 27E for LL (step S13).

Next, unlike in Example 1, the receiving device 20 according to Example 2 calculates, as a series of operations following the decoding of the UL signal in the error correction decoding unit 24, a log-likelihood ratio (LLR) for each received symbol (received LDM symbol) using only the Euclidean distance selected for LL decoding by the Euclidean distance selection unit 26E by the LLR calculation unit 27E (step S14).

That is, the UL LLR calculation unit 23 calculates the log likelihood ratio (LLR) by replacing the Euclidean distance (r1 to r16) with a probability. For example, the probability P10 that the first bit a1 in the bit sequence [a1 a2 a3 a4] constituting the received symbol (received LDM symbol) #1 is ₀ and the probability _P11 that the first bit a1 is 1 can be expressed as in equations (1) and (2). If the LLR (likelihood) of the first bit a1 of the received symbol #1 shown in FIG. 10(a) is LLR _{1_#1} , it can be expressed as in equation (3). The LLR (likelihood) of the second bit a2 of the received symbol #1 can also be obtained in the same way as equations (1) to (3). In equations (1) and (2), σ represents the variance of white noise added to the received symbol calculated from the received C/N, and ^σ2 represents the noise power.

Next, in the receiving device 20 according to the second embodiment, as in the first embodiment, the error correction decoding unit 28 performs a decoding process corresponding to an error correction coding process (such as an LDPC code) predetermined for the LL on the transmitting side on the UL signal obtained from the LLR calculation unit 27R, and generates decoded bits of the LL signal as the decoding result, which are output to the outside.

In this way, the UL decoding process in Example 2, similar to Example 1, uses all possible symbol positions of the LDM symbol as reference points, calculates the Euclidean distance (r1 to r16) between the received symbol and each reference point (see Figure 10(a)), converts it to a probability, calculates the log-likelihood ratio (LLR) for UL in the LLR calculation unit 23, and generates decoded bits for UL in the error correction decoding unit 24.

However, in Example 2, the Euclidean distances (r1 to r16) calculated for all reference points corresponding to all possible symbol positions of the LDM symbol are stored in the Euclidean distance storage unit 25E. Then, as shown in FIG. 10(b), the Euclidean distance selection unit 26E selects from the Euclidean distance storage unit 25E the Euclidean distance required to calculate the LL log-likelihood ratio (LLR), which indicates the likelihood of the bit as an LL signal, based on the UL decoding result. The LL LLR calculation unit 27E calculates the LL log-likelihood ratio (LLR) using only this selected Euclidean distance, and the error correction decoding unit 28 generates decoded bits for LL.

In other words, the LL LLR calculation unit 27E can calculate the LL log-likelihood ratio (LLR) based on the Euclidean distance used to calculate the LLR in the UL LLR calculation unit 23 and information on the UL decoded bits indicating the UL decoding result. In particular, the Euclidean distance required to calculate the LL log-likelihood ratio (LLR) can be selected from the Euclidean distances (r1 to r16) used to calculate the UL log-likelihood ratio (LLR) based on the UL decoding result, thereby avoiding the use of Euclidean distances unnecessary for LL in probability calculations. In other words, the Euclidean distance accumulation unit 25E can narrow down the LL reference points from the UL decoding result among the Euclidean distances for all reference points accumulated for the received symbol, thereby eliminating the need to calculate the Euclidean distance again.

To explain the reduction in the calculation load more specifically, in the conventional "batch demodulation" process, the probability _P'30 that the third bit a3 is 0 and the probability P'31 that the third bit a3 is ₁ are expressed as in equations (4) and (5). Note that in equations (4) and (5), σ represents the variance of white noise added to the received symbol calculated from the received C/N, and ^σ2 represents the noise power.

On the other hand, in this embodiment, the Euclidean distance required to calculate the log-likelihood ratio (LLR) for LL is selected based on the UL decoding result. For example, when UL=00 (a1=0, a2=0) of received symbol #1, the probability _P30 that the third bit a3 is 0 and the probability _P31 that the third bit a3 is 1 can be expressed as in equations (6) and (7). If the LLR (likelihood) of the third bit a3 of received symbol #1 is _{LLR3_} #1, it can be expressed as in equation (8). Note that the LLR (likelihood) of the fourth bit a4 of received symbol #1 can also be obtained in the same way as in equations (6) to (8). Note that in equations (6) and (7), σ represents the variance of white noise added to the received symbol calculated from the received C/N, and ^σ2 represents the noise power.

In this way, in Example 2, compared to Example 1, information on the Euclidean distance for each reference point used in the calculation of the UL log-likelihood ratio (LLR) by the LLR calculation unit 23 for the received symbol is stored in the Euclidean distance storage unit 25E. This makes it possible to omit the Euclidean distance calculation involved in the calculation of the LL log-likelihood ratio (LLR), thereby reducing the processing cost.

Furthermore, the reception performance of Example 2 is equivalent to that of Example 1 (i.e., see Figures 6 and 7), and has the advantage that at least the amount of calculation required for Euclidean distance calculation can be reduced compared to Example 1.

Therefore, according to the receiving device 20 of the second embodiment of the present invention, in the receiving process of the Layer Division Multiplexing (LDM) system, the reference point for LL decoding is narrowed down from the UL decoding result, and further, the recalculation of the Euclidean distance required for LL decoding is omitted, thereby achieving good receiving performance with high noise resistance equivalent to or better than SIC, and further enabling a significant reduction in the cost of receiving processing compared to SIC.

While the above-described embodiments have been described as representative examples, it will be apparent to those skilled in the art that many modifications and substitutions can be made within the spirit and scope of the present invention. For example, in the above-described embodiments, only the LDM symbols shown in Figures 2 and 10(a) are described in which the UL and LL modulation schemes are QPSK, and only examples are described in which LDPC codes are used as error correction codes. However, the UL and LL modulation schemes may be different, and other modulation schemes besides QPSK may be π/2-shift BPSK, 8PSK, 16APSK, etc. Furthermore, various error correction codes other than LDPC codes may be used. Therefore, the present invention is not limited to the above-described embodiments, but is limited only by the scope of the claims.

The present invention can achieve excellent reception performance with noise resistance equivalent to or greater than that of SIC as reception processing for hierarchical division multiplexing (LDM), while also significantly reducing the cost of reception processing compared to SIC, making it useful for use in LDM transmission systems.

10 LDM transmitter 11 UL error correction encoder 12 UL mapping unit 13 LL error correction encoder 14 LL mapping unit 15 power adjustment unit 16 combiner 17 orthogonal modulator 18 transmitter antenna 20 LDM receiver 20S according to the present invention SIC receiver 20B according to the prior art Batch demodulation receiver 21 receiver antenna 22 orthogonal demodulator 23 UL LLR calculator 24 UL error correction decoder 25 replica signal generator 25R reference point determiner 25E according to embodiments 1 and 2 of the present invention Euclidean distance storage unit 26 according to embodiment 2 of the present invention subtractor 26E Euclidean distance selector 27, 27B according to embodiment 2 of the present invention LL LLR calculator 27R LL LLR calculator 27E according to embodiment 1 of the present invention LL LLR calculator 28 according to embodiment 2 of the present invention Error correction decoding unit for LL

Claims (6)

A receiving device of a hierarchical division multiplexing (LDM) system,
an orthogonal demodulation unit that receives an LDM signal in which two modulated signals, a high layer (UL) and a low layer (LL), are multiplexed with different power levels and orthogonally modulated in the same frequency band and transmitted, and performs orthogonal demodulation processing to extract an LDM symbol;
an LLR calculation unit for UL that regards the LDM symbol as a symbol for UL, calculates a log-likelihood ratio using all symbol positions that the LDM symbol can take or a reference point that indicates a symbol position according to a modulation method predetermined for UL, and generates a UL signal that indicates the log-likelihood ratio;
an error correction decoding unit for UL that performs a predetermined error correction decoding process for UL on the UL signal and generates decoded bits of the UL signal;
a reference point determination unit that determines a reference point to be used in calculating a log-likelihood ratio for LL decoding according to the decoded bits of the UL signal by narrowing down reference points that indicate symbol positions that LL symbols in the received LDM symbols can take from among reference points that indicate all symbol positions that LDM symbols can take, based on the decoded bits of the UL signal;
an LLR calculation unit for LL that calculates a log-likelihood ratio for a received LDM symbol corresponding to the UL signal based only on the reference point determined by the reference point determination unit and generates an LL signal indicating the log-likelihood ratio;
an error correction decoding unit for LL that performs a predetermined error correction decoding process for LL on the LL signal to generate decoded bits of the LL signal;
A receiving device comprising: The receiving device described in claim 1, characterized in that the UL error correction decoding unit attaches identification information that enables the reference point determination unit to identify the LDM symbol corresponding to the UL signal, and feeds back to the reference point determination unit the decoding result indicating the decoded bits of the UL signal. The receiving device described in claim 2, characterized in that, in order to generate an LL signal corresponding to the UL signal, the LL LLR calculation unit performs delay adjustment on the received LDM symbol variably based on the identification information, calculates a log-likelihood ratio for the received LDM symbol corresponding to the UL signal using only the reference point determined by the reference point determination unit, and generates an LL signal indicating the log-likelihood ratio. The receiving device described in claim 1, characterized in that the LL LLR calculation unit performs delay adjustment on the received LDM symbol for a fixed time period required to generate an LL signal corresponding to the UL signal, and then calculates a log-likelihood ratio for the received LDM symbol corresponding to the UL signal using only the reference point determined by the reference point determination unit, and generates an LL signal indicating the log-likelihood ratio. the LLR calculation unit for UL calculates a log-likelihood ratio using reference points indicating all possible symbol positions of the LDM symbol;
a Euclidean distance storage unit that temporarily stores, while updating, information on the Euclidean distance to each reference point for each received LDM symbol used in calculating the log-likelihood ratio indicated as the UL signal, for a predetermined number of symbols;
the reference point determination unit includes a Euclidean distance selection unit that selects, for each LDM symbol decoded for the UL signal by the UL error correction decoding unit, a Euclidean distance of a reference point to be used for decoding the LL signal from among the Euclidean distances of the reference points stored in the Euclidean distance storage unit, based on the decoded bits of the UL signal;
2. The receiving device according to claim 1, wherein the LLR calculation unit for LL calculates a log-likelihood ratio for the received LDM symbol corresponding to the UL signal using only the Euclidean distance selected by the Euclidean distance selection unit for decoding the LL signal, and generates an LL signal indicating the log-likelihood ratio. A receiving method for a layer division multiplexing (LDM) system, comprising:
receiving an LDM signal in which two modulated signals, a high layer (UL) and a low layer (LL), are multiplexed with different power levels and orthogonally modulated in the same frequency band and transmitted; and performing orthogonal demodulation processing to extract an LDM symbol;
a step of regarding the LDM symbol as a UL symbol, calculating a log-likelihood ratio using all possible symbol positions of the LDM symbol or a reference point indicating a symbol position according to a predetermined modulation method for UL, and generating a UL signal indicating the log-likelihood ratio;
performing a predetermined error correction decoding process for UL on the UL signal to generate decoded bits of the UL signal;
determining a reference point used for calculating a log-likelihood ratio for LL decoding according to the decoded bits of the UL signal by narrowing down reference points indicating symbol positions that can be taken by LL symbols in the received LDM symbols from among reference points indicating all symbol positions that can be taken by LDM symbols;
calculating a log-likelihood ratio for a received LDM symbol corresponding to the UL signal based solely on the reference point determined by the step, and generating an LL signal indicative of the log-likelihood ratio;
a step of performing an error correction decoding process predetermined for LL on the LL signal to generate decoded bits of the LL signal;
A receiving method comprising: