JP2003224538A - Transmitter, method for generating transmission signal, receiver and method for processing reception signal - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transmitter enabling to use a low bit digital-to-analog conversion circuit and a low bit analog-to-digital conversion circuit with low required accuracy, making the circuits unregulated and highly accurate and miniaturizing the scale of the circuits and to provide a transmission signal generating method, a receiver and a reception signal processing method. <P>SOLUTION: Oversampling processing can be performed by grounding a plurality of subcarriers without increasing a sampling frequency of a digital-to- analog converter and an analog-to-digital converter. This manner makes it possible to use low bit (e.g. one-bit) digital-to-analog conversion circuits 40a and 41a and low bit (e.g. one-bit) analog-to-digital conversion circuits 138a and 139a with low required accuracy to be able to make the circuits unregulated and highly accurate. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmitting device, a transmitting signal generating method, a receiving device and a receiving signal processing method, and more particularly to a transmitting device, a transmitting signal generating method, a receiving device and a receiving signal processing method of a multi-carrier transmission system. .

[0002]

2. Description of the Related Art In recent years, in the transmission of video signals or audio signals, digital modulation of high quality and high wave number utilization efficiency has been developed. Particularly in mobile communication, orthogonal frequency division multiplexing (hereinafter referred to as OF
DM (Orthogonal Frequency Division Multiplexing)
A method of modulating the data to be transmitted by modulation is considered.

In this OFDM, a large number of carrier waves (hereinafter, referred to as subcarriers) orthogonal to each other are used for transmitting data.
Is a modulation method in which each of the dispersed values is modulated. In addition to the characteristics that OFDM is less susceptible to multipath interference, OFDM has the advantage of high frequency utilization efficiency.

[0004] Conventionally, as an OFDM type transmitting apparatus for transmitting data by multicarrier, there is one having a configuration shown in FIG. In FIG. 24, the transmission device 10 transmits information data to an IDFT (Inverse Discrete Fourier Transfo
rm) Received by the calculation unit 11. The IDFT calculation unit 11 determines the N-point ID for each N samples with respect to the input information data.
By performing the FT operation, the information data is converted into a baseband signal composed of in-phase and quadrature components having N subcarriers, and each of the converted baseband signals is converted into a DA (digital analog) converter 12 or 13. Supply to.

The DA converters 12 and 13 perform digital-analog conversion on each baseband signal supplied from the IDFT operation unit 11, and then perform the digital-analog conversion on the baseband signal.
Supply to.

The quadrature modulator 14 quadrature-modulates a baseband signal consisting of in-phase and quadrature components with a carrier of a predetermined frequency, thereby performing IF (Intermediate frequen
cy) Frequency conversion to band signal. Thus, as shown in FIG. 25, the number N of subcarriers having the center frequency f _IF
A transmission IF signal (for example, N = 16) is obtained.

The quadrature-modulated transmission IF signal in the IF band is
After that, an RF (Radi
o Frequency) signal is frequency converted to a signal and transmitted via an antenna.

Further, as a receiver for receiving the data modulated by the OFDM system, there is one having a structure shown in FIG. In FIG. 26, the receiving device 20 uses a frequency converter (not shown) to convert the RF signal received through the antenna.
The frequency is converted into a reception IF signal in the IF band by the, and this is received by the quadrature detection unit 21.

The quadrature detector 21 quadrature-detects the input received IF signal with a carrier corresponding to the center frequency f _IF , thereby performing frequency conversion into a baseband signal composed of in-phase and quadrature components.

The baseband signal obtained by the quadrature detector 21 is supplied to AD (analog-digital) converters 22 and 23. The AD converters 22 and 23 AD-convert the input baseband signal at a sampling frequency equal to the subcarrier frequency interval, and supply this to a subsequent DFT (Discrete Fourier Transform) operation unit 24. The DFT calculation unit 24 executes an N-point DFT calculation for every N sample inputs. As a result, the quadrature detection unit 2
N received symbols are extracted from the baseband signal obtained in 1.

[0011]

However, in the conventional transmitting apparatus 10, when the number of subcarriers N increases, the required number of bits of the DA converters 12 and 13 increases, and the operation of the DA converters 12 and 13 is reduced. The required accuracy also increases. For example, taking the peak amplitude as an example, assuming that the number of subcarriers is N, the peak amplitude is N times that of one carrier, so that the required number of bits in the DA converters 12 and 13 is log ₂ N [bit]. ] Only increase. That is, N
= 1024 means an increase of 10 [bit].

OFDM is often used for wide band transmission, and a high sampling frequency of several MHz or higher is generally required for a DA converter. There is a problem that it is difficult to realize such a high-speed and high-precision DA converter.

When the number of subcarriers increases, I
There is a problem that the circuit scale of the DFT calculation unit 11 increases, and it becomes difficult to realize the miniaturization of the transmission device 10.

Further, in the conventional receiving apparatus 20, when the number N of subcarriers becomes large, the AD converters 22 and 2 are provided.
The number of required bits of 3 increases, and the accuracy required for the operation of the AD converters 22 and 23 also increases. For example, taking the peak amplitude as an example, assuming that the number of subcarriers is N, the peak amplitude is N times that of one carrier, so that the required number of bits in the AD converters 22 and 23 is log ₂ N [bit]. ] Only increase. That is, when N = 1024, it means an increase of 10 [bit].

The OFDM is often used for wide band transmission, and a high sampling frequency of several MHz or higher is generally required for the AD converter. There is a problem that it is difficult to realize such a high-speed and high-precision AD converter.

When the number of subcarriers increases, D
Since the circuit scale of the FT calculation unit 24 increases, there is a problem that it is difficult to realize the miniaturization of the receiving device 20.

The present invention has been made in view of the above points, and it is possible to use a low-bit digital-analog conversion circuit and a low-bit analog-digital conversion circuit with low required accuracy, and to make the circuit unadjusted and highly accurate. It is an object of the present invention to provide a transmitting device, a transmitting signal generating method, a receiving device, and a received signal processing method that can obtain the circuit size.

[0018]

A transmission apparatus of the present invention generates transmission data by inversely orthogonally transforming input data for each predetermined number of samples, thereby generating a plurality of groups of transmission data each consisting of a predetermined number of subcarriers. Means and
A plurality of low-bit digital-analog conversion means for digital-analog converting the transmission data of each group in parallel at a conversion frequency that can be restored according to the bandwidth when all the transmission data of each group are orthogonal frequency division multiplexed, Transmission signal generating means for generating an orthogonal frequency division multiplex signal composed of the plurality of groups of subcarriers as a transmission signal by adding the output signals of the plurality of low bit digital-analog converting means. .

According to this structure, by dividing the subcarriers of the orthogonal frequency division multiplexed signal to be transmitted into a plurality of groups, it is possible to narrow the signal bandwidth in each group. As a result, a predetermined oversampling ratio can be obtained without increasing the sampling frequency (conversion frequency) at the time of DA conversion, so that it is possible to use a low-bit digital-analog conversion circuit with low required accuracy, and the circuit No adjustment and higher accuracy can be achieved.

In the transmitting apparatus of the present invention having the above-mentioned configuration, the transmission data generating means includes a plurality of calculating means for performing inverse orthogonal transform on the plurality of input data at the processing points corresponding to the number of subcarriers of each group. Adopt a configuration that has.

According to this structure, by dividing the subcarriers of the orthogonal frequency division multiplexed signal to be transmitted into a plurality of groups, it is possible to narrow the signal bandwidth in each group. As a result, a predetermined oversampling ratio can be obtained without increasing the sampling frequency (conversion frequency) at the time of DA conversion, so that it is possible to use a low-bit digital-analog conversion circuit with low required accuracy, and the circuit No adjustment and higher accuracy can be achieved.

In the transmitting apparatus of the present invention having the above-mentioned configuration, the transmission data generating means includes an arithmetic means for performing an inverse orthogonal transform on the input data at a processing point corresponding to the number of subcarriers in each group, and the arithmetic means. And a time-division processing means for allocating the output data of 1 to each of the groups.

According to this structure, by dividing the subcarriers of the orthogonal frequency division multiplexed signal to be transmitted into a plurality of groups, the processing time in the arithmetic means for performing the inverse orthogonal transform becomes shorter according to the number of divisions. As a result, it is possible to use only one arithmetic unit configured to perform the inverse orthogonal processing with the number of processing points according to the number of subcarriers allocated to the group,
Since the result of the inverse discrete Fourier transform can be distributed to each group, the configuration of the calculation means, that is, the circuit scale can be reduced.

The transmission signal generating method of the present invention performs inverse orthogonal transform on the input data for each predetermined number of samples,
Transmission data of each group is generated by generating a plurality of groups of transmission data each consisting of a predetermined number of subcarriers, at a recoverable conversion frequency corresponding to the bandwidth when all the transmission data of each group are orthogonal frequency division multiplexed. Are subjected to digital-analog conversion processing in parallel, and output signals of the plurality of low-bit digital-analog conversion means are added to generate an orthogonal frequency division multiplexed signal composed of the plurality of groups of subcarriers as a transmission signal. .

According to this method, by dividing the subcarriers of the orthogonal frequency division multiplexed signal to be transmitted into a plurality of groups, it is possible to narrow the signal bandwidth in each group. As a result, a predetermined oversampling ratio can be obtained without increasing the sampling frequency (conversion frequency) at the time of DA conversion, so that it is possible to use a low-bit digital-analog conversion circuit with low required accuracy, and the circuit No adjustment and higher accuracy can be achieved.

The receiving apparatus of the present invention comprises a received signal dividing means for dividing the received orthogonal frequency division multiplexed signal into a predetermined number of groups for each subcarrier, and each divided received signal before division. A plurality of low bit analog-to-digital conversion means for performing analog-to-digital conversion in parallel at a predetermined oversampling ratio by sampling at a sampling frequency according to the bandwidth, and output data of the plurality of low-bit analog-to-digital conversion means. Then, the received data processing means for generating a reception symbol according to the number of subcarriers is subjected to orthogonal transformation with the number of processing points according to the number of divided subcarriers.

According to this structure, by dividing the subcarriers of the received orthogonal frequency division multiplexed signal into a plurality of groups, the signal bandwidth in each group can be narrowed. As a result, a predetermined oversampling ratio can be obtained without increasing the sampling frequency at the time of AD conversion, so that it is possible to use a low bit analog-to-digital conversion circuit with low required accuracy, and the circuit can be adjusted without adjustment. High accuracy can be achieved.

In the receiving apparatus of the present invention, in the above-mentioned configuration, the received data processing means comprises a plurality of arithmetic means for executing the orthogonal transformation with the number of processing points according to the number of subcarriers of each group. take.

According to this structure, by dividing the subcarriers of the received orthogonal frequency division multiplexed signal into a plurality of groups, it is possible to narrow the signal bandwidth in each group. As a result, a predetermined oversampling ratio can be obtained without increasing the sampling frequency at the time of AD conversion, so that it is possible to use a low bit analog-to-digital conversion circuit with low required accuracy, and the circuit can be adjusted without adjustment. High accuracy can be achieved.

In the receiver of the present invention having the above-mentioned structure, the reception data processing means includes time division multiplexing means for sequentially time division multiplexing the output data of the plurality of low bit analog-digital conversion means, and the time division multiplexing means. An arithmetic means for sequentially executing the orthogonal transformation on the multiplexed output data with the number of processing points according to the number of subcarriers in each group is adopted.

According to this structure, by dividing the subcarriers of the received orthogonal frequency division multiplex signal into a plurality of groups, the processing time in the arithmetic means for performing orthogonal transformation becomes shorter according to the number of divisions. As a result, the subcarriers divided into each group can be sequentially subjected to the discrete Fourier transform by using only one arithmetic unit configured to perform the orthogonal transform processing with the number of processing points according to the number of subcarriers assigned to the group. The configuration of the calculating means, that is, the circuit scale can be reduced.

The received signal processing method of the present invention divides the received orthogonal frequency division multiplexed signal into a predetermined number of groups for each subcarrier, and divides each of the divided received signals into a bandwidth before division. Analog to digital conversion processing is performed in parallel at a predetermined oversampling ratio by sampling at a corresponding sampling frequency, and orthogonal conversion is performed on the result of the analog to digital conversion processing at the number of processing points according to the number of divided subcarriers. By doing so, the received symbols according to the number of subcarriers are generated.

According to this method, it is possible to narrow the signal bandwidth in each group by dividing the subcarriers of the received orthogonal frequency division multiplexed signal into a plurality of groups. As a result, a predetermined oversampling ratio can be obtained without increasing the sampling frequency at the time of AD conversion, so that it is possible to use a low bit analog-to-digital conversion circuit with low required accuracy, and the circuit can be adjusted without adjustment. High accuracy can be achieved.

[0034]

BEST MODE FOR CARRYING OUT THE INVENTION The essence of the present invention is to divide a plurality of subcarriers into groups so that a DA converter or an A converter is provided.
It is to realize oversampling processing in the DA converter or the AD converter without increasing the sampling frequency of the D converter.

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

(Embodiment 1) FIG.1 is a block diagram showing the configuration of transmitting apparatus 30 according to Embodiment 1 of the present invention. In the first embodiment, the number of subcarriers N (N = 1
6) is divided into the number of groups K (K = 4) and DA (digital / analog) conversion is performed. That is, in FIG.
(N = 16) is divided into the number of groups K (K = 4) for DA conversion.

In the case of the first embodiment, a high speed data of one channel to be transmitted is switched by a switch circuit (not shown).
Depending on the symbol, it is divided into 4 groups,
The IDFT calculator 31a of the block that processes the first group g1 of subcarriers, the IDFT calculator 31b of the block that processes the second group g2, and the third
IDFT operation unit 3 of the block that processes the group g3 of
ID of the block that processes 1c and the fourth group g4
It is input to the FT calculation unit 31d. These 4 groups g1,
g2, g3, and g4 IDFT operation units 31a and 31
b, 31c, and 31d are 4-point IDFT for every 4 samples (= N / K) with respect to the input information data.
By performing the calculation, a baseband signal composed of N / K subcarriers is obtained for each group. Incidentally, different data of four channels may be input to each of the IDFT operation units 31a, 31b, 31c and 31d of each of the groups g1, g2, g3 and g4. Incidentally, the IDFT calculation unit 31a,
The IDFT calculation in 31b, 31c, and 31d means the inverse function of the orthogonal transformation process, respectively, and the IDFT calculation shows an example of the inverse orthogonal transformation process. Therefore, the present invention is not limited to the IDFT calculation, and may use, for example, a calculation unit that executes an IFFT (Inverse Fast Fourier Transform) calculation.

In the block for processing the first group g1 of subcarriers, the IDFT calculator 31a outputs 4 samples for every 4 samples (N / K = 4 samples) for the input first channel information data. By performing the point IDFT, the information data is converted into a baseband signal including in-phase and quadrature components of four subcarriers, and these are supplied to the quadruple interpolation circuits 32a and 33a.

The quadruple interpolation circuits 32a and 33a are IDF
The baseband signal including the in-phase and quadrature components supplied from the T calculation unit 31 is subjected to quadruple interpolation processing, thereby returning the baseband signal to the original sampling frequency. In the case of this embodiment, the sampling frequency f _S for the input information data is decimated to 1 / K (K = 4) in the IDFT calculation unit 31a, and thus the number of processing points in the IDFT calculation unit 31a is 1 as well. / K (K = 4) times. Figure 2 (A)
[Fig. 4] is a signal waveform diagram showing the baseband signals of the first group g1 output from the IDFT calculation unit 31a, and the sampling frequency f _S is ¼.

As a result, the number of complex multiplications performed by the IDFT calculation unit 31a is (N / K) ² <N ² / K (instead of the IDFT calculation unit 31a, an IFFT (Inverse Fast Fourier T
ransform), (N / K) log ₂ (N / K)
<Nlog ₂ N / K), the required calculation time can be reduced to less than 1 / K, and the circuit scale of the IDFT calculation unit 31a can be reduced accordingly.

As shown in FIG. 3A, the baseband signals of the first group g1 whose sampling frequency f _S has been restored in the quadruple interpolation circuits 32a and 33a are the noise shaping circuit 34a (FIG. 1). ) Is supplied to. The noise shaping circuit 34a includes a subtractor 35a,
36a, digital low pass filter (LPF) 37
a, 38a, 1-bit DA converters 40a, 41a, 1-bit AD (analog-digital) converters 39a, 42
It is an oversampling circuit having a loop configuration including a and amplifiers 43a and 44a having a gain of G times.

However, this noise shaping circuit 34a
Then, the conversion frequency (sampling frequency) f _S for the input baseband signal remains the frequency f _S returned to the original value in the quadruple interpolation circuits 32a and 33a. This means that the signal band to be finally transmitted is 1 /
The division into K means that the oversampling ratio K can be obtained without increasing the conversion frequency in the noise shaping circuit 34a, because the frequency band is narrowed.

As a result, the oversampling process similar to the case where the conversion frequency is increased can be performed without actually increasing the conversion frequency, and the quantization noise generated during DA conversion can be shifted to a high band. it can.

Therefore, the low bit conversion circuit (1 bit DA converters 40a, 41a, 1 bit A) with large quantization noise is used.
Even if DA conversion is performed using the D converters 39a and 42a), the quantization noise is shifted to a high band, and the analog low-pass filter (LPF) of the quadrature modulation circuit 45a, which will be described later, easily causes the quantization noise. It becomes possible to remove the quantization noise.

Low bit conversion circuit (1 bit DA converter 4
0a, 41a, 1-bit AD converters 39a, 42a)
Has only one determination threshold because it quantizes with 1 bit, and as a result, can be operated without adjustment and with high accuracy.

The quantization noise shifted to a high band in the noise shaping circuit 34a is suppressed by the analog low pass filter (LPF) inside the quadrature modulation circuit 45a provided at the subsequent stage of the noise shaping circuit 34a.

Further, the quadrature modulation circuit 45a quadrature modulates the baseband signal having the in-phase and quadrature components output from the noise shaping circuit 34a with the carrier having the center frequency f _IF1 assigned to the first group g1. In this way, the IF (Intermediate frequenc
y) Convert frequency to band signal. Thus, as shown in FIG. _3E , the number N / K of subcarriers having the center frequency f _IF1 (for example, N = 16 and K = 4, N / K =
4) Obtain the transmission IF signal of the first group g1.

Also, the second group g2 of subcarriers
Similarly, the IDFT calculation unit 31b, the four-fold interpolation circuits 32b and 33b, and the subtracters 35b and 36b
Digital low pass filter (LPF) 37b, 38
b, 1-bit DA converter 40b, 41b, 1-bit AD
Converters 39b and 42b and an amplifier 43 having a gain of G times
3B through the process shown in FIGS. 2B and 3B by a processing block including a noise shaping circuit 34b having a loop configuration including b and 44b, and a quadrature modulation circuit 45b. To obtain the transmission IF signal of the second group g2 having the center frequency f _IF2 .

Also, the third group g3 of subcarriers
Similarly, the IDFT calculation unit 31c, the four-fold interpolation circuits 32c and 33c, and the subtractors 35c and 36c
Digital low pass filter (LPF) 37c, 38
c, 1-bit DA converter 40c, 41c, 1-bit AD
Converters 39c and 42c and an amplifier 43 having a gain of G times
3 (E) through the process shown in FIGS. 2 (C) and 3 (C) by a processing block including a noise shaping circuit 34c having a loop configuration including c and 44c, and a quadrature modulation circuit 45c. To obtain the transmission IF signal of the third group g3 _having the center frequency f _IF3 .

Also, the fourth group g4 of subcarriers
Similarly, for the IDFT calculation unit 31d, the four-fold interpolation circuits 32d and 33d, and the subtractors 35d and 36d,
Digital low pass filter (LPF) 37d, 38
d, 1-bit DA converter 40d, 41d, 1-bit AD
Converters 39d and 42d, and an amplifier 43 having a gain of G times
3E is performed by the processing block including the noise shaping circuit 34d having a loop configuration including d and 44d, and the quadrature modulation circuit 45d through the processes shown in FIGS. 2D and 3D. To obtain the transmission IF signal of the fourth group g4 _having the center frequency f _IF4 .

In this way, the analog signals having different center frequencies obtained in the blocks for processing the groups of the subcarriers are sequentially added by the adders 48, 47 and 46, and the center signals shown in FIG. Frequency f _IF
Is converted into an RF (Radio Frequency) signal by a frequency converter (not shown) and transmitted via an antenna.

Incidentally, FIGS. 4A and 4B are block diagrams showing the principle of the noise shaping circuits 34a to 34d of the transmitter 30 described above with reference to FIG. As shown in FIG. 4 (A), the noise shaping circuit is a mixture of analog circuits and sample value circuits. This is achieved by replacing the 1-bit DA converter 41a with quantization noise addition. 4 (B) can be represented by a sample value equivalent circuit.

In this sample value equivalent circuit, as represented by the following equation (1), the signal X (z)
Is output via the transfer function H _X (z), and the quantization noise Q (z) is output via the transfer function H _Q (z). Y (z) = H _X (z) X (z) + H _Q (z) Q (z) (1)

Therefore, for the transfer function H (z) of the digital low pass filter, the transfer functions H _X (z) and H
_Q (z) has a low-pass characteristic and a high-pass characteristic as expressed by the following equations (2) and (3), respectively, and as a result, a low-distortion signal and a high-distortion signal are output. Quantization noise in which components are concentrated in the range is included. H _X (z) = (z ⁻¹ GH (z)) / (1 + z ⁻¹ GH (z)) (2) H _Q (z) = (z ⁻¹ G) / (1 + z ^{− 1} GH (z)) (3)

Incidentally, FIG. 5 shows the transfer characteristic of the digital low-pass filter in the sample value equivalent circuit (see FIG. 5).
(A)), input signal and quantization noise (FIG. 5B), input signal pass characteristic and quantization noise suppression characteristic in digital low-pass filter (FIG. 5C), output signal and quantization Each represents noise (FIG. 5D).

As described above, according to the principle of noise shaping, the quantization noise comes to concentrate in a high band, and as a result, the band of high SN (Signal Noise) ratio in which the transmission signal exists is quadrature-modulated in the subsequent stage. By taking out by the analog filter of the circuit 45a, it is possible to obtain a transmission IF signal in which noise is sufficiently suppressed. Incidentally, when the noise shaping circuits 34a to 34d shown in FIG. 1 are used, the conversion frequency (sampling frequency) should be changed to a high value although it depends on the characteristics of the digital low pass filters 37a to 37d and 38a to 38d. 1-bit DA converters 40a-40d and 41a-4 which are easy to adjust
By using 1d, accuracy of about 5 bits can be realized.

In the above configuration, the transmitter 30
When performing multi-carrier transmission, multiple sub-carriers are K
The noise shaping circuits 34a to 34d perform DA conversion on the respective groups in parallel. As a result, in each DA conversion process, the oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal because the target signal band is narrowed to 1 / K.

That is, in multi-carrier transmission, since the band is widened, it is necessary to adopt an extremely high frequency as the sampling frequency f _S when attempting to oversample without subdividing each subcarrier. In the transmitting device 30 of the first embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency as much as the frequency band is narrowed. Become.

Therefore, it is possible to carry out a practically sufficient oversampling process, and as a result, a low bit converter (1 bit DA converter, 1 bit AD converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the transmitting device 30 of the present embodiment
According to this, it is possible to reduce the accuracy required for the DA converter, and to realize no adjustment and high accuracy of the circuit. In particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, which is advantageous in realizing no adjustment of the circuit and higher accuracy. Become.

(Embodiment 2) FIG.6 is a block diagram showing the configuration of transmitting apparatus 60 according to Embodiment 2 of the present invention. However, the same components as those in FIG. 1 are denoted by the same reference numerals as those in FIG. 1, and detailed description thereof will be omitted.

In the second embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) as in the case of the transmitting device 30 described above with reference to FIG.
(Digital analog) Convert. That is, in FIG. 6, the transmitting device 60 has the number of subcarriers N (N = 16).
Is divided into the number of groups K (K = 4) and transmitted.

This transmitter 60 includes noise shaping circuits 34a-34 of the transmitter 30 described above with reference to FIG.
Second-order ΔΣ DA converters 61a to 61d and 62a to 62d, which are noise shaping circuits having a configuration different from that of d, are used. The second-order ΔΣ DA converters 61a to 61d and 62a to 62d are also for performing oversampling processing similar to the noise shaping circuits 34a to 34d in the transmission device 30.

Each ΔΣ type DA converter 61a shown in FIG.
-61d and 62a-62d have the same structure, respectively. FIG. 7 is a block diagram showing the configuration of the ΔΣ DA converter 61a. As shown in FIG. 7, ΔΣ
The type DA converter 61a DA-converts the baseband signal output from the quadruple interpolation circuit 32a in the 1-bit DA converter 71, and then AD-converts it in the 1-bit AD converter 72 in the feedback loop.

The signal converted into the digital signal by this is fed back to the subtracters 65 and 66, and the difference from the baseband signal inputted here is obtained. This difference is integrated in integrators 67 and 68. Then, the 1-bit DA converter 71 is fed back so that the integral value becomes the minimum.
The quantizing noise included in the output signal of is unevenly distributed in the high frequency band.

In this way, the ΔΣ DA converter 61a
61a to 61d and 62a to 62d, it is possible to obtain the oversampling effect by noise shaping similar to the noise shaping circuits 34a to 34d of the transmission device 30 described above with reference to FIG.

Further, the ΔΣ DA converters 61a to 61 are also provided.
If a to 61d and 62a to 62d are realized with high accuracy by a sample value circuit or the like, the transmitting device 30 described above with reference to FIG.
Compared with the configuration of the noise shaping circuits 34a to 34d, the digital low-pass filters (LPF) 37a, 3
By not interposing 8a, there is an advantage that filter distortion is not received.

Further, in the ΔΣ DA converters 61a to 61d and 62a to 62d, generally, the input signal (baseband signals output from the quadruple interpolation circuits 32a to 32d and 33a to 33d) has an all-pass characteristic. With respect to the quantization noise, the low-pass component of the quantization noise can be sufficiently reduced by having the high-pass characteristic.

Therefore, the ΔΣ type DA converters 61a to 61d.
, 62a to 62d, the quantization noise component Q of the output signal Y is sufficiently suppressed in the low frequency region, as shown in FIG.
A good signal of SN can be obtained.

Incidentally, the 1-bit D described above with reference to FIG.
In the case of the second-order ΔΣ type DA converter 61a using the A converter 71, the oversampling ratio m (= N / K) is about 5 lo.
Since the accuracy of g ₂ (m) −4.14 [bit] can be obtained, the accuracy of 5.8 bits can be obtained when m = 4.

In the above configuration, the transmitter 60 is
When performing multi-carrier transmission, multiple sub-carriers are K
Are divided into groups, each of which is connected in parallel with ΔΣ
DA conversion is performed by the type DA converters 61a to 61d and 62a to 62d. As a result, in each DA conversion process, the oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal because the target signal band is narrowed to 1 / K.

That is, in multi-carrier transmission, since the band is widened, it is necessary to use an extremely high frequency as the sampling frequency f _S when attempting to oversample without subdividing each subcarrier. In the transmitting device 60 of the second embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency as much as the frequency band is narrowed. Become.

Therefore, it is possible to carry out practically sufficient oversampling processing, and as a result, a low bit converter (1 bit DA converter, 1 bit AD converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the transmitting device 60 of the present embodiment
According to this, it is possible to reduce the accuracy required for the DA converter, and to realize no adjustment and high accuracy of the circuit. In particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, which is advantageous in realizing no adjustment of the circuit and higher accuracy. Become.

(Embodiment 3) FIG. 9 is a block diagram showing the configuration of transmitting apparatus 80 according to Embodiment 3 of the present invention. However, the same components as those in FIG. 1 are denoted by the same reference numerals as those in FIG. 1, and detailed description thereof will be omitted.

In the third embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) as in the case of the transmitting device 30 described above with reference to FIG.
(Digital analog) Convert. That is, in FIG. 9, the transmission device 80 has the number of subcarriers N (N = 16).
Is divided into the number of groups K (K = 4) and transmitted.

This transmitting apparatus 80 has the quadruple interpolation circuits 32a and 32a in the first subcarrier group g1.
The baseband signals of the in-phase and quadrature components output from b are input to the quadrature modulation circuit 45a and quadrature-modulated by the carrier of the center frequency f _IF1 assigned to the first group g1 to obtain the IF band signal. The frequency is converted into a signal and then supplied to the noise shaping circuit 84a.

The noise shaping circuit 84a includes a subtractor 85a and a digital band pass filter (BPF) 86.
a, 1-bit DA converter 87a, 1-bit AD converter 8
8a, and an oversampling circuit having a loop configuration including an amplifier 89a having a gain of G times.

In this noise shaping circuit 84a,
Noise shaping is performed on the transmission IF signal output from the quadrature modulation circuit 45a by using the digital bandpass filter 86a in which the center frequency is adjusted to the center frequency f _IF1 of the first group g1 in the loop.

In this noise shaping circuit 34a,
The conversion frequency (sampling frequency) f _S for the input transmission IF signal remains the frequency f _S returned to the original value in the quadruple interpolation circuits 32a and 33a. this is,
Since the signal band to be finally transmitted is divided into 1 / K, the oversampling ratio K can be obtained even if the conversion frequency is not increased in the noise shaping circuit 84a because the frequency band is narrowed. It means that.

As a result, the oversampling process similar to the case where the conversion frequency is raised can be performed without actually raising the conversion frequency, and the quantization noise generated during DA conversion can be shifted out of the band. it can.

Therefore, even if DA conversion is performed using the low bit conversion circuit (1-bit DA converter 87a, 1-bit AD converter 88a) having a large quantization noise, the quantization noise is still generated by the digital bandpass filter 86a. As the band is limited, it is sufficiently removed.

The sample value equivalent circuit in this case is shown in FIG.
Similar to the case described above with respect to (B), as shown in FIG. 10, the transfer characteristic of the digital band pass filter in the sample value equivalent circuit (FIG. 10 (A)),
Input signal X and quantization noise Q (FIG. 10 (B)), input signal pass characteristic and quantization noise suppression characteristic (FIG. 10 (C)) in the digital band filter, and output signal and quantization noise (FIG. 10). (D)) is the above-mentioned formula (1),
These are expressed based on the equations (2) and (3), respectively.

As described above, by using the noise shaping circuit 84a using the digital bandpass filter 86a, noise shaping can be performed on the transmission IF signal, and as a result, compared with the transmission device 30 of FIG. The number of noise shaping circuits can be halved compared to the case where noise shaping is performed on in-phase and quadrature components.

Also, the group g2 of the second subcarriers
In the same manner, the noise shaping circuit 84b
Is a subtractor 85b, a digital bandpass filter (BP
F) 86b, 1-bit DA converter 87b, 1-bit AD
This is an oversampling circuit having a loop configuration including a converter 88b and an amplifier 89b having a gain of G times. In this noise shaping circuit 84b, the center frequency is set to the center frequency f _IF2 of the second group g2 in the loop. By using the combined digital bandpass filter 86b, noise shaping is performed on the transmission IF signal output from the quadrature modulation circuit 45b.

Also, the third subcarrier group g3
Similarly, the noise shaping circuit 84c
Is a subtractor 85c, a digital band pass filter (BP
F) 86c, 1-bit DA converter 87c, 1-bit AD
This noise sampling circuit 84c is a loop-configuration oversampling circuit including a converter 88c and an amplifier 89c having a gain of G times. In the noise shaping circuit 84c, the center frequency is set to the center frequency f _IF3 of the third group g3. By using the combined digital band pass filter 86c, noise shaping is performed on the transmission IF signal output from the quadrature modulation circuit 45c.

Also, the fourth subcarrier group g4
Similarly, the noise shaping circuit 84d
Is a subtractor 85d, a digital band pass filter (BP
F) 86d, 1-bit DA converter 87d, 1-bit AD
This noise sampling circuit 84d is an oversampling circuit having a loop structure including a converter 88d and an amplifier 89d having a gain of G times. In the noise shaping circuit 84d, the center frequency is set to the center frequency f _IF4 of the fourth group g4. By using the combined digital band pass filter 86d, noise shaping is performed on the transmission IF signal output from the quadrature modulation circuit 45d.

Thus, each noise shaping circuit 84
The analog signals with different center frequencies output from a, 84b, 84c and 84d are added by adders 48, 47.
And 46, the signals are sequentially added to form a multi-carrier signal having the center frequency f _IF shown in FIG. 3 (E), and thereafter, a RF (RadioFre
quency) signal is frequency-converted and transmitted via an antenna.

In the above configuration, the transmitter 80
When performing multi-carrier transmission, multiple sub-carriers are K
Noise shaping circuits 84a, 84b, 84c and 84d in parallel with each of the groups.
DA conversion is performed by. As a result, in each DA conversion process, since the target signal band is narrowed to 1 / K,
The oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal.

That is, in multi-carrier transmission, since the band is widened, it is necessary to adopt an extremely high frequency as the sampling frequency f _S when attempting to oversample without subdividing each subcarrier. In the transmitting device 80 of the third embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency by the amount of the narrowed frequency band. Become.

Therefore, it is possible to perform oversampling processing that is practically sufficient, and as a result, a low bit converter (1 bit DA converter, 1 bit AD converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy. Also,
In the transmitting device 80 according to the third embodiment, the quadrature modulation circuit 4
Transmission I output from 5a, 45b, 45c and 45d
By performing noise shaping on each of the F signals, the number of noise shaping circuits can be halved compared to the case of FIG. 1 in which noise shaping is performed in the preceding stage of the quadrature modulation circuit.

As described above, the transmitting device 80 of the present embodiment
According to this, it is possible to more easily realize a configuration in which the circuit can be adjusted and the accuracy can be improved by lowering the accuracy required for the DA converter. In particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, which is advantageous in realizing no adjustment of the circuit and higher accuracy. Become.

(Embodiment 4) FIG.11 is a block diagram showing the configuration of transmitting apparatus 100 according to Embodiment 4 of the present invention. However, the same components as those in FIG. 1 are denoted by the same reference numerals as those in FIG. 1, and detailed description thereof will be omitted.

In the fourth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) in the same manner as in the case of the transmitting device 30 described above with reference to FIG.
Convert. That is, in FIG. 11, the transmitting device 100
Is the number of subcarriers N (N = 16) and the number of groups K (K
= 4) and is transmitted separately.

Compared to the transmitter 30 described above with reference to FIG. 1, this transmitter 100 performs IDFT calculation on each subcarrier group g1, g2, g3 and g4 using only one IDFT calculator 31a. The difference is what is done.

That is, as described above with reference to FIG.
The IDFT calculation unit 31a uses the 4-point DFT for every 4 samples (N / K = 4 samples) for the input information data.
By performing the above, the information data is converted into a baseband signal composed of in-phase and quadrature components of four subcarriers.

In this case, in the IDFT operation unit 31a, the sampling frequency f _S for the input information data is 1 / K (K = 4) as compared with the case where the input information data is converted into N subcarriers. As a result, the number of processing points in the IDFT calculation unit 31a is also 1 / K (K = 4) times. As a result, the number of complex multiplications performed by the IDFT calculation unit 31a is (N /
K) ² <N ² / K (By the way, when IFFT (Inverse Fast Fourier Transform) is used instead of the IDFT operation unit 31a, (N / K) log ₂ (N / K) <Nlog ₂ N /
K), and the required calculation time is less than 1 / K.

Therefore, the information data input to the IDFT operation unit 31a is IDFT in a processing time of 1 / K as compared with the case where it is converted into N (= 16) subcarriers.
By being output from the calculation unit 31a, the calculation result can be distributed to the groups g1, g2, g3, and g4 by the time division processing circuit (switch circuit) 101.

The baseband signals distributed to the respective groups g1, g2, g3 and g4 are
For each of 1, g2, g3, and g4, the quadruple interpolation circuit 32a to
By subjecting 32d and 33a to 33d to quadruple interpolation processing, respectively, the sampling frequency is quadrupled and restored to the original frequency.

In the above configuration, the IDFT calculation unit 31
Each channel output from a (each group g1, g
The information data of 2, g3 and g4) is thinned out by the amount that the sampling frequency is grouped,
Also, the number of processing points of the IDFT calculation is reduced by the amount of grouping. Thereby, the IDFT calculation unit 31a
In the configuration of, the circuit scale is reduced as compared with the case where the IDFT processing is performed at N points, and even if compared with the case described above with reference to FIG. 1, only one IDFT operation unit 31a is required. The circuit scale has been reduced by this amount.

The result of the IDFT operation in the IDFT operation section 31a is distributed to the blocks of the corresponding group in the time division processing circuit 101. And
Similar to the case of the transmission apparatus 30 described above with reference to FIG. 1, in the transmission apparatus 100, a plurality of subcarrier groups g1, g2, g3, and g4 when performing multicarrier transmission divided into K groups. On the other hand, the noise shaping circuits 34a to 34d connect the DAs in parallel.
The conversion is done. As a result, in each DA conversion process, since the target signal band is narrowed to 1 / K, the conversion frequency (sampling frequency) of the target signal is not increased,
An oversampling ratio K is obtained.

Therefore, it is possible to perform practically sufficient oversampling processing, and as a result, a low bit converter (1 bit DA converter, 1 bit AD converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the transmitting apparatus 10 of the present embodiment
According to 0, a configuration that can make the circuit unadjusted and highly accurate by lowering the required accuracy for the DA converter,
This can be more easily realized by reducing the configuration of the IDFT calculation unit. In addition, in particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, thereby achieving no adjustment of the circuit and higher accuracy. Be advantageous.

(Embodiment 5) FIG.12 is a block diagram showing the configuration of transmitting apparatus 110 according to Embodiment 5 of the present invention. However, the same components as those in FIG. 6 are designated by the same reference numerals as those in FIG. 6, and detailed description thereof will be omitted.

In the fifth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) in the same manner as in the case of the transmission device 60 described above with reference to FIG.
Convert. That is, in FIG. 12, the transmitter 110
Is the number of subcarriers N (N = 16) and the number of groups K (K
= 4) and is transmitted separately.

Compared to the transmitter 60 described above with reference to FIG. 6, the transmitter 110 performs IDFT calculation on each subcarrier group g1, g2, g3 and g4 using only one IDFT calculator 31a. The difference is what is done.

That is, as described above with reference to FIG.
The IDFT calculation unit 31a uses the 4-point DFT for every 4 samples (N / K = 4 samples) for the input information data.
By performing the above, the information data is converted into a baseband signal composed of in-phase and quadrature components of four subcarriers.

In this case, the sampling frequency f _S for the input information data is decimated to 1 / K (K = 4) in the IDFT calculation unit 31a, which also reduces the number of processing points in the IDFT calculation unit 31a to 1 / K. K
(K = 4) times. As a result, the number of complex multiplications of the IDFT calculation unit 31a is (N / K) ² <N ² / K (instead of the IDFT calculation unit 31a, the IFFT (Invers
When using e Fast Fourier Transform, (N /
K) log ₂ (N / K) <Nlog ₂ N / K), and the required calculation time is less than 1 / K.

Therefore, the information data input to the IDFT operation section 31a is 1 / K in processing time compared with the case where it is converted into N (= 16) subcarriers.
By being output from the calculation unit 31a, the calculation result can be distributed to the groups g1, g2, g3, and g4 by the time division processing circuit (switch circuit) 101.

In the above configuration, the IDFT calculation unit 31
Each channel output from a (each group g1, g
2, g3 and g4) information data is processed by the time division processing circuit 10
In 1, the blocks are assigned to the corresponding groups.

Then, in the transmitting device 110, as shown in FIG.
Similarly to the case of the transmission device 60 described above with respect to the above, a plurality of subcarriers when performing multicarrier transmission are divided into K groups, and ΔΣ DA converters 61a to 61d and 62a are parallel to each of the subgroups. DA conversion is performed according to .about.62d. As a result, in each DA conversion process, the oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal because the target signal band is narrowed to 1 / K.

Therefore, it is possible to carry out practically sufficient oversampling processing, and as a result, a low bit converter (1 bit DA converter, 1 bit AD converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the transmitting device 11 of the present embodiment
According to 0, a configuration that can make the circuit unadjusted and highly accurate by lowering the required accuracy for the DA converter,
This can be more easily realized by reducing the IDFT calculation unit. Further, in particular, as the number N of subcarriers increases, the number N of subcarriers is divided,
The oversampling process can be performed more easily, which is advantageous in realizing no adjustment of the circuit and higher accuracy.

(Embodiment 6) FIG.13 is a block diagram showing the configuration of transmitting apparatus 120 according to Embodiment 6 of the present invention. However, the same components as those in FIG. 9 are denoted by the same reference numerals as those in FIG. 9, and detailed description thereof will be omitted.

In the sixth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) in the same manner as in the case of the transmitting device 80 described above with reference to FIG.
Convert. That is, in FIG. 13, the transmitting device 120
Is the number of subcarriers N (N = 16) and the number of groups K (K
= 4) and is transmitted separately.

Compared to the transmitter 80 described above with reference to FIG. 9, the transmitter 120 performs IDFT calculation on each of the subcarrier groups g1, g2, g3 and g4 using only one IDFT calculator 31a. The difference is what is done.

That is, as described above with reference to FIG.
The IDFT calculation unit 31a uses the 4-point DFT for every 4 samples (N / K = 4 samples) for the input information data.
By performing the above, the information data is converted into a baseband signal composed of in-phase and quadrature components of four subcarriers.

In this case, the sampling frequency f _S for the input information data is decimated to 1 / K (K = 4) in the IDFT calculation unit 31a, which also reduces the number of processing points in the IDFT calculation unit 31a to 1 / K. K
(K = 4) times. As a result, the number of complex multiplications of the IDFT calculation unit 31a is (N / K) ² <N ² / K (instead of the IDFT calculation unit 31a, the IFFT (Invers
When using e Fast Fourier Transform, (N /
K) log ₂ (N / K) <Nlog ₂ N / K), and the required calculation time is less than 1 / K.

Therefore, the information data input to the IDFT operation unit 31a is IDFT in 1 / K processing time as compared with the case where it is converted into N (= 16) subcarriers.
By being output from the calculation unit 31a, the calculation result can be distributed to the groups g1, g2, g3, and g4 by the time division processing circuit (switch circuit) 101.

In the above configuration, the IDFT calculation unit 31
Each channel output from a (each group g1, g
2, g3 and g4) information data is processed by the time division processing circuit 10
In 1, the blocks are assigned to the corresponding groups.

Then, in the transmitting device 120, as shown in FIG.
Similarly to the case of the transmitting device 80 described above with respect to the above, a plurality of subcarriers when performing multicarrier transmission are divided into K groups, and DA conversion is performed in parallel for each of them by the noise shaping circuits 84a to 84d. Done. As a result, in each DA conversion process, the oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal because the target signal band is narrowed to 1 / K.

Therefore, it is possible to perform practically sufficient oversampling processing, and as a result, a low bit converter (1 bit DA converter, 1 bit AD converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the transmission device 12 of the present embodiment
According to 0, a configuration that can make the circuit unadjusted and highly accurate by lowering the required accuracy for the DA converter,
This can be more easily realized by reducing the IDFT calculation unit. Further, in particular, as the number N of subcarriers increases, the number N of subcarriers is divided,
The oversampling process can be performed more easily, which is advantageous in realizing no adjustment of the circuit and higher accuracy.

(Embodiment 7) FIG.14 is a block diagram showing the configuration of receiving apparatus 130 according to Embodiment 7 of the present invention. In the first embodiment, the number N (N = N) of subcarriers of the reception signal transmitted by the multicarrier transmission method.
A case will be described where 16) is divided into the number of groups K (K = 4) and AD conversion is performed. That is, in FIG. 14, the receiving device 130 divides the number of subcarriers N (N = 16) into the number of groups K (K = 4) and performs AD conversion.

In the receiver 130, the received signal received via the antenna (not shown) is frequency-converted by the frequency converter (not shown) from the RF band signal to the IF band received IF signal, and the sub-signal is received. Carrier first group g1, second group g2, third group g3
And the quadrature detection circuits 131a to 1 of the fourth group g4.
31d is supplied.

In the block for processing the first group g1 of subcarriers, the quadrature detection circuit 131a quadratures the received IF signal with the carrier ω _C of the center frequency f _IF1 assigned to the first group g1. To detect.
That is, FIG. 15 is a block diagram showing the configuration of the quadrature detection circuit 131a, and the input received IF signal is supplied to the multipliers 151a and 152a, respectively.

The multiplier 151a outputs the center frequency f assigned to the first group g1 to the received IF signal.
The _IF1 of the carrier as omega _C, as well as multiplied by cos .omega _C t, the multiplier 152a is the first for the received IF signal
By assuming that the carrier of the center frequency f _IF1 assigned to the group g1 of ω _C is ω _C , and multiplying by sin ω _C t,
From the multipliers 151a and 152a, baseband signals composed of in-phase and quadrature components are output. These baseband signals are analog low pass filters (LPF) 1
53a and 154a suppress subcarriers of other groups, and extract only the subcarriers of the first group as a baseband signal.

Also, the second group g2 of subcarriers
Quadrature detection circuit 131b
_Performs quadrature detection on the received IF signal with the carrier ω _C having the center frequency f _IF2 assigned to the second group g2. This quadrature detection circuit 131b is also the quadrature detection circuit 1
31a has the same configuration as that of 31a, performs quadrature detection on the received IF signal by the carrier ω _C having the center frequency f _IF2 assigned to the second group g2, and also performs the quadrature detection on the baseband. The band limitation of the signal is performed by the analog low pass filter (LPF), the sub-carriers of the other groups are suppressed, and only the sub-carriers of the second group are extracted as the baseband signal.

Also, the third group g3 of subcarriers
Quadrature detection circuit 131c
_Performs quadrature detection on the received IF signal with the carrier ω _{C having} the center frequency f _IF3 assigned to the third group g3. This quadrature detection circuit 131c is also the quadrature detection circuit 1
31a has the same configuration as that of 31a, and quadrature detection is performed on the received IF signal by the carrier ω _C of the center frequency f _IF3 assigned to the third group g3, and the quadrature-detected base band is also performed. By subjecting signals to band limitation by an analog low-pass filter (LPF), subcarriers in other groups are suppressed, and only subcarriers in the third group are extracted as baseband signals.

Also, the fourth group g4 of subcarriers
Quadrature detection circuit 131d
_Performs quadrature detection on the received IF signal with the carrier ω _{C having} the center frequency f _IF4 assigned to the fourth group g4. This quadrature detection circuit 131d is also the quadrature detection circuit 1
31a, the quadrature detection is performed on the received IF signal by the carrier ω _{C having} the center frequency f _IF4 assigned to the fourth group g4, and the quadrature-detected baseband is also performed. By subjecting the signal to band limitation by an analog low pass filter (LPF), subcarriers in another group are suppressed, and only the subcarriers in the fourth group are extracted as a baseband signal.

As a result, as shown in FIG. 16A, the received IF signals of the number of subcarriers N (N = 16) in each of the groups g1, g2, g3, and g4 are equal to each group g.
Quadrature detection circuits 131a to 131 of 1, g2, g3, and g4
1d, FIG. 16 (B), (C), (D), respectively.
And (E), each group g1, g2,
It is divided into g3 and g4 subcarriers and extracted as a baseband signal.

The subcarriers (baseband signals) of the groups g1, g2, g3 and g4 extracted by the quadrature detection circuits 131a to 131d are supplied to the following noise shaping circuits 132a to 132d, respectively.
The noise shaping circuit 132a includes a subtractor 133a,
134a, analog low pass filter (LPF) 135
a, 136a, 1-bit AD converters 138a, 139
a, a 1-bit DA converter 137a, 140a, and an oversampling circuit having a loop configuration including amplifiers 141a, 142a having a gain of G times.

However, this noise shaping circuit 132
In a, the sampling frequency f _S for the input baseband signal remains the sampling frequency in the received IF signal as shown in FIGS. 16 (A) and 16 (B). In this case, the signal band of the reception IF signal is the analog low pass filter 153a of the quadrature detection circuit 132A,
By being divided into 1 / K (K = 4) by 154A, the oversampling ratio K can be obtained without increasing the sampling frequency in the noise shaping circuit 132a by the amount of the narrowed frequency band. .

As a result, the oversampling process similar to the case where the sampling frequency is raised can be performed without actually raising the sampling frequency.
Quantization noise generated during D conversion can be shifted to a high band by the analog low pass filters 135a and 136a.

Therefore, even if the AD conversion is performed using the low bit conversion circuits (1-bit AD converters 138a, 139a, 1-bit DA converters 137a, 142a) having large quantization noise, the quantization noise has a high band. Will be shifted to 1/4 thinning circuits 145a and 146a described later.
It is possible to easily remove the quantization noise by a digital low-pass filter (not shown) provided in the preceding stage.

Low bit conversion circuit (1 bit AD converter 1
38a, 139a, 1-bit DA converters 137a, 14
2a) has only one determination threshold because it quantizes with 1 bit, and as a result, it can be operated without adjustment and with high accuracy.

The quarter thinning circuits 145a and 146a are
After suppressing the quantization noise shifted to the high frequency side by the digital low-pass filters provided inside, for each baseband signal composed of the in-phase and quadrature components supplied from the noise shaping circuit 132a, The quarter-thinning process is performed by adopting the sample value only once in four samples. As a result, as shown in FIG. 2, the sampling frequency of the subcarriers of the first group g1 becomes a low frequency f _S / K (K = 4) in the range required to restore the subcarriers. Thinned out.

The DFT calculator 147a performs 4-point DFT for every 4 samples (N / K = 4 samples) on the first group of subcarriers input from the 1/4 decimation circuits 145a, 146a. Thus, four received symbols corresponding to the four subcarriers of the first group g1 are obtained.

Also, the second group g2 of subcarriers
Similarly, the noise shaping circuit 13
After suppressing the quantization noise shifted to the high-frequency side by the digital low-pass filter for each baseband signal composed of the in-phase and quadrature components output from 2b, the sample value is adopted only once in four samples By doing so, 1/4 thinning-out processing is performed. As a result, as described above with reference to FIG. 2, the sampling frequency of the subcarriers of the second group g2 is a low frequency f _S / K (K = K) in the range necessary to restore the subcarriers.
4) is thinned out.

The DFT calculator 147b performs 4-point DFT for every 4 samples (N / K = 4 samples) for the second group of subcarriers input from the 1/4 decimation circuits 145b, 146b. Thus, four received symbols corresponding to the four subcarriers of the second group g2 are obtained.

Also, the third group g3 of subcarriers
Similarly, the noise shaping circuit 13
After suppressing the quantization noise shifted to the high band side by the digital low-pass filter for each baseband signal composed of in-phase and quadrature components output from 2c, the sample value is adopted only once in 4 samples By doing so, 1/4 thinning-out processing is performed. Thus, as described above for FIG. 2, the sampling frequency of the subcarrier of the third group g3 is lower frequency range needed to restore the sub-carrier f _S / K (K =
4) is thinned out.

The DFT operation unit 147c performs 4-point DFT for every 4 samples (N / K = 4 samples) for the third group of subcarriers input from the 1/4 decimation circuits 145c, 146c. Thus, four received symbols corresponding to the four subcarriers of the third group g3 are obtained.

Also, the fourth group g4 of subcarriers
Similarly, the noise shaping circuit 13
After suppressing the quantization noise shifted to the high band side by the digital low pass filter for each baseband signal consisting of in-phase and quadrature components output from 2d, sample value is adopted only once in 4 samples By doing so, 1/4 thinning-out processing is performed. As a result, as described above with reference to FIG. 2, the sampling frequency of the subcarriers of the fourth group g4 is a low frequency f _S / K (K = K) in the range necessary for restoring the subcarriers.
4) is thinned out.

The DFT operation unit 147d performs 4-point DFT for every 4 samples (N / K = 4 samples) for the fourth group of subcarriers input from the 1/4 decimation circuits 145d, 146d. Thus, four received symbols corresponding to the four subcarriers of the fourth group g4 are obtained. Incidentally, in the present invention, the DFT operation in the DFT operation units 147a to 147d only needs to be an orthogonal transform process. Therefore, the DFT operation unit 1
Instead of 47a to 147d, for example, FFT (Fast Fou
It is also possible to use an arithmetic unit or the like that executes the carrier transform) operation.

FIGS. 17A and 17B show the noise shaping circuit 13 of the transmitter 130 described above with reference to FIG.
It is a block diagram which shows the principle of 2a-132d. As shown in FIG. 17 (A), the noise shaping circuit is a mixture of analog circuits and sample value circuits. This is achieved by replacing the 1-bit AD converter 139a with quantization noise addition. This can be represented by a sample value equivalent circuit shown in FIG.

In this sample value equivalent circuit, the signal X (z) is expressed by the equation (1) above.
Is output via the transfer function H _X (z), and the quantization noise Q (z) is output via the transfer function H _Q (z).

Therefore, with respect to the transfer function H (z) of the digital low-pass filter, the transfer functions H _X (z) and H
_Q (z) has a low-pass characteristic and a high-pass characteristic as expressed by the above equations (2) and (3), respectively, and as a result, a low-distortion signal and a high-pass signal are output. Quantization noise in which components are concentrated in the range is included.

Incidentally, also in this case, it has the same characteristics as those shown in FIG. That is, FIG. 5 shows the transfer characteristics of the digital low pass filter (FIG. 5 (A)), the quadrature detection output signal and the quantization noise (FIG. 5 (B)) in the sample value equivalent circuit, and the input of the digital low pass filter. Signal passing characteristics and quantization noise suppression characteristics (FIG. 5C), and output signals and quantization noise (FIG. 5C).
(D)) respectively.

As described above, according to the principle of noise shaping, the quantization noise is concentrated in a high band, and as a result, the band having a high SN (Signal Noise) ratio in which a transmission signal exists is reduced to 1 / second of the latter stage. 4 thinning circuits 145a to 145
of the transmission I in which the noise is sufficiently suppressed by extracting with the digital filters d, 146a to 146d.
An F signal can be obtained. Incidentally, when the noise shaping circuits 132a to 132d shown in FIG. 14 are used, it depends on the characteristics of the analog low-pass filters 135a to 135d and 136a to 136d, but the conversion frequency (sampling frequency) does not need to be changed high. , 1-bit AD converters 138a to 138d and 1 which can be easily adjusted
By using 39a to 139d, accuracy of about 5 bits can be realized.

In the configuration described above, in receiving apparatus 130, a plurality of subcarriers for multicarrier transmission are divided into K groups, and noise shaping circuits 132a to 132d are provided in parallel for each group.
D conversion is performed. As a result, in each AD conversion process, since the target signal band is narrowed to 1 / K, the conversion frequency (sampling frequency) of the target signal is not increased,
An oversampling ratio K is obtained.

That is, in multi-carrier transmission, since the band is widened, if it is attempted to oversample without subdividing each subcarrier, it is necessary to adopt an extremely high frequency as the sampling frequency f _S. In the receiving device 130 according to the seventh embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency by the amount of the narrowed frequency band. Become.

Therefore, it is possible to carry out practically sufficient oversampling processing, and as a result, a low bit converter (1 bit AD converter, 1 bit DA converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the receiving device 13 of the present embodiment
According to 0, the accuracy required for the AD converter is lowered,
It is possible to realize adjustment-free and highly accurate circuits.
In particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, which is advantageous in realizing no adjustment of the circuit and higher accuracy. Become.

(Embodiment 8) FIG.18 is a block diagram showing the configuration of receiving apparatus 160 according to Embodiment 8 of the present invention. However, components having the same configurations as those in FIG. 14 are denoted by the same reference numerals as those in FIG. 14, and detailed description thereof will be omitted.

In the eighth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) and AD-converted, as in the case of the receiving apparatus 130 described above with reference to FIG. That is, in FIG. 18, the receiving device 1
60 is the number of subcarriers N (N = 16) and the number of groups K
AD conversion is performed separately for (K = 4).

This receiving apparatus 160 includes the noise shaping circuit 132 of the receiving apparatus 130 described above with reference to FIG.
a to 132d, second-order ΔΣ AD converters 161a to 161d and 1 which are noise shaping circuits having different configurations.
62a to 162d are used. The second-order ΔΣ AD converters 161a to 161d and 162a to 162d are also noise shaping circuits 132a to 1 in the receiving apparatus 130.
This is for performing oversampling processing similar to 32d.

Each ΔΣ type AD converter 16 shown in FIG.
1a-161d and 162a-162d have the same structure, respectively. FIG. 19 shows a ΔΣ type AD converter 16
It is a block diagram which shows the structure of 1a. As shown in FIG. 19, the ΔΣ AD converter 161a AD-converts the baseband signal output from the quadrature detection circuit 131a in the 1-bit AD converter 171, and then performs 1-bit DA conversion in the feedback loop. DA conversion is performed in the device 172.

The signal thus converted back to the analog signal is fed back to the subtracters 165 and 166, and the difference from the baseband signal input here is obtained. This difference is integrated in integrators 167 and 168. Then, by feeding back so that the integrated value becomes the minimum, the quantization noise included in the output signal of the 1-bit AD converter 171 is unevenly distributed in the high frequency band.

In this way, the ΔΣ type AD converter 161
a-161a-161d and 162a-162d, respectively, the receiving device 130 described above with reference to FIG.
It is possible to obtain the oversampling effect by the noise shaping similar to the noise shaping circuits 132a to 132d.

Further, the ΔΣ type AD converters 161a to 161a
When 61a to 161d and 162a to 162d are realized with high precision by a sample value circuit or the like, the noise shaping circuits 132a to 132a of the receiving device 130 described above with reference to FIG.
Compared to the 32d configuration, an analog low pass filter (L
The absence of the PFs 135a and 136a has an advantage that no filter distortion is caused.

Further, the ΔΣ type AD converters 161a to 161
In general, d and 162a to 162d have an all-pass characteristic for their input signals (baseband signals output from the quadrature detection circuits 131a to 131d) and a high-pass characteristic for quantization noise. As a result, the low frequency component of the quantization noise is sufficiently reduced.

Therefore, the ΔΣ type AD converters 161a to 161a
As for the output signals Y of 1a and 162a to 162d, the quantization noise component Q is sufficiently suppressed in the low frequency region, and a signal with a good SN can be obtained, as in the case shown in FIG.

Incidentally, a secondary ΔΣ type AD converter 161 using the 1-bit AD converter 171 described above with reference to FIG.
In the case of a, for the oversampling ratio m (= N / K),
Since an accuracy of about 5 log ₂ (m) -4.14 [bit] can be obtained, an accuracy of 5.8 bits can be obtained when m = 4.

In the configuration described above, in receiving apparatus 160, a plurality of subcarriers for multicarrier transmission are divided into K groups, and ΔΣ type AD converters 161a to 161d and 162a are parallel to each of them. ~ 1
AD conversion is performed by 62d. As a result, each AD
In the conversion process, the oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal because the target signal band is narrowed to 1 / K.

That is, in multi-carrier transmission, since the band is widened, if it is attempted to oversample without subdividing each subcarrier, it is necessary to adopt an extremely high frequency as the sampling frequency f _S. In receiving apparatus 160 of Embodiment 8, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency by the amount that the frequency band is narrowed. Become.

Therefore, it is possible to carry out practically sufficient oversampling processing, and as a result, a low bit converter (1 bit AD converter, 1 bit DA converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the receiving device 16 of the present embodiment
According to 0, the accuracy required for the AD converter is lowered,
It is possible to realize adjustment-free and highly accurate circuits.
In particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, which is advantageous in realizing no adjustment of the circuit and higher accuracy. Become.

(Embodiment 9) FIG.20 is a block diagram showing the configuration of receiving apparatus 180 according to Embodiment 9 of the present invention. However, components having the same configurations as those in FIG. 14 are denoted by the same reference numerals as those in FIG. 14, and detailed description thereof will be omitted.

In the ninth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) and AD-converted, as in the case of the receiving apparatus 130 described above with reference to FIG. That is, in FIG. 20, the receiving device 1
80 is the number of subcarriers N (N = 16) and the number of groups K
AD conversion is performed separately for (K = 4).

The receiving device 180 receives the reception signal from the RF band from the signal of the RF band by the frequency converter (not shown) by using the reception signal received through the antenna (not shown).
After frequency conversion into the F signal, the noise shaping circuits 18 of the first group g1, the second group g2, the third group g3, and the fourth group g4 of subcarriers
4a to 184d.

The noise shaping circuit 184a includes a subtractor 185a and an analog band pass filter (BPF) 18
6a is a 1-bit AD converter 187a, a 1-bit DA converter 188a, and an oversampling circuit having a loop configuration including an amplifier 189a having a gain of G times.

In the noise shaping circuit 184a, the center frequency is set in the loop within the first group g.
By using the analog band-pass filter 186a matched with the center frequency f _{IF1 of} 1, the input reception IF
Perform noise shaping on the signal.

In the noise shaping circuit 184a, the conversion frequency (sampling frequency) f _S for the input received IF signal remains the frequency f _S of the received received IF signal. This is because the signal band to be finally obtained is divided into 1 / K, so that the noise shaping circuit 184 is reduced by the amount corresponding to the narrowed frequency band.
This means that the oversampling ratio K can be obtained without increasing the conversion frequency at a.

As a result, oversampling processing similar to the case where the conversion frequency is increased can be performed without actually increasing the conversion frequency, and the quantization noise generated during AD conversion can be shifted out of the band. it can.

Therefore, even if the AD conversion is performed by using the low bit conversion circuit (1 bit AD converter 187a, 1 bit DA converter 188a) having a large quantization noise, the quantization noise is converted by the analog band pass filter 186a. As the band is limited, it is sufficiently removed.

The sample value equivalent circuit in this case is shown in FIG.
This is similar to the case described above with respect to (B), and similarly to the case shown in FIG. 10, the transfer characteristic of the analog bandpass filter in the sample value equivalent circuit (FIG. 10).
(A)), the input signal X and the quantization noise Q (FIG. 10).
(B)), the pass characteristic of the input signal and the suppression characteristic of the quantization noise in the digital bandpass filter (FIG. 10C), and the output signal and the quantization noise (FIG. 10D) are described above in (1). These are expressed based on the equation, the equation (2), and the equation (3).

As described above, the analog bandpass filter 1
By using the noise shaping circuit 184a using the 86a, it is possible to perform noise shaping on the received IF signal, and as a result, perform noise shaping on the in-phase and quadrature components.
It is possible to reduce the number of noise shaping circuits by half as compared with.

Also, the group g2 of the second subcarriers
In the same manner, the noise shaping circuit 184
b is a subtractor 185b, an analog bandpass filter (B
PF) 186b, 1-bit AD converter 187b, 1-bit DA converter 188b, and amplifier 189 whose gain is G times
The noise shaping circuit 184b is an oversampling circuit having a loop configuration including b. The noise shaping circuit 184b has an analog band pass filter 186 in which the center frequency is adjusted to the center frequency f _IF2 of the second group g2.
By using b, noise shaping is performed on the input received IF signal.

Also, the group g3 of the third subcarriers
In the same manner, the noise shaping circuit 184
c is a subtractor 185c, an analog band pass filter (B
PF) 186c, 1-bit AD converter 187c, 1-bit DA converter 188c, and amplifier 189 having a gain of G times
The noise shaping circuit 184c is an over-sampling circuit having a loop configuration including c. The noise shaping circuit 184c has an analog band pass filter 186 in which the center frequency is _adjusted to the center frequency f _IF3 of the third group g3.
By using c, noise shaping is performed on the input received IF signal.

Also, the fourth subcarrier group g4
In the same manner, the noise shaping circuit 184
d is a subtractor 185d, an analog bandpass filter (B
PF) 186d, 1-bit AD converter 187d, 1-bit DA converter 188d, and amplifier 189 having a gain of G times
The noise shaping circuit 184d is an oversampling circuit having a loop configuration including d. The noise shaping circuit 184d has an analog bandpass filter 186 whose center frequency is _adjusted to the center frequency f _IF4 of the fourth group g4.
By using d, noise shaping is performed on the input received IF signal.

Thus, each noise shaping circuit 18
The analog signals with different center frequencies output from 4a to 184d are supplied to the quadrature detection circuits 131a to 131d in the blocks that process the groups g1 to g4. These quadrature detection circuits 131a to 131d
Then, a baseband signal is obtained by _performing quadrature detection on the input received IF signals with the carriers ω _C having the center frequencies f _{IF1 to} f _IF4 assigned to the respective groups. The quadrature-detected baseband signal is the subsequent 1/4 decimation circuits 145a to 145d and 146a to 1.
By performing band limitation by the digital low-pass filter (LPF) of 46d, subcarriers of other groups are suppressed and only subcarriers of each group are extracted.

The subcarriers of each group extracted in this way are subjected to 4-point DFT every 4 samples (N / K = 4 samples) in DFT operation units 147a to 147d, so that in each group. , Four received symbols corresponding to each of the four subcarriers are obtained.

In the configuration described above, in receiving apparatus 180, a plurality of subcarriers for multicarrier transmission are divided into K groups, and noise shaping circuits 184a to 184d are provided in parallel for each of the groups.
D conversion is performed. As a result, in each AD conversion process, since the target signal band is narrowed to 1 / K, the conversion frequency (sampling frequency) of the target signal is not increased,
An oversampling ratio K is obtained.

That is, in multicarrier transmission, since the band is widened, it is necessary to adopt an extremely high frequency as the sampling frequency f _S when attempting to oversample without subdividing each subcarrier. In the receiving device 180 according to the ninth embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency by the amount of the narrowed frequency band. Become.

Therefore, it is possible to perform oversampling processing which is practically sufficient, and as a result, a low bit converter (1 bit AD converter, 1 bit DA converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy. Also,
In the receiving device 180 according to the ninth embodiment, the reception I is provided in a stage before being input to the quadrature detection circuits 131a to 131d.
Compared to the case of FIG. 14 in which noise shaping is performed on the F signals, noise shaping is performed in the subsequent stage of the quadrature detection circuits 131a to 131d.
The number of the noise shaping circuits can be halved.

As described above, the receiving device 18 of the present embodiment
According to 0, it is possible to more easily realize a configuration in which the circuit can be adjusted and the accuracy can be improved by lowering the accuracy required for the AD converter. In particular, as the number N of subcarriers increases, the number N of subcarriers can be divided to more easily perform oversampling processing, which is advantageous in realizing no adjustment of the circuit and higher accuracy. Become.

(Embodiment 10) FIG. 21 is a block diagram showing the configuration of receiving apparatus 200 according to Embodiment 10 of the present invention. However, components having the same configurations as those in FIG. 14 are denoted by the same reference numerals as those in FIG. 14, and detailed description thereof will be omitted.

In the tenth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) and AD-converted, as in the case of the receiving apparatus 130 described above with reference to FIG. That is, in FIG. 21, the receiving device 200 divides the number of subcarriers N (N = 16) into the number of groups K (K = 4) and performs AD conversion.

Compared to the receiver 130 described above with reference to FIG. 14, the receiver 200 has one DFT for each group g1, g2, g3 and g4 of subcarriers.
The difference is that the DFT calculation is performed using only the calculation unit 147a.

That is, the receiving apparatus 200 supplies the subcarriers of each group output from the 1/4 decimation circuits 145a to 145d and 146a to 146d to the time division multiplexing circuit 201, and provides the time division multiplexing circuit 201. The data is temporarily stored in the memory for each group thus created.

The time division multiplexing circuit 201 sequentially supplies the received data composed of N / K subcarriers for each group stored in the memory for each group to the DFT operation unit 147a. The DFT operation unit 147a performs 4-point DFT for every 4 samples (N / K = 4 samples) on the reception data for each group that is sequentially supplied from the time division multiplexing circuit 201, thereby performing 4 points of the group. Obtain four received symbols corresponding to one subcarrier.

In this case, in the DFT calculation unit 147a, the sampling frequency f _S for the input received data is the DFT with the number of processing points of N points for the input N subcarriers (for example, N = 16). It is decimated to 1 / K (K = 4) compared to the case of performing calculation,
Further, the number of processing points in the DFT calculation unit 147a is also 1 / K.
(K = 4) times. As a result, the DFT operation unit 1
The number of complex multiplications of 47a is (N / K) ² <N ² / K (Incidentally, instead of the DFT operation unit 147a, FFT (FastFour)
(N / K) log ₂ (N
/ K) <Nlog ₂ N / K) and the required calculation time is 1 / K
It is less than.

Therefore, compared with the case where the received data input to the DFT operation unit 147a is composed of N (= 16) subcarriers, the DFT operation unit 1 can be processed in 1 / K.
It will be output from 47a. Therefore, the reception data of each group once stored in the time division multiplexing circuit 201 is read in 1 / K times the time required for the reception data to be input to the time division multiplexing circuit 201, and overflow or The received data of each group is sequentially subjected to DFT operation in one DFT operation unit 147a without becoming an underflow state.

In the above configuration, the DFT operation unit 147
The received data of each channel (each group g1, g2, g3, and g4) for which the DFT operation is performed in a is decimated by the sampling frequency, and the number of processing points of the DFT operation is also grouped. It has been reduced by the amount. As a result, the DFT operation unit 1
The configuration of 47a has a reduced circuit scale as compared to the case where DFT processing is performed at N points, and only one DFT operation unit 147a is required as compared with the case described above with reference to FIG. Therefore, the circuit scale is reduced accordingly.

The received data, which has been divided into groups and is input to the DFT operation unit 147a, is AD-converted in parallel by the noise shaping circuits 132a to 132d in the preceding stage. As a result, in each AD conversion process, since the target signal band is narrowed to 1 / K, the conversion frequency (sampling frequency) of the target signal is not increased,
An oversampling ratio K is obtained.

Therefore, practically sufficient oversampling processing can be performed, and as a result, a low bit converter (1 bit AD converter, 1 bit DA converter) can be employed. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the receiving device 20 of the present embodiment
According to No. 0, a configuration that can make the circuit unadjusted and highly accurate by lowering the accuracy required for the AD converter,
This can be realized more easily by reducing the DFT calculation unit. Further, in particular, as the number N of subcarriers increases, by dividing the number N of subcarriers, oversampling processing can be more easily performed,
This is advantageous in realizing no adjustment of the circuit and higher accuracy.

(Embodiment 11) FIG.22 is a block diagram showing the configuration of receiving apparatus 210 according to Embodiment 11 of the present invention. However, components having the same configurations as those in FIG. 18 are denoted by the same reference numerals as those in FIG. 18, and detailed description thereof will be omitted.

In the eleventh embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) and AD-converted, as in the case of the receiving apparatus 160 described above with reference to FIG. That is, in FIG. 22, the reception device 210 divides the number of subcarriers N (N = 16) into the number of groups K (K = 4) and performs AD conversion.

Compared to the receiver 160 described above with reference to FIG. 18, this receiver 210 has one DFT for each group g1, g2, g3 and g4 of subcarriers.
The difference is that the DFT calculation is performed using only the calculation unit 147a.

That is, the receiving apparatus 210 supplies the subcarriers of each group output from the 1/4 decimation circuits 145a to 145d and 146a to 146d to the time division multiplexing circuit 201, and provides the time division multiplexing circuit 201 with the subcarriers. The data is temporarily stored in the memory for each group thus created.

The time division multiplexing circuit 201 sequentially supplies the received data composed of N / K subcarriers for each group stored in the memory for each group to the DFT calculation unit 147a. The DFT operation unit 147a performs 4-point DFT for every 4 samples (N / K = 4 samples) on the reception data for each group that is sequentially supplied from the time division multiplexing circuit 201, thereby performing 4 points of the group. Obtain four received symbols corresponding to one subcarrier.

In this case, in the DFT calculation unit 147a, the sampling frequency f _S for the input received data is the DFT with the number of processing points N for the input N subcarriers (for example, N = 16). It is decimated to 1 / K (K = 4) compared to the case of performing calculation,
Further, the number of processing points in the DFT calculation unit 147a is also 1 / K.
(K = 4) times. As a result, the DFT operation unit 1
The number of complex multiplications of 47a is (N / K) ² <N ² / K (Incidentally, instead of the DFT operation unit 147a, FFT (FastFour)
(N / K) log ₂ (N
/ K) <Nlog ₂ N / K) and the required calculation time is 1 / K
It is less than.

Therefore, compared with the case where the received data input to the DFT operation unit 147a is composed of N (= 16) subcarriers, the DFT operation unit 1 can be processed in 1 / K.
It will be output from 47a. Therefore, the reception data of each group once stored in the time division multiplexing circuit 201 is read in 1 / K times the time required for the reception data to be input to the time division multiplexing circuit 201, and overflow or The received data of each group is sequentially subjected to DFT operation in one DFT operation unit 147a without becoming an underflow state.

In the above configuration, the DFT operation unit 147
The received data of each channel (each group g1, g2, g3, and g4) for which the DFT operation is performed in a is decimated by the sampling frequency, and the number of processing points of the DFT operation is also grouped. It has been reduced by the amount. As a result, the DFT operation unit 1
In the configuration of 47a, the circuit scale is reduced as compared with the case where the DFT processing is performed at N points, and only one DFT operation unit 147a is required as compared with the case described above with reference to FIG. Therefore, the circuit scale is reduced accordingly.

The reception data, which has been divided into groups and is input to the DFT operation unit 147a, is parallel to the ΔΣ AD converters 161a to 161d and 162a in the preceding stage.
AD conversion is performed by 162d. As a result, each A
In the D conversion process, since the target signal band is narrowed to 1 / K, the conversion frequency (sampling frequency) of the target signal
The oversampling ratio K can be obtained without increasing.

That is, in multi-carrier transmission, since the band is widened, if it is attempted to oversample without subdividing each subcarrier, it is necessary to adopt an extremely high frequency as the sampling frequency f _S. In the receiving apparatus 210 according to the eleventh embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency as much as the frequency band is narrowed. Become.

Therefore, it is possible to perform oversampling processing which is practically sufficient, and as a result, a low bit converter (1 bit AD converter, 1 bit DA converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

As described above, the receiving device 21 of the present embodiment
According to No. 0, a configuration that can make the circuit unadjusted and highly accurate by lowering the accuracy required for the AD converter,
This can be realized more easily by reducing the DFT calculation unit. Further, in particular, as the number N of subcarriers increases, by dividing the number N of subcarriers, oversampling processing can be more easily performed,
This is advantageous in realizing no adjustment of the circuit and higher accuracy.

(Embodiment 12) FIG.23 is a block diagram showing the configuration of receiving apparatus 220 according to Embodiment 12 of the present invention. However, the same components as those in FIG. 20 are designated by the same reference numerals as those in FIG. 20, and detailed description thereof will be omitted.

In the twelfth embodiment, the number of subcarriers N (N = 16) is divided into the number of groups K (K = 4) and AD-converted, as in the case of the receiving apparatus 180 described above with reference to FIG. That is, in FIG. 23, the receiving device 220 divides the number of subcarriers N (N = 16) into the number of groups K (K = 4) and performs AD conversion.

Compared to the receiver 180 described above with reference to FIG. 20, this receiver 220 has one DFT for each group g1, g2, g3 and g4 of subcarriers.
The difference is that the DFT calculation is performed using only the calculation unit 147a.

That is, the receiving apparatus 220 supplies the subcarriers of each group output from the 1/4 decimation circuits 145a to 145d and 146a to 146d to the time division multiplexing circuit 201, and provides them to the time division multiplexing circuit 201. The data is temporarily stored in the memory for each group thus created.

The time division multiplexing circuit 201 sequentially supplies the received data, which is stored in the memory for each group and includes N / K subcarriers for each group, to the DFT operation unit 147a. The DFT operation unit 147a performs 4-point DFT for every 4 samples (N / K = 4 samples) on the reception data for each group that is sequentially supplied from the time division multiplexing circuit 201, thereby performing 4 points of the group. Obtain four received symbols corresponding to one subcarrier.

In this case, in the DFT calculation unit 147a, the sampling frequency f _S for the input received data is the DFT with the number of processing points N for the input N subcarriers (for example, N = 16). It is decimated to 1 / K (K = 4) compared to the case of performing calculation,
Further, the number of processing points in the DFT calculation unit 147a is also 1 / K.
(K = 4) times. As a result, the DFT operation unit 1
The number of complex multiplications of 47a is (N / K) ² <N ² / K (Incidentally, instead of the DFT operation unit 147a, FFT (FastFour)
(N / K) log ₂ (N
/ K) <Nlog ₂ N / K) and the required calculation time is 1 / K
It is less than.

Therefore, compared to the case where the received data input to the DFT operation unit 147a consists of N (= 16) subcarriers, the DFT operation unit 1 can be processed in 1 / K processing time.
It will be output from 47a. Therefore, the reception data of each group once stored in the time division multiplexing circuit 201 is read in 1 / K times the time required for the reception data to be input to the time division multiplexing circuit 201, and overflow or The received data of each group is sequentially subjected to DFT operation in one DFT operation unit 147a without becoming an underflow state.

In the above configuration, the DFT operation unit 147
The received data of each channel (each group g1, g2, g3, and g4) for which the DFT operation is performed in a is decimated by the sampling frequency, and the number of processing points of the DFT operation is also grouped. It has been reduced by the amount. As a result, the DFT operation unit 1
In the configuration of 47a, the circuit scale is reduced as compared with the case where the DFT processing is performed at N points, and only one DFT operation unit 147a is required as compared with the case described above with reference to FIG. Therefore, the circuit scale is reduced accordingly.

The received data, which has been divided into groups and is input to the DFT operation unit 147a, is AD-converted in parallel by the noise shaping circuits 184a to 184d in the preceding stage. As a result, in each AD conversion process, the oversampling ratio K can be obtained without increasing the conversion frequency (sampling frequency) of the target signal because the target signal band is narrowed to 1 / K.

That is, in multicarrier transmission, since the band is widened, if it is attempted to oversample without subdividing each subcarrier, it is necessary to adopt an extremely high frequency as the sampling frequency f _S. In the receiving device 220 according to the twelfth embodiment, by dividing the subcarrier into 1 / K, the sampling frequency is relatively high without changing the sampling frequency by the amount of the narrowed frequency band. Become.

Therefore, it is possible to carry out practically sufficient oversampling processing, and as a result, a low bit converter (1 bit AD converter, 1 bit DA converter) can be adopted. Thus, by low bit conversion,
It is possible to realize no adjustment and high accuracy.

Also, the receiving device 22 of the twelfth embodiment
At 0, the quadrature detection circuit 131a is performed by performing noise shaping on the reception IF signal in the preceding stage input to the quadrature modulation circuits 145a to 145d.
It is possible to reduce the number of the noise shaping circuits by half as compared with the case of FIG. 14 in which noise shaping is performed in the subsequent stage of 131 d to 131 d.

As described above, the receiving device 22 of the present embodiment
According to No. 0, a configuration that can make the circuit unadjusted and highly accurate by lowering the accuracy required for the AD converter,
This can be realized more easily by reducing the DFT calculation unit. Further, in particular, as the number N of subcarriers increases, by dividing the number N of subcarriers, oversampling processing can be more easily performed,
This is advantageous in realizing no adjustment of the circuit and higher accuracy.

[0223]

As described above, according to the present invention,
It is possible to use a low-bit digital-analog conversion circuit and a low-bit analog-digital conversion circuit with low required accuracy, which can make the circuit unadjusted and highly accurate, and also can reduce the circuit scale.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a transmission apparatus according to Embodiment 1 of the present invention.

FIG. 2 is a signal waveform diagram for explaining the operation of the transmitting apparatus according to the first embodiment.

FIG. 3 is a signal waveform diagram for explaining the operation of the transmitting apparatus according to the first embodiment.

FIG. 4 is a block diagram showing an equivalent circuit configuration of a DA converter of the transmitting device according to the first embodiment of the present invention.

FIG. 5 is a signal waveform diagram for explaining the operation of the transmitting apparatus according to the first embodiment.

FIG. 6 is a block diagram showing a configuration of a transmission apparatus according to Embodiment 2 of the present invention.

FIG. 7 is a block diagram showing an equivalent circuit configuration of a DA converter of the transmitting device according to the second embodiment.

FIG. 8 is a signal waveform diagram for explaining the operation of the transmitting apparatus according to the second embodiment.

FIG. 9 is a block diagram showing a configuration of a transmission apparatus according to Embodiment 3 of the present invention.

FIG. 10 is a signal waveform diagram for explaining the operation of the transmitting apparatus according to the third embodiment.

FIG. 11 is a block diagram showing a configuration of a transmission apparatus according to Embodiment 4 of the present invention.

FIG. 12 is a block diagram showing a configuration of a transmission apparatus according to Embodiment 5 of the present invention.

FIG. 13 is a block diagram showing a configuration of a transmission apparatus according to Embodiment 6 of the present invention.

FIG. 14 is a block diagram showing the configuration of a receiving apparatus according to Embodiment 7 of the present invention.

FIG. 15 is a block diagram showing the configuration of a quadrature detection circuit of the receiving device according to the seventh embodiment.

FIG. 16 is a signal waveform diagram for explaining the operation of the device according to the seventh embodiment.

FIG. 17 is an AD of the receiving device according to the seventh embodiment of the present invention.
Block diagram showing the equivalent circuit configuration of the converter

FIG. 18 is a block diagram showing the configuration of a receiving apparatus according to Embodiment 8 of the present invention.

FIG. 19 is an AD of the receiving device according to the eighth embodiment of the present invention.
Block diagram showing the equivalent circuit configuration of the converter

FIG. 20 is a block diagram showing the configuration of a receiving apparatus according to Embodiment 9 of the present invention.

FIG. 21 is a block diagram showing the configuration of a receiving apparatus according to Embodiment 10 of the present invention.

FIG. 22 is a block diagram showing the configuration of a receiving apparatus according to Embodiment 11 of the present invention.

FIG. 23 is a block diagram showing the configuration of a receiving apparatus according to Embodiment 12 of the present invention.

FIG. 24 is a block diagram showing the configuration of a conventional transmitter.

FIG. 25 is a signal waveform diagram showing each subcarrier in the multicarrier transmission system.

FIG. 26 is a block diagram showing a configuration of a conventional receiving device.

[Explanation of symbols]

30, 60, 80, 100, 110, 120 Transmitters 31a to 31d IDFT operation units 32a to 32d, 33a to 33d Quadruple interpolation circuits 34a to 34d, 84a to 84d, 132a to 132
d, 184a to 184d Noise shaping circuit 45a to 45d Quadrature modulation circuit 61a to 61d, 62a to 62d ΔΣ DA converter 101 Time division processing circuit 130, 160, 180, 200, 210, 220 Receiver 131a to 131d Quadrature detection circuit 145a to 145d 1/4 thinning circuits 147a to 147d DFT operation unit

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Mitsuru Uesugi             3-1, Tsunashima-Higashi 4-chome, Kohoku-ku, Yokohama-shi, Kanagawa             Matsushita Communication Industry Co., Ltd. F-term (reference) 5C063 AA20 AB06 AB15 AC01 AC05                       CA09                 5K022 DD01 DD23 DD33

Claims (8)

[Claims] 1. Transmission data generation means for generating transmission data of a plurality of groups each consisting of a predetermined number of subcarriers by inverse orthogonal transforming input data for each predetermined number of samples, and all of the transmission data of each group. A plurality of low-bit digital-analog conversion means for performing parallel digital-analog conversion of the transmission data of each group at a conversion frequency that can be restored according to the bandwidth when orthogonal frequency division multiplexing is performed; Transmission signal generating means for generating, as a transmission signal, an orthogonal frequency division multiplex signal composed of the subcarriers of the plurality of groups by adding the output signals of the means. 2. The transmission data generating means comprises a plurality of calculating means for performing an inverse orthogonal transform on a plurality of the input data at a processing point corresponding to the number of subcarriers in each group. The transmission device described. 3. The transmission data generating means includes an arithmetic means for performing an inverse orthogonal transform on the input data at a processing point corresponding to the number of subcarriers in each group, and an output data of the arithmetic means for allocating to each group. The transmission device according to claim 1, further comprising: a division processing unit. 4. The transmission data of a plurality of groups each consisting of a predetermined number of subcarriers is generated by performing inverse orthogonal transform on the input data for each predetermined number of samples, and all the transmission data of each group are orthogonal frequency division multiplexed. By performing digital-analog conversion processing on the transmission data of each group in parallel at a conversion frequency that can be restored according to the bandwidth at the time of adding, and adding the output signals of the low-bit digital-analog conversion means, A transmission signal generation method, characterized in that an orthogonal frequency division multiplex signal composed of subcarriers of a group is generated as a transmission signal. 5. A received signal division means for dividing a received orthogonal frequency division multiplexed signal into a predetermined number of groups for each subcarrier, and for each of the divided received signals, a bandwidth before division is determined. A plurality of low-bit analog-digital conversion means for performing analog-digital conversion in parallel at a predetermined oversampling ratio by sampling at a sampling frequency respectively, and the divided output data of the plurality of low-bit analog-digital conversion means. And a reception data processing unit that generates a reception symbol according to the number of subcarriers by performing orthogonal transformation with the number of processing points according to the number of subcarriers. 6. The receiving apparatus according to claim 5, wherein the received data processing means comprises a plurality of arithmetic means for executing the orthogonal transformation with the number of processing points according to the number of subcarriers of each group. . 7. The reception data processing means comprises time division multiplexing means for sequentially time-division-multiplexing each output data of the plurality of low-bit analog-digital conversion means, and each of the time-division-multiplexed output data, The receiving device according to claim 5, further comprising: an arithmetic unit that sequentially executes the orthogonal transformation with the number of processing points according to the number of subcarriers in each group. 8. The received orthogonal frequency division multiplexed signal is divided into a predetermined number of groups for each subcarrier, and each of the divided received signals is sampled at a sampling frequency according to the bandwidth before division. By performing analog-digital conversion processing in parallel at a predetermined oversampling ratio, the analog-digital conversion processing result is subjected to orthogonal conversion at the number of processing points according to the number of divided subcarriers, so that the subcarrier A received signal processing method characterized by generating received symbols according to the number.