JP2008148212A - Communication apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a communication apparatus reduced in power consumption while achieving noise reduction. <P>SOLUTION: This transmitting-side communication apparatus which adopts a multi-carrier system is provided with: a carrier waveform generation unit 4 for generating sampling data by sampling a modulation signal at a predetermined frequency; a delta/sigma modulator 6 for converting the sampling data into 1-bit digital data; an amplifier 7 for amplifying the digital data by the switching operation of a transistor; and a low-pass filter 8 for removing a frequency component other than a required frequency domain from the amplified data. An operating frequency of the delta/sigma modulation means is calculated on the basis of a signal-to-noise ratio required for transmission data. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a communication apparatus that performs multicarrier communication, and more particularly, to a communication apparatus that performs multicarrier communication that achieves power saving.

  In conventional multi-carrier communication equipment using a plurality of orthogonal carriers, a modulated waveform is generated by digital circuit processing, converted to an analog signal by a DA (digital-analog) converter, and transmitted to a transmission line. It is general to take a configuration to perform the above. As an example of such a communication device, for example, there is a multicarrier power line carrier communication device described in Patent Document 1 below. In this power line carrier communication device, after the digital transmission waveform is generated, DA conversion is performed using, for example, an 8-bit DA converter, the analog transmission waveform is amplified, the aliasing noise is removed by a low-pass filter, and the coupling circuit is Output to the transmission line. In this power line carrier communication device, an analog amplifier is used to amplify the analog transmission waveform. As in this example, an analog amplifier is often used in a multi-carrier communication device. However, generally, an analog amplifier is required to always apply a bias voltage, so that power consumption tends to increase.

  As one method for reducing the power consumption of the transmission amplifier, there is a method of amplifying the transmission waveform using a digital amplifier that does not require a bias voltage by using delta-sigma modulation (hereinafter referred to as ΔΣ modulation). ΔΣ modulation is a technique for converting an analog or multi-bit signal into digital data of a small number of bits such as 1 bit. In addition, ΔΣ modulation can reduce the quantization noise by oversampling, and can obtain a high S / N ratio in the use band by biasing the quantization noise to the high frequency side by noise shaping characteristics. It is used in. By using ΔΣ modulation, a digital amplifier can be used instead of an analog amplifier. Since the digital amplifier performs signal amplification by switching operation of the transistor, it can save power without requiring a bias voltage.

  As an example of using a digital amplifier by using ΔΣ modulation for AD (analog-digital) conversion, there is a technique described in Patent Document 2 below. In this example, ΔΣ modulation is performed on an audio input signal from a microphone or the like to obtain 1-bit digital data, which is amplified and then input to a bone conduction speaker. The quantization noise component generated by the ΔΣ modulation contains a lot of ultrasonic components, and these ultrasonic components work effectively as energy for driving the bone conduction speaker.

  Further, in the technology described in Patent Document 3 below, in an audio device that generates a tone signal generated when a key is touched or the like, data generated by sampling the generated tone signal at a sampling frequency higher than an audible frequency band is generated. Is subjected to DA conversion using ΔΣ modulation and then output to a speaker amplifier. By adopting such a configuration, it is possible to realize an audio device that generates a tone signal sound that is not affected by aliasing noise in the audible frequency band and that does not cause a sense of incongruity.

JP 2000-165304 A JP 2006-217088 A JP 2000-332611 A

  However, according to the technology of the conventional communication device, since an analog amplifier is used for amplification of a transmission signal, it is necessary to always apply a bias voltage, and there is a problem that power consumption increases.

  In addition, when a digital amplifier is applied using ΔΣ modulation, power saving of the transmission amplifier can be achieved. However, conventional AD conversion using ΔΣ modulation is used to transmit an audio signal. In order to obtain a high S / N ratio, the sampling frequency of the input signal is increased (for example, 128 times the maximum use frequency). However, in communication equipment where the frequency is higher than the audio signal, for example, the maximum usable frequency is several MHz, sampling the signal input to the ΔΣ modulation process with the same oversampling ratio achieves a sampling frequency of several GHz. Is difficult.

  Furthermore, when ΔΣ modulation is performed on an audio signal, it is not necessary to suppress especially high-frequency quantization noise and aliasing noise that occur during ΔΣ modulation if they are outside the audible frequency band. For example, in Patent Document 2, high-frequency quantization noise is positively used as energy transmitted by bone conduction. However, in communication equipment, it is necessary to remove quantization noise and aliasing noise outside the used frequency band so that they are not output to the transmission line, and the ΔΣ modulation technique relating to the audio signal described in Patent Document 2 is used as it is. It cannot be applied.

  Moreover, in the said patent document 3, although the auditory problem which arises by aliasing noise is solved by using the data sampled with the high sampling frequency for the production | generation of a tone signal sound, about the noise outside an audio frequency range, No particular mention is made, and no mention is made of quantization noise and aliasing noise generated during ΔΣ modulation.

  The present invention has been made in view of the above, and an object thereof is to provide a communication device with low power consumption while realizing low noise by applying a digital amplifier to a multi-carrier communication device.

  In order to solve the above-described problems and achieve the object, the present invention is a communication apparatus on the transmission side that employs a multi-carrier scheme, in which sampling data is generated by sampling a modulation signal at a predetermined frequency. Generating means; delta-sigma modulation means for converting the sampling data into 1-bit digital data; amplification means for amplifying the digital data by switching operation of a transistor; and frequency components other than the necessary frequency region from the amplified data. A low-pass filter for removing the delta-sigma modulation means, calculating an operating frequency of the delta-sigma modulation means based on a signal-to-noise ratio required for transmission data, and setting the predetermined frequency as the operating frequency.

  According to the present invention, the operating frequency of the ΔΣ modulator is determined so as to satisfy the SN ratio required for the system, and waveform data sampled at the same frequency as the operating frequency of the ΔΣ modulator is generated to generate the ΔΣ modulator. Since the digital amplifier amplifies the signal after conversion into 1-bit data, the effect of lowering the power consumption as compared with the conventional communication device is achieved while satisfying the required SN ratio.

  Embodiments of a communication apparatus according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
FIG. 1 is a diagram illustrating a functional configuration example of a communication device according to a first embodiment of the present invention. The communication apparatus according to the present embodiment adds header information and the like to data to be transmitted, and forms the frame forming unit 1 that forms the frame to be output to the transmission path, and transmits the frame-formed data to each carrier. A mapper 2 for assigning data to be processed, a phase converter 3 for converting the data assigned to each carrier into index data indicating the phase of the waveform, and generating a waveform of each carrier according to the index data, and performing predetermined sampling A carrier waveform generation unit 4 that generates sampling data sampled at a rate, and a carrier synthesis unit that adds the waveform of each carrier to the sampling data of each carrier waveform output from the carrier waveform generation unit 4 and outputs the result. 5. Converts multi-bit digital data to 1-bit digital data Σ modulator 6, amplifier 7 for amplifying 1-bit digital data by switching operation of transistor, and transmission path for extra frequency components such as quantization noise generated by ΔΣ modulator 6 for the amplified data And a low-pass filter 8 for suppressing the output of. Output data 14, 15, 16, and 18 are output data of the carrier waveform generation unit 4, the carrier synthesis unit 5, the ΔΣ modulator 6, and the low-pass filter 8, respectively.

The communication apparatus according to the present embodiment performs transmission by a multi-carrier scheme in which the basic frequency f _C is 1.28 MHz and the maximum number of carriers is 7. Further, the carrier frequency of the carrier #N which is the Nth carrier is N × f _C (N = 1 to 7), and the symbol frequency f _MOD which is a unit of modulation / demodulation is 40 kHz. The modulation method is DQPSK (Differential Quadrature Phase-Shift Keying).

For example, when the sampling frequency f _S is 20.48 MHz, f _S is 16 times f _C , and each carrier constitutes a sine wave for N cycles with 16 samples. One symbol is composed of 32 times (f _C / f _MOD ) of a sine wave for N periods. Note that f _C , N, and f _MOD need not be limited to this, and can be set to arbitrary numerical values according to the communication system.

  Next, the operation of the present embodiment will be described. First, the frame shaping section 1 adds header information containing frame length, address information, and other information necessary for demodulation to the inputted transmission data, and shapes the frame to be output to the transmission path.

  Next, the mapper 2 distributes the frame-formed transmission data to each carrier. FIG. 2 shows an example of distribution to each carrier. FIG. 2 illustrates an example in which transmission data is “123456h” (hexadecimal), and is distributed to three carriers, carriers # 1, # 2, and # 3, respectively. In DQPSK modulation, since each carrier can transmit information of 2 bits per symbol, transmission data is distributed to each carrier by 2 bits in order from the front. For example, in the first symbol (symbol S0 in FIG. 2), among the first 4 bits “0001” (binary) of the transmission data, the preceding 2 bits “00” is the carrier # 1 and the last 2 bits “01”. "Is distributed to carrier # 2. Then, among the next 4 bits “0010” (binary), the previous 2 bits “00” are distributed to carrier # 3, and the last 2 bits “10” are distributed to carrier # 1 of the next symbol (symbol S1). The Similarly, transmission data is distributed to each carrier in order of 2 bits.

  Next, the phase converter 3 converts the data distributed to each carrier into index data that is phase information based on DQPSK modulation. In DQPSK modulation, a 2-bit data value is associated with the phase difference between the previous symbol and the current symbol, and the phase corresponding to the current symbol is calculated based on the phase difference associated with the phase of the previous symbol as a reference. For example, when the transmission data is “00”, the phase is the same as that of the previous symbol, when it is “01”, the phase advanced by π / 2 radians from the previous symbol, and when it is “10”, it is π from the previous symbol. A phase delayed by / 2 radians is associated with the current symbol a phase different by π radians in the case of “11”. On the receiving side, if the phase difference between the current symbol and the previous symbol of the received frame is in the range of −π / 4 to π / 4, it is “00”, and if it is in the range of π / 4 to 3π / 4. If “01” is in the range of −π / 4 to −3π / 4, it is demodulated to “10”, and if it is in the range of 3π / 4 to π or −3π / 4 to −π, it is demodulated to “11”. To do.

  The phase converter 3 associates the data distributed for each carrier with a phase based on the above-described DQPSK modulation, and converts the phase into index data shown in association with a 2-bit numerical value. FIG. 3 shows an example of correspondence between index data and phase. FIG. 4 shows an example in which the phase corresponding to the transmission data shown in FIG. 2 is calculated based on the DQPSK modulation method, and the calculated phase is converted into index data by the correspondence shown in FIG. In order to obtain the phase of the current symbol and convert it to index data, the phase of the previous symbol is required. However, in FIG. 2, there is no previous symbol for symbol S0. Here, the symbols before symbol S0 in FIG. 4 are converted into index data on the assumption that the phase is 0 (index data is 0) in all carriers.

  For example, since the data of symbol S0 in carrier # 1 in FIG. 2 is “00” and has the same phase as the previous symbol (phase is 0), symbol S0 index data in carrier # 1 in FIG. 4 is “00”. . In the next symbol S1, the data of the symbol S1 in the carrier # 1 in FIG. 2 is “10”, so the phase difference from the previous symbol (symbol S0 in the carrier # 1) is −π / 2. Since the phase of the symbol S0 in the previous symbol carrier # 1 is 0, the phase of the symbol S1 is −π / 2, and the index data is “11” according to the correspondence in FIG.

  Next, the carrier waveform generation unit 4 uses the index data converted for each carrier, samples at a predetermined sampling frequency, and generates a digital waveform. Specifically, the carrier waveform generation unit 4 internally stores waveform data for each carrier and each phase (for each index value) as a waveform table, and sequentially outputs waveform data corresponding to the index data for each carrier. The waveform data is a data string (hereinafter referred to as sampling data) sampled at a predetermined sampling frequency.

The sampling frequency f _S of the waveform data needs to be at least twice the maximum frequency (hereinafter referred to as the maximum use band) f _max to be reproduced by the sampling theorem. Here, in order to simplify the device configuration, it is assumed that f _S and f _max are integer multiples of f _C. For this reason, f _max is the minimum value that exceeds the maximum frequency N × 1.28 = 8.96 MHz (N = 7) of the carrier and is an integral multiple of f _C , and is 1.28 × 8 = 10.24 MHz. To do. Therefore, in order to reproduce f _max , f _S needs to be 2 × f _max = 20.48 MHz or more according to the sampling theorem. Note that f _max need only be equal to or higher than the maximum frequency of the carrier (in this case, 8.96 MHz), and need not be limited to this value.

In the present embodiment, since the sampling frequency of the waveform data generated by the carrier waveform generation unit 4 is set to the operating frequency f _SD of the ΔΣ modulator 6, oversampling is performed to set the frequency necessary for the input of the ΔΣ modulator 6. Do. In the present embodiment, the oversampling is intended to mean that samples the waveform data beyond the sampling frequency required to reproduce the _{f max (2 × f max)} . In this embodiment, f _S is set to 327.68 MHz, which is 16 times 20.48 MHz (2 × f _max ), based on a server sampling ratio calculation method for the ΔΣ modulator 6 described later.

FIG. 5 shows an example of sampled data generated and sampled by the carrier waveform generation unit 4. In FIG. 5, the solid line corresponds to the waveform data before sampling, and the data for 256 samples corresponding to one cycle of f _C is shown. In FIG. 5, the points where the sampling points are thinned out to 1/16 are shown as black sampling points as thinned sampling points D1, D2,..., D16 for simplification. The density is 16 times that of D2,. Since the sampling frequency is 327.68 MHz, the interval between D1, D2,..., D16 is 1 / (327.68 / 16) = 1 / 20.48 [μs]. Since the carrier frequency of the carrier #N is (1.28 × N) MHz, in 256 samples, the data corresponds to N (= 256 × (327.68 / 1.28 × N)) cycles (carrier # 1 is for one cycle, carrier # 2 is for two cycles, and carrier # 3 is for three cycles).

FIG. 5 shows an example of sampling data when the phase is 0, but similar sampling data is also generated when the phase is π / 2, −π / 2, and π. The carrier waveform generation unit 4 sequentially samples 256 samples of the corresponding phase according to the index data for each carrier at an interval of 1 / f _S (in this embodiment, 1 / 327.68 [μs]). Output. Then, (f _C / f _MOD ) times of sampling data for 256 samples (in this embodiment, 32 times) is output as sampling data for one symbol.

  In FIG. 5, for example, the sampling data of carrier # 2 is repeated at every eight thinned sampling points, and the data is inverted at every four thinned sampling points. For this reason, the number of data held as a waveform table may be reduced by controlling the output order of sampling data and performing control with a positive / negative inversion circuit. For example, the sample data of carrier # 2 is a data string obtained by thinning out carrier # 1 sample data one by one. Similarly, the number of data may be reduced by controlling the output order.

  Next, the carrier synthesizing unit 5 synthesizes carriers by adding sampling data for each carrier and each phase output from the carrier waveform generating unit 4 for each sample point. 6 shows sampling data (2 × 256 = 512 samples) of 2N cycles of each carrier of the output data 14 of the carrier waveform generation unit 4 and the waveform of the output data 15 obtained by synthesizing them by the carrier synthesis unit 5. It is the figure which showed the concept.

FIG. 7 shows a conceptual diagram of the spectrum of the output data 15 of the carrier synthesizing unit 5. As shown in FIG. 7, this spectrum does not include extra frequency components other than the carrier frequency component up to f _S / 2. As described above, in the present embodiment, the sampling frequency f _S of the carrier waveform generation unit 4 and the carrier synthesis unit 5 is the same as the operating frequency f _SD of the ΔΣ modulator 6. Even if a waveform is generated at a sampling rate twice as high as f _max (10.24 MHz in this embodiment) by the sampling theorem, the waveform can be restored. In that case, it is necessary to perform upsampling before the input of the ΔΣ modulator 6 in order to adjust to f _SD . Note that up-sampling means increasing the sampling rate to be output by inserting “0” between the sampled data.

However, when upsampling is performed, aliasing noise is generated, and it is necessary to remove it. Figure 8-1 use band is F ₁ to F _2, a conceptual diagram showing the spectrum of data sampled at a higher sampling frequency than twice the F ₂ F _SP. FIG. 8-2 is a conceptual diagram showing the spectrum of the data after up-sampling by generating up-sampling data 2F _SP by up-sampling the sampled data shown in FIG. is there. As shown in FIG. 8-2, F _SP / 2 is the center of the folding, and folding noise is generated in the range of (F _SP -F ₂ ) to (F _SP -F ₁ ). Therefore, in order to remove out-of-band noise, it is necessary to remove aliasing noise when the frequency is adjusted to the input frequency to the ΔΣ modulator 6 by upsampling. In contrast, in the present embodiment, as described above, in order to have a sampling frequency subjected to oversampling by a carrier wave generator 4 the same as f _SD, it is not necessary to perform up-sampling, also aliasing There is no need to remove.

  Next, the ΔΣ modulator 6 converts the combined sampling data output from the carrier combining unit 5 into 1-bit data. Delta-sigma modulation can reduce the quantization noise with a high oversampling ratio, and can reduce the quantization noise in the use band by biasing the quantization noise to the high frequency side by noise shaping characteristics, thereby obtaining a high S / N ratio. Therefore, this technology is used mainly in the audio field.

  FIG. 9 is a diagram showing a configuration example of a secondary ΔΣ modulator as a configuration example of the ΔΣ modulator 6. As shown in FIG. 9, the delta-sigma modulator 6 includes difference units 60 and 62 that take a difference between input data and a delay unit 65, integrators 61 and 63 that add the previous sample output values, and an input according to the input. Based on the output of the quantizer 64 that outputs 1-bit data of “0” or “1”, and the output of the quantizer 64, when the output is “0”, the negative maximum value of the input amplitude value is When the output is “1”, the delay unit 65 holds and outputs a positive maximum value for one cycle.

  The operation of the ΔΣ modulator 6 will be described with reference to FIG. The subtractor 60 outputs a positive or negative input corresponding to the output of the delay unit 65 (the output of the quantizer 64 in the previous cycle described later) with respect to the combined sampling data (multi-bit) output from the carrier combining unit 5. And the maximum amplitude value) is output to the integrator 61. The integrator 61 sequentially adds the input data and outputs the result to the difference unit 62. The difference unit 62 takes the difference from the output of the delay unit 65 again for the data output from the integrator 61 and outputs the difference to the integrator 63. The integrator 63 adds the outputs of the difference unit 62 and outputs the result to the quantizer 64. The quantizer 64 converts the data output from the integrator 63 into 1-bit data. The quantizer 64 outputs “1” when the output of the integrator 63 is positive or “0”, and outputs “0” when it is negative. Based on the output of the quantizer 64, the delay unit 65 holds the negative maximum value of the input amplitude value when the output is “0”, and holds the positive maximum value when the output is “1”. Then, it is output to the differentiators 60 and 62 as the value of the previous cycle. The ΔΣ modulator 6 is not limited to this configuration, and any ΔΣ modulator that converts input data into 1-bit data can be used.

The quantization noise generated by ΔΣ modulation has a characteristic that it is low on the low frequency side and becomes higher as it approaches half f _SD / 2 of the operating frequency f _SD . Thus, the ratio of f _SD for _{f max (f SD / f max} ), that is, by increasing the oversampling ratio, can be mentioned SN ratio in the use band. However, increasing the oversampling ratio means that the operating frequency becomes high. If the frequency is high, circuit implementation becomes difficult and power consumption increases. Therefore, it is desirable that the operating frequency be as low as possible within a range that satisfies the S / N ratio requirement.

Here, a method of calculating the oversampling ratio in the present embodiment will be described. The bit error rate in a certain modulation system is determined by the SN ratio. Therefore, if the bit error rate required for the system is determined, the S / N ratio necessary to satisfy the bit error rate is determined. For example, when the modulation method is DQPSK, the SN ratio needs to be about 15 dB or more in order to satisfy the bit error rate of 10 ⁻⁵ or less. On the other hand, the amount and characteristics of quantization noise generated in ΔΣ modulation depend on the order and configuration of the ΔΣ modulator. For example, in the second-order ΔΣ modulator, if the oversampling ratio is 16 or more, that is, if the operating frequency is 32 times or more of the use band f _max , the S / N ratio exceeds 15 dB.

For this reason, in this embodiment, the oversampling ratio is set to 16, and the operating frequency of the ΔΣ modulator is set to 327.68 MHz, which is 32 times f _max . As described above, in this embodiment, since upsampling is not performed, the sampling frequency f _S in the carrier waveform generation unit 4 is set to be the same as the operating frequency f _SD of the ΔΣ modulator (327.68 MHz). This numerical value of 32 times is a numerical value lower than the oversampling ratio (for example, 128 times) of ΔΣ modulation used in conventional audio. In this way, by adopting the above-described oversampling ratio calculation method, it is possible to set a value smaller than the oversampling ratio of ΔΣ modulation used in conventional audio. The bit error rate, SN ratio, and oversampling ratio described here are examples. If the oversampling ratio is obtained from the required bit error rate by the same calculation method according to the communication apparatus and communication system to be applied. Good.

  Next, the amplifier 7 amplifies the output of the ΔΣ modulator 6 by the switching operation of the transistor. Unlike a linear amplification operation with an analog amplifier, a bias voltage is not required, so that power consumption can be suppressed. The low pass filter 8 removes quantization noise and aliasing noise from the output of the amplifier 7.

10-1 is a diagram conceptually showing the output data 15 of the carrier synthesizing unit, FIG. 10-2 is a conceptual diagram showing the output data 16 of the ΔΣ modulator, and FIG. 10-3 is the low-pass filter output data 18. . The output 15 of the carrier synthesizing unit 5 shown in FIG. 10-1 is converted into a 1-bit signal as shown in FIG. 10-2 by the ΔΣ modulator. FIG. 11 shows a spectrum when the 1-bit signal shown in FIG. 10-2 is considered as an analog signal. As shown in FIG. 11, this spectrum is for quantization noise closer to f _S / 2 is increased, it contains aliasing noise caused by the folding around an axis of f _S / 2. Since the output from the amplifier 7 also includes this quantization noise and aliasing noise, in the present embodiment, these are removed by the low-pass filter 8 as described above. FIG. 12 is a diagram showing a spectrum of the output data 18 of the low-pass filter 8. As shown in FIG. 12, quantization noise and aliasing can be removed by a low-pass filter. Further, as shown in FIG. 10-3, the output data 18 of the low-pass filter 8 has a waveform similar to that shown in FIG.

  As described above, in the present embodiment, the operating frequency of the ΔΣ modulator 6 is determined so as to satisfy the SN ratio required to satisfy the bit error rate required for the system, and the operating frequency of the ΔΣ modulator 6 is determined. Waveform data sampled at the same frequency as in FIG. 1, converted into 1-bit data by the ΔΣ modulator 6, amplified by the transistor switching operation in the amplifier 7, and then the quantization error and aliasing noise are removed in the low-pass filter. I did it. For this reason, power consumption can be reduced compared with the conventional communication apparatus, satisfy | filling required SN ratio.

Embodiment 2. FIG.
FIG. 13 is a diagram illustrating a functional configuration example of the communication apparatus according to the second embodiment of the present invention. As shown in FIG. 13, the communication apparatus of the present embodiment removes the aliasing component generated by upsampling 9 and upsampling that inserts “0” between input data in the functional configuration example of the first embodiment. The interpolation filter 10 which is a digital filter is added, and the carrier waveform generation unit 4 and the carrier synthesis unit 5 of the first embodiment are replaced with a carrier waveform generation unit 4a and a carrier synthesis unit 5a, respectively. The output data 14a, 15a, and 19 are output data of the carrier waveform generation unit 4a, the carrier synthesis unit 5a, and the upsampler 9, respectively. Components having the same functions as those of the first embodiment are denoted by the same reference numerals and description thereof is omitted. The carrier frequency and symbol frequency are the same as in the first embodiment.

  In the present embodiment, the carrier waveform generation unit 4a generates sampling data at a sampling frequency higher than twice the use band, performs upsampling by the upsampler 9, and then performs upsampling using a digital filter. The resulting aliasing is removed and input to the ΔΣ modulator 6.

  Next, the operation of the present embodiment will be described. Since the process from the transmission data to the index data indicating the phase by the phase converter 3 is the same as that of the first embodiment, the description thereof is omitted.

The carrier waveform generation unit 4a generates a digital waveform using the index data converted for each carrier. Specifically, the carrier waveform generation unit 4a internally holds a waveform table composed of waveform data for each carrier and each phase, and sequentially outputs waveform data corresponding to the index for each carrier. The waveform data is sampling data sampled at a predetermined sampling frequency f _S.

In the present embodiment, unlike Embodiment 1, f _{S is set} to 20.48 MHz, which is twice the maximum usable bandwidth of 10.24 MHz. Therefore, in the present embodiment, the waveform for N cycles of the sampling data is composed of 16 sample points every 1 / 20.48 [μs]. Therefore, sampling is performed at a period 16 times that of the first embodiment, and the points indicated by D1, D2,..., D16 of the waveform data shown in FIG. To do. Then, (f _C / f _MOD ) times of sampling data for 16 samples (32 times in this embodiment) is output as sampling data for one symbol.

  Next, the carrier synthesizer 5a synthesizes the carrier by adding the sampling data for each carrier output from the carrier waveform generator 4a for each sample point. The operation of the carrier synthesizing unit 5 of the first embodiment is the same except that the sampling period is different. FIG. 14 shows a conceptual diagram of the spectrum of the output data 15a of the carrier synthesizer 5a.

Next, the upsampler 9 performs an upsampling process by inserting “0” between the data so that a predetermined upsampling ratio is obtained with respect to the sampling data after synthesis of the output of the carrier synthesis unit 5a. Also in the present embodiment, as in the case of the first embodiment, the operating frequency f _SD of the ΔΣ modulator 6 is set to 32 times or more of f _max in order to satisfy the bit error rate necessary for the system. Therefore, as in the first embodiment, the operating frequency f _SD of the ΔΣ modulator 6 is set to 327.68 MHz which is 32 times f _max . Therefore, the upsampling ratio f _SD / f _S is 16, and in the upsampling process, 15 “0” s are inserted between the data of the combined sampling data.

FIG. 15 shows a conceptual diagram of the spectrum of the output data 19 of the upsampler 9. In FIG. 15, in order to make it easy to understand the return state, the use spectrum region is indicated by a triangle instead of each band shown in FIG. 14 from carrier # 1 to carrier # 7. A triangle indicated by a solid line is an original use spectrum region. As shown in FIG. 15, aliasing noise indicated by a dotted line occurs in a range from f _max (= f _S / 2) to f _SD / 2 (= f _max × 16).

  Next, the interpolation filter 10 removes aliasing noise generated by the upsampling process of the upsampler 9.

  Next, the ΔΣ modulator 6 converts the output of the interpolation filter 10 into a 1-bit signal. The processing of the ΔΣ modulator 6 and the subsequent processing are the same as in the first embodiment.

As described above, in the present embodiment, the operating frequency f _SD of the ΔΣ modulator 6 is determined so as to satisfy the SN ratio necessary to satisfy the bit error rate required for the system, and is twice the maximum use frequency. The waveform of the transmission waveform is generated at the above frequency f _S , the up sampler 9 performs the up sampling process with the up sampling ratio of f _SD / f _S , the aliasing noise is removed by the interpolation filter 10, and the ΔΣ modulator 6 It was made to input in. Therefore, f _S can be set to the minimum frequency, and the sampling circuit scale and power consumption can be reduced as compared with the first embodiment.

Embodiment 3 FIG.
FIG. 16 is a diagram illustrating a functional configuration example of the communication apparatus according to the third embodiment of the present invention. As shown in FIG. 16, the communication apparatus according to the present embodiment adds an upsampler 9a that inserts “0” between input data to the functional configuration example of the first embodiment. The generation unit 4 and the carrier synthesis unit 5 are replaced with a carrier waveform generation unit 4b and a carrier synthesis unit 5b, respectively. The output data 14b, 15b, and 19a are output data of the carrier waveform generation unit 4b, the carrier synthesis unit 5b, and the upsampler 9a, respectively. Components having the same functions as those in the first or second embodiment are denoted by the same reference numerals and description thereof is omitted. The carrier frequency and symbol frequency are the same as in the first embodiment.

  In the present embodiment, the upsampling ratio is determined so that the aliasing noise generated by the upsampling process is within a range that can be removed by the low-pass filter 8, and the sampling frequency of the carrier waveform generation unit 4b is based on the upsampling ratio and the oversampling ratio. To decide.

  Next, the operation of the present embodiment will be described. Since the process from the transmission data to the index data indicating the phase by the phase converter 3 is the same as that of the first embodiment, the description thereof is omitted.

  The carrier waveform generation unit 4b generates a digital waveform using the index data converted for each carrier. Specifically, the carrier waveform generation unit 4b internally holds a waveform table composed of waveform data for each carrier and each phase, and sequentially outputs waveform data corresponding to the index for each carrier. The waveform data is sampling data sampled at a predetermined sampling frequency.

In the present embodiment, unlike the first embodiment, the sampling frequency f _{S is set} to 81.96 MHz, which is four times the double of the maximum usable band of 10.24 MHz, based on a request for an upsampling ratio described later. Therefore, in the present embodiment, the waveform of N cycles of the sampling data is composed of 64 sample points every 1 / 81.96 [μs]. Then, (f _C / f _MOD ) times of sampling data for 64 samples (in this embodiment, 32 times) is output as sampling data for one symbol.

Next, the carrier synthesizing unit 5b synthesizes the carrier by adding the sampling data for each carrier output from the carrier waveform generating unit 4b for each sample point. FIG. 17 shows a conceptual diagram of the spectrum of the output data 15b of the carrier synthesizing unit 5b. As shown in FIG. 17, the output data 15b of the carrier synthesizing unit 5b has a spectrum that does not include an extra frequency component other than the carrier frequency in the frequency domain up to f _S / 2.

Next, the upsampler 9a performs an upsampling process by inserting “0” between the data so as to obtain a predetermined upsampling ratio with respect to the sampling data after the synthesis of the output of the carrier synthesizing unit 5b. Also in the present embodiment, the operating frequency f _SD of the ΔΣ modulator 6 is set to 32 times or more of f _max in order to satisfy the bit error rate necessary for the system, as in the first embodiment. Therefore, as in the first embodiment, the operating frequency f _SD of the ΔΣ modulator 6 is 327.68 MHz, which is 32 times f _max , and the frequency of the output data from the upsampler 9a is also f _SD . The upsampling ratio is determined based on a request for an upsampling ratio described later. In the present embodiment, since it is 4, in the upsampling process, three “0” s are inserted between each piece of combined sampling data.

FIG. 18 shows a conceptual diagram of the spectrum of the output data 19a of the upsampler 9a. In FIG. 18, in order to make it easy to understand the folding state, the use spectrum region is represented by a triangle instead of each band shown in FIG. 17 from carrier # 1 to carrier # 7. A triangle indicated by a solid line is an original use spectrum region. As shown in FIG. 18, aliasing noise indicated by a dotted line occurs in a range from f _max (= f _S / 8) to f _SD / 2 (= f _max × 16).

  Next, the ΔΣ modulator 6 converts the output of the upsampler 9a into a 1-bit signal. The processing of the ΔΣ modulator 6 and the subsequent processing are the same as in the first embodiment.

  FIG. 19 is a conceptual diagram of the spectrum of the output data 16 of the ΔΣ modulator 6. Unlike the second embodiment, the present embodiment does not remove the aliasing component caused by upsampling, and causes aliasing noise as shown in the figure. However, unlike the second embodiment, the aliasing noise is generated at a frequency position away from the original use spectrum region, and therefore can be removed together with the quantization noise by the low-pass filter 8. That is, in consideration of the characteristics of the low-pass filter 8, the sampling ratio in the upsampler 9 a is determined so that aliasing noise is generated in a range that can be removed by the low-pass filter 8. The poration filter 10 becomes unnecessary.

  Therefore, the requirement for the upsampling ratio of the present embodiment is to consider the characteristics of the low-pass filter 8 and to set a numerical value that causes aliasing noise in a range that can be removed by the low-pass filter 8. In the present embodiment, the upsampling ratio is set to 4 from the characteristics shown in FIG. 19, but it can be set to a larger value when the filter characteristics are steeper, for example.

As described above, in the present embodiment, the operating frequency f _SD of the ΔΣ modulator 6 is determined so as to satisfy the SN ratio necessary to satisfy the bit error rate required for the system, and further, it is generated by upsampling. The upsampling ratio is determined so that aliasing noise is removed by the low-pass filter 8, and sampling is performed by determining the frequency f _S so as to satisfy the upsampling ratio. For this reason, f _S can be made lower than that in the first embodiment, and since it is not necessary to provide an interpolation filter, the configuration can be simplified and the power consumption can be reduced as compared with the second embodiment.

  As described above, the communication apparatus according to the present invention is useful for multicarrier communication, and is particularly suitable for communication with low power consumption while realizing low noise.

It is a figure which shows the function structural example of Embodiment 1 of the communication apparatus concerning this invention. It is a figure which shows the example of distribution of the data to a carrier. It is a figure which shows the example of a response | compatibility of index data and a phase. It is a figure which shows the example which converted transmission data into index data. It is a figure which shows the example of the sampling data produced | generated by the carrier waveform production | generation part. It is a figure which shows the sampling data for 2N period of each carrier. It is a figure which shows the conceptual diagram of the spectrum of the output of a carrier synthetic | combination part. It is a conceptual diagram showing the spectrum of the data sampled with a sampling frequency higher than twice the maximum use band. It is a conceptual diagram showing the spectrum of the data which performed upsampling. It is a figure which shows the structural example of a secondary delta-sigma modulator. It is a conceptual diagram of the output of a carrier synthetic | combination part. It is a conceptual diagram of the output of a ΔΣ modulator. It is a conceptual diagram of the output of a low-pass filter. It is a conceptual diagram of the spectrum of the output of a ΔΣ modulator. It is a conceptual diagram of the spectrum of the output of a low-pass filter. It is a figure which shows the function structural example of Embodiment 2 of the communication apparatus concerning this invention. It is a conceptual diagram of the spectrum of the output of a carrier synthetic | combination part. It is a conceptual diagram of the spectrum of the output of an up sampler. It is a figure which shows the function structural example of the communication apparatus of Embodiment 3 concerning this invention. FIG. 10 is a conceptual diagram of an output spectrum of a carrier synthesis unit according to the third embodiment. It is a conceptual diagram of the spectrum of the output of an up sampler. It is a conceptual diagram of the spectrum of the output of a ΔΣ modulator.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Frame shaping part 2 Mapper 3 Phase converter 4, 4a, 4b Carrier waveform production | generation part 5, 5a, 5b Carrier synthetic | combination part 6 Delta-sigma modulator 7 Amplifier 8 Low pass filter 9, 9a Upsampler 10 Interpolation filter 14, 14a, 14b, 15, 15a, 15b, 16, 18, 19, 19a Output data 60, 62 Differentiator 61, 63 Integrator 64 Quantizer 65 Delay device

Claims (7)

A communication device on the transmission side adopting a multi-carrier method,
Sampling data generating means for generating sampling data obtained by sampling the modulation signal at a predetermined frequency;
Delta-sigma modulation means for converting the sampling data into 1-bit digital data;
Amplifying means for amplifying the digital data by a switching operation of a transistor;
A low-pass filter for removing frequency components other than the necessary frequency region from the amplified data;
With
A communication apparatus characterized in that an operating frequency of the delta-sigma modulation means is calculated based on a signal-to-noise ratio required for transmission data, and the predetermined frequency is set as the operating frequency. A communication device on the transmission side adopting a multi-carrier method,
Sampling data generating means for generating sampling data obtained by sampling the modulation signal at a predetermined frequency;
An upsampler that generates upsampling data upsampled at a predetermined upsampling ratio by inserting 0 between each piece of the sampling data;
An interpolation filter that generates filtering data obtained by removing aliasing noise from the upsampling data;
Delta-sigma modulation means for converting the filtering data into 1-bit digital data;
Amplifying means for amplifying the digital data by a switching operation of a transistor;
A low-pass filter for removing frequency components other than the necessary frequency region from the amplified data;
With
An operating frequency of the delta-sigma modulation means is calculated based on a signal-to-noise ratio required for transmission data, the predetermined frequency is set to be twice or more of a maximum use band, and the upsampling ratio is set to the operating frequency and the predetermined frequency A communication device characterized by having a ratio to the frequency of A communication device on the transmission side adopting a multi-carrier method,
Sampling data generating means for generating sampling data obtained by sampling the modulation signal at a predetermined frequency;
An upsampler that generates upsampling data upsampled at a predetermined upsampling ratio by inserting 0 between each piece of the sampling data;
Delta-sigma modulation means for converting the up-sampling data into 1-bit digital data;
Amplifying means for amplifying the digital data by a switching operation of a transistor;
A low-pass filter for removing frequency components other than the necessary frequency region from the amplified data;
With
The operating frequency of the delta-sigma modulation means is calculated based on a signal-to-noise ratio required for transmission data, and the up-sampling ratio is a numerical value at which aliasing noise generated by the up-sampling process can be removed by the low-pass filter. The communication apparatus is characterized in that the predetermined frequency is obtained based on the operating frequency and the upsampling ratio.   The communication apparatus according to claim 3, wherein the upsampling ratio is four times or more.   The communication apparatus according to claim 1, wherein the operating frequency is 32 times or more of a maximum use band.   6. The communication apparatus according to claim 1, wherein a modulation method of the modulation signal is a DQPSK (Differential Quadrature Phase-Shift Keying) method.   The communication apparatus according to claim 1, wherein the delta-sigma modulator is a second-order delta-sigma modulator.

Images