JP2018019246A - Peak suppression circuit and peak suppression method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve a modulation accuracy characteristic, to satisfy a severe error vector magnitude standard with 256 QAM introduction and to implement peak factor reduction equal to the prior arts.SOLUTION: A peak suppression circuit comprises: a pulse generation part 100 for generating a first pulse signal by multiplying a value resulting from subtracting a predetermined threshold from an amplitude value of an input signal in which the band of a spectrum is limited, by phase information of the input signal; a band limit filter part 101 having multiple tap coefficients and outputting a second pulse signal of a band-limited pulse signal by limiting a frequency band of the first pulse signal in accordance with filter properties that are set based on carrier information from the outside and determined by the tap coefficients; a subtraction part 103 for subtracting the second pulse signal from the input signal; and a tap coefficient calculation part 102 for determining the tap coefficients in such a manner that a pass band of the band limit filter part 101 becomes wider than the spectrum bandwidth of the input signal within a predetermined range.SELECTED DRAWING: Figure 12

Description

  The present invention relates to a peak suppression circuit and a peak suppression method for reducing a peak factor of a transmission signal.

In mobile communications, a peak suppression circuit that reduces the peak factor of a transmission signal is employed to achieve a low back-off operation of the power amplifier for the purpose of reducing the size of the transmitter and reducing power consumption. . The peak suppression circuit performs signal processing for suppressing a peak component while allowing a slight signal quality degradation to the transmission signal.
As an example of the peak suppression circuit, Patent Document 1 describes a power limiting circuit to which a technique called PC-CFR (Peak Cancellation-Crest Factor Reduction) is applied. In this power limiting circuit, a peak suppression pulse signal for canceling a peak value equal to or higher than a preset threshold value is generated. Then, the frequency band of the peak suppressing pulse signal is limited so that it falls within the spectrum band (channel band) of the transmission signal, and the peak suppressing pulse signal with the band limited is subtracted from the input signal.

According to the power limiting circuit, it is possible to eliminate the spread of the spectrum outside the band of the transmission signal by using the band-limited peak suppressing pulse signal. Furthermore, since peak suppression processing can be performed in a narrow range in time compared to other peak suppression techniques, signal quality degradation can be suppressed as much as possible.
However, all of the existing peak suppression circuits including the power limiting circuit described above change the output signal of the peak suppression circuit with respect to the original input signal as a result of suppressing the peak component from the input signal. To do. For this reason, in the output of the peak suppression circuit, the trade-off between the deterioration amount of the modulation accuracy (EVM: Error Vector Magnitude) determined by the RMS (Root Mean Square) value of the error vector from the original input signal and the peak suppression amount is necessary.

Also, in the recent field of mobile communication, OFDM (Orthogonal Frequency Division Multiplex) modulation is adopted in the downlink of the LTE (Long Term Evolution) system in order to improve frequency utilization efficiency.
In Release 8 to Release 11 of 3GPP (3rd Generation Partnership Project), the primary modulation scheme of LTE was up to 64QAM. However, in order to improve the peak data rate of downlink, 256QAM was newly introduced in Release 12. It was done. Along with it, 3GPP TS 36.104: Technical Specification Group Number of Radio Q's for the Technical Group of Radio Group Access Q and the number of QQs for the standard QQ of the Evolved Universal Random Access Base Radiation As a standard of modulation accuracy (Error Vector Magnet), 3.5%, which is stricter than 8% at 64QAM, is defined.

上述したように、２５６ＱＡＭの変調精度の規格は、６４ＱＡＭの変調精度の規格よりも厳しい値になっているため、ピーク抑圧回路で生じる変調精度を改善する必要がある。 The modulation accuracy (EVM) of QPSK, 16QAM, 64QAM, and 256QAM will be described below.

As described above, since the 256QAM modulation accuracy standard is stricter than the 64QAM modulation accuracy standard, it is necessary to improve the modulation accuracy generated in the peak suppression circuit.

International Publication No. 2010/074187

  When the modulation accuracy (EVM) standard accompanying the introduction of 256QAM has been reviewed, the influence of deterioration due to the phase noise of the local oscillator in the transmission apparatus, which did not greatly contribute in the case of standard 8%, is the case where the new standard is 3.5% Since the contribution becomes large, it is necessary to consider up to several percent deterioration distribution due to the phase noise.

ここで、図２における入力ベクトルの振幅ｒｍｓ値をＲｉｎとすると、エラーベクトルの振幅ＲＭＳ値のＲｅｒｒｏｒは、次の（２）式の通りである。

FIG. 2 is a conceptual diagram illustrating the influence of modulation accuracy (EVM) due to the phase noise of the local oscillator in the transmission apparatus.
When the single sideband power ratio of the phase-modulated wave to the carrier is set as phase noise PN (f) [dBc], the square (power) of the RMS value Δθrms of the phase error is expressed by the following equation (1).

Here, when the amplitude rms value of the input vector in FIG. 2 is Rin, the error RMS value of the error vector is represented by the following equation (2).

Therefore, the RMS value of the modulation accuracy due to the error vector caused by the phase noise can be approximately expressed by the following equation (3) from the above equation (1) and the above equation (2).

より）、２．４５％までしか許容されないことから現実的ではない。 From the above equation (3), if the integrated value of the phase noise PN (f) is −35 dBc, the modulation accuracy of the error vector caused by the phase noise is 2.5%. Since the modulation accuracy of the transmission device is roughly determined by the phase noise of the peak suppression circuit and the local oscillator, the degradation distribution to the peak suppression circuit that satisfies the modulation accuracy standard of 3.5% of the transmission device in 256QAM is (

It is not realistic because only 2.45% is allowed.

より）、３．３５％まで許容されるが、それでも６４ＱＡＭの場合に比べて大幅に小さくなる。
代表的なＰＣ−ＣＦＲ方式を用いたピーク抑圧回路は、７．２〜７．３ｄＢのＰＡＰＲ（Ｐｅａｋ ｔｏ Ａｖｅｒａｇｅ Ｐｏｗｅｒ Ｒａｔｉｏ）のピーク抑圧時に、変調精度（ＥＶＭ）が４．５〜５．０％の特性を有している。以下に、このピーク抑圧回路における、変調精度が４．５％の場合の２５６ＱＡＭ時の影響を説明する。 Even when it is assumed that the integral value of the phase noise PN (f) is improved to −43 dBc, the modulation accuracy of the error vector caused by the phase noise is 1.0%, so that the modulation accuracy standard 3.5 at 256QAM is 3.5. % Degradation distribution to the peak suppression circuit that satisfies (

To 3.35%, but still much smaller than 64QAM.
A peak suppression circuit using a typical PC-CFR system has a modulation accuracy (EVM) of 4.5 to 5.0% during peak suppression of 7.2 to 7.3 dB PAPR (Peak to Average Power Ratio). It has the characteristics. In the following, the influence of 256QAM when the modulation accuracy is 4.5% in this peak suppression circuit will be described.

FIG. 3A shows a constellation comparison result in the case of adding an error in which the modulation accuracy becomes 4.5% and 3.5% at 64 QAM. FIG. 3B shows a constellation comparison result when an error that causes modulation accuracy to be 4.5% and 3.5% at 256QAM is added.
As shown in FIG. 3A, in the case of 64QAM, the adjacent signal points in the constellation are sufficiently separated from each other in both the modulation accuracy of 3.5% error and 4.5% error. On the other hand, in the case of 256QAM, since the number of bits per symbol is larger than that of 64QAM, as shown in FIG. 3B, the adjacent signal points in the constellation in the case of an error with a modulation accuracy of 4.5% are blurred. Yes. For this reason, in order to ensure a sufficient interval between adjacent signal points, the error must be reduced to a modulation accuracy of 3.5%.

FIG. 4 is a conceptual diagram for explaining the modulation accuracy (EVM) degradation that occurs in the peak suppression circuit. The vertical axis represents the Q component, and the horizontal axis represents the I component.
If the amplitude threshold for limiting the peak amplitude with respect to the amplitude R of the PC-CFR input vector is Th, the amplitude of the peak suppression signal vector to be subtracted from the input vector is (R-Th). Here, since the peak suppression subtraction signal vector for obtaining the output vector after peak suppression is an error vector for the input signal, the modulation accuracy is deteriorated. That is, to improve the modulation accuracy generated in the peak suppression circuit is nothing but mitigating the peak suppression amount by reducing the peak suppression subtraction signal corresponding to the error vector.

Even in the RGP-131472 and R4-134065 contributions in 3GPP RAN WG4 until the modulation accuracy standard of 3.5% in 256QAM is determined, peak suppression is lower in 256QAM than in the current modulation accuracy standard of 8% in 64QAM. It has been argued that there is an effect that must be relaxed and the back-off amount of the power amplifier must be increased by 1 dB to several dB.
From the conditions described above, if the existing technology is used as it is, the peak suppression amount in the peak suppression circuit has to be relaxed, and as a result, it is necessary to secure an operation back-off amount of the power amplifier that is larger than the conventional one. There was a problem that the power consumption of the apparatus would increase.

  An object of the present invention is to provide a peak suppression circuit and a peak suppression method capable of improving modulation accuracy characteristics, satisfying strict modulation accuracy standards accompanying the introduction of 256QAM, and realizing a peak factor reduction equivalent to that of the existing technology. It is to provide.

In order to achieve the above object, according to one aspect of the present invention,
A pulse generator for generating a first pulse signal by multiplying a value obtained by subtracting a predetermined threshold value from an amplitude value of an input signal whose spectrum is band-limited, and phase information of the input signal;
The first pulse signal is a band-limited pulse signal having a plurality of tap coefficients, limiting the frequency band of the first pulse signal according to a filter characteristic determined by the tap coefficient set based on carrier information from the outside, A band limiting filter unit that outputs a pulse signal of 2;
A subtractor for subtracting the second pulse signal from the input signal;
There is provided a peak suppression circuit comprising: a tap coefficient calculation unit that determines the plurality of tap coefficients so that a pass band of the band limiting filter unit is wider than a spectrum bandwidth of the input signal within a predetermined range. The

According to another aspect of the invention,
A value obtained by subtracting a predetermined threshold from the amplitude value of the input signal whose spectrum is band-limited is multiplied by the phase information of the input signal to generate a first pulse signal,
Using a band limiting filter that has a plurality of tap coefficients and whose filter characteristics are determined by the tap coefficients set based on carrier information from the outside, limits the frequency band of the first pulse signal,
A peak suppression method that subtracts a second pulse signal that is a pulse signal band-limited by the band-limiting filter from the input signal,
A peak suppression method is provided in which the plurality of tap coefficients are set such that a pass band of the band limiting filter unit is wider than a spectrum bandwidth of the input signal within a predetermined range.

  According to the present invention, it is possible to improve the modulation accuracy characteristics and realize the peak factor reduction equivalent to the existing technology.

It is a block diagram which shows the structure of the peak suppression circuit by one Embodiment of this invention. It is a conceptual diagram for demonstrating the modulation precision (EVM) influence by the phase noise of the local oscillator in a transmitter. It is a figure which shows the constellation comparison result at the time of adding the error that modulation accuracy becomes 4.5% and 3.5% at 64QAM. It is a figure which shows the constellation comparison result at the time of adding the error from which modulation accuracy becomes 4.5% and 3.5% at 256QAM. It is a conceptual diagram for demonstrating degradation of the modulation accuracy (EVM) which arises in a peak suppression circuit. It is a figure which shows each spectrum of the pulse signal for peak suppression in the comparative example at the time of 20 MHz of LTE, the characteristic of a band limitation filter, and the peak suppression signal after a band limitation filter. It is a figure which shows each spectrum of the LTE carrier signal after a LTE channel filter part in the comparative example at the time of LTE 20MHz, and the peak suppression signal after a band-limiting filter, and an operating band unnecessary emission standard. It is a figure explaining the pass-band determination method of the band-limiting filter part performed with the peak suppression circuit shown in FIG. It is a figure which shows the difference of the filter gain accompanying the band-limiting filter characteristic of the comparative example, the band-limiting filter characteristic of the peak suppression circuit shown in FIG. 1, and those pass-band differences. It is a figure which shows the tap coefficient of the band-limiting filter in a comparative example. It is a figure which shows the tap coefficient of the band-limiting filter of the peak suppression circuit shown in FIG. FIG. 3 is a diagram illustrating each spectrum of a peak suppression pulse signal at 20 MHz LTE, the characteristics of a band limit filter, and a peak suppression signal after the band limit filter in the peak suppression circuit illustrated in FIG. 1. FIG. 3 is a diagram illustrating each spectrum of an LTE carrier signal after channel filtering and a peak suppression signal after band limiting filtering in the peak suppression circuit shown in FIG. 1 and an operating band unnecessary emission standard. It is a block diagram which shows the structure of the peak suppression circuit by other embodiment of this invention.

Next, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a peak suppression circuit according to an embodiment of the present invention.
Referring to FIG. 1, a PC-CFR circuit 3 that is a peak suppression circuit includes an amplitude calculation unit 4, a threshold calculation unit 5, a delay adjustment unit 6, 10, a pulse generation unit 7, a tap coefficient calculation unit 8, and a band limiting filter unit. 9 and a subtracting unit 11.

  The transmission apparatus includes an LTE channel filter unit 1, a digital up-converter unit 2, a PC-CFR circuit 3, a distortion compensation circuit 12, and the like. The LTE channel filter unit 1 limits the spectrum of the side lobe of the OFDM signal so as to satisfy each carrier signal requirement of the transmission system. The baseband IQ signal input to the transmission apparatus is converted into a signal whose spectrum is band-limited by the LTE channel filter unit 1, and then converted to a required frequency by the digital up-converter unit 2 to perform rate conversion. Is called. The CFR input IQ signal that is the output of the digital up-converter unit 2 is input to the PC-CFR circuit 3.

In the PC-CFR circuit 3, the CFR input IQ signal is supplied to the amplitude calculation unit 4 and the delay unit adjustment unit 6. The amplitude calculator 4 calculates the amplitude value R of the input IQ signal. The threshold value subtracting unit 5 subtracts an arbitrarily set threshold value Th from the amplitude value R to obtain (R−Th). The threshold value subtracting unit 5 performs processing for setting a signal point of (R−Th) <0 among the subtraction results to a 0 (zero) level. The threshold subtraction unit 5 supplies the subtraction result (R-Th) to the pulse generation unit 7.
The delay adjusting unit 6 gives a predetermined delay to the input IQ signal. Here, the predetermined delay corresponds to the time required for processing from the amplitude calculation unit 4 to the pulse generation unit 7. The IQ signal that is the output of the delay adjustment unit 6 is supplied to the pulse generation unit 7 and the delay adjustment unit 10.

The pulse generation unit 7 multiplies (R−Th) from the threshold subtraction unit 5 by the phase information exp (jθ) of the input IQ signal from the delay adjustment unit 6 to obtain (R−Th) · exp (jθ). Generate a pulse. Here, exp (jθ) = cos θ + jsin θ (j is a subscript representing an imaginary number). In this pulse generation process, a vector (pulse) serving as a peak suppression signal for limiting only the amplitude to the threshold Th without changing the phase of the signal point having the amplitude equal to or greater than the threshold Th on the complex plane is calculated. It is processing.
The peak suppression pulse signal that is the output of the pulse generator 7 is input to the band limiting filter unit 9. The band limiting filter unit 9 is a filter circuit including a plurality of tap coefficients, for example, an FIR filter. The band limiting filter unit 9 limits the frequency band of the peak suppressing pulse signal from the pulse generating unit 7 in accordance with filter characteristics determined by a plurality of tap coefficients. The band-limited peak suppressing pulse signal output from the band-limiting filter unit 9 is supplied to the subtracting unit 11.

The delay adjustment unit 10 gives a predetermined delay to the input IQ signal from the delay adjustment unit 6. Here, the predetermined delay corresponds to the time required for processing from the pulse generation unit 7 to the band limiting filter unit 9. The IQ signal that is the output of the delay adjustment unit 10 is supplied to the subtraction unit 11. By the delay adjustment of the delay adjustment unit 6 and the delay adjustment unit 10, the timing of the center tap of the band limiting filter unit 9 coincides with the timing of the peak detection point detected in the CFR input IQ signal and subjected to suppression.
The subtracting unit 11 subtracts the output signal of the band limiting filter unit 9 from the CFR input IQ signal after delay adjustment, which is the output signal of the delay adjusting unit 10. Thereby, the peak of the CFR input IQ signal is suppressed to the threshold Th. In order to realize this, the gain of the tap coefficient of the band limiting filter unit 9 is not calculated so that the input and output are at the same level, but instead of the band limiting filter unit 9 at the timing corresponding to the peak to be suppressed. It is necessary to design the tap coefficient center tap to be (1.0 + j0.0) times so that the output pulse amplitude becomes equal to the difference between the peak amplitude and the threshold value (R-Th).

The tap coefficient calculation unit 8 limits the band so that the pass band of the band limiting filter unit 9 becomes wider within a predetermined range than the channel bandwidth of the input IQ signal (the channel bandwidth limited by the LTE channel filter unit 1). The tap coefficient of the filter unit 9 is calculated. In order to realize this tap coefficient calculation processing, the distortion compensation circuit 12 is used.
The distortion compensation circuit 12 includes an FFT (Fast Fourier Transform) calculation unit 13 that observes a spectrum at the output of the transmission apparatus. The FFT operation unit 13 performs an FFT operation on a feedback IQ signal from a power amplifier (not shown), and calculates out-of-band power based on a peak suppression pulse signal component after the band limiting filter unit 9 observed on the frequency axis. The FFT calculation unit 13 supplies out-of-band power information including the calculation result of out-of-band power and carrier information to the tap coefficient calculation unit 8. The carrier information includes information indicating a channel bandwidth (LTE channel bandwidth) and can be set from the outside. Based on the carrier information and out-of-band power information, the tap coefficient calculation unit 8 expands the pass band of the band limiting filter unit 9 to a limit that can satisfy the operating band unnecessary emission standard that defines the output spectrum of the transmission device. The tap coefficient of the band limiting filter unit 9 is determined.

Note that the FFT calculation unit 13 does not need to be newly added, and the distortion detection circuit used in the distortion compensation circuit 12 in the existing transmission apparatus can be used as the FFT calculation unit 13. Thereby, an increase in circuit scale and cost increase can be suppressed.
The distortion compensation circuit 12 to which the output of the power amplifier is supplied as a feedback IQ signal is described in, for example, International Publication No. 2010/073483 (hereinafter referred to as a reference). The main configuration of the circuit described in the reference will be briefly described below.

The CFR output IQ signal that is the output of the PC-CFR circuit 3 is supplied to the power amplifier via the predistorter unit, the D / A converter, and the frequency conversion unit. The predistorter unit performs complex multiplication of the I and Q compensation signals, which are distortion compensation values, and the input IQ signal (output of the PC-CFR circuit 3), and applies the predistorted signal to the D / A converter. Supply. The D / A converter converts the signal from the predistorter unit from a digital signal to an analog signal. The frequency converter converts the analog signal from the D / A converter into a radio frequency (RF) signal. The power amplifier amplifies a radio frequency (RF) signal from the frequency converter and supplies the amplified signal to the antenna.
A directional coupler is provided at the output stage of the power amplifier, and the output of the power amplifier is taken out as a feedback signal (feedback IQ signal) by this directional coupler. A feedback signal from the directional coupler is supplied to the distortion compensation circuit 12.

The distortion compensation circuit 12 includes a frequency conversion unit, an A / D converter, an FFT unit, and a distortion band power calculation unit. The feedback signal is supplied to the FFT unit via the frequency conversion unit and the A / D converter. The frequency conversion unit converts the feedback signal from the directional coupler into an intermediate frequency (IF) band signal. The A / D converter converts the signal from the frequency conversion unit from an analog signal to a digital signal. The FFT unit performs an FFT process on the signal from the A / D converter. By this FFT processing, the signal from the A / D converter is converted from time axis data to frequency axis data. The distortion band power calculation unit calculates a value of distortion component power (distortion power) included in the feedback signal based on the frequency axis data from the FFT unit.
The FFT calculation unit 13 performs the FFT calculation of the feedback IQ signal on the part composed of the above FFT unit and the distortion band power calculation unit, and uses the peak suppression pulse signal component after the band limiting filter unit 9 observed on the frequency axis. It is configured to calculate out-of-band power.

Next, the operation of the PC-CFR circuit 3 of this embodiment will be described. In the following, the characteristic operation will be described with emphasis, and the description of the same operation as that of the existing PC-CFR circuit (comparative example) will be omitted.
In the following description, the PC-CFR circuit of the comparative example includes an amplitude calculation unit 4a, a threshold calculation unit 5a, delay adjustment units 6a and 10a, a pulse generation unit 7a, a band limiting filter unit 9a, and a subtraction unit 11a. The amplitude calculation unit 4a, threshold calculation unit 5a, delay adjustment units 6a and 10a, pulse generation unit 7a, band limiting filter unit 9a and subtraction unit 11a are respectively the amplitude calculation unit 4, threshold calculation unit 5 and delay adjustment unit described above. 6, 10, pulse generation unit 7, band limiting filter unit 9 and subtraction unit 11.

  According to the 3GPP regulations, for example, when the channel bandwidth reserved as a radio carrier signal band is 20 MHz, the transmission bandwidth (Transmission Bandwidth) of the effective signal band in which the resource block (Resource Block) exists is 18 MHz. However, the application of the operating band unnecessary emissions standard as the spectrum definition of the carrier signal requirement is for the outside of the band having a channel bandwidth of 20 MHz.

FIG. 5 shows each spectrum of the peak suppression pulse signal, the characteristics of the band limiting filter, and the peak suppression signal after the band limiting filter in the comparative example at 20 MHz of LTE.
For the above spectrum requirement, in the comparative example, the band limiting filter unit 9a for band limiting the pulse signal for peak suppression from the pulse generating unit 7a is substantially the same as the spectrum band of the LTE carrier signal by the LTE channel filter unit 1, Alternatively, tap coefficients having filter characteristics in a frequency band that falls within the range are set. This is because the peak suppression pulse signal is an impulse, and the frequency characteristic thereof is flat across the entire band as in the peak suppression pulse signal shown in FIG. For this reason, a peak suppression signal that is band-limited so as to be within the spectrum band of the LTE carrier signal by the LTE channel filter unit 1 is obtained by the filter characteristics of the band-limiting filter unit 9a.

FIG. 6 shows an example of each spectrum of the LTE carrier signal after the LTE channel filter unit 1 and the peak suppression signal after the band limiting filter unit 9a in the comparative example at 20 MHz of LTE, and an example of an operating band unnecessary emission standard. Indicates. Since the operating band unnecessary emission standard is the absolute power regulation, here, as an example, the power regulation is converted into the ratio with the carrier power under the condition that the output of the transmission apparatus is +37 dBm.
As shown in FIG. 6, the application of the spectrum-defined operating band unnecessary emission standard to the carrier signal is intended to be performed outside the band having a channel bandwidth of 20 MHz. On the other hand, the modulation accuracy (EVM) is determined by the characteristics within the transmission bandwidth 18 MHz of the effective signal band in which the resource block exists. Therefore, if the peak suppression signal within the transmission bandwidth can be reduced by expanding the band of the peak suppression signal within a range that can satisfy the operating band unnecessary emission standard applied outside the channel bandwidth, the above-described problems can be solved. it can.

  In the PC-CFR circuit 3 of the present embodiment, an FFT operation unit 13 that observes a spectrum at the output of the transmission device is used. The FFT operation unit 13 performs an FFT operation on the feedback IQ signal output from the power amplifier, and calculates the out-of-band power due to the peak suppression signal component after the band limiting filter unit 9 observed on the frequency axis. Based on the out-of-band power information from the FFT calculation unit 13 and the carrier information (LTE channel band) set from the outside, the band limiting filter unit in the tap coefficient calculation unit 8 can satisfy the operating band unnecessary emission standard. The tap coefficient is calculated so as to widen the 9 passband. The calculated tap coefficient is set in the band limiting filter unit 9.

In FIG. 7, the conceptual diagram for demonstrating the pass-band determination method of the band-limiting filter part 9 in the PC-CFR circuit 3 of this embodiment is shown. FIG. 8 shows the filter characteristics of the band limiting filter unit 9a in the PC-CFR circuit of the comparative example, the filter characteristics of the band limiting filter unit 9 of the PC-CFR circuit 3 of the present embodiment, and the pass band difference between these band limiting filters. The figure for demonstrating the difference of the filter gain accompanying with is shown. FIG. 9A shows tap coefficients of the band limiting filter of the PC-CFR circuit of the comparative example, and FIG. 9B shows tap coefficients of the band limiting filter unit 9 of the PC-CFR circuit 3 of the present embodiment.
A pass band determination method of the band limiting filter unit 9 of the PC-CFR circuit 3 of the present embodiment will be described with reference to FIGS. 7, 8, 9A, and 9B.

だけ低減できる。以下に、その理由を説明する。 The pass band of the band limiting filter unit 9a in the PC-CFR circuit of the comparative example is BW1, and the pass band of the band limiting filter unit 9 of the PC-CFR circuit 3 of the present embodiment is BW2. In the PC-CFR circuit 3 of the present embodiment, the pass band of the band limiting filter unit 9 is expanded by the ratio of BW2 / BW1. In this case, the center tap of the tap coefficient is set to 1.0 + j0.0 so that the output pulse amplitude of the band limiting filter unit 9 at the timing corresponding to the peak to be suppressed is equal to the difference between the peak amplitude and the threshold (R−Th). As shown in FIGS. 7 and 8, while maintaining the peak suppression amount due to the center tap amplitude, the filter gain for the input amplitude of the band limiting filter unit 9 determined by the sum of the tap coefficients is set.

Can only be reduced. The reason will be described below.

とし、図９Ｂに示したＢＷ２の通過帯域を有する帯域制限フィルタ部９のタップ係数（タップ数：２ｍ−１、センタータップ：Ｃ_２ｍ＝１．０）の総和を

とすると、あるピーク点での、比較例におけるピーク抑圧信号振幅総和に対する、本実施形態における通過帯域拡張に伴うピーク抑圧信号振幅総和の低減量［ｄＢ］は

である。分母と分子をピーク振幅と閾値の差（Ｒ−Ｔｈ）で除算すると、タップ係数の総和で決定されるフィルタゲインの低減量［ｄＢ］の

に等しい。ここで、センタータップにおける係数がＣ_１ｍ＝Ｃ_２ｍ＝１．０である場合は

である。 The peak suppression amount is a characteristic determined by (R−Th) × (1.0 + j0.0) based on the difference between the peak amplitude and the threshold and the center tap amplitude. On the other hand, the amount of modulation accuracy degradation in the PC-CFR circuit 3 is a characteristic caused by the RMS value of the error vector amplitude generated in the PC-CFR circuit 3, that is, the RMS value of the peak suppression signal amplitude.
The sum of the tap coefficients (number of taps: 2m−1, center tap: C ₁ m = 1.0) of the band limiting filter having the passband of BW1 shown in FIG.

And the sum of the tap coefficients (number of taps: 2m−1, center tap: C ₂ m = 1.0) of the band limiting filter unit 9 having the BW2 pass band shown in FIG. 9B.

Then, the reduction amount [dB] of the peak suppression signal amplitude sum accompanying the passband extension in this embodiment with respect to the peak suppression signal amplitude sum in the comparative example at a certain peak point is

It is. When the denominator and numerator are divided by the difference between the peak amplitude and the threshold (R-Th), the filter gain reduction amount [dB] determined by the sum of the tap coefficients is

be equivalent to. Here, when the coefficient at the center tap is C ₁ m = C ₂ m = 1.0

It is.

だけ低減できる。その結果、ピーク抑圧信号振幅のＲＭＳ値に起因する変調精度特性を改善することが可能となる。 Therefore, since the coefficient at the center tap remains the same condition of C ₁ m = C ₂ m = 1.0, the peak suppression signal amplitude summation is accompanied with the passband expansion while maintaining the peak suppression amount due to the center tap amplitude. The

Can only be reduced. As a result, it is possible to improve the modulation accuracy characteristic due to the RMS value of the peak suppression signal amplitude.

から約３．４％

に２ｄB改善させる必要がある。この場合、ΔＡＴＴ［ｄＢ］＝２ｄBより、本実施形態のＰＣ−ＣＦＲ回路３の帯域制限フィルタ部９の通過帯域ＢＷ２を、比較例の通過帯域ＢＷ１の１．２６倍

以上に拡張すればよい。通過帯域の拡張可能範囲は、タップ係数演算部８において、ＦＦＴ演算部１３からのキャリア情報（ＬＴＥチャネル帯域）と帯域外電力情報とに基づき、オペレーティングバンド不要発射規格を満足できるかを判断することで決定する。さらに、タップ係数演算部８において、決定した帯域制限フィルタ部９の通過帯域に応じたタップ係数を演算し、その結果を帯域制限フィルタ部９に設定する。 For example, assuming that the modulation accuracy of an error vector caused by phase noise is 1.0%, when the modulation accuracy of the transmission apparatus is improved from 4.5% to 3.5%, the modulation accuracy by the peak suppression circuit is about 4%. .3%

3.4%

It is necessary to improve 2 dB. In this case, since ΔATT [dB] = 2 dB, the passband BW2 of the band limiting filter unit 9 of the PC-CFR circuit 3 of the present embodiment is 1.26 times the passband BW1 of the comparative example.

What is necessary is just to extend above. The passband expandable range is determined by the tap coefficient calculation unit 8 based on carrier information (LTE channel band) and out-of-band power information from the FFT calculation unit 13 to determine whether the operating band unnecessary emission standard can be satisfied. To decide. Further, the tap coefficient calculation unit 8 calculates a tap coefficient corresponding to the determined pass band of the band limiting filter unit 9 and sets the result in the band limiting filter unit 9.

As described above, based on the carrier information and the out-of-band power information from the FFT calculation unit 13, the tap coefficient calculation unit 8 determines the pass band of the band limiting filter unit 9 within a range that can satisfy the operating band unnecessary emission standard. By calculating a tap coefficient corresponding to the expanded pass band and setting it in the band limiting filter unit 9, the peak suppression within the transmission bandwidth that affects the modulation accuracy while maintaining the peak suppression amount due to the center tap amplitude is maintained. It becomes possible to reduce the RMS value of the signal amplitude.
As for the adjustment method of the pass band of the band limiting filter unit 9 in the tap coefficient calculation unit 8, the pass band is rewritten to a tap coefficient stored in advance in a non-volatile memory or the pass band is changed with respect to a basic filter characteristic. What is necessary is just to determine a realization means by the necessity of an adaptive process, such as calculating | requiring the characteristic to adjust by a convolution calculation.

The PC-CFR circuit 3 of the present embodiment has the following effects.
In order to improve the modulation accuracy due to the introduction of 256QAM, with the existing technology, it is necessary to reduce the error vector generated in the peak suppression circuit, that is, the amplitude RMS value of the peak suppression signal, to alleviate the peak suppression.

  FIG. 10 shows each spectrum of the peak suppression pulse signal at 20 MHz of LTE, the characteristics of the pulse signal band limiting filter unit 9, and the peak suppression signal after the band limiting filter unit 9. FIG. 11 shows each spectrum of the LTE carrier signal after the LTE channel filter unit 1 and the peak suppression signal after the band limiting filter unit 9 at the time of LTE 20 MHz, and the operating band unnecessary emission standard. The operating band unnecessary emission standard is the same as the condition shown in FIG. In the example shown in FIGS. 10 and 11, the out-of-band power due to the peak suppression signal component after the band limiting filter unit 9 is actually calculated for the pass band BW2 of the band limiting filter unit 9 within a range that can satisfy the operating band unnecessary emission standard. The results are based on the conditions shown in FIGS.

Compared to the comparative example shown in FIGS. 5 and 6, according to the example shown in FIGS. 10 and 11, the bandwidth limitation within the 18 MHz transmission bandwidth that affects the modulation accuracy due to the filter gain reduction accompanying the bandwidth extension. The level of the peak suppression signal spectrum after the filter unit 9 can be reduced.
As described above, according to the peak suppression circuit of this embodiment, the RMS value of the peak suppression signal amplitude within the transmission bandwidth that affects the modulation accuracy can be reduced while maintaining the peak suppression amount due to the center tap amplitude. Thus, it is possible to improve the modulation accuracy characteristic and to realize a peak factor reduction equivalent to that of the existing one.

The above-described embodiment is an example of the present invention, and the present invention is not limited to the configuration. For example, an improvement using an equivalent circuit or an approximate method with a different configuration may be applied.

In the configuration shown in FIG. 1, the tap coefficient calculation unit 8 and the FFT calculation unit 13 may be configured by software or hardware. Similarly, the portions other than the tap coefficient calculation unit 8 and the FFT calculation unit 13 may be configured by software or hardware.

(Other embodiments)
FIG. 12 is a block diagram showing a configuration of a peak suppression circuit according to another embodiment of the present invention.
Referring to FIG. 12, the peak suppression circuit includes a pulse generation unit 100, a band limiting filter unit 101, a tap coefficient calculation unit 102, and a subtraction unit 103.

The pulse generation unit 100 generates a first pulse signal by multiplying the value obtained by subtracting a predetermined threshold from the amplitude value of the input signal whose spectrum is band-limited by the phase information of the input signal. The band limiting filter unit 101 has a plurality of tap coefficients, limits the frequency band of the first pulse signal according to the filter characteristics determined by the tap coefficient set based on external carrier information, and is band-limited. A second pulse signal that is a pulse signal is output. The subtracting unit 103 subtracts the second pulse signal from the input signal.
The tap coefficient calculation unit 102 determines the plurality of tap coefficients so that the pass band of the band limiting filter unit 101 is wider within a predetermined range than the spectrum bandwidth of the input signal.

In another embodiment, the peak suppression circuit uses the second pulse signal component observed on the frequency axis after the transmission signal is input as a feedback signal and the feedback signal is subjected to FFT (Fast Fourier Transform) processing. You may have further the FFT calculating part which calculates out-of-band electric power. In this case, the tap coefficient calculation unit 102 determines the tap coefficient of the band limiting filter unit 101 based on the out-of-band power value calculated by the FFT calculation unit and the carrier information.
The predetermined range is a range that satisfies a spectrum-defined operating band unnecessary emission standard for the transmission signal.

  The pulse generation unit 100 may correspond to, for example, a configuration including the amplitude calculation unit 4, the threshold calculation unit 5, the delay adjustment units 6, 10 and the pulse generation unit 7 illustrated in FIG. The band limiting filter unit 101, the tap coefficient calculating unit 102, and the subtracting unit 103 may correspond to the band limiting filter unit 9, the tap coefficient calculating unit 8, and the subtracting unit 11 shown in FIG. The FFT operation unit may be the FFT operation unit 13 illustrated in FIG.

DESCRIPTION OF SYMBOLS 1 LTE channel filter part 2 Digital up-converter part 3 PC-CFR circuit 4 Amplitude calculating part 5 Threshold value subtracting part 6, 10 Delay adjustment part 7, 100 Pulse generation part 8, 102 Tap coefficient calculating part 9, 101 Band-limiting filter part 11 103 Subtractor 12 Distortion compensation circuit 13 FFT calculator

Claims (4)

A pulse generator for generating a first pulse signal by multiplying a value obtained by subtracting a predetermined threshold value from an amplitude value of an input signal whose spectrum is band-limited, and phase information of the input signal;
The first pulse signal is a band-limited pulse signal having a plurality of tap coefficients, limiting the frequency band of the first pulse signal according to a filter characteristic determined by the tap coefficient set based on carrier information from the outside, A band limiting filter unit that outputs a pulse signal of 2;
A subtractor for subtracting the second pulse signal from the input signal;
A peak suppression circuit, comprising: a tap coefficient calculation unit that determines the plurality of tap coefficients so that a pass band of the band limiting filter unit is wider than a spectrum bandwidth of the input signal within a predetermined range. An FFT operation unit that receives the transmission signal as a feedback signal, performs FFT (Fast Fourier Transform) on the feedback signal, and calculates out-of-band power based on the component of the second pulse signal observed on the frequency axis; Have
The peak suppression circuit according to claim 1, wherein the tap coefficient calculation unit determines the plurality of tap coefficients based on the out-of-band power value calculated by the FFT calculation unit and the carrier information.   The peak suppression circuit according to claim 1, wherein the predetermined range is a range that satisfies a spectrum-defined operating band unnecessary emission standard for the transmission signal. A value obtained by subtracting a predetermined threshold from the amplitude value of the input signal whose spectrum is band-limited is multiplied by the phase information of the input signal to generate a first pulse signal,
Using a band limiting filter that has a plurality of tap coefficients and whose filter characteristics are determined by the tap coefficients set based on carrier information from the outside, limits the frequency band of the first pulse signal,
A peak suppression method that subtracts a second pulse signal that is a pulse signal band-limited by the band-limiting filter from the input signal,
The peak suppression method, wherein the plurality of tap coefficients are set such that a pass band of the band limiting filter unit is wider than a spectrum bandwidth of the input signal within a predetermined range.

Images