JP2012165261A - Power amplifier circuit and communication device including this circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to implement a CFR circuit whose circuit scale is relatively small in an EER-type power amplifier circuit 5.SOLUTION: The present invention relates to an EER-type power amplifier circuit 5. The power amplifier circuit 5 comprises: a calculation unit 12 for calculating a scalar quantity (for example, instantaneous power P=I+Q) corresponding to the amplitude component of an IQ baseband signal; a phase modulation unit 13 for modulating a phase component of the IQ baseband signal to a high frequency by extracting the phase component; a CFR processing unit 14 for limiting the upper limit of the scalar quantity to a value corresponding to a predetermined threshold value; and a power amplifier 16 which uses a phase modulation signal output by the phase modulation unit 13 as an input signal and operates using an amplitude modulation signal corresponding to the scalar quantity with limited the upper limit as a power source voltage.

Description

  The present invention relates to an envelope removal / restoration (EER) type power amplification circuit to which a CFR (Crest Factor Reduction) function is added, and a communication apparatus having this circuit.

For example, in a method of modulating a transmission signal using a plurality of carriers, such as an OFDM (Orthogonal Frequency Division Multiplex) method and a W-CDMA (Wideband Code Division Multiple Access) method, the phases of the carriers overlap. May have a large peak power.
On the other hand, the power amplifier (power amplifier) is required to have excellent linearity. However, when a signal having a level exceeding the maximum output is input, the output is saturated and nonlinear distortion increases.

For this reason, when a signal having a large peak power is input to the nonlinear amplifier, nonlinear distortion occurs in the output signal, which causes deterioration of reception characteristics and out-of-band radiation on the reception side.
In order not to increase nonlinear distortion with respect to peak power, a power amplifier with a wide dynamic range is required, but if the dynamic range is widened due to peak power that does not appear frequently, the average power of the waveform on the time axis And the peak power for a short time (PAPR: Peak to Average Power Ratio) are increased, and the power efficiency is deteriorated.

Therefore, it is more reasonable to suppress a signal having a large peak power with a low appearance frequency before being input to the amplifier as it is. Therefore, in order to suppress the peak power of the IQ baseband signal before power amplification, there is a case where a clipping process is performed in which an IQ baseband signal having a peak power exceeding a predetermined threshold is instantaneously given an opposite amplitude.
Since this process is a process of applying an impulse signal in the reverse direction on the time axis, it is the same as applying noise in a wide frequency band on the frequency axis. For this reason, there is a problem that noise is generated outside the band when only clipping is performed.

Therefore, in order to deal with the problem of out-of-band radiation, peak power suppression circuits called NS-CFR (Noise Shaping-Crest Factor Reduction) and PC-CFR (Peak Cancellation-Crest Factor Reduction) are known.
Among these, NS-CFR performs band limitation by filtering the peak component of the IQ baseband signal whose instantaneous power exceeds the threshold by using a low-pass filter, FIR (Finite Impulse Response) filter, and the like. The peak component is subtracted from the original IQ baseband signal (see Patent Document 1).

  The PC-CFR has a basic function that does not cause out-of-band radiation even when clipping, and is obtained by multiplying the peak component of an IQ baseband signal whose instantaneous power exceeds a threshold by the basic function. Subtract from the original IQ baseband signal (see Patent Documents 2 and 3).

  On the other hand, by changing the drain voltage of the power amplifier according to the power on the input side, the power loss that occurs during operation with a fixed voltage is reduced to improve the efficiency of the high-frequency amplifier. Elimination and Restoration (hereinafter referred to as “EER”) type power amplification circuit and envelope tracking (hereinafter referred to as “ET”) type power amplification circuit are known (see Patent Document 4). .

Among these, in the EER system, an input modulation signal is separated into a phase component and an amplitude component (polar modulation), and the phase component is input to the power amplifier with a constant amplitude to operate near saturation at which efficiency is always maximized. On the other hand, the amplitude component is supplied to the power amplifier as power (modulated power supply) that has been power amplified and intensity-modulated with high efficiency.
By operating in this way, the power amplifier operates as a multiplier to combine the phase component and the amplitude component, and an output modulation signal amplified with high efficiency without backoff is obtained.

The ET method is the same as the EER method in that the amplitude component of the input modulation signal is amplified with high efficiency and supplied to the power amplifier as intensity-modulated power.
The difference between the two is that in the EER system, the power amplifier is saturated by inputting only a phase-modulated signal having a constant amplitude to the power amplifier, whereas in the ET system, the input includes both the amplitude component and the phase component. The modulation signal is input to the power amplifier as it is, and the power amplifier is operated linearly.

In the ET system, although the power amplifier operates linearly, the efficiency is lower than that of the EER system that performs saturation operation. However, since the minimum necessary power corresponding to the amplitude component is supplied, the power amplifier is independent of the amplitude of the input modulation signal. Higher efficiency than when using a fixed voltage.
The ET method also has an advantage that it is easier to implement than the EER method because the time lag for combining and restoring the amplitude component and the phase component is reduced.

Japanese Patent No. 3954341 Japanese Patent No. 3853509 Japanese Patent Laid-Open No. 2004-135087 (FIGS. 1 to 6) JP 2008-193633 A

If the large peak power generated in the IQ baseband signal is suppressed by the CFR circuit, and the input modulation signal generated from the IQ baseband signal after the suppression is input to the power amplification circuit of the EER method or the ET method, Power efficiency can be increased synergistically.
However, in this case, the circuit configuration is such that opposite amplitudes are given to the I component and the Q component of the IQ baseband signal. Therefore, the components of the CFR circuit such as a multiplier, an adder, and a delay circuit are the I component and the Q component. There is a problem that the circuit scale becomes large because it is necessary for each component.

  In view of such conventional problems, the present invention enables a CFR circuit having a relatively small circuit scale to be mounted on an EER type power amplifier circuit, thereby reducing the power amplifier circuit that can effectively increase the efficiency of the power amplifier at low cost. It aims to be realized.

  (1) A power amplifier circuit according to the present invention is an EER-type power amplifier circuit, and calculates a scalar amount corresponding to an amplitude component of an IQ baseband signal, and extracts a phase component of the IQ baseband signal. And the phase modulation unit that modulates to a high frequency, the CFR processing unit that limits the upper limit of the scalar amount to a predetermined threshold, and the phase modulation signal output by the phase modulation unit as an input signal, and the upper limit is limited And a power amplifier that operates using an amplitude modulation signal corresponding to the scalar quantity as a power supply voltage.

  According to the power amplifier circuit of the present invention, the CFR processing unit limits the upper limit of the scalar amount corresponding to the amplitude component of the IQ baseband signal to a predetermined threshold value, and the power amplifier outputs the phase modulation output from the phase modulation unit. Since the signal operates as an input signal and the amplitude modulation signal corresponding to the scalar amount whose upper limit is limited as the power supply voltage, the power supply voltage of the power amplifier does not fluctuate greatly with the peak fluctuation of the IQ baseband signal. The power efficiency of the power amplifier can be effectively increased.

In addition, according to the power amplifier circuit of the present invention, the CFR processing unit that limits the upper limit of the scalar quantity corresponding to the amplitude component of the IQ baseband signal to a predetermined threshold value is adopted, so that the inverse of one scalar quantity is obtained. A circuit configuration that provides a direction canceling signal is sufficient.
Therefore, compared to the conventional CFR circuit that limits the peak power by subtracting each of the I component and the Q component of the IQ baseband signal, there are fewer components such as a filter, a multiplier, an adder / subtractor, and a delay circuit, A CFR circuit having a small circuit scale can be employed.

(2) In the power amplifier circuit of the present invention, the scalar amount calculated by the calculation unit is not particularly limited as long as it corresponds to the amplitude component of the IQ baseband signal. For example, the instantaneous amount of the IQ baseband signal Electric power (P = I ² + Q ² ) can be employed.
(3) In the power amplifier circuit of the present invention, the CFR processing unit performs band limitation by filtering the peak component of the scalar amount exceeding a threshold value, and the peak component after the band limitation is converted into the original peak component. An NS-CFR circuit that subtracts from the scalar quantity can be used.

  (4) Further, in the power amplifier circuit of the present invention, the CFR processing unit holds in advance a basic function for preventing out-of-band radiation due to clipping, and the basic component is added to the peak component of the scalar quantity exceeding a threshold value. A PC-CFR circuit that subtracts the product multiplied by the function from the original scalar quantity can also be used.

(5) When the PC-CFR circuit is employed, the basic function is preferably composed of canceling pulses having frequency components both inside and outside the frequency band of the IQ baseband signal.
The reason is that a canceling pulse that also has an out-of-band frequency component can be a sharp pulse-like signal that changes in a shorter time than a canceling pulse that has only a frequency component in the band. is there. In this case, it is possible to prevent a new peak waveform from being generated by the CFR process, and it is possible to more reliably suppress the scalar peak.

(6) When the PC-CFR circuit is employed, it is preferable that the basic function is a canceling pulse that can cancel the IQ baseband signal corresponding to the average power for each frequency.
The reason for this is that if such canceling pulses are employed, the power in each frequency band is prevented from unnecessarily fluctuating due to CFR processing, and even in the case of an IQ baseband signal having a different average power for each frequency band, the SNR This is because the amplitude of the scalar quantity can be appropriately limited without deteriorating the value.

(7) The power amplifier circuit according to the present invention further includes a threshold update unit that updates the threshold used in the CFR processing unit for each predetermined control period in which the average power of the IQ baseband signal may fluctuate. Preferably it is.
In this case, since the threshold update unit updates the threshold used in the CFR processing unit for each predetermined control cycle, even if the average power of the IQ baseband signal fluctuates depending on the time zone, the amplitude limitation of the scalar amount in the CFR processing unit Can be executed reliably.

Therefore, even when the average power of the IQ baseband signal fluctuates greatly every predetermined control period, the PAPR of the amplitude modulation signal input to the power amplifier circuit can be reduced, and the power efficiency of the power amplifier can be reliably improved. it can.
(8) The communication device of the present invention includes a transmitter on which the power amplifier circuit of the present invention (the power amplifier circuit described in any one of (1) to (7) above) is mounted. The same effects as the power amplifier circuit of the present invention are achieved.

  As described above, according to the present invention, since a CFR circuit having a relatively small circuit scale can be mounted on an EER power amplifier circuit, a power amplifier circuit that can effectively increase the efficiency of the power amplifier can be realized at low cost. it can.

1 is an overall configuration diagram of a wireless communication system according to a first embodiment. It is a functional block diagram which shows the principal part of the OFDM transmitter of a base station apparatus. It is a functional block diagram of the power amplifier circuit which concerns on 1st Embodiment. It is a functional block diagram of the CFR processing unit (NS-CFR circuit) according to the first embodiment. It is a coordinate diagram on the IQ plane showing the relationship between IQ baseband signals and thresholds. It is a graph which shows an example of the time change of the threshold value updated sequentially with the instantaneous electric power of IQ baseband signal. FIG. 2 is a frame configuration diagram of an LTE downlink frame. It is a whole block diagram which shows the modification of a radio | wireless communications system. It is a functional block diagram of the CFR processing part (PC-CFR circuit) concerning a 2nd embodiment. It is a wave form diagram which shows the production | generation method of the pulse for cancellation. It is a whole block diagram of the radio | wireless communications system which concerns on 3rd Embodiment. It is a functional block diagram which shows the principal part of the OFDM transmitter of a base station apparatus. It is a functional block diagram of a CFR processing unit (PS-CFR circuit) according to a third embodiment. It is a functional block diagram of a pulse generation part. It is a time-domain graph which shows the variation of a basic pulse.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[First Embodiment]
[Wireless communication system]
FIG. 1 is an overall configuration diagram of a radio communication system according to the first embodiment to which the present invention can be preferably applied.
As shown in FIG. 1, a wireless communication system according to the present embodiment includes a base station device (BS: Base Station) 1 and a plurality of mobile terminals (MSs) that perform wireless communication with the device 1 in the cell of the device 1. Mobile Station) 2.

  In this radio communication system, an OFDM scheme is adopted as a modulation scheme between the base station apparatus 1 and the mobile terminal 2. This method is a multi-carrier digital modulation method in which transmission data is carried on a large number of carrier waves (subcarriers). Since the subcarriers are orthogonal to each other, the data can be arranged so densely as to overlap on the frequency axis. There are advantages.

The wireless communication system of the present embodiment is a mobile phone system to which the LTE (Long Term Evolution) system is applied, and communication based on the LTE system is performed between each base station apparatus 1 and the mobile terminal 2. Done.
But the radio | wireless communications system which can apply this invention is not restricted to LTE, You may employ | adopt W-CDMA.

[LTE downlink frame]
FIG. 7 is a diagram illustrating a structure of an LTE downlink frame. In the figure, the vertical axis direction represents frequency, and the horizontal axis direction represents time.
As shown in FIG. 7, a total of ten subframes (subframe # 0 to # 9) constituting a downlink (DL) frame are each composed of two slots (slot # 0 and slot # 1). One slot is composed of 7 OFDM symbols (in the case of Normal Cyclic Prefix).

Also, in the figure, a resource block (RB: Resource Block) which is a basic unit in data transmission is defined as 12 subcarriers in the frequency axis direction and 7 OFDM symbols (1 slot) in the time axis direction.
Therefore, for example, when the frequency bandwidth of the DL frame is set to 5 MHz, since 300 subcarriers are arranged, 25 resource blocks are arranged in the frequency axis direction.

  Note that the transmission time of one subframe is 1 ms, and in this embodiment, since two slots constituting one subframe each include seven OFDM symbols, the transmission period (symbol period) of one OFDM symbol. ) Is 1/14 ms (= about 0.071 ms).

As shown in FIG. 7, a control channel for transmitting information necessary for downlink communication from the base station apparatus 1 to the mobile terminal 2 is assigned to the head of each subframe.
In this control channel, DL control information, resource allocation information of the subframe, reception success notification (ACK: Acknowledgement) by hybrid automatic retransmission request (HARQ: Hybrid Automatic Report Request), reception failure notification (NACK: Negative Acknowledgement) Etc. are stored.

  In the DL frame shown in FIG. 7, PBCH (Physical Broadcast CHannel) is a broadcast channel for notifying the terminal device of the system bandwidth and the like by broadcast transmission, and is 0th (# 0) and 6th (# 5). ) Includes a first synchronization signal (P-SCH: Primary Synchronization CHannel) and a second synchronization signal (S-SCH: Secondary Synchronization CHannel), which are signals for identifying the base station apparatus 1 and the cell. Assigned.

Also, resource blocks in other areas to which the above channels are not assigned (areas without hatching in FIG. 7) are DL shared communication channels (PDSCHs) for storing user data and the like. Used.
The allocation of user data stored in the PDSCH is defined by the resource allocation information in the control channel allocated at the head of each subframe, and the mobile terminal 2 uses the resource allocation information to It can be determined whether data is stored in a subframe.

[Configuration of transmitter]
FIG. 2 is a functional block diagram illustrating a main part of the OFDM transmitter 3 of the base station apparatus 1.
The transmitter 3 includes a transmission processor 4 and a power amplification circuit 5. The transmission processor 4 is configured by, for example, an FPGA (Field Programmable Gate Array) having one or a plurality of memories and a CPU therein.

The FPGA can set (configure) configuration information for various logic circuits in advance at the time of shipment of the processor, manufacture of the base station apparatus 1, and the like. By performing such setting work, the functional units 6 to 8 shown in FIG. 2 are configured.
That is, the transmission processor 4 of this embodiment includes an S / P converter 6, a mapping unit 7, and an IFFT (Inverse Fast Fourier Transform) unit 8 in order from the left.

The serial signal sequence input to the transmission processor 4 is converted into a plurality of signal sequences by an S / P (serial parallel) converter 6. Each converted parallel signal sequence is converted by the mapping unit 7 into a plurality of subcarrier signals f1, f2,.
Each of the subcarrier signals f1, f2,... Fn is converted into an I signal and a Q signal as baseband signals orthogonal to each other on the time axis by the IFFT unit 8. The IQ signal (Iin, Qin) is input to the power amplifier circuit 5 at the subsequent stage.

[Configuration of power amplifier circuit]
FIG. 3 is a functional block diagram of the power amplifier circuit 5 according to the first embodiment.
The power amplifying circuit 5 of the present embodiment employs the EER method, and extracts the amplitude information (envelope) from the IQ signal on the input side and applies it as the power supply voltage of the power amplifier 16, thereby The amplifier 16 is operated in a state almost close to saturation.

As shown in FIG. 3, the power amplification circuit 5 specifically includes a polar modulation unit 11, a CFR processing unit 14, a drain voltage modulation unit 15, and a power amplifier 16.
Among these, the polar modulation unit 11 separates the IQ baseband signal in the orthogonal coordinate format into an amplitude component and a phase component, and is composed of an IQ baseband signal power calculation unit 12 and a phase modulation unit 13. .

The power calculation unit 12 calculates an instantaneous power P (= I ² + Q ² ) of an IQ baseband signal that is a sum of squares of the I signal and the Q signal (Iin, Qin) on the input side. The calculated instantaneous power P is input to the subsequent CFR processing unit 14.
The phase modulation unit 13 includes an extraction circuit that extracts a phase component of the IQ baseband signal, a modulator that modulates the extracted phase component to a high frequency of RF (Radio Frequency), a limiter that makes the amplitude of the modulation signal constant, and the like. Thus, the phase modulation signal output from the phase modulation unit 13 is input to the power amplifier 16 through the RF line.

The CFR processing unit 14 includes a CFR circuit that limits the upper limit so that the instantaneous power P input from the power calculation unit 12 does not exceed a predetermined threshold value Pth, the details of which will be described later.
The drain voltage modulation unit 15 includes an amplifier (for example, a class D amplifier) that amplifies the instantaneous power P of the IQ base burst signal whose upper limit is limited by the CFR processing unit 14, and a peak hold circuit that holds the maximum value of the amplified signal. Thus, an amplitude modulation signal corresponding to the instantaneous power P is output.

The power amplifier 16 is composed of, for example, a FET (Field Effect Transistor) type transistor.
The power amplifier 16 always operates in the vicinity of saturation with the phase modulation signal from the phase modulation unit 13 as an input signal to the gate terminal and the amplitude modulation signal from the drain voltage modulation unit 15 as the power supply voltage to the drain terminal. As a result, the power amplifier 16 operates as a multiplier, and the phase modulation signal and the amplitude modulation signal are combined to obtain an output modulation signal amplified with high efficiency.

[Configuration of CFR processing unit]
FIG. 4 is a functional block diagram of the CFR processing unit (NS-CFR circuit) 14 according to the first embodiment.
As shown in FIG. 4, the CFR processing unit 14 includes a correction signal calculation unit 20, a filter 21, a scaling unit 22 including a multiplier, an adder / subtractor 23, a delay unit 24, an average calculation unit 25, and a threshold update unit 26. .

The correction signal calculation unit 20 compares the instantaneous power P input from the power calculation unit 12 with the threshold value Pth held at that time, and when the instantaneous power P is larger than the threshold value Pth, the correction ΔP is calculated. Output. Further, the correction signal calculation unit 20 outputs zero when the instantaneous power P is equal to or less than the threshold value Pth.
The difference ΔP output from the correction signal calculation unit 20 is band-limited (noise shaping) by a filter 21 made of a subsequent low-pass filter, FIR filter, or the like, and is further amplitude-adjusted by a subsequent scaling unit 22 to add / subtractor 23. Is input.

The delay unit 24 delays the signal output timing of the instantaneous power P by the time of calculation processing in the correction signal calculation unit 20 and other calculation units.
Then, the adder / subtracter 23 outputs the instantaneous power output signal Pout by subtracting the band-limited correction signal ΔP from the delayed instantaneous power P. By this subtraction, the instantaneous power P exceeding the threshold value Pth is corrected to the instantaneous power P corresponding to the threshold value Pth. Further, the instantaneous power P that is equal to or less than the threshold value Pth is output as it is without being corrected.

FIG. 5 is a coordinate diagram on the IQ plane showing the relationship between the IQ baseband signal and the threshold value Pth when the above signal processing is performed.
As shown in FIG. 5, the signal processing by the CFR processing unit 14 of the present embodiment is a clipping process for cutting the outer peripheral side of the instantaneous power P of the IQ baseband signal. For this reason, since the PAPR of the drain voltage applied to the power amplifier 16 of the power amplifier circuit 5 is reduced, the power efficiency of the power amplifier 16 can be improved.

[Effect of power amplifier circuit]
According to the power amplifier circuit 5 of the present embodiment, the CFR processing unit 14 sets the upper limit of the instantaneous power P (= I ² + Q ² ), which is a scalar amount corresponding to the amplitude component of the IQ baseband signal, corresponding to the predetermined threshold value Pth. The power amplifier 16 operates using the phase modulation signal from the phase modulation unit 13 as an input signal and the amplitude modulation signal corresponding to the instantaneous power P whose upper limit is limited as a power supply voltage.
Therefore, even if there is a large peak fluctuation in the IQ baseband signal, the power supply voltage of the power amplifier 16 does not fluctuate greatly along with it, and the power efficiency of the power amplifier 16 can be effectively increased.

Further, according to the power amplifier circuit 5 of the present embodiment, as shown in FIG. 3, the CFR processing unit 14 is arranged between the power calculation unit 12 and the drain voltage modulation unit 15, and the instantaneous power P of the IQ baseband signal is set. The circuit arrangement is such that the instantaneous power Pout after the restriction is supplied to the drain voltage modulation unit 15 after the upper limit of is restricted to a predetermined threshold value Pth.
Therefore, as shown in FIG. 4, it is sufficient to mount only one filter 21, multiplier 22, adder / subtractor 23, and delay circuit 24 as a circuit configuration for giving a reverse cancellation signal to the instantaneous power P.

  Therefore, compared to a CFR circuit that limits the peak power by subtracting each of the I component and Q component of the IQ baseband signal, it is possible to use a CFR circuit that requires fewer components and has a smaller circuit scale. Thus, the power amplifier circuit 5 that can effectively increase the efficiency of the power amplifier 16 can be realized at low cost.

[About the threshold update unit]
Returning to FIG. 4, the average calculation unit 25 acquires the symbol period of the OFDM symbol, which is the minimum time unit in which the transmission power can vary greatly, as a control period for calculating the average power.
The average calculation unit 25 acquires the instantaneous power P of the IQ baseband signal from the power calculation unit 12, and averages the instantaneous power P within the symbol period to thereby obtain an IQ baseband for each symbol period. The average power Pave of the signal is calculated and output to the threshold update unit 26.

The threshold update unit 26 employs a value obtained by multiplying the average power Pave for each symbol period acquired from the average calculation unit 25 by a predetermined magnification as the threshold Pth in the symbol period. For example, when the ratio of the peak power Ppeak and the average power Pave of the IQ baseband signal is reduced to 6 dB, the predetermined magnification is doubled.
The threshold update unit 26 calculates the threshold value Pth for each symbol period as described above, dynamically updates the threshold value Pth, and outputs the updated threshold value Pth to the correction signal calculation unit 20.

Then, the correction signal calculation unit 20 determines the magnitude of the instantaneous power P calculated by the power calculation unit 12 using the threshold value Pth acquired from the threshold update unit 26, and the instantaneous power P exceeds the updated threshold value Pth. It is determined whether or not.
FIG. 6 is a graph showing an example of temporal changes in the instantaneous power P of the IQ baseband signal and the threshold value Pth that is sequentially updated.

As shown in FIG. 6, in the present embodiment, the threshold value Pth used in the CFR processing unit 14 is sequentially calculated based on the average power Pave calculated for each symbol period (1/14 ms) and updated for each symbol period. The
For this reason, for example, the clipping process is appropriately performed even if the average power Pave of the IQ baseband signal fluctuates in response to fluctuations in the call volume by the mobile terminal 2, so that the power amplifier 16 by reducing the PAPR. It is possible to effectively ensure the improvement in power efficiency.

Further, according to the present embodiment, since the OFDM symbol period, which is the minimum time unit in which the transmission power can be varied, is employed as the control period for updating the threshold value Pth, the threshold value Pth can be updated accurately and quickly. There is also an advantage.
However, in LTE, the resource block (see FIG. 7) is the minimum unit for user allocation, and therefore 7 OFDM symbols (1 slot), which is the transmission period of this resource block, is adopted as the control period for updating the threshold value Pth. You may decide.

[Modification of wireless communication system]
FIG. 8 is an overall configuration diagram showing a modification of the radio communication system.
As shown in FIG. 8, in the radio communication system of this modification, an RRH 27 is connected to the base station apparatus 1 via a CPRI (Common Public Radio Interface). The base station apparatus 1 transmits a synchronization signal 28 for establishing synchronization with the RRH 27 to the RRH 27 through an optical fiber. The synchronization signal 28 is a clock signal having a 1 ms period synchronized with the OFDM symbol period.

The transmitter 3 having the power amplifier circuit 5 according to the first embodiment or each of the embodiments described later may be mounted on the RRH 27.
In this case, a symbol period is generated from the synchronization signal 28 acquired from the base station apparatus 1 which is an external apparatus, and the generated symbol period is output to the average calculation unit 25 (see FIG. 4), whereby the threshold value Pth for each symbol period. Can be obtained.

[Second Embodiment]
In the first embodiment, the CFR processing unit 14 of the power amplifier circuit 5 is configured by an NS-CFR circuit. However, the CFR processing unit 14 may be configured by a PC-CFR circuit.
FIG. 9 is a functional block diagram of the CFR processing unit (PC-CFR circuit) 14 according to the second embodiment. As shown in FIG. 9, the CFR processing unit 14 includes a correction signal calculation unit 30, a comparison unit 31, a pulse holding unit 32, a multiplier 33, an adder / subtractor 34, a delay unit 35, an average calculation unit 36, and a threshold update unit 37. including.

The correction signal calculation unit 30 compares the instantaneous power P input from the power calculation unit 12 with the threshold value Pth held at that time, and when the instantaneous power P is larger than the threshold value Pth, the correction ΔP is calculated. Output to the multiplier 33.
The comparison unit 31 compares the instantaneous power P calculated by the power calculation unit 12 with the threshold value Pth, and issues an output command of the cancellation pulse S to the pulse holding unit 32 when the instantaneous power P is larger than the threshold value Pth. .

The pulse holding unit 32 has a memory made up of a dual port RAM or the like that temporarily holds a canceling pulse S (see FIG. 10) composed of a synthetic pulse described later, and receives a command from the comparison unit 31 The held cancellation pulse S is output to the multiplier 33.
Further, the pulse holding unit 32 outputs zero to the multiplier 33 when no command is received from the comparison unit 31. Therefore, for the instantaneous power P exceeding the threshold value Pth, the canceling signal Pc calculated as Pc = ΔP × S is input to the adder / subtractor 34.

The delay unit 35 delays the signal time of the instantaneous power P by the time of calculation processing in the correction signal calculation unit 30 and other calculation units.
The adder / subtractor 34 subtracts the canceling signal Pc from the delayed instantaneous power P to output an output signal Pout of the instantaneous power. By this subtraction, the instantaneous power P exceeding the threshold value Pth is corrected to the instantaneous power P corresponding to the threshold value Pth. Further, the instantaneous power P that is equal to or less than the threshold value Pth is output as it is without being corrected.

  As shown in FIG. 9, the CFR processing unit 14 of the second embodiment is also provided with an average calculation unit 36 and a threshold update unit 37. These are the CFRs of the first embodiment (FIG. 4). This is a functional unit that performs the same processing as the average calculating unit 25 and the threshold updating unit 26 provided in the processing unit 14.

[About canceling pulse]
FIG. 10 is a waveform diagram showing a generation method of the cancellation pulse S.
As shown in FIG. 10, the cancellation pulse S of the present embodiment is composed of a synthesized pulse obtained by synthesizing a basic pulse Sa and an auxiliary pulse Sb.
In FIG. 10, the time waveform and frequency spectrum of the basic pulse Sa are shown in the upper left frame, and the time waveform and frequency spectrum of the auxiliary pulse Sb are shown in the lower left frame. The time waveform and frequency spectrum of the canceling pulse S are shown in the right frame.

  As in the case of Patent Document 3 (Japanese Patent Laid-Open No. 2004-135078), the basic pulse Sa is a plurality of bands included in a band (hereinafter sometimes referred to as “used band”) B used for transmission of a downlink signal. This is a sinc waveform obtained by inputting the number of carrier waves (for example, N) into the IFFT unit 8 with an amplitude of 1 / N and a phase of 0. In this case, only the real part I appears in the output of the IFFT unit 8 and the imaginary part Q becomes zero.

Thus, the basic pulse Sa is obtained by performing inverse Fourier transform on the plurality of subcarriers included in the frequency band B of the IQ baseband signal by the same IFFT unit 8 as that of the signal. I waveform (Sinc waveform).
Accordingly, the frequency band of the basic pulse Sa matches the use band B, and even if the instantaneous power P is clipped using the cancellation signal obtained by multiplying the increment ΔP of the instantaneous power P exceeding the threshold Pth by the basic pulse a, it is used. No unnecessary frequency component is generated outside the band B.

However, since only the signal component within the frequency band B used for transmission is used in the basic pulse Sa, the pulse width on the time axis is made too narrow as shown in the time waveform in the upper left frame of FIG. Can not do it.
For this reason, when only the basic pulse Sa is adopted as the cancellation pulse S and the instantaneous power P is canceled by the cancellation signal Pc obtained by multiplying it by the increment ΔP, the cancellation signal Pc interferes with the power waveform after the peak time. As a result, a new peak waveform is generated, and the peak power may not be reliably suppressed.

Therefore, in this embodiment, in order to appropriately suppress a certain amount of peak power even in a band outside the use band B, an auxiliary pulse Sb is defined in addition to the basic pulse Sa, and the auxiliary pulse Sb is used as the basic pulse. A composition combined with Sa is used as the canceling pulse S.
As shown in the time waveform in the lower left frame of FIG. 10, the auxiliary pulse Sb rises sharply at the time when the peak of the basic pulse Sa rises, and has a pulse waveform close to a delta function that is narrower than the basic pulse Sa. It becomes more.

Therefore, the frequency spectrum of the auxiliary pulse Sb is a very wide band including the use band B (hereinafter, also referred to as “broad band”) Bw.
Thus, the cancellation pulse S of the present embodiment is composed of a synthesized pulse obtained by synthesizing the auxiliary pulse Sb with the conventional basic pulse Sa. Therefore, as shown in the frequency spectrum in the right frame of FIG. It has not only the frequency band B of the signal but also a frequency component of a wide band Bw outside the band B.

[Peak level of each pulse]
Further, in the present embodiment, when the peak level of the basic pulse Sa is α and the peak level of the auxiliary pulse Sb is β, those levels α and so that 0.03 ≦ β / α ≦ 0.1 are satisfied. The ratio of β is set. The reason will be described below.

As described above, since the frequency component of the auxiliary pulse Sb extends over a wide band Bw that deviates from the use band B of the IQ baseband signal, the cancellation signal Pc obtained using the synthesized pulse including the auxiliary pulse Sb is used as the instantaneous power. Subtracting from P gives the same result as applying noise over a wide frequency band Bw.
For this reason, if the levels of the basic pulse Sa and the auxiliary pulse Sb are not set appropriately, the communication quality (EVM) in the use band B may be deteriorated, or high-level noise may be generated outside the use band B. There is sex.

Here, for example, in LTE, it is assumed that the EVM in the use band B is required to secure 40 dB, and the adjacent channel leakage power ratio is required to be secured 60 dB according to the Radio Law.
Therefore, the power reduction allowed in the use band B by the synthesis of the auxiliary pulse Sb is 20 dB at the maximum, and becomes 0.1 when converted into a voltage. Therefore, the peak level (voltage) ratio β / α of the auxiliary pulse Sb is allowable up to 0.1.

On the other hand, if the peak level of the auxiliary pulse Sb is too small, there is a possibility that a new peak waveform generated by subtraction of the cancellation signal Pc cannot be canceled appropriately. However, the ratio β / α of the peak level (voltage) of the auxiliary pulse Sb is It has been found that a new peak waveform can be canceled with a minimum of about 0.03.
From the above, regarding the peak levels α and β of the pulses S1 and S2, the ratio of the levels α and β may be set so as to satisfy 0.03 ≦ β / α ≦ 0.1. Become.

According to the CFR processing unit 14 of the present embodiment, the canceling pulse S (see FIG. 10) having not only the frequency component within the frequency band B of the IQ baseband signal but also the frequency component outside the band from the threshold Pth. The cancellation signal Pc obtained by multiplying the increment ΔP is subtracted from the instantaneous power P.
For this reason, it is possible to prevent a new peak waveform from being generated by subtraction of the cancellation signal Pc that affects the frequency component outside the band, and it is possible to more reliably suppress the peak power of the IQ baseband signal.

[Third Embodiment]
FIG. 11 is an overall configuration diagram of a wireless communication system according to the third embodiment. FIG. 12 is a functional block diagram showing a main part of the OFDM transmitter 3 of the base station apparatus 1 in that case.
Also in this embodiment, a radio communication system based on the LTE scheme is employed. In this type of base station apparatus 1, for example, the frequency band of the downlink frame can be set in units of 5 MHz, and when transmitting a downlink signal to each mobile terminal 2 in the cell, the transmission power is changed for each frequency band. It is possible.

The base station apparatus 1 of the present embodiment exemplifies a case where a downlink frame is transmitted in two types of frequency bands B1 and B2, and the transmission power of the first band B1 having a smaller frequency is set to be large, and the frequency The transmission power in the second band B2 having a larger value is set smaller.
For this reason, as shown by a broken line in FIG. 11, the communication area A1 in which the downlink signal of the first band B1 with high transmission power reaches is farther than the communication area A2 in which the downlink signal of the second band B2 with low transmission power reaches. And it has become widespread.

  In the area where the communication areas A1 and A2 overlap, the mobile terminal 2 can communicate in both the first and second bands B1 and B2, so that the mobile terminal 2 can reliably communicate even when the call volume is large. It will be.

In this embodiment, since it is assumed that the base station apparatus 1 transmits a downlink signal in two types of frequency bands B1 and B2, as shown in FIG. 12, subcarriers are included in the first band B1. 1 signal I 1, Q 1 and second signal I 2, Q 2 including subcarriers included in second band B 2 are output from IFFT section 8.
The first signals I1 and Q1 and the second signals I2 and Q2 are input to the power amplifier circuit 5 at the subsequent stage, and predetermined signal processing is performed in the circuit 5.

[Configuration of CFR processing unit]
FIG. 13 is a functional block diagram of the CFR processing unit (PC-CFR circuit) 14 according to the third embodiment. As shown in FIG. 13, the CFR processing unit 14 (FIG. 13) of the present embodiment differs from the CFR processing unit (FIG. 9) of the second embodiment in that the instantaneous power P corresponds to the average power for each frequency band. Is provided with a pulse generator 40 that generates a canceling pulse S that can cancel out the above.

Hereinafter, configurations and functions common to the second embodiment will be denoted by the same reference numerals in the drawings, description thereof will be omitted, and differences from the second embodiment will be mainly described.
In the following, the combined signal of the first signal I1, Q1 and the second signal I2, Q2 is simply referred to as “IQ baseband signal” or “IQ signal”.
The instantaneous power of the first signals I1 and Q1 is P1, the instantaneous power of the second signals I2 and Q2 is P2, and the instantaneous power of the IQ baseband signal is P (= P1 + P2).

A first calculation unit 41 and a second calculation unit 42 are provided in front of the CFR processing unit 14 of the present embodiment, in addition to the power calculation unit 12 that calculates the instantaneous power P.
Of these, the first calculating section 41 calculates the instantaneous power P1 of the first signal I1, Q1 consisting sum of squares of the I and Q components of the first signal I1, Q1 and ^{(= I1 2 + Q1 2)} . The second calculator 42 calculates the instantaneous power P2 (= I2 ² + Q2 ² ) of the second signals I2 and Q2 that is the sum of squares of the I component and Q component of the second signals I2 and Q2.

[Configuration of pulse generator]
FIG. 14 is a functional block diagram of the pulse generator 40.
The pulse generator 40 multiplies the composite pulses S1 and S2 obtained in advance for each of the first and second bands B1 and B2 by the relative ratios C1 and C2 of the average power for each of the bands B1 and B2, respectively. By taking the above, the cancellation pulse S is generated, and has a ratio calculation unit 43, a waveform storage unit 44, and a multiplication / addition unit 45.

Among these, the waveform memory | storage part 44 consists of memory | storage devices, such as a memory which memorize | stores the synthetic | combination pulses S1, S2 for every frequency band B1, B2. The synthesized pulses S1 and S2 are obtained by synthesizing the basic pulse Sa and the auxiliary pulse Sb (FIG. 4) as in the case of the second embodiment.
That is, the composite pulse S1 for the first band B1 is a real part obtained by performing inverse Fourier transform on the plurality of subcarriers included in the first band B1 by the same IFFT unit 8 as in the case of the transmission signal. The auxiliary pulse Sb is combined with the basic pulse Sa having the I waveform.

The synthesized pulse S2 for the second band B2 is a real part obtained by performing inverse Fourier transform on the plurality of subcarriers included in the second band B2 by the same IFFT unit 8 as in the case of the transmission signal. The auxiliary pulse Sb is combined with the basic pulse Sa having the I waveform.
However, in the present embodiment, the pulses S1 and S2 of the frequency bands B1 and B2 may be configured only from the basic pulse Sa without considering the auxiliary pulse Sb.

The ratio calculator 43 receives the instantaneous power P1 of the first signals I1 and Q1 calculated by the first calculator 41 and the instantaneous power P2 of the second signals I2 and Q2 calculated by the second calculator 42, respectively. .
The ratio calculation unit 43 uses these instantaneous powers P1 and P2 to calculate the relative ratios C1 and C2 of the average power for each of the frequency bands B1 and B2 based on the following equation.
C1 = Σ√P1 / (Σ√P1 + Σ√P2)
C2 = Σ√P2 / (Σ√P1 + Σ√P2)

  As shown in the above calculation formula, the relative ratios C1 and C2 of the average power for each of the frequency bands B1 and B2 are the square roots √P1 and √P2 of the instantaneous powers P1 and P2 for each of the frequency bands B1 and B2, and a predetermined sampling period. And the accumulated values Σ√P1, Σ√P2 are divided by the sum of the accumulated values of each frequency band B1, B2 (Σ√P1 + Σ√P2).

The ratio calculation unit 43 acquires the symbol period of the OFDM symbol, which is the minimum time unit in which the transmission power can fluctuate greatly, as the control period, and executes the calculation of the relative ratios C1 and C2 within this symbol period. It has become.
In this way, since the relative ratios C1 and C2 can be calculated in a stable state where the average power of the IQ baseband signal does not change much, there is an effect that accurate relative ratios C1 and C2 can be obtained.

  However, in LTE, the resource block (see FIG. 7) is the minimum unit for user allocation, and therefore 7 OFDM symbols (1 slot), which is the transmission period of this resource block, are used to calculate the relative ratios C1 and C2. You may decide to employ | adopt as a control period.

Multiplier / addition unit 45 includes two multipliers 46 and 47 and one adder 48. Among them, the multiplier 46 multiplies the relative ratio C1 corresponding to the first band B1 by the pulse S1 for the band B1, and the multiplier 47 multiplies the relative ratio C2 corresponding to the second band B2 for the band B2. Is multiplied by the pulse C2.
The adder 48 adds the multiplication results of the multipliers 46 and 47 to generate a cancellation pulse S, and outputs the pulse S to the pulse holding unit 32 of the CFR processing unit 14. That is, the multiplier / adder 45 generates the cancellation pulse S based on the following equation.
S = C1 * S1 + C2 * S2

The multiplication / addition unit 45 of this embodiment compares the relative ratios C1 and C2 calculated by the ratio calculation unit 43 with a predetermined threshold value to determine the variation, and the relative ratios C1 and C2 exceed the threshold value. Only when it fluctuates, multiplication and summation using the relative ratios C1 and C2 after the fluctuation are executed, and the cancellation pulse S generated as a result is output to the pulse holding unit 32.
Therefore, unless the relative ratios C1 and C2 change to some extent, the multiplication / addition unit 45 does not perform multiplication and summation, and the pulse holding unit 32 maintains the previous canceling pulse S. Therefore, the calculation load of the circuit can be reduced as compared with the case where the cancellation pulse S is generated every time.

The cancellation pulse S corresponds to the second band B2 obtained by multiplying the relative ratio C1 of the average power of the first signals I1 and Q1 corresponding to the first band B1 by the combined pulse S1 for the band B1. This is the sum of the relative ratio C2 of the average power of the second signals I2 and Q2 multiplied by the composite pulse S2 for the band B2.
Therefore, even if the cancellation signal Pc obtained by multiplying the cancellation pulse S by the increment ΔP is subtracted from the original instantaneous power P, the instantaneous power P of the first and second bands B1 and B2 corresponds to the average power. The amplitude will be canceled out.

  Therefore, according to the CFR processing unit 14 of the present embodiment, even when there is a difference in the average power in the first and second bands B1 and B2, the transmission power in the band B2 with the smaller average power is reduced in the cancellation signal Pc. Even in the case of the instantaneous power P of the IQ baseband signal having a different average power for each of the frequency bands B1 and B2, the clipping process can be appropriately performed without deteriorating the SNR without being reduced more than necessary due to the subtraction.

[Basic pulse variations]
FIG. 15 is a time-domain graph showing variations of the basic pulse Sa. In FIG. 15, (a) is a Sinc waveform, (b) is a Chebyshev waveform, and (c) is a Taylor waveform.
These waveforms can be expressed mathematically by the following equation (1), and in the case of a Sinc waveform, an = nπ.

Here, if the section including the maximum absolute value and the amplitude value becomes zero (section indicated by hatching in FIG. 15) is referred to as a main lobe section, the side lobe amplitude is compared in the case of the Sinc waveform. Therefore, the energy localization rate in the main lobe section cannot be improved much.
On the other hand, in the Chebyshev waveform, the amplitude of the side lobe can be reduced by adjusting the value of the numerical sequence an constituting the solution of x where the amplitude value = 0, but in this case, the amplitude is not attenuated.

Therefore, the tailor waveform uses Chebyshev waveform values for the first few points (for example, a1 and a2) of the sequence an, and Sinc waveform values for the subsequent points. Both sidelobe amplitude suppression and attenuation characteristics are achieved.
Therefore, when the energy localization rate in the main lobe section is compared with the total energy (the square of the amplitude) in the predetermined time section T defining the basic pulse Sa, it is 91% in the case of the Sinc waveform, and the Chebyshev waveform 93% in the case, and about 95% in the case of the tailor waveform, and the tailor waveform is the most advantageous.

  The predetermined time interval T is a sample time of the waveform recorded on the memory and is a time corresponding to the upper limit number of sample points. For example, in the case of LTE, the number of samples included in one symbol period (1/14 ms) is 2048. Therefore, assuming that oversampling is performed four times in the time domain, the waveform of the basic pulse Sa is defined. The upper limit number of sample points necessary for this is 2048 × 4 = 8192.

When the basic pulse Sa that can be used in the PC-CFR circuit is specified in the numerical range of the energy localization rate with respect to the predetermined time interval T of the main lobe interval, the energy localization rate is preferably 85% to 99%. .
The reason is that when the energy localization rate becomes 100%, the basic pulse Sa becomes an impulse (delta function) and cannot be applied to the present invention with band limitation. When the localization rate is less than 85%, the pulse This is because the shape is too slow to be used.

From the above, the technical features of the basic pulse Sa that can be used in the PC-CFR circuit are listed as follows.
Feature 1: The basic pulse Sa can be configured by a waveform having an energy localization rate of 85% to 99% in the main lobe period with respect to the total energy (amplitude squared) in a predetermined time period T (for example, one symbol period). .

Characteristic 2: When the basic pulse Sa is mathematically described, it consists of a waveform represented by the formula (1) having symmetry in the time domain.
Feature 3: More specifically, the basic pulse Sa is composed of a Sinc waveform, a Chebyshev waveform, or a Taylor waveform. Among these, the Sinc waveform is a waveform of a real part (I signal) obtained by inverse Fourier transform of a plurality of carriers in the band with the same amplitude and zero phase.

[Other variations]
The embodiment disclosed this time is illustrative and not restrictive. The scope of the present invention is defined by the terms of the claims, and includes all modifications that are within the scope of the claims and equivalents.
For example, the power amplifier circuit 5 of the present invention can be employed not only in the LTE system but also in a communication device compliant with the W-CDMA system.

  In this W-CDMA system, the transmission power of the base station apparatus 1 is controlled by closed-loop transmission power control, and this control cycle is the minimum time unit for transmission control. Specifically, this control cycle is 1/15 (= about 0.667 ms) of one radio frame cycle of 10 ms. Therefore, when the power amplifier circuit 5 of the present invention is used in a W-CDMA transmitter, a closed-loop transmission power control cycle may be employed as a control cycle when the threshold value Pth is updated.

Moreover, the input signal of the drain voltage modulation | alteration part 15 should just be a scalar quantity corresponding to the amplitude component of IQ baseband signal, and is not limited to the instantaneous electric power P of the signal.
For example, the scalar quantity may be defined by the square root of the instantaneous power P that is the amplitude information itself of the IQ baseband signal, or may be defined by the sum of the absolute value of the I component and the absolute value of the Q component. In this case, the CFR processing unit 14 may be configured by a CFR circuit that performs clipping processing on the defined scalar quantity.

DESCRIPTION OF SYMBOLS 1 Base station apparatus 2 Mobile terminal 3 Transmitter 4 Transmission processor 5 Power amplifier circuit 12 Power calculation part 13 Phase modulation part 14 CFR process part 15 Drain voltage modulation part 16 Power amplifier

Claims (8)

An EER (Envelope Elimination and Restoration) type power amplifier circuit,
A calculator that calculates a scalar amount corresponding to the amplitude component of the IQ baseband signal;
A phase modulation unit that extracts a phase component of the IQ baseband signal and modulates the phase component to a high frequency;
A CFR (Crest Factor Reduction) processing unit for limiting the upper limit of the scalar amount to a predetermined threshold value;
A power amplifier that operates using a phase modulation signal output from the phase modulation unit as an input signal, and an amplitude modulation signal corresponding to the scalar quantity with an upper limit limited as a power supply voltage;
A power amplifying circuit comprising:   The power amplification circuit according to claim 1, wherein the scalar quantity calculated by the calculation unit is an instantaneous power of the IQ baseband signal.   The CFR processing unit performs band limitation by filtering the peak component of the scalar amount exceeding a threshold value, and subtracts the peak component after the band limitation from the original scalar amount, NS-CFR (Noise Shaping- 3. The power amplifier circuit according to claim 1, comprising a Crest Factor Reduction circuit.   The CFR processing unit holds in advance a basic function for preventing out-of-band radiation due to clipping, and subtracts a value obtained by multiplying the peak component of the scalar quantity exceeding a threshold by the basic function from the original scalar quantity. The power amplifier circuit according to claim 1, comprising a PC-CFR (Peak Cancellation-Crest Factor Reduction) circuit.   The power amplification circuit according to claim 4, wherein the basic function includes a canceling pulse having frequency components both inside and outside the frequency band of the IQ baseband signal.   The power amplification circuit according to claim 4, wherein the basic function includes a canceling pulse that can cancel the scalar quantity corresponding to an average power of each frequency band of the IQ baseband signal.   The threshold value update part which updates the threshold value used by the said CFR process part for every predetermined | prescribed control period with which the average electric power of the said IQ baseband signal may fluctuate is further provided. The power amplifier circuit described in 1.   A communication apparatus comprising a transmitter on which the power amplifier circuit according to claim 1 is mounted.

Images