JP5175751B2 - Peak factor reduction device and base station - Google Patents

Description

  The present invention relates to a peak factor reduction apparatus and a base station, and more particularly to a peak factor reduction apparatus and a base station that employ an OFDM scheme.

  As a third-generation mobile communication system, a CDMA system with high use efficiency of frequency resources has been widely used. The CDMA system uses a code multiplexing system in which a plurality of baseband signals are spread into uncorrelated signals by orthogonal spreading codes, and the spread signals are multiplexed. Here, when the spread uncorrelated signals are multiplexed, the signal characteristics approach a normal distribution as the number of multiplexed signals increases. A signal having a normal distribution characteristic generates a peak power that is 10 dB or more higher than the average power of the signal, although the generation probability is low. In general, the ratio between the peak power and the average power is called a peak factor.

  When a signal having a normal distribution characteristic is transmitted by a wireless transmitter, the transmission spectrum is distorted unless the linearity of the transmitter is secured even with respect to peak power. For this reason, unnecessary radio waves are emitted outside the transmission frequency band. As a result, the system using the band is disturbed. Therefore, the amount of spectrum generated outside the band is strictly regulated by the Radio Law.

  For the reasons described above, it is difficult for the power amplifier used for the wireless transmitter to operate the average power near the saturation power of the amplifier, and it is difficult to sufficiently increase the power efficiency. As a result, there arises a problem that the apparatus size and running cost increase due to an increase in power consumption of the apparatus.

  As one method for solving this problem, a so-called peak factor reduction process is known in which the generation of peak power is suppressed by changing the distribution of the baseband signal. This peak factor reduction processing is described in Patent Document 1.

  With the technique of Patent Document 1, the peak factor that has been a problem in the CDMA system can be suppressed. For this reason, the average power of the power amplifier can be operated close to the saturated power, and as a result, the power efficiency can be increased. Further, an impulse signal is added in the vicinity of the peak amplitude to perform the peak factor reduction process, and the peak factor reduction process has an influence only in the vicinity of the peak amplitude. For this reason, it is possible to minimize degradation of signal quality. Furthermore, since the band limiting filter is applied after the impulse signal for reducing the peak factor is added to the transmission signal, the spread of the spectrum due to the peak factor reducing process can also be suppressed.

  Patent Document 2 discloses a peak factor reduction device in which the peak factor reduction effect is not impaired even if interpolation is performed after the peak factor reduction processing in Patent Document 1.

JP 2003-124824 A JP 2008-103881 A

  As a fourth generation mobile communication system, an OFDM system that is resistant to multipath delay waves has been widely studied. There is OFDMA as a multiple access system using this OFDM system. OFDMA positions multiple subcarriers as subchannels and performs multiple access by adaptively assigning subchannels to each user at arbitrary time timings. In OFDMA, since communication is performed using an arbitrary subchannel, subcarriers allocated to subchannels that are not used for communication are stopped. For this reason, the shape of the transmission spectrum varies greatly depending on the communication state.

  In the prior art, filter characteristics corresponding to arbitrarily changing spectrum shapes cannot be realized. For this reason, the transmission signal after the peak factor reduction results in an unnecessary spectrum in the subcarrier region not used for communication as shown in FIG. Here, FIG. 1 is a graph for explaining the spectrum shape of the OFDM transmission signal. In FIG. 1, a transmission spectrum with PFR of the conventional method has a spectrum generated in a stopped subcarrier region compared to a transmission spectrum without PFR (Peak Factor Reduction).

  The present invention has been made to solve the above-described problems of the prior art. In the present invention, the filter coefficient to be filtered is generated from the spectrum mask obtained from the FFT result of the transmission signal for the impulse signal added to the main signal for amplitude limitation. As a result, when the frequency arrangement of the transmission signal changes, generation of a frequency mask and generation of a filter coefficient following the change are performed. For this reason, generation | occurrence | production of the unnecessary spectrum to the area | region without a carrier arrangement | positioning can be suppressed.

  According to the present invention, it is possible to suppress generation of an unnecessary spectrum in a subcarrier region that is not used for communication.

It is a figure explaining the spectrum shape of an OFDM transmission signal. It is a block diagram of a radio base station. It is a block diagram of a peak factor reduction device. It is the spectrum and CCDF characteristic of the OFDM transmission signal explaining the OFDM signal peak factor reduction processing result. It is a block diagram of a filter coefficient generation circuit. It is a mapping waveform of a subcarrier. It is a block diagram of an impulse generation circuit. It is a figure explaining operation | movement of a dead zone circuit. It is a block diagram of a maximum value detection circuit. It is another block diagram of the maximum value detection circuit. It is another block diagram of a filter coefficient generation circuit. It is an IFFT output signal waveform. It is a window function circuit output signal waveform. It is an FFT waveform of a window function circuit output signal. FIG. 10 is still another block diagram of a filter coefficient generation circuit. It is another block diagram of a peak factor reduction apparatus. It is another block diagram of a radio base station.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings using examples. The same reference numerals are assigned to substantially the same parts, and the description will not be repeated.

  A first embodiment will be described with reference to FIGS. 2 to 9. First, the configuration of the radio base station will be described with reference to FIG. In FIG. 2, a radio base station 500 includes a transmitter 510, a receiver 520, a duplexer (DUP) 530, and an antenna 540. The transmitter 510 includes a digital unit 550 and an analog unit 560. The digital unit 550 includes a modem unit (MDM) 551, a baseband filter (BBFIL) 552, and a peak factor reduction device (PFR) 100. The analog unit 560 includes a DA converter (DAC) 561, a quadrature modulator (MOD) 562, a filter (FIL) 563, and a power amplifier (PA) 564.

  The modem unit 551 converts the transmission data into a baseband signal suitable for wireless communication. Baseband filter 552 limits the transmission band of the baseband signal. The peak factor reduction device 100 is mounted between the baseband filter 552 and the AD converter 561, and reduces the peak signal by digital signal processing.

  The DA converter 561 performs digital-analog conversion. The quadrature modulator 562 performs frequency conversion of the transmission signal. Filter 563 limits the band of the transmission signal. The power amplifier 564 performs power amplification of the transmission signal. A duplexer 530 following the transmitter 510 separates a transmission signal and a reception signal.

  Since the radio base station 500 performs the peak reduction processing by the peak factor reduction apparatus 100 in the previous stage of the power amplifier 564, the peak factor of the transmission signal input to the power amplifier 564 becomes low. Therefore, the power amplifier 564 can be operated close to the saturated power. For this reason, the power efficiency at the time of operation of the radio base station 500 can be improved.

  With reference to FIG. 3, the configuration of the peak factor reduction device will be described. As the spectrum shape of the arbitrarily arranged transmission wave, the spectrum without PFR in FIG. 4A is assumed. In FIG. 3, the peak factor reduction device 100 includes a baseband filter 110, a delay circuit 120, an impulse generation circuit 130, a filter coefficient generation circuit 140, a baseband filter 150, and a subtraction circuit 160.

  The transmission I and Q signals are input to the digital filter 110 to limit the band of the transmission signal. The baseband filter 110 is provided for suppressing spurious out of band and is realized by a digital low-pass filter. The band-limited transmission signal is input to the delay circuit 120, the impulse generation circuit 130, and the filter coefficient generation circuit 140.

  The signal input to the impulse generation circuit 130 generates an amplitude limit signal of the transmission signal by the impulse generation circuit 130 and the baseband filter 150. The filter coefficient generation circuit 140 generates filter coefficients for the baseband filter 150. The transmission I and Q signals input to the delay circuit 120 are delayed by the processing of the impulse generation circuit 130 and the baseband filter 150. The signal delayed by the delay circuit 120 is input to the subtraction circuit 160. The subtraction circuit 160 performs a subtraction process on the complex impulse signal band-limited by the baseband filter 150 from the transmission I and Q signals delayed by the delay circuit 120.

  The configuration of the filter coefficient generation circuit will be described with reference to FIG. In FIG. 5, a filter coefficient generation circuit 140 includes a symbol synchronization circuit 141, a CP (Cyclic Prefix) removal circuit 142, a Fast Fourier Transform (FFT) circuit 143, a frequency mask generation circuit 144, an inverse fast Fourier transform (IFFT). : Inverse Fast Fourier Transform) circuit 145.

  The band-limited transmission signal is input to the symbol synchronization circuit 141, where the symbol head is detected. This is realized by detecting a preamble signal in a transmission signal by a matched filter or using a reference signal given to the transmission signal. After the symbol head is detected by the symbol synchronization circuit 141, the CP removal circuit 142 removes the CP added to the transmission signal. The signal from which the CP has been removed is input to the fast Fourier transform circuit 143 and converted from a time axis waveform to a frequency axis waveform. The signal converted into the frequency axis waveform is input to the frequency mask generation circuit 144. The frequency mask generation circuit 144 outputs 1 when the amplitude of the frequency axis-converted signal is equal to or greater than a certain value, and outputs 0 when it is equal to or less than the certain value. That is, the frequency mask generation circuit 114 acquires a subcarrier map. A filter coefficient of the baseband filter 150 is generated by performing an inverse fast Fourier transform on the subcarrier map by an inverse fast Fourier transform circuit 145.

  With reference to FIG. 6, the subcarrier map acquired by the frequency mask generation circuit will be described. In FIG. 6, the subcarrier map is 1 for the subcarrier number corresponding to the spectrum without PFR in FIG.

  The operation of the impulse generation circuit will be described with reference to FIG. 7, FIG. 8, and FIG. In FIG. 7, the impulse generation circuit 130 includes an absolute value calculation circuit 131, a dead zone circuit 132, a maximum value detection circuit 133, two delay circuits 134 and 135, a division circuit 136, and a multiplier 137.

  First, the absolute value calculation shown in Expression (1) is performed in the absolute value calculation circuit 131 by the transmission I and Q signals input to the impulse generator circuit 130.

| X | = | I ^ 2 + Q ^ 2 | (1)
The absolute value calculation result | X | is input to the dead zone circuit 132. The dead zone circuit 132 outputs an amplitude greater than the set value A0 from the input | X | based on input / output characteristics described later with reference to FIG. The input / output result of the dead zone circuit 132 is expressed by Expression (2).

DEAD (X) = 0 [| X | <= A0],
DEAD (X) = | X | −A0 [| X |> A0] (2)
The dead zone circuit output signal is input to the maximum value detection circuit 133. The maximum value detection circuit 133 outputs an impulse signal proportional to the peak signal. The output signal of the maximum value detection circuit 133 is divided by the division circuit 136 and the absolute value arithmetic circuit output signal in which a delay equivalent to the delay amount generated by the dead zone circuit 132 and the maximum value detection circuit 133 is performed by the delay circuit 134. And normalization with the transmission power is performed.

  The output signal of the division circuit 136 is input to the impulse generation circuit 130 in which a delay equivalent to the delay amount generated in the absolute value calculation circuit 131, the dead zone circuit 132, and the maximum value detection circuit 133 is performed in the delay circuit 135. The I and Q signals are multiplied by the multiplier 137, and the impulse signal is complexed.

  The input / output characteristics of the dead zone circuit will be described with reference to FIG. In FIG. 8, the horizontal axis is input, and the vertical axis is output. The dead zone circuit 132 sets the output to 0 when the input is less than A0, and outputs (input−A0) when the input is A0 or more.

  The configuration of the maximum value detection circuit will be described with reference to FIG. In FIG. 9, the maximum value detection circuit 133 includes a differentiation circuit 1331, two delay circuits 1332 and 1335, two multipliers 1333 and 1336, and a negative value detection circuit 1334. The signal input to the maximum value detection circuit 133 is input to the differentiation circuit 1331 and the delay circuit 1335. The differentiation circuit 1331 calculates the difference between two consecutive samples. This is realized by a simple digital FIR filter having a filter coefficient of [1-1]. When x [n−1] is a signal one sample before x [n], the differentiation circuit output Yout is expressed by Equation (3).

Yout <0 (x [n] <x [n−1]),
Yout> 0 (x [n]> x [n−1]),
Yout = 0 (x [n] = x [n−1]) (3)
The output signal of the differentiation circuit 1331 is delayed by one sample by the delay circuit 1332, and the differentiation circuit output signal and the differentiation circuit output signal delayed by one sample by the delay circuit 1332 are input to the multiplier 1333, and the differentiation circuit output continues. Multiplication of data for two samples is performed. Here, since the output of the differentiation circuit is expressed as in Expression (3), only the sample in which the peak signal exists is negative, and positive or 0 is output otherwise.

  The output of the multiplier 1333 is input to the negative value detection circuit 1334. The negative value input circuit 1334 outputs 1 if it is negative, and 0 otherwise. As a result, the output waveform of the negative value detection circuit 1334 outputs an impulse signal at the timing of the peak of the transmission signal.

  The impulse signal generated by the negative value detection circuit 1334 is multiplied by the multiplier 1336 with the input signal of the maximum value detection circuit 133 obtained by delaying the delay amount generated until the impulse signal is generated by the delay circuit 1335.

  With reference to FIG. 4 again, the effect of the first embodiment will be described. In FIG. 4A, the horizontal axis represents frequency and the vertical axis represents relative intensity. Also, the presence or absence of PFR is compared as a parameter. In contrast to the transmission spectrum without PFR in FIG. 4A, the transmission spectrum with PFR slightly generates a spectrum in the stopped subcarrier region. However, the degree is clearly improved compared to FIG.

  FIG. 4B shows a CCDF (Complementally Cumulative Distribution Function) representing the frequency distribution of peak signals. In FIG. 4B, the horizontal axis represents the deviation intensity (dB) from the average power, and the vertical axis represents the frequency. As is apparent from FIG. 4B, the peak factor without PFR is 9.98 dB. On the other hand, the peak factor of Example 1 is 8.10 dB.

  Example 2 will be described with reference to FIG. The maximum value detection circuit described with reference to FIG. 9 can also be realized by the delay device, the comparator, the AND circuit, and the selector circuit shown in FIG. In FIG. 10, the maximum value detection circuit 133 </ b> A includes six delay devices 10 connected in series, six comparators 20, an AND circuit 30, and a selector circuit 40. The maximum value detection circuit 133A compares the three samples before and after and outputs a peak signal at the timing of the maximum value. The signal input to the maximum value detection circuit 133A is compared with the signals of three samples before and after in the comparator 20. The comparison result is input to the AND circuit 30. In the AND circuit 30, when all the comparison signals are larger than the data of the three samples before and after, all the outputs of the comparator 20 are 1, so the output of the AND circuit 30 is also 1. On the other hand, when a value larger than the comparison signal exists within three samples before and after, all the inputs of the AND circuit 30 do not become 1, and therefore the output of the AND circuit 30 becomes 0. The selector circuit 40 can output an impulse signal only at the timing of the peak signal by outputting a comparison signal when the output of the AND circuit 30 is 1, and outputting 0 when it is 0.

  A third embodiment will be described with reference to FIGS. 11 to 14. In FIG. 11, the filter coefficient generation circuit 140A includes a symbol synchronization circuit 141, a CP removal circuit 142, an FFT circuit 143, a frequency mask generation circuit 144, an IFFT circuit 145, and a window function circuit 146.

  In the filter frequency generation circuit 140A, a window function circuit 146 is added to the final stage. The signal shape of the subcarrier map created by FFT of the transmission I and Q signals is a square wave as shown in FIG. FIG. 12 shows filter coefficients created by IFFT of a square wave. From FIG. 12, it can be seen that the energy of the filter coefficient is locally distributed in the center of the coefficient. Therefore, even if the coefficients at both ends are cut off, the filter characteristics are not greatly affected. Therefore, the window function circuit 146 cuts off the filter coefficient of the subsequent baseband filter using the window function.

With reference to FIG. 13, the filter coefficient truncated by the Hamming window and the Hamming window is shown. Further, FIG. 14 shows frequency characteristics before and after the filter coefficient is cut off. In FIG. 14, it can be seen that the band cut-off characteristics are slightly deteriorated due to the truncation of the coefficients. If this deterioration satisfies the spectrum characteristics, the filter coefficient of the filter can be reduced by using the third embodiment, and the mounting hardware capacity can be reduced. In the third embodiment, the filter coefficient is reduced using a Hamming window, but a generally known Hanning window, Blackman window, or the like may be used.
According to the third embodiment, the number of filter taps can be reduced.

  Example 4 will be described with reference to FIG. In FIG. 15, the filter coefficient generation circuit 140B includes a symbol synchronization circuit 141, a CP removal circuit 142, an FFT circuit 143, a frequency mask generation circuit 144, an IFFT circuit 145, and a low-pass filter (LPF) 147. In the fourth embodiment, a low-pass filter 147 is added to the final stage of the filter coefficient generation circuit 140 of the peak factor reduction device shown in the first and third embodiments. In OFDM, the subcarrier map is switched for each OFDM symbol. Therefore, the coefficient of the baseband filter 150 changes abruptly between OFDM symbols. As a result, the output waveform is distorted at the change of the symbol. In order to suppress this rapid fluctuation of the filter coefficient, a low-pass filter 147 is added to the final stage of the filter coefficient generation circuit 140B. As a result, since rapid fluctuations in the filter coefficient can be suppressed, distortion of the output waveform at the change of the symbol can be suppressed.

  Example 5 will be described with reference to FIG. In FIG. 16, the peak factor reduction apparatus 100A includes a symbol synchronization circuit 110, a delay circuit 120, an impulse generation circuit 130, a filter coefficient generation circuit 140, a baseband filter 150, a subtraction circuit 160, a coefficient generation unit 170, and a coefficient selection unit 180. It is configured. In FIG. 16, in contrast to FIG. 3, a coefficient generation unit (TAP GEN) 170 that generates the same filter coefficients as the baseband filter 110 and a coefficient selection unit (TAP SELECT) 180 that selects the filter coefficients of the baseband filter 150. Has been added. When the transmission signal is a CDMA signal, the conventional peak factor reduction processing can be realized by selecting the filter coefficient of the baseband filter 150 as the coefficient of the TAP GEN 170 in the TAP SELECT 180. Further, when the transmission signal is OFDM, the peak factor reduction processing of the first embodiment can be realized by selecting the filter coefficient of the baseband filter 150 as the output of the filter coefficient generation circuit 140 by TAP SELECT 180, and CDMA can be realized by one apparatus. It is possible to realize peak factor reduction processing of both the OFDM and OFDM signals.

Example 6 will be described with reference to FIG. In FIG. 17, the radio base station 500A includes a transmitter 510A, a receiver 520, a duplexer 530, and an antenna 540. The transmitter 510A further includes a digital unit 550A and an analog unit 560A. The digital unit 550A includes a modem unit 551, a baseband filter 552, a peak factor reduction device 100, and a digital predistortion (DPD) circuit 553. The analog unit 560A includes a DA converter 561, an orthogonal transformer 562, a filter 563, a power amplifier 564, an attenuator (ATT) 565, a mixer (MIX) 566, and an AD converter (ADC) 567. 17, the power amplifier output signal for row of cormorants D P D circuitry 553 and DPD 553 the distortion compensation of the power amplifier 564 to the radio base station 500 plus the peak factor reduction apparatus 100 shown in Embodiment 1 This is a radio base station 500A to which a detection circuit (565 to 567) is added.

  Attenuator 565 attenuates the output of power amplifier 564. The mixer 566 down-converts the output of the attenuator 565. The AD converter 567 AD converts the output of the mixer 566. The DPD circuit 553 compares the input signal from the peak factor reduction device 100 with the input signal from the AD converter 567 and compensates for distortion of the power amplifier 564.

  The power amplifier 564 used in the radio base station 500A has a nonlinear characteristic in input / output characteristics from a level considerably lower than the saturation region. For this reason, the output waveform is distorted even if the output level is not near the saturated power of the power amplifier 564. The DPD circuit 553 compensates for the characteristics of the non-linear region existing up to the saturation region. The DPD circuit 553 cannot compensate for the saturation region because of its operating characteristics. Therefore, by combining with the peak factor reduction device 100, the power amplifier 564 can be operated closer to the saturated power.

  DESCRIPTION OF SYMBOLS 100 ... Peak factor reduction device, 110 ... Baseband filter, 120 ... Delay circuit, 130 ... Impulse generation circuit, 131 ... Absolute value circuit, 132 ... Dead zone circuit, 133 ... Maximum value detection circuit, 134 ... Delay circuit, 135 ... Delay circuit, 136 ... division circuit, 137 ... multiplier, 140 ... filter coefficient generation circuit, 141 ... symbol synchronization circuit, 142 ... CP elimination circuit, 143 ... fast Fourier transform circuit, 144 ... frequency mask generation circuit, 145 ... inverse high speed Fourier transform circuit, 146 ... window function circuit, 147 ... low pass filter, 150 ... baseband filter, 160 ... subtraction circuit, 170 ... coefficient generator, 180 ... TAP SELECT, 500 ... radio base station, 510 ... transmitter, 520 ... Receiver 530 ... Duplexer 540 ... Antenna 550 ... De Digital unit 551 Modem unit 552 Baseband filter 553 Digital predistortion circuit 560 Analog unit 561 DA converter 562 Quadrature modulator 563 Filter 564 Power amplifier 566 Attenuation , 566 ... mixer, 567 ... AD converter.

Claims (6)

A first baseband filter for filtering a signal outside the transmission band, and a complex impulse signal having an amplitude proportional to the excess when the amplitude component of the output signal of the first baseband filter exceeds a set value an impulse generator for a second baseband filter for performing a filtering process of the complex impulse signal at the output of the impulse generator, and a pre-Symbol filter coefficient generation circuit for generating a filter coefficient of the second baseband filter, wherein A delay device that delays the output signal of the first baseband filter by a processing time of the impulse generation circuit and the second baseband filter, and an output signal of the second baseband filter from the output signal of the delay device A peak factor reduction device comprising a subtractor for subtracting
The filter coefficient generation circuit subjects the filter coefficient of the second baseband filter to symbol synchronization, CP removal, and FFT processing on the output signal of the first baseband filter that is a transmission signal, Frequency mask circuit by converting to waveform, inputting to frequency mask generation circuit, and outputting 1 when the amplitude of the frequency axis converted signal is greater than or equal to a predetermined value, and 0 when less than a certain value A peak factor reduction apparatus characterized by generating a filter coefficient that follows a change in frequency arrangement of a transmission signal by performing subcarrier map generation by performing IFFT processing on the subcarrier map . The peak factor reduction device according to claim 1,
A peak factor reduction device comprising a window function circuit for reducing a filter coefficient by a window function at a final stage of the filter coefficient generation circuit. The peak factor reduction device according to claim 1,
A peak factor reduction device having a low-pass filter for suppressing a rapid fluctuation of a filter coefficient at a final stage of the filter coefficient generation circuit. The peak factor reduction device according to claim 1,
The filter coefficient generation unit includes a coefficient generation unit that generates the same filter coefficient as the first baseband filter and a coefficient selection unit that selects a coefficient of the second baseband filter. Factor reduction device. A modem unit that converts transmission data into a baseband signal suitable for wireless communication, a baseband filter that limits transmission band of the baseband signal, a DA converter that performs digital-analog conversion, and frequency conversion of the transmission signal In a radio base station including a transmitter including a quadrature modulator, a power amplifier that performs power amplification of a transmission signal, and a filter that performs band limitation of the transmission signal,
Between the baseband filter and the DA converter,
A first baseband filter for filtering a signal outside the transmission band, and a complex impulse signal having an amplitude proportional to the excess when the amplitude component of the output signal of the first baseband filter exceeds a set value an impulse generator which, the impulse output and a second baseband filter for performing a filtering process of the complex impulse signal generating circuit, before Symbol first baseband the said output signal of the filter impulse generating circuit and the second A peak factor reduction device comprising: a delay device that delays the processing time of the baseband filter; and a subtractor that subtracts the output signal of the second baseband filter from the output signal of the delay device ,
Further, the output signal of the first baseband filter, which is a transmission signal, is used as an input signal, symbol synchronization, CP removal, and FFT processing are performed on the output signal of the first baseband filter to obtain a frequency axis waveform Is output to the frequency mask generation circuit, and the frequency mask circuit outputs 1 when the amplitude of the frequency-axis converted signal is equal to or larger than a predetermined value, and 0 when the amplitude is equal to or smaller than the predetermined value. A filter coefficient generation circuit for generating a subcarrier map, generating a filter coefficient following a change in frequency arrangement of a transmission signal by performing IFFT processing on the subcarrier map, and applying the generated filter coefficient to the second baseband filter; A radio base station. The radio base station according to claim 5, wherein
A radio base station comprising a digital predistortion circuit that performs distortion compensation of an amplifier downstream of the peak factor reduction device.

Images