JP2017118483A - Communication device and transmission method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the deteriorated quality of each signal in a communication system in which a plurality of signals are multiplexed and transmitted from a communication device to a remote device.SOLUTION: A communication device includes: a signal generation circuit which generates a plurality of intermediate frequency signals; a synthesizer which synthesizes the plurality of intermediate frequency signals to generate a multi-channel signal; a reduction unit which reduces a peak-to-average power ratio of the multi-channel signal; and a converter which converts the multi-channel signal having the reduced peak-to-average power ratio by the reduction unit into an optical signal. The reduction unit reduces the power of the multi-channel signal when the power of the multi-channel signal is higher than a predetermined threshold level, using an auxiliary signal whose frequency is different from either of the plurality of intermediate frequency signals.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a communication apparatus and a transmission method for transmitting an optical signal in which a plurality of signals are multiplexed.

  As one of the techniques for reducing the cost for constructing a wireless communication system, a distributed antenna system (DAS) has been put into practical use. In the distributed antenna system, a signal processing device that processes transmission signals and a wireless device that outputs radio signals are separated. In the following description, the signal processing apparatus may be referred to as a “digital processing unit”. In addition, the wireless device may be referred to as “remote radio unit (RRU)” or “remote radio head (RRH)”.

  Transmission between the digital processing unit and the remote radio unit is realized by, for example, optical fiber radio (RoF). In optical fiber radio, a radio frequency signal or an intermediate frequency signal is transmitted through an optical fiber. A configuration in which an intermediate frequency signal is transmitted through an optical fiber is sometimes called IFoF (Intermediate Frequency over Fiber). IFoF is a form of RoF. Note that RoF or IFoF is described in Non-Patent Documents 1 and 2, for example.

  For example, the digital processing unit up-converts the data signal to generate an intermediate frequency signal (hereinafter referred to as IF signal). In this case, the digital processing unit converts the IF signal into an optical IF signal and transmits the optical IF signal to the remote wireless unit via the optical fiber. The remote radio unit converts the received optical IF signal into an electrical signal and amplifies it. Then, the remote radio unit transmits the amplified signal to the mobile terminal via the radio link.

  In order to increase the capacity of a radio link, a multi-antenna system that transmits a plurality of radio signals using a plurality of antennas has been put into practical use. As an example of a multi-antenna system, for example, in a multiple input multiple output (MIMO) system, a plurality of radio signals transmitted from a plurality of transmitting antennas are received by a plurality of receiving antennas.

  Furthermore, a configuration in which a plurality of transmission antennas are provided in the above-described remote wireless unit has been proposed. In this configuration, a plurality of signals are multiplexed and transmitted from the digital processing unit to the remote radio unit via an optical fiber. Then, the remote radio unit amplifies each signal and then outputs it through the antenna.

  However, when a plurality of signals are multiplexed and transmitted, the peak-to-average power ratio (PAPR) of the multiplexed signal becomes high. When the peak-to-average power ratio is high, the signal waveform may be distorted due to the saturation of the circuit elements in the transmitter. For example, in the digital processing unit of the above-described RoF system or IFoF system, when the peak power of the multiplexed signal reaches the saturation region of the laser element, the waveform of the optical signal is distorted. In this case, communication quality may be deteriorated.

  For this reason, conventionally, techniques for reducing the peak-to-average power ratio have been proposed (for example, Non-Patent Documents 3 to 6). Patent Documents 1 to 7 describe related technologies.

JP 2008-085379 A JP 2006-005390 A JP 2001-237800 A Special table 2013-515424 gazette JP 2009-290493 A JP 2009-055558 A JP2013-153479A

Charles H. Cox III et. Al. “Limits on the Performance of RF-Over-FiberLinks and Their Impact on Device Design,” IEEE Translations On Microwave Theory and Techniques, vol. 54, no 2, pp.906-920, Feb . 2006. Changyo Han, Seung-Hyun Cho, Hwan Seok Chung, Sang Soo Lee and Jonghyun Lee, “Experimental Comparison of the Multi-IF Carrier Generation Methods in IF-over-Fiber System Using LTE Signals,” MWP 2014, Sapporo, Japan. Jose Tellado and John M. Cioffi, “Efficient Algorithms for Reducing PAR in Multicarrier Systems,” ISIT 1998. Cambridge, MA, USA. H. Han and J. H. Lee, “An overview of peak-to-average power ratio reduction techniques for multicarrier transmission,” IEEE Wireless Communications, vol. 12, no. 2, pp. 56-65, April 2005. Mohamad Mroue, Amor Nafkha, Jacques Palicot, Benjamin Gavalda, and Nelly Dagorne, “Performance and Implementation Evaluation of TR PAPR Reduction Methods for DVB-T2”, Hindawi Publishing Corporation International Journal of Digital Multimedia Broadcasting Volume 2010, Article ID 797393, 10 pages doi: 10.1155 / 2010/797393 Hou-Tzu Huang et al. “W-band DD-OFDM-RoF System Employing Pilot-aided PAPR Reduction” Wireless Microwave Photonics (WMP) 2014

  In the prior art, the peak-to-average power ratio is reduced, for example, by “clipping” the peak power of the electrical signal within the communication device that transmits the optical signal. In this case, an optical signal is generated from the clipped electrical signal and transmitted to the destination device (eg, remote radio unit). However, with this method, the waveform of the signal reproduced in the destination device is distorted. That is, the waveform distortion due to the saturation of the laser element can be eliminated, but the waveform of the signal reproduced on the receiving side may be distorted.

  An object according to one aspect of the present invention is to suppress deterioration in quality of each signal in a communication system in which a plurality of signals are multiplexed and transmitted from a communication device to a remote device.

  A communication apparatus according to one aspect of the present invention includes a signal generation circuit that generates a plurality of intermediate frequency signals, a combiner that combines the plurality of intermediate frequency signals to generate a multichannel signal, and a peak of the multichannel signal. A reduction unit that reduces an average power ratio; and a converter that converts a multi-channel signal whose peak-to-average power ratio is reduced by the reduction unit into an optical signal. The reduction unit reduces the power of the multichannel signal by using an auxiliary signal having a frequency different from any of the plurality of intermediate frequency signals when the power of the multichannel signal is higher than a predetermined threshold level.

  According to the above aspect, in the communication system in which a plurality of signals are multiplexed and transmitted from the communication device to the remote device, deterioration of the quality of each signal is suppressed.

It is a figure which shows an example of the communication system which uses an optical fiber radio. It is a figure which shows an example of the digital processing unit concerning embodiment of this invention. It is a figure which shows the example of a multichannel signal. It is a figure which shows an example of the sampling of a multichannel signal. It is a figure which shows an example of operation | movement of a clipping circuit and a subtractor. It is a flowchart which shows an example of a process of a PAPR reduction circuit. It is a figure which shows typically the signal processing of a PAPR reduction circuit. It is a figure which shows an example of a remote radio | wireless unit. It is a figure explaining the filter which removes a tone signal. It is a figure which shows the example of the spectrum of the multichannel signal produced | generated in a digital processing unit. It is a figure which shows the simulation result about suppression of a peak-to-average power ratio. It is a figure which shows an example of the time change of the amplitude of a tone signal. It is a figure explaining the mutual phase modulation distortion of a multi-channel signal. It is a figure which shows an example of arrangement | positioning of adjustment of a carrier frequency, and a tone signal. It is a figure which shows an example of the digital processing unit concerning other embodiment. It is a flowchart which shows an example of the process which determines arrangement | positioning of IF signal and a tone signal. It is FIG. (1) which shows an example of the other method which arrange | positions an IF signal and a tone signal. It is FIG. (2) which shows an example of the other method which arrange | positions an IF signal and a tone signal. It is a figure which shows an example of the simulation result which calculated PAPR with respect to arrangement | positioning of IF signal.

  FIG. 1 shows an example of a communication system using optical fiber radio (RoF). The communication system shown in FIG. 1 includes a digital processing unit 100, a remote radio unit 200, and an optical fiber cable 300 that connects the digital processing unit 100 and the remote radio unit 200.

  The digital processing unit 100 includes a plurality of modulators 1-1 to 1-n, a plurality of oscillators 2-1 to 2-n, a combiner 3, and an E / O conversion circuit 4. Baseband region data signals CH1 to CHn are applied to modulators 1-1 to 1-n, respectively. The data signals CH1 to CHn are generated by, for example, OFDM (Orthogonal Frequency Division Multiplexing). In OFDM, data is transmitted using a number of subcarriers orthogonal to each other. In FIG. 1, the data signals CH1 to CHn are represented by an I component and a Q component, respectively. The modulators 1-1 to 1-n are supplied with oscillation signals IF1 to IFn generated by the corresponding oscillators 2-1 to 2-n, respectively. The frequencies of the oscillation signals IF1 to IFn are different from each other. Further, the oscillation signals IF1 to IFn are arranged in the intermediate frequency band in this embodiment.

  Modulators 1-1 to 1-n generate IF signals CH1 to CHn by modulating oscillation signals IF1 to IFn with data signals CH1 to CHn, respectively. Note that IF signals CH1 to CHn are modulation signals that transmit data signals CH1 to CHn, respectively. The combiner 3 adds the IF signals CH1 to CHn. Then, the E / O conversion circuit 4 converts the output signal of the combiner 3 into an optical signal. The E / O conversion circuit 4 includes, for example, a laser element. The optical signal output from the E / O conversion circuit 4 is transmitted to the remote radio unit 200 via the optical fiber cable 300. That is, the data signals CH1 to CHn are transmitted to the remote radio unit 200 by IFoF (Intermediate Frequency over Fiber). Here, the frequencies of the carriers (oscillation signals IF1 to IFn) transmitting the data signals CH1 to CHn are different from each other. Therefore, data signals CH1 to CHn are transmitted to remote radio unit 200 by frequency division multiplexing. That is, a multi-channel signal in which a plurality of IF signals are multiplexed is transmitted from the digital processing unit 100 to the remote radio unit 200.

  The remote radio unit 200 includes an O / E conversion circuit 11, a plurality of demodulators 12-1 to 12-n, a plurality of oscillators 13-1 to 13-n, a plurality of mixers 14-1 to 14-n, and a plurality of amplifiers. 15-1 to 15-n and a plurality of antennas 16-1 to 16-n are provided. The remote wireless unit 200 receives the optical signal output from the digital processing unit 100 via the optical fiber cable 300.

  The O / E conversion circuit 11 converts the received optical signal into an electrical signal. This electric signal includes IF signals CH1 to CHn. Note that the O / E conversion circuit 11 includes, for example, a photodiode. The demodulators 12-1 to 12-n use the oscillation signals IF1 to IFn generated by the oscillators 13-1 to 13-n, respectively, and include IFs included in the output signals of the O / E conversion circuit 11. The signals CH1 to CHn are demodulated. Note that the frequencies of the oscillation signals IF1 to IFn generated by the oscillators 2-1 to 2-n of the digital processing unit 100 and the oscillation signals IF1 to IFn generated by the oscillators 13-1 to 13-n of the remote radio unit 200 are described. Are substantially the same. Therefore, the data signals CH1 to CHn in the baseband region are reproduced by the demodulators 12-1 to 12-n.

  The mixers 14-1 to 14-n upconvert the regenerated data signals CH1 to CHn using the radio frequency oscillation signal LO to generate RF signals CH1 to CHn, respectively. A modulator may be provided between each of the demodulators 12-1 to 12-n and the mixers 14-1 to 14-n. In this case, the mixers 14-1 to 14-n upconvert the corresponding modulation signals using the oscillation signals LO, respectively, to generate RF signals CH1 to CHn. The amplifiers 15-1 to 15-n amplify the RF signals CH1 to CHn, respectively. The amplifiers 15-1 to 15-n are high power amplifiers (HPAs) in this embodiment. The antennas 16-1 to 16-n output RF signals CH1 to CHn amplified by the amplifiers 15-1 to 15-n, respectively.

  The communication system according to the embodiment of the present invention can be applied to, for example, a MIMO system. The digital processing unit 100 is an example of a communication device that transmits a multi-channel signal in which a plurality of signals (for example, intermediate frequency signals) are multiplexed. The remote radio unit 200 is an example of a remote device that receives a multi-channel signal transmitted from the digital processing unit 100.

  Thus, the digital processing unit converts the multi-channel signal into an optical signal and transmits it to the remote radio unit. Here, the multi-channel signal is generated by combining a plurality of IF signals as shown in FIG. For this reason, even if the peak-to-average power ratio of each IF signal is reduced, the peak-to-average power ratio of the multichannel signal may regrow. When the peak-to-average power ratio of the multichannel signal is large, the waveform of the optical signal may deteriorate due to saturation of the E / O conversion circuit. In the example shown in FIG. 1, the E / O conversion circuit 4 may be saturated. Further, if the amplitude of the input signal to the E / O conversion circuit is simply reduced in order to avoid saturation of the E / O conversion circuit, the signal-to-noise ratio of the optical signal is deteriorated. Thus, the digital processing unit comprises a PAPR reduction circuit for reducing the peak-to-average power ratio of the multichannel signal.

  FIG. 2 shows an example of a digital processing unit according to an embodiment of the present invention. As illustrated in FIG. 2, the digital processing unit 500 according to the embodiment includes a plurality of modulators 1-1 to 1-n, a plurality of oscillators 2-1 to 2-n, a combiner 3, an E / O conversion circuit 4, A PAPR reduction circuit 20 is provided. Note that the digital processing unit 500 shown in FIG. 2 can be used in place of the digital processing unit 100 in the communication system shown in FIG. That is, the digital processing unit 500 can transmit an optical signal to the remote radio unit 200 via the optical fiber cable 300.

  The modulators 1-1 to 1-n, the oscillators 2-1 to 2-n, and the combiner 3 are substantially the same in FIGS. That is, modulators 1-1 to 1-n and oscillators 2-1 to 2-n generate IF signals CH1 to CHn. Note that the carrier frequencies of IF signals CH1 to CHn are different from each other. The combiner 3 also adds the IF signals CH1 to CHn to generate a multichannel signal.

  The PAPR reduction circuit 20 includes a switch 21, a subtractor 22, a clipping circuit 23, a subtractor 24, a tone signal generator 25, multipliers 26 and 27, a delay circuit 28, an output circuit 29, and a controller 30. Then, the PAPR reduction circuit 20 reduces the peak-to-average power ratio of the multichannel signal as necessary. As an example, when the power of the multi-channel signal generated by the combiner 3 is higher than a predetermined threshold level, the PAPR reduction circuit 20 includes a plurality of IF signals CH1 to CHn multiplexed in the multi-channel signal. The auxiliary signal (tone signal in this embodiment) having a frequency different from any one of them is used to reduce the power of the multichannel signal.

  The PAPR reduction circuit 20 performs signal processing in the frequency domain, for example. In this case, the digital processing unit 500 includes a Fourier transform (FFT) circuit for converting a time domain multi-channel signal into a frequency domain multi-channel signal between the combiner 3 and the PAPR reduction circuit 20. May be. When the combiner 3 outputs an analog multichannel signal, the digital processing unit 500 may further include an A / D (Analog-to-Digital) converter between the combiner 3 and the Fourier transform circuit. In addition, the digital processing unit 500 includes an inverse Fourier transform (IFFT) circuit for converting a frequency domain multichannel signal into a time domain multichannel signal between the PAPR reduction circuit 20 and the E / O conversion circuit 4. And a D / A (Digital-to-Analog) converter.

  The multichannel signal output from the combiner 3 is guided to the switch 21 of the PAPR reduction circuit 20. As shown in FIG. 3A, IF signals CH1 to CHn are multiplexed in the multichannel signal output from the combiner 3. IF signals CH1 to CHn are all arranged in the intermediate frequency region. The IF signals CH1 to CHn are not particularly limited, but are arranged at intervals of 60 MHz, for example.

  The multi-channel signal is generated by multiplexing a plurality of IF signals CH1 to CHn having different carrier frequencies. For this reason, the power of the multi-channel signal varies with time as shown in FIG. The PAPR reduction circuit 20 samples and processes the multi-channel signal output from the combiner 3 at a predetermined time interval. In the example shown in FIG. 4, the PAPR reduction circuit 20 includes S1, S2, S3. . . Process the multi-channel signal sampled at.

  The PAPR reduction circuit 20 compares the power of the multi-channel signal with a predetermined threshold level for each sampling period. When the power of the multichannel signal is larger than the threshold level, the PAPR reduction circuit 20 reduces the power of the multichannel signal. The threshold level is determined based on the characteristics of the E / O conversion circuit 4, for example. As an example, the threshold level is determined so that the power of the output optical signal is not saturated with respect to the amplitude of the input electric signal in the E / O conversion circuit 4.

  The PAPR reduction circuit 20 executes an iterative process for reducing the peak power of the multichannel signal for each sampling. The number of executions of the iterative process is assumed to be specified in advance. By this iterative process, the peak power of the multichannel signal gradually approaches the threshold level.

  The switch 21 selects an input multichannel signal or an output signal of the delay circuit 28 in accordance with an instruction given from the controller 30. Specifically, the switch 21 selects the input multichannel signal in the initial state of the iterative process for each sampling. Thereafter, the switch 21 selects the output signal of the delay circuit 28 during a period in which the iterative process is executed. The output signal of the switch 21 is led to the subtractor 22. The subtracter 22 subtracts the output signal of the multiplier 27 from the output signal of the switch 21. The calculation result by the subtracter 22 is used in the next calculation in an iterative process described later. Therefore, in the following description, the multi-channel signal output from the subtractor 22 (that is, the multi-channel signal being processed in the PAPR reduction circuit 20) may be referred to as “multi-channel signal X”. Multichannel signal X includes IF signals CH1 to CHn shown in FIG. The multi-channel signal X may include a tone signal described later.

  The clipping circuit 23 compares the power of the input signal with a predetermined threshold level. A multi-channel signal X generated by the subtractor 22 is input to the clipping circuit 23. As described above, the threshold level is determined based on the characteristics of the E / O conversion circuit 4. This threshold level is sometimes called a “clipping level”. When the power of the input signal is higher than the threshold level, the clipping circuit 23 removes a signal component corresponding to the difference between the power of the input signal and the threshold level from the input signal. The subtractor 24 subtracts the output signal of the clipping circuit 23 from the multichannel signal X.

  FIG. 5 shows an example of the operation of the clipping circuit 23 and the subtractor 24. When the power of the multichannel signal X is lower than the threshold level, the clipping circuit 23 does not change the multichannel signal X. In this case, the output signal of the subtractor 24 is “zero”. On the other hand, when the power of the multi-channel signal X is higher than the threshold level, the clipping circuit 23 generates a signal component corresponding to the difference between the power of the multi-channel signal X and the threshold level from the multi-channel signal X. Remove. Then, the subtractor 24 calculates the difference between the multichannel signal X and the output signal of the clipping circuit 23. Therefore, the output signal of the subtractor 24 represents the signal component removed by the clipping circuit 23 as shown in FIG. In the following description, the output signal of the subtractor 24 may be referred to as a “clipping noise signal cn”. Note that the intensity of the clipping noise signal cn is substantially constant over a wide frequency range.

  The tone signal generator 25 generates a tone signal having a predetermined frequency. In this embodiment, the tone signal generator 25 generates a plurality of tone signals. In this case, the frequencies of the plurality of tone signals are different from each other. Further, as shown in FIG. 3B, the frequency of the tone signal is different from any of the carrier frequencies of the plurality of IF signals CH1 to CHn included in the multichannel signal. When each IF signal CH1 to CHn is generated by OFDM, the tone signal is generated regardless of the subcarrier of each IF signal CH1 to CHn.

  For example, tone signals are placed between frequency channels. In the example shown in FIG. 3B, three tone signals are arranged between IF signal CH1 and IF signal CH2, and three tone signals are arranged between IF signal CH2 and IF signal CH3. Yes. Further, the tone signal may be arranged outside the frequency band of the multichannel signal. In the example shown in FIG. 3B, the tone signal is arranged on the lower frequency side than the IF signal CH1, and the tone signal is arranged on the higher frequency side than the IF signal CHn. The tone signal is arranged so as not to overlap with the spectrum of each IF signal. The shape and width of the spectrum of the IF signal depends on the bit rate, the modulation scheme, and the like.

  In this embodiment, the amplitudes of the plurality of tone signals generated by the tone signal generator 25 are the same. However, the amplitudes of the plurality of tone signals generated by the tone signal generator 25 may not be the same. Note that the tone signal is an example of an auxiliary signal used to reduce the peak-to-average power ratio of the multichannel signal.

  The multiplier 26 multiplies the output signal of the subtractor 24 by the tone signal. That is, the multiplier 26 multiplies the clipping noise signal cn by the tone signal. Thus, multiplier 26 provides frequency masking with a tone signal for clipping noise signal cn. That is, the multiplier 26 outputs the clipping noise signal cn at the frequency where the tone signal is arranged, and outputs “zero” at other frequencies. In other words, the signal C output from the multiplier 26 includes only the frequency component of the tone signal. Note that when the power of the multi-channel signal X is lower than the threshold level, the clipping noise signal cn is “zero”, and the signal C output from the multiplier 26 is also “zero”.

  The multiplier 27 multiplies the signal C output from the multiplier 26 by the parameter μ. The parameter μ is a real number greater than zero and less than one. That is, the multiplier 27 reduces the amplitude of the signal C. Then, as described above, the subtracter 22 subtracts the output signal of the multiplier 27 from the output signal of the switch 21 to generate the multichannel signal X.

  The multi-channel signal X generated by the subtracter 22 is guided to the clipping circuit 23, the delay circuit 28, and the output circuit 29. The clipping circuit 23 clips the multi-channel signal X as described above. The delay circuit 28 delays the multichannel signal X. Specifically, the delay circuit 28 temporarily stores the multichannel signal X. Then, the multi-channel signal X stored in the delay circuit 28 is read for the next calculation in the iterative process and guided to the switch 21. That is, the multichannel signal X obtained by the previous calculation in the iterative process is led to the subtractor 22 via the switch 21. The output circuit 29 stores the latest multi-channel signal X in the iterative process. When the output instruction indicating the end of the iterative process is received from the controller 30, the output circuit 29 reads and outputs the latest multi-channel signal X.

  Next, the operation of the PAPR reduction circuit 20 will be described. In the following, it is assumed that the multi-channel signal shown in FIG. As shown in FIG. 3A, IF signals CH1 to CHn are multiplexed in the multichannel signal. The PAPR reduction circuit 20 samples and processes the multichannel signal at a predetermined period.

The PAPR reduction circuit 20 executes an iterative process for reducing the peak-to-average power ratio for the sampled multi-channel signal. In the following description, the number of executions of the calculation in the iterative process is represented by “i”. For example, the operation result of the subtractor 22 in the i-th operation is represented by X _i .

The multi-channel signal sampled in the sampling period S1 is guided to the switch 21. The switch 21 selects the output signal of the combiner 3 (that is, the input multichannel signal) in the initial state of the iterative process. The multi-channel signal input to the PAPR reduction circuit 20 may be referred to as “original multi-channel signal”. Then, subtractor 22, by subtracting the output signal of the multiplier 27 from the input multi-channel signal, generating a multi-channel signal X _0. Here, in this embodiment, the signal C is “zero” in the initial state of the iterative process. Accordingly, the clipping circuit 23, as a multi-channel signal X _0, based on the multi-channel signal (i.e., a multi-channel signal input to the PAPR reduction circuit 20) is input. At this time, the multi-channel signal X ₀ is stored in the delay circuit 28.

The clipping circuit 23 compares the power of the input signal with a threshold level. Here, the power of the multi-channel signal sampled in the sampling period S1 is lower than the threshold level. In this case, the clipping noise signal cn ₀ output from the subtractor 24 is zero, and the signal μC ₀ generated by the multipliers 26 and 27 is also zero.

Subsequently, an iterative process for the sampling period S1 is executed. When the iterative process is executed, the switch 21 selects a signal stored in the delay circuit 28. Here, the delay circuit 28, the multi-channel signal X ₀ is stored. Therefore, the switch 21 selects the multi-channel signal X ₀ stored in the delay circuit 28 and guides it to the subtractor 22. Then, the subtracter 22 subtracts the output signal of the multiplier 27 from the output signal of the delay circuit 28. That is, the subtractor 22 subtracts the signal μC ₀ from the multichannel signal X ₀ . As a result, a new multichannel signal X ₁ is generated.

In this way, the PAPR reduction circuit 20 executes X _i = X _i−1 −μC _i−1 in the iterative process. However, the power of the multichannel signal sampled in the sampling period S1 is lower than the threshold level. In this case, the clipping noise signal cn is always zero, and the signal μC is always zero. Accordingly, in the iterative process for the sampling period S1, the multichannel signal X _i output from the subtractor 22 remains the multichannel signal shown in FIG.

When the number of executions of the iterative process reaches N, the controller 30 gives an output instruction indicating the end of the iterative process to the output circuit 29. N is assumed to be designated in advance. N is, for example, about 10 to 20. When receiving the output instruction, the output circuit 29 reads and outputs the latest multi-channel signal X (in this example, the multi-channel signal X _N ). That is, in the sampling period S1, the PAPR reduction circuit 20 does not change the input multichannel signal. As described above, when the power of the input multichannel signal is lower than the threshold level, the tone signal is not added to the multichannel signal.

  When the sampling period S1 ends, the multichannel signal sampled in the sampling period S2 is guided to the switch 21. In this embodiment, as shown in FIG. 4, the power of the multichannel signal sampled in the sampling period S2 is also lower than the threshold level. Therefore, the PAPR reduction circuit 20 guides the multichannel signal to the E / O conversion circuit 4 without adding a tone signal to the multichannel signal.

  The power of the multichannel signal sampled in the sampling period S3 is higher than the threshold level as shown in FIG. In this case, when the above-described repetitive processing is executed by the PAPR reduction circuit 20, the power of the multichannel signal is reduced. At this time, a tone signal may be added to the multi-channel signal by the following processing.

The multi-channel signal sampled in the sampling period S3 is guided to the switch 21. The switch 21 selects the output signal of the combiner 3 (that is, the input multichannel signal) in the initial state of the iterative process. Then, subtractor 22, by subtracting the output signal of the multiplier 27 from the multi-channel signal, generating a multi-channel signal X _0. Accordingly, the clipping circuit 23, as a multi-channel signal X _0, based on the multi-channel signal is input. The multi-channel signal X ₀ is stored in the delay circuit 28.

The clipping circuit 23 compares the power of the input signal with a threshold level. Here, the power of the multi-channel signal sampled in the sampling period S3 is higher than the threshold level. In this case, the clipping noise signal cn ₀ corresponding to the difference between the power of the multichannel signal and the threshold level is generated by the clipping circuit 23 and the subtractor 24, and the signal μC ₀ is generated by the multipliers 26 and 27. The signal μC ₀ includes only the frequency component of the tone signal. The amplitude of the signal μC ₀ depends on the intensity of the clipping noise signal cn ₀ . Therefore, the signal μC ₀ substantially represents a tone signal having an amplitude that depends on the strength of the clipping noise signal cn ₀ . This signal μC ₀ is given to the subtractor 22.

Subsequently, an iterative process for the sampling period S3 is executed. That is, the PAPR reduction circuit 20 executes an iterative process “X _i = X _i−1 −μC _i−1 ”. For example, in the first calculation of the iterative process, a new multichannel signal X ₁ is generated by subtracting the signal μC ₀ from the multichannel signal X ₀ . Here, the signal μC ₀ represents a tone signal having an amplitude depending on the intensity of the clipping noise signal cn ₀ . Therefore, when the subchannel 22 generates the multichannel signal X ₁ from the multichannel signal X ₀ , the frequency component of the tone signal may be added to the multichannel signal. As a result, a multi-channel signal X ₁ including IF signals CH1 to CHn and a tone signal is generated. The multi-channel signal X ₁ is guided to the clipping circuit 23 and stored in the delay circuit 28. Note that when the subtractor 22 subtracts the signal μC from the multichannel signal X, it is considered that a part of the side lobes of each IF signal CH1 to CHn is removed.

In the second calculation of the iterative process, a new multi-channel signal X ₂ is generated by subtracting the signal μC ₁ from the multi-channel signal X ₁ . Here, the signal μC ₁ represents a tone signal having an amplitude depending on the intensity of the clipping noise signal cn ₁ . Therefore, even when the multi-channel signal X ₂ is generated from the multi-channel signal X _1, there is the frequency component of the tone signal to the multi-channel signal X ₁ is added.

However, the multi-channel signal X ₁ is obtained by subtracting the signal μC ₀ from the multi-channel signal X ₀ . Therefore, the power of the multichannel signal X ₁ is smaller than the power of the multichannel signal X ₀ . That is, the difference between the power of the multichannel signal X ₁ and the threshold level is smaller than the difference between the power of the multichannel signal X ₀ and the threshold level. Therefore, the strength of the clipping noise signal cn ₁ is smaller than the intensity of the clipping noise signal cn _0. Therefore, the amplitude of the tone signal of the signal μC ₁ is smaller than the amplitude of the tone signal of the signal μC ₀ .

Thereafter, the PAPR reduction circuit 20 repeatedly performs the process “X _i = X _i−1 −μC _i−1 ” until i = N. In this iterative process, a tone signal is added to the multichannel signal X. In this iterative process, the power of the multi-channel signal X gradually approaches the threshold level. Note that the speed at which the power of the multi-channel signal X gradually approaches the threshold level is specified by the parameter μ.

When N operations are completed in the repetitive processing, the PAPR reduction circuit 20 outputs the latest multi-channel signal X (multi-channel signal X _N ). That is, the PAPR reduction circuit 20 outputs the multichannel signal shown in FIG. In this case, the amplitude of the tone signal added to the multichannel signal depends on the difference between the power of the input multichannel signal to the PAPR reduction circuit 20 and the threshold level.

  Thus, the PAPR reduction circuit 20 reduces the peak power of the multichannel signal using the tone signal when the power of the multichannel signal is higher than the threshold level. At this time, the frequency component of the tone signal may be added to the multi-channel signal. Here, the tone signal is arranged at a frequency that does not overlap with the IF signals CH1 to CHn multiplexed in the multichannel signal. Therefore, the IF signals CH1 to CHn are not affected by the tone signal, and the quality of the IF signals CH1 to CHn is not deteriorated by the tone signal.

  The PAPR reduction circuit 20 is realized by a digital signal processor including a processor and a memory, for example. In this case, the PAPR reduction circuit 20 provides the above-described function by executing a given program. However, a part of the function of the PAPR reduction circuit 20 may be realized by a hardware circuit.

  FIG. 6 is a flowchart illustrating an example of processing of the PAPR reduction circuit 20. The process of this flowchart is executed every time the multi-channel signal output from the combiner 3 is sampled.

  In S1, the controller 30 initializes a variable i representing the number of repetitions to zero. In S <b> 2, the controller 30 controls the switch 21 to select the multichannel signal generated by the combiner 3. As a result, the multi-channel signal generated by the combiner 3 is selected by the switch 21 and guided to the subtractor 22. In the initial state of the iterative process, the clipping noise signal cn, the signal C, and the signal μC are zero.

In S3, the subtractor 22 calculates the multichannel signal X _i . The multichannel signal X _i is calculated by the following equation.
X _i = X _i-1 -μC _i-1
When the power of the input multichannel signal is higher than the threshold level, a tone signal is added to the multichannel signal and the peak power of the multichannel signal is reduced in this calculation.

In S4, the controller 30 determines whether or not the variable i has reached N. N is an integer specified in advance. In S5, the clipping circuit 23 and the subtractor 24 generate the clipping noise signal cn _i by comparing the power of the multichannel signal X _i with a predetermined threshold level. Note that the clipping noise signal cn _i is zero when the power of the input multi-channel signal (ie, the multi-channel signal X _{i at} i = 0) is lower than the threshold level.

In S6, the multiplier 26 generates a signal C _i by multiplying the clipping noise signal cn _i by the tone signal. The signal C _i includes only the frequency component of the tone signal. In S7, the multiplier 27 generates the signal μC _i by multiplying the signal C _i by the parameter μ. In S8, the controller 30 increments the variable i. Thereafter, the process of the PAPR reduction circuit 20 returns to S3.

  The PAPR reduction circuit 20 repeatedly executes the processes of S3 to S8 until the variable i reaches N. When the variable i reaches N, the output circuit 29 outputs the latest multi-channel signal X according to the output instruction given from the controller 30.

  As described above, when the power of the input multi-channel signal is higher than the threshold level, the tone signal is added to the multi-channel signal and the peak power of the multi-channel signal when the processes of S3 to S8 are executed N times. Is gradually reduced. On the other hand, when the power of the input multi-channel signal is lower than the threshold level, since the clipping noise signal cn is zero, the output signal of the PAPR reduction circuit 20 is the same as the input signal of the PAPR reduction circuit 20.

  FIG. 7 schematically shows signal processing of the PAPR reduction circuit 20. In this example, the multi-channel signal includes IF signals CH1 and CH2. Further, the power of the multi-channel signal is assumed to be higher than the threshold level set in the clipping circuit 23. In the following, the parameter μ is “1” in order to simplify the description.

In this case, the PAPR reduction circuit 20 generates a clipping noise signal cn ₀ representing the difference between the power of the input multi-channel signal and the threshold level in the first iterative process. The multiplier 26 generates a signal C ₀ . The signal C ₀ has only the frequency component of the tone signal. Here, the tone signal is arranged at a different frequency from the IF signals CH1 and CH2. However, the tone signal is allowed to overlap with the side lobes of the IF signals CH1 and CH2. The amplitude of the signal C ₀ is proportional to the intensity of the clipping noise signal cn ₀ . Then, the subtracter 22 subtracts the signal C ₀ from the input multi-channel signal (that is, the multi-channel signal X ₀ ), thereby generating the multi-channel signal X ₁ .

Compared with the input multi-channel signal, the side lobe intensity of the multi-channel signal X ₁ is reduced. Thus, towards the multi-channel signal X ₁ power it is lower than the power of the input multi-channel signal.

In the second iterative process, a clipping noise signal cn ₁ representing the difference between the power of the multichannel signal X ₁ and the threshold level is generated, and the signal C ₁ is generated by the multiplier 26. Like the signal C ₀ , the signal C ₁ has only the frequency component of the tone signal. However, since the input multi-channel signal towards a multi-channel signal X ₁ power is lower than the power, the strength of the clipping noise signal cn ₁ is lower than the clipping noise signal cn _0. For this reason, the amplitude of the signal C ₁ is smaller than that of the signal C ₀ . The subtractor 22 subtracts the signal C ₁ from the multi-channel signal X ₁ to generate a multi-channel signal X ₂ . Compared to multi-channel signal X _1, the intensity of the side lobes of a multi-channel signal X ₂ is further reduced.

  Thereafter, the above-described processing is repeatedly executed a predetermined number of times. By this iterative process, the side lobes of the IF signals CH1 and CH2 are gradually reduced, and the power of the multichannel signal is gradually reduced. That is, the peak-to-average power ratio of the multichannel signal is reduced. At this time, since the tone signal is arranged at a frequency different from that of the IF signals CH1 and CH2, the tone signal does not deteriorate the quality of the IF signals CH1 and CH2.

  FIG. 8 shows an example of a remote radio unit that receives an optical signal transmitted from a digital processing unit. In this embodiment, the remote radio unit 200 receives an optical signal transmitted from the digital processing unit 500 shown in FIG.

  As shown in FIG. 8, the remote radio unit 200 includes an O / E conversion circuit 41, oscillators 42-1 to 42-n, demodulators 43-1 to 43-n, and a low-pass filter (LPF) 44-1. 44-n, an oscillator 45, modulators 46-1 to 46-n, amplifiers 47-1 to 47-n, and antennas 48-1 to 48-n. O / E conversion circuit 41, oscillators 42-1 to 42-n, demodulators 43-1 to 43-n, modulators 46-1 to 46-n, amplifiers 47-1 to 47-n, and antenna shown in FIG. 48-1 to 48-n are respectively an O / E conversion circuit 11, an oscillator 13-1 to 13-n, a demodulator 12-1 to 12-n, a mixer 14-1 to 14-n, an amplifier shown in FIG. It corresponds to 15-1 to 15-n and antennas 16-1 to 16-n. The remote radio unit 200 may include other circuit elements not shown in FIG.

  The O / E conversion circuit 41 converts an optical signal received from the digital processing unit 500 via the optical fiber cable 300 into an electrical signal. This electrical signal represents the multi-channel signal shown in FIG. That is, this electrical signal includes IF signals CH1 to CHn. When the tone signal shown in FIG. 3B is added to the multi-channel signal in the digital processing unit 500, the output signal of the O / E conversion circuit 41 includes the tone signal in addition to the IF signals CH1 to CHn. . Note that the O / E conversion circuit 41 includes, for example, a photodiode.

  The oscillators 42-1 to 42-n generate oscillation signals IF1 to IFn, respectively. The frequencies of the oscillation signals IF1 to IFn generated by the transmitters 42-1 to 41-n are substantially the same as the oscillation signals generated by the oscillators 2-1 to 2-n of the digital processing unit 500, respectively. That is, the frequencies of the oscillation signals IF1 to IFn generated by the transmitters 42-1 to 42-n are the same as the carrier frequencies of the IF signals CH1 to CHn, respectively.

  Demodulators 43-1 to 43-n demodulate IF signals CH1 to CHn using oscillation signals IF1 to IFn generated by transmitters 42-1 to 42-n, respectively. The configurations of the demodulators 43-1 to 43-n are substantially the same. In this example, each demodulator 43-1 to 43-n includes a phase circuit 43p and mixers 43i and 43q.

  For example, in the demodulator 43-1, the phase circuit 43p generates a set of oscillation signals whose phases are different from each other by 90 degrees from the oscillation signal IF1 generated by the oscillator 42-1. The mixer 43 i generates an I component signal by multiplying the output signal of the O / E conversion circuit 41 by one oscillation signal. The mixer 43q generates a Q component signal by multiplying the output signal of the O / E conversion circuit 41 by the other oscillation signal. Here, the frequency of the oscillation signal IF1 generated by the oscillator 42-1 is the same as the carrier frequency of the IF signal CH1 multiplexed in the multichannel signal. Therefore, in the output signal of demodulator 43-1, data signal CH1 is arranged in the baseband as shown in FIG.

  The operations of the demodulators 43-1 to 43-n are substantially the same. Therefore, in the output signals of the demodulators 43-1 to 43-n, the corresponding data signals are arranged in the baseband, respectively. For example, in the output signal of the demodulator 43-n, as shown in FIG. 9B, the data signal CHn is arranged in the baseband.

  The low-pass filters 44-1 to 44-n remove high frequency components from the output signals of the demodulators 43-1 to 43-n, respectively. The cut-off frequency of each low-pass filter 44-1 to 44-n is designed to remove adjacent channels and tone signals. Therefore, for example, as shown in FIG. 9A, the low-pass filter 44-1 removes the data signals CH2 to CHn and the tone signal from the output signal of the demodulator 43-1 and passes the data signal CH1. . Further, as shown in FIG. 9B, the low-pass filter 44-n removes the data signals CH1 to CHn-1 and the tone signal from the output signal of the demodulator 43-n and passes the data signal CHn. .

  The oscillator 45 generates an oscillation signal RF having a predetermined radio frequency. The oscillation signal RF generated by the oscillator 45 is supplied to the modulators 46-1 to 46-n.

  The modulators 46-1 to 46-n modulate the oscillation signal RF with the data signals CH1 to CHn reproduced by the demodulators 43-1 to 43-n and the low-pass filters 44-1 to 44-n, respectively. To generate remodulated RF signals CH1 to CHn. In this embodiment, each of the modulators 46-1 to 46-n includes a phase circuit 46p, mixers 46i and 46q, and a combiner 46c. In this case, the phase circuit 46p generates a set of oscillation signals whose phases are different from each other by 90 degrees from the oscillation signal RF generated by the oscillator 45. The mixer 46i multiplies the I component of the corresponding data signal by one oscillation signal to generate the I component of the remodulated RF signal. Also, the mixer 46q generates the Q component of the remodulated RF signal by multiplying the Q component of the corresponding data signal by the other oscillation signal. Then, the combiner 46c adds the generated I component and Q component to generate a remodulated RF signal.

  The amplifiers 47-1 to 47-n amplify the remodulated RF signals CH1 to CHn, respectively. In this embodiment, each of the amplifiers 47-1 to 47-n is a high power amplifier. Remodulated RF signals CH1 to CHn amplified by the amplifiers 47-1 to 47-n are guided to antennas 48-1 to 48-n, respectively. Therefore, the remodulated RF signals CH1 to CHn are output via the antennas 48-1 to 48-n, respectively.

  In the remote radio unit 200 described above, the demodulators 43-1 to 43-n down-convert the received signals, and the low-pass filters 44-1 to 44-n pass the data signals of the target channels, respectively. Here, the tone signal is arranged at a different frequency from IF signals CH1 to CHn. Therefore, the low-pass filters 44-1 to 44-n can easily remove the tone signal added to the multichannel signal. That is, the remote radio unit 200 can remove the tone signal without degrading the waveform of the data signal of each channel. Thereafter, the demodulated data signal is up-converted to a radio frequency band and output. Therefore, the remodulated RF signals CH1 to CHn output via the antenna do not include a tone signal.

  FIG. 10 shows an example of the spectrum of a multi-channel signal generated in the digital processing unit. In this embodiment, IF signals CH1 and CH2 are multiplexed on the multichannel signal. The frequency interval between the IF signals CH1 and CH2 is 60 MHz. The horizontal axis represents the offset frequency, and the vertical axis represents the signal strength.

  The input multichannel signal is guided to the PAPR reduction circuit 20 as described above. When the power of the input multichannel signal is higher than the threshold level, a tone signal may be added to the multichannel signal by the iterative processing shown in FIG. 2 or FIG. The example shown in FIG. 10 represents the spectrum of a multi-channel signal obtained after 12 iterations have been performed.

  FIG. 11 shows simulation results for suppression of peak-to-average power ratio. The horizontal axis of the graph represents the peak-to-average power ratio (PAPR). The vertical axis represents a complementary cumulative distribution function (CCDF).

Characteristic A represents the PAPR for one IF signal. Characteristics BD represent PAPR of a multi-channel signal in which two IF signals are multiplexed. However, characteristic B represents PAPR when each IF signal is not clipped. A characteristic C represents PAPR when each IF signal is clipped at a predetermined power level. Characteristic D represents the PAPR when each IF signal is clipped at a predetermined power level and then the PAPR reduction circuit 20 performs the iterative process on the multichannel signal. The repeated process is executed 12 times. In this simulation, when the characteristic C and the characteristic D are compared, the PAPR reduction circuit 20 improves the PAPR by about 0.5 dB at, for example, CCDF = 10 ⁻⁶ .

  Thus, according to the configuration of the embodiment of the present invention, the peak-to-average power ratio of the multichannel signal is reduced. In addition to this, the tone signal used to reduce the power of the multi-channel signal is arranged at a different frequency from each IF signal multiplexed in the multi-channel signal. For this reason, the tone signal does not affect each IF signal, and the tone signal is easily removed at the receiver. Therefore, the characteristic of each IF signal is not deteriorated by the tone signal. For example, even if a tone signal is added to a multi-channel signal, the error vector amplitude (EVM) and adjacent channel leakage ratio (ACLR) of each IF signal do not increase.

  FIG. 12 shows an example of the change over time of the amplitude of the tone signal. In this example, five tone signals are used. As described above, the amplitude of the tone signal is adjusted by iterative processing according to the difference between the power of the input multichannel signal and the threshold level. For example, when the power of the input multichannel signal is higher than the threshold level and the difference between the power of the input multichannel signal and the threshold level is large, the amplitude of the tone signal is large.

<Layout of IF signal and tone signal>
As described above, in a communication system that transmits a multi-channel signal in which a plurality of data signals are multiplexed, the use of tone signals suppresses peak-to-average power without degrading the quality of each data signal. be able to. Hereinafter, arrangement of tone signals for appropriately suppressing peak-to-average power will be described. Further, depending on the frequency arrangement of each data signal, the peak-to-average power may not be appropriately suppressed using the tone signal. Therefore, the arrangement of data signals is also examined.

FIG. 13 is a diagram for explaining cross-phase modulation distortion of a multichannel signal. In this example, it is assumed that a multi-channel signal including IF signals CH1 and CH2 shown in FIG. The carrier frequencies of IF signals CH1 and CH2 are f ₁ and f ₂ , respectively.

  It is assumed that IF signals CH1 and CH2 are transmitted from the digital processing unit 500 shown in FIG. 2 to the remote radio unit. In this case, the IF signals CH 1 and CH 2 are combined by the combiner 3 and converted into an optical signal by the E / O conversion circuit 4.

  When the power of the input signal (the combined IF signals CH1 and CH2) of the E / O conversion circuit 4 is smaller than a predetermined threshold value, the power of the output signal is equal to that of the input signal as shown in FIG. Proportional to power. In this case, Inter Modulation Distortion (IMD) does not occur or is sufficiently small.

  However, when the power of the input signal is large, an intermodulation distortion component can be generated. That is, when the voltage of the input signal increases to the non-linear region, an intermodulation distortion component is generated as shown in FIG. In the following description, the n-th order intermodulation distortion is expressed as “IMDn”. For example, IMD3 represents third-order intermodulation distortion.

When the carrier frequencies of the IF signals CH1 and CH2 are f ₁ and f ₂ , respectively, the frequencies of the IMD 3 are | 2f ₁ −f ₂ | and | 2f ₂ −f ₁ |. Here, when receiving a signal at a receiving station (for example, a remote radio unit), it is preferable that the IMD 3 is suppressed or removed. Therefore, the digital processing unit 500 uses the PAPR reduction circuit 20 shown in FIG. 2 to process the input signal so that IMD3 is suppressed. Here, the PAPR reduction circuit 20 subtracts the frequency component of the tone signal from the input signal. Therefore, if the tone signal is arranged at the same frequency as that of IMD 3, IMD 3 is suppressed in PAPR reduction circuit 20. That is, if the PAPR reduction circuit 20 processes the input signal using the tone signal having the frequency | 2f ₁ −f ₂ | and the tone signal having the frequency | 2f ₂ −f ₁ |, a multi-channel signal in which IMD3 is suppressed is generated. Is done.

Note that when the power of the input signal is large, second-order distortion components can also be generated. As shown in FIG. 13A, the second-order distortion component occurs at frequencies | f ₁ −f ₂ |, 2f ₁ , f ₁ + f ₂ , and 2f ₂ . However, the frequency of the second-order distortion component is generally far away from the carrier frequency of the main signal (here, IF signals CH1 and CH2). That is, the second-order distortion component can be easily removed using a bandpass filter in the receiver. Therefore, the digital processing unit 500 does not need to generate a tone signal for suppressing the second-order distortion component.

  Further, when the power of the input signal is large, distortion components of the fourth order or higher can be generated. Here, when compared with IMD3, the fourth-order or higher-order distortion component is small. Accordingly, the digital processing unit 500 may generate a tone signal for suppressing a fourth-order or higher distortion component. However, the digital processing unit 500 may generate a tone signal for suppressing a fourth-order or higher-order distortion component as necessary.

However, depending on the arrangement of IF signals multiplexed in a multi-channel signal, a tone signal may not be arranged at the same frequency as IMD3. For example, it is assumed that a multi-channel signal including main signals F1, F2, and F3 shown in FIG. Here, to simplify the description, it is assumed that the frequency of the main signal F2 is twice the frequency of the main signal F1, and the frequency of the main signal F3 is three times the frequency of the main signal F1. That is, 2F1 = F2 and 3F1 = F3. In this case, six IMD3 components (IMD3_A to IMD3_F) are generated. The frequency of each IMD3 component is as follows.
IMD3_A: 2F1-F3
IMD3_B: 2F1-F2
IMD3_C: 2F2-F3
IMD3_D: 2F2-F1
IMD3_E: 2F3-F2
IMD3_F: 2F3-F1

  In order to suppress these IMD3 components, the PAPR reduction circuit 20 may arrange tone signals for the respective IMD3 components. However, the frequency of IMD3_C is the same as the frequency of main signal F1, and the frequency of IMD3_D is the same as the frequency of main signal F3. For this reason, when the tone signal for suppressing IMD3_C is generated, not only IMD3_C but also the main signal F1 is suppressed. Similarly, when a tone signal for suppressing IMD3_D is generated, not only IMD3_D but also the main signal F3 is suppressed.

  Therefore, the digital processing unit 500 has a function of controlling the oscillation frequency of the oscillators 2-1 to 2-n and a function of controlling the frequency of the tone signal, as shown in FIG. These functions are realized by the controller 30. Further, the oscillation frequency of the oscillators 2-1 to 2-n and the frequency of the tone signal are determined before data communication is started, for example.

  FIG. 16 is a flowchart illustrating an example of processing for determining the arrangement of IF signals and tone signals. In this example, a multi-channel signal including N IF signals is transmitted. N is an integer of 3 or more. 16 may be executed in the digital processing unit 500, or may be executed by a computer connected to the digital processing unit 500. In the following description, it is assumed that the processing of the flowchart shown in FIG.

  In S11, the controller 30 determines each frequency F (k) of N IF carriers used for data transmission. N IF carriers are arranged at a constant frequency interval, for example. Thereafter, the controller 30 executes the processes of S12 to S16 for all combinations of the variables k (k = 1 to N), n (n = 1 to N), and m (m = 1 to N).

In S12, the controller 30 compares the frequency of IMD3 (n, m) with the IF carrier frequency F (k). The frequency of IMD3 (n, m) is expressed by the following equation. However, n and m are different from each other.
IMD3 (n, m) = 2F (n) -F (m)

  If the frequency of the IMD3 (n, m) and the IF carrier frequency F (k) are different, the controller 30 performs the processing of S12 to S16 for all combinations of the variables k, n, and m in S13. Determine if it is finished. If there remains a combination for which the processing of S12 to S16 has not been completed, the controller 30 selects the next combination in S14. Thereafter, the process of the controller 30 returns to S12.

  When the frequency of IMD3 (n, m) matches the IF carrier frequency F (k), the controller 30 shifts the IF carrier frequency F (k) by ΔF in S15. ΔF is preferably sufficiently smaller than the frequency interval at which the IF signal is arranged. Further, ΔF is determined so that the frequency F (k) + ΔF is sufficiently removed when the frequency F (k) is passed through the filter (LPF 44-1 to 44-n in FIG. 8) provided in the receiver. It is preferable. In S16, the controller 30 selects the next combination. Thereafter, the process of the controller 30 returns to S12.

  After the processing of S12 to S16 is completed for all combinations, the controller 30 determines the arrangement of tone signals in S17. Specifically, the tone signal is arranged so as to match the frequency of each IMD 3.

The processing of the flowchart shown in FIG. 16 will be described with reference to the embodiment shown in FIG. In this embodiment, it is assumed that IF carrier frequencies F (1) to F (3) are set in S11 as follows.
F (1) = 60MHz
F (2) = 120MHz
F (3) = 180MHz

When k = 3, n = 3, and m = 1 are selected, the following calculation is executed in S12.
F (k = 3) = 180MHz
IMD3 (3,1) = 360MHz-60MHz = 300MHz
That is, the IF carrier frequency F (3) does not match the frequency of IMD3 (3,1). In this case, the controller 30 does not change the IF carrier frequency F (3).

When k = 3, n = 3, and m = 2 are selected, the following calculation is executed in S12.
F (k = 3) = 180MHz
IMD3 (3,2) = 360MHz-120MHz = 240MHz
That is, the IF carrier frequency F (3) does not match the frequency of IMD3 (3,2). In this case, the controller 30 does not change the IF carrier frequency F (3).

When k = 3, n = 2, and m = 1 are selected, the following calculation is executed in S12.
F (k = 3) = 180MHz
IMD3 (2,1) = 240MHz-60MHz = 180MHz
That is, the IF carrier frequency F (3) matches the frequency of IMD3 (2,1). In this case, the controller 30 shifts the IF carrier frequency F (3) by ΔF. In this embodiment, ΔF is 30 MHz. Then, IF carrier frequencies F (1) to F (3) are rearranged as follows.
F (1) = 60MHz
F (2) = 120MHz
F (3) = 210M

  When IF carriers F1 to F3 are rearranged, as shown in FIG. 14B, the frequency of each IF carrier does not match any of the IMD3 components. Therefore, the process of the controller 30 proceeds to S17. In S17, a tone signal is assigned to each IMD3 component.

  Thereafter, the controller 30 gives a frequency instruction to the oscillators 2-1 to 2-n. This frequency instruction controls the oscillation frequency of the oscillators 2-1 to 2-n. In the above example, the frequencies of the three oscillators among the oscillators 2-1 to 2-n are controlled to 60 MHz, 120 MHz, and 210 MHz, respectively.

  Further, the controller 30 provides arrangement information to the tone signal generator 25. This arrangement information represents the arrangement of tone signals. In the above example, the arrangement information representing the six tone signals corresponding to IMD3_A to IMD3_F shown in FIG.

  The PAPR reduction circuit 20 processes the input signal according to the above settings. Specifically, when the power of the input signal is larger than the threshold level of the clipping circuit 23, the tone signal component is subtracted from the input signal. As a result, each IMD3 component is suppressed in the process of reducing the peak-to-average power ratio of the input signal.

  17 to 18 show an example of another method for arranging IF signals and tone signals. In this example, a multi-channel signal including two IF signals Ch1 and Ch2 is transmitted. The difference between the carrier frequencies of the IF signals Ch1 and Ch2 is Fs / 2 as shown in FIG. The frequency Fs represents the sampling frequency of the digital processing unit 500. The broken arrow represents a tone signal.

  When the input power of the multi-channel signal is larger than the threshold level, IMD3 is generated as described above. Here, it is assumed that the IMD3 component overlaps with the IF signals Ch1 and Ch2, as shown in FIG. In this case, the PAPR reduction circuit 20 cannot suppress the IMD3 component using the tone signal. Therefore, the PAPR reduction circuit 20 shifts the carrier frequencies of the IF signals Ch1 and Ch2 so that the IMD3 component can be suppressed using the tone signal.

  The sampling frequency Fs is 30.72 MHz. That is, the difference between the carrier frequencies of the IF signals Ch1 and Ch2 is 61.44 MHz. Further, the tone signal is arranged approximately in the middle of the IF signals Ch1 and Ch2.

  As shown in FIGS. 18A to 18C, when the frequency interval between the IF signals Ch1 and Ch2 is widened, the frequency of the IMD3 component approaches the intermediate frequency of the IF signals Ch1 and Ch2. That is, when the frequency interval between the IF signals Ch1 and Ch2 is widened, the frequency of the IMD3 component approaches the frequency of the tone signal. Here, when the frequency of the IMD3 component matches the frequency of the tone signal, the PAMD reduction circuit 20 suppresses the IMD3 component. Therefore, the PAPR reduction circuit 20 determines the carrier frequencies of the IF signals Ch1 and Ch2 so that the frequency of the IMD3 component substantially matches the frequency of the tone signal. In this example, as shown in FIG. 18C, the carrier frequency of the IF signal Ch1 is shifted by about 19 MHz, and the carrier frequency of the IF signal Ch2 is shifted by about 19 MHz.

FIG. 19 shows an example of a simulation result obtained by calculating the PAPR for the IF signal arrangement. In this simulation, as shown in FIGS. 17 to 18, a plurality of tone signals are arranged at substantially the intermediate frequency of the IF signals Ch1 and Ch2. The horizontal axis of the graph shown in FIG. 19 represents the frequency interval between IF signals Ch1 and Ch2. The vertical axis represents the peak-to-average power ratio (PAPR). Characteristics X, Y and Z represent the peak-to-average power ratio obtained for CCDF = 10 ⁻⁴ , CCDF = 10 ⁻⁶ and CCDF = 10 ⁻⁷ , respectively. CCDF represents a complementary cumulative distribution function.

  In this simulation, the peak-to-average power ratio is minimized when the frequency interval between the IF signals Ch1 and Ch2 is set to about 100 MHz. At this time, as shown in FIG. 18C, the frequency of the IMD3 component substantially matches the frequency of the tone signal. That is, it is considered that the peak-to-average power ratio is reduced by suppressing the IMD3 component by the tone signal.

In the above-described embodiment, the IMD3 component is suppressed, but the PAPR reduction circuit 20 may suppress the fourth-order or higher-order intermodulation distortion component. For example, it is assumed that the carrier frequencies of N IF signals included in the multichannel signal are f1 to fN. In this case, an intermodulation distortion component can be generated at the following frequencies.
｜ K × fn−L × fm |
K = 2, 4, 6,. . .
L = 1, 3, 5,. . .
n = 1 to N
m = 1 to N
When K = 2 and L = 1 are given, the frequency of the IMD3 component is obtained.

  The PAPR reduction circuit 20 arranges tone signals at frequencies at which intermodulation distortion components are generated. Thereby, each intermodulation distortion component is suppressed, and the peak-to-average power ratio is also suppressed. At this time, the PAPR reduction circuit 20 does not place a tone signal at the carrier frequency of the IF signal. Therefore, when the carrier frequency of the IF signal matches the frequency at which the intermodulation distortion component is generated, the carrier frequency of the IF signal is shifted. When the frequency of the tone signal is fixedly determined in advance, the PAPR reduction circuit 20 does not match the carrier frequency of each IF signal with the frequency at which the intermodulation distortion component is generated, and the IF signal The carrier frequency of each IF signal may be adjusted so that the carrier frequency does not match the frequency of the tone signal. Further, the PAPR reduction circuit 20 may arrange the tone signal only at the frequency at which the intermodulation distortion component is generated.

1-1 to 1-n Modulator 2-1 to 2-n Oscillator 3 Combiner 4 E / O conversion circuit 20 PAPR reduction circuit 21 Switch 22 Subtractor 23 Clipping circuit 24 Subtractor 25 Tone signal generators 26 and 27 Multipliers 30 Controller 44-1 to 44-n Low-pass filter (LPF)
200 Remote wireless unit 500 Digital processing unit

Claims (11)

A signal generation circuit for generating a plurality of intermediate frequency signals;
A combiner that combines the plurality of intermediate frequency signals to generate a multi-channel signal;
A reduction unit for reducing a peak-to-average power ratio of the multi-channel signal;
A converter that converts a multi-channel signal having a peak-to-average power ratio reduced by the reduction unit into an optical signal, and
The reduction unit reduces the power of the multichannel signal using an auxiliary signal having a frequency different from any of the plurality of intermediate frequency signals when the power of the multichannel signal is higher than a predetermined threshold level. A communication device. The reduction unit reduces the power of the multichannel signal by subtracting the auxiliary signal from the multichannel signal in the frequency domain when the power of the multichannel signal is higher than the threshold level. The communication device according to claim 1. When the power of the multi-channel signal is higher than the threshold level, the reduction unit repeatedly performs a process of reducing the power of the multi-channel signal using the auxiliary signal, thereby reducing the power of the multi-channel signal. The communication device according to claim 1, wherein the communication device gradually approaches the threshold level. The communication device according to claim 1, wherein the reduction unit controls the amplitude of the auxiliary signal based on a difference between the power of the multi-channel signal and the threshold level. When the multi-channel signal includes a first intermediate frequency signal of a first carrier frequency and a second intermediate frequency signal of a second carrier frequency, the auxiliary signal includes the first carrier frequency and the second carrier frequency. The communication apparatus according to claim 1, wherein the communication apparatus is disposed between the carrier frequency and the carrier frequency. When the multi-channel signal includes a first intermediate frequency signal of a first carrier frequency and a second intermediate frequency signal of a second carrier frequency higher than the first carrier frequency, the auxiliary signal is Arranged in a low frequency side frequency band of one carrier frequency, a frequency band between the first carrier wave number signal and the second carrier frequency signal, and a high frequency side frequency band of the second carrier frequency The communication device according to claim 1, wherein: The reduction unit is
An auxiliary signal generator for generating the auxiliary signal;
A difference calculation unit for calculating a difference between the power of the input signal and the threshold level;
A multiplier for multiplying the auxiliary signal by the difference;
A subtractor for giving a new multi-channel signal obtained by subtracting the output signal of the multiplication circuit from the multi-channel signal to the difference calculation unit,
The difference calculation unit, the multiplier circuit, and the subtractor repeatedly execute a process of subtracting the output signal of the multiplier circuit from the multichannel signal a predetermined number of times when the power of the multichannel signal is higher than the threshold level. The communication apparatus according to claim 1, wherein power of the multi-channel signal is reduced. The signal generation circuit adjusts the carrier frequency of each intermediate frequency signal so that the frequency of the intermodulation distortion component of the plurality of intermediate frequency signals does not match the carrier frequency of each intermediate frequency signal,
The communication device according to claim 1, wherein the reduction unit arranges the auxiliary signal at a frequency of the intermodulation distortion component. The communication device according to claim 8, wherein the reduction unit arranges the auxiliary signal so as not to match any carrier frequency of the plurality of intermediate frequency signals. A communication system including a communication device and a remote device for receiving an optical signal generated by the communication device,
The communication device
A signal generation circuit for generating a plurality of intermediate frequency signals from a plurality of baseband signals;
A combiner that combines the plurality of intermediate frequency signals to generate a multi-channel signal;
A reduction unit for reducing a peak-to-average power ratio of the multi-channel signal;
An E / O converter that converts a multi-channel signal having a peak-to-average power ratio reduced by the reduction unit into an optical signal;
The reduction unit reduces the power of the multichannel signal by using an auxiliary signal having a frequency different from any of the plurality of intermediate frequency signals when the power of the multichannel signal is higher than a predetermined threshold level.
The remote device is:
An O / E converter that converts the optical signal into an electrical signal;
A demodulation circuit for reproducing the plurality of baseband signals from the electrical signal;
A plurality of filters that respectively remove frequency components of the auxiliary signal from a plurality of baseband signals reproduced by the demodulation circuit;
An up-conversion circuit that up-converts a plurality of baseband signals output from the plurality of filters to generate a plurality of radio frequency signals. Generate multiple intermediate frequency signals,
Combining the plurality of intermediate frequency signals to generate a multi-channel signal;
When the power of the multi-channel signal is higher than a predetermined threshold level, the power of the multi-channel signal is reduced using an auxiliary signal having a frequency different from any of the plurality of intermediate frequency signals,
Convert multi-channel signals with reduced power into optical signals,
Transmitting the optical signal to a remote device.

Images