JP2002300132A - Method and instrument for measuring nonlinear characteristic of amplifier - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and an instrument for measuring a nonlinear characteristic of an amplifier, that can measure both amplitude and phase change characteristics of the amplifier only from an output signal of the ampli fier when the amplifier amplifies an OFDM signal. SOLUTION: In the method and instrument for measuring the nonlinear characteristic of the amplifier in the OFDM(Orthogonal Frequency Division Multiplex) transmission, the nonlinear characteristic is acquired by comparing the amplitude and the phase of a signal, which is obtained by applying Fourier transform to a multi-carrier signal and also applying inverse Fourier transform to the resulting signal, with those of a signal, which is obtained by applying correction with respect to signal space diagram to a signal resulting from applying inverse Fourier transform to the multi-carrier signal. Furthermore, averaging pilot signals obtained through the processing of a plurality of symbols is used as a pilot for the application of the correction.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and a device for measuring nonlinear characteristics of an amplifier used for orthogonal frequency division multiplexing transmission performed in digital terrestrial television broadcasting.

[0002]

2. Description of the Related Art In the transmission of terrestrial television broadcasts, particularly in an urban area where high-rise buildings are in a forest, interference (multipath interference) due to radio waves that are reflected from buildings and arrive with a delay is a serious problem. This disturbance is a well-known problem as a ghost for broadcast waves based on analog modulation. In a broadcast wave by digital modulation, a symbol that should have arrived in the past is delayed with respect to a symbol being received by a receiver, resulting in intersymbol interference. The delay time that the jammer received is compared to the length of one symbol,
For example, if it is a very small value such as one tenth,
The effect of such intersymbol interference is small. That is,
If the same data is transmitted, the influence of inter-symbol interference can be reduced by using a scheme in which the symbol length is as large as possible.

One of such systems is OFDM (Orthogonal
Frequency Division Multiplexing). this is,
This is a method in which the transfer of data is shared by multiplexing using a plurality of carriers and the symbol length is increased, so that interference due to delayed arriving waves can be reduced. Due to such properties, it has been adopted for the terrestrial digital television broadcasting system in Japan.

In the OFDM system, hundreds to thousands of digitally modulated waves are transmitted at regular intervals on the frequency axis. this,
Compared to general FDM (Frequency Division Multiplexing), there is a difference in the spectrum interval of the modulated wave.
In a general FDM system, the spectrums of the modulated waves are arranged at frequency intervals so as not to overlap with each other, and it is necessary to use a filter for extracting the individual modulated waves on the receiving side. However, in the OFDM system, the frequency interval is set to the reciprocal of the symbol length, and all symbol timings are matched. Under this condition, multiplexing is performed so that a part of the spectrum overlaps, but it can be completely separated on the receiving side. Thus, in normal FDM,
A filter is required both before multiplexing and at the time of reception, but is not used in OFDM.

Further, in practice, the symbol length of the OFDM signal is often set to be longer than the length for reducing the influence of the intersymbol interference described above. This excess portion is called a guard interval, and has a configuration in which the influence of intersymbol interference during demodulation can be suppressed even if an interfering wave whose delay time is shorter than this length is mixed.

The generation of a signal is performed, for example, as shown in FIG. First, the input digital data is accumulated by the number of bits transmitted in one symbol and assigned to each carrier. In other words, PSK (Phase Shift Keying) or QAM (Quadrature
Complex signal point data mapped in the same manner as in Amplitude Modulation) is generated. At this time, based on a predetermined rule, known complex signal point data is assigned to some carriers to a receiving side serving as a pilot (a reference point in demodulation). The above data group is subjected to inverse Fourier transform, and generation and synthesis of individual carriers are performed simultaneously. The output signal is a signal corresponding to the time-series complex data of OFDM. When a guard interval is added, an operation of adding a copy of the last part of the symbol to the beginning of the symbol is performed.
The complex data is subjected to frequency conversion by means such as DA conversion and quadrature modulation in a predetermined combination, and is converted to a center frequency used for broadcasting. Transmission is realized by amplifying the converted signal and supplying it to the antenna.

The principle of reception will be described below with reference to FIG. After appropriate amplification, the signal input from the antenna is subjected to frequency conversion and AD conversion in a predetermined combination, and is converted into digital complex data of a predetermined frequency. Here, the subsequent FFT
Fine adjustment of the sampling frequency and the center frequency needs to be performed by a predetermined method so as not to hinder the processing by (Fast Fourier Transform). Also used for demodulation
A section for one symbol is cut out by a predetermined method. The extracted data is input to the FFT, and the output of the FFT becomes complex signal point data of each carrier. A pilot is extracted from these data and processed, the reference of the amplitude and phase of each carrier is calculated, and the position in the complex signal point space is corrected. Digital data corresponding to the corrected position is found, and these data are rearranged in a procedure reverse to that at the time of modulation, whereby a bit string for one symbol can be obtained.

[0008] The probability density distribution of the instantaneous power of this orthogonal frequency division multiplexing (OFDM) signal substantially follows a distribution generally known as a ^{２ 2} distribution with ^two degrees of freedom, and is between the average power and the maximum value of the instantaneous power. It is known that a difference of several tens of times occurs. For this reason, it is also known that in the broadcasting of the OFDM system, a transmission device having a linear amplification degree up to several tens times the transmission power generally represented by the average power is required.

[0009] If a device having a sufficient margin cannot be used as described above, distortion is affected by the nonlinearity of the device in a portion where the power is large. As a simple example of nonlinearity, it is assumed that a third-order intermodulation product has occurred between multicarrier inputs, which is a feature of the OFDM system. For example, 3 between two carriers whose frequencies are f ₁ and f ₂
As the next intermodulation product, the frequency is 2 · f ₁ −f ₂ or 2 · f ₂
A component that becomes −f ₁ is generated. If other carriers exist at these frequencies, they become interference signals and cause transmission errors. An actual amplifier often shows a more complicated behavior including phase rotation due to nonlinearity.

Therefore, in order to avoid the transmission error as described above, it is necessary to grasp the non-linearity of the amplifier, and a measuring means therefor is indispensable.

As a conventional method for measuring these characteristics, a method of inputting a measurement signal reflecting the statistical properties of an OFDM signal, and simultaneously measuring and comparing the output signal is disclosed in Reference 1 (Nagatsuka et al. Degradation of OFDM Signal by Amplifier and Measurement Method of Nonlinear Characteristics "Journal of the Institute of Image Information and Television Engineers 1999
November issue, p.1550-1556).

Here, two types of measurement methods are described. For the purpose of measuring a region where the instantaneous power is larger than the average power, a signal generated so that the envelope becomes a SINC function is inputted. Focusing on the measuring method, there are the following three disadvantages. Disadvantage 1) Cannot be measured during operation. Disadvantage 2) A device for generating a special signal is required. Disadvantage 3) It is necessary to measure both the input side and the output side and then match the time relationship between the measured data.

A method for correcting nonlinear distortion using an OFDM signal is described in Japanese Patent Application Laid-Open No. 2000-22659. In the configuration in the measurement method described here, OF
The DM modulator includes quadrature demodulation means for demodulating a transmission signal after passing through a high-power amplifier having nonlinear characteristics. Further, the demodulated signal output from the quadrature demodulation means and the output of the pre-distortion circuit for pre-distorting the transmission signal are provided to the modeling means. Then, the modeling means models the nonlinear characteristics of the high-power amplifier based on the given signal, and reflects the modeling result on the predistortion circuit. That is, in order to favorably remove nonlinear distortion from a transmission signal, a non-linear characteristic that changes every moment is dynamically obtained, and predistortion is performed according to the nonlinear characteristic.

Since this method uses a signal being broadcast, the above disadvantages 1 and 2 can be overcome. However, it is still the same as measuring both input and output signals, and the above disadvantage 3 cannot be solved. (In the configuration described in Japanese Patent Application Laid-Open No. 2000-22659, a measurement function is included in the modulator. For this reason, the output is apparently measured only. Considering that it goes out of the modulator, the disadvantage 3 is found to occur again). Further, when measuring in an actual operation state, since different signals are always input to the amplifier, it is necessary to perform simultaneous simultaneous measurement of input and output. In this case, it is impossible to use a single system of measurement devices for input and output.

[0015]

In the conventional method and apparatus for measuring nonlinear characteristics of an amplifier, as described above, both the input side and the output side are measured, and the time relationship between the measured data is matched. Operation was required.

The present invention has been proposed in view of the above,
An object of the present invention is to provide a method and a device for measuring nonlinear characteristics of an amplifier, which can be measured only from an output signal during an operation of amplifying an OFDM signal.

[0017]

To achieve the above object, a first aspect of the present invention is a method for measuring the characteristics of an amplifier in orthogonal frequency division multiplexing transmission, which converts a multicarrier signal into first and second multicarrier signals. A signal obtained by dividing the first multicarrier signal through the first digital signal processing means and a signal obtained by passing the second multicarrier signal through the second digital signal processing means; It is characterized in that a nonlinear characteristic is obtained by comparing each amplitude with each phase.

According to a second aspect of the present invention, in addition to the method of the first aspect, the first digital signal processing means is means for removing noise, and the second digital signal processing means is a Fourier signal. It is characterized by being a digital processing means for adding a correction relating to a signal space diagram to a signal obtained by the conversion and further performing an inverse Fourier transform.

According to a third aspect of the present invention, there is provided a method for measuring the characteristics of an amplifier in orthogonal frequency division multiplex transmission, wherein the multicarrier signal is converted into first and second multicarrier signals. Divided into
The amplitude and the phase of each of a signal obtained by passing the first multicarrier signal through a band-limiting filter and a signal obtained by passing the second multicarrier signal through a filter including a Fourier transform are defined as follows. It is characterized in that a non-linear characteristic is obtained by comparison.

According to a fourth aspect of the present invention, in addition to the method of the third aspect, the Fourier transform filter further performs a Fourier inverse transform on the signal obtained by adding a correction relating to the signal space diagram to the signal obtained by the Fourier transform. The filter is characterized by applying

In the second or fourth invention,
It is necessary to provide a method for correcting a signal space diagram to a signal obtained by performing a Fourier transform on a multicarrier signal. In the present invention, the following method is used to solve the problem by the fifth invention which generates a reliable pilot. are doing.

According to a fifth aspect of the present invention, in addition to the configuration of the second or fourth aspect described above, a pilot obtained by processing a plurality of symbols is used as a pilot for use in performing the above-described correction on the signal space diagram. Is characterized by using an average of

According to a sixth aspect of the present invention, there is provided an apparatus for measuring the characteristics of an amplifier in orthogonal frequency division multiplex transmission, wherein the multicarrier signal is converted into first and second multicarrier signals. Means for dividing the first multi-carrier signal into means for limiting the frequency band to obtain a first signal; and means for passing the second multi-carrier signal through digital signal processing means including Fourier transform. 2 and a means for obtaining nonlinear characteristics by comparing the amplitude and the phase of each of the first signal and the second signal.

According to a seventh aspect of the present invention, in addition to the configuration of the sixth aspect, the digital signal processing means including the Fourier transform corrects a signal obtained by performing the Fourier transform with respect to a signal space diagram. It is a digital signal processing means for performing inverse Fourier transform.

According to an eighth aspect of the present invention, in addition to the configuration of the seventh aspect of the present invention, there is provided means for generating a pilot for use in making a correction on the signal space diagram, wherein the means comprises a plurality of symbols. And means for averaging and generating the result obtained by the above processing.

[0026]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The outline of the present invention will be described below,
Next, an embodiment will be described. An object of the present invention is to establish a method for estimating an input signal before being subjected to nonlinear distortion by an amplifier from a measured output signal. Since the signal to be measured is a signal immediately before transmission, it is assumed that the signal to be measured has a quality that enables demodulation without bit error even in a state where no error correction code is used.

First, if a technique similar to that of a general demodulator is used, an effective symbol section is cut out from a signal under measurement and processed to obtain a complex signal point for each carrier.
This complex signal point is shifted from its original position due to nonlinear distortion. However, since the quality is such that a bit error does not occur, the closest original signal point can be considered as a correct signal point before being subjected to nonlinear distortion. In addition, if the signal uses an error correction code, the error correction code is solved, and if the signal re-encoded is regarded as a correct signal point, more reliable signal point determination can be performed.

Using the complex signal points from which the influence of nonlinear distortion has been removed in this way, OFDM modulation is performed again to generate a signal. At this time, using the amplitude and phase reference for each carrier obtained from the pilot at the time of demodulation, each carrier of the remodulated OFDM signal has the same amplitude and phase reference as the signal obtained by the measurement. Convert. The signal obtained as a result of the above processing is obtained by estimating the signal waveform before the amplifier receives nonlinear distortion.

The problem with the above procedure is that the pilot inserted to obtain the reference of the amplitude and phase of each carrier is also affected by the nonlinear distortion. Looking at the effect of nonlinear distortion on a carrier basis,
If the number of carriers exceeds 1,000, the average is 0
Can be treated in the same way as additive Gaussian noise (see Document 1 above). Therefore, the effect of nonlinear distortion can be reduced by collecting pilots in many symbols and performing averaging between symbols for each carrier. This averaging process needs to be performed for all carriers where pilots exist. Therefore, in the following description, it is assumed that the influence of the nonlinear distortion is reduced by such a method for the pilot.

The carrier frequency is ω ₀ , and r (t)
And ψ (t) are the envelope amplitude and phase angle of the modulated signal, respectively. At this time, the input signal is expressed as x (t) = r (t) cos (ω ₀ t + ψ (t)). Correspondingly, the output signal is y (t) = A (t) cos (ω ₀ t + ψ (t) + Φ
(T)),. At this time, r (t) and A
AM-A describes the time-independent relationship of (t).
The AM-PM conversion characteristic describes the M-conversion characteristic and describes the time-independent relationship between r (t) and Φ (t). The non-linear characteristics of the amplifier are based on these AM-AM conversion characteristics and A
What can be understood from the M-PM conversion characteristics is that
well known.

Next, an embodiment will be described with reference to FIGS. Here, the Japanese terrestrial digital television broadcasting system that adopted the OFDM system (Telecommunications Technology Council report “Technical conditions of terrestrial digital television broadcasting system”
An example of implementing this measurement method using a signal based on May 24, 1999) will be described. Also, here, the broadcast mode Mode
Signal with 13 segments used in 3 (5617 carriers)
Book).

FIG. 1 is a diagram showing a preferred configuration of the present invention, and is a diagram showing a signal flow in a block diagram. An OFDM signal to be measured is input to the signal input portion of FIG. In the AD conversion / frequency correction unit, the input analog signal is frequency-converted into an appropriate frequency band and then AD (analog-digital) converted. In the AD conversion, the sampling frequency is determined so that the effective symbol length of the input signal corresponds to 65536 sample points. The number of sample points per effective symbol length must be a power of 2 in order to use the FFT (Fast Fourier Transform) algorithm. If the number of carriers is 5617,
In normal demodulation, 8192 points are sufficient, but if an OFDM signal is subjected to nonlinear distortion, its frequency bandwidth is widened, and therefore it is necessary to set a higher sampling frequency.

As shown in this example, 6553 per effective symbol length
The sampling frequency for sampling six points can be obtained as follows. First, carrier interval = bandwidth per segment (6/14 MHz) / number of carriers per segment (432) ≒ 0.99 kHz, and 1
The symbol length is the reciprocal of the carrier interval. This one
Since sampling is performed at 65536 points during the symbol length, sampling is performed at a carrier interval × 65536 points per second. Therefore, the sampling frequency is approximately 0.99 kHz × 65536/65 MHz. Note that reproduction of a signal having a sampling frequency can be realized by a general method already developed for a demodulator, which is described in, for example, Reference 2 (Kageyama, “Elemental Technology of Digital Terrestrial Television Broadcasting Receivers”). About "1998 Annual Conference of the Institute of Image Information and Television Engineers, S3-3), utilizing the fact that the guard interval signal is a cyclic copy of the signal after the effective symbol period, A block diagram of a method for simultaneously estimating a frequency error and a carrier frequency error within a carrier frequency interval is described.

The center frequency of the OFDM signal at this time is represented by f _C
Then the sampled digital data is exp
(-j2πf _C t), and converted to complex data. This complex data contains O
An FDM signal component and its accompanying signal, where the same signal is inverted on the frequency axis, and the center carrier frequency is -2f _C
Is generated. The reproduction of the center frequency can be realized by a well-known method described in Document 2, for example.

Of the two frequency components, only the one whose center frequency is 0 is used. In order to easily separate these two components, it is necessary to determine the center frequency (f _C ) of the analog signal immediately before digital sampling. And the necessary components center frequency is zero, unnecessary components associated to the generated center frequency is -2f _C thereto, to such can be easily separated, there is desirable as far as possible position. For this purpose, if the center frequency (f _C ) of the analog signal is set to 1/4 of the sampling frequency, the center of the unnecessary component will be the Nyquist frequency (the sampling frequency of the original signal frequency is guaranteed to be able to be sampled without distortion). It is ideal because it appears at the portion corresponding to the maximum value.)

The sampling frequency and center frequency used in the above processing can be used by obtaining a frequency reference signal from a generator of an OFDM signal to be measured, in addition to a method of reproducing from an input signal.

The OFDM signal of the terrestrial digital television assumed here has a configuration including a guard interval as shown in FIG. The guard interval is provided mainly to reduce the effect of signal delay due to multipath, and a part of the effective signal is used repeatedly.

The symbol section determining section shown in FIG. 1 performs a process of cutting out a section corresponding to the effective symbol length from the same symbol section. When one processing symbol section is determined, data of a time corresponding to the guard interval length is discarded, and data corresponding to the subsequent effective symbol length is determined as data of the next processing symbol section. Thus, the processing symbol sections are determined one after another.

For the determination of the processing symbol section, a general method already developed can be used. For example, in Reference 2,
It is described that by calculating a correlation value, for example, a time reference of a symbol cutout position (FFT window position) is extracted at a correlation peak.

When the amplifier to be measured constitutes a broadcast wave relay system for receiving and retransmitting a broadcast wave, there is a possibility that a delayed arriving wave is mixed in the input signal. In this case, it is necessary to determine a processing symbol section so that delayed arrival waves of different symbol sections do not mix in the processing symbol section. For this purpose, JP-A-20
An impulse response detection circuit described in JP-A-00-022657 detects an impulse response of a transmission line from a received signal, and thereby detects a state of occurrence of a delayed wave in the transmission line. Next, OF time is generated by a time window position control signal generation circuit.
A control signal for controlling the position of the time window for performing the discrete Fourier transform for demodulating the DM according to the content of the impulse response is generated. The discrete Fourier transformer is
A discrete Fourier transform is performed at a time window position on a time axis controlled by the control signal. ], Etc. can be used.

As another method, the measured O
There is also a method of obtaining and using a symbol synchronization signal from an FDM signal generator.

The carrier component decomposing unit performs an FFT process on the signal having the symbol section length obtained in the preceding stage. In this example, since one symbol length is 65536 samples, 65536-stage complex FFT operation is performed. Since the frequency conversion has been performed so that the center frequency becomes 0, 56
The result of 17 outputs is a complex number representing a complex signal point of each carrier obtained as a result of demodulation. If the bandwidth of the OFDM signal is widened due to the non-linearity, a component due to the non-linearity is generated in the output group adjacent to the center 5617. The processed signal is a system A for providing information about the pilot,
The signals are supplied to a system B for providing a corrected signal and a system C for providing a signal not to be corrected. Here, the carrier component decomposition unit, the system C, and the time domain signal generation unit 2 described below constitute first digital signal processing means, which is first digital processing means and means for limiting a frequency band. Further, the carrier component decomposing unit, the system B, and the time domain signal generating unit 1 described below constitute a second digital signal processing unit.

First, the system A will be described. Practically, the processing symbol section at the time of reception does not exactly match the original effective symbol section, but is determined over the guard interval section of the symbol. In this case, the position of the complex signal point of each carrier after FFT is rotated at a different angle for each carrier from what is intended on the transmission side. In addition, in the case of a signal in which a delayed wave is mixed, a carrier in which the desired wave is strengthened and a carrier in which the desired wave is weakened are generated. As described above, since the reference of the amplitude and the phase is different for each carrier, it is specified by the standard that a pilot for obtaining the information is inserted.

Assuming that 64QAM is used,
The pilot arrangement is defined as shown in FIG. FIG.
FIG. 3 is a schematic diagram showing a configuration of a signal as a multicarrier along a time axis and a frequency axis. A pilot corresponds to a particular complex signal point in each signal space diagram, regardless of the symbol (regardless of time), depending only on the carrier position. The pilot extraction unit in FIG. 1 extracts a pilot signal from the complex signal points output from the carrier component decomposition unit and outputs it to the pilot averaging unit.

The pilot averaging unit averages the results output by the pilot extraction unit for each carrier for a large number of processing symbol sections. For example, if the number of symbols processed by the pilot extraction unit is 400, 100 data are averaged for each pilot except for the carrier shown at the right end in FIG.

Next, a system B for providing a corrected signal will be described. In the amplitude / phase reference calculator,
The pilot obtained by the pilot averaging unit and OFDM
Using the data of the original complex signal points of the pilot defined by the signal standard, the amplitude and phase references of all carriers are estimated. For this part, a well-known method used for a demodulator can be used.

The amplitude / phase reference correction section corrects the amplitude / phase reference for the output data of the carrier component decomposition section in each processing symbol section using the result of the amplitude / phase reference calculation section. When this process is completed, the complex signal point becomes data in which a deviation due to nonlinear distortion is superimposed on one of the ideal signal point arrangements of the 64QAM signal space diagram shown in FIG.

The signal point arrangement shown in FIG. 4 is based on the pilot (points A and B) in 64 QAM of the terrestrial digital broadcasting standard.
Are represented by coordinates normalized to (√42 × 4/3, 0).

The complex signal point correction unit converts the signal points output from the amplitude / phase reference correction unit into 64QAM signals shown in FIG.
Correct the constellation to the closest point on the signal space diagram. If the signal point is a pilot, one of the points A and A is used.
Correct to any signal point of M.

The amplitude / phase reference recorrection unit converts the complex signal point output from the complex signal point correction unit to a complex signal point having the same amplitude / phase reference as before the amplitude / phase reference correction unit performs processing. Convert. This is the reverse operation of the amplitude / phase reference correction unit.

The time domain signal generator 1 generates a signal in the time domain from the data of the complex signal points. The same operation as the inverse Fourier transform (IFFT) in the modulation operation shown in the configuration of FIG. However, it is necessary to perform the same number of IFFT operations as the carrier component decomposition unit. In this example, 5617 data centered on DC is data corresponding to a complex signal point for each carrier, and (0, 0) must be inserted into other inputs to perform 65536-stage operations. The output signal of this portion corresponds to a signal obtained by correcting the nonlinear distortion of the signal cut out by the symbol section determination unit.

Next, a system C for providing a signal that is not subjected to correction will be described. In the out-of-band component suppression unit, unnecessary frequency components of the output of the carrier component decomposition unit are converted into (0, 0), and the time domain signal generation unit 2 converts the frequency components into time domain signals again. This part includes unnecessary frequency components generated in the AD conversion / frequency correction unit,
It serves as a filter for removing noise generated by frequency conversion or the like before conversion. The frequency components to be left without being converted to (0, 0) can be limited to, for example, a range of five times the original bandwidth of the OFDM signal around the center frequency of the OFDM signal. This multiple is determined by the degree of distortion, and it is desirable to increase the magnification when dealing with large distortion. Unwanted frequency components must be suppressed, but since the OFDM signal distorted by the amplifier has a wide bandwidth, it is desirable to pass through as wide a band as possible to obtain an accurate time-domain signal. .

[0053] In the signal comparing unit, a signal is reproduced by receiving the correction system B x '(t) = r ' (t) cos (ω 0 t + ψ '(t)), reproduced in not subject to correction system C Signal y ′ (t) = A ′ (t) cos (ω ₀ t + ψ ′ (t) + Φ ′
(T)), and outputs the result to the characteristic accumulation unit. Also, x ′ (t) is re-combined after removing elements that cause distortion, and thus can be considered to be very close to the original x (t).

The characteristic integration section has a function of integrating signals obtained from a large number of symbols and integrating them as amplifier characteristics. This storage number does not always match the storage number of the pilot averaging unit or the storage number of the amplitude / phase reference calculation unit. In this characteristic accumulation unit, the above x ′ (t)
And y ′ (t). AM-AM conversion characteristics are:
A ′ can be obtained as a function of r ′, and the AM-PM conversion characteristic can be obtained by determining Φ ′ as a function of r ′.

The AM-AM conversion characteristics and the AM-AM
What is important in the measurement of the PM conversion characteristic is mainly the portion where the envelope amplitude is large, and it is often sufficient to grasp only the point where the envelope is maximum in each symbol and the data in the temporal vicinity thereof. . In this case, it is also possible to reduce the data processing amount in the characteristic accumulation unit by selecting only the data of the corresponding part in the signal comparison unit and outputting the selected data to the characteristic accumulation unit.

[0056]

Since the present invention has the above-described configuration,
The following effects can be obtained.

It is now possible to measure only the signal on the output side of the amplifier and measure its nonlinear characteristic. In addition, since a single measurement point is used to extract a signal, it is not necessary to perform the process of adjusting the time relationship between the signals in the measurement of a plurality of signals. Furthermore, even when a linear element having a frequency characteristic such as a filter is inserted before the amplifier, the pilot estimation operation is performed, so that different attenuation and phase rotation for each carrier by the linear element are performed. Because it can be removed, only the non-linearity can be measured.

[Brief description of the drawings]

FIG. 1 is a diagram showing a desirable configuration of the present invention, and is a block diagram showing a flow of signals.

FIG. 2 is a schematic diagram showing a configuration of a symbol along a time axis of a signal.

FIG. 3 is a schematic diagram illustrating a configuration of a signal as a multicarrier along a time axis and a frequency axis.

FIG. 4 is a diagram showing a signal point arrangement of a 64QAM signal space diagram.

FIG. 5 is a block diagram showing an example of a configuration used for transmitting and receiving a signal of the OFDM system, showing the configuration on the (a) modulation side and (b) the demodulation side.

Claims (8)

[Claims] 1. A method for measuring characteristics of an amplifier in orthogonal frequency division multiplex transmission, wherein a multicarrier signal is divided into first and second multicarrier signals, and the first multicarrier signal is transmitted to a first digital signal processing means. Non-linear characteristics are obtained by comparing the amplitude and the phase of a signal obtained through the second multi-carrier signal with a signal obtained through the second digital signal processing means. A method for measuring non-linear characteristics of an amplifier. 2. The first digital signal processing means is means for removing noise, and the second digital signal processing means is
2. The method for measuring nonlinear characteristics of an amplifier according to claim 1, wherein the signal is a digital signal processing means for adding a correction related to a signal space diagram to a signal obtained by performing Fourier transform, and further performing inverse Fourier transform. 3. A method for measuring characteristics of an amplifier in orthogonal frequency division multiplexing transmission, comprising dividing a multicarrier signal into first and second multicarrier signals and passing the first multicarrier signal through a band-limiting filter. A non-linear characteristic is obtained by comparing the amplitude and the phase of the signal obtained by passing the second multi-carrier signal through a filter including a Fourier transform with the second multi-carrier signal. A method for measuring nonlinear characteristics of an amplifier. 4. The amplifier according to claim 3, wherein the Fourier transform filter is a filter for performing a Fourier inverse transform on a signal obtained by adding a correction related to a signal space diagram to a signal obtained by performing a Fourier transform. Non-linear characteristic measurement method. 5. The pilot according to claim 2 or 4, wherein an average obtained from processing of a plurality of symbols is used as a pilot used when making a correction on the signal space diagram. Method for measuring nonlinear characteristics of amplifier. 6. An apparatus for measuring characteristics of an amplifier in orthogonal frequency division multiplex transmission, comprising: means for dividing a multicarrier signal into first and second multicarrier signals; and means for limiting a frequency band of the first multicarrier signal. Means for obtaining a first signal through
Means for passing the second multicarrier signal through digital signal processing means including Fourier transform to obtain a second signal;
And a means for obtaining a nonlinear characteristic by comparing each amplitude of the signal and the second signal with each phase. 7. A digital signal processing means including a Fourier transform is a digital signal processing means for performing a correction on a signal space diagram to a signal obtained by the Fourier transform, and further performing an inverse Fourier transform. 7. The device for measuring nonlinear characteristics of an amplifier according to item 6. 8. A means for generating a pilot for use in making a correction relating to the signal space diagram, wherein the means for averaging and generating a result obtained by processing a plurality of symbols. The apparatus for measuring nonlinear characteristics of an amplifier according to claim 7, wherein: