JPH09116589A - Multi-valued number variable modem and radio communication equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate code errors generated accompanying the changeover of a modulation system and to switch the modulation system without hit by providing an amplitude correction means for performing adjustment to a signal point level corresponding to the respective modulation systems. SOLUTION: In a multi-valued number variable modulator 100, the amplitude correction means 130 is inserted between an operating point correction circuit 110 and a digital modulator 120. The amplitude correction means 130 corrects amplitude by using a correction coefficient corresponding to the modulation system specified by modulation system control signals for base band signals outputted from the operating point correction circuit 110 and sends them out to the digital modulator 120. The amplitude correction means 230 of a multi- valued number variable demodulator 200 corrects the amplitude by using the correction coefficient (the inverse number of the correction coefficient in the multi-valued number variable modulator 100) corresponding to the modulation system specified by the modulation system control signals for the base band signals outputted from a digital demodulator 210 and sends them out to an error signal judgement circuit 220.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a modulator / demodulator used in digital wireless communication. In particular, the present invention relates to a modulator / demodulator capable of varying the number of multilevels in a multilevel QAM modulation system classified into a phase modulation system (PSK) and an amplitude phase modulation system (APSK). Further, the present invention relates to a wireless communication device configured using the multilevel variable modulator / demodulator.

[0002]

2. Description of the Related Art FIG. 7 shows the configuration of a conventional digital wireless communication device. In the figure, a multilevel variable modulator 100 and a multilevel variable demodulator 200 are disclosed in Japanese Patent Application Laid-Open No. 3-25425.
This is based on the configuration disclosed in Japanese Patent Laid-Open No. 6 (multilevel variable modulator / demodulator). The multilevel variable modulator 100 is
The multi-value variable demodulator 200 includes an operating point correction circuit 110 and a digital modulator 120, and a multi-level variable demodulator 200 includes a digital demodulator 210 and an error signal determination circuit 220.

The n-bit (n = 1, 2, ...) Digital signal of each of the I-channel and the Q-channel and a modulation method control signal for designating a modulation method are input to the multilevel variable modulator 100. The operating point correction circuit 110 is composed of digital filters 111 and 112 corresponding to each channel, and operating point shift circuits 113 and 114 that shift the operating point of each digital filter output according to a modulation method control signal. When n-bit input signals are converted into 2 ^n- value baseband signals, the signal points shown in FIG. 8A are obtained. This signal point is shifted to the state of FIG. 8 (b) according to the modulation system, and the operating points are made to match in all modulation systems. The digital modulator 120 includes a digital quadrature modulator 121 that quadrature modulates the baseband signal output from the operating point correction circuit 110 with a carrier generation clock, and a digital / analog converter (D) that converts the modulated output to an analog signal. /
A) 122 and a band pass filter (BPF) 123, and outputs a 2 ²ⁿ QAM signal.

The transmitter 10 has an automatic level control (ALC).
A circuit 11, a frequency converter including a multiplier 12 and an oscillator 13, a band pass filter (BPF) 14, and an amplifier 15. The ALC circuit 11 includes a multi-valued variable modulator 100 to optimize the operating points of the frequency converter and the amplifier.
The power of the 2 ²ⁿ QAM signal input from is detected, and the average power is adjusted to be constant. The 2 ²ⁿ QAM signal with a constant average power is converted to the RF band by the frequency converter, amplified by the amplifier 15, and transmitted.

The receiver 20 includes a band pass filter (BP).
F) 21, automatic gain control (AGC) circuit 22, multiplier 2
3 and an oscillator 24, and a band pass filter (BPF) 25. 2 ²ⁿ QAM received
The signal is input to the AGC circuit 22 via the BPF 21. The AGC circuit 22 detects the received power in order to correct an amplitude error due to a propagation path condition such as fading.
The average power of the ²ⁿ QAM signal is adjusted to be constant. A 2 ²ⁿ QAM signal with a constant average power is converted into an IF band by a frequency converter and output via the BPF 25.

The digital demodulator 210 of the multi-valued variable demodulator 200 is a low pass filter (LPF) 211, an analog / digital converter (A / D) 212, and a digital detector for performing synchronous detection with a carrier recovery clock. A quadrature detector 213, retiming circuits (R) 214 and 215 for adjusting the timing of the detection output according to the reproduction clock,
It is composed of multipliers 216 and 217 which adjust the signal point level of the retiming circuit to a constant level by the control coefficient for AGC, demodulates the 2 ²ⁿ QAM signal from the receiver 20,
The baseband signal of m bits (m> n) for each of the channel and the Q channel is output.

The error signal judgment circuit 220 is composed of error bit selection circuits 221 and 222 corresponding to each channel, and calculates the value of n from the modulation method control signal which indicates the multi-valued number of 2 ²ⁿ QAM signals, and determines the value of m bits. Of these, the upper n bits are selected as the decoded signal and the (n + 1) th bit is selected as the error signal and output. Since the amplitude error and the carrier phase error can be detected from the decoded signal and the error signal, these error signals are given to the multipliers 216 and 217 of the digital demodulator 210 by AG.
It is used as a control coefficient for C and a signal for controlling the carrier recovery circuit.

[0008]

The 2 ²ⁿ QAM signal (n is output from the conventional multilevel variable modulator 100 shown in FIG.
= 1, 2, ...) The average power of each modulation method is different. 2 ²ⁿ
FIG. 9 shows the relationship between the average power of the QAM signal and the modulation method.
The horizontal axis shows the modulation method, and the vertical axis shows the average power when the average power of the QPSK modulation method is 0 dBm. As shown in the figure, it can be seen that the average power of the 2 ²ⁿ QAM signal changes before and after switching the modulation method. Since the control voltage of the ALC circuit 11 of the transmitter 10 changes accordingly,
The values of the parasitic elements such as the parasitic capacitance and the parasitic resistance of the amplifier circuit or the variable resistance circuit change, and as a result, the delay time changes. Furthermore, the AGC circuit 22 of the receiver 20 is also 2 ²ⁿ QAM.
Since the average power of the signal is adjusted to be constant, when the transmission power fluctuates, the delay time of the AGC circuit 22 changes like the ALC circuit 11.

Further, when the 2 ²ⁿ QAM signal whose average power is adjusted to be constant regardless of the modulation system is input from the receiver 20 to the multilevel variable demodulator 200, if the modulation system is switched, the baseband signal The signal point deviates from the optimum level. The reason is 2 ²ⁿ obtained from the signal shown in Fig. 8 (b).
Since the average power of the QAM signal is different for each modulation method as shown in FIG. 9, if the average power is adjusted to be constant,
This is because the signal point interval changes as shown in FIG.
The multipliers 216 and 217 of the digital demodulator 210 function as a level AGC circuit that adjusts the deviation of the signal point from the optimum level and adjusts the signal point level to be constant. The AGC circuit 22 installed in the receiver 20
Functions as a power AGC circuit for adjusting the average power of the signal to a constant value. As a result, like the ALC circuit 11,
The delay time of the AGC circuit including the multipliers 216 and 217 changes.

When the delay time of the entire wireless communication device changes each time the modulation method is switched, the carrier phase plane of the 2 ²ⁿ QAM signal rotates. As a result, the phase error of the carrier increases in the detector, and a code error easily occurs. It is an object of the present invention to provide a multi-level variable modulator / demodulator and a wireless communication device capable of eliminating a code error caused by switching a modulation system and switching the modulation system without interruption.

[0011]

The multilevel variable modulator includes:
Amplitude correction means for adjusting the average power of the output 2 ²ⁿ QAM signal to a constant value is provided. The amplitude correction means corrects the amplitude of the baseband signal, the carrier wave for modulation, or the quadrature-modulated 2 ²ⁿ QAM signal with a correction coefficient corresponding to the modulation method.

In the conventional multilevel variable modulator shown in FIG. 7,
The minimum signal point interval 2d _n of a 2 ^n- valued baseband signal is Q
If the minimum signal point interval of PSK is represented by 2δ, then 2d _n = 2δ / 2 ^n-1 (1). The average power P _n of 2 ²ⁿ QAM signal obtained the signal quadrature modulation to ^{_{the, P n = (d n 2}} /3) to become ^{(2 2n -1) ... (2} ).

Let N be the maximum value of n that the multilevel variable modulator can take, and let P _{N be} the average power of the 2 ^2N QAM signal. At this time, in order for the average power of the 2 ²ⁿ QAM signal to match P _N regardless of the modulation method, the minimum signal point interval is 2d _n = (2δ / 2 ^n-1 ) (P _N / P _n ) ^{1 / 2} (n = 1,2, ..., N) (3) Therefore, as shown in FIG. 5, for example, the baseband signal output from the operating point correction circuit is multiplied by the correction coefficient α = (P _N / P _n ) ^1/2 (4) according to each modulation method to obtain the amplitude. It is understood that the correction should be made. By this, even if the modulation method is switched, it is possible to perform correction so that the average power becomes constant with reference to the 2 ^2N QAM signal.

Since the modulation method is switched within 1 symbol, it is difficult for the analog circuit to follow the amplitude correction accompanying this. Therefore, all the above processes are performed digitally. Furthermore, this amplitude correction can be performed not only on the baseband signal but also on the carrier wave for modulation or the 2 ²ⁿ QAM signal after quadrature modulation, and in this case also, α in Eq. (4) is used as the correction coefficient. .

On the other hand, a multilevel variable demodulator to which a 2 ²ⁿ QAM signal whose average power is adjusted to a constant value is input from the receiver regardless of the modulation system, performs the reverse operation of the multilevel variable modulator.
Amplitude correction means for adjusting the signal point level at the time of error signal determination to an optimum value is provided. This amplitude correction means uses the reciprocal 1 / α of the correction coefficient used in the multi-valued variable modulator to calculate the amplitude of the 2 ²ⁿ QAM signal before quadrature detection, the carrier wave for detection, or the baseband signal after retiming. To correct.

For example, as shown in FIG. 6, 2 ^2N QAM
The reciprocal 1 / α of the correction coefficient corresponding to each modulation method is applied to the baseband signal controlled so that the average power is constant with the signal as a reference, and the signal point level is adjusted to the optimum level. Similarly, this amplitude correction may be applied not only to the baseband signal, but also to the 2 ²ⁿ QAM signal before quadrature modulation or the carrier wave for detection by the reciprocal 1 / α of the correction coefficient. This processing is also performed by digital processing like the modulator.

By the above operation, the average power of the 2 ²ⁿ QAM signal output from the multilevel variable modulator can be adjusted to be constant even if the modulation method is switched. Further, even if the 2 ²ⁿ QAM signal adjusted to have a constant average power regardless of the modulation method is input from the receiver to the multilevel variable demodulator, adjustment is performed so that the signal point level is optimum according to the modulation method. be able to.

[0018]

1 shows an embodiment of a multilevel variable modulator / demodulator of the present invention. In the figure, (1) shows an embodiment of the multilevel variable modulator 100, and (2) shows an embodiment of the multilevel variable demodulator 200. In the multilevel variable modulator 100, the amplitude correction means 130 is inserted between the operating point correction circuit 110 and the digital modulator 120. Amplitude correction means 1
Reference numeral 30 corrects the amplitude of the baseband signal output from the operating point correction circuit 110 using a correction coefficient corresponding to the modulation method specified by the modulation method control signal, and sends it to the digital modulator 120. As a result, even if the modulation method is switched, the average power of the 2 ²ⁿ QAM signal output from the multilevel variable modulator 100 can be controlled to be constant.

In the multilevel variable demodulator 200, an amplitude correction means 230 is inserted between the digital demodulator 210 and the error signal determination circuit 220. The amplitude correction means 230 uses, for the baseband signal output from the digital demodulator 210, a correction coefficient (the reciprocal of the correction coefficient in the multilevel variable modulator 100) corresponding to the modulation method specified by the modulation method control signal. Error signal determination circuit 2
Send to 20. As a result, even if a 2 ²ⁿ QAM signal whose average power is adjusted to a constant value is input to the multilevel variable demodulator 200 regardless of the modulation method, it is possible to adjust the signal point level to an optimum level.

[0020]

【Example】

(First Embodiment) FIG. 2 shows the first embodiment of the amplitude correcting means in the multilevel variable modulator / demodulator of the present invention. In the figure,
(1) shows a first embodiment of the amplitude correction means 130 provided in the multilevel variable modulator 100, and (2) shows a first embodiment of the amplitude correction means 230 provided in the multilevel variable demodulator 200. Show. The operating point shift circuits 113 and 114 of the operating point correction circuit 110 shown in FIG. 1 include the correction coefficient ROM 115 for operating point correction, the digital filter 111 and
The output of the correction coefficient ROM 115 is added to the output of the adder 112 and the adders 116 and 117.

The amplitude correction means 130 includes a correction coefficient ROM 131 that outputs a correction coefficient for amplitude correction, and a multiplier 132 that multiplies the baseband signal of each channel 2 ⁿ value output from the operating point correction circuit 110 by the correction coefficient. , 133
It consists of. The amplitude correction means 230 is a correction coefficient ROM 231 that outputs a correction coefficient for amplitude correction, and a multiplication that multiplies the correction coefficient by the baseband signal of each channel 2 ⁿ value input from the digital demodulator 210 to the error signal determination circuit 220. It is configured by the vessels 232 and 233.

Correction coefficient ROM 115, 131, 231
, The correction coefficient is updated based on the modulation method specified by the modulation method control signal. The correction coefficient is updated in synchronization with the clock so that the switching timing of the modulation method and the data switching timing of the baseband signal are matched. Here, Table 1 shows an example of the correction coefficient for amplitude correction when the modulation method is switched from QPSK to 256QAM.

[0023]

[Table 1]

In this embodiment, the amplitude correction process is performed by using a multiplier, but the multiplier may be replaced with a ROM, for example. Further, the tap coefficients of the digital filters 111 and 112 may be multiplied by the correction coefficient of Table 1. Also, in the configuration for correcting the amplitude of the 2 ²ⁿ QAM signal output from the digital quadrature modulator 121 and the 2 ²ⁿ QAM signal input to the digital quadrature detector 213,
The same action and effect as in the case of correcting the amplitude of the baseband signal in this embodiment can be obtained.

(Second Embodiment) FIG. 3 shows a second embodiment of the amplitude correcting means in the multilevel variable modulator / demodulator of the present invention.
This embodiment is the first example of the amplitude correction means 230 shown in FIG.
This is an alternative to the embodiment and shows the configuration included in the AGC circuit of the digital demodulator 210. Here, the multipliers 216 and 217 that form the AGC circuit multiply the baseband signal by the AGC control coefficient based on the amplitude error obtained from the decoded signal and the error signal, and adjust the signal point level.

The amplitude correction means includes a correction coefficient ROM 234 that outputs a correction coefficient for amplitude correction, and multipliers 216 and 21.
7 is composed of multipliers 235 and 236 for multiplying the AGC control coefficient by the correction coefficient. The correction coefficient ROM 234 is updated with the correction coefficient based on the modulation method specified by the modulation method control signal. The correction coefficient is updated in synchronization with the clock so that the switching timing of the modulation method and the data switching timing of the baseband signal are matched.

(Third Embodiment) FIG. 4 shows a third embodiment of the amplitude correction means in the multilevel variable modulator / demodulator of the present invention.
In this embodiment, QPSK, 16QAM, 64QAM, 256QA
Digital quadrature modulator 121 corresponding to each M modulation method
Also, in the digital quadrature detector 213, the amplitudes of the modulation carrier wave and the detection carrier wave are corrected.

In the figure, (1) shows a configuration example of a digital quadrature modulator 121 'including amplitude correction means, and (2) shows a configuration example of a digital quadrature detector 213' including amplitude correction means. A normal digital quadrature modulator 121 includes a counter 141 for inputting a carrier wave generation clock and a waveform ROM 14 for outputting a carrier wave waveform corresponding to the counter output.
2, multipliers 143 and 144 that multiply the carrier waveform of each channel and the baseband signal of each channel, and an adder 145 that adds the outputs of the multipliers. The normal digital quadrature detector 213 includes a counter 241 that inputs a carrier wave reproduction clock, a waveform ROM 242 that outputs a carrier wave waveform corresponding to the counter output, and multipliers 243 and 24 that multiply the carrier wave waveform and the 2 ²ⁿ QAM signal.
4 and a shift register 245 that gives a predetermined delay to the multiplication result of one of them (Q channel in this case).

Digital quadrature modulator 1 in this embodiment
Reference numeral 21 'includes a plurality of ROMs corresponding to each modulation system as the waveform ROM 142, and a selector 146 for selecting the output of the ROM corresponding to the modulation system designated by the modulation system control signal. The waveform ROM 142 stores a sine wave and a cosine wave whose amplitude is corrected by the correction coefficient in Table 1, that is, a value obtained by multiplying the amplitude of the modulation carrier wave by the correction coefficient corresponding to each modulation method, and stores the value of the counter 141 as an address. It is read as a value. The selector 146 switches the output of the waveform ROM 142 in synchronization with the clock based on the modulation method control signal, and outputs the carrier wave for modulation. Using this signal, the multipliers 143, 144 and the adder 145 perform quadrature modulation of the baseband signal, and the 2 ²ⁿ QAM signal (n = 1
~ 4) is output.

Digital quadrature detector 2 in this embodiment
13 'includes a similar waveform ROM 242 and a selector 246. The multipliers 243 and 244 and the shift register 245 perform quadrature detection of the baseband signal using the detection carrier whose amplitude is corrected corresponding to the modulation method output via the selector 246, and output the baseband signal.

(Other Embodiments) In the embodiment described above, in the multilevel variable modulator, the amplitude of the baseband signal is corrected (FIG. 2 (1)). The amplitude of the 2 ²ⁿ QAM signal after quadrature modulation. A configuration for correcting the amplitude (not shown) A configuration for correcting the amplitude of the carrier wave for modulation (FIG. 4 (1)) is shown. In addition, in the multilevel variable demodulator, the configuration for correcting the amplitude of the baseband signal (Fig. 2 (2),
3) A configuration for correcting the amplitude of the 2 ²ⁿ QAM signal before quadrature modulation (not shown) A configuration for correcting the amplitude of the carrier wave for detection (FIG. 4 (2)) is shown. The combination of the configuration of the multilevel variable modulator and the configuration of the multilevel variable demodulator is arbitrary, and all combinations are possible as the multilevel variable modulator / demodulator.

[0032]

As described above, in the multilevel variable modulator / demodulator and wireless communication apparatus of the present invention, even if the modulation system is switched, the average power of the 2 ²ⁿ QAM signal output by the multilevel variable modulator is changed. It can be adjusted to a certain level. Further, even if the 2 ²ⁿ QAM signal adjusted to have a constant average power by the receiver is input to the multilevel variable demodulator, the signal point level can be adjusted to the optimum signal level according to the modulation method. . As a result, the ALC circuit of the wireless communication device, the AG
The control of the feedback circuit such as the C circuit and the detector does not change, and no code error occurs. Therefore, by using the multilevel variable modulator / demodulator and the wireless communication device of the present invention, the modulation method can be switched without interruption.

Further, when the modulation method is designated other than the maximum multi-valued number, it can be considered that the amplitude of the baseband signal is expanded by the modulator and contracted by the demodulator.
Since this is equivalent to an increase in the number of bits of the baseband signal, the influence of quantization error can be reduced by adopting the configuration of the present invention.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a multilevel variable modulator / demodulator of the present invention.

FIG. 2 is a block diagram showing a first embodiment of amplitude correction means in the multilevel variable modulator / demodulator of the present invention.

FIG. 3 is a block diagram showing a second embodiment of the amplitude correction means in the multilevel variable modulator / demodulator of the present invention.

FIG. 4 is a block diagram showing a third embodiment of the amplitude correction means in the multilevel variable modulator / demodulator of the present invention.

FIG. 5 is a diagram showing the operating principle of the level correction function of the multilevel variable modulator of the present invention.

FIG. 6 is a diagram showing the operating principle of the level correction function of the multilevel variable demodulator of the present invention.

FIG. 7 is a block diagram showing a configuration of a conventional digital wireless communication device.

FIG. 8 is a diagram showing an operating principle of an operating point correction circuit of a conventional multilevel variable modulator.

FIG. 9: 2 ²ⁿ QAM output from a conventional multilevel variable modulator
The figure which shows the average power of a signal, and the relationship of a modulation system.

FIG. 10 is a diagram showing a relationship between a signal point level after detection when the average power of a 2 ²ⁿ QAM signal is constant in a conventional multilevel variable demodulator and an optimum signal point level for each modulation scheme.

[Explanation of symbols]

10 Transmitter 11 Automatic Level Control (ALC) Circuit 12 Multiplier 13 Oscillator 14 Band Pass Filter (BPF) 15 Amplifier 20 Receiver 21 Band Pass Filter (BPF) 22 Automatic Gain Control (AGC) Circuit 23 Multiplier 24 Oscillator 25 Band Pass filter (BPF) 100 Multilevel variable modulator 110 Operating point correction circuit 111, 112 Digital filter 113, 114 Operating point shift circuit 115 Correction coefficient ROM 116, 117 Adder 120 Digital modulator 121 Digital quadrature modulator 122 Digital / Analog converter (A / D) 123 Band pass filter (BPF) 130, 230 Amplitude correction means 131, 231, 234 Correction coefficient ROM 132, 133, 232, 233, 235, 236 Multiplier 141, 241 Counter 142, 242 Waveform ROM 143, 144, 243, 244 Multiplier 145 Adder 146, 246 Selector 200 Multi-level variable demodulator 210 Digital demodulator 211 Low pass filter (LPF) 212 Analog / digital converter (D / A) 213 Digital quadrature detector 214,215 Retiming circuit (R) 216,217 Multiplier 220 Error signal determination circuit 221,222 Error bit selection circuit 245 Shift register

Claims (3)

[Claims] 1. A baseband signal according to a modulation method designated by the modulation method control signal, by inputting each n-bit (n is an integer of 1 or more) digital signal of the I channel and Q channel and a modulation method control signal. Shift the operating point of
Furthermore a multi-level number variable modulator quadrature modulation and outputs a 2 ²ⁿ QAM signal, inputs the 2 ²ⁿ QAM signal and the modulation scheme control signal,
The 2 ²ⁿ QAM signal is subjected to quadrature detection, and the obtained baseband signal is determined according to the modulation scheme specified by the modulation scheme control signal.
In a multi-level variable modulator / demodulator including a multi-level variable demodulator that outputs a bit-decoded signal, the multi-level variable modulator includes a correction coefficient corresponding to a modulation method specified by the modulation method control signal. To correct and output the amplitude of the baseband signal, the carrier wave for modulation, or the 2 ²ⁿ QAM signal after quadrature modulation, and output 2 ²ⁿ QA
An amplitude correction unit that adjusts the average power of the M signal to a constant value is provided, and the multilevel variable demodulator is provided with a 2 ²ⁿ QAM signal before quadrature detection by the reciprocal of the correction coefficient used in the multilevel variable modulator A multilevel variable modulator / demodulator characterized by comprising amplitude correction means for correcting the amplitude of the carrier wave for detection or the baseband signal and adjusting it to a signal point level according to each modulation method. 2. A digital filter for inputting an n-bit (n is an integer of 1 or more) digital signal and a modulation method control signal for each of the I channel and the Q channel, and shaping and band limiting the digital signal, and a digital filter. An operating point correction circuit including an operating point shift circuit that shifts the operating point of the output baseband signal according to the modulation method specified by the modulation method control signal, and matches the operating points in all modulation methods, and an operating point a multi-level number variable modulator comprising a digital modulator that baseband signal quadrature modulation and outputs a 2 ²ⁿ QAM signal output from the correction circuit, and enter the 2 ²ⁿ QAM signal and the modulation scheme control signal,
A digital demodulator that orthogonally detects 2 ²ⁿ QAM signals and outputs a baseband signal, and the obtained baseband signal is determined according to the modulation scheme specified by the modulation scheme control signal, and the I-channel and the Q-channel are determined. In a multilevel variable modulator / demodulator including a multilevel variable demodulator including an error signal determination circuit that outputs each n-bit decoded signal, the multilevel variable modulator is designated by the modulation method control signal. A correction coefficient corresponding to the modulation method to be used, a tap coefficient of the digital filter, a baseband signal output from the operating point correction circuit, a carrier wave for modulation given to the digital modulator, or 2 after quadrature modulation.
Multiplying the amplitude of the ²ⁿ QAM signal, and outputs 2 ²ⁿ QAM
Amplitude correction means for adjusting the average power of the signal to a constant value is provided, and the multi-value variable demodulator is provided with the inverse of the correction coefficient used in the multi-value variable modulator, and the quadrature detection is input to the digital demodulator Amplitude correction means is provided for multiplying the amplitude of the preceding 2 ²ⁿ QAM signal, the carrier wave for detection given to the digital demodulator, or the baseband signal to adjust the signal point level according to each modulation method. A multi-level variable modulator / demodulator that does. 3. A multilevel variable modulator / demodulator comprising the multilevel variable modulator and the multilevel variable demodulator according to claim 1, or 2 output from said multilevel variable modulator. ^A transmitter that transmits a ²ⁿ QAM signal as a radio signal, and a receiver that receives the radio signal transmitted from the transmitter and sends a 2 ²ⁿ QAM signal whose average power is adjusted to a constant value to the multilevel variable demodulator. And a wireless communication device.