JP2004173249A - Transmitter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transmitter of an EER (envelope elimination and restoration) method with wide band and high efficiency. <P>SOLUTION: Amplitude components of a modulation signal are inputted in a power supply terminal of a high frequency power amplifier 130, an IQ quadrature signal is inputted in a high frequency input terminal of the high frequency power amplifier 130 and an original modulation signal is obtained from the output of the high frequency power amplifier 130. Collector voltage is supplied from a group 615 of DC-DC converters with successively different output voltage to an emitter follower 729 via a group 621 of switches. As for the collector voltage, efficiency of the emitter follower 729 is enhanced by reducing difference between emitter voltage of the emitter follower 729 and collector voltage of the emitter follower 729 by selecting any one output of DC-DC converters 616 to 620 by the group 621 of switches corresponding to the level of the amplitude components and giving the output to the emitter follower and a wide band operation is made possible by performing voltage conversion of power voltage of the high frequency power amplifier 130 by the emitter follower 729. <P>COPYRIGHT: (C)2004,JPO

Description

  The present invention relates to a radio transmitter used in a communication system using subcarriers such as OFDM (Orthogonal Frequency Division Multiplex).

  In general, a modulated signal involving voltage conversion, particularly a modulated signal involving multi-level modulation such as QAM (quadrature voltage conversion) requires a high-frequency power amplifier for transmitting power to an antenna to perform a linear operation. For this reason, class A or class AB has been used as an operation class of the high-frequency power amplifier.

  However, with the development of broadband communication, communication systems using subcarriers such as orthogonal frequency division multiplexing (OFDM) have begun to be used, and higher efficiency is expected in conventional class A and class AB high frequency power amplifiers. Can not. That is, in the OFDM, large power is generated instantaneously and completely at random by superposition of subcarriers. That is, the ratio between the average power and the instantaneous maximum power, PAPR (Peak to Average Power Ratio), is large. Therefore, it is necessary to always maintain a large DC power so that a high-frequency signal having such a large power can also be linearly amplified. In class A operation, the power efficiency is only 50% at the maximum, and especially in the case of OFDM, the power efficiency is about 10% due to the large PAPR.

  For this reason, for example, in a portable wireless device using a battery as a power supply, the continuous usable time is short, and a practical problem occurs.

In order to solve such a problem, a conventional EER method (Envelope Elimination and Restoration) known as the Kahn method has been proposed (for example, see Patent Document 1).
FIG. 6 is a block diagram schematically showing the EER method. In FIG. 6, a high-frequency modulation signal 46, for example, a QAM signal input to a terminal 40 is branched into two. In one branch, the modulated wave 46 is envelope-detected by a detector 41, thereby generating an amplitude component signal. The power supply voltage Vdd is voltage-converted by a voltage converter (an amplifier for amplifying an amplitude component) 42. At this time, an S class amplifier (such as a switching regulator) capable of high-efficiency operation (up to 95%) is used as the voltage converter 42. In the other branch, the amplitude of the modulated wave 46 is controlled by the amplitude control amplifier (limiter 43), whereby a modulated wave having only phase information is obtained. The modulated wave having the phase information is input to the RF input terminal of the switch amplifier 44, and modulates the gate voltage of, for example, a field effect transistor, which is a component of the switch amplifier 44.

  Here, the switch-type amplifier is a class F amplifier whose harmonics are controlled so that the drain voltage waveform becomes rectangular, a class E amplifier or a class D whose load conditions are optimized so that the drain voltage waveform and the drain current waveform do not overlap. Point to the amplifier.

  In the conventional class A amplifier, a period occurs in which a drain voltage and a drain current are simultaneously generated, and power is consumed. On the other hand, the switch amplifier 44 minimizes the period during which the drain current and the drain voltage are simultaneously generated, so that the power consumption can be suppressed.

  For example, if DC power of 200 mA and 3 V is supplied, DC power is 600 mW. In the switch-type amplifier 44, no current flows at the time of OFF, and only the voltage Vdd is applied, so that the DC power consumption is zero. On the other hand, a current of 200 mA flows at the time of ON, but since the transistor is completely conducting, the drain-source voltage VDS can be assumed to be about 0.3 V at most of the saturation voltage. In this case, 0.3 × 0.2 = 0.06, that is, 60 mW of DC power is consumed in the transistor. The power efficiency indeed reaches (600-60) / 600 = 90%. This effect is significant because the power efficiency of the class A amplifier reaches only 50% at the maximum.

  That is, high power efficiency is realized by using the switch-type amplifier. However, since the switch-type amplifier is a non-linear amplifier, it is necessary to linearly amplify a modulated signal such as a QAM signal in which the amplitude level of a modulated wave changes, so that a switch-type amplifier cannot be used.

  In order to solve this problem, the EER method separates a signal containing amplitude information into an amplitude component and a phase component, and a switch amplifier amplifies only the phase component. Here, if the amplitude component is input to the power supply terminal of the switch-type amplifier, an output power proportional to the amplitude component can be obtained. As a result, a signal including the original amplitude information is reproduced.

With such a configuration, a non-linear but high-efficiency amplifier such as a switch-type amplifier can be used, so that high efficiency can be achieved.
U.S. Pat. No. 6,256,482 B1 (3 page drawing, FIG. 6)

  However, since the bandwidth of the voltage converter 42 that modulates the amplitude component, for example, a switching regulator) is at most 5 MHz, for example, a modulation wave bandwidth of 20 MHz according to the IEEE802.11a standard, which is a wireless LAN standard, and a conventional EER The law cannot be used.

  To widen the band, it is necessary to reduce the inductance of the low-pass filter built into the output of the voltage converter 42. However, since the Q value of the inductance decreases, the amount of heat consumed by the inductance cannot be ignored, and the efficiency of the voltage converter 42 decreases. Also, noise increases.

  When a series regulator is used as the voltage converter 42, the power consumption is the product of the voltage conversion amount (difference between the power supply voltage and the amplitude component voltage) and the drain current of the high-frequency power amplifier. In OFDM, the average value of the voltage of the amplitude component is equal to or less than half of the power supply voltage.

  Therefore, an object of the present invention is to provide a transmitter capable of realizing a wideband EER method without reducing efficiency.

  In order to solve the above-mentioned problem, a transmitter according to a first aspect of the present invention includes a modulation signal generating means for generating a modulation signal, and a phase and amplitude component for separating the modulation signal generated by the modulation signal generation means into a phase component and an amplitude component. Separating means, amplitude slicing means for slicing the amplitude components separated by the phase and amplitude separating means at a plurality of voltage levels different from each other, and a plurality of switching regulators for converting a power supply voltage into a plurality of voltages having different values step by step A switch group for selecting any one of output voltages of the plurality of switching regulators, a switch driver for selectively turning on each switch of the switch group according to slice data of an amplitude component sliced by the amplitude slicing means, and a switch. The output voltage of any of the switching regulators selected by the group is used as the A linear voltage converting means for converting, a phase component is input to a high frequency input terminal, an amplitude component obtained by voltage conversion by the linear voltage converting means is input to a power supply terminal, and as a result, a modulated wave whose amplitude and phase are multiplied is output. And a high-frequency power amplifier.

  According to this configuration, a plurality of switching regulators for converting the power supply voltage into a plurality of voltages having different values in a stepwise manner are provided, the switching regulator is selected according to the level of the amplitude component, and the output voltage of the selected switching regulator is output. A configuration is adopted in which the linear voltage converter converts the amplitude component into a voltage as the power supply voltage. Therefore, the voltage drop by the linear voltage converter can be suppressed to be small, and the loss due to the switching regulator can be reduced, and the power loss due to the linear voltage converter can be suppressed to be small. In addition, since a linear voltage converter is used for voltage conversion, and a low-pass filter does not need to be used for an output unit, a wider band can be achieved. Therefore, a wideband EER method can be realized without lowering the efficiency.

  According to a second aspect of the present invention, there is provided a transmitter that generates a modulation signal, a phase / amplitude separation unit that separates a modulation signal generated by the modulation signal generation unit into a phase component and an amplitude component, and a phase / amplitude separation unit. Amplitude slicing means for slicing the amplitude components separated by a plurality of voltage levels stepwise, a plurality of switching regulators stepwise converting a power supply voltage into a plurality of voltages having different values, and outputs of the plurality of switching regulators A plurality of linear voltage converting means for converting an amplitude component into a voltage using each voltage as a power supply voltage; a group of switches for transmitting the amplitude component to the plurality of linear voltage converting means; and a switch in accordance with slice data of the amplitude component sliced by the amplitude slicing means. A switch driver that selectively turns on each switch in the group, and the phase component is input to the high-frequency input terminal , And a high frequency power amplifier inputs the voltage transformed amplitude component to a power supply terminal, and outputs a modulated wave resulting amplitude and phase is multiplied by a plurality of linear voltage converting means.

  According to this configuration, a plurality of switching regulators for converting the power supply voltage into a plurality of voltages having different values in a stepwise manner are provided, and the output voltages of the plurality of switching regulators are used as the power supply voltage, and the plurality of linear voltage conversion means respectively convert the amplitude components. In addition to the voltage conversion, one of the plurality of linear voltage conversion means is selectively enabled according to the level of the amplitude component. Therefore, it is possible to reduce the voltage drop due to the linear voltage conversion unit when performing the voltage conversion, and to reduce the loss due to the switching regulator and the power loss due to the linear voltage conversion unit. In addition, since a linear voltage converter is used for voltage conversion, and a low-pass filter does not need to be used for an output unit, a wider band can be achieved. Therefore, a wideband EER method can be realized without lowering the efficiency. Further, since only the linear voltage converting means enters between the switching regulator and the high-frequency power amplifier and the switching means is removed from the path, the power loss can be further reduced as compared with the configuration of the first invention. it can.

  In the transmitter according to the first or second aspect of the present invention, the transmitter may have a frequency converter between the output terminal of the phase component of the phase and amplitude separator and the input terminal of the high-frequency power amplifier.

  According to this configuration, the following operation and effect can be obtained. Since the band of the phase / amplitude separating means is at most several hundred MHz, when the carrier wave exceeds GHz, it cannot be processed. However, by using a frequency converting means such as a quadrature modulator, it can be easily processed. Carrier frequency can be up-converted.

  In the transmitter according to the first or second aspect of the present invention, a feedback unit provided at an output end of the high-frequency power amplifier for feeding back the high-frequency output power, and a timing shift between the phase and the amplitude is calibrated based on a signal from the feedback unit. Timing calibration means for generating a first calibration signal for receiving the first calibration signal from the first timing calibration means, and determining the timing of the amplitude component and the phase component output from the phase / amplitude separation means. It is preferable to add a first timing correcting means for correcting.

  According to this configuration, the following operation and effect can be obtained. If the timing of the phase component and the amplitude component deviates due to the delay due to the layout from the input of each modulated wave component to the output of the high-frequency power amplifier, or the delay due to the wiring length or parasitic component of the transistor, the correct modulation at the high-frequency power amplifier output Waves cannot be formed. However, the provision of the feedback means, the timing correction means, and the timing correction means makes it possible to accurately correct the timing of the phase component and the voltage-converted wave component, and to form a correct modulated wave at the output of the high-frequency power amplifier.

  In the transmitter according to the first or second aspect, the first voltage detecting means provided between the output terminal of the switching regulator and the power supply voltage input terminal of the linear voltage converting means for detecting the output voltage of the switching regulator. And second voltage detecting means provided at the amplitude component input terminal of the linear voltage converting means for detecting the voltage of the amplitude component, and voltage amplitude data obtained from the first and second voltage detecting means. A second timing calibrator for outputting a second calibration signal for calibrating a timing deviation between the amplitude component and the slice data; and a second calibration signal from the second timing calibrator receiving the second calibration signal from the second component. Preferably, a second timing correction means for correcting timing is added.

  According to this configuration, the following operation and effect can be obtained. If the timing of the amplitude slice data and the amplitude component is shifted due to the delay due to the layout from the input of the amplitude slice data and the amplitude component to the output of the high-frequency power amplifier, or the delay due to the wiring length of the transistor or the parasitic component, the slice data causes The output of the switching regulator turned on by the driven switch deviates from the value of the amplitude component, and an unnecessarily large voltage drop occurs in the linear voltage conversion means, thereby reducing power supply efficiency or turning off the linear voltage conversion means. However, by providing the first and second voltage detecting means, the second timing calibrating means, and the second timing correcting means, it is possible to accurately correct the timing of the amplitude slice data and the amplitude component, and to provide an ideal power supply. Efficiency can be realized.

  In the transmitter according to the first or second aspect, the linear voltage conversion means is constituted by, for example, an emitter follower.

  According to this configuration, the amplitude component is converted to a voltage lower by a certain voltage level (for example, 0.7 V) of the emitter-base voltage determined by the built-in potential of the PN junction, and a feedback loop is formed. Since it is not provided, there is no band limitation by the loop, and the configuration is simplified.

  In the transmitter according to the first or second aspect, the linear voltage converter may be constituted by a linear regulator.

  According to this configuration, the voltage level can be accurately controlled by the feedback loop, and the amplitude component can be correctly converted into a voltage.

  In the transmitter according to the first or second aspect, the amplitude component is input to the operational amplifier, the output of the operational amplifier is connected to the input of the emitter follower, and the output of the emitter follower is negatively fed back to the operational amplifier. Is preferred.

  According to this configuration, the nonlinearity and the temperature characteristic of the emitter follower can be compensated, and the amplitude component can be correctly transmitted to the high-frequency power amplifier.

  In the transmitter according to the first or second aspect, the emitter follower is constituted by a push-pull circuit, the amplitude component is input to the operational amplifier, and the output of the operational amplifier is connected to the input of the push-pull circuit. May be negatively fed back to the operational amplifier.

  According to this configuration, the non-linearity and temperature characteristics of the emitter follower are compensated, and the transient characteristics of the operational amplifier prevent the voltage from being held at the positive power supply voltage or the negative power supply voltage applied to the operational amplifier, and the amplitude is reduced. The component can be correctly transmitted to the high frequency power amplifier.

  According to a third aspect of the present invention, there is provided a transmitter that generates a modulation signal, an amplitude extraction unit that extracts an amplitude component from the modulation signal generated by the modulation signal generation unit, and an amplitude component that is extracted by the amplitude extraction unit. Amplitude slicing means for slicing the power supply voltage stepwise at a plurality of different voltage levels, a plurality of switching regulators for converting a power supply voltage to a plurality of voltages having different values stepwise, and any one of output voltages of the plurality of switching regulators , A switch driver for selectively turning on each switch of the switch group according to the slice data of the amplitude component sliced by the amplitude slicing means, and an output voltage of one of the switching regulators selected by the switch group. Linear voltage conversion means for converting the amplitude component as a power supply voltage, and the modulation signal Input to a high frequency input terminal, the apply voltage transformed amplitude component to the power supply terminal by a linear voltage converting means, and a high-frequency power amplifier for outputting a modulated wave as a result.

  According to this configuration, a plurality of switching regulators for converting the power supply voltage into a plurality of voltages having different values in a stepwise manner are provided, the switching regulator is selected according to the level of the amplitude component, and the output voltage of the selected switching regulator is output. The voltage conversion is performed by the linear voltage conversion means converting the amplitude component as the power supply voltage. Therefore, it is possible to reduce the voltage drop due to the linear voltage conversion unit when performing the voltage conversion, and to reduce the loss due to the switching regulator and the power loss due to the linear voltage conversion unit. In addition, since a linear voltage converter is used for voltage conversion, and a low-pass filter does not need to be used for an output unit, a wider band can be achieved. Therefore, a wideband EER method can be realized without lowering the efficiency.

  Further, since the modulation signal is used as it is instead of the phase component, it is possible to avoid deterioration of the modulation accuracy (ErrorVectorMagnitude: EVM) which cannot be avoided in the EER method in which the amplitude and phase components are separated. That is, when the phase component is used, the phase component is filtered to the extent permitted by the band of the digital-to-analog converter and to such an extent that the EVM is not affected. When the phase component was combined with the amplitude component at the output of the amplifier, the EVM was significantly deteriorated. In addition, since the required bandwidth of the modulated signal is about 1/6 smaller than the phase component separated from the modulated signal, the bandwidth of a digital-to-analog converter or an anti-alias filter that suppresses spurious components generated by the digital-to-analog conversion is used. Can be narrowed. This is advantageous for reducing the power consumption of the digital-to-analog converter and for reducing the size and cost of the inductor used for the filter.

  Further, in the conventional EER method, even when peak power is input, an input level enough to saturate the high-frequency power amplifier is injected. Therefore, the isolation characteristic (output) when the high-frequency power amplifier is OFF (amplitude component 0) is set. If the ratio of the leakage from the input power in the power is not good, power higher than the expected level is output, and as a result of being multiplied by the amplitude component, a correct modulated wave cannot be formed at the output of the high-frequency power amplifier (EVM performance). Degradation). However, in this configuration, when the high-frequency power amplifier is OFF (amplitude component 0), the power input to the high-frequency power amplifier is also 0, so that a correct modulated wave can be output at the high-frequency power amplifier output without depending on the isolation characteristics. Can be formed.

  According to a second aspect of the present invention, there is provided a transmitter that generates a modulation signal, an amplitude extraction unit that extracts an amplitude component from the modulation signal generated by the modulation signal generation unit, and an amplitude component that is extracted by the amplitude extraction unit. Amplitude slicing means for slicing the power supply voltage stepwise at a plurality of different voltage levels, a plurality of switching regulators for converting a power supply voltage into a plurality of voltages having different values stepwise, and an output voltage of the plurality of switching regulators as a power supply voltage. A plurality of linear voltage converting means for converting the amplitude component into a voltage, a group of switches for transmitting the amplitude signal to the plurality of linear voltage converting means, and selecting each switch of the switch group according to the slice data of the amplitude component sliced by the amplitude slicing means A switch driver for electrically conducting the signal and the modulation signal to a high frequency input terminal, And a high frequency power amplifier inputs the voltage transformed amplitude component to a power supply terminal, and outputs the modulated wave as a result by the pressure transducer means.

  According to this configuration, a plurality of switching regulators for converting the power supply voltage into a plurality of voltages having different values in a stepwise manner are provided, and the output voltages of the plurality of switching regulators are used as the power supply voltage, and the plurality of linear voltage conversion means respectively convert the amplitude components. The voltage conversion is performed by voltage conversion, and any one of the plurality of linear voltage conversion means is selectively enabled according to the level of the amplitude component. Therefore, it is possible to reduce the voltage drop due to the linear voltage conversion unit when performing the voltage conversion, and to reduce the loss due to the switching regulator and the power loss due to the linear voltage conversion unit. In addition, since a linear voltage converter is used for voltage conversion, and a low-pass filter does not need to be used for an output unit, a wider band can be achieved. Therefore, a wideband EER method can be realized without lowering the efficiency. Also, since only the linear voltage conversion means is inserted between the switching regulator and the high-frequency power amplifier and the switching means is removed from the path, the power loss can be further reduced as compared with the configuration of the third invention. it can.

  Further, since the modulation signal is used as it is instead of the phase component, it is possible to avoid deterioration of the modulation accuracy (ErrorVectorMagnitude: EVM) which cannot be avoided in the EER method in which the amplitude and phase components are separated. That is, when the phase component is used, the phase component is filtered to the extent permitted by the band of the digital-to-analog converter and to the extent that the EVM is not affected. Significant degradation of the EVM occurs when the phase component is combined with the amplitude component at the output of the amplifier. In addition, since the required bandwidth of the modulated signal is about 1/6 smaller than the phase component separated from the modulated signal, the bandwidth of a digital-to-analog converter or an anti-alias filter that suppresses spurious components generated by the digital-to-analog conversion is used. Can be narrowed. This is advantageous for reducing the power consumption of the digital-to-analog converter and for reducing the size and cost of the inductor used for the filter.

  Further, in the conventional EER method, even when peak power is input, the input level sufficient to saturate the high-frequency power amplifier is injected, so that the isolation characteristics when the high-frequency power amplifier is OFF (amplitude component 0) are good. If not, power higher than the expected level is output and multiplied by the amplitude component, and as a result, a correct modulated wave cannot be formed at the output of the high-frequency power amplifier (degradation of EVM performance has been caused). When the high-frequency power amplifier is OFF (amplitude component 0), the power input to the high-frequency power amplifier is also 0, so that a correct modulated wave can be formed at the output of the high-frequency power amplifier without depending on the isolation characteristics.

  In the transmitter according to the third or fourth invention, the linear voltage conversion means is constituted by, for example, an emitter follower.

  According to this configuration, the amplitude component is converted to a voltage lower by a certain voltage level (for example, 0.7 V) of the emitter-base voltage determined by the built-in potential of the PN junction, and a feedback loop is formed. Since it is not provided, there is no band limitation by the loop, and the configuration is simplified.

  In the transmitter according to the third or fourth invention, the linear voltage conversion means may be constituted by a linear regulator.

  According to this configuration, the voltage level can be accurately controlled by the feedback loop, and the level of the amplitude component can be correctly converted.

  In the transmitter according to the third or fourth aspect, the amplitude component is input to the operational amplifier, the output of the operational amplifier is connected to the input of the emitter follower, and the output of the emitter follower is negatively fed back to the operational amplifier. Is preferred.

  According to this configuration, the nonlinearity and the temperature characteristic of the emitter follower can be compensated, and the amplitude component can be correctly transmitted to the high-frequency power amplifier.

  In the transmitter according to the third or fourth aspect, the emitter follower is constituted by a push-pull circuit, the amplitude component is input to the operational amplifier, and the output of the operational amplifier is connected to the input of the push-pull circuit. May be negatively fed back to the operational amplifier.

  According to this configuration, the non-linearity and temperature characteristics of the emitter follower are compensated, and the transient characteristics of the operational amplifier prevent the voltage from being held at the positive power supply voltage or the negative power supply voltage applied to the operational amplifier, and the amplitude is reduced. The component can be correctly transmitted to the high frequency power amplifier.

  As described above in detail, according to the present invention, a wideband and highly efficient operation can be performed in the EER method in which the high-frequency power amplifier can be operated as a switch type.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
Hereinafter, Embodiment 1 of the present invention will be described with reference to the drawings. In the present embodiment, a wireless LAN system using the IEEE 802.11a standard using a wideband modulated signal will be described as an example. In the wireless LAN system, modulation of 64 QAM is applied to each of the 52 orthogonal subcarriers, and the sum is added to obtain a modulated signal. The 52 subcarriers are separated by 312.5 kHz, and occupy 52 × 312.5 = 16.25 MHz.

  FIG. 1 is a circuit diagram of a transmitter according to a first embodiment of the present invention that implements the EER method. As shown in FIG. 1, the transmitter includes an OFDM signal generation unit 111, a phase / amplitude separation unit 112, an amplitude slicing unit 113, a switching regulator group 115, a switch group 121, a switch driver 114, a quadrature modulation , A series regulator 129, and a switch type high frequency power amplifier 130.

  The above-mentioned OFDM signal generation unit 111 generates an OFDM signal, and corresponds to a modulation signal generation unit that generates a modulation signal.

  The phase / amplitude separating means 112 receives, for example, a power supply voltage of 5 V as an input and separates the OFDM signal generated by the OFDM signal generating means 111 into a phase component and an amplitude component.

  The amplitude slicing means 113 slices the amplitude component separated by the phase / amplitude separating means 112 at a plurality of appropriate voltage levels that differ stepwise. As this voltage level, for example, 0.5 V, 1.0 V, 1.5 V, 2.0 V, and 2.5 V are set. FIG. 1 shows an amplitude component input to the amplitude slicing means 113, that is, a source signal, and an output signal of the amplitude slicing means 113, that is, a slice signal.

  Here, the relationship between the source signal and the slice signal shown in FIG. 1 will be described. The amplitude slicing means 113 detects the level of the amplitude component as shown in FIG. 1, compares the level with a preset voltage level, and slices the amplitude component as shown in FIG.

  The method of the amplitude slicing is as follows. For example, if the amplitude component is 0.5V <amplitude component ≦ 1.0V, it is rounded to 1V, and if 1V <amplitude component ≦ 1.5V, it is rounded to 1.5V. Round levels. In the figure, since there are a total of seven levels, these are assigned to 3-bit data, and 3-bit slice data is output to the switch driver 114.

  The switching regulator group 115 includes a plurality of, for example, four switching regulators to which a power supply voltage of, for example, 3 V is input, that is, four DC-DC converters 116 to 120. The DC-DC converters 116 to 120 convert the power supply voltage stepwise into a plurality of voltages having different values. Specifically, the DC-DC converters 116 to 120 convert a voltage of 3V into respective voltages of 2.5V, 2.0V, 1.5V, 1.0V, and 0.5V.

  The switch group 121 includes, for example, five switches 122 to 127, one of which is selectively conductive. The switch group 121 has a power supply voltage of 3V, 2.5V which is an output voltage of the plurality of DC-DC converters 116 to 120, One of the voltages of 2.0 V, 1.5 V, 1.0 V, and 0.5 V is selected. Note that the switches 122 to 127 are configured by, for example, MOS transistors.

  The switch driver 114 selectively turns on each of the switches 122 to 127 of the switch group 121 according to the slice data of the amplitude component sliced by the amplitude slicing means 113.

  The quadrature modulator 128 converts a phase component (quadrature component and in-phase component) output from the phase / amplitude separating means 112 into a high-frequency signal, and corresponds to a frequency converting means.

  The series regulator (linear regulator) 129 converts the amplitude component of the OFDM signal into a voltage using the 3V power supply voltage selected by the switch group 121 or the output voltage of any of the DC-DC converters 116 to 120 as a power supply voltage. It corresponds to linear voltage conversion means.

  The high-frequency power amplifier (PA) 130 is of a switch type, inputs a high-frequency signal (a phase component of which has been subjected to high-frequency conversion) input from the quadrature modulator 128 to a high-frequency input terminal, and is voltage-converted by the series regulator 129. The amplitude component is input to a power supply terminal, and as a result, a modulated wave in which both the phase and the amplitude are modulated, that is, the amplitude and the phase are multiplied is output.

  The operation will be described below. In the present embodiment, a system with a power supply voltage of 3 V is assumed.

  The OFDM signal generated by the OFDM signal generation unit 111 is separated into an amplitude component and a phase component by a phase / amplitude separation unit 112 and output. Based on the output amplitude component, the amplitude slicing means 113 generates drive information for driving on / off of each of the switches 122 to 127 of the switch group 121. The drive information is hereinafter referred to as slice data.

The method of amplitude slicing is, for example, that the amplitude component is 0 V <the amplitude component ≦ 0.5 V
Then round it to 0.5V,
0.5V <amplitude component ≦ 1.0V
Then round it to 1.0V,
1.0V <amplitude component ≦ 1.5V
Then round it to 1.5V,
1.5V <amplitude component ≦ 2.0V
Then round it to 2.0V,
2.0V <amplitude component ≦ 2.5V
Then round to 2.5V,
2.5V <amplitude component ≦ 3.0V
Then, a threshold range in which the amplitude component is included is detected, such as rounding to 3.0 V, and the level is rounded to the maximum value of the included range.

  The rounding is performed as follows. The DC-DC converters 116 to 120 are prepared to output the same output voltage (2.5 V, 2.0 V, 1.5 V, 1.0 V, 0.5 V) as the rounded voltage level. In accordance with the level of the amplitude component, the amplitude slicing means 113 gives the switch driver 114 information on which DC-DC converter (116, 117, 118, 119 or 120) to activate the output. According to the given information, the switch driver 114 selectively turns on / off the switches 122 to 127 provided in the output stages of the DC-DC converters 116 to 120, and outputs a voltage corresponding to the rounded voltage.

  To describe a specific example, when the amplitude component is 1.2 V, the path of the DC-DC converter 118 is turned on, and a voltage of 1.5 V is supplied to the voltage input terminal of the series regulator 129. Similarly, when the amplitude component is 1.6 V, the path of the DC-DC converter 117 is turned on, and a voltage of 2.0 V is supplied to the voltage input terminal of the series regulator 129.

  The amplitude component output from the phase / amplitude separation unit 112 is input to the reference input terminal of the series regulator 129, and modulates the output voltage of the series regulator 129. At this time, since the series regulator 129 has a feedback loop inside, there is no need to offset the amplitude component.

  Further, it is desirable that the amplitude component is output in a form synchronized with the slice data.

  At this time, if the amplitude component is not synchronized with the slice data, an unnecessarily large voltage drop appears and the power loss deteriorates.

  By realizing such an operation, the voltage drop (potential difference between the output of the DC-DC converter and the output of the series regulator) of the series regulator 129 is kept at a small value, and the power loss due to the series regulator 129 is suppressed to a small value.

  Further, since the phase component needs to be frequency-converted into a modulated wave, it is input to the quadrature modulator 128 as an I (in-phase) signal and a Q (quadrature) signal, and is multiplied by the carrier.

  The high-frequency power amplifier 130 receives an amplitude component output from the series regulator 129 from a power supply terminal, and a phase component (modulated wave) output from the quadrature modulator 128 from a high-frequency signal input terminal. The result of multiplication of the phase component and the amplitude component is output to the output of the high-frequency power amplifier 130, and a correct OFDM output is obtained.

  When the amplitude component and the phase component are multiplied by the high-frequency power amplifier 130, it is desirable that there is no timing shift.

  The effects expected from the operation described above will be described below.

  It is assumed that the power loss in the DC-DC converters 116 to 120 is 96% and the voltage drop of the switches 122 to 127 is 0.1V. These values are based on data on parts actually available on the market. It is also assumed that the efficiency of the switch type high frequency power amplifier 130 is 80%.

    In the case of the wireless LAN IEEE802.11a standard, for example, the average output power can be assumed to be 13 dBm (20 mW). At this time, the peak power is 20 dBm (100 mW) at the average power of +7 dB. Therefore, the high-frequency power amplifier 130 needs to output a peak power of 20 dBm. Assuming that the power efficiency (RF output power / added DC power) of the high-frequency power amplifier 130 is 80%, when the AC power PAC is 100 mW (20 dBm) peak power, the DC power PDC is 125 mW. At this time, if the power supply is set to 3 V, a current of 41.7 mA is required at the peak. At the time of the average power, the power supply voltage required for the high-frequency power amplifier 130 is 1.3 V. However, the average output power of the AC power PAC is 20 mW (13 dBm), and the DC power PDC is 25 mW. Required.

  Hereinafter, the efficiency at the time of average power, that is, the efficiency at the output of 20 mW will be discussed.

  When examining the power loss of the power supply unit, first, since the slice data is cut every 0.5 V, the voltage drop in the series regulator 129 is 0.5 V at the maximum, and each switch (NPN) constituting the switch group 121 is further reduced. Assuming that the collector-emitter saturation voltage VCE of the transistors 122 to 127 is 0.1 V, the power loss in the switch group 121 and the series regulator 129 is calculated as 19.2 mA × 0.6 V = 11.5 mW.

If the power loss of the DC-DC converters 116 to 120 is 4%, the power loss of the DC-DC converters 116 to 120 is 25 mW × 0.04 = 1.0 mW.
It becomes.

Therefore, the total power loss of the switch group 121, the series regulator 129, and the DC-DC converters 116 to 120 is 11.5 mW + 1.0 mW = 12.5 mW
It becomes. As a result, the total efficiency at the time of peak power is 20 mW / (25 mW + 12.5 mW) = 53.3%
It becomes.

  When a normal linear amplifier is used, only 10% of the efficiency can be obtained at the maximum, but the efficiency can be greatly improved.

  Further, the voltage conversion unit that conventionally modulates the DC-DC converter is replaced with a switching regulator group (DC-DC converters 116 to 120) that outputs a constant voltage and a series regulator 129, thereby providing a DC-DC converter. Broadbanding, which was difficult with a DC converter alone, can be realized. The reason is as follows.

  That is, in the series regulator 129, it is not necessary to provide a low-pass filter for limiting the band, and the problem that the band is necessarily limited by the low-pass filter is solved, and other factors such as the transistor characteristics of the series regulator 129 or Is limited only by a band determined by a phase delay or the like caused by a feedback loop.

  These limiting elements can realize a band greatly exceeding the conventional band of 5 MHz, and can sufficiently cover a modulation band of 20 MHz such as a wireless LAN.

  Further, the output of the high frequency power amplifier 130 may have a band limiting filter.

  Further, the DC-DC converters 116 to 120 indicate those whose output includes a low-pass filter. In this configuration, a low-pass filter that suppresses spurious components outside the modulation wave band may be provided between the output of the series regulator 129 and the power supply terminal of the high-frequency power amplifier 130.

  Although it is desirable that the amplitude component and the amplitude slice data are synchronized, the output voltage of the series regulator 129 is adjusted so that the output voltage of the DC-DC converters 116 to 120 is not increased. If there is no problem. Further, for example, the slice data may be given a time margin in advance so that the above-mentioned state is not obtained even if there is a slight timing shift.

  Further, it is desirable that the amplitude component and the phase component are input in a state synchronized with the high-frequency power amplifier 130. However, if the timing is shifted, the vector error amount (Error Vector Magnitude) of the transmission output deteriorates, Do not satisfy wireless standards. Therefore, it is necessary to adjust the timing as much as possible by the following method.

  The first is a method of adjusting timing only at the time of manufacturing. This method does not require the provision of a feedback circuit or the like in the wireless circuit, and can be simplified. However, depending on the usage environment, synchronization may not be achieved.

  The second method is to adjust the timing only when the power is turned on. According to this method, it is possible to cope with the environment in which the power is turned on, and the synchronization can be obtained more reliably than the first method. However, there is a problem that communication cannot be performed for the time required for calibration.

  Further, as a third method, for example, in the case of TDD (time division multiplexing) such as a wireless LAN, transmission and reception are repeated alternately. In such wireless communication, a switching time between transmission and reception is used. There is a method of adjusting the timing. This is the most ideal because it can be adapted successively to the environment, but the calibration must be completed within the transmission / reception switching time specified by the wireless standard. Since the time is 1 μs or less in a wireless LAN, it is necessary to devise a device that ends in such a short time.

  Further, as a fourth method, the receiver is turned on at the time of transmission, and the transmission wave wrapping around from the antenna switch to the receiving unit is received and demodulated, and the timing of the amplitude component and the phase component is adjusted so that the bit error amount is minimized. There is a method to correct. In this method, if the isolation of the antenna switch is not sufficient, a large amount of power is input to the receiving unit. Therefore, it is necessary to increase the linearity of the receiving unit.

  A combination of these is also conceivable.

  In the present embodiment, the modulation circuit employs the direct modulation method of directly up-converting the baseband IQ signal into a high-frequency signal. However, a voltage-variable capacitance section of a voltage-controlled oscillator used as a local oscillation signal source, for example, a varactor A direct modulation method in which a baseband signal is shaped by a diode or a capacitor that realizes a variable capacitance by combining a fixed capacitance having a large number of capacitance values with MOS transistor switches, and may be directly modulated may be used.

  The direct modulation method simplifies the circuit form and can reduce current consumption, but is not suitable when the modulation accuracy is severe. Further, there is a method in which an IQ signal is not directly up-converted into a high-frequency signal, but is up-converted into a high-frequency signal via an intermediate frequency. In this method, since the frequency of the local oscillation signal source and the frequency of the transmission wave are different, the problem that the local oscillation signal source is fluctuated by the transmission wave can be avoided. However, it is disadvantageous in terms of current consumption and spurious.

  As described above, according to this embodiment, a plurality of DC-DC converters 116 to 120 for converting a power supply voltage into a plurality of voltages having different values in a stepwise manner are provided. And the series regulator 129 converts the amplitude component into a voltage using the output voltage of the selected DC-DC converter as a power supply voltage. Therefore, the voltage drop by the series regulator 129 during voltage conversion can be reduced, and the loss due to the DC-DC converter can be reduced, and the power loss due to the series regulator 129 can be reduced. Further, since the series regulator 129 is used for voltage conversion, and it is not necessary to use a low-pass filter in the output unit, a wider band can be achieved. Therefore, a wideband EER method can be realized without lowering the efficiency.

  Further, since the series regulator 129 is used as the linear voltage conversion means, the voltage level can be accurately controlled by the feedback loop, and the amplitude component can be correctly converted in voltage.

  Further, since the quadrature modulator 128 as the frequency converting means is provided between the output terminal of the phase component of the phase and amplitude separating means 112 and the input terminal of the high frequency power amplifier 130, the following effects can be obtained. Since the band of the phase / amplitude separating means 112 is at most several hundred MHz, when the carrier wave exceeds GHz, it cannot be processed. However, by using a frequency converting means such as the quadrature modulator 128, The carrier frequency can be easily up-converted.

(Embodiment 2)
FIG. 2 shows a block diagram of a transmitter according to Embodiment 2 of the present invention. This embodiment is different from the first embodiment in that an emitter follower 229 is used instead of the series regulator 129. Otherwise, the configuration and operation are the same as those of the sixth embodiment. The emitter follower 229 inputs the output of the switch group 121 to the collector, and supplies a power supply voltage to the high-frequency power amplifier 130 based on the voltage of the phase and amplitude separating means 112 input to the base.

  Hereinafter, the effect added by using the emitter follower 229 will be described. In a series regulator, a sufficient band may not be obtained due to a phase delay caused by a feedback loop. On the other hand, when the emitter follower 229 is used, it is limited in a band determined by transistor characteristics. However, this limiting element can realize a band greatly exceeding the conventional band of 5 MHz, and can sufficiently cover a modulation band of 20 MHz such as a wireless LAN.

(Embodiment 3)
FIG. 3 shows a block diagram of a transmitter according to Embodiment 8 of the present invention. In this embodiment, the outputs from the power supply and the DC-DC converter group 115 are directly connected to the collectors of the emitter follower group 329 of the same number as the output, and the bus connected to the base terminal of the emitter follower group 329 is connected to the emitter follower group 329. The second embodiment differs from the first and second embodiments in that the switching is performed by the same number of switch groups 121. The same components as those in the first and second embodiments are denoted by the same reference numerals, and description thereof is omitted. Note that the emitter follower group 329 corresponds to a linear voltage converter. Note that the switch group 121 is desirably constituted by NMOS transistors. Further, in the present embodiment, a plurality of emitter followers are used, but instead of this, an embodiment using a series regulator can be considered similarly to the above.

  An additional effect expected in the third embodiment is that only the emitter followers 330 to 335 are inserted between the DC-DC converter group 115 and the high-frequency power amplifier 130, and the switch group 121 is connected to the power supply path (from the power supply to the high-frequency power Since the connection is removed from the path to the amplifier 130), the power loss can be further reduced as compared with the configuration of the first embodiment.

(Embodiment 4)
Hereinafter, a fourth embodiment of the present invention will be described with reference to the drawings.

  FIG. 4 is a circuit diagram of a transmitter for realizing the EER method according to the fourth embodiment of the present invention.

  In the present embodiment, the following configuration is newly added to the configuration described in the first embodiment. That is, by adding a directional coupler 431 for extracting, for example, high-frequency power to the output of the high-frequency power amplifier 130, and extracting the amplitude component of the power extracted by the directional coupler 431 as feedback means by, for example, diode detection, A timing calibrating unit 433 is provided for comparing the amplitude component from the phase / amplitude separating unit 112 and calibrating the timing of the phase component and the amplitude component so as to minimize the error, based on the calibration data output from the timing calibrating unit 433. For example, a timing correction means 432 for correcting the timing of the phase component, for example, a delay circuit is provided.

  Other configurations and operations are the same as those in the first embodiment, and thus detailed description is omitted. 4, reference numeral 111 denotes an OFDM signal generation unit, reference numeral 113 denotes an amplitude slicing unit, reference numeral 114 denotes a switch driver, reference numeral 115 denotes a switching regulator group, and reference numerals 116 to 120 denote DC-DC. Reference numeral 121 denotes a switch group, reference numerals 122 to 127 denote switches, reference numeral 129 denotes a series regulator, and reference numeral 128 denotes a quadrature modulator.

  According to this embodiment, the following effects can be obtained. If the timings of the phase component and the amplitude component are shifted by the delay due to the layout from the input of each modulated wave component to the output of the high-frequency power amplifier 130, or the delay due to the wiring length of the transistor or the parasitic component, the original modulated wave is correctly corrected. Cannot be reproduced. However, by providing the directional coupler 431, the timing calibrating means 433, and the timing correcting means 432, the timing of the phase component and the amplitude component can be accurately corrected, and a correct modulated wave can be reproduced at the output of the high-frequency power amplifier. Other effects are the same as in the first embodiment.

(Embodiment 5)
Hereinafter, a fifth embodiment of the present invention will be described with reference to the drawings.

  FIG. 5 is a circuit diagram of a transmitter for realizing the EER method according to the fifth embodiment of the present invention.

  In the present embodiment, the following configuration is newly added to the configuration described in the fourth embodiment. That is, means for detecting a voltage, for example, a resistor 534 of several KΩ is added to a terminal of the series regulator 129 to which a DC-DC converter output is input, and means for detecting a voltage, for example, a resistor of several KΩ to a terminal to which an amplitude component is input. 535 is added.

  The resistance components 534 and 535 detect the level of the amplitude component and the output level of the DC-DC converter selected by the slice data. These voltage differences are calculated by the timing calibration means 433, and a timing adjustment signal is generated so that the output of the DC-DC converter becomes an appropriate voltage with respect to the amplitude component.

  Further, a timing correction unit 536 for correcting the timing to the amplitude slicing unit 113 based on the calibration data (timing adjustment signal) output from the timing correction unit 433, for example, a delay circuit is newly added.

  Other configurations and operations are the same as those in the first and fourth embodiments, and thus description thereof is omitted. In FIG. 5, reference numeral 111 denotes an OFDM signal generation unit, reference numeral 113 denotes an amplitude slicing unit, reference numeral 114 denotes a switch driver, reference numeral 115 denotes a switching regulator group, and reference numerals 116 to 120 denote DC-DC. Reference numeral 121 indicates a switch group, reference numerals 122 to 127 indicate switches, reference numeral 128 indicates a quadrature modulator, reference numeral 130 indicates a high-frequency power amplifier, reference numeral 431 indicates a directional coupler, and reference numeral Reference numeral 432 denotes a timing correction unit.

  According to this embodiment, the following effects can be obtained. If the timing of the amplitude slice data and the amplitude component is shifted by the delay due to the layout from the input of the amplitude slice data and the amplitude component to the output of the high-frequency power amplifier 130, or the delay due to the wiring length of the transistor or the parasitic component, the slice The output of the switching regulator group 115 turned on by the switch driven by the data greatly deviates from the value of the amplitude component, and the efficiency is reduced or the series regulator 129 is turned off. However, the provision of the resistors 534, 535, the timing calibrator 433, and the timing corrector 536 can correct the amplitude slice data and the timing of the amplitude component accurately, and can achieve ideal efficiency. Other effects are the same as those of the first or fourth embodiment.

(Embodiment 6)
Hereinafter, a sixth embodiment of the present invention will be described with reference to the drawings. In the present embodiment, a wireless LAN system using the IEEE802.11a standard using a wideband modulated signal will be described as an example. In a wireless LAN system, modulation of 64 QAM is applied to each of 52 orthogonal subcarriers, and this is added to obtain a modulated signal. The 52 subcarriers are separated by 312.5 kHz and occupy 52 × 312.5 = 16.25 MHz.

  FIG. 7 shows a circuit diagram of a transmitter for realizing the EER method according to the sixth embodiment of the present invention. As shown in FIG. 7, the transmitter includes OFDM signal generation means 611, amplitude extraction means 612, amplitude slicing means 613, switching regulator group 615, switch group 621, switch driver 614, quadrature modulator 628, a series regulator 629, and a switch type high frequency power amplifier 630.

  The above-described OFDM signal generation unit 611 corresponds to a modulation signal generation unit that generates a modulation signal.

  The amplitude extracting unit 612 extracts an amplitude component from the modulated signal generated by the OFDM signal generating unit 611.

  The amplitude slicing means 613 slices the amplitude component extracted by the amplitude extracting means 612 at a plurality of appropriate voltage levels that differ stepwise. As this voltage level, for example, 0.5 V, 1.0 V, 1.5 V, 2.0 V, 2.5 V, and 3.0 V are set. FIG. 7 shows an amplitude component input to the amplitude slicing means 613, ie, a source signal, and an output signal of the amplitude slicing means 613, ie, a slice signal.

  Here, the relationship between the source signal and the slice signal shown in FIG. 7 will be described. The amplitude slicing means 613 detects the amplitude level of the amplitude component as shown in FIG. 7, compares the level with a preset voltage level, and slices the amplitude component as shown in FIG.

  The method of the amplitude slicing is as follows. For example, the amplitude component is rounded to 1 V when 0.5 V <amplitude component ≦ 1.0 V, and rounded to 1.5 V when 1.0 V <amplitude component ≦ 1.5 V. Round level to value. In the figure, since there are a total of seven levels, these are assigned to 3-bit data, and 3-bit slice data is output to the switch driver 614.

  The switching regulator group 615 includes a plurality of, for example, five switching regulators to which a power supply voltage of, for example, 3 V is input, that is, five DC-DC converters 616 to 620. The DC-DC converters 616 to 620 convert the power supply voltage stepwise into a plurality of voltages having different values. Specifically, the DC-DC converters 616 to 620 convert a voltage of 3 V into respective voltages of 2.5 V, 2.0 V, 1.5 V, 1.0 V, and 0.5 V.

  The switch group 621 includes, for example, six switches 622 to 627, one of which is selectively turned on. The power supply voltage is 3 V, the output voltage of the plurality of DC-DC converters 616 to 620 is 2.5 V, One of the voltages of 2.0 V, 1.5 V, 1.0 V, and 0.5 V is selected. The switches 622 to 627 are formed of, for example, NPN transistors, and their base voltages are switched by, for example, MOS transistors, and switch the output of the DC-DC converter group 615.

  The switch driver 614 selectively turns on each of the switches 622 to 627 of the switch group 621 according to the slice data of the amplitude component sliced by the amplitude slicing unit 613.

  The quadrature modulator 628 multiplies the modulated signal output from the OFDM signal generation unit 611 by a carrier and converts the multiplied signal into a modulated wave, and corresponds to a frequency conversion unit.

  The series regulator (linear regulator) 629 converts the amplitude component of the modulation signal into a voltage using the 3V power supply voltage selected by the switch group 621 or the output voltage of any of the switching regulators 616 to 620 as a power supply voltage. It corresponds to a conversion means.

  The high-frequency power amplifier (PA) 630 is a switch type, inputs a modulated wave input from the quadrature modulator 628 to a high-frequency input terminal, inputs an amplitude component voltage-converted by the series regulator 629 to a power supply terminal, As a result, the amplified modulated wave is output.

  The operation will be described below. In the present embodiment, a system with a power supply voltage of 3 V is assumed.

  The amplitude component of the modulated signal generated by the OFDM signal generation unit 611 is extracted by the amplitude extraction unit 612 and output. Based on the output amplitude component, the amplitude slicing unit 613 generates drive information for driving on / off of each of the switches 622 to 627 of the switch group 621. The drive information is hereinafter referred to as slice data.

In the method of amplitude slicing, for example, the amplitude level of the amplitude component is 0 V <amplitude level ≦ 0.5 V
Then round it to 0.5V,
0.5V <amplitude level ≦ 1.0V
Then round it to 1.0V,
1.0V <amplitude level ≦ 1.5V
Then round it to 1.5V,
1.5V <amplitude level ≦ 2.0V
Then round it to 2.0V,
2.0V <amplitude level ≦ 2.5V
Then round to 2.5V,
2.5V <amplitude level ≦ 3.0V
Then, a threshold range in which the amplitude level is included is detected, such as rounding to 3.0 V, and the level is rounded to the maximum value of the included range.

  The rounding is performed as follows. The DC-DC converters 616 to 620 are prepared to output the same output voltage (2.5 V, 2.0 V, 1.5 V, 1.0 V, 0.5 V) as the rounded voltage level. According to the level of the amplitude component, the slicing means 613 gives the switch driver 614 information on which DC-DC converter 616 to 620 or the output of the 3V power supply is to be activated. According to the given information, the switch driver 614 selectively turns on / off the switches 622 to 627 provided in the output stage of the DC-DC converters 616 to 620 or the 3V power supply, and outputs a voltage corresponding to the rounded voltage. I do.

  To explain a specific example, when the amplitude component is 1.1 V, the path of the DC-DC converter 618 is turned on, and a voltage of 1.5 V is supplied to the voltage input terminal of the series regulator 629. Similarly, when the amplitude component is 1.6V, the path of the DC-DC converter 617 is turned on, and a voltage of 2.0V is supplied to the voltage input terminal of the series regulator 629.

  The amplitude component output from the amplitude extracting means 612 is input to the reference input terminal of the series regulator 629, and modulates the output voltage of the series regulator 629.

  Further, it is desirable that the amplitude component is output in a form synchronized with the slice data.

  At this time, if the amplitude component is not synchronized with the slice data, an unnecessarily large voltage drop appears and the power loss deteriorates.

  By implementing such an operation, the voltage drop (potential difference between the output of the DC-DC converter and the output of the series regulator) of the series regulator 627 is kept at a small value, and the power loss due to the series regulator 629 is suppressed to be small.

  Further, since the modulated signal needs to be frequency-converted into a modulated wave, it is input to the quadrature modulator 628 as an I (in-phase) signal and a Q (quadrature) signal, and is multiplied by the carrier.

  The high-frequency power amplifier 630 receives an amplitude component output from the series regulator 629 from a power supply terminal, and receives a modulated wave output from the quadrature modulator 628 from a high-frequency signal input terminal. Since a high-frequency power amplifier 630 uses a saturation type amplifier, its output is saturated. At the output of the high-frequency power amplifier 630, the amplitude component of the modulated wave is equalized and only the phase component is extracted. As a result, at the output of the high-frequency amplifier, a modulated output in which the phase component and the amplitude component are multiplied is output, and the original modulated wave is obtained.

  When the amplitude component and the phase component are multiplied by the high-frequency power amplifier 630, it is desirable that there is no timing shift.

  The effects expected from the operation described above will be described below. It is assumed that the power loss in the DC-DC converters 616 to 620 is 96% and the voltage drop of the switches 622 to 627 is 0.1V. These values are based on data on parts actually available on the market. It is also assumed that the efficiency of the switch type high frequency power amplifier 630 is 80%.

  In the case of the wireless LAN IEEE802.11a standard, for example, the average output power can be assumed to be 13 dBm (20 mW). At this time, the peak power is 20 dBm (100 mW) at the average power of +7 dB. Therefore, the high-frequency power amplifier 630 needs to output a peak power of 20 dBm. Assuming that the power efficiency (RF output power / added DC power) of the high-frequency power amplifier 630 is 80%, when the AC power PAC is 100 mW (20 dBm) peak power, the DC power PDC is 125 mW. At this time, if the power supply is set to 3 V, a current of 41.7 mA is required at the peak. At the time of average power, the power supply voltage required for the high-frequency power amplifier 630 is 1.3 V, but the average output power of the AC power PAC is 20 mW (13 dBm), and the DC power PDC is 25 mW, so a current of 19.2 mA is required. Become.

  Hereinafter, the efficiency at the time of average power, that is, the efficiency at the output of 20 mW will be discussed.

  When examining the power loss of the power supply unit, first, since the slice data is cut every 0.5 V, the voltage drop in the series regulator 629 is 0.5 V at the maximum, and each switch (NPN) constituting the switch group 621 is further reduced. Assuming that the collector-emitter saturation voltage VCE of the transistors 622 to 627 is 0.1 V, the power loss in the switch group 621 and the series regulator 629 is calculated as 19.2 mA × 0.6 V = 11.5 mW.

If the power loss of the DC-DC converters 616 to 620 is 4%, the power loss of the DC-DC converters 616 to 620 is 25 mW × 0.04 = 1.0 mW.
It becomes.

Therefore, the power loss of the switch group 621, the series regulator 629, and the DC-DC converters 616 to 620 is 11.5 mW + 1.0 mW = 12.5 mW
It becomes. As a result, the total efficiency at the time of peak power is 20 mW / (25 mW + 12.5 mW) = 53.3%
It becomes.

  When a normal linear amplifier is used, only 10% of the efficiency can be obtained at the maximum, but the efficiency can be greatly improved.

  Further, the DC-DC converter, which conventionally modulates the DC-DC converter, is configured by a switching regulator group (DC-DC converters 616 to 620) and a series regulator 629 that output a constant voltage, thereby providing a DC-DC converter. Broadbanding, which was difficult with a converter alone, can be realized. The reason is as follows.

  That is, in the series regulator 629, it is not necessary to provide a low-pass filter for limiting the band, and the problem that the band is necessarily limited by the low-pass filter is solved, and other factors such as the transistor characteristics of the series regulator 629 or feedback. It is limited by a band determined by a phase delay caused by a loop.

  These limiting elements can realize a band greatly exceeding the conventional band of 5 MHz, and can sufficiently cover a modulation band of 20 MHz such as a wireless LAN.

  Further, there may be a band limiting filter at the output of the high frequency power amplifier 630.

  Further, the DC-DC converters 616 to 620 refer to those including a low-pass filter in the output, and suppress spurious components outside the modulation wave band between the output of the series regulator 629 and the power supply terminal of the high-frequency power amplifier 630. There may be a low-pass filter.

  The amplitude component and the amplitude slice data are desirably synchronized. However, if the output voltage of the series regulator 629 is adjusted so as not to be higher than the output voltage of the DC-DC converters 616 to 620, it is preferable. There is no problem, and the slice data may be given a time allowance in advance so that the above-mentioned state is not achieved even if there is a slight timing shift.

  Further, it is desirable that the amplitude component and the phase component are input to the high-frequency power amplifier 630 in a synchronized state. However, if the timing is shifted, the vector error amount (Error Vector Magnitude) of the transmission output deteriorates, Do not satisfy wireless standards. Therefore, it is necessary to adjust the timing as much as possible by the following method.

  The first is a method of adjusting timing only at the time of manufacturing. With this method, it is not necessary to provide a feedback circuit or the like in the wireless circuit, and the wireless circuit can be simplified. However, depending on the usage environment, synchronization may not be achieved.

  The second method is to adjust the timing only when the power is turned on. According to this method, it is possible to cope with the environment in which the power is turned on, and the synchronization can be obtained more reliably than the first method. However, there is a problem that communication cannot be performed for the time required for calibration.

  Further, as a third method, in wireless communication in which transmission and reception are alternately performed, such as a wireless LAN, there is a method of adjusting timing using a preamble period at the start of transmission. This is the most ideal because it can be adapted successively to the environment, but the calibration must be completed within a sufficiently short time compared to the preamble. In the case of a wireless LAN, it needs to be completed in about 1 μs.

  Further, as a fourth method, the receiver is turned on at the time of transmission, and a transmission wave wrapping around from the antenna switch to the reception unit is received and demodulated, and the amplitude component and the phase are adjusted so that the vector error and the bit error are minimized. There is a method of correcting the component timing. In this method, if the isolation of the antenna switch is not sufficient, a large amount of power is input to the receiving unit. Therefore, it is necessary to increase the linearity of the receiving unit.

  A combination of these is also conceivable.

  As described above, according to this embodiment, a plurality of DC-DC converters 616 to 620 for converting a power supply voltage into a plurality of voltages having different values in a stepwise manner are provided. And the series regulator 629 converts the amplitude component into a voltage using the output voltage of the selected DC-DC converter as a power supply voltage. Therefore, a voltage drop by the series regulator 629 at the time of performing voltage conversion can be reduced, and a loss due to the DC-DC converter can be reduced, and a power loss due to the series regulator 629 can be reduced. Further, since the series regulator 629 is used for the amplitude, it is not necessary to use a low-pass filter in the output part, so that the band can be widened. Therefore, a wideband EER method can be realized without lowering the efficiency.

  Further, by using the series regulator 629 as the linear voltage conversion means, the voltage level can be accurately controlled by the feedback loop, and the level of the amplitude component can be correctly converted.

  Further, since the modulation signal is used as it is instead of the phase component, it is possible to avoid deterioration of the modulation accuracy (ErrorVectorMagnitude: EVM) which cannot be avoided in the EER method in which the amplitude and phase components are separated. That is, when the phase component is used, the phase component is filtered to the extent that the band of the digital-to-analog converter allows and does not affect the EVM. However, the partial amplitude reduction of the phase component caused by the filtering is as follows. When the phase component is combined with the amplitude component at the output of the high-frequency amplifier, the EVM is significantly deteriorated. In addition, since the required bandwidth of the modulated signal is about 1/6 smaller than the phase component separated from the modulated signal, the bandwidth of a digital-to-analog converter or an anti-alias filter that suppresses spurious components generated by the digital-to-analog conversion is used. Can be narrowed. This is advantageous in reducing the power consumption of the digital-to-analog converter and the cost of subsequent circuits.

  Further, in the conventional EER method, even when peak power is input, the input level sufficient to saturate the high-frequency power amplifier is injected, so that the isolation characteristics when the high-frequency power amplifier is OFF (amplitude component 0) are good. Otherwise, the multiplication with the amplitude component was not performed accurately, and the original modulated wave could not be restored (it caused deterioration of the EVM performance). In this configuration, when the high-frequency power amplifier is OFF (amplitude component is 0), the power input to the high-frequency power amplifier is also 0, so that a correct modulated wave can be restored without depending on the isolation characteristics.

  In the present configuration, the modulated signal is converted into a modulated wave using the quadrature modulator 628, but the quadrature modulator 628 becomes unnecessary when the OFDM signal generation unit 611 outputs a modulated wave. In this case, the amplitude extracting means 612 detects the amplitude of the modulated wave and extracts the amplitude component.

(Embodiment 7)
FIG. 8 shows a block diagram of a transmitter according to Embodiment 7 of the present invention. This embodiment is different from the sixth embodiment in that an emitter follower 729 is used instead of the series regulator 629. Otherwise, the configuration and operation are the same as those of the sixth embodiment. The emitter follower 729 inputs the output of the switch group 621 to the collector, and supplies a power supply voltage to the high-frequency power amplifier 630 based on the voltage of the amplitude extracting means 612 input to the base.

  Hereinafter, effects added by using the emitter follower 729 will be described. In a series regulator, a sufficient band may not be obtained due to a phase delay caused by a feedback loop. On the other hand, when the emitter follower 729 is used, it is limited in a band determined by transistor characteristics. However, this limiting element can realize a band greatly exceeding the conventional band of 5 MHz, and can sufficiently cover a modulation band of 20 MHz such as a wireless LAN.

(Embodiment 8)
FIG. 9 shows a block diagram of a transmitter according to Embodiment 8 of the present invention. In the present embodiment, the outputs from the power supply and the DC-DC converter group 615 are directly connected to the collectors of the same number of emitter follower groups 829, and the bus connected to the base terminal of the emitter follower group 829 is connected to the emitter follower group 829. This embodiment is different from the sixth and seventh embodiments in that the switching is performed by the same number of switch groups 621. The same components as those in the sixth and seventh embodiments are denoted by the same reference numerals, and description thereof is omitted. Note that the emitter follower group 829 corresponds to a linear voltage converter. Note that the switch group 621 is desirably formed of an NMOS transistor. Further, in the present embodiment, a plurality of emitter followers are used, but instead of this, an embodiment using a series regulator can be considered similarly to the above.

  An additional effect expected in the eighth embodiment is that only the emitter followers 830 to 835 are inserted between the DC-DC converter group 615 and the high-frequency power amplifier 630, and the switch group 621 is connected to the power supply path (from the power supply to the high-frequency power supply). Since it is removed from the path to the amplifier 630), the power loss can be further reduced as compared with the configuration of the sixth embodiment.

(Embodiment 9)
FIG. 10 is a block diagram of a main part of a transmitter according to Embodiment 9 of the present invention. In the transmitter of this embodiment, as shown in FIG. 10, the amplitude component output from the phase / amplitude separation means 112 is input to the operational amplifier 901, and the output of the operational amplifier 901 is input to the input (base) of the emitter follower 229 The output (emitter) of the emitter follower 229 is negatively fed back to the operational amplifier 901. Reference numerals 902 and 903 indicate bypass capacitors, respectively. Other configurations are the same as those in FIG.

  According to this embodiment, the nonlinearity and temperature characteristics of the emitter follower 229 can be compensated, and the amplitude component can be transmitted to the high-frequency power amplifier 130 correctly.

(Embodiment 10)
FIG. 11 shows a block diagram of a main part of a transmitter according to Embodiment 10 of the present invention. In the transmitter of this embodiment, as shown in FIG. 11, the emitter follower 910 is not constituted by only one transistor, but is constituted by a push-pull circuit. The transmitter inputs the amplitude component output from the phase / amplitude separating means 112 to the operational amplifier 901, connects the output of the operational amplifier 901 to the input of the emitter follower 910 which is a push-pull circuit, and connects the output of the emitter follower 910. The output is negatively fed back to the operational amplifier 901. Reference numerals 902 and 903 indicate bypass capacitors, respectively. Reference numerals 911 and 912 represent transistors, reference numerals 913 and 914 represent resistors, and reference numerals 915 and 916 represent diodes. These constitute an emitter follower. Other configurations are the same as those in FIG.

  According to this embodiment, the nonlinearity and temperature characteristics of the emitter follower 910 are compensated, and the voltage is held by the positive power supply voltage or the negative power supply voltage applied to the operational amplifier due to the transient characteristics of the operational amplifier 901. And the amplitude component can be transmitted to the high-frequency power amplifier 233 correctly.

(Embodiment 11)
FIG. 12 is a block diagram of a main part of a transmitter according to Embodiment 11 of the present invention. In the transmitter of this embodiment, as shown in FIG. 12, the emitter follower 920 is configured by a push-pull circuit. The transmitter inputs the amplitude component output from the phase / amplitude separating means 112 to the operational amplifier 901, connects the output of the operational amplifier 901 to the input of the emitter follower 920 which is a push-pull circuit, and connects the output of the emitter follower 920. The output is negatively fed back to the operational amplifier 901. Reference numerals 902 and 903 indicate bypass capacitors, respectively. Reference numerals 921 and 922 represent transistors, reference numeral 923 represents resistors, and reference numerals 924 and 925 represent diodes, which constitute an emitter follower. Other configurations are the same as those in FIG.

  According to this embodiment, the same functions and effects as those of the tenth embodiment are obtained.

  The configurations described in the ninth, tenth, and eleventh embodiments can also be applied to the transmitter circuit shown in FIG. 8, respectively. In this case, the ninth, tenth, and eleventh embodiments can be applied. The same effect as in the embodiment can be obtained.

  The transmitter according to the present invention has the effect of enabling wideband and high-efficiency operation in the EER method in which a high-frequency power amplifier can be operated as a switch type, and has an OFDM (Orthogonal Frequency Division Multiplex). ) Is useful as a transmitter of a communication method using subcarriers.

FIG. 2 is a block diagram illustrating a configuration of a transmitter according to Embodiment 1 of the present invention. FIG. 9 is a block diagram illustrating a configuration of a transmitter according to a second embodiment of the present invention. FIG. 11 is a block diagram illustrating a configuration of a transmitter according to a third embodiment of the present invention. FIG. 13 is a block diagram illustrating a configuration of a transmitter according to a fourth embodiment of the present invention. FIG. 14 is a block diagram illustrating a configuration of a transmitter according to a fifth embodiment of the present invention. FIG. 11 is a block diagram illustrating a configuration of a conventional transmitter. FIG. 17 is a block diagram illustrating a configuration of a transmitter according to a sixth embodiment of the present invention. FIG. 21 is a block diagram illustrating a configuration of a transmitter according to Embodiment 7 of the present invention. FIG. 19 is a block diagram illustrating a configuration of a transmitter according to an eighth embodiment of the present invention. FIG. 21 is a block diagram showing a configuration of a main part of a transmitter according to Embodiment 9 of the present invention. FIG. 21 is a block diagram showing a configuration of a main part of a transmitter according to Embodiment 10 of the present invention. FIG. 21 is a block diagram showing a configuration of a main part of a transmitter according to an eleventh embodiment of the present invention.

Explanation of reference numerals

111 OFDM signal generating means 112 Phase / amplitude separating means 113 Amplitude slicing means 114 Switch driver 115 Switching regulator group 116-120 DC-DC converter 121 Switch group 122-127 Switch 128 Quadrature modulator 129 Series regulator 130 High frequency power amplifier 229 Emitter follower 329 Emitter follower group 330 to 335 Emitter follower 431 Directional coupler 432 Timing correction means 433 Timing calibration means 534 Resistor 535 Resistor 536 Timing correction means 611 OFDM signal generation means 612 Amplitude extraction means 613 Amplitude slicing means 614 Switch driver 615 Switching regulator Group 616 to 620 DC-DC converter 621 switch group 622 to 6 7 switch 628 quadrature modulator 629 series regulator 630 high-frequency power amplifier 729 emitter follower 829 emitter follower groups 830-835 emitter follower 40 terminal 41 detector 42 voltage converter 43 the limiter 44 switched amplifier

Claims (15)

A modulation signal generating means for generating a modulation signal,
Phase and amplitude separating means for separating the modulated signal generated by the modulated signal generating means into a phase component and an amplitude component,
Amplitude slicing means for slicing the amplitude component separated by the phase and amplitude separating means at a plurality of voltage levels that differ stepwise;
A plurality of switching regulators for converting a power supply voltage into a plurality of voltages having different values in a stepwise manner;
A switch group for selecting any one of the output voltages of the plurality of switching regulators,
A switch driver for selectively conducting each switch of the switch group according to slice data of an amplitude component sliced by the amplitude slicing means;
Linear voltage conversion means for converting the amplitude component into a voltage using an output voltage of any of the switching regulators selected by the switch group as a power supply voltage;
A high-frequency power amplifier that inputs the phase component to a high-frequency input terminal, inputs the amplitude component voltage-converted by the linear voltage conversion means to a power supply terminal, and outputs a modulated wave obtained by multiplying the amplitude and the phase. Equipped transmitter. A modulation signal generating means for generating a modulation signal,
Phase and amplitude separating means for separating the modulated signal generated by the modulated signal generating means into a phase component and an amplitude component,
Amplitude slicing means for slicing the amplitude component separated by the phase and amplitude separating means at a plurality of voltage levels that differ stepwise;
A plurality of switching regulators for converting a power supply voltage into a plurality of voltages having different values in a stepwise manner;
A plurality of linear voltage conversion means for performing voltage conversion of the amplitude component using output voltages of the plurality of switching regulators as power supply voltages,
A switch group for transmitting the amplitude component to the plurality of linear voltage conversion means,
A switch driver for selectively conducting each switch of the switch group according to slice data of an amplitude component sliced by the amplitude slicing means;
A high-frequency power amplifier for inputting the phase component to a high-frequency input terminal, inputting the amplitude component voltage-converted by the plurality of linear voltage conversion means to a power supply terminal, and outputting a modulated wave whose amplitude and phase are multiplied as a result; And a transmitter with.   3. The transmitter according to claim 1, further comprising a frequency conversion unit between an output terminal of the phase component of the phase and amplitude separation unit and an input terminal of the high-frequency power amplifier. Feedback means provided at an output end of the high-frequency power amplifier to feed back high-frequency output power;
First timing calibration means for generating a first calibration signal for calibrating a phase shift between a phase and an amplitude based on a signal from the feedback means;
A first timing correction means for receiving a first calibration signal from the first timing correction means and correcting the timing of the amplitude component and the phase component output from the phase / amplitude separation means has been added. The transmitter according to claim 1, 2 or 3, wherein First voltage detection means provided between an output terminal of the switching regulator and a power supply voltage input terminal of the linear voltage conversion means to detect an output voltage of the switching regulator;
Second voltage detection means provided at an amplitude component input terminal of the linear voltage conversion means for detecting a voltage of the amplitude component;
A second timing for outputting a second calibration signal for calibrating a timing deviation between the amplitude component and the slice data by comparing the voltage amplitude data obtained from the first and second voltage detection means; Calibration means;
4. The apparatus according to claim 1, further comprising second timing correction means for receiving a second calibration signal from said second timing correction means and correcting the amplitude component and the timing of said slice data. 3. The transmitter according to 3, or 4.   6. The transmitter according to claim 1, wherein said linear voltage conversion means is an emitter follower.   6. The transmitter according to claim 1, wherein said linear voltage converter is a linear regulator.   The transmitter according to claim 6, wherein the amplitude component is input to an operational amplifier, an output of the operational amplifier is connected to an input of the emitter follower, and an output of the emitter follower is negatively fed back to the operational amplifier.   The emitter follower comprises a push-pull circuit, inputs the amplitude component to an operational amplifier, connects an output of the operational amplifier to an input of the push-pull circuit, and negatively feeds back an output of the push-pull circuit to the operational amplifier. The transmitter according to claim 6. A modulation signal generating means for generating a modulation signal,
Amplitude extraction means for extracting an amplitude component from the modulation signal generated by the modulation signal generation means,
Amplitude slicing means for slicing the amplitude component extracted by the amplitude extracting means at a plurality of voltage levels that differ stepwise;
A plurality of switching regulators for converting a power supply voltage into a plurality of voltages having different values in a stepwise manner;
A switch group for selecting any one of the output voltages of the plurality of switching regulators,
A switch driver for selectively conducting each switch of the switch group according to slice data of an amplitude component sliced by the amplitude slicing means;
Linear voltage conversion means for converting the amplitude component into a voltage using an output voltage of any of the switching regulators selected by the switch group as a power supply voltage;
A transmitter comprising: a high-frequency power amplifier that inputs the modulation signal to a high-frequency input terminal, inputs the amplitude component voltage-converted by the linear voltage conversion unit to a power terminal, and outputs a modulated wave as a result. A modulation signal generating means for generating a modulation signal,
Amplitude extraction means for extracting an amplitude component from the modulation signal generated by the modulation signal generation means,
Amplitude slicing means for slicing the amplitude component extracted by the amplitude extracting means at a plurality of voltage levels that differ stepwise;
A plurality of switching regulators for converting a power supply voltage into a plurality of voltages having different values in a stepwise manner;
A plurality of linear voltage conversion means for performing voltage conversion of the amplitude component using output voltages of the plurality of switching regulators as power supply voltages,
A switch group for transmitting the amplitude signal to the plurality of linear voltage conversion means,
A switch driver for selectively conducting each switch of the switch group according to slice data of an amplitude component sliced by the amplitude slicing means;
A transmitter comprising: a high-frequency power amplifier that inputs the modulation signal to a high-frequency input terminal, inputs the amplitude components voltage-converted by the plurality of linear voltage conversion units to a power supply terminal, and outputs a modulated wave as a result.   The transmitter according to claim 10 or 11, wherein the linear voltage conversion means is an emitter follower.   The transmitter according to claim 10 or 11, wherein the linear voltage conversion means is a linear regulator.   The transmitter according to claim 12, wherein the amplitude component is input to an operational amplifier, an output of the operational amplifier is connected to an input of the emitter follower, and an output of the emitter follower is negatively fed back to the operational amplifier. The emitter follower comprises a push-pull circuit, inputs the amplitude component to an operational amplifier, connects an output of the operational amplifier to an input of the push-pull circuit, and negatively feeds back an output of the push-pull circuit to the operational amplifier. The transmitter according to claim 12.

Images