Mixer device including an amplifier circuit

[0001] This application is a divisional of U.S. application Ser. No. 09/950,625 filed on Sep. 13, 2001, now U.S. Pat. No. 6,768,379, and is based upon and claims the benefit of priority from the prior Japanese Patent Application No. 2000-298278, filed Sep. 29, 2000, the entire contents of which are incorporated herein by reference.

[0002] 1. Field of the Invention

[0003] The present invention relates to an amplifier circuit, particularly to an amplifier circuit including a problem such that a distortion occurs in spread spectrum communication.

[0004] 2. Description of the Related Art

[0005] There will now be described a linear amplifier as one example of a conventional amplifier circuit. The linear amplifier inputs a low level signal, linearly amplifies the signal, and subsequently outputs a desired signal. In general, the linear amplifier comprises a differential amplifier. When the differential amplifier is supplied with a voltage signal and outputs a current signal, a load resistor converts the current signal to a voltage. A current gain ΔI₁ of the differential amplifier in voltage-to-current conversion is represented by the following equation using V_in as an input signal amplitude. $\begin{array}{cc}\Delta I_1=A·\tan h\left(\frac{V_{in}}{2V_T}\right)& \left(1\right)\end{array}$
Here A denotes a current value of a constant current source of a differential pair of transistors, and VT denotes a thermal voltage. When tanh_␣x is approximated, tanh_␣x≈x-x³/3 can be represented. Therefore, the equation (1) can be represented as follows. $\begin{array}{cc}\Delta I_1=A\left(\frac{V_{in}}{2V_T}-\frac{1}{3}{\left(\frac{V_{in}}{2V_T}\right)}^3\right)& \left(2\right)\end{array}$
Here a second term indicates a distortion component. Particularly a third-order intermodulation distortion (IM3) poses a largest problem as an adjacent channel leakage power of the signal in a spread spectrum radio system. The distortion is generated by presence of the term. Since VT is 26 mV at room temperature, and when IM3 is lowered to −60 dBc or less as a condition for distortion reduction, V_in needs to be 2.8 mV or less. Therefore, the aforementioned linear amplifier has a reduced distortion only when the input signal amplitude is very small. The linear amplifier cannot be utilized in an amplifier in which a signal to be handled is large in a range of 10 mVpp to 1 Vpp, particularly in a power amplifier.

[0006] The third-order intermodulation distortion as a main factor of distortion of the amplifier increases when an output power is increased. Therefore, in order to amplify the signal at the reduced distortion, a method of reducing the output power per one stage of the amplifier and obtaining a gain by a multi-stage structure of an amplifier is used. However, this poses problems such as an increase of power consumption, increase of a mounting area by an increase of the number of chips, and cost increase.

[0007] It is an object of the present invention to provide an amplifier circuit in which the third-order intermodulation distortion is inhibited without suppressing the output power.

[0008] According to a first aspect of the invention, there is provided an amplifier circuit comprising: a differential amplifier configured by a differential pair of transistors; a common emitter amplifier connected in parallel to the differential amplifier and configured by a pair of common-emitter configuration transistors; input and output terminals which are common to the differential amplifier and the common emitter amplifier, an input signal being input to the input terminals and an output signal output from the output terminal; and a bias controller configured to control a bias of at least one of the differential amplifier and the common emitter amplifier.

[0009] According to a second aspect of the invention, there is provided an amplifier circuit comprising: a first amplifier whose input-to-output characteristic indicates a hyperbolic tangent function characteristic; a second amplifier whose input-to-output characteristic indicates an exponential characteristic, the second amplifier being connected in parallel to the first amplifier; input and output terminals which are common to the differential amplifier and the common emitter amplifier; and a bias controller configured to control a bias of at least one of the first and second amplifiers.

[0026] According to an amplifier circuit of the present invention, the voltage signal input-to-current signal output characteristic of one of two signal amplifiers connected in parallel with each other indicates the exponential characteristic. The voltage signal input-to-current signal output characteristic of the other of the signal amplifiers indicates a hyperbolic tangent function characteristic. The amplifier of the hyperbolic tangent function characteristic can be realized by a differential amplifier as represented by the equation (1).

[0027] On the other hand, the amplifier of the exponential characteristic can be realized by common-emitter transistors. In this case, when a current gain during a certain biasing in the common-emitter amplifier is ΔI₂, the following equation results. $\begin{array}{cc}\Delta I_2=B·\exp \left(\frac{V_{in}}{2V_T}\right)& \left(3\right)\end{array}$
Here, B denotes a constant. When the common-emitter amplifier is used for a differential signal, and the exponential function is developed, the current gain is represented by the following equation. $\begin{array}{cc}\Delta I_2=2B\left(\frac{V_{in}}{2V_T}+\frac{1}{6}{\left(\frac{V_{in}}{2V_T}\right)}^3\right)& \left(4\right)\end{array}$
A second term of the equation (4) denotes IM3, but a coefficient is positive, while the coefficient is negative in the equation (2).

[0028] Since the signal amplifier circuit having the exponential characteristic and the signal amplifier circuit having the hyperbolic tangent function characteristic are connected in parallel with each other, a combined output current ΔI is represented as follows from the equations (2) and (4). $\begin{array}{cc}\Delta I=\Delta I_1+\Delta I_2=\left(A+2B\right)\left(\frac{V_{in}}{2V_T}\right)+\left(\frac{B-A}{3}\right)\left(\frac{V_{in}}{2V_T}\right)& \left(5\right)\end{array}$
On a condition of B=A, the following equation results. $\begin{array}{cc}\Delta I=3A\left(\frac{V_{in}}{2V_T}\right)& \left(6\right)\end{array}$
An ideal power amplifier is obtained in which a distortion term is completely cancelled and a desired signal is amplified.

[0029] Embodiments of the present invention will be described hereinafter with reference to the drawings.

[0030] FIG. 1 is a block diagram showing a basic configuration of an amplifier circuit according to a first embodiment. In FIG. 1 a signal is inputted via an input signal terminal V_in, and outputted via an output signal terminal V_out. The amplifier circuit of the first embodiment comprises two signal amplifier circuits connected in parallel with each other, and a signal input terminal and signal output terminal are common with the circuits. One of the two signal amplifiers is an amplifier (hereinafter referred to as “exponential circuit”) 101 whose signal input/output characteristic (voltage signal input current signal output characteristic) is substantially an exponential characteristic, and the other is an amplifier (hereinafter referred to as “tanh circuit”) 102 which substantially has a hyperbolic tangent function characteristic.

[0031] FIG. 2 is a circuit diagram showing a concrete configuration of the amplifier circuit of FIG. 1. A differential amplifier of bipolar transistors Q_T1 and Q_T2 corresponds to the tanh circuit, and a variable current source I₁ comprises a common current source of the differential pair of transistors. Common-emitter bipolar transistors Q_E1 and Q_E2 correspond to the exponential circuit, and respective emitters thereof are grounded via a variable voltage source V₁₁. Base terminals of the transistors Q_T1 and Q_E1 are connected to a signal input terminal D₁, and the bases of the transistors Q_T2 and Q_E2 are connected to a signal input terminal D₂. A differential signal is inputted via the terminals D₁ and D₂. Collector terminals of the transistors Q_T1 and Q_E1 are connected as a common terminal to a signal output terminal I_out1, and the collectors of the transistors Q_T2 and Q_E2 are connected as the common terminal to a signal output terminal I_out2.

[0032] It is assumed that a current value of the variable current source I₁ of the differential pair of transistors Q_T1 and Q_T2 is I_E1. In this case, a coefficient A of equation (1) substantially indicates a value of I_E1. On the other hand, it is assumed that a voltage value of the variable voltage source V₁₁ for determining emitter potentials of the common-emitter transistors Q_E1 and Q_E2 is V_E1. Moreover, the signal input terminals D₁ and D₂ are fixed at a predetermined bias potential V_B1, and a coefficient B of equation (3) indicates a value determined by V_B1 and −V_E1. The potential V_E1 is adjusted so as to be B=A. A distortion term therefore is cancelled as shown in equation (6), and a current characteristic having an inhibited third-order intermodulation distortion is obtained.

[0033] As described above, the voltage value V_E1 of the variable voltage source V₁₁ is changed, and the value is adjusted to be optimum so that the equation (6) is obtained. In an alternate method, an optimum value of an emitter size or an optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2, Q_E1 and Q_E2 in a design stage, and I₁, V₁₁ may be designed as fixed sources.

[0034] Additionally, the first embodiment has been described using the bipolar transistors Q_T1 and Q_T2 as the transistors comprising the differential amplifier corresponding to the tanh circuit, but the transistors comprising the differential amplifier are not limited to the bipolar transistors, and MOS type field effect transistors or Schottky junction type field effect transistors may be used. Moreover, the current source I₁ and voltage source V₁₁ correspond to bias controllers configured to control bias levels of the tanh circuit and exponential circuit, respectively.

[0035] FIG. 3 is a block diagram showing the basic configuration of the amplifier circuit according to a second embodiment. The same structure elements as those of FIG. 1 are denoted with the same reference numerals as those of FIG. 1. The second embodiment is different from the embodiment of FIG. 1 in that a bias level control circuit (bias controller) 201 is connected to the exponential circuit 101 and tanh circuit 102.

[0036] FIG. 4 shows another embodiment for realizing the concrete circuit configuration of FIG. 3. The transistors Q_T1 and Q_T2 comprise a differential transistor circuit, and the constant current source I₁ comprises a common current source of the differential pair of transistors. The transistors Q_E1 and Q_E2 are common-emitter transistors whose emitters are grounded. The bases of the transistors Q_T1 and Q_E1 are connected to the signal input terminal D₁ via capacitors C₁ and C₄, respectively, and the bases of the transistors Q_T2 and Q_E2 are connected to the signal input terminal D₂ via capacitors C₂ and C₃, respectively. The differential signal is inputted via the terminals D₁ and D₂. The collectors of the transistors Q_T1 and Q_E1 are connected as the common terminal to a signal output terminal O₂ via a cascode connection transistor Q_C1, and the collectors of the transistors Q_T2 and Q_E2 are connected as the common terminal to a signal output terminal O₁ via a cascode connection transistor Q_C2.

[0037] Respective bias potentials of the transistors Q_T1, Q_T2, Q_E1 and Q_E2 are supplied from the bias level control circuit 201. It is assumed that the bias potential supplied to the transistors Q_T1 and Q_T2 is V_B2, and the bias potential supplied to the transistors Q_E1 and Q_E2 is V_B3. The coefficient A of the equation (1) is determined by the value of the constant current source I₁ regardless of the value of V_B2. On the other hand, the coefficient B of the equation (3) depends on the value of V_B3. Therefore, the value of V_B3 is adjusted so as to be B=A. As a result, the current characteristic causing no third-order intermodulation distortion is obtained from the equation (6). Additionally, the current source I₁ may be a variable current source. In this case, the current value of the current source I₁ may be adjusted so as to be B=A, or both V_B3 and I₁ may be adjusted so as to be B=A.

[0038] As described above the voltage value V_B or the current value of the current source I₁ is changed by the bias level control circuit. The value is adjusted to be optimum so that the equation (6) is obtained. However, in the alternative method, the optimum value of the emitter size or the optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2, Q_E1 and Q_E2 in the design stage, and the bias level or I₁ is designed as the fixed source.

[0039] FIG. 5 is a circuit diagram showing a concrete configuration of the bias level control circuit 201 of FIG. 4. The bias level control circuit 201 comprises a voltage source V₂, variable current source I₂, resistors R₃, R₄, R₅, R₆ and R₇, and transistor Q_B1. The transistors Q_T1 and Q_T2 are biased at fixed voltages of V₂, R₆ and R₇. The bias levels of the transistors Q_E1 and Q_E2 are generated by a mirror circuit comprising the variable current source I₂, and resistors R₃, R₄ and R₅, and transistor Q_B1. The collector bias currents of the transistors Q_E1 and Q_E2 are a current proportional to the current value I_C1 of the variable current source I₂. The coefficient A of the equation (1) is determined by the current value of the constant current source I₁.

[0040] On the other hand, the coefficient B of the equation (3) depends on the current value I_C1. Therefore, the current value I_C1 is adjusted so as to be B=A, whereby the current characteristic causing no third-order intermodulation distortion is obtained from the equation (6). Additionally, the current source I₁ may be a variable current source. In this case, the current value of the current source I₁ may be adjusted so as to be B=A, or both V_B3 and I₁ may be adjusted so as to be B=A.

[0041] As described above the current value I_C1 or the value of the current source I_C1 is changed by the bias level control circuit, and the value is adjusted to be optimum so that the equation (6) is obtained. However, in the alternative method, the optimum value of the emitter size or the optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2, Q_E1 and Q_E2 in the design stage, and the bias level or I₁ is designed as the fixed source.

[0042] FIG. 6 is a circuit diagram showing another example of the amplifier circuit according to the second embodiment. Here, an inductor L₁ is used as a common load of the transistors Q_T1 and Q_E1, and connected to the output terminal O₁ via a capacitor C₂₂. A capacitor C₂₁ is used as the common load of the transistors Q_T2 and Q_E2, and connected to the output terminal O₁ via an inductor L₂. That is, O₁ is a common output terminal with respect to a differential signal, and a differential to single-phase conversion is achieved.

[0043] Assuming that a signal frequency is f, values of the inductance L₁ and L₂ and capacitance C₂₁ and C₂₂ satisfying the following equation are selected. $\begin{array}{cc}f=\frac{1}{2\pi \sqrt{L_1·C_{22}}}=\frac{1}{2\pi \sqrt{L_2·C_{21}}}& \left(7\right)\end{array}$
Then, a phase shifts by +/−90 degrees at the frequency f. As a result, an amplified signal with the same phase is outputted via the output terminal O₁.

[0044] FIG. 7 shows an output signal spectrum by a solid line when a two-tone high-frequency signal is inputted to the circuit of the present invention described with reference to FIG. 4. Moreover, a major signal spectrum of a general differential circuit as a conventional circuit is shown by a broken line. Here, it is assumed that the circuit (FIG. 4) of the present invention and the conventional circuit have the same power consumption. The input signal includes two tones of 2.05 GHz and 2.055 GHz, and an input signal power is −10 dBm. In the circuit (FIG. 4) of the present invention, the output signal indicates 350 mV. This is an output power of 0 dBm or more. The third-order intermodulation distortion is generated in frequencies of 2.045 GHz and 2.06 GHz. However, the third-order intermodulation distortion is of the order of 40 uV in the circuit (FIG. 4) of the present invention. Therefore, the distortion indicates −78 dBc, and is very small for a large signal output.

[0045] On the other hand, in the conventional circuit, the output is of the order of 150 mV and the output power is of the order of −6 dBm.

[0046] In this case, the third-order intermodulation distortion of −23 dBc is generated at 10 mV, and this is a level which causes a practical trouble. These results reveal that the circuit (FIG. 4) of the present invention increases the output power with respect to the desired signal, and provides an effect of remarkably reducing the distortion as compared with the conventional circuit.

[0047] FIG. 8 shows signal currents passed through the differential amplifier and common-emitter amplifier of the amplifier circuit when an output signal spectrum during input of two tones of high-frequency signal into the circuit of FIG. 4 is measured, and a signal spectrum of a current obtained by combining the signal currents by a common collector. Similarly as FIG. 7, assuming that the frequencies of the input signal are 2.05 GHz and 2.055 GHz, the third-order intermodulation distortion is measured at 2.045 GHz.

[0048] When the third-order intermodulation distortion is noted, large values of both an output current (1) of the common-emitter amplifier and an output current (2) of the differential amplifier are generated, but these values are substantially equal to each other. On the other hand, the value of the combined current of these currents is a value extremely lower than the value of each of the output currents. This result is similar to a principle in which the distortion is offset by parallel connection of the amplifier having the exponential characteristic and the amplifier having the hyperbolic tangent function characteristic in equation (5). The measurement result shows that the amplifier circuit can be realized in a practically useful state. Furthermore, for the spectrum of 2.05 GHz of the desired signal, the combined current indicates an added value of the respective signal currents of the common-emitter amplifier and differential amplifier, and both amplifiers effectively strengthen each other with respect to the desired signal.

[0049] FIG. 9 shows signal input/output characteristics of the circuit (FIG. 4) of the present invention and the conventional general differential circuit with the same power consumption. A solid line shows the desired signal and third-order intermodulation distortion characteristic in the circuit (FIG. 4) of the present invention, and a broken line shows the desired signal and third-order intermodulation distortion characteristic in the conventional circuit. The circuit (FIG. 4) of the present invention shows a result of adjustment of a bias in order to obtain an optimum third-order intermodulation distortion characteristic for each signal input power. According to the result, in the circuit of the present invention, a gain of 10 dB is obtained, and an output power of 10 dBm or more is obtained. Even in this case, the third-order intermodulation distortion is −70 dBc or less and a very satisfactory characteristic is obtained.

[0050] On the other hand, in the conventional circuit, the gain is as low as about 4 dB, the distortion is −10 dBc during output of 0 dBm, and the result is very bad as compared with the present invention.

[0051] FIG. 10 is a circuit diagram showing a circuit configuration in which the second embodiment is applied to a mixer circuit. The transistors Q_T1 and Q_T2 comprises the differential transistor circuit, and the constant current source I₁ comprises the common current source of the differential pair of transistors. The transistors Q_E1 and Q_E2 are common-emitter transistors and respective emitters thereof are grounded. The bases of the transistors Q_T1 and Q_E1 are connected to the signal input terminal D₁ via the capacitors C₁ and C₄, respectively, and the bases of the transistors Q_T2 and Q_E2 are connected to the signal input terminal D₂ via the capacitors C₂ and C₃, respectively. The differential signal is inputted via the terminals D₁ and D₂.

[0052] Transistors Q₁, Q₂ and Q₃, Q₄ comprise respective differential pair of transistors, the bases of the transistors Q₁ and Q₄ are connected to a local oscillation signal input terminal LO₁, and the bases of the transistors Q₂ and Q₃ are connected to a local oscillation signal input terminal LO₂. The differential signal is inputted via LO₁ and LO₂. The collectors of the transistors Q_T1 and Q_E1 are connected as the common terminal to the signal output terminals O₁ and O₂ via common emitter of the transistors Q₁ and Q₂, and the collectors of the transistors Q_T2 and Q_E2 are connected as the common terminal to the signal output terminals O₁ and O₂ via the common emitter of the transistors Q₃ and Q₄.

[0053] The respective bias potentials of the transistors Q_T1, Q_T2, Q_E1 and Q_E2 are supplied from the bias level control circuit 201. It is assumed that the bias potential supplied to the transistors Q_T1 and Q_T2 is V_B2, and the bias potential supplied to the transistors Q_E1 and Q_E2 is V_B3. The coefficient A of the equation (1) is determined by the value of the constant current source I₁ regardless of the value of V_B2. On the other hand, the coefficient B of the equation (3) depends on the value of V_B3. Therefore, the value of V_B3 is adjusted so as to be B=A, and the current characteristic causing no third-order intermodulation distortion is therefore obtained from the equation (6). Additionally, the current source I₁ may be a variable current source. In this case, the value of I₁ may be adjusted so as to be B=A, or both V_B3 and I₁ may be adjusted so as to be B=A.

[0054] As described above, the voltage value V_B3 or the value of the current source I₁ is changed by the bias level control circuit, and the value is adjusted to be optimum so that the equation (6) is obtained. However, in the alternative method, the optimum value of the emitter size or the optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2, Q_E1 and Q_E2 in the design stage, and the bias level or I₁ is designed as the fixed source.

[0055] FIG. 11 shows a circuit diagram of another embodiment comprising the circuit configuration of FIG. 2. The transistors Q_T1 and Q_T2 comprise the complementary transistor circuit, the emitters of the transistors Q_T1 and Q_T2 are connected to impedance elements Zd₁ and Zd₂, respectively, and the other ends of the impedance elements Zd₁ and Zd₂ are connected to the variable current source I₁. The variable current source I₁ comprises the common current source of the differential pair of transistors. The transistors Q_E1 and Q_E2 are common-emitter transistors, the respective emitters thereof are connected to impedance elements Zd₃ and Zd₄, respectively, and the other ends of the impedance elements Zd₃ and Zd₄ are grounded via the variable voltage source V₁₁. The bases of the transistors Q_T1 and Q_E1 are connected to the signal input terminal D₁, and the bases of the transistors Q_T2 and Q_E2 are connected to the signal input terminal D₂. The differential signal is inputted via the terminals D₁ and D₂. The collectors of the transistors Q_T1 and Q_E1 are connected as the common terminal to the signal output terminal I_out1, and the collectors of the transistors Q_T2 and Q_E2 are connected as the common terminal to the signal output terminal I_out2.

[0056] The current value of the variable current source I₁ of the differential pair of transistors Q_T1 and Q_T2 is set to I_E1 In this case, the coefficient A of the equation (1) substantially indicates a value of I_E1. On the other hand, the voltage value of the variable voltage source V₁₁ for determining the emitter potentials of the common-emitter transistors Q_E1 and Q_E2 is set to V_E1. Moreover, the signal input terminals D₁ and D₂ are fixed at the predetermined bias potential VB1, and the coefficient B of the equation (3) indicates a value determined by V_B1 and −V_E1. The value of V_E1 is adjusted so as to be B=A, and the current characteristic causing no third-order intermodulation distortion is obtained in the equation (6).

[0057] The impedance elements Zd₁, Zd₂, Zd₃ and Zd₄ function as degeneration elements in a used frequency band, VT in denominators of the equations (1) and (3) increases by a voltage generated at the degeneration elements, and therefore a fluctuation of ΔI to V_in can be reduced. As a result, even when V_in is used together with the system of the present invention and further increases, the distortion can be inhibited. Additionally, only the impedance elements Zd₁ and Zd₂ or Zd₃ and Zd₄ may be disposed.

[0058] As described above, the voltage value V_E1 of the variable current source V₁₁ is changed, and the value is adjusted to be optimum so that the equation (6) is obtained. However, in the alternative method, the optimum value of the emitter size or the optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2, Q_E1 and Q_E2 in the design stage, and I₁ and V₁₁ are designed as the fixed sources.

[0059] FIG. 12 is a circuit diagram of another embodiment comprising the circuit configuration of FIG. 4. The transistors Q_T1, Q_T2 comprise the complementary transistor circuit, the emitters of the transistors Q_T1, Q_T2 are connected to the impedance elements Zd₁ and Zd₂, respectively, and the other ends of the impedance elements Zd₁ and Zd₂ are connected to the constant current source I₁. The constant current source I₁ comprises the common current source of the differential pair of transistors. The transistors Q_E1 and Q_E2 are common-emitter transistors, and the respective emitters thereof are grounded via the impedance elements Zd₃ and Zd₄. The bases of the transistors Q_T1 and Q_E1 are connected to the signal input terminal D₁ via the capacitors C₁ and C₄, and the bases of the transistors Q_T2 and Q_E2 are connected to the signal input terminal D₂ via the capacitors C₂ and C₃. The differential signal is inputted via the terminals D₁ and D₂. The collectors of the transistors Q_T1 and Q_E1 are connected as the common terminal to the signal output terminal O₁ via the cascode connection transistor Q_C1, and the collectors of the transistors Q_T2 and Q_E2 are connected as the common terminal to the signal output terminal O₂ via the cascode connection transistor Q_C2.

[0060] The respective bias potentials of the transistors Q_T1, Q_T2, Q_E1 and Q_E2 are supplied from the bias level control circuit 201. The bias potential supplied to the transistors Q_T1 and Q_T2 is V_B2, and the bias potential supplied to the transistors Q_E1 and Q_E2 is V_B3. The coefficient A of the equation (1) is determined by the value of the constant current source I₁ regardless of the value of V_B2. On the other hand, the coefficient B of the equation (3) depends on the value of V_B3. Therefore, the value of V_B3 is adjusted so as to be B=A, and the current characteristic causing no third-order intermodulation distortion is therefore obtained from the equation (6). Additionally, the current source I₁ may be the variable current source. In this case, the value of I₁ may be adjusted so as to be B=A, or both V_B3 and I₁ may be adjusted so as to be B=A. The impedance elements Zd₁, Zd₂, Zd₃ and Zd₄ function as degeneration elements in the used frequency band, VT in denominators of the equations (1) and (3) increases by the voltage generated at the degeneration elements, and therefore the fluctuation of ΔI to V_in can be reduced. As a result, even when V_in is used together with the system of the present invention and further increases, the distortion can be cancelled.

[0061] As described above the voltage value V_B3 or the value of the current source I₁ is changed by the bias level control circuit, and the value is adjusted to be optimum so that the equation (6) is obtained. However, in the alternative method, the optimum value of the emitter size or the optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2, Q_E1 and Q_E2 in the design stage, and the bias level or I₁ is designed as the fixed source.

[0062] FIG. 13 is a circuit diagram of another embodiment in which the present invention is applied to the single-phase signal input circuit. The transistors Q_T1 and Q_T2 comprise the complementary transistor circuit, and the constant current source I₁ comprises the common current source of the differential pair of transistors. The transistor Q_E1 is the common-emitter transistor and the emitter thereof is grounded. The bases of the transistors Q_T1 and Q_E1 are connected to the signal input terminal D₁ via the capacitors C₁ and C₂, respectively, and the potential of the base of the transistor Q_T2 is fixed by the constant voltage source V₁. A single-phase signal is inputted via the terminals D₁. The collector of the transistors Q_T1 and Q_E1 is the common terminal having a load of an inductor L₁, and connected to the signal output terminal O1 via the capacitor C₂₂.

[0063] The respective bias potentials of the transistors Q_T1 and Q_E1 are supplied from the bias level control circuit 201. The bias potential supplied to the transistor Q_T1 is set to V_B2, and the bias potential supplied to the transistor Q_E1 is set to V_B3. The value of V_B2 is set to be substantially the same as that of the bias potential V₁ of the transistor Q_T2. The coefficient A of the equation (1) is determined by the value of the current flowing through a resistor R_T1 regardless of the value of V_B2. On the other hand, the coefficient B of the equation (3) depends on the value of V_B3. Therefore, the value of V_B3 is adjusted so as to be B=A, and the current characteristic causing no third-order intermodulation distortion is therefore obtained from the equation (6).

[0064] As described above, the voltage value V_B3 is changed by the bias level control circuit 201, and the value is adjusted to be optimum so that the equation (6) is obtained. However, in the alternative method, the optimum value of the emitter size or the optimum value of the number of transistors is estimated with respect to the transistors Q_T1, Q_T2 and Q_E1 in the design stage, and the bias level is designed as the fixed source.

[0065] FIG. 14 is a circuit diagram of another embodiment showing the concrete circuit configuration of FIG. 3. The transistors Q_T1 and Q_T2 comprise the differential transistor circuit, and a transistor Q10 for the current source and a resistor R8 comprise the common current source of the differential pair of transistors. The transistors Q_E1 and Q_E2 are common-emitter transistors, and the respective emitters thereof are grounded via transistors Q₉ and Q₁₁ connected via the diode. Moreover, the emitters of the transistors Q_E1 and Q_E2 are grounded via capacitors C₅ and C₈ in a high frequency. The bases of the transistors Q_T1 and Q_E1 are connected in common to the emitter of an emitter follower transistor Q₃. The bases of the transistors Q_T2 and Q_E2 are similarly connected in common to the emitter of the emitter follower transistor Q₁. The transistors Q₁ and Q₂, and resistor R₁ comprise an emitter follower circuit, and the base of the transistor Q₁ is connected to the signal input terminal D₂ via the capacitor C₂. Similarly, the transistors Q₃ and Q₄, and resistor R₂ comprise the emitter follower circuit, and the base of the transistor Q₃ is connected to the signal input terminal D₁ via the capacitor C₁. The differential signal is inputted via the terminals D₁ and D₂. The collectors of the transistors Q_T2 and Q_E1 are connected as the common terminal to the signal output terminal I_out1. The inductor L₁, capacitor C₆, and resistor R₇ are connected as a load in parallel with one another between the terminal I_out1 and power source V_CC. The collectors of the transistors Q_T2 and Q_E2 are connected as the common terminal to the signal output terminal I_out2 The inductor L₂, capacitor C₇, and resistor R₉ are connected as the load in parallel with one another between the terminal I_out2 and power source V_CC.

[0066] The respective bias potentials of the paired transistors Q_T1 and Q_E1, and Q_T2 and Q_E2 are controlled by base potentials of the emitter follower transistors Q₁ and Q₃, respectively. The base potential is controlled by the current I₂ flowing through a current mirror circuit connected to the transistors via resistors R₅ and R₄. Here, the current mirror circuit comprises the current source I₂, resistor R₆, transistors Q₆, Q₇ and Q₈, and capacitor C₄.

[0067] The current flowing through the transistors Q_T1 and Q_T2 is controlled by the base potential of the transistor Q10 for the current source, and the base of the transistor is connected to the current mirror circuit configured by a current source I₃, transistor Q₁₂, resistor R₁₀, and capacitor C₉. Therefore, the current flowing through the transistors Q_T1 and Q_T2 is controlled by the current value of the current source I₃. The coefficient A of the equation (1) is determined by a control current value of the current source I₃, and the coefficient B of the equation (3) is determined by the control current value of the current source I₂. Therefore, the current source I₃ is adjusted in order to obtain a desired output power, and the current source I₂ is adjusted in order to remove the distortion by the output power. Thereby, the condition of B=A is satisfied during outputting of a desired signal, the distortion term is cancelled from the equation (6), and the current characteristic can be obtained in which the third-order intermodulation distortion is cancelled.

[0068] The method of adjusting the current sources I2 and I3, changing the base bias levels of the transistors Q_T1, Q_E1, Q_T2 and Q_E2 and the current values of the transistors QT1 and QT2 and obtaining the optimum value in the equation (6) has been described above. However, there is another method of estimating the optimum value of the emitter size or the optimum value of the number of transistors with respect to the transistors Q_T1, Q_E1, Q_T2 and Q_E2 in the design stage, and using the current sources I₂ and I₃ as the fixed sources.

[0069] FIG. 15 is a block diagram showing the basic configuration of the amplifier circuit according to a third embodiment. In FIG. 15 the signal is inputted via the terminal V_in, and outputted via the terminal V_out. The amplifier circuit of the third embodiment comprises two amplifiers connected in parallel with each other, and the signal input terminal and signal output terminal are common with the circuits. One of the two amplifiers is the exponential circuit 101 whose signal input/output characteristic (voltage signal input-to-current signal output characteristic) is substantially the exponential characteristic, and the other is the tanh circuit 102 which substantially has the hyperbolic tangent function characteristic.

[0070] The present circuit configuration comprises the bias level control circuit 201 configured to control the bias voltages to be supplied to the two amplifiers, and a signal peak detector 301 that detects a peak value of the signal. At least one of a signal input section and signal output section inputs a signal value to the signal peak detector 301, and a control signal is inputted to the bias level control circuit 201 in accordance with the signal value. Based on this result, the bias voltages to be supplied to the two amplifiers are determined. According to the third embodiment, for the bias potential of the amplifier, even when the input signal power changes, an optimum bias potential is automatically supplied, and a satisfactory output signal with a suppressed distortion can be obtained.

[0071] FIG. 16 is a block diagram showing the basic configuration of the amplifier circuit according to a fourth embodiment. In FIG. 16 the signal is inputted via the terminal V_in, and outputted via the terminal V_out. The amplifier circuit of the fourth embodiment comprises two amplifiers connected in parallel with each other, in which the signal input terminal and signal output terminal are common with the circuits; and a variable amplifier 402 connected in cascode (tandem) with outputs of two amplifiers. One of the two amplifiers is the exponential circuit 101 whose signal input/output characteristic (voltage signal input current signal output characteristic) is basically the exponential characteristic, and the other is the tanh circuit 102 which basically has the hyperbolic tangent function characteristic.

[0072] The present circuit configuration comprises the bias level control circuit 201 configured to control the bias voltages to be supplied to the two amplifiers, and a gain controller 401 configured to control a gain. The control signal is inputted via a control signal input terminal CTRL_in, and a gain control signal is inputted to the bias level control circuit 201 and variable amplifier 402 in response to the inputted control signal. As a result, the bias voltage to be supplied to the two amplifiers is determined in accordance with the gain control. According to the fourth embodiment, for the bias potential of the amplifier, even when the output signal power changes, an optimum bias potential is constantly supplied, and the satisfactory output signal with the suppressed distortion can be obtained.

[0073] Additional advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.