JP2003008355A - Multiband frequency converting circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to make conversion gains in a plurality of different frequency bands almost uniform without increasing a consumption current and deteriorating NF remarkably, in a frequency converting circuit including a plurality of mixers constituted of dual gate FETs. SOLUTION: This frequency converting circuit is equipped with a first mixer 61 for performing frequency conversion from a first RF signal to an IF signal, and a second mixer 62 for performing frequency conversion from a second RF signal different in frequency from the first RF signal to an IF signal whose frequency is identical to the above IF signal. Dual gate FETs (FET1, FET2) constituting the first and second mixers are made different in a gate length Lg, thereby making mutual conductances of the respective dual gate FETs different from each other. The conversion gains of frequency conversion of the first and second mixers are made almost equal. As a result, the almost equal conversion gains can be realized in different frequency bands without causing large deterioration of the NF, and adjustment is enabled by only forming process of a gate aperture of the dual gate FET, so that the number of manufacturing processes is not increased but merit is obtained from a process viewpoint.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a frequency conversion circuit used in a communication device, a mobile phone terminal, etc., and particularly to a multi-band frequency used for frequency conversion of high frequency signals of different frequency bands into the same intermediate frequency signal. It relates to a conversion circuit.

[0002]

2. Description of the Related Art At present, a plurality of mobile phone systems coexist in Europe and North America, and mobile phone terminals are shifting to dual band machines or tri-mode machines corresponding thereto. 80 of PDC in Japan
There is a movement to integrate the 0 MHz band and the 1.5 GHz band, and along with that, the development of the RF (high frequency) -IC (integrated circuit) that constitutes the front end part of the mobile terminal is changed from single mode to dual band 1 chip IC. It is progressing. FIG. 7 shows a block diagram of the RF receiver system of the dual band machine. RF received by antenna (ANT) 1
Signal is antenna SW2 at 800MHz band and 1.5GHz
Divided into bands, each with a low noise amplifier (LNA)
Amplified by 3 and further input to the dual band 1st mixer 6 via the filter system 4. In addition, 800MHz
In the band, the filter is switched by the switch 5, but the detailed description is omitted here. Then, each RF input to the dual band 1st mixer 6
The signals are frequency-converted into 1st intermediate frequency (IF) signals by mixers 61 and 62, respectively. The mixer 6
Reference numerals 1 and 62 denote VCOs (voltage controlled oscillators) not shown in the figure, respectively.
As a local oscillation (LO) signal from 680MHz,
The LO signal of 1.36 GHz is input through the buffers 63 and 64, and these LO signals and the above 800 MH
Each of the RF signals of z and 1.5 GHz is frequency mixed and 13
The frequency is converted into a 0 MHz signal. Then, the frequency-converted signal is input to a 2nd mixer (not shown) through the common filter 7 and frequency-converted into a 2nd intermediate frequency signal.

As described above, the conventional dual band 1st
In the mixer 6, the independent mixers 61 and 62 are used to
The frequency conversion is performed independently for each of the 0 MHz band and the 1.5 GHz band, but after the frequency conversion to 130 MHz, it is processed by a common single system. Therefore, it is necessary to set the levels of the IF signals converted in the 800 MHz band and the 1.5 GHz band to the same level, and to set the gains of the LNA 3 and the dual band 1st mixer 6 to the same level.

By the way, each of the mixers 61, 6 of the dual band 1st mixer 6 constituted by such a chip IC.
As 2, an active mixer having an equivalent circuit illustrated in the mixer 61 in FIG. 8A is used. This mixer is a field effect transistor (FE) having a dual gate structure.
T) is used, and the frequency is converted by utilizing the non-linearity of the dual gate FET. That is, the RF signal is input to the first gate G1 of the dual gate FET (FET), the LO signal is input to the second gate G2, the power supply voltage Vdd is input to the drain side, and the gate bias resistance Rg1 is input to the gates G1 and G2. By applying the gate biases Vg1 and Vg2 through Rg2, the IF signal is output through the matching circuit 65 connected to the drain side. In addition, the source side capacitor C and the resistor R
This constitutes a self-bias circuit that sets the source potential of the dual gate FET.

[0005]

However, since the mixer composed of such a dual gate FET has a frequency characteristic in conversion gain, the mixer 61, 62 of the above-mentioned dual band 1st mixer is used. When applied as, 800MHz band and 1.5GHz
There is a problem that the conversion gain differs between the band and the band. For example, in FIG. 8B, "HJ (heterojunction) -dual gate FET" manufactured by NEC Corporation is used, and the same condition (power supply voltage, bias, FET structure, measurement condition, etc.) is used.
It is the result of measuring the frequency characteristics when the signals of the 0 MHz band and the 1.5 GHz band are frequency-converted. From this result, the frequency characteristics of the mixer are 800MHz band and 1.5GH.
It can be seen that there is a difference in conversion gain of about 7 dB from the z band. Therefore, in the dual band 1st mixer using this dual band FET, the 800 MHz band and 1.
It is required that the conversion gain of each mixer in the 5 GHz band be similar.

In order to eliminate the difference in conversion gain with respect to such a frequency band and make the conversion gain approximately the same, the conversion gain in the 800 MHz band on the lower conversion gain side is lowered, or the 1.5 GHz band on the higher conversion gain side is formed. There are two ways to increase the conversion gain in. An example of arranging the conversion gains of two bands by increasing the conversion gain of the 1.5 GHz band is an IC called TQ9222 manufactured by TriQuint. This IC
Then, the bias of the mixer 62 in the 1.5 GHz band and 800M
The bias of the mixer 61 in the Hz band is changed to 1.5 GH.
800 by increasing the bias of the mixer 62 in the z band
The conversion gain is 17.5 dB in both the MHz band and the 1.5 GHz band. However 800M
While the current consumption in the Hz band is 10 mA, it is 1.5
The current consumption is 22 mA in the GHz band, and the current consumption is doubled. This is not a problem even in a system in which the frequency of use in the 1.5 GHz band is low, but in the system of PDC800M / 1.5G, voice is assigned to the 1.5 GHz band, and in this case, the current consumption in the 1.5 GHz band is large. The increase in the number becomes an obstacle to the realization of the system.

On the other hand, as a method of lowering the gain in the 800 MHz band, as shown in FIG.
A method of damping the gate on the side where the F signal is input can be considered. That is, the first gate G1 of the RF signal of the dual gate FET that constitutes the mixer of 800 MHz band.
A damping resistor RD of about several hundreds Ω is connected to this to reduce the bias. 800 MH with this technology
It is considered that the conversion gain of the mixer 61 in the z band can be lowered to eliminate the difference in the conversion gain from the 1.5 GHz band.
However, in actuality, there is a problem that the noise figure (NF) is deteriorated as much as the gain is lowered due to the damping on the RF input side, and the NF is significantly deteriorated. As described above, there are a number of methods of making the conversion gains of the 800 MHz band and the 1.5 GHz band to be approximately the same, but they have disadvantages such as a difference in current consumption or deterioration of NF. Is the current situation.

An object of the present invention is, in a frequency conversion circuit including a mixer composed of dual gate FETs, the conversion gains in a plurality of different frequency bands are substantially equal to each other without an increase in current consumption and a significant deterioration in NF. The present invention provides a multi-band frequency conversion circuit that can be aligned.

[0009]

According to the present invention, a plurality of mixers composed of dual gate FETs are provided, and signals of different frequency bands are frequency-converted to the same frequency by the plurality of mixers and output to the same output end. In the multi-band frequency conversion circuit, the dimensions of the dual gate FETs of the mixers are different from each other,
It is characterized in that the conversion gain of the frequency conversion of each dual gate FET is configured to be substantially the same. For example, the gate lengths of the plurality of dual gate FETs are different from each other, and the mutual conductances of the dual gate FETs are different from each other.

According to the present invention, a first mixer for frequency-converting a first RF signal into an IF signal and a second RF signal having a frequency different from that of the first RF signal are used as an IF signal having the same frequency as the IF signal. A second mixer for performing frequency conversion into a signal, wherein the dual-gate FETs constituting the first and second mixers have different transconductances, and
And the second mixer is preferably configured as a dual-band frequency conversion circuit in which the conversion gain of the frequency conversion of the second mixer is made substantially the same.

According to the present invention, the dual gate FET constituting the mixer in the frequency band having a high conversion gain is increased in gate length to reduce the mutual conductance, while the dual gate FET constituting the mixer in the frequency band having a low conversion gain is constituted. By decreasing the gate length of the gate FET and increasing the transconductance, it is possible to make the change gain of each mixer in both frequency bands approximately the same, and in a different frequency band without substantially degrading NF. Similar conversion gain can be realized. Further, in the present invention, the dual gate FE
Since the transconductance is adjusted only by changing the size of the gate length of T to adjust the conversion gain, the adjustment can be performed only by the opening process of the gate opening of the dual gate FET in the manufacturing process. It is also advantageous from the point of view of the process without causing an increase in

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 illustrates the present invention at 800 MHz / 1.
It is a block diagram configured as a dual band mixer 6 as a 5 GHz dual band frequency conversion circuit. This dual-band mixer 6 has an 800 MHz band and 1.
5 GHz band mixers 61 and 62 and the mixers 6
1 and 62 are configured as a one-chip IC including LO buffer amplifiers 63 and 64 for inputting LO signals, respectively. Then, 80 is applied to the first RF input terminal RF1.
Input the RF signal of 0MHz, and input the second RF input terminal RF2.
To the IF output terminal IF by inputting an RF signal of 1.5 GHz into the first LO input terminal LO1, inputting an LO signal of 680 MHz into the second LO input terminal LO2, and inputting a LO signal of 1.36 GHz into the second LO input terminal LO2. 130M
It is configured to output an IF signal of Hz. Here, the IF signal output from the IF output terminal IF is 80
It is configured such that the gain is about the same when frequency-converting either the 0 MHz band RF signal or the 1.5 GHz band RF signal.

FIG. 2 is an equivalent circuit diagram of the mixers 61 and 62 of the dual band mixer 6. 800M
The mixer 61 in the Hz band is the first dual gate FET (F
ET1), and the first gate G1 has the first
The second gate G2 is connected to the RF input terminal RF1 by the first LO.
Each is connected to the input terminal LO1. Further, the first and second gates G1 and G2 have resistors Rg1 and
Gate biases Vg1 and Vg2 are applied via Rg2, and a self-bias circuit in which a bypass capacitor C1 and a resistor R1 are connected in parallel is connected to the source side.
Similarly, the mixer 62 of the 1.5 GHz band is composed of a second dual gate FET (FET2),
The gate G1 is connected to the second RF input terminal RF2, and the second gate G2 is connected to the second LO input terminal LO2. In addition, the first and second gates G1 and G
2 to the gate bias V via resistors Rg3 and Rg4.
g3 and Vg4 are applied, and a self-bias circuit in which a bypass capacitor C2 and a resistor R2 are connected in parallel is connected to the source side. The drain sides of the first and second dual gate FETs (FET1, FET2) are connected in common and then connected to the IF output terminal IF via a matching circuit 65, and the power supply voltage Vd
1 is applied.

3 (a) and 3 (b) respectively show the above 800
The first dual-gate FET (FET) constituting the mixers 61 and 62 for the MHz band and the 1.5 GHz band, respectively.
FIG. 1 is a schematic cross-sectional view of 1) and a second dual gate FET (FET2). 800 MHz as shown in FIG.
The first dual-gate FET (FET1) in the strip has an i-AlGaAs layer 102, n-A on a semi-insulating substrate 101.
lGaAs layer 103, i-AlGaAs layer 104, i-
GaAs layer 105, i-AlGaAs layer 106, n ⁺ -
The GaAs layer 107 is laminated, and the i-GaAs
A gate opening 109 having a predetermined width is formed in the i-GaAs layer 105 at predetermined intervals so as to expose the surface of the i-AlGaAs layer 104 in the recess 108. Are provided, and two gate electrodes 110 and 111, which are Schottky-connected, are formed in these gate openings 109 to form a first gate G1 and a second gate G2. Also,
An ohmic-connected source electrode 112 and drain electrode 113 are provided on the surface of the n ⁺ -GaAs layer 107 that sandwiches the recess 108. Further, as shown in FIG. 3B, the second dual-gate FET in the 1.5 GHz band has the same configuration, and the same parts are denoted by the reference numeral in which the hundreds digit is replaced with "2". is there.

Here, in the 800 MHz band first dual gate FET (FET1), the gate length Lg1 of each of the gate electrodes 110 and 111 of the first gate G1 and the second gate G2 is set to 1.6 μm. But 1.5 GHz
In the second dual gate FET (FET2) in the band, the gate electrodes 2 of the first gate G1 and the second gate G2 are
The gate length Lg2 of 10,211 is set to 0.8 μm. By thus changing the dimensions of the dual gate FETs (FET1, FET2) forming the mixers 61, 62, the dual gate FETs (FE
By controlling the mutual conductance gm of T1, FET2), each dual gate FET (FET1, FET2)
Adjust the change gain of 2). Generally, in a FET, the transconductance gm decreases as the gate-source capacitance increases and the channel resistance increases as the gate length becomes longer when the gate widths are made equal. Therefore, the first dual-gate FET (FET1 ), It is possible to increase the transconductance gm of the second dual gate FET (FET2) having a small gate length, and consequently increase the conversion gain.

That is, the mixer using the dual gate FET operates like an amplifier having a SW (switch) that opens and closes with the voltage amplitude of the LO signal. FIG. 4 shows the transfer characteristics of the first dual gate FET (FET1). The horizontal axis represents Vg2 and the vertical axis represents the drain current Idd. The figure also shows the characteristics of the mutual conductance gm. If the LO signal is large enough, Vg2
The potential of fluctuates greatly with the voltage amplitude of the LO signal, and g
m is the pinch-off area and gm · as shown by the arrow in the figure.
Travel back and forth between the max regions. Generally, the conversion gain is g
It is known to depend on m · max, and the conversion gain can be controlled by controlling the gm curve. Therefore, the above-mentioned first dual gate FET (FET
In 1) and the second dual gate FET (FET2), the FE of the characteristic that the gm curve of FIG.
Since it is configured as T, both dual gate FETs (F
It becomes possible to control the conversion gain in ET1, FET2). The fact that the conversion gain depends on gm · max is also described in the following documents 1, 2, and 3. 1. Aikawa, Ohira et al., “Monolithic Microwave Integrated Circuit”, P116, IEICE, 1997. 2. Ueda, Ito et al., “High-frequency / optical semiconductor devices”, P1
29, Institute of Electronics, Information and Communication Engineers, 1999. 3. Yoichiro Takayama, “Microwave Transistor”, P23
2, IEICE, 1998.

Thus, the high conversion gain of 800 MHz
The gate length Lg of the first dual gate FET (FET1) in the band is increased to lower the transconductance gm, while the gate length Lg of the second dual gate FET (FET2) in the 1.5 GHz band having a low conversion gain is reduced. To increase the transconductance gm by increasing the first dual gate FET (FET1) for each frequency band.
It is possible to make the change gain of the second dual gate FET (FET2) substantially the same. FIG. 5 (a),
(B) shows the measurement results of an actual prototype. 800 MH without significant deterioration of NF with the same drain current
It has been confirmed that almost the same conversion gain is realized in the z band and the 1.5 GHz band.

Further, in the present invention, each dual gate FET is
Since the transconductance gm is adjusted to adjust the conversion gain only by changing the size of the gate length of (FET1, FET2), the photolithography process in the opening process of the gate opening to be opened in the semiconductor substrate in the manufacturing process is performed. Since the adjustment can be performed only by the mask size, it is advantageous from the viewpoint of the process without increasing the number of manufacturing steps.

In the first embodiment, since the RF signal of each frequency of 800 MHz or 1.5 GHz is converted into the IF intermediate frequency of 130 MHz, the first and second dual gate FETs (FET1 and FET2) are converted. Each of the bypass capacitors C1 and C2 of the self-bias circuit connected to the source side must appear as a short circuit at 130 MHz, and therefore each of the capacitors C1 and C2 is
Requires a large capacitance of about 100 pF, and a total capacitance of about 200 pF is required. When a capacitor having such a capacitance is formed on a chip IC, a 200 pF capacitor requires a considerably large area even if a high dielectric film is used, which is an obstacle in realizing a small chip IC. Therefore, in the second embodiment, the dual band mixer 6 has the circuit configuration shown in FIG. In this circuit, the first and second dual gate FETs (FE
T1, FET2) is a differential pair, and the source sides of both dual gate FETs (FET1, FET2) are connected in common, and the current source Ix and the bypass capacitor Cx are connected to the common source side. ing.

In the second embodiment as well, 8
00MHz band first dual gate FET (FET
The gate length of 1) and the gate length of the second dual gate FET (FET2) in the 1.5 GHz band are set to appropriate dimensions in consideration of transconductance gm and conversion gain. The gate biases Vg1, Vg2, V
g3 and Vg4 are connected to a control logic circuit, and switched to the first or second dual gate FET by the control terminal. This is 8 on the system
This is because the 00 MHz band and the 1.5 GHz band are not used at the same time. With this configuration, the conversion gain of the frequency conversion of each RF signal in the 800 MHz band and the 1.5 GHz band can be made approximately the same as in the first embodiment, and a large area is required on the chip IC. The capacity of the bypass capacitor Cx can be reduced to half, and the dual band mixer can be downsized.

Here, in each of the above embodiments, 800 MH
An example is shown in which the present invention is applied to a dual-band mixer that frequency-converts z-band and 1.5 GHz-band RF signals, but the present invention is also applied to a dual-band frequency conversion circuit that frequency-converts an RF signal different from this embodiment. The same can be applied. Further, it can be applied to a part or the whole of a multi-band frequency conversion circuit that frequency-converts three or more different RF signals.
In this case, it goes without saying that the gate length dimension of the dual gate FET as a mixer in each frequency band is set to an appropriate dimension corresponding to the frequency of each RF signal. Further, it is also possible to adopt an FET having a configuration different from that of the above-described embodiment regarding the configuration of the dual gate FET.

[0022]

As described above, the present invention is provided with a plurality of mixers composed of dual-gate FETs, and the signals of different frequency bands are frequency-converted to the same frequency by the plurality of mixers, respectively, and output to the same output terminal. In the output multi-band frequency conversion circuit, the mutual conductance changes corresponding to the gate lengths of the dual gate FETs that form a plurality of mixers to take advantage of the high conversion gain frequency band and the low conversion gain frequency band. It is possible to make each transconductance different and, as a result, make the conversion gain of the frequency conversion by each dual gate FET substantially the same. As a result, a large increase in operating current and NF
It is possible to make the conversion gain of the frequency conversion the same in a plurality of different frequency bands without causing deterioration of the device, and the number of steps for manufacturing the device does not increase, and there is no process demerit.

[Brief description of drawings]

FIG. 1 is a block diagram of a dual band mixer IC according to the present invention.

FIG. 2 is a circuit diagram of a dual band mixer IC according to the first embodiment of the present invention.

FIG. 3 is a cross-sectional view of a dual gate FET.

FIG. 4 is a transfer characteristic diagram of a dual gate FET.

FIG. 5 is a diagram showing comparison of conversion gains in respective mixers.

FIG. 6 is a circuit diagram of a dual band mixer IC according to a second embodiment of the present invention.

FIG. 7 is a block diagram of an RF receiving unit of a dual band frequency conversion circuit.

FIG. 8 is a diagram showing a mixer circuit diagram and a conversion gain.

FIG. 9 is a circuit diagram illustrating a method of controlling a conversion gain of a mixer.

[Explanation of symbols]

1 antenna 2 antenna SW 3 Low noise amplifier 4 filter system 5 switches 6 dual band mixer 7 filters 61 800MHz band mixer 62 1.5 GHz band mixer 63, 64 buffer amplifier 65 Matching circuit FET1 1st dual gate FET FET2 Second dual gate FET

Claims (6)

[Claims] 1. A multi-band frequency conversion circuit comprising a plurality of mixers configured of dual gate FETs, wherein signals of different frequency bands are frequency-converted by the plurality of mixers to the same frequency and output to the same output end,
The multi-band frequency conversion circuit is characterized in that the plurality of mixers have different dimensions of respective dual gate FETs, and the conversion gains of frequency conversion of the respective dual gate FETs are substantially the same. 2. The dual-gate FET has a first gate connected to a frequency signal input end, a second gate connected to a local oscillation signal input end, and a drain-side output of a frequency-converted signal. The multi-band frequency conversion circuit according to claim 1, wherein the ends are connected to each other, and the dual gate FETs of the plurality of mixers are commonly connected to the output ends on the respective drain sides. 3. The multiband frequency conversion circuit according to claim 1, wherein the plurality of dual gate FETs have different gate lengths, and the dual gate FETs have different transconductances. 4. A frequency signal input to one mixer of the plurality of mixers has a higher frequency than a frequency signal input to another mixer, and a gate length of a dual gate FET of the one mixer is the above-mentioned. The multi-band frequency conversion circuit according to claim 3, wherein the multi-band frequency conversion circuit is formed longer than the gate length of a dual gate FET of another mixer. 5. The dual gate F of each of the plurality of mixers
5. The multiband frequency conversion circuit according to any one of claims 2 to 4, wherein the ETs are commonly connected on the source side and shared with a bias circuit including a capacitor connected to the source side. 6. A first mixer for frequency-converting a first RF signal into an IF signal, and a second RF signal having a frequency different from that of the first RF signal by an I having the same frequency as the IF signal.
A second mixer for performing frequency conversion into an F signal, and each dual gate FET constituting the first and second mixers
6. The dual-band frequency conversion circuit according to claim 1, wherein the first and second mixers have different transconductances and the conversion gains of the frequency conversions of the first and second mixers are substantially the same. The multi-band frequency conversion circuit described in.