Patent Application
20090088121
This disclosure relates to mixer circuits, for example, in radio frequency (RF) applications.
In general, in some aspects, a mixer circuit includes a first transistor including first and second terminals, in which the first terminal is configured to handle an input RF signal. There is a second transistor including a first terminal coupled to the second terminal of the first transistor, a second terminal configured to handle an input oscillator signal, and a third terminal configured to output an intermediate frequency (IF) signal, in which the IF signal includes a mixed product of the input RF signal and the input oscillator signal. A gate oxide thickness of the first transistor is less than a gate oxide thickness of the second transistor.
This and other implementations can optionally include one or more of the following features, which also may optionally be in any combination. The mixer circuit can include a single balanced architecture, which may be a folded mixer circuit. The mixer circuit can include at least one element to match an impedance of the first transistor with an RF source impedance, in which at least one element includes a capacitor, an inductor, a resistor and/or an impedance. The mixer circuit can have a first impedance-matching inductor and a second impedance-matching inductor coupled to the first transistor, where the first and second impedance matching inductors can be configured to match an impedance of the first transistor with an RF source impedance. The first terminal of the first transistor can be coupled to a first impedance-matching inductor, in which the first transistor includes a third terminal. The third terminal can be coupled to the second impedance-matching inductor, and the second impedance-matching inductor may be coupled to a ground terminal. The mixer circuit can include a double balanced architecture. The mixer circuit can be a folded mixer circuit. The mixer circuit can have at least one element to match an impedance of the first transistor with an RF source impedance. The mixer circuit can have a third transistor coupled to the first transistor to form a differential input for the input RF signal. The mixer circuit can include at least two transistors configured to handle a differential input oscillator frequency signal. The mixer circuit can include at least two pairs of transistors to handle a differential input RF signal. The at least two pairs of transistors can include a first set of cross-coupled terminals for handling the differential input RF signal, and a second set of cross-coupled terminals to output a differential intermediate frequency (IF) signal. The mixer circuit can have at least two power supply voltages, and the first transistor may be coupled to a lower power supply voltage than the second transistor. The mixer circuit can include at least two pairs of transistors configured to handle a differential input oscillator frequency signal. The gate oxide thickness of the first transistor can be less than 80 Angstroms. The mixer circuit can have metal-oxide-semiconductor field-effect transistor (MOSFET) devices. A gate oxide thickness of the first transistor can be less than a gate oxide thickness of any other transistors in the mixer circuit.
In other implementations, some aspects include a method for manufacturing a mixer circuit. The method includes coupling a first terminal of a first transistor to an input RF signal, where the first terminal of the first transistor is configured to handle the input RF signal. The method coupling a second terminal of the first transistor to at least one terminal of a second transistor, coupling a first terminal of the second transistor to the second terminal of the first transistor, coupling a second terminal of the second transistor to an input oscillator frequency signal. The second terminal of the second transistor is configured to handle the input oscillator frequency signal. The method involves coupling a third terminal of the second transistor to output an intermediate frequency (IF) signal, in which the third terminal of the second transistor is configured to handle the IF signal, and the IF signal includes a mixed product of the input RF signal and the input oscillator frequency signal. A gate oxide thickness of the first transistor is less than a gate oxide thickness of the second transistor.
This and other implementations can optionally include one or more of the following features, which also may optionally be in any combination. The method can include forming a gate oxide of the first transistor for a thickness of less than 80 Angstroms. The method can include forming the gate oxide of the first transistor for a thickness that is less than a gate oxide thickness of any other transistors in the mixer circuit. The method can include coupling a bias transistor to one or more nodes of the mixer circuit, and forming the bias transistor with a gate oxide thickness that is greater than the gate oxide thickness of the second transistor. The method can include coupling the transistors to form an architecture comprising one of a single balanced architecture, a single balanced folded architecture, a double balanced architecture, and/or a double balanced folded architecture. The method can include coupling the first transistor to a power supply terminal that is different from a power supply terminal of the second transistor. The method may involve coupling the first terminal of the first transistor to a first impedance-matching inductor, in which the first transistor includes a third terminal, coupling the third terminal of the first transistor to the second impedance-matching inductor, and coupling the second impedance-matching inductor to a ground terminal.
In general, some other implementations involve aspects for a method for a mixer circuit including at least one radio frequency (RF) device coupled to at least one switching device. The method involves receiving an input RF signal at a first RF device terminal of at least one RF device, in which the RF device includes a second RF device terminal coupled to a first switching device terminal of at least one switching device. The method involves receiving an input oscillator frequency signal at a second switching device terminal of at least one switching device, mixing the input RF signal and the input oscillator frequency signal to generate an intermediate frequency (IF) signal, and outputting the IF signal at a third switching device terminal of the switching device. A gate oxide thickness of the RF device is less than a gate oxide thickness of the switching device. The RF device and the switching device include metal-oxide-semiconductor field-effect transistor (MOSFET) devices.
This and other implementations can optionally include one or more of the following features, which also may optionally be implemented in various combinations. The gate oxide thickness of the RF device is less than 80 Angstroms. The mixer circuit can include an architecture including one of a single balanced architecture, a single balanced folded architecture, a double balanced architecture, and/or a double balanced folded architecture. The gate oxide thickness of the RF device can be less than a gate oxide thickness of any other device in the mixer circuit. The method can include utilizing the switching device as a switch when mixing the signals. The method can include utilizing a first power supply voltage for the RF device, and utilizing a second power supply voltage for the switching device, in which he first power supply voltage can be less than the second power supply voltage.
In general, some implementations include aspects for a mixer circuit that involves at least one radio frequency (RF) device that includes a first RF device terminal for handling an input RF signal, and a second RF device terminal coupled to at least one terminal of at least one switching device. The at least one switching device includes a first switching device terminal coupled to the second RF device terminal, a second switching device terminal for handling an input oscillator frequency signal, and a third switching device terminal to output an intermediate frequency (IF) signal, in which the IF signal includes a mixed product of the input RF signal and the input oscillator frequency signal. A gate oxide thickness of the at least one RF device is less than a gate oxide thickness of the at least one switching device. An impedance of the at least one RF device is matched with an RF source impedance. The at least one RF device and the at least one switching device comprise metal-oxide-semiconductor field-effect transistor (MOSFET) devices.
This and other implementations can optionally include one or more of the following features, which also may optionally be implemented in various combinations. The mixer circuit can include at least one element to match the impedance of the at least one RF device with the RF source impedance, in which at least the one element includes a capacitor, an inductor, a resistor and/or an impedance. The mixer circuit can have a first impedance-matching inductor and a second impedance-matching inductor, in which the first and second impedance matching inductors includes impedances to match an impedance of the at least one RF device with an RF source impedance. The first RF device terminal can be coupled to a first impedance-matching inductor. The at least one RF device can include a third RF device terminal, and the third RF device terminal can be coupled to the second impedance-matching inductor. The second impedance-matching inductor can be further coupled to a ground terminal. The mixer circuit can include at least two RF devices for a differential input RF signal, at least two switching devices for a differential input oscillator frequency signal, at least two pairs of switching devices, in which the at least two pairs of switching devices can include a first set of cross-coupled terminals for handling a differential input oscillator frequency signal, and a second set of cross-coupled terminals to output a differential intermediate frequency (IF) signal. The at least one radio frequency (RF) device can include a gate oxide thickness of less than 80 Angstroms, and a gate oxide thickness of the at least one RF device can be less than a gate oxide thickness of any other devices in the mixer circuit.
In general, other implementations include aspects for a method for manufacturing a mixer circuit. The method for manufacturing the mixer circuit includes coupling a first RF device terminal of at least one radio frequency (RF) device to handle an input RF signal, coupling a second RF device terminal of the at least one RF device to at least one terminal of at least one switching device, coupling a first switching device terminal of the at least one switching device to the second RF device terminal, coupling a second switching device terminal of the at least one switching device to handle an input oscillator frequency signal. The method also includes coupling a third switching device terminal of the at least one switching device to output an intermediate frequency (IF) signal, in which the IF signal includes a mixed product of the input RF signal and the input oscillator frequency signal, and forming a gate oxide of the at least one RF device for a thickness of less than 80 Angstroms. The gate oxide thickness of the at least one RF device is less than a gate oxide thickness of the at least one switching device.
In general, other implementations includes aspects for a receiver that includes an antenna to receive a radio frequency (RF) signal, an RF filter to filter the RF signal, at least one oscillator to generate an oscillator signal, at least one low noise amplifier (LNA) to amplify the filtered RF signal, at least one mixer to mix the amplified filtered RF signal with the oscillator signal, and a low pass filter to receive an output of the mixer and generate a baseband input signal. The mixer includes a first transistor that includes first and second terminals, in which the first terminal is configured to handle an input RF signal. The mixer includes a second transistor that includes a first terminal coupled to the second terminal of the first transistor, a second terminal configured to handle an input oscillator signal, and a third terminal configured to output an intermediate frequency (IF) signal. The IF signal includes a mixed product of the input RF signal and the input oscillator signal. A gate oxide thickness of the first transistor is less than a gate oxide thickness of the second transistor.
In general, other implementations includes aspects for a receiver. The receiver includes an antenna to receive a radio frequency (RF) signal, an RF filter coupled to the antenna to filter the RF signal, a low noise amplifier (LNA) coupled to the RF filter, a first oscillator coupled with a first mixer. The first mixer includes at least one first mixer input transistor and at least one first mixer switching transistor, and the at least one first mixer input transistor includes a gate oxide thickness that is thinner than a gate oxide thickness of the at least one first mixer switching transistor. The first mixer is configured to perform image rejection and mix an output signal of the LNA with an output signal of the first oscillator. The receiver also includes an intermediate frequency (IF) filter coupled to an output of the first mixer, an IF amplifier coupled to an output of the IF filter, and a second oscillator coupled to a second mixer. The second mixer includes at least one second mixer input transistor and at least one second mixer switching transistor, in which the at least one second mixer input transistor includes a gate oxide thickness that is thinner than a gate oxide thickness of the at least one second mixer switching transistor. The second mixer is configured to mix an output signal of the IF amplifier with an output signal of the second oscillator. The receiver also includes a low pass filter to filter an output of the second mixer and to generate a baseband input signal.
These and other implementations can optionally include one or more of the following features. The system can include any combination of one or more components for receivers, transmitters, and transceivers, in which a mixer can be coupled to any of the one or more components or their sub-components. Control circuits may include a digital circuit or a microprocessor. Any of the methods, designs, and techniques described herein can also be implemented in a system, an apparatus, a circuit, device, a machine, or in any combination thereof.
The details of one or more implementations of the disclosure are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the disclosure can be apparent from the description, drawings, and claims.
Many RF transceivers may require a mixer when converting frequencies. A “dynamic range” can be one characteristic that is addressed in mixer designs. In particular, the dynamic range may be referred to as a measurement or ratio between the maximum and minimum signals the mixer can process without significant distortion. Dynamic range may also refer to a range of power over which the mixer provides useful operation, or an amplitude range over which the mixer can operate without degradation of performance. In some implementations, the dynamic range can be bounded by a conversion compression point for high input signals, and by the noise figure (NF) of the mixer for low level input signals. Generally, the minimum input signal power that a mixer can process can be determined by the mixer's NF while the maximum input signal power a mixer can process is determined by the mixer's linearity.
The NF can be defined as the ratio of input signal-to-noise-ratio to output signal-to-noise-ratio, which may be expressed as shown below in equation (1):
In equation (1), S can represent the power of the signal, N can represent the power of the noise, and G can represent the circuit gain. Particularly, Nin can represent the input noise, and Nadded can represent the noise generated by the mixer itself. As Nadded decreases, the total NF also decreases.
Linearity may refer to a measurement of how precisely the output signal, either in terms of voltage or power, can track the input signal as a linear function of the input signal. There can be multiple parameters that may be used to evaluate the linearity of the mixer, such as, the 3rd-order input intercept point (IIP3), or the 1 dB gain compression point (P1dB). The linearity of the input device and of the output load together with the power supply may limit the overall linearity of the circuit. Due to a channel carriers' velocity saturation and mobility degradation, using short channel transistors for the input stage can result in better linearity. Also, with a fixed channel length, a higher power supply can result in a higher output swing while maintaining sufficient linearity.
Noise in Complementary Metal Oxide Semiconductor (CMOS) devices may be considered as being primarily attributable to thermal noise and flicker noise. Thermal noise can be caused by the random thermal motion of the channel carriers in the CMOS device. The power spectral density of the thermal noise may be expressed as shown below in equation (2):
In equation (2), K can represent Boltzmann's Constant (1.38×10−23 V·C/K), T can represent the absolute temperature, gd0 can represent the drain-source conductance at zero Vds, and y can represent a constant that has a value of about ⅔ for long channel devices. Thermal noise may sometimes be referred to as white noise because its power spectral density can be “flat” up to about 1012 Hz. Depending on where the CMOS device is located in the circuit, the operating transconductance can be altered to reduce the thermal noise.
Flicker noise can result from the release of trapped electrons or holes. In particular, dangling bonds may be present at the interface of the gate oxide at the end of the crystal lattice and the substrate of the CMOS transistor. When electrons or holes travel through this area, they can be trapped and then later released, introducing “flicker” noise in the drain current. The power of the flicker noise current may be expressed as shown below in equation (3):
In equation (3), Kp can represent a process-dependent constant, gm can represent the transconductance of the MOS transistor, W and L can represent the width and length of the transistor, respectively, COX can represent the gate oxide capacitance per unit area, and f can represent the flicker frequency. A spectral density of flicker noise can be inversely proportional to frequency.
At a given frequency, the power density of flicker noise is inversely proportional to the area of the MOS device. Thus, beyond altering frequency, the area of the MOS device may be altered to reduce flicker noise. However, increasing device size may introduce unwanted effects, such as, for example, increasing parasitic capacitance or slowing circuit operation. Other techniques, such as, chopper stabilization or correlated double sampling can reduce flicker noise at the price of increased complexity, high power consumption, or increasing thermal noise.
Communication systems with a direct conversion architecture and lower power supplies can have a small circuit size and low power consumption in RF transceivers, and therefore, can lower an overall cost. In direct conversion, the RF signal can be converted to the baseband signal directly to avoid the use of off-chip intermediate frequency (IF) filters. However, direct conversion systems can exhibit higher flicker noise than alternative systems. Moreover, the increased flicker noise may corrupt a weak signal before the signal is amplified. Lowering power supply may increase the possibility of creating frequency harmonics in the output signal, and thus, degrading linearity
For CMOS mixers, the input devices of the mixer can be the main contributor of noise to the mixer. As shown by equation (3) above, when Cox, the gate oxide capacitance, is increased, flicker noise decreases. Therefore, decreasing the thickness of the gate oxide layer results in increasing COX and decreasing flicker noise. Thus, by fabricating a thinner device at the input stage, the noise figure of a mixer can be lowered. CMOS devices in mixers may be fabricated with different gate oxide thickness in the same process. For example, in one fabrication technology, there can be at least three gate oxide thicknesses (e.g., on the order of 30 Angstroms, 40 Angstroms or 60 Angstroms) by some semiconductor manufactures. Other manufactures may have one or more gate oxide thicknesses in a fabrication process. In another technology, there can be gate oxide thickness of 30, 40, and 80 Angstroms, where the 30 Angstrom gate oxide is considered the thinnest gate oxide. When using, designing, or manufacturing multiple gate oxide thicknesses in the same circuit or silicon chip, the power applied to the thin gate oxide transistors can be used within the parameters specified by the manufacturer to avoid damage to the transistors. One example parameter that can be taken into consideration is the breakdown voltage limit of the thin oxide device.
The mixer 100 can be considered a single, balanced mixer 100 that mixes a single-ended input RF signal VRF and a differential local oscillator (LO) signal, (VLO+−VLO−). In the mixer 100, the Radio Frequency (RF) input signal VRF, which has a frequency of ωRF is applied to the input terminal 110 for the single ended input signal VRF. The MOS device 120 generates a current of the same frequency ωRF at the drain terminal 130 of device 120. The current from the device 120 is added to the bias current of device 150 and device 160 at a shared node 140. Devices 150 and 160 can be switching transistors with gate terminals driven by, for example, a differential square wave LO signal at frequency ωLO. The differential output signal is a differential intermediate frequency (IF) signal (VF+−VIF−). The differential output signal at output terminals 190, 195 from device 150 and device 160, respectively, are the mixed products of the RF signal VRF and the local oscillator (LO) signal (VLO+−VLO−) using mixed frequencies (ωRF+ωLO), (ωRF−ωLO) and ωLO. The output currents can be converted to voltages VIF+, and VIF− by resistors R1 170 and R2 180 at output terminals 190 and 195, respectively. In some implementations, a filter can follow the mixer 100 to filter unwanted signals.
In some implementations, the power supply voltage VDD of the mixer 100 may be designed to have a specified voltage breakdown value of the thinner oxide transistor, e.g., device 120. This requirement may be a factor in the performance of the devices 150 and 160. Device 185 can be a bias transistor controlled by the reference voltage Vbias.
The device 120 of the mixer input stage can be a device that has a shorter channel and a thinner gate oxide than the other devices 150, 160, 185 of the mixer to reduce the 1/f flicker noise in the signal that propagates through the mixer. The devices 150 and 160 can have a gate oxide thickness that is thicker than the gate oxide of device 120 to increase a gain of the output voltage. Devices 150 and 160 are switching transistors that may be less affected by the flicker noise, and thus, may be thicker oxide transistors such that higher power supply can be used to improve the mixer linearity. In some implementations, a 30 Angstrom gate oxide (e.g., a thin gate oxide) can be used for device 120 and a 60 Angstrom gate oxide (e.g., a thicker gate oxide) can be used for devices 150, 160, and 185. In other implementations, gate oxide thicknesses of 30, 40 and/or 60 Angstroms can be used to implement the mixer, in which the bias device 185 can have a different gate oxide thickness than the switching devices 150 and 160, while device 120 is implemented with 30 Angstrom gate oxide thickness.
The operation of the mixer circuit 300 can be similar to that of the mixer 100 of
The operation of the mixer 400 can be similar to that of the mixer circuit 300 of
The operation of the mixer 500 can be similar to that of the mixer circuit 200 of
In most of the above implementations, in order to improve the linearity at the output load, the power supply voltage can be selected according to the thick gate oxide devices because the thick gate oxide devices can have a higher breakdown voltage than the thinner gate oxide devices provided the voltage applied to the thinner gate oxide devices are lower than a breakdown voltage of the thinner gate oxide device. In some implementations, the thinner gate oxide devices can be coupled to a lower voltage power supply voltage than the thicker gate oxide devices. For example, in one technology, devices having a 60 Angstrom gate oxide thickness can operate at 2.5 V, and the devices having a 30 Angstrom gate oxide thickness can operate at 1.8V.
The disclosed techniques can be used with wireless communication systems. For example, the disclosed techniques can be used with receivers, transmitters, and transceivers for both wireless and wireline applications.
In operation, an RF signal arriving at the antenna 736 passes through the RF filter 737, and to the LNA 738. In some implementations, an image filter may be coupled between the LNA 738 and the first mixer 740 to produce a band-limited RF signal. The RF signal enters the first mixer 740, which translates the RF signal down to an intermediate frequency by mixing the RF signal with a signal produced by the first LO 741. Some or all undesired mixer products in the IF signal are rejected by an IF filter 742. The filtered IF signal then enters an IF amplifier 743, after which the outputs from the IF amplifier 743 feed into the second mixer 744. The second mixer 744 translates that signal down to yet another intermediate frequency by mixing the signal with the signal produced by a second LO 745. The first mixer 740 and/or the second mixer 744 can use the mixer techniques described above. The signal from the second mixer 744 is low-pass filtered in a low pass filter 739, and that filtered signal is sent to a baseband circuit 755 for processing. Tuning into a particular channel within the band-limited RF signal can be accomplished by varying the frequency of the local oscillators 741, 745.
In operation, the antenna 846 couples an input RF signal through the first bandpass RF filter 847 and into the LNA 848. The signal from the LNA 848 can proceed through a second RF filter 849. The RF filter 849 can generate a band-limited RF signal, which can enter the mixer 850 and to mix with a LO frequency produced by the LO 851. The mixer 850 may use any of the mixer techniques described above. The signal from the mixer 850 can be coupled into a low pass filter 852 before proceeding into a baseband circuit 855. Information sent to the baseband circuit 855 can be used by other components of a communications system. In some implementations, the low pass filter 852 may be an analog filter.
In some implementations, the positions of switches, resistors, or other components can be exchanged from the disclosed figures with minimal change in circuit functionality. Various topologies for circuit models can also be used, other than what is shown in the figures. The exemplary designs shown are not limited to CMOS process technology, but may also use other process technologies, such as BiCMOS (Bipolar-CMOS) process technology, or Silicon Germanium (SiGe) technology. In some implementations, switches can be implemented as transmission gate switches. The circuits can be single-ended or fully-differential circuits.
In some implementations, the techniques described in this disclosure can be used with radio architectures that support multiple communication standards, such as GSM/EDGE/WEDGE, and emerging standards, such as WiMAX, LTE, and UMB. The techniques in this disclosure can also be used with multi-band radios, GPS, RX Diversity, WLAN, and FM/DTV receivers.
Generally, implementations may involve using different gate oxide thickness in the input transistor for a mixer to improve its linearity and noise figure. In addition, the techniques described here can be used with any other analog/RF or digital algorithms used in communication systems. In one implementation, a circuit for processing radio frequency signals may include a first transistor with a first gate oxide thickness and a second transistor with a gate oxide thickness which is greater than the gate oxide thickness of the first transistor. The first transistor can be used in an input transconductance stage and the second transistor is used as part of a switched pair of transistors. In some implementations, the gate oxide thickness of the thinner gate oxide device can be in any range at or below 80 Angstroms (e.g., a range of 60 Angstroms and less, a range of 40 Angstroms and less).
The system can include other components, where the circuit can couple with those components. Some of the components may include computers, processors, clocks, radios, signal generators, counters, test and measurement equipment, function generators, oscilloscopes, phase-locked loops, frequency synthesizers, phones, wireless communication devices, and components for the production and transmission of data. The number and order of variable gain and filter stages can vary. In addition the number of controllable steps, as well as the steps sizes of each of the stages of gain can also vary. Other implementations can be within the scope of the following claims.
This application claims priority to U.S. Provisional Application No. 60/975,771, filed on Sep. 27, 2007, the disclosure of which is incorporated by reference.
| Number | Date | Country | |
|---|---|---|---|
| 60975771 | Sep 2007 | US |
