CA2690442A1 - Distortion cancellation method and circuit - Google Patents
Distortion cancellation method and circuit Download PDFInfo
- Publication number
- CA2690442A1 CA2690442A1 CA2690442A CA2690442A CA2690442A1 CA 2690442 A1 CA2690442 A1 CA 2690442A1 CA 2690442 A CA2690442 A CA 2690442A CA 2690442 A CA2690442 A CA 2690442A CA 2690442 A1 CA2690442 A1 CA 2690442A1
- Authority
- CA
- Canada
- Prior art keywords
- intermodulation distortion
- circuit
- output
- primary
- intermodulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 239000000872 buffer Substances 0.000 claims description 33
- 230000001965 increasing effect Effects 0.000 claims description 6
- 230000006866 deterioration Effects 0.000 abstract 1
- 238000013461 design Methods 0.000 description 13
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 230000006872 improvement Effects 0.000 description 5
- 229910000577 Silicon-germanium Inorganic materials 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 230000010363 phase shift Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 229910001218 Gallium arsenide Inorganic materials 0.000 description 1
- 101001091538 Homo sapiens Pyruvate kinase PKM Proteins 0.000 description 1
- 102100034911 Pyruvate kinase PKM Human genes 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000005669 field effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000003071 parasitic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
- H03F3/21—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only
- H03F3/211—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only using a combination of several amplifiers
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
- H03F1/08—Modifications of amplifiers to reduce detrimental influences of internal impedances of amplifying elements
- H03F1/22—Modifications of amplifiers to reduce detrimental influences of internal impedances of amplifying elements by use of cascode coupling, i.e. earthed cathode or emitter stage followed by earthed grid or base stage respectively
- H03F1/223—Modifications of amplifiers to reduce detrimental influences of internal impedances of amplifying elements by use of cascode coupling, i.e. earthed cathode or emitter stage followed by earthed grid or base stage respectively with MOSFET's
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
- H03F1/32—Modifications of amplifiers to reduce non-linear distortion
- H03F1/3205—Modifications of amplifiers to reduce non-linear distortion in field-effect transistor amplifiers
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/189—High-frequency amplifiers, e.g. radio frequency amplifiers
- H03F3/19—High-frequency amplifiers, e.g. radio frequency amplifiers with semiconductor devices only
- H03F3/193—High-frequency amplifiers, e.g. radio frequency amplifiers with semiconductor devices only with field-effect devices
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/50—Amplifiers in which input is applied to, or output is derived from, an impedance common to input and output circuits of the amplifying element, e.g. cathode follower
- H03F3/505—Amplifiers in which input is applied to, or output is derived from, an impedance common to input and output circuits of the amplifying element, e.g. cathode follower with field-effect devices
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03D—DEMODULATION OR TRANSFERENCE OF MODULATION FROM ONE CARRIER TO ANOTHER
- H03D2200/00—Indexing scheme relating to details of demodulation or transference of modulation from one carrier to another covered by H03D
- H03D2200/0041—Functional aspects of demodulators
- H03D2200/0088—Reduction of intermodulation, nonlinearities, adjacent channel interference; intercept points of harmonics or intermodulation products
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- Nonlinear Science (AREA)
- Amplifiers (AREA)
Abstract
This invention provides methods and circuits for reducing or cancelling intermodulation distortion in a preceding circuit, using the output of the preceding circuit.
In one embodiment the method includes generating, from the circuit output (i) a primary output including the intermodulation distortion and a fundamental signal of the preceding circuit;
and (ii) an auxiliary output, including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs; such that the portion of the intermodulation distortion is reduced or cancelled. In one embodiment the methods and circuits improve third-order linearity and attenuate or cancel odd-order intermodulation distortion products with minimal deterioration in the noise figure of the preceding circuit.
In one embodiment the method includes generating, from the circuit output (i) a primary output including the intermodulation distortion and a fundamental signal of the preceding circuit;
and (ii) an auxiliary output, including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs; such that the portion of the intermodulation distortion is reduced or cancelled. In one embodiment the methods and circuits improve third-order linearity and attenuate or cancel odd-order intermodulation distortion products with minimal deterioration in the noise figure of the preceding circuit.
Description
DISTORTION CANCELLATION METHOD AND CIRCUIT
Field of the Invention This invention relates generally to distortion cancellation in electronic circuits. In particular, this invention relates to distortion cancellation applied to a circuit output, to reduce or cancel intermodulation distortion products in the circuit.
Background The intermodulation distortion (IMD) of an electronic circuit such as an amplifier, mixer, or other circuit is an important consideration in many applications. The third-order intermodulation product intercept point (IP3) is commonly used to characterize IMD. This is based on the typical situation where the desired signal and an interference signal are nearby in the frequency domain. Non-linearities in the circuit generate new signals that can cause interference with the desired signal. Third-order nonlinearities, in particular, are often the most important because they generate intermodulation distortion products that are close to the desired signal and cause interference, and often cannot be removed by filtering.
Several methods have been suggested that can potentially improve the linearity of a circuit such as an amplifier. These include predistortion, feedback, feedforward, optimal biasing, and derivative superposition.
Predistortion techniques rely on the ability to distort the input signal to a circuit such that it is the inverse of the circuit's nonlinearity, and are often used in power amplifiers (e.g., [1]) and mixers. Active feedback has been used to linearize a low-noise amplifier (LNA) [2], but in some cases the noise figure may be seriously degraded using this technique.
Feedforward distortion cancellation was used in both [3] and [4]. While a high IP3 was obtained in [3], only simulation results awere shown. In [4], a duplication of the original amplifier is required, which doubles the power consumption, and also requires different input signal power levels for the main and auxiliary amplifier circuits. The optimal biasing technique relies on accurate biasing of the transistors such that the third-order nonlinearity coefficient is zero. While this technique has been demonstrated with good results (e.g., a 10.5 dB IP3 improvement in [5]), it is very sensitive to bias variations. The derivative superposition technique uses parallel transistors with different biases such that when the signals are summed, the distortion terms are cancelled. Two or more FETs are used in parallel with different bias points and widths in order to control distortion at the output. This technique has been used in [6]-[8] with measured IP3 improvements of 10 dB or more.
A typical derivative superposition circuit with two FETs is shown in Figure 1.
This circuit is most commonly used to control third-order intermodulation (1M3) distortion by separately biasing the two FETs such that one has a positive third-order coefficient and the other a negative third-order coefficient with equal magnitudes. This can result in a significant improvement in the IP3 of an amplifier, for instance. In [6], derivative superposition was used in a cascode amplifier design by adding an additional common source transistor in parallel with the original common-source transistor (each with appropriate gate bias and device width).
A drawback common to all of the distortion cancellation techniques described above is that they usually must be integrated within the amplification circuitry. This means that they must be designed together with the amplifier, which limits design flexibility.
Summary Described herein is a method for reducing or cancelling intermodulation distortion in a circuit, the circuit having an output including a fundamental signal and intermodulation distortion, comprising: generating, from the circuit output: (i) a primary output including the intermodulation distortion and the fundamental signal; (ii) an auxiliary output, including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled; such that intermodulation distortion of the circuit is reduced or cancelled.
The portion of the intermodulation distortion may be at least one odd-order intermodulation product. The portion of the intermodulation distortion may be a third-order intermodulation product.
In one embodiment, the fundamental signals of the primary and auxiliary outputs are summed, such that output power of the fundamental signal is increased. The contribution of the auxiliary output to the fundamental signal output power may be smaller than the contribution from the primary output.
In another embodiment, generating the primary output may comprise using a buffer, and generating the auxiliary output may comprise using a buffer. The buffers may be of substantially the same type.
The method may include a derivative superposition technique.
In another embodiment the method may comprise generating, from the circuit output: two or more auxiliary outputs, each auxiliary output including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal;
and summing the primary and auxiliary outputs such that the portions of the intermodulation distortion are reduced or cancelled. The portions of the intermodulation distortion may be odd-order intermodulation products.
Also described herein is an intermodulation distortion reducing or cancelling circuit, comprising: a primary path circuit that receives a preceding circuit output including a fundamental signal and intermodulation distortion, and generates a primary output including the fundamental signal and the intermodulation distortion; an auxiliary path circuit that receives the preceding circuit output and generates an auxiliary output including the fundamental signal and a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output; and means that sums the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled; wherein intermodulation distortion of the preceding circuit is reduced or cancelled.
Field of the Invention This invention relates generally to distortion cancellation in electronic circuits. In particular, this invention relates to distortion cancellation applied to a circuit output, to reduce or cancel intermodulation distortion products in the circuit.
Background The intermodulation distortion (IMD) of an electronic circuit such as an amplifier, mixer, or other circuit is an important consideration in many applications. The third-order intermodulation product intercept point (IP3) is commonly used to characterize IMD. This is based on the typical situation where the desired signal and an interference signal are nearby in the frequency domain. Non-linearities in the circuit generate new signals that can cause interference with the desired signal. Third-order nonlinearities, in particular, are often the most important because they generate intermodulation distortion products that are close to the desired signal and cause interference, and often cannot be removed by filtering.
Several methods have been suggested that can potentially improve the linearity of a circuit such as an amplifier. These include predistortion, feedback, feedforward, optimal biasing, and derivative superposition.
Predistortion techniques rely on the ability to distort the input signal to a circuit such that it is the inverse of the circuit's nonlinearity, and are often used in power amplifiers (e.g., [1]) and mixers. Active feedback has been used to linearize a low-noise amplifier (LNA) [2], but in some cases the noise figure may be seriously degraded using this technique.
Feedforward distortion cancellation was used in both [3] and [4]. While a high IP3 was obtained in [3], only simulation results awere shown. In [4], a duplication of the original amplifier is required, which doubles the power consumption, and also requires different input signal power levels for the main and auxiliary amplifier circuits. The optimal biasing technique relies on accurate biasing of the transistors such that the third-order nonlinearity coefficient is zero. While this technique has been demonstrated with good results (e.g., a 10.5 dB IP3 improvement in [5]), it is very sensitive to bias variations. The derivative superposition technique uses parallel transistors with different biases such that when the signals are summed, the distortion terms are cancelled. Two or more FETs are used in parallel with different bias points and widths in order to control distortion at the output. This technique has been used in [6]-[8] with measured IP3 improvements of 10 dB or more.
A typical derivative superposition circuit with two FETs is shown in Figure 1.
This circuit is most commonly used to control third-order intermodulation (1M3) distortion by separately biasing the two FETs such that one has a positive third-order coefficient and the other a negative third-order coefficient with equal magnitudes. This can result in a significant improvement in the IP3 of an amplifier, for instance. In [6], derivative superposition was used in a cascode amplifier design by adding an additional common source transistor in parallel with the original common-source transistor (each with appropriate gate bias and device width).
A drawback common to all of the distortion cancellation techniques described above is that they usually must be integrated within the amplification circuitry. This means that they must be designed together with the amplifier, which limits design flexibility.
Summary Described herein is a method for reducing or cancelling intermodulation distortion in a circuit, the circuit having an output including a fundamental signal and intermodulation distortion, comprising: generating, from the circuit output: (i) a primary output including the intermodulation distortion and the fundamental signal; (ii) an auxiliary output, including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled; such that intermodulation distortion of the circuit is reduced or cancelled.
The portion of the intermodulation distortion may be at least one odd-order intermodulation product. The portion of the intermodulation distortion may be a third-order intermodulation product.
In one embodiment, the fundamental signals of the primary and auxiliary outputs are summed, such that output power of the fundamental signal is increased. The contribution of the auxiliary output to the fundamental signal output power may be smaller than the contribution from the primary output.
In another embodiment, generating the primary output may comprise using a buffer, and generating the auxiliary output may comprise using a buffer. The buffers may be of substantially the same type.
The method may include a derivative superposition technique.
In another embodiment the method may comprise generating, from the circuit output: two or more auxiliary outputs, each auxiliary output including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal;
and summing the primary and auxiliary outputs such that the portions of the intermodulation distortion are reduced or cancelled. The portions of the intermodulation distortion may be odd-order intermodulation products.
Also described herein is an intermodulation distortion reducing or cancelling circuit, comprising: a primary path circuit that receives a preceding circuit output including a fundamental signal and intermodulation distortion, and generates a primary output including the fundamental signal and the intermodulation distortion; an auxiliary path circuit that receives the preceding circuit output and generates an auxiliary output including the fundamental signal and a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output; and means that sums the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled; wherein intermodulation distortion of the preceding circuit is reduced or cancelled.
The portion of the intermodulation distortion may beat least one odd-order intermodulation product. The portion of the intermodulation distortion may be a third-order intermodulation product.
In one embodiment, the primary path circuit may comprise a buffer and the auxiliary path circuit may comprise a buffer. The buffers may be of substantially the same type. The buffers may be source-follower buffers.
In another embodiment, the intermodulation distortion reducing or cancelling circuit may comprise two or more auxiliary path circuits that receive the preceding circuit output and generate auxiliary outputs including the fundamental signal and portions of the intermodulation distortion that are substantially equivalent in magnitude and inverted relative to the intermodulation distortion present in the primary output; and means that sums the primary and auxiliary outputs such that the portions of the intermodulation distortion are reduced or cancelled. The portions of the intermodulation distortion may be odd-order intermodulation products.
In the intermodulation distortion reducing or cancelling methods and circuits described herein, the preceding circuit may be an amplifier, a mixer, an active filter, an active attenuator, or a buffer.
Brief Description of the Drawings For a better understanding of the invention, and to show more clearly how it may be carried into effect, the invention will be described, by way of example, with reference to the accompanying drawings, wherein:
Figure 1 is schematic diagram of a derivative superposition circuit using two field effect transistors (FETs) according to the prior art.
Figure 2(a) is a schematic diagram of a conventional cascode amplifier; Figure 2(b) is a schematic diagram of a cascode amplifier with third-order IMD cancellation according to an embodiment; Figure 2(c) is a schematic diagram of a cascode amplifier with third-order IMD
cancellation according to another embodiment.
In one embodiment, the primary path circuit may comprise a buffer and the auxiliary path circuit may comprise a buffer. The buffers may be of substantially the same type. The buffers may be source-follower buffers.
In another embodiment, the intermodulation distortion reducing or cancelling circuit may comprise two or more auxiliary path circuits that receive the preceding circuit output and generate auxiliary outputs including the fundamental signal and portions of the intermodulation distortion that are substantially equivalent in magnitude and inverted relative to the intermodulation distortion present in the primary output; and means that sums the primary and auxiliary outputs such that the portions of the intermodulation distortion are reduced or cancelled. The portions of the intermodulation distortion may be odd-order intermodulation products.
In the intermodulation distortion reducing or cancelling methods and circuits described herein, the preceding circuit may be an amplifier, a mixer, an active filter, an active attenuator, or a buffer.
Brief Description of the Drawings For a better understanding of the invention, and to show more clearly how it may be carried into effect, the invention will be described, by way of example, with reference to the accompanying drawings, wherein:
Figure 1 is schematic diagram of a derivative superposition circuit using two field effect transistors (FETs) according to the prior art.
Figure 2(a) is a schematic diagram of a conventional cascode amplifier; Figure 2(b) is a schematic diagram of a cascode amplifier with third-order IMD cancellation according to an embodiment; Figure 2(c) is a schematic diagram of a cascode amplifier with third-order IMD
cancellation according to another embodiment.
Figure 3 is a block diagram showing an embodiment of a distortion cancellation circuit with multiple auxiliary paths.
Figure 4 is a plot showing typical third-order nonlinearity coefficient as a function of gate-source voltage VGS.
Figure 5 is a plot showing measured S-parameters for a conventional cascode amplifier and for the embodiment of Figure 2(b).
Figure 6 is a plot showing output power spectra for a conventional cascode amplifier and for the embodiment of Figure 2(b).
Figure 7 is a plot showing IP3 measurements for a conventional cascode amplifier and for the embodiment of Figure 2(b).
Figure 8 is a photomicrograph of a chip including a conventional cascode amplifier (top) and the embodiment of Figure 2(b) (bottom).
Detailed Description of Embodiments Described herein is a distortion cancellation technique which is applied to the output of a circuit that generates nonlinear distortion (i.e., intermodulation distortion), particularly odd-order intermodulation distortion. For example, distortion cancellation as described herein may be employed to improve third-order linearity and attenuate third-order intermodulation distortion products, without significantly increasing the noise figure of the circuit. As will be appreciated by those of ordinary skill in the art, the embodiments described herein may be applied to circuits such as, but not limited to, amplifiers, buffers, active attenuators, active filters, and mixers.
Techniques conventionally employed to reduce distortion of a circuit, such as predistortion, feedback, feedforward, optimal biasing, and derivative superposition, are generally integrated into the circuit design and do not use the circuit output. In contrast, the technique described herein implements distortion cancellation after a preceding circuit to reduce or cancel intermodulation distortion products generated by the preceding circuit. In particular, a distortion cancellation circuit as described herein receives the output of a preceding circuit and uses the output to generate signals to reduce or cancel distortion. This opens the possibility of enhancing the linearity of a pre-existing circuit by following it with a distortion cancellation circuit.
Furthermore, since the distortion cancellation circuit is connected to the output of the preceding circuit, it has no effect or only a minor effect on the gain and noise figure of the preceding circuit.
In some embodiments of the distortion cancellation circuit described herein, the design is based on a unity gain buffer or source follower. Such design is well-suited to distortion cancellation because of its inherently high linearity. In addition, unity gain, or very low gain (e.g., 0.1 to 0.3 dB), of this design render the distortion cancellation circuit substantially transparent when connected to a preceding circuit and/or a following circuit.
As such, a distortion cancellation circuit as described herein can be inserted between virtually any circuit blocks of a system. Measurement results of an embodiment of the technique described herein, connected to the output of a cascode amplifier, show a very minor effect on the gain and noise figure characteristics of the original amplifier, but a significant improvement in the IP3.
As described herein, distortion cancellation employs a derivative superposition technique.
This technique includes feeding the output of a preceding circuit into a distortion cancellation circuit including a primary path and an auxiliary path. As used herein, the term "preceding circuit" refers to the circuit for which distortion is to be reduced or cancelled. The output of the preceding circuit includes a fundamental signal (i.e., the signal of interest) and intermodulation distortion. For example, the output of the preceding circuit may include odd-order intermodulation distortion products, even-order intermodulation distortion products, or both. In applications not involving extremely wide bandwidths, odd-order intermodulation distortion may be reduced or cancelled. Of the odd-order intermodulation distortion products, the third-order products typically have the greatest magnitude and can be reduced or cancelled as described herein.
In the following paragraph, operation of the distortion cancellation circuit is described by way of an example in which the third-order intermodulation product in the output of a preceding circuit is reduced or cancelled. The primary path of the distortion cancellation circuit produces an output that is substantially the same as the output of the preceding circuit; that is, it includes the fundamental signal and the third-order intermodulation product. The auxiliary path is tuned so as to invert the third-order intermodulation product, relative to the primary path, and to leave the fundamental signal in phase with the primary path. In one embodiment, this is accomplished by biasing transistor(s) in the auxiliary path to have a negative gri3 coefficient whose absolute value is the same as the 9,,,3 of the primary path, as shown in Figure 4. As an example, the auxiliary path may produce a third-order intermodulation product current that is equal in magnitude to that of the primary path, but substantially 180 out of phase.
The resulting third-order intermodulation products of the primary path and the auxiliary path are summed at the output of the distortion cancellation circuit, substantially cancelling the third-order intermodulation products. The fundamental signals in the primary path and the auxiliary path are substantially in phase, and are added, thus potentially increasing the fundamental output power.
However, the contribution of the auxiliary path to the fundamental output power may be smaller than from the primary path.
The primary path and the auxiliary path may be tuned as to their gain on the magnitude of the fundamental signal and on third-order intermodulation products, so as to optimize performance of the distortion cancellation circuit. For example, such tuning may maximize the difference in magnitude between the fundamental signal and the third-order intermodulation products, thereby substantially reducing or cancelling intermodulation distortion (IMD).
However, it will be appreciated that such tuning of the primary path and the auxiliary path has substantially no effect on the phase of the fundamental signal.
Reduction or cancellation of further odd-order intermodulation products (e.g., fifth or seventh order) may be accomplished by adding one or more further auxiliary path circuits, each auxiliary path circuit tuned to invert a specific odd-order product. For example, Figure 3 shows an embodiment of a distortion cancellation circuit 10 connected to the output of a preceding circuit 20. The distortion cancellation circuit includes a unity gain buffer 30 as the primary path circuit and multiple auxiliary path circuits 40 - 70 each corresponding to an odd-order intermodulation distortion product. The outputs of the primary path circuit and the auxiliary path circuits are summed in a summer 80. Tuning of the primary path circuit and auxiliary path circuits is required to maximize the difference in magnitude between the fundamental signal and the odd-order intermodulation products, thereby substantially reducing or cancelling IMD.
Figure 4 is a plot showing typical third-order nonlinearity coefficient as a function of gate-source voltage VGS.
Figure 5 is a plot showing measured S-parameters for a conventional cascode amplifier and for the embodiment of Figure 2(b).
Figure 6 is a plot showing output power spectra for a conventional cascode amplifier and for the embodiment of Figure 2(b).
Figure 7 is a plot showing IP3 measurements for a conventional cascode amplifier and for the embodiment of Figure 2(b).
Figure 8 is a photomicrograph of a chip including a conventional cascode amplifier (top) and the embodiment of Figure 2(b) (bottom).
Detailed Description of Embodiments Described herein is a distortion cancellation technique which is applied to the output of a circuit that generates nonlinear distortion (i.e., intermodulation distortion), particularly odd-order intermodulation distortion. For example, distortion cancellation as described herein may be employed to improve third-order linearity and attenuate third-order intermodulation distortion products, without significantly increasing the noise figure of the circuit. As will be appreciated by those of ordinary skill in the art, the embodiments described herein may be applied to circuits such as, but not limited to, amplifiers, buffers, active attenuators, active filters, and mixers.
Techniques conventionally employed to reduce distortion of a circuit, such as predistortion, feedback, feedforward, optimal biasing, and derivative superposition, are generally integrated into the circuit design and do not use the circuit output. In contrast, the technique described herein implements distortion cancellation after a preceding circuit to reduce or cancel intermodulation distortion products generated by the preceding circuit. In particular, a distortion cancellation circuit as described herein receives the output of a preceding circuit and uses the output to generate signals to reduce or cancel distortion. This opens the possibility of enhancing the linearity of a pre-existing circuit by following it with a distortion cancellation circuit.
Furthermore, since the distortion cancellation circuit is connected to the output of the preceding circuit, it has no effect or only a minor effect on the gain and noise figure of the preceding circuit.
In some embodiments of the distortion cancellation circuit described herein, the design is based on a unity gain buffer or source follower. Such design is well-suited to distortion cancellation because of its inherently high linearity. In addition, unity gain, or very low gain (e.g., 0.1 to 0.3 dB), of this design render the distortion cancellation circuit substantially transparent when connected to a preceding circuit and/or a following circuit.
As such, a distortion cancellation circuit as described herein can be inserted between virtually any circuit blocks of a system. Measurement results of an embodiment of the technique described herein, connected to the output of a cascode amplifier, show a very minor effect on the gain and noise figure characteristics of the original amplifier, but a significant improvement in the IP3.
As described herein, distortion cancellation employs a derivative superposition technique.
This technique includes feeding the output of a preceding circuit into a distortion cancellation circuit including a primary path and an auxiliary path. As used herein, the term "preceding circuit" refers to the circuit for which distortion is to be reduced or cancelled. The output of the preceding circuit includes a fundamental signal (i.e., the signal of interest) and intermodulation distortion. For example, the output of the preceding circuit may include odd-order intermodulation distortion products, even-order intermodulation distortion products, or both. In applications not involving extremely wide bandwidths, odd-order intermodulation distortion may be reduced or cancelled. Of the odd-order intermodulation distortion products, the third-order products typically have the greatest magnitude and can be reduced or cancelled as described herein.
In the following paragraph, operation of the distortion cancellation circuit is described by way of an example in which the third-order intermodulation product in the output of a preceding circuit is reduced or cancelled. The primary path of the distortion cancellation circuit produces an output that is substantially the same as the output of the preceding circuit; that is, it includes the fundamental signal and the third-order intermodulation product. The auxiliary path is tuned so as to invert the third-order intermodulation product, relative to the primary path, and to leave the fundamental signal in phase with the primary path. In one embodiment, this is accomplished by biasing transistor(s) in the auxiliary path to have a negative gri3 coefficient whose absolute value is the same as the 9,,,3 of the primary path, as shown in Figure 4. As an example, the auxiliary path may produce a third-order intermodulation product current that is equal in magnitude to that of the primary path, but substantially 180 out of phase.
The resulting third-order intermodulation products of the primary path and the auxiliary path are summed at the output of the distortion cancellation circuit, substantially cancelling the third-order intermodulation products. The fundamental signals in the primary path and the auxiliary path are substantially in phase, and are added, thus potentially increasing the fundamental output power.
However, the contribution of the auxiliary path to the fundamental output power may be smaller than from the primary path.
The primary path and the auxiliary path may be tuned as to their gain on the magnitude of the fundamental signal and on third-order intermodulation products, so as to optimize performance of the distortion cancellation circuit. For example, such tuning may maximize the difference in magnitude between the fundamental signal and the third-order intermodulation products, thereby substantially reducing or cancelling intermodulation distortion (IMD).
However, it will be appreciated that such tuning of the primary path and the auxiliary path has substantially no effect on the phase of the fundamental signal.
Reduction or cancellation of further odd-order intermodulation products (e.g., fifth or seventh order) may be accomplished by adding one or more further auxiliary path circuits, each auxiliary path circuit tuned to invert a specific odd-order product. For example, Figure 3 shows an embodiment of a distortion cancellation circuit 10 connected to the output of a preceding circuit 20. The distortion cancellation circuit includes a unity gain buffer 30 as the primary path circuit and multiple auxiliary path circuits 40 - 70 each corresponding to an odd-order intermodulation distortion product. The outputs of the primary path circuit and the auxiliary path circuits are summed in a summer 80. Tuning of the primary path circuit and auxiliary path circuits is required to maximize the difference in magnitude between the fundamental signal and the odd-order intermodulation products, thereby substantially reducing or cancelling IMD.
As noted above, distortion cancellation as described herein may be applied to a variety of circuits, such as amplifiers, buffers, active attenuators, active filters, and mixers. In the case of an amplifier, it may be of a design such as, but not limited to, cascode, common-source, common-gate, class A, class AB, and travelling wave. It will be appreciated that, because distortion cancellation is applied to the output of, e.g., an amplifier, it may easily be adapted to an existing amplifier design or other circuit designs, avoiding the need for a complete reworking of the design. In addition, because distortion cancellation as described herein is applied to the output of an amplifier, it may be implemented as a retrofit to existing hardware.
Distortion cancellation as described herein may be implemented in any suitable technology, such as, CMOS, Bipolar, BiCMOS, SiGe, MESFET, HEMT, PHEMT, etc.
using, for example, an integrated circuit based on substrates such as GaAs, Si, SiGe, GaN, GaN, SiC, InP, etc. Distortion cancellation as described herein is suitable for broadband applications because the technique used to invert the nonlinearity is not frequency-dependent. Thus, substantially the same bias settings work from approximately DC to the maximum frequency of the circuit(s) to which it may be applied. Accordingly, the technique may be applied to the signal path of circuits operating at frequencies in the audible range (e.g., audio systems) and higher, for example, at radio frequencies (RF) and millimeter wave and microwave frequencies, as in, for example, radio and telecommunications systems. In contrast, a conventional approach such as a phase shift technique (e.g., using a delay line) is narrowband and operates only within a narrow frequency range. Furthermore, a phase shift technique inevitably affects the fundamental signal, whereas the methods and circuits described herein may be configured so as to have substantially no effect on the fundamental signal.
In one embodiment, a distortion cancellation circuit as described herein is implemented without or substantially without input and output matching networks. Such an embodiment is suitable for combining with an existing preceding circuit and/or an existing following circuit. In another embodiment, a distortion cancellation circuit as described herein is implemented in combination with a following circuit, and includes an input matching network.
In another embodiment, a distortion cancellation circuit as described herein is implemented in combination with a preceding circuit, and includes an output matching network. In a further embodiment, a distortion cancellation circuit as described herein is implemented in combination with preceding and following circuits, and includes an input matching network and an output matching network.
This last embodiment is a stand-alone distortion cancellation (i.e., linearization) block. A
distortion cancellation circuit as described herein may be provided as a module, with or without input and/or output matching networks, to be inserted between circuit blocks of existing systems.
Matching networks may induce some signal loss, and this may be compensated by tuning of the distortion cancellation circuit.
The primary path and the auxiliary path may each be an amplifier, mixer, or buffer circuit, such as, for example, a source-follower buffer. In one embodiment, the primary path and auxiliary path circuits may be connected in parallel. In some embodiments, the primary and auxiliary path circuits may be the same. Differences in performance of the circuits in the primary and auxiliary paths may be achieved by appropriately adjusting independently the biasing of those circuits.
Exemplary embodiments will be described below as applied to amplifiers.
However, as noted above, distortion cancellation as described herein may be applied to other circuits as well.
A conventional, general-purpose cascode amplifier is shown in Figure 2(a).
This amplifier uses inductors Lg and LS for matching and the parallel tank circuit, C1 and L1, in addition to any parasitic capacitance, is designed to resonate at a desired frequency.
Figure 2(b) shows an embodiment of a distortion cancellation circuit, which provides third-order intermodulation distortion cancellation, applied to the cascode amplifier of Figure 2(a). Operation of this circuit will be described below.
In general, the output drain current of a FET amplifier can be described by the power series:
id = 9rn1 'gs + 9m2V9S + gm v , +- ...
where CHID
gnaw &117 n G, S
are the coefficients of the fundamental (n = 1) and higher order (n > 1) terms that lead to nonlinearities. The IP3 performance is predominantly determined by the third-order nonlinearity coefficient, gm3. The input IP3 (IIP3), which is the point at which the power in the third-order intermodulation products is extrapolated to intersect the input power, for an amplifier with a 50 ohm input impedance is given by [3]:
IIP3 = 10 lob 40 dBm (3) 3 g:
It is well known in the art that the gm3 coefficient can be either positive or negative, depending on the biasing conditions (e.g., see [5]). In fact, gm3 changes from positive to negative as the FET transitions from the weak to strong inversion regions, as shown in Figure 4. The ability to control the sign of the third order coefficient is the property used in this work to design an amplifier with third-order IMD cancellation.
In the embodiment of Figure 2(b) the distortion cancellation circuit includes two parallel source-follower buffers. The load resistor, RL, may be designed to be somewhat greater than 50 ohms so that the resulting increased voltage amplitude is equal to the loss through the buffer consisting of transistors M3 and M4. The result is a fundamental output that is substantially the same as the output of the amplifier in Figure 2(a). The auxiliary path through the buffer M5 and M6 is designed with appropriate biasing and device sizes such that the third-order nonlinearity current that is generated is equal in magnitude to the primary path, but 180 out of phase. The phase inversion is achieved by biasing M3 and M5 independently in the negative and positive gm3 regions, respectively. For the primary path, through M3 and M4, the required VGS will typically be in the strong inversion region, and thus will have a negative gm3. The auxiliary path through M5 and M6 is therefore biased through Vg2 and Vg3 to be in the weak inversion region where there is a positive gm3. The resulting third-order nonlinearity currents are summed at the output, resulting in distortion cancellation. Since the fundamental signal maintains the same sign for both buffer circuits, it is added in-phase, thus increasing the fundamental output power.
However, since a positive gm3 occurs in the weak or moderate inversion regions, the contribution of the auxiliary path to the fundamental output power is much smaller than from the primary path (since gm, is small in the weak and moderate inversion regions).
-H_ Consider two input signals to the amplifier shown in Figure 2(b) with different frequencies, fl and f2. At the output of the conventional amplifier block are the fundamental signals, fl and f2, as well as the IM3 signals generated in the cascode amplifier (2fl + f2 and 212 fl). The IM3 products at the output of M3-M4 are the sum of the IM3 signals generated by the conventional amplifier and those generated in M3-M4 from the amplified fl and 12 signals. The M3-M4 buffer is biased such that the gm3 coefficient is negative. The auxiliary path M5-M6 buffer is designed to be in the positive gm3 region, and it therefore produces IM3 products that are opposite in phase relative to those generated by the M3-M4 buffer. By designing each source-follower buffer in the distortion cancellation circuit to generate IM3 currents that are equal in magnitude, but 180 out of phase, the IP3 performance of the amplifier may be significantly improved by summing these currents at the output.
In simulations, existing BSIM3 FET models were used in order to choose bias points that resulted in optimal distortion cancellation, and the simulations produced accurate predictions. If accurate device models are not available, this technique can still be used, but IM3 measurements for the chosen technology may be necessary to achieve optimal performance.
In the embodiment of Figure 2(b) there are two transistors shown in parallel which generate intermodulation components that cancel distortion from the preceding circuit. It is also possible to have more than two transistors in parallel. Having more than two transistors in parallel provides more flexibility and accuracy in the distortion cancellation signals that are generated, and reduces sensitivity to bias voltages. In particular, by using more than two transistors in parallel, the overall response with respect to bias voltage (see Figure 4) can be "flattened", making the circuit performance less susceptible to degradation from voltage fluctuations.
Figure 2(c) shows an embodiment similar to that of Figure 2(b), implemented with current sources.
Embodiments are further described by way of the following non-limiting example.
Example A chip was fabricated using a standard CMOS process with a silicon substrate, based on the circuits of Figures 2(a) and 2(b). A photomicrograph of the fabricated chip is shown in Figure 8. The dimensions of the chip were 1 mm x 1 mm with each amplifier using half of this space (0.5 mm2). The additional chip area required by the distortion cancellation circuitry was approximately 300 pm x 200 pm (0.06 mm2).
The supply voltage for both amplifiers was set to 1.8 V and the bias voltages on the third-order IMD cancellation amplifier were Vgl = 1.0 V, Vg2 = 0.52 V, and Vg3 = 1.5 V. To characterize the performance of both of the amplifiers, a full two-port calibration was performed and the S-parameters were measured using a vector network analyzer. The results are shown in Figure 5. It is clear that both amplifiers had very similar S 11 and S21 characteristics. Note that the amplifier with third-order IMD cancellation achieved a slightly higher gain and a slightly larger bandwidth. For both amplifiers the gain and return loss at 2.4 GHz were approximately 16 dB and -10 dB, respectively.
To determine the linearity of each amplifier, a two-tone input signal (2.400 GHz and 2.401 GHz) was used and the output power of the third-order distortion was measured. Shown in Figure 6 is the output spectrum for both amplifiers superimposed so that the difference in third-order intermodulation distortion is clear.
The input power used for each tone was -30 dBm. It was not possible to distinguish between the two amplifiers with regard to the fundamental outputs. For both amplifiers, the two fundamental outputs showed a power of approximately -14 dBm as expected since the measured S21 was approximately 16 dB. The third-order intermodulation distortion products occurred at 2.399 GHz and 2.402 GHz (2fl - f2 and 2f2 - fl) for this two-tone test. In the case of the conventional amplifier, the output power of the third-order intermodulation products was approximately -67 dBm (dashed line). In the case of the amplifier with distortion cancellation, the third-order intermodulation products were attenuated by approximately 21 dB (solid line).
The plots of Figures 4 and 5 clearly illustrate that the distortion cancellation method provides much improved third-order linearity performance. The IP2 performance of the amplifier was degraded somewhat with this technique; however, since the second-order distortion signals were far out of band (1 MHz and 4.801 GHz in this example), this would not be a major concern in most cases.
The input power of the two input tones was swept for each amplifier while the fundamental and third-order distortion output powers were measured. The results are shown in Figure 7. The IIP3 of the conventional amplifier was approximately -2 dBm, whereas the IIP3 of the amplifier with distortion cancellation was approximately 10 dBm.
Therefore, a +12 dB
improvement in IIP3 was measured using the embodiment of Figure 2(b). With the distortion cancellation amplifier, the 1-dB compression point was reduced by approximately 3 dB;
however, this may not be a significant concern, for example in cases where there is a low input power level (e.g., in a low-noise amplifier).
The noise figure of each amplifier was measured to determine the effect of the distortion cancellation circuit on the noise performance. At 2.4 GHz, the noise figure of the conventional amplifier was 4.02 dB, while the noise figure of the third-order IMD
cancellation amplifier was 4.18 dB, an increase of just 0.16 dB. Therefore, there was not a significant degradation in the noise figure with the circuit of Figure 2(b). The DC current used by the conventional amplifier was 21 mA, and for the distortion cancellation amplifier it was 31 mA. The added current in the proposed amplifier was due almost entirely to the M3-M4 source-follower circuit shown in Figure 2(b).
A commonly used figure of merit (FOM) [10] for low-noise amplifiers is defined as:
(F - 1) PDc where OIP3 is the output-referred IP3 point, F is the noise factor, and PDC is the DC power consumption. The embodiment of Figure 2(b) achieved a FOM of 308, which compares favourably to most LNAs in the literature.
Equivalents Those skilled in the art will recognize, or will be able to ascertain, equivalents to the embodiments described herein. Such equivalents are within the scope of the invention and are covered by the appended claims.
References [1] G. Hau, T. B. Nishimura, and N. Iwata, "A highly efficient linearized wide-band CDMA
handset power amplifier based on predistortion under various bias conditions", IEEE
Transactions on Microwave Theory and Techniques, Vol. 49, pp. 1194-1201, June 2001.
[2] K. W. Kobayashi, D. C. Streit, A. K. Oki, D. K. Umemoto, and T. R. Block, "A novel monolithic linearized HEMT LNA using HBT tunable active feedback", IEEE MTT-S
International Microwave Symposium Digest, Vol. 3, pp. 1217-1220, June 1996.
[3] M. Lin, Y. Li, and H. Chen, "A novel IP3 boosting technique using feedforward distortion cancellation method for 5 GHz CMOS LNA", IEEE Radio Freq. Integrated Circuits Symp., pp.
699-702, June 2003.
[4] Y. Ding and R. Harjani, "A +18 dBm IIP3 LNA in 0.35 m CMOS", IEEE
International Solid-State Circuits Conference, pp. 162-164, 2001.
[5] V. Aparin, G. Brown, and L. E. Larson, "Linearization of CMOS LNA's via optimum gate biasing", IEEE International Symposium on Circuits and Systems, Vol. 4, pp.
748-751, May 2004.
[6] V. Aparin and L. E. Larson, "Modified derivative superposition method for linearizing FET
low-noise amplifiers", IEEE Trans. on Microwave Theory and Techniques, Vol.
53, No. 2, pp.
571-581, February 2005.
[7] D. Webster, J. Scott, and D. Haigh, "Control of circuit distortion by derivative superposition method", IEEE Microwave and Guided Wave Letters, Vol. 6, No. 3, pp. 123-125, March 1996.
[8] D. R. Webster, et. al., "Low-distortion MMIC power amplifier using a new form of derivative superposition", IEEE Trans. on Microwave Theory and Techniques, Vol. 49, No. 2, pp. 328-332, Feb. 2001.
[9] T. T. Ha, Solid-State Microwave Amplifier Design, New York: Wiley, ch. 6, 1981.
[10] D. C. Ahlgren, N. King, G. Freeman, R. Groves, and S. Subbanna, "SiGe BiCMOS
technology for RF devices and design applications", Proc. IEEE Radio Wireless Conf., pp. 281-284, 1999.
Distortion cancellation as described herein may be implemented in any suitable technology, such as, CMOS, Bipolar, BiCMOS, SiGe, MESFET, HEMT, PHEMT, etc.
using, for example, an integrated circuit based on substrates such as GaAs, Si, SiGe, GaN, GaN, SiC, InP, etc. Distortion cancellation as described herein is suitable for broadband applications because the technique used to invert the nonlinearity is not frequency-dependent. Thus, substantially the same bias settings work from approximately DC to the maximum frequency of the circuit(s) to which it may be applied. Accordingly, the technique may be applied to the signal path of circuits operating at frequencies in the audible range (e.g., audio systems) and higher, for example, at radio frequencies (RF) and millimeter wave and microwave frequencies, as in, for example, radio and telecommunications systems. In contrast, a conventional approach such as a phase shift technique (e.g., using a delay line) is narrowband and operates only within a narrow frequency range. Furthermore, a phase shift technique inevitably affects the fundamental signal, whereas the methods and circuits described herein may be configured so as to have substantially no effect on the fundamental signal.
In one embodiment, a distortion cancellation circuit as described herein is implemented without or substantially without input and output matching networks. Such an embodiment is suitable for combining with an existing preceding circuit and/or an existing following circuit. In another embodiment, a distortion cancellation circuit as described herein is implemented in combination with a following circuit, and includes an input matching network.
In another embodiment, a distortion cancellation circuit as described herein is implemented in combination with a preceding circuit, and includes an output matching network. In a further embodiment, a distortion cancellation circuit as described herein is implemented in combination with preceding and following circuits, and includes an input matching network and an output matching network.
This last embodiment is a stand-alone distortion cancellation (i.e., linearization) block. A
distortion cancellation circuit as described herein may be provided as a module, with or without input and/or output matching networks, to be inserted between circuit blocks of existing systems.
Matching networks may induce some signal loss, and this may be compensated by tuning of the distortion cancellation circuit.
The primary path and the auxiliary path may each be an amplifier, mixer, or buffer circuit, such as, for example, a source-follower buffer. In one embodiment, the primary path and auxiliary path circuits may be connected in parallel. In some embodiments, the primary and auxiliary path circuits may be the same. Differences in performance of the circuits in the primary and auxiliary paths may be achieved by appropriately adjusting independently the biasing of those circuits.
Exemplary embodiments will be described below as applied to amplifiers.
However, as noted above, distortion cancellation as described herein may be applied to other circuits as well.
A conventional, general-purpose cascode amplifier is shown in Figure 2(a).
This amplifier uses inductors Lg and LS for matching and the parallel tank circuit, C1 and L1, in addition to any parasitic capacitance, is designed to resonate at a desired frequency.
Figure 2(b) shows an embodiment of a distortion cancellation circuit, which provides third-order intermodulation distortion cancellation, applied to the cascode amplifier of Figure 2(a). Operation of this circuit will be described below.
In general, the output drain current of a FET amplifier can be described by the power series:
id = 9rn1 'gs + 9m2V9S + gm v , +- ...
where CHID
gnaw &117 n G, S
are the coefficients of the fundamental (n = 1) and higher order (n > 1) terms that lead to nonlinearities. The IP3 performance is predominantly determined by the third-order nonlinearity coefficient, gm3. The input IP3 (IIP3), which is the point at which the power in the third-order intermodulation products is extrapolated to intersect the input power, for an amplifier with a 50 ohm input impedance is given by [3]:
IIP3 = 10 lob 40 dBm (3) 3 g:
It is well known in the art that the gm3 coefficient can be either positive or negative, depending on the biasing conditions (e.g., see [5]). In fact, gm3 changes from positive to negative as the FET transitions from the weak to strong inversion regions, as shown in Figure 4. The ability to control the sign of the third order coefficient is the property used in this work to design an amplifier with third-order IMD cancellation.
In the embodiment of Figure 2(b) the distortion cancellation circuit includes two parallel source-follower buffers. The load resistor, RL, may be designed to be somewhat greater than 50 ohms so that the resulting increased voltage amplitude is equal to the loss through the buffer consisting of transistors M3 and M4. The result is a fundamental output that is substantially the same as the output of the amplifier in Figure 2(a). The auxiliary path through the buffer M5 and M6 is designed with appropriate biasing and device sizes such that the third-order nonlinearity current that is generated is equal in magnitude to the primary path, but 180 out of phase. The phase inversion is achieved by biasing M3 and M5 independently in the negative and positive gm3 regions, respectively. For the primary path, through M3 and M4, the required VGS will typically be in the strong inversion region, and thus will have a negative gm3. The auxiliary path through M5 and M6 is therefore biased through Vg2 and Vg3 to be in the weak inversion region where there is a positive gm3. The resulting third-order nonlinearity currents are summed at the output, resulting in distortion cancellation. Since the fundamental signal maintains the same sign for both buffer circuits, it is added in-phase, thus increasing the fundamental output power.
However, since a positive gm3 occurs in the weak or moderate inversion regions, the contribution of the auxiliary path to the fundamental output power is much smaller than from the primary path (since gm, is small in the weak and moderate inversion regions).
-H_ Consider two input signals to the amplifier shown in Figure 2(b) with different frequencies, fl and f2. At the output of the conventional amplifier block are the fundamental signals, fl and f2, as well as the IM3 signals generated in the cascode amplifier (2fl + f2 and 212 fl). The IM3 products at the output of M3-M4 are the sum of the IM3 signals generated by the conventional amplifier and those generated in M3-M4 from the amplified fl and 12 signals. The M3-M4 buffer is biased such that the gm3 coefficient is negative. The auxiliary path M5-M6 buffer is designed to be in the positive gm3 region, and it therefore produces IM3 products that are opposite in phase relative to those generated by the M3-M4 buffer. By designing each source-follower buffer in the distortion cancellation circuit to generate IM3 currents that are equal in magnitude, but 180 out of phase, the IP3 performance of the amplifier may be significantly improved by summing these currents at the output.
In simulations, existing BSIM3 FET models were used in order to choose bias points that resulted in optimal distortion cancellation, and the simulations produced accurate predictions. If accurate device models are not available, this technique can still be used, but IM3 measurements for the chosen technology may be necessary to achieve optimal performance.
In the embodiment of Figure 2(b) there are two transistors shown in parallel which generate intermodulation components that cancel distortion from the preceding circuit. It is also possible to have more than two transistors in parallel. Having more than two transistors in parallel provides more flexibility and accuracy in the distortion cancellation signals that are generated, and reduces sensitivity to bias voltages. In particular, by using more than two transistors in parallel, the overall response with respect to bias voltage (see Figure 4) can be "flattened", making the circuit performance less susceptible to degradation from voltage fluctuations.
Figure 2(c) shows an embodiment similar to that of Figure 2(b), implemented with current sources.
Embodiments are further described by way of the following non-limiting example.
Example A chip was fabricated using a standard CMOS process with a silicon substrate, based on the circuits of Figures 2(a) and 2(b). A photomicrograph of the fabricated chip is shown in Figure 8. The dimensions of the chip were 1 mm x 1 mm with each amplifier using half of this space (0.5 mm2). The additional chip area required by the distortion cancellation circuitry was approximately 300 pm x 200 pm (0.06 mm2).
The supply voltage for both amplifiers was set to 1.8 V and the bias voltages on the third-order IMD cancellation amplifier were Vgl = 1.0 V, Vg2 = 0.52 V, and Vg3 = 1.5 V. To characterize the performance of both of the amplifiers, a full two-port calibration was performed and the S-parameters were measured using a vector network analyzer. The results are shown in Figure 5. It is clear that both amplifiers had very similar S 11 and S21 characteristics. Note that the amplifier with third-order IMD cancellation achieved a slightly higher gain and a slightly larger bandwidth. For both amplifiers the gain and return loss at 2.4 GHz were approximately 16 dB and -10 dB, respectively.
To determine the linearity of each amplifier, a two-tone input signal (2.400 GHz and 2.401 GHz) was used and the output power of the third-order distortion was measured. Shown in Figure 6 is the output spectrum for both amplifiers superimposed so that the difference in third-order intermodulation distortion is clear.
The input power used for each tone was -30 dBm. It was not possible to distinguish between the two amplifiers with regard to the fundamental outputs. For both amplifiers, the two fundamental outputs showed a power of approximately -14 dBm as expected since the measured S21 was approximately 16 dB. The third-order intermodulation distortion products occurred at 2.399 GHz and 2.402 GHz (2fl - f2 and 2f2 - fl) for this two-tone test. In the case of the conventional amplifier, the output power of the third-order intermodulation products was approximately -67 dBm (dashed line). In the case of the amplifier with distortion cancellation, the third-order intermodulation products were attenuated by approximately 21 dB (solid line).
The plots of Figures 4 and 5 clearly illustrate that the distortion cancellation method provides much improved third-order linearity performance. The IP2 performance of the amplifier was degraded somewhat with this technique; however, since the second-order distortion signals were far out of band (1 MHz and 4.801 GHz in this example), this would not be a major concern in most cases.
The input power of the two input tones was swept for each amplifier while the fundamental and third-order distortion output powers were measured. The results are shown in Figure 7. The IIP3 of the conventional amplifier was approximately -2 dBm, whereas the IIP3 of the amplifier with distortion cancellation was approximately 10 dBm.
Therefore, a +12 dB
improvement in IIP3 was measured using the embodiment of Figure 2(b). With the distortion cancellation amplifier, the 1-dB compression point was reduced by approximately 3 dB;
however, this may not be a significant concern, for example in cases where there is a low input power level (e.g., in a low-noise amplifier).
The noise figure of each amplifier was measured to determine the effect of the distortion cancellation circuit on the noise performance. At 2.4 GHz, the noise figure of the conventional amplifier was 4.02 dB, while the noise figure of the third-order IMD
cancellation amplifier was 4.18 dB, an increase of just 0.16 dB. Therefore, there was not a significant degradation in the noise figure with the circuit of Figure 2(b). The DC current used by the conventional amplifier was 21 mA, and for the distortion cancellation amplifier it was 31 mA. The added current in the proposed amplifier was due almost entirely to the M3-M4 source-follower circuit shown in Figure 2(b).
A commonly used figure of merit (FOM) [10] for low-noise amplifiers is defined as:
(F - 1) PDc where OIP3 is the output-referred IP3 point, F is the noise factor, and PDC is the DC power consumption. The embodiment of Figure 2(b) achieved a FOM of 308, which compares favourably to most LNAs in the literature.
Equivalents Those skilled in the art will recognize, or will be able to ascertain, equivalents to the embodiments described herein. Such equivalents are within the scope of the invention and are covered by the appended claims.
References [1] G. Hau, T. B. Nishimura, and N. Iwata, "A highly efficient linearized wide-band CDMA
handset power amplifier based on predistortion under various bias conditions", IEEE
Transactions on Microwave Theory and Techniques, Vol. 49, pp. 1194-1201, June 2001.
[2] K. W. Kobayashi, D. C. Streit, A. K. Oki, D. K. Umemoto, and T. R. Block, "A novel monolithic linearized HEMT LNA using HBT tunable active feedback", IEEE MTT-S
International Microwave Symposium Digest, Vol. 3, pp. 1217-1220, June 1996.
[3] M. Lin, Y. Li, and H. Chen, "A novel IP3 boosting technique using feedforward distortion cancellation method for 5 GHz CMOS LNA", IEEE Radio Freq. Integrated Circuits Symp., pp.
699-702, June 2003.
[4] Y. Ding and R. Harjani, "A +18 dBm IIP3 LNA in 0.35 m CMOS", IEEE
International Solid-State Circuits Conference, pp. 162-164, 2001.
[5] V. Aparin, G. Brown, and L. E. Larson, "Linearization of CMOS LNA's via optimum gate biasing", IEEE International Symposium on Circuits and Systems, Vol. 4, pp.
748-751, May 2004.
[6] V. Aparin and L. E. Larson, "Modified derivative superposition method for linearizing FET
low-noise amplifiers", IEEE Trans. on Microwave Theory and Techniques, Vol.
53, No. 2, pp.
571-581, February 2005.
[7] D. Webster, J. Scott, and D. Haigh, "Control of circuit distortion by derivative superposition method", IEEE Microwave and Guided Wave Letters, Vol. 6, No. 3, pp. 123-125, March 1996.
[8] D. R. Webster, et. al., "Low-distortion MMIC power amplifier using a new form of derivative superposition", IEEE Trans. on Microwave Theory and Techniques, Vol. 49, No. 2, pp. 328-332, Feb. 2001.
[9] T. T. Ha, Solid-State Microwave Amplifier Design, New York: Wiley, ch. 6, 1981.
[10] D. C. Ahlgren, N. King, G. Freeman, R. Groves, and S. Subbanna, "SiGe BiCMOS
technology for RF devices and design applications", Proc. IEEE Radio Wireless Conf., pp. 281-284, 1999.
Claims (20)
1. A method for reducing or cancelling intermodulation distortion in a circuit, the circuit having an output including a fundamental signal and intermodulation distortion, comprising:
generating, from the circuit output:
(i) a primary output including the intermodulation distortion and the fundamental signal;
(ii) an auxiliary output, including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled;
such that intermodulation distortion of the circuit is reduced or cancelled.
generating, from the circuit output:
(i) a primary output including the intermodulation distortion and the fundamental signal;
(ii) an auxiliary output, including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled;
such that intermodulation distortion of the circuit is reduced or cancelled.
2. The method of claim 1, wherein the portion of the intermodulation distortion is at least one odd-order intermodulation product.
3. The method of claim 1, wherein the portion of the intermodulation distortion is a third-order intermodulation product.
4. The method of claim 1, wherein the fundamental signals of the primary and auxiliary outputs are summed, such that output power of the fundamental signal is increased.
5. The method of claim 1, wherein the contribution of the auxiliary output to the fundamental signal output power is smaller than the contribution from the primary output.
6. The method of claim 1, wherein generating the primary output comprises using a buffer, and generating the auxiliary output comprises using a buffer.
7. The method of claim 6, wherein the buffers are of substantially the same type.
8. The method of claim 1, wherein the method includes a derivative superposition technique.
9. The method of claim 1, comprising generating, from the circuit output: two or more auxiliary outputs, each auxiliary output including a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output, and the fundamental signal; and summing the primary and auxiliary outputs such that the portions of the intermodulation distortion are reduced or cancelled.
10. The method of claim 9, wherein the portions of the intermodulation distortion are odd-order intermodulation products.
11. The method of claim 1, wherein the circuit is selected from an amplifier, a mixer, an active filter, an active attenuator, and a buffer.
12. An intermodulation distortion reducing or cancelling circuit, comprising:
a primary path circuit that receives a preceding circuit output including a fundamental signal and intermodulation distortion, and generates a primary output including the fundamental signal and the intermodulation distortion;
an auxiliary path circuit that receives the preceding circuit output and generates an auxiliary output including the fundamental signal and a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output; and means that sums the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled;
wherein intermodulation distortion of the preceding circuit is reduced or cancelled.
a primary path circuit that receives a preceding circuit output including a fundamental signal and intermodulation distortion, and generates a primary output including the fundamental signal and the intermodulation distortion;
an auxiliary path circuit that receives the preceding circuit output and generates an auxiliary output including the fundamental signal and a portion of the intermodulation distortion that is substantially equivalent in magnitude and inverted relative to the intermodulation distortion in the primary output; and means that sums the primary and auxiliary outputs such that the portion of the intermodulation distortion is reduced or cancelled;
wherein intermodulation distortion of the preceding circuit is reduced or cancelled.
13. The intermodulation distortion reducing or cancelling circuit of claim 12, wherein the portion of the intermodulation distortion is at least one odd-order intermodulation product.
14. The intermodulation distortion reducing or cancelling circuit of claim 12, wherein the portion of the intermodulation distortion is a third-order intermodulation product.
15. The intermodulation distortion reducing or cancelling circuit of claim 12, wherein the primary path circuit comprises a buffer and the auxiliary path circuit comprises a buffer.
16. The intermodulation distortion reducing or cancelling circuit of claim 15, wherein the buffers are of substantially the same type.
17. The intermodulation distortion reducing or cancelling circuit of claim 15, wherein the buffers are source-follower buffers.
18. The intermodulation distortion reducing or cancelling circuit of claim 12, comprising two or more auxiliary path circuits that receive the preceding circuit output and generate auxiliary outputs including the fundamental signal and portions of the intermodulation distortion that are substantially equivalent in magnitude and inverted relative to the intermodulation distortion present in the primary output; and means that sums the primary and auxiliary outputs such that the portions of the intermodulation distortion are reduced or cancelled.
19. The intermodulation distortion reducing or cancelling circuit of claim 18, wherein the portions of the intermodulation distortion are odd-order intermodulation products.
20. The intermodulation distortion reducing or cancelling circuit of claim 12, wherein the preceding circuit is selected from an amplifier, a mixer, an active filter, an active attenuator, and a buffer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA2690442A CA2690442A1 (en) | 2010-01-19 | 2010-01-19 | Distortion cancellation method and circuit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA2690442A CA2690442A1 (en) | 2010-01-19 | 2010-01-19 | Distortion cancellation method and circuit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2690442A1 true CA2690442A1 (en) | 2011-07-19 |
Family
ID=44303617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA2690442A Abandoned CA2690442A1 (en) | 2010-01-19 | 2010-01-19 | Distortion cancellation method and circuit |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA2690442A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014037027A1 (en) * | 2012-09-04 | 2014-03-13 | Telefonaktiebolaget L M Ericsson (Publ) | A sub-harmonic mixer |
| TWI617131B (en) * | 2016-12-30 | 2018-03-01 | 立積電子股份有限公司 | Amplifier circuit |
-
2010
- 2010-01-19 CA CA2690442A patent/CA2690442A1/en not_active Abandoned
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014037027A1 (en) * | 2012-09-04 | 2014-03-13 | Telefonaktiebolaget L M Ericsson (Publ) | A sub-harmonic mixer |
| US9478204B2 (en) | 2012-09-04 | 2016-10-25 | Telefonaktiebolaget L M Ericsson (Publ) | Sub-harmonic mixer |
| TWI617131B (en) * | 2016-12-30 | 2018-03-01 | 立積電子股份有限公司 | Amplifier circuit |
| CN108270403A (en) * | 2016-12-30 | 2018-07-10 | 立积电子股份有限公司 | Amplifying circuit |
| US10291189B2 (en) | 2016-12-30 | 2019-05-14 | Richwave Technology Corp. | Amplification circuit |
| CN108270403B (en) * | 2016-12-30 | 2023-05-02 | 立积电子股份有限公司 | Amplifying circuit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7944298B2 (en) | Low noise and low input capacitance differential MDS LNA | |
| KR101237565B1 (en) | Amplifier with improved linearization | |
| KR100686785B1 (en) | Impedance matching networks for non-linear circuits | |
| KR101092264B1 (en) | Amplifier with active post-distortion linearization | |
| US6943628B2 (en) | Methods and apparatus for substantially reducing nonlinear distortion | |
| JP5612034B2 (en) | Differential amplifier with active post-distortion linearization | |
| Aparin et al. | Linearization of monolithic LNAs using low-frequency low-impedance input termination | |
| Kim et al. | Highly linear CMOS RF MMIC amplifier using multiple gated transistors and its Volterra series analysis | |
| JP2004533135A (en) | Circuits for linearizing electronic devices | |
| KR20070033352A (en) | Tunable Circuitry for Neutralization of Third-order Modulation | |
| KR19990053981A (en) | Gain Control Circuit of Low Noise Amplifier | |
| CA2690442A1 (en) | Distortion cancellation method and circuit | |
| Brinkhoff et al. | Baseband impedance and linearization of FET circuits | |
| Parvizi et al. | Improved derivative superposition scheme for simultaneous second-and third-order distortion cancellation in LNAs | |
| KR20000037693A (en) | Integrated radio frequency circuit for improving linear characteristics of small signal using balun of active element | |
| US7940126B2 (en) | Signal amplification apparatus with advanced linearization | |
| Najari et al. | Wideband inductorless LNA employing simultaneous 2 nd and 3 rd order distortion cancellation | |
| Jackson et al. | A CMOS amplifier with third-order intermodulation distortion cancellation | |
| Kim et al. | A low-power highly linear cascoded multiple-gated transistor CMOS RF amplifier with 10 dB IP3 improvement (Revised) | |
| CN114759886B (en) | Radio frequency amplifying circuit | |
| JPH07240638A (en) | High frequency amplifier | |
| KR102449479B1 (en) | Power Amplifier | |
| Zhang et al. | Linearity improvement using a modified derivative superposition method in cascaded stages amplifier | |
| Watanabe et al. | IP3 boost circuitry for highly linear CDMA low noise amplifiers (LNA) | |
| Umoh et al. | Digital post-linearization of a wideband low noise amplifier for ultra-wideband wireless receivers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FZDE | Dead |
Effective date: 20130121 |