Frequency conversion circuits

Abstract

A frequency conversion circuit having first and second input signal terminals, and first and second output terminals includes a plurality of non-linear elements, each one having a pair of input electrodes and an output electrode. The first input electrode of each non-linear element is successively interconnected by a two port input coupling means, said input coupling means having a first end coupled to the first input signal terminal and a second end terminated in a terminating impedance. The output electrode of each non-linear device is successively interconnected by a two port output coupling means coupled between the pair of output terminals. An input signal is fed to the first input signal terminal and is coupled to the input coupling means and a portion of such signal is fed to the first input electrode of each non-linear device. Each portion of the input signal is provided with a successive phase shift as it propagates along the input coupling means. The second input signal terminal is fed by a second signal which is fed to the second input electrode of each non-linear element. In response to the pair of input signals, an output signal is provided at the output electrode of each one of said non-linear devices having frequency components equal to the sum and difference of the frequencies of the first and second input signals. In a preferred embodiment, the frequency propagation characteristics of the output coupling means are selected to support propagation of a first one of said sum and difference frequency components of the output signal.

Description

BACKGROUND OF THE INVENTION

This invention relates generally to radio frequency circuits and more particularly to frequency conversion circuits.

As is known in the art, frequency conversion circuits provide an output signal in response to an input signal having a frequency which is higher or lower than the frequency of the input signal. One type of frequency conversion circuit is the so-called down converter or a mixer. Radio frequency mixer circuits are widely employed in superheterodyne receivers which include in addition to the mixer an intermediate frequency (IF) amplifier tuned to a predetermined frequency provided from the mixer and a fixed frequency detector which is fed the amplified signal from the IF amplifier. Generally, a received input signal and a local oscillator (LO) signal are fed to the mixer circuit to provide an output signal having a pair of frequency components equal to the sum and difference of the frequencies of the input signal and the local oscillator signal. Typically, the sum frequency component is filtered from the signal and the signal having the difference frequency component is fed to the IF amplifier.

One type of mixer is the so-called single-ended mixer. The single-ended mixer generally includes a nonlinear device such as a transistor or a diode which is fed by the input signal and the local oscillator signal. In response an output signal is provided having the sum and difference pair of frequencies (i.e., .omega..sub.rf .+-..omega..sub.LO).

One problem with this type of mixer, however, is that the output signal generally includes undesirable frequency components such as the input signal frequency .omega..sub.rf, the local oscillator signal frequency .omega..sub.LO, harmonics (n.omega..sub.rf, m.omega..sub.LO) of the original input signals, intermodulation products of the harmonics (m .omega..sub.LO .+-.n.omega..sub.rf), and a DC output level. It is generally required to suppress the undesired frequency components, since their presence may cause frequency ambiguity in the receiver. A second problem with single-ended mixers is that the local oscillator signal terminal and the radio frequency input signal terminal generally are not isolated. Thus, a portion of the local oscillator signal may feed through to the radio frequency terminal causing radiation of the local oscillator signal and additional interference problems.

A further problem with the single-ended mixer is the so-called image response of the mixer. The desired IF frequency .omega..sub.IF can be produced by an r.f. signal having a frequency above or below the frequency of the local oscillator (.omega..sub.rf =.omega..sub.LO .+-..omega..sub.IF). If one of the IF frequencies is the desired frequency, then the other frequency is termed the image frequency. If one input signal produces the desired IF signal, then the other input signal produces a signal referred to as the image signal. In many applications, it is necessary to either distinguish between the desired and image signals or eliminate the image signal. With a single-ended mixer, there is generally no way to distinguish between the desired signal and the image signal; therefore, filtering is employed to eliminate the image signal. Fixed frequency filtering, however, is possible only when there is no overlap in the bandwidth of the desired signal and the image signal and tunable frequency filtering are generally difficult to fabricate over wide bandwidths, particularly, as integrated circuits. Therefore, for broadband applications, filtering is generally not successfully employed with single-ended mixers.

A third problem with the single-ended mixer is that the mixer not only responds to a signal at the image frequency, it may produce a signal at the image frequency. The mixer generally produces such a signal in one of two ways: First, an r.f. signal having a frequency higher or above the local oscillator frequency (given by .omega..sub.rf =.omega..sub.LO +.omega..sub.IF) if mixed with the second harmonic of the local oscillator signal 2 .sub.LO will produce a signal at the image frequency (.omega..sub.IM) (given by .omega..sub.IM =2.omega..sub.LO -.omega..sub.rf).

Alternatively, an impedance mismatch at the IF output terminal will produce a reflected signal which will propagate back towards the mixer. This reflected signal mixes with the local oscillator signal to produce a signal either above or below the local oscillator signal (.omega..sub.LO .+-..omega..sub.IF), one of which is at the image frequency. Nevertheless, in either case, the signal produced at the image frequency increases the conversion loss of the mixer, or in other words, reduces the efficiency by which the mixer converts the input signal to the desired IF signal.

A second type of mixer is the so-called balanced mixer. A balanced mixer generally includes a pair of single-ended mixers and a 3 db hybrid coupler. The inputs of the coupler are fed the input signal and local oscillator signal and outputs of the hybrid are coupled to the input of each single-ended mixer. The outputs of the mixer elements are combined in a common junction to provide the output IF signal. The balanced mixer generally has a relatively high isolation between input and local oscillator signals if the outputs of the hybrid are properly terminated thereby reducing re-radiation of the LO signal. The hybrid provides a predetermined phase shift between the input signal and local oscillator signal generally equal to 90.degree. or 180.degree.. Balanced mixers having a 90.degree. phase shift between the local oscillator signal and the input signal are generally used to suppress harmonics and intermodulation products for both the input signal and the local oscillator signal. Balanced mixers having a 180.degree. phase shift between the signals are generally used to suppress the even harmonics of the local oscillator frequency signal. One problem with either of the balanced mixers is that while the above-recited balanced mixer configurations suppress some of the unwanted frequency components in the output, certain others of those frequency components are not suppressed. A second problem is that the balanced mixer includes a hybrid coupler, a generally difficult circuit to fabricate, particularly as an integrated circuit. Furthermore, the balanced mixer may produce energy at the image frequency. Also, the balanced mixer cannot differentiate between input signals having frequencies either above or below that of the local oscillator signal. Further still, since the balanced mixer requires the use of a hybrid coupler, the bandwidth of the balanced mixer is generally limited due to the limited bandwidth of the hybrid coupler.

A third type of mixer is the so-called double balanced mixer. The double balanced mixer generally includes two balanced mixers, each fed by the local oscillator signal and one of a pair of input signals having a 180.degree. differential phase shift provided by a hybrid coupler or balun. The outputs of each balanced mixer are combined together by an IF hybrid coupler. One problem with double balanced mixers is that the couplers or baluns are generally difficult to fabricate as monolithic integrated circuits. Accordingly, the double balanced mixer is not easily fabricated as a monolithic integrated circuit. A second problem with the double balanced mixer is a relatively narrow frequency bandwidth of operation of the mixer due to the presence of the couplers and baluns. Further, while certain configurations of double balanced mixers may distinguish between signals having a frequency either higher or lower than that of the local oscillator, those configurations of the double balanced mixers may still produce a signal at the image frequency.

As is also known in the art, a second type of frequency conversion circuit, an up converter, provides an output signal having a frequency which is higher than the frequency of the input signal. As mentioned above, a nonlinear device when fed input and local oscillator signals, provides an output signal having frequency components equal to the sum and difference between the frequencies of said signals. Therefore, by filtering the difference frequency component, the remaining sum frequency component provides a signal having a frequency which is higher than the frequency of the input signal.

SUMMARY OF THE INVENTION

In accordance with the present invention, a frequency conversion circuit having first and second input signal terminals, and first and second output terminals includes a plurality of nonlinear elements, each one having a pair of input electrodes and an output electrode. The first input electrode of each nonlinear element is successively interconnected by a two port input coupling means, said input coupling means having a first end coupled to the first input signal terminal and a second end terminated in a terminating impedance. The output electrode of each nonlinear device is successively interconnected by a two port output coupling means coupled between the pair of output terminals. An input signal is fed to the first input signal terminal and is coupled to the input coupling means and a portion of such signal is fed to the first input electrode of each nonlinear device. Each portion of the input signal is provided with a successive phase shift as it propagates along the input coupling means. The second input signal terminal is fed by a second signal which is fed to the second input electrode of each nonlinear element. In response to the pair of input signals, an output signal is provided at the output electrode of each one of said nonlinear devices having frequency components equal to the sum of the frequencies of the first input signal and the difference in frequency between the input signals. In a preferred embodiment, the frequency propagation characteristics of the output coupling means are selected to support propagation of a first one of said sum and difference frequency components of the output signal. With such an arrangement, a frequency conversion circuit is provided having a frequency characteristic selected in accordance with the frequency propagation characteristics of the output coupling means. This generally results in elimination or reduction of external filtering requirements.

In accordance with a further aspect of the present invention, the plurality of nonlinear elements include a transistor biased, for example, to operate as a half-wave vth law device, such as a half-wave square law device. Preferably, each transistor is a field effect transistor having the pair of input electrodes, the output electrode and a reference electrode. Each transistor has the first input electrode interconnected by the input coupling means and the output electrode successively interconnected by the output coupling means. The input coupling means is fed the first input signal and provides signals each having a successively increasing phase shift which are fed to the first input electrode of each transistor. The second input electrode of each transistor is fed an equal amplitude, in-phase portion of the second or local oscillator signal. In response, output signals from each output electrode are coupled to the output coupling means having frequency components equal to the sum of and difference between the frequencies of the input and local oscillator signals. The propagation characteristics of the output coupling means are selected to support propagation of one of the sum and difference frequency components and provide a predetermined phase shift to each of the output signals. The phase shifts from the input and output coupling means are selected to have the output signals from each output electrode add in-phase at a first output terminal and provide a null or attenuated signal at the second output terminal, in accordance with, the frequency of the input signal with respect to the frequency of the local oscillator signal. Accordingly, the mixer circuit discriminates between the desired frequency signal and image frequency signal. Further, a signal produced at the image frequency by each FET has a phase shift characteristic which changes in accordance with changes in phase of the input signal. The output signals produced when the signal at the image frequency is down-converted to the IF frequency likewise have phase shifts which change in accordance with the desired IF signal. Therefore, a signal produced at the image frequency, by the mixer circuit, will propagate toward the same output terminal as the desired signal, thereby reducing the conversion loss of the mixer circuit.

In accordance with a further aspect of the present invention, the phase and amplitude characteristics of output signals fed to the first and second outputs are tailored, such that, at a first output terminal the output signals combine in-phase, and at a second output terminal the signals are out-of-phase such that a null or attenuated signal is provided. Signals are fed to the output terminals in accordance with the frequency of the input signal relative to the frequency of the local oscillator, over a wide bandwidth of input and intermediate signal frequencies. In one embodiment, the amplitude tailoring is provided by capacitively coupling a selected portion of the input signal to each input signal gate electrode. Phase shift tailoring is provided by selecting the phase shift characteristics of each one of the distributed input signals fed to first input electrodes and distributed output signals fed from the output electrodes. With such an arrangement, amplitude tailoring and phase tailoring permit discrimination between an input signal having a frequency either above or below the local oscillator signal over a wide band of input signal and intermediate signal frequencies.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing features of this invention, as well as the invention itself, may be more fully understood from the following detailed description of the drawings, in which:

FIG. 1 is a schematic diagram of a frequency conversion circuit, here a mixer circuit;

FIG. 2 is a plan view of the circuit of FIG. 1 fabricated as a microwave monolithic integrated circuit;

FIG. 3 is an exploded view of a portion of FIG. 2;

FIG. 4 is a cross-sectional view taken along line 4--4 of FIG. 3;

FIG. 5A is a simulated plot of input voltage signal distribution along the input coupling means;

FIG. 5B is a simulated plot of the input circuit relative phase differences between succeeding adjacent pairs of non-linear device:

FIG. 6 is a simulated plot of the magnitude of the transfer characteristic of the circuit of FIG. 1 in accordance with the input signal profile of FIG. 5A and relative phase difference of FIG. 5B;

FIG. 7 is a typical noise spectrum produced by a local oscillator frequency signal;

FIG. 8 is a schematic diagram of an alternate embodiment of a frequency conversion circuit, here a mixer circuit;

FIGS. 9-11 are schematic diagrams for alternate embodiments of non-linear elements for use with the circuits described in conjunction with FIGS. 1 and 8;

FIG. 12 is a schematic diagram of a further alternate embodiment of a frequency conversion circuit, here a frequency multiplier;

FIG. 13 is a schematic diagram of an alternate embodiment of a non-linear device for use in the invention as shown in FIG. 12;

FIG. 14 is a schematic diagram of a further alternate embodiment of a frequency conversion circuit; and

FIG. 15 is a schematic diagram for an alternate embodiment of the non-linear elements for use with the circuit of FIG. 14.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Referring now to FIG. 1, a frequency conversion circuit, here a mixer circuit 10 is shown to include a plurality of here four non-linear elements 15a-15d, each one having input electrodes 15a.sub.1 -15d.sub.1, 15a.sub.2 -15d.sub.2 and output electrodes 15a.sub.3 -15d.sub.3. Here each non-linear element 15a-15d includes a dual-gate field effect transistor FET 1-FET 4, biased to operate as a non-linear device, for example FET 1-FET 4, may be biased to operate as a half wave vth law device such as a half wave square law device. Each FET 1-FET 4 is successively coupled between an input terminal 13, and a pair of output terminals 19a, 19b, and have first gate electrodes G.sub.1a -G.sub.4a, second gate electrodes G.sub.1b --G.sub.4b, drain electrodes D.sub.1 -D.sub.4 connected to respective electrodes of non-linear elements 15a-15d,respectively, as shown, and source electrodes S.sub.1 -S.sub.4. A transmission line T.sub.1, here a microstrip transmission line, is coupled at a first end to the input terminal 13 via a d.c. blocking capacitor C.sub.1 and a second end to an RF signal source 11. First gate electrodes G.sub.1a -G.sub.4a are shown successively interconnected together by input coupling means 16, here a traveling wave structure comprising a plurality of two port phase shift elements such as an artificial line or a distributed line. Here the two port elements comprise a plurality of distributed transmission lines T.sub.2 -T.sub.5, here each distributed transmission line T.sub.2 -T.sub.5 being a microstrip transmission line. Transmission line T.sub.5 is coupled at a first end to transmission line T.sub.4 and at a second end to a gate line termination matching network 22. Transmission line T.sub.1 is fed an input signal (V.sub.rf) which propagates along transmission lines T.sub.1 -T.sub.5. Selected portions V.sub.rf1 -V.sub.rf4 of input signal V.sub.rf are coupled to respective gate electrodes G.sub.1a -G.sub.4a preferably through a plurality of coupling capacitors C.sub.2 -C.sub.5. Each transmission line T.sub.2 -T.sub.5 has a selected impedance and length to provide in combination with coupling capacitors C.sub.2 -C.sub.5 a predetermined phase shift with respect to signal V.sub.rf to each signal portion V.sub.rf1 -V.sub.rf4 of the input signal V.sub.rf. The coupling capacitors C.sub.2 -C.sub.5 in addition to contributing to the phase shift characteristics imparted to the signals V.sub.rf1 -V.sub.rf4 also provide in combination with the intrinsic capacitance (not shown) between gates G.sub.1a -G.sub.4a and source electrodes S.sub.1 -S.sub.4 input signals having predetermined amplitudes. Suffice it here to say that the impedance and length of transmission lines T.sub.2 -T.sub.5 and capacitance of coupling capacitors C.sub.2 -C.sub.5 are selected to provide predetermined phase and amplitude relationships to input signals V.sub.rf1 -V.sub.rf4.

Second gate electrodes G.sub.2a -G.sub.2d are each fed a second input signal, here an in-phase equal amplitude signal from a common signal source V.sub.L. Signal source V.sub.L feeds an input terminal 18a to a power divider circuit 18 which provides four equal amplitude in-phase signals V.sub.LO (herein referred to as a local oscillator signal) at ports 18b-18e.

Drain electrodes D.sub.1 -D.sub.4 are successively interconnected together by a second, here output or drain coupling means 17, here a traveling wave structure with said output coupling means 17 comprising a plurality of two port phase shifting elements such as an artificial line or a plurality of distributed transmission lines T.sub.7 -T.sub.11 and a plurality of capacitors C.sub.8 -C.sub.10, as shown. The transmission lines T.sub.8 -T.sub.10 are here mutually coupled microstrip transmission lines. The mutual coupling of the transmission lines T.sub.8 -T.sub.10, and capacitors C.sub.8 -C.sub.10 here provide the phase shifting elements having electrical pathlengths to provide the requisite phase shift at the frequency of the IF signals. Alternatively, artificial lines comprising a plurality of lumped inductors and capacitors may be used. The particular choice of construction for the phase shift elements for either coupling means 16 or 17 is determined in accordance with the frequency characteristic of signals which propagate through such coupling means 16, 17.

The mixer circuit 10 is shown to further include the gate line termination network 22 coupled to transmission line T.sub.5, including a resistor R.sub.1, capacitor C.sub.6, and a transmission line T.sub.6, as shown. A direct current path (DC path) to couple a gate bias source V.sub.GG to gates G.sub.1a -G.sub.1d is provided through resistors R.sub.G1 -R.sub.G4, as shown. Drain bias is here provided by an external biasing arrangement (not shown).

In operation, input signal portion V.sub.rf1 is coupled to gate electrode G.sub.1a through coupling capacitor C.sub.2. As previously mentioned, capacitor C.sub.2 has a capacitance selected in accordance with the inherent or intrinsic capacitance (not shown) of FET 1 to provide a voltage divider and hence to provide a selected amplitude to signal V.sub.rf1. Succeeding portions V.sub.rf2 -V.sub.rf4 of the input signal V.sub.rf are coupled to gate electrodes G.sub.2a -G.sub.4a through coupling capacitors C.sub.3 -C.sub.5. In a similar manner, capacitors C.sub.3 -C.sub.5 each have a capacitance selected in accordance with the intrinsic gate-source capacitance of FET 2-FET 4, respectively, to provide signals coupled to said gate electrodes having a selected or tailored amplitude. Each one of said signals V.sub.rf1 -V.sub.rf4 has a predetermined electrical phase shift related to the number of transmission line sections through which the signal propagates beyond transmission line T.sub.2, as well as, the capacitance of the respective coupling capacitors.

Local oscillator signals V'.sub.LO here of equal amplitude and equal phase are fed to the second gate electrodes G.sub.1b -G.sub.4b and in combination with signals V.sub.rf1 -V.sub.rf4 produce output signals V.sub.IF1 -V.sub.IF4 at drain electrodes D.sub.1 -D.sub.4. The output signals V.sub.IF1 -V.sub.IF4 each include frequency components corresponding to the sum frequency (.omega..sub.rf +.omega..sub.LO), the difference frequency .vertline..omega..sub.RF -.omega..sub.LO .vertline., as well as, the frequencies of the original signals .omega..sub.rf, .omega..sub.LO, harmonics of the input signal N.omega..sub.rf and the local oscillator signal M.omega..sub.LO, as well as, intermodulation products of the components (N.omega..sub.rf .+-.M.omega..sub.LO). Here the output coupling means comprising transmission lines T.sub.7 -T.sub.11 has characteristics selected to support propagation of the difference frequency signal .vertline..omega..sub.rf -.omega..sub.LO .vertline., and preferably, filters or blocks the frequency components of the RF signal, the local oscillator signal, the sum frequency component, the intermodulation products, the harmonics of the local oscillator frequency signal, and the harmonics of the radio frequency signal.

For the case where the input signal V.sub.rf has a frequency .omega..sub.rf less than the frequency .omega..sub.LO of the local oscillator signal, at a first one of the pair of terminals 19a, 19b, here terminal 19a, will be provided the intermediate frequency signal having a frequency corresponding to .omega..sub.IF =.omega..sub.LO -.omega..sub.rf, with the second one of said pair of terminals 19a, 19b, here terminal 19b providing a null or attenuated signal. For the second case, where the frequency of input signal .omega..sub.RF is greater than the frequency .omega..sub.LO of the local oscillator signal, at the second output terminal terminal 19.sub.b will be provided an intermediate frequency signal having a frequency corresponding to .omega..sub.IF =.omega..sub.rf -.omega..sub.LO and at terminal 19a will be provided a null or attenuated signal. With this arrangement, the radio frequency mixer discriminates between radio frequency signals having a frequency either above or below that of the local oscillator signal. That is, the mixer discriminates between the desired signal and the image signal. The following discussion and accompanying Table 1 are useful in understanding how the mixer circuit 10 discriminates between the desired IF signal and image IF signal.

TABLE I__________________________________________________________________________.omega..sub.rf > .omega..sub.LO .omega..sub.rf < .omega..sub.LOFET .omega..sub.rf phase shift IF freq. phase shift .omega..sub.rf phase shift IF freq. phase shift__________________________________________________________________________1 .omega..sub.LO + .omega..sub.IF 0 .omega..sub.IF 0 .omega..sub.LO - .omega..sub.if 0 .omega..sub.IF 02 .omega..sub.LO + .omega..sub.IF .theta. .omega..sub.IF .theta. .omega..sub.LO - .omega..sub.if .theta. .omega..sub.IF -.theta.3 .omega..sub.LO + .omega..sub.IF 2.theta. .omega..sub.IF 2.theta. .omega..sub.LO - .omega..sub.if 2.theta. .omega..sub.IF -2.theta.4 .omega..sub.LO + .omega..sub.IF 3.theta. .omega..sub.IF 3.theta. .omega..sub.LO - .omega..sub.if 3.theta. .omega. .sub.IF -3.theta.__________________________________________________________________________ As an illustrative example, consider the case where each one of lines T.sub.3 -T.sub.5 and T.sub.8 -T.sub.10 provide phase shifts of .theta..sub.rf and .theta..sub.IF, respectively, where .theta..sub.rf and .theta..sub.IF are the same phase shift equal to 90.degree. at the respective rf and IF frequencies where this condition is satisfied, where the phase shift contribution of the FETS, as well as, the capacitors can be ignored, and where C.sub.2 =C.sub.3 =C.sub.4 =C.sub.5 =.infin..

Over a narrow frequency band, for .omega..sub.rf <.omega..sub.LO, signals V.sub.1a -V.sub.4a will substantially add in-phase at terminal 19a and will provide substantially a null or attenuated signal at terminal 19b. Similarly, over a narrow band where .omega..sub.rf >.omega..sub.LO, signals V.sub.1b -V.sub.4b will add in-phase at terminal 19b and will provide a null or attenuated signal at terminal 19a.

This may be demonstrated as follows: V.sub.rf1 has a relative phase shift of 0.degree. with respect to itself and thus V.sub.IF1 similarly has a relative phase shift of 0.degree.. Thus, signal component V.sub.IF1a has a phase shift of 0.degree. at terminal 19a, whereas signal component V.sub.IF1b has a phase shift of 3.theta..sub.IF at terminal 19b. The intermediate signal V.sub.IF2 has a phase shift relative to V.sub.IF1 corresponding to .theta..sub.rf. Signal component V.sub.IF2a which propagates toward terminal 19a has a composite phase shift of .theta..sub.rf +.theta..sub.IF =2.theta. and the signal component V.sub.IF2b which propagates toward terminal 19b has a composite phase shift of .theta..sub.rf +2.theta..sub.IF =3.theta.. Thus, at terminal 19a signal component V.sub.IF2a in combination with the signal component V.sub.IF1a provides a null or strongly attenuated signal since components V.sub.IFla, V.sub.IF2a are substantially of equal magnitude and 180.degree. out-of-phase (as assumed .theta.=90.degree.). The second signal component V.sub.2b combines in-phase at terminal 19b with signal V.sub.IF1b. Similarly, the intermediate signal V.sub.IF3 has a phase of 2.theta..sub.rf. Signal component V.sub.IF3a propagates toward terminal 19a having a composite phase shift of 2.theta..sub.rf +2.theta..sub.IF =2.pi.+0.degree. and signal component V.sub.IF3b propagates toward terminal 19b having a composite phase shift of 2.theta..sub.rf +.theta..sub.IF =3.theta.. Signal V.sub.IF3b combines in-phase at terminal 19b with the signals V.sub.IF1b, V.sub.IF2b. The fourth signal V.sub.4IF has a phase shift of 3.theta..sub.rf. Signal component V.sub.IF4a propagates towards terminal 19a having a composite phase shift of 3.theta..sub.rf +3.theta..sub.IF =2.pi.+2.theta. and signal component VIF4b propagates toward terminal 19b having a composite phase shift of 3.theta..sub.rf. Signal V.sub.IF4a arrives at terminal 19a having a phase shift equivalent to 2.theta. thus out-of-phase with signal component V.sub.IF3a. Thus, a null signal is provided at terminal 19a. On the other hand, signal V.sub.IF4b arrives at terminal 19b having a relative phase shift of 3.theta..sub.IF or 3.theta. and therefore adds in-phase at terminal 19b with the signals V.sub.IF3b, V.sub.IF2b and V.sub.IF1b previously coupled to terminal 19b. Accordingly, it is now clear that by selecting the phase shifts between input signals, and the IF output signals from adjacent pairs of transistors (i.e., FET 1, FET 2 and FET 3, FET 4) the IF signals will be out-of-phase at terminal 19a, but will be in-phase at terminal 19b.

For the case where .omega..sub.LO >.omega..sub.rf, the difference frequency signal (.omega..sub.LO -.omega..sub.rf) will provide a signal at terminal 19a signal having the frequency .omega..sub.IF =.omega..sub.LO -.omega..sub.rf and a null or strongly attenuated signal at terminal 19b. This can be seen by a similar analysis as described above. From each drain electrode a signal will propagate in each of two directions toward terminals 19a, 19b. However, the initial phase shifts from signals V.sub.IF1 -V.sub.IF4 which couple from drain electrodes D.sub.1 -D.sub.4 will have a phase lag rather than phase lead, as also shown in Table I, that is, will have a negative phase shift. Therefore, the phases of the signals coupled from each one of the drain electrodes D.sub.1 -D.sub.4 will couple to output terminal 19a having a relative phase shift of 0.degree. and will couple to terminal 19b having relative phase shifts of -.theta. or +.theta.. Therefore, at terminal 19a will be provided the intermediate frequency signal .omega..sub.IF equal to (.omega..sub.LO -.omega..sub.rf) and at terminal 19b will be provided the null or strongly attenuated signal.

Preferably, however, the capacitors C.sub.2 -C.sub.5, rather than being equal, as assumed above, are selected in accordance with the inherent capacitance (not shown) between the first gate and source of each FET to provide selected input signal amplitude tailoring to the signals fed to each first gate electrode G.sub.1a -G.sub.4a. The transmission lines T.sub.3 -T.sub.5 and/or T.sub.7 -T.sub.9 have selected impedances and electrical pathlengths, as described above to provide selected phase shift tailoring. With this arrangement, capacitors C.sub.2 -C.sub.5 are here used to tailor the input voltage to each FET as a function of frequency. Further, the capacitors C.sub.2 -C.sub.5 and transmission lines T.sub.3 -T.sub.5, T.sub.7 -T.sub.9 provide tailored electrical pathlengths to signals V.sub.rf1 -V.sub.rf4 or V.sub.IF1 -V.sub.IF2 in order to provide the high directivity of propagation of IF signal components Vla-V4a toward terminal 19a and V.sub.1b -V.sub.4b toward terminal 19b over a wide bandwidth.

Referring now to FIGS. 2-4, the mixer 10 (FIG. 1) is shown fabricated as a monolithic integrated circuit 10'. The mixer 10' is formed on a substrate 40, here comprised of gallium arsenide or other suitable Group III-V material or other semiconductor material. The substrate 40 has formed on a bottom surface portion thereof, a ground plane conductor and on an upper surface portion thereof, a mesa-shaped epitaxial layer 44 (FIGS. 3, 4). The mesa-shaped layer 44 provides active regions for field effect transistors FET 1-FET 4. Here, each FET is identical in construction and includes a pair of source electrodes and a common drain electrode spaced by one gate electrode of each of a pair of gate electrodes (not numbered), as shown. Thus considering an exemplary one of such field effect transistors, here FET 4, such transistor is shown in FIGS. 3 and 4 to have a common drain pad D.sub.4 connected by strip conductors T.sub.s10, T.sub.s11 and a pair of source contacts S.sub.4a, S.sub.4b connected to ground plane conductor 42 through plated via holes 47 (FIG. 2). FET 4 is shown to further include a gate pad G.sub.4a having connected thereto a pair of gate fingers G.sub.4a1, G.sub.4a2 and a second gate pad region G.sub.4b having a second pair of gate fingers G.sub.4b1, G.sub.4b2. Source electrodes S.sub.4a and S.sub.4b are spaced from drain electrode D.sub.4 by one gate finger of each pair G.sub.4a1, G.sub.4a2 and G.sub.4b1, G.sub.4b2 of gate fingers. A conductor 45 is formed on the substrate 40 to feed the gate bias signal from gate bias source (not shown) to each one of the gate electrodes. The conductor 45 is coupled to each one of the gate electrodes through resistors R.sub.G1 -R.sub.G4, as shown. Here resistors R.sub.G1 -R.sub.G4 are provided by open or floating gate MESFETS, as shown in FIGS. 3 and 4 for resistor R.sub.G4. Resistors R.sub.G1 -R.sub.G4 each provide a relatively high resistance approximately equal to 2K ohms. The drain and source electrodes D, S provide the ohmic contact terminals for resistor R.sub.G4. Coupling capacitor C5, as shown in FIGS. 3, 4, includes a first electrode C.sub.5a disposed on the semi-insulating substrate 40, a dielectric C.sub.5b disposed over the first electrode C.sub.5a, and a second electrode contact C.sub.5c disposed over a portion of the dielectric C.sub.5b. The second electrode contact C.sub.5c is directly connected to strip conductor T.sub.s5 and resistor R.sub.1, here a metal film resistor. The bottom contact C.sub.5a of capacitor C.sub.5 is connected to gate electrode fingers G.sub.4a1 and G.sub.4a2. The bottom contact C.sub.5a is connected to the aforesaid mentioned gate contact G.sub.4a of field effect transistor FET 4. Thus, an input signal propagating along transmission lines T.sub.2 -T.sub.6 is coupled to gate electrodes G.sub.4a1, G.sub.4a2 through capacitor C.sub.5, whereas, the bias signal provided from resistor R.sub.G1 is coupled directly to gate electrode fingers G.sub.4a1, G.sub.4a2. A similar arrangement is provided with field effect transistor FET 1. The gate electrodes G.sub.2a, G.sub.3a of field effect transistors FET 2 and FET 3 are here connected directly to the transmission line sections T.sub.2, T.sub.3 and T.sub.4.

As further shown in FIG. 2, the strip conductors T.sub.s8 -T.sub.s10 are arranged to provide a mutually coupled line, that is, the strip conductor is arranged in a coiled fashion having upper portions bridging over underlying portions of the strip conductors T.sub.s8 -T.sub.s10. Capacitors C.sub.8 -C.sub.10 are shunted to ground by interconnections to source pads, as shown. The width of the strip conductors, the gap between strip conductors and the length of the strip conductors are chosen for each one of said strip conductors T.sub.s8 -T.sub.s10 to provide a predetermined electrical pathlength at the intermediate frequency signal of the mixer, as shown for example in Table 2. Here mutually coupled transmission line sections T.sub.8 -T.sub.10 and capacitors C.sub.8 -C.sub.10 are provided to selectively increase the electrical pathlength of each one of the transmission lines over the corresponding frequency bandwidth of the intermediate frequency signals, here 2 to 8 GHz. With this arrangement, the output coupling means comprises substantially distributed transmission line sections, whereas, the relatively large electrical pathlengths required for the distributed transmission lines between each one of the field effect transistors is provided, in part, by the mutual coupling between the lines, as well as, the capacitors C.sub.8 -C.sub.10 shunted to ground.

A mixer 10, as described in FIGS. 1-4, was designed the following characteristics: The circuit was modeled to be fabricated on a 4 mil thick GaAs substrate having disposed thereon four dual gate MESFETs, each MESFET having a 200 .mu.m gate periphery. The mixer was designed to operate over an input signal bandwidth of 14 GHz to 20 GHz, IF bandwidth 2 Ghz to 8 Ghz at a local oscillator frequency of 12 GHz. All component values given below in Table 2 are modeled at 17 Ghz for input circuit components and 5 GHz for output circuit components. The phase shift of input transmission lines includes the phase shift provided by capacitors C.sub.2 and C.sub.5, and is measured with respect to the phase of the signal V.sub.1rf at the gate electrode G.sub.1a.

TABLE 2______________________________________Transmission Lines CapacitorsWidth of ElectricalStrip Conductors Pathlength Capacitance______________________________________Input circuitT.sub.3 .049 mm .theta..sub.3 = 55.degree. C.sub.2 = 0.15 pfT.sub.4 .036 mm .theta..sub.4 = 61.degree. C.sub.3 = .alpha.T.sub.5 .026 mm .theta..sub.5 = 59.degree. C.sub.4 = .alpha. C.sub.5 = 0.37 pfOutput circuit (W) (G) (L) width gap lengthT.sub.7 .025 mm .028 mm 1.2 mm C.sub.8 = 0.5 pfT.sub.8 .025 mm .018 mm 1.0 mm C.sub.9 = 0.5 pfT.sub.9 .025 mm .028 mm 1.2 mm C.sub.10 = 0.5 pf______________________________________

Referring to FIGS. 5A, 5B and 6, the input voltage profile at each FET as a function of frequency (FIG. 5A), the relative differential phase shift between succeeding adjacent pairs of input electrodes as a function of frequency (FIG. 5B), and the IF frequency output response (FIG. 6) at terminals 19a, 19b (FIG. 1) are shown, respectively, for the circuit designed in accordance with the parameters of Table 2. FIG. 5A shows the general shape of the input voltage amplitude distribution chosen to provide the relatively flat, desired output response (FIG. 6) at the IF frequency at terminal 19b and provides a null or attenuated response at terminal 19a. The attenuated input excitations to FET 1 and FET 4 are here provided by coupling capacitors C.sub.2 and C.sub.5, respectively. Furthermore, the varied transmission line capacitance also contributes to the general shape of the voltage distribution at each FET. Here the input coupling means 16 was optimized to provide the desired flat output response (FIG. 6) at terminal 19b (curve 68) and a null or attenuated signal at terminal 19b (curve 70) in response to r.f. input signals having a frequency greater than that of the local oscillator. However, given a desired output response at terminal 19a for input signals having a frequency less than that of the local oscillator, the input circuit could be re-optimized, as will now become apparent to those skilled in the art, by changing the values of capacitors C.sub.2 -C.sub.5, capacitances of lines T.sub.2 -T.sub.5 and electrical pathlengths of lines T.sub.2 -T.sub.5. Moreover, the input coupling means may be further re-optimized to provide an output signal response at a first one of terminals 19a, 19b and a null or attenuated response at the second one of terminals 19a, 19b in accordance with the frequency of the input signal as described above.

As shown in FIG. 5B, the differential phase (expressed as a phase lag) between the first gate electrodes G.sub.1a -G.sub.4a of successive adjacent pairs of FETS (FET 2 and FET 1, curve 62, FET 3 and FET 2, curve 64, FET 4 and FET 3, curve 66) was selected to be approximately 90.degree. at the midband frequency (here 17 GHz). However, even though the phase shift changed to 70.degree. at 14 GHz and to about 115.degree. between FET 2 and FET 3 at 20 GHz, the output response (FIG. 6) still shows relatively good directivity of the IF signal produced by the desired input signal at the extremes (2 and 8 GHz) of the IF frequency band.

Preferably, the impedance of transmission lines T.sub.2 -T.sub.5 is also selected in accordance with capacitors C.sub.2 -C.sub.5, and the intrinsic capacitance (not shown) between the gate electrodes G.sub.1a -G.sub.4a and source electrodes S.sub.1 -S.sub.4 to provide the mixer 10 with a predetermined input impedance, preferably related to the impedance of transmission line T.sub.1. In a similar manner, the characteristic impedance of transmission lines T.sub.8 -T.sub.10 is selected in accordance with the intrinsic capacitance (not shown) between drain electrodes D.sub.1 -D.sub.4 and source electrodes S.sub.1 -S.sub.4, respectively, to provide the mixer 10 with a predetermined output impedance. Such an arrangement to provide selected input and output impedances is described in U.S. Pat. No. 4,456,888 issued June 26, 1984, and assigned to the same assignee as the present invention.

The field effect transistor when biased as a non-linear device generates spurious or undesirable frequency components, one of which is the second harmonic of the local oscillator signal. The mixing of the second harmonic of the LO and the input signal (.omega..sub.rf -2.omega..sub.LO) produces a signal at the image frequency .omega..sub.IM. This signal is particularly undesirable because the signal produced at the image band frequency increases the conversion loss of the mixer. However, with the above arrangement, because the phase of the input signal changes at each field effect transistor in accordance with the propagation of the input signal through transmission lines T.sub.2 -T.sub.5 and the IF signals through transmission lines T.sub.8 -T.sub.10, the phase of the image signal created by mixing the second harmonic of the LO and the input signal also changes as does the phase of the IF frequency when the signal produced at the image frequency is down-converted to the IF frequency. The phases of the image rf and IF signals are shown in Table 3 below. The direction of the phase progression indicates that the signal produced at the image frequency will propagate back in the direction of the signal source. Along this propagation, some of this signal is fed back to FET 1-FET 4 and down-converted to the IF image frequency. The resulting IF image signal will propagate in the same direction as the desired IF signal provided in response to the input signal, and therefore, the signal produced at the image frequency will be provided at the same terminal as the desired IF signal. This will reduce conversion loss of the mixer circuit.

TABLE 3______________________________________.omega..sub.rf > .omega..sub.LO Image signal IF from Image phase phase phaseFET rf shift freq. shift freq. shift______________________________________1 .omega..sub.LO + .omega..sub.IF 0 .omega..sub.LO - .omega..sub.IF 0 .omega..sub.IF 02 .omega..sub.LO + .omega..sub.IF .theta. .omega..sub.LO - .omega..sub.IF .theta. .omega..sub.IF .theta.3 .omega..sub.LO + .omega..sub.IF 2.theta. .omega..sub.LO - .omega..sub.IF 2.theta. .omega..sub.IF 2.theta.4 .omega..sub.LO + .omega..sub.IF 3.theta. .omega..sub.LO - .omega..sub.IF 3.theta. .omega..sub.IF 3.theta.______________________________________

The mixer circuit 10 (FIG. 1) also provides substantially complete cancellation of noise introduced into the IF band by the local oscillator signal. As seen in FIG. 7, noise associated with the local oscillator frequency signal is in general clustered in a relatively narrow band about the local oscillator frequency f.sub.LO. However, while this condition is not a prerequisite for noise cancellation to occur in the circuit described in FIG. 1, assumption of this condition makes the following analysis easier.

For noise cancellation to occur in the circuit of FIG. 1, it is generally required that the local oscillator frequency excitation of each field effect transistors FET 1-FET 4 be in-phase and of equal amplitude. Each frequency component of the noise, therefore, will mix with the local oscillator carrier frequency and be down-converted to the IF band. This includes noise in both the upper and lower side bands of the local oscillator signal. The phase difference between the noise components of the IF ports, IF.sub.1 and IF.sub.2, will be substantially equal to zero since the local oscillator excitation of each field effect transistor FET 1-FET 4 arises from input signals provided from the same signal source having substantially equal phase and amplitude.

Assuming that the IF noise voltage component at the nth IF port is denoted by e.sub.n (f.sub.m) where f.sub.m is the frequency offset of the noise component from the carrier such as shown in FIG. 8, and, n is the number of IF ports (not necessarily equal to four), then because all e.sub.n signals are from the same local oscillator frequency source, these components are fully correlated and can be added algebraically. Taking the phase shift introduced between the IF ports by the IF circuit into consideration, the noise voltage corresponding to the frequency component f.sub.m at the two IF ports labeled 19a and 19b in FIG. 1 is given by: ##EQU1## For balanced mixers, e.sub. =e.sub.k for all n and k. Thus, ##EQU2## Therefore, for all other frequency components about the local oscillator carrier frequency, the above relationship is true. It is evident that complete noise cancellation of the frequency noise occurs when .theta. is chosen to satisfy the condition .theta.=2.pi./N for (N>1).

For example, for four field effect transistors, as shown in FIG. 1, N=4, the required phase shift between field effect transistor and IF terminals is 90.degree.. This simple case may be shown by a vector diagram (not shown) of the noise voltage components at each IF port. Because of the 90.degree. phase shift, the components are added vectorially as the four-sides of a square to a null vector.

Referring now to FIG. 8, an alternate embodiment of a frequency conversion circuit, here a mixer circuit 60 is shown. Mixer circuit 60 includes the aforementioned output coupling means 17, input coupling means 16, and a plurality of non-linear elements 15a-15d, here each including the field effect transistors FET 1-FET 4 as described in conjunction with FIG. 1. Mixer circuit 60 is shown to further include a second input traveling wave structure 62, here comprising a plurality of distributed transmission lines T.sub.12 -T.sub.16, here each one of said transmission lines being a microstrip transmission line. A first end of transmission line T.sub.16 is terminated in a characteristic impedance, here shown as R.sub.2, and a first end of said distributed transmission line T.sub.12 is coupled to a local oscillator signal V.sub.LO. Accordingly, local oscillator signal V.sub.LO propagates along transmission line sections T.sub.12 -T.sub.16, with portions of said signals V.sub.LO '.sub.1 -V.sub.L0 '.sub.4 being fed from said transmission lines to each one of the corresponding gate electrodes G.sub.1b -G.sub.4b. The local oscillator signal is fed to terminal 18a' and propagates along transmission mission lines T.sub.14 -T.sub.16 having progressively or successively increasing phase shifts. As before, the input signal V.sub.rf propagates along transmission lines T.sub.2 -T.sub.5 and is coupled to corresponding gate electrodes G.sub.1a -G.sub.4a. Since the local oscillator V.sub.LO and the input signal V.sub.rf propagate in opposite directions, the relative phase shift difference between the pair of signals will be the sum of the respective phase shifts at the input electrodes of each one of the field effect transistors. Accordingly, the pathlength of the traveling wave structure 62 for the local oscillator signal may be adjusted to provide additional control of the phase shift of the signals coupled to the output or drain electrodes of each one of the field effect transistors. Accordingly, the phase shift of the IF signal at the output of each one of the field effect transistors becomes the sum of the phase shift from the r.f. input signal line, and the phase shift from the local oscillator signal line. Preferably, as shown in FIG. 7, coupling capacitors C.sub.2 -C.sub.5 are used to provide amplitude tailoring to the input signals as described in conjunction with FIG. 1. Further, by providing a traveling wave structure 62 to feed the local oscillator signal to each FET, the local oscillator signal may be swept or varied over a broad range of frequencies.

Referring now to FIGS. 9-11, alternate embodiments for the mixer elements 15a-15d (FIG. 1) are shown. As an illustrative example, element 15a is shown in FIG. 9 having dual-gate field effect transistor FET 1 (FIG. 1) replaced by a diode D.sub.1 and a capacitor C coupled in shunt between a cathode of diode D.sub.1 and ground. The local oscillator signal V.sub.LO and the radio frequency input signal (V.sub.rf) are fed via terminal 15.sub.a1 to an anode of diode Dl with the cathode providing the output signal V.sub.IF1 at terminal 15.sub.a3. Similarly, as shown in FIG. 10, the dual-gate field effect transistor FET 1 (FIG. 1) may be replaced by a single gate field effect transistor, here FET 1', having a first one of drain and source electrodes D', S' being fed via terminal 15.sub.a1 by an RF signal (V.sub.rf) and the gate electrode, for example, being fed via terminal 15.sub.a2 by the local oscillator signal (V.sub.LO) with the second one of the drain and source electrodes providing the output signal V.sub.IFl at terminal 15a.sub.3. Similarly, as shown in FIG. 11, dual-gate field effect transistor FET 1 (FIG. 1) may be replaced by a pair of field effect transistors FET 2', FET 3'. The first field effect transistor FET 2' connected in a common source configuration has a gate electrode G.sub.2 ' fed, via terminal 15a.sub.1, the r.f. input signal V.sub.rf and provides an output signal to a first one of drain and source electrodes D.sub.3 ', S.sub.3 ' of the field effect transistor FET 3'. The field effect transistor FET 3' is fed, via terminal 15a.sub.2, the local oscillator signal (V.sub.LO) which modulates the transconductance of field effect transistor FET 3' to produce the mixed output signal V.sub.IFl at terminal 15a.sub.3. It can be seen, therefore, by substituting the dual-gate field effect transistors FET 1-FET 4 of FIG. 1 with any one of the above devices between the input coupling means 16 and output coupling means 17, alternate embodiments of the frequency conversion circuit may be provided.

Referring now to FIG. 12, a further alternate embodiment of a frequency conversion circuit, here a distributed frequency multiplier 80 is shown to include a first plurality or set of here four field effect transistors FET 11-FET 14 successively coupled between an input terminal 82 and an output terminal 83, via a plurality of two port phase shift elements such as an artificial line or a distributed line. Here again non-linear elements are provided with element 15a only shown for clarity. Here the two port elements comprise a plurality of distributed input transmission lines T.sub.21 -T.sub.25 and a plurality of output transmission lines T.sub.26 -T.sub.29 and T.sub.30 -T.sub.34. Here transmission lines T.sub.21 -T.sub.34 are microstrip transmission lines. The field effect transistors FET 11 to FET 14 have input electrodes, here gate electrodes G.sub.11 to G.sub.14, respectively, successively electrically interconnected via the first plurality of transmission lines T.sub.21 -T.sub.25. The output electrodes, here drain electrodes D.sub.11 -D.sub.14 are successively electrically interconnected via transmission lines T.sub.26 -T.sub.29 and common output lines T.sub.30 -T.sub.34. Source electrodes S.sub.11 to S.sub.14 of transistors FET 11 to FET 14, respectively, are connected to a reference potential, here ground, through a common RF and DC path, as shown. The gate electrode of the first one of the field effect transistors, here gate electrode G.sub.11 of FET 11 is coupled to input terminal 82 through transmission line T.sub.21, whereas the drain electrode D.sub.11 of the first field effect transistor FET 11 is coupled to an RF output terminal 83 through transmission lines T.sub.26 and T.sub.30. A succeeding one of the field effect transistors, here the last or fourth field effect transistor FET 14 has an input or gate electrode G.sub.14 coupled to a gate DC bias circuit 92 through transmission line T.sub.25, as shown, and has the output or drain electrode D.sub.4 of such field effect transistor FET 4 coupled via transmission lines T.sub.29 and T.sub.34 to the output or drain bias circuit 90, as shown.

The distributed frequency multiplier circuit 80 is shown to further include a second plurality of here four field effect transistors FET 15-FET 18 successively coupled between an input terminal 82' and the output terminal 83 via a second plurality of two port phase shift elements such as an artificial line or a distributed line. Here again non-linear elements are provided as shown in FIGS. 1, 8 and as shown for FET 11, as described above. Here the two port elements comprise a plurality of distributed input transmission lines T.sub.21 ' to T.sub.25 ', and a second plurality of two port elements or drain transmission lines T.sub.26 ' to T.sub.29 ' and the common output drain transmission lines T.sub.30 to T.sub.34. The field effect transistors FET 15-FET 18 have input electrodes, here gate electrodes G.sub.15 -G.sub.18, respectively, successively electrically interconnected via the second plurality of transmission lines T.sub.21 ' to T.sub.25 ', here each line being a microstrip transmission line. The output electrodes, here drain electrodes D.sub.15 -D.sub.18 of field effect transistors FET 15-FET 18, respectively, are successively electrically interconnected via the drain electrode transmission lines T.sub.26 ' to T.sub.29 ' and the common transmission lines T.sub.30 ', T.sub.34 '. The source electrodes S.sub.15 -S.sub.18 of transistors FET 15-FET 18, respectively, are connected to a reference potential, here ground, through a common DC and RF path, as shown. The gate electrode of the first one of the second set of field effect transistors, here gate electrode G.sub.15 of FET 15 is coupled to the input terminal 82', via transmission line T.sub.21 ', whereas the drain electrode D.sub.15 of the first field effect transistor FET 15 of the second set of field effect transistors is coupled to the output terminal 83 via transmission line T.sub.26 ' and T.sub.30. A succeeding one of the field effect transistors, here the last or fourth field effect transistor, here FET 18 of the second set of field effect transistors has an input or gate electrode G.sub.18 coupled to a gate DC bias circuit 92' via transmission line T.sub.25 ', as shown, and the output or the drain electrode D.sub.18 of such field effect transistor FET 18 is coupled to the output or drain bias circuit 90 via transmission lines T.sub.29 ' and T.sub.34, as shown.

The drain bias circuit 90 is here a ladder network having three shunt paths to ground via capacitors C.sub.13, C.sub.14, C.sub.15 with transmission lines, here microstrip transmission lines T.sub.36, T.sub.37 and T.sub.38 providing serial elements of such network. A pair of bias terminals 91a, 91b are adapted for coupling to a grounded DC bias source V.sub.DD, such terminal 21a being connected to capacitor C.sub.15 and transmission line section T.sub.38, as shown. Capacitors C.sub.13, C.sub.14 and C.sub.15 provide a relatively low impedance path to radio frequency (RF) signals and thus shunt such radio frequency signals to ground to prevent such RF signals from being coupled to the DC bias source V.sub.DD. A resistor R.sub.12 is shown coupled in shunt with ground and the connection between transmission line T.sub.36 and transmission line section T.sub.34. Resistor R.sub.2 in combination with the composite impedance of transmission line sections T.sub.36, T.sub.37 and T.sub.38 and capacitors C.sub.13, C.sub.14 and C.sub.15 provide a complex impedance for matched termination of transmission line section T.sub.34.

Gate bias circuits 92 and 92' are substantially identical and therefore the description of gate bias circuit 92 is likewise applicable to gate bias circuit 92'. The gate bias circuit 92 is also a ladder network and includes a serially connected resistor R.sub.11 and transmission line section T.sub.35. Transmission line section T.sub.35 and resistor R.sub.11 provide a direct current path between input bias terminal 26 and the microwave transmission lines T.sub.21 -T.sub.25. Radio frequency bypass capacitors C.sub.11 and C.sub.12 are coupled in shunt between the ends of transmission line section T.sub.35 and ground. Again, such radio frequency bypass capacitors C.sub.11, C.sub.12 provide low impedance paths to radio frequency signals shunting such radio frequency signals to ground and thereby isolate such radio frequency signals from a voltage source V.sub.GG which is coupled between ground terminal 23b and terminal 23a.

The characteristics of drain and gate bias circuits are selected to place each of the field effect transistors in the non-linear or square law region of their respective transfer characteristics.

A signal is fed from a source V.sub.rf to a directional coupler 86 which provides at its output terminals (not numbered) a pair of input signals V.sub.S, V.sub.S ' having a 180.degree. differential phase shift. Alternatively, the input signal V.sub.rf may be fed to a common junction including a pair of transmission lines having a differential pathlength corresponding to a phase shift of 180.degree.. Thus, the first signal V.sub.S is coupled to input terminal 82 and propagates along transmission lines T.sub.21 to T.sub.25. A portion of signal V.sub.S is coupled to each one of the gate electrodes G.sub.11 to G.sub.14 of FET 11-FET 14. At the output of the corresponding drain electrodes D.sub.11 to D.sub.14, an output signal appears having frequency components equal to .omega..sub.o, 2.omega..sub.o, 3.omega..sub.o, . . . n.omega..sub.o where n is an integer greater than 1 and .omega..sub.o is the fundamental frequency of the radio frequency input source V.sub.rf, and where 2.omega..sub.o, 3.omega..sub.o, . . . n.omega..sub.o are the even and odd harmonics of the fundamental frequency .omega..sub.o. Similarly, the second signal V.sub.S ' is fed to the input terminal 82' and such signal propagates along transmission lines T.sub.21 ' to T.sub.25 ' and a portion of such signal is coupled to gate electrodes G.sub.15 to G.sub.18. In response, an output signal is produced at each of the drain electrodes D.sub.15 to D.sub.18, having frequency components .omega..sub.o, 2.omega..sub.o, 3.omega..sub.o, . . . n.omega..sub.o, again where n is equal to an integer greater than 1 and .omega..sub.o is the fundamental frequency of the radio frequency source V.sub.rf. Signals from drain electrodes D.sub.11 to D.sub.14 and D.sub.15 to D.sub.18 are coupled to radio frequency transmission lines T.sub.30 -T.sub.34 at corresponding junctions 86a-86d. The same electrical pathlengths are provided between each one of said input terminals 82, 82' and output terminal 83 through respective ones of each of said pairs of field effect transistors (i.e., FET 11, FET 15; FET 12, FET 16; FET 13, FET 17; and FET 14, FET 18). Since the input signals V.sub.S, V.sub.S ' are fed to terminals 82 and 82' having a 180.degree. differential phase shift, signals coupled from respective pairs of drain electrodes D.sub.11 and D.sub.15, D.sub.12 and D.sub.16, D.sub.13 and D.sub.17, D.sub.14 and D.sub.18 will have the fundamental frequency components having the 180.degree. differential phase shift, and the odd harmonics of the fundamental will likewise have a 180.degree. differential phase shift. Therefore, the initial phase relationship of 180.degree. provided by the coupler 30 is maintained in each frequency component of the signal having an odd multiple harmonic relationship with the fundamental. With this arrangement, all the odd harmonics (2n+1).omega..sub.o of the fundamental frequency component of the signal including the fundamental component coupled from the output terminal 83 will cancel at terminal 83. On the other hand, since the devices are half-wave vth, more preferably, square law, non-linear devices, all the even harmonics (2n).omega..sub.o will have a phase difference corresponding to multiples of 0.degree. and therefore all of the even harmonics will be summed in-phase and will appear at the output terminal 83, as a composite signal V.sub.O.

At the output terminal 83 the composite signal is provided having even harmonic frequency components. A selected frequency component may be recovered from such signal in one of several ways. One approach is to provide the output coupling lines T.sub.26 -T.sub.29, T.sub.26 '-T.sub.29 ' and T.sub.30 -T.sub.34 with a bandpass filter characteristic permitting only the selected even harmonic to propagate therethrough. An alternative way is to provide a lumped filter arrangement at the output of the radio frequency terminal 83 to pass the desired even harmonic.

Preferably, in order to provide relatively broadband performance, that is, operation over a relatively broad range of input signal frequencies, the characteristic impedance of each one of the transmission lines T.sub.21 -T.sub.25, T.sub.21 '-T.sub.25 ' is selected in accordance with the inherent input reactance between gate and source of the field effect transistors FET 11-FET 14 and FET 15-FET 18 to provide an input network having a predetermined characteristic impedance. In a similar manner, the characteristic impedance of the output transmission lines T.sub.26 -T.sub.29, T.sub.26 '-T.sub.29 ' and T.sub.30 -T.sub.34 are selected in accordance with the inherent output reactance between drain and source of such field effect transistors FET 11-FET 14 and FET 15-FET 18 to provide an output network having a predetermined output impedance. Such an arrangement is described in the U.S. Pat. No. 4,456,888 issued June 26, 1984, and assigned to the same assignee as the present invention.

Referring now to FIG. 13, an alternate embodiment for a non-linear device, here a diode such as Schottky diodes 96, 96', are shown. The Schottky diode would be used to replace the field effect transistors (FET 11, FET 15) as illustratively shown in FIG. 13. The diode being a non-linear device would provide similar signal cancellation by proper phasing of the odd harmonics of the output signals and signal enhancement of the even harmonics, as described above.

Referring now to FIG. 14, a further alternate embodiment of a frequency conversion circuit, here a mixer 110, is shown to include the first plurality of here four non-linear elements 15a-15d, here each one of said elements including one of the active elements described in conjunction with FIGS. 9-11 or the dual-gate field effect transistor shown in FIG. 15. Here said non-linear elements are, for example, half wave vth law devices, such as a half wave square law device. Each one of said non-linear elements 15a-15d is coupled to a corresponding one of a plurality of output phase shifting elements 117a-117d. Each one of said phase shifting elements 117a-117d includes respective pairs of any generalized two port elements 117a'-117d' and 117a"-117d", here each one of said two port phase shifting elements may include an artificial line or a distributed transmission line. Alternatively, the pairs of two port elements may be active devices such as an amplifier or other non-reciprocal two port elements. Here one of said two port distributed lines has a common connection which is coupled to respective output ports 15a.sub.3 -15d.sub.3 of respective ones of the non-linear elements 15a-15d and has a pair of input ports connected to one of a pair of output terminals 19a, 19b of the circuit. The circuit 110 is shown to further include an input coupling means 116, here comprising a plurality of transmission lines 116a-116d, each one of said lines having a predetermined electrical pathlength, to provide a differential pathlength or phase shift between adjacent successive pairs of said lines 116a-116d. An input signal V.sub.rf is fed to an input terminal 113 of the circuit and said signal is distributed to each one of said transmission lines 116a-116d from a common connection with terminal 113. The outputs of said transmission lines 116a-116d are fed preferably through one of a corresponding plurality of capacitors C.sub.12 -C.sub.15 with the output of said capacitors feeding input terminals 15a.sub.1 -15d.sub.1 of the respective non-linear elements 15a-15d. A second input signal, here a local oscillator signal, is coupled to a second input terminal 118a, and said signal is fed to a local oscillator coupling means 118. Local oscillator coupling means 118 here comprises a plurality of here four, two port phase shifting elements 118b-118e, here said elements comprising a distributed transmission line. The phase shifts .theta..sub.LO1 -.theta..sub.LO4 provided by said distributed transmission lines 118b-118e in a first embodiment of the local oscillator means 118 have substantially equal pathlengths and in a second embodiment of the means, the electrical pathlengths .theta..sub.LO1 -.theta..sub.LO4 have successively increasing electrical pathlengths. The outputs of said local oscillator coupling means feed a portion of the local oscillator signal to corresponding ones of said second input terminals 15a.sub.2 -15d.sub.2 of respective non-linear elements 15a-15d.

In operation, input signals fed to terminals 113 and 118 are coupled via the input coupling means 116 and local oscillator coupling means 118 to respective input terminals 15a.sub.1 -15d.sub.1 and 15a.sub.2 -15d.sub.2 of the non-linear elements 15a-15d and produce at the outputs of said elements 15a.sub.3 -15d.sub.3 signals having frequency components equal to the sum and difference of the input signals, as well as, harmonics of the input signals. The signals propagate through one of the pair of two port phase shifting elements of each one of the output coupling means to provide at a first one of the output terminals 119a, 119b, an output signal having one of the aforementioned frequency components and at the second one of the output terminals 119a, 119b, a null or attenuated output signal in accordance with the relative frequency of the pair of input signals. Radio frequency circuit 110 is a non-distributed, non-successive interconnected version of the radio frequency conversion circuit shown in conjunction with FIGS. 1, 7 and 12. By adjusting the electrical pathlengths of the two port phase shifting elements comprising the output coupling means 117, input coupling means 116, and local oscillator coupling means 118, any one of the aforementioned radio frequency conversion circuits, the mixer 10 (FIG. 1), the mixer 60 (FIG. 8) and the multiplier 80 (FIG. 12) may be fabricated in a non-distributed version using two port phase shifting elements.

Having described preferred embodiments of this invention, it will now be apparent to one of skill in the art that other embodiments incorporating its concept may be used. It is felt, therefore, that this invention should not be restricted to the disclosed embodiment, but rather should be limited only by the spirit and scope of the appended claims.

Claims

1. A radio frequency circuit having first and second input terminals fed by first and second input signals having first and second frequencies and first and second output terminals comprising:

(i) a plurality of nonlinear elements, each having an input electrode and an output electrode;

(ii) input means, fed by the first input signal having a first frequency, comprising at least one two port phase shift element connected to the first input terminal of the circuit and the input electrode of each non-linear element, for providing a phase shift difference between portions of said signal coupled to the input electrodes of adjacent, successive ones of said nonlinear elements;

(iii) means, having a first end coupled to the second one of the input terminals of the circuit for feeding a second input signal having the second frequency to each non-linear element; and

(iv) output means, connected between the pair of output terminals of the circuit, for successively electrically connecting the output electrodes of each nonlinear element;

wherein in response to said first and second input signals, output signals are provided from said output electrodes and said signals propagate in first and second directions along said output means providing first and second output signal portions to provide a composite output signal at one of said terminals having a frequency component equal to at least a first one of the sum of and difference between of the frequencies of said first and second input signals.

2. The circuit of claim 1 wherein the means for feeding the second signal comprises means for providing substantially identical electrical pathlengths between the second input terminal and each one of the plurality of nonlinear elements.

3. The circuit of claim 2 wherein the non-linear elements each comprise a dual-gate field effect transistor.

4. The circuit of claim 3 wherein the composite output signal is provided at a first output terminal when the frequency of the first signal is higher than that of the second signal, and said composite output signal is provided at the second output terminal when the frequency of the first signal is lower than that of the second signal.

5. The circuit of claim 4 further comprising:

means for selectively tailoring as a function of frequency the amplitude and phase characteristics of each of the output signal portions propagating along the output means.

6. The circuit of claim 5 wherein the signal tailoring means comprises at least one capacitor connected between the input means and a first input electrode of one of the plurality of field effect transistors.

7. The circuit of claim 6 wherein the input means further comprises:

a first plurality of distributed transmission lines, each having a predetermined electrical pathlength;

the output coupling means further comprises:

a second plurality of distributed transmission lines, each having a predetermined electrical pathlength; and

wherein the predetermined electrical pathlengths of the first and second pluralities of distributed transmission lines are selected to impart predetermined phase shift characteristics to said first and second output signal portions provided from each one of the field effect transistors, to provide from each of the first output signal portions at first one of the first and second output terminals of the circuit, the composite output signal having a frequency equal to a first one of a sum of the frequencies of the first and second signals and difference between the frequencies of the first and second signals, and for providing from each of the second output signal portions at a second one of the first and second output terminals said null or attenuated signal.

8. The circuit of claim 3 further comprising:

means for selectively tailoring as a function of frequency the amplitude and phase characteristics of each of the signal portions propagating along the output means.

9. The circuit of claim 8 wherein the signal tailoring means comprises at least one capacitor connected between the input means and the input electrode of each one of the plurality of nonlinear elements.

10. The circuit as recited in claim 1 wherein the nonlinear elements include a second input electrode fed by the second input signal; and

wherein the means for feeding the second input signal comprises means for successively electrically interconnecting each one of said second input electrodes to said second input terminal, and for providing successively increasing electrical pathlengths between said input terminal and each successive one of said second input electrodes.

11. The circuit of claim 10 wherein the nonlinear elements each comprise a dual-gate field effect transistor.

12. The circuit of claim 11 wherein the composite output signal is provided at the first output terminal when the frequency of the first signal is higher than that of the second signal, and said composite output signal is provided at the second output terminal when the frequency of the first signal is lower than that of the second signal.

13. The circuit of claim 12 further comprising:

means for selectively tailoring as a function of frequency the amplitude and phase characteristics of each of the signal portions propagating along the output means.

14. The circuit of claim 13 wherein the signal tailoring means further comprises at least one capacitor connected between the input means and a first input electrode of one of the plurality of field effect transistors.

15. The circuit of claim 14 wherein the input means further comprises:

a first plurality of distributed transmission lines, each having a predetermined electrical pathlength;

the output means further comprises:

a second plurality of distributed transmission lines, each having a predetermined electrical pathlength; and

wherein the predetermined electrical pathlengths of the first and second pluralities of distributed transmission lines are selected to impart predetermined phase shift characteristics to said first and second output signal portions provided from each one of the field effect transistors to provide from each of the first output signal portions at a first one of the first and second output terminals of the circuit, the composite output signal having a frequency equal to a first one of a sum of the frequencies of the first and second input signals and difference between the frequencies of the first and second input signals, and for providing from each of the second output signal portions at a second one of the first and second output terminals said null or attenuated signal.

16. The circuit of claim 10 further comprising:

means for selectively tailoring as a function of frequency the amplitude and phase characteristics of each of the output signal portions propagating along the output means.

17. A radio frequency circuit having a pair of input terminals, each fed by one of a pair of input signals and a pair of output terminals comprising:

means, including a plurality of transistors, each having a pair of input electrodes and an output electrode, responsive to the pair of input signals for providing a corresponding plurality of output signals having a frequency component related to the frequency of each input signal;

means for providing a composite output signal responsive to the pair of input signals, which propagates towards one of the pair of output terminals in accordance with the difference in frequency between the pair of signals comprising:

(i) input means having a first end connected to a first one of the input terminals of the circuit for coupling successive portions of the input signal to a first one of the input electrodes of each transistor, said successive portions having a progressively increasing phase shift;

(ii) means connected to the second input terminal of the circuit for coupling a portion of the second input signal to a second one of the pair of input electrodes of each transistor; and

(iii) means, fed by each of said corresponding plurality of output signals, for providing the composite output signal.

18. A radio frequency circuit having a pair of input terminals, each fed by one of a pair of signals, and a pair of output terminals, comprising:

means, including a plurality of nonlinear elements, each having a pair of input electrodes and an output electrode for providing, in response to the pair of input signals, a corresponding plurality of output signals, each output signal having a frequency component related to the frequency of each of the pair of input signals, and means further comprising:

(i) input means, having a first end connected to a first one of the pair of terminals of the circuit, for distributively coupling a portion of the first input signal fed to said terminal to a first one of the pair of input electrodes of each non-linear element;

(ii) means connected to a second one of the pair of input terminals of the circuit, for coupling a portion of the second input signal fed to said second terminal of the circuit to a second one of the pair of input electrodes of each nonlinear element; and

(iii) output means, connected between the pair of output terminals of the circuit, for coupling the plurality of output signals from the non-linear elements to provide at a first one of the pair of output terminals of the circuit a composite output signal having a frequency component related to the frequency of each one of the input signals, with a second one of the pair of terminals having a null signal.

19. The circuit of claim 18 wherein said input means comprises:

a distributed transmission line connected to the input terminal and a plurality of capacitors each one disposed between the input electrode of each respective nonlinear element and a portion of the distributed transmission line to tailor the amplitude and phase characteristics to the input signals feed to each of the first input electrodes;

said means for coupling the second signal provides equal electrical pathlengths to each of said second signal portions fed to the second one of the pair of input electrodes of each nonlinear element; and

said output means comprises a distributed transmission line connected between the pair of output terminals and a plurality of capacitors connected between said line and a reference potential to increase the electrical pathlength of said distributed transmission line in accordance with the frequency of said signals coupled to the output means.

20. The circuit as recited in claim 19 wherein the phase and amplitude characteristics of said signals fed to the first input electrodes and the phase and amplitude characteristics of the output means are selected to provide at the first output terminal the input signal when the frequency of the first input signal is higher than the frequency of the second input signal, with a null or attenuated signal at the first output terminal and to provide at the second output terminal a signal when the frequency of the first input signal is lower than the frequency of the second input signal with a null or attenuated signal at the first output terminal.

21. The circuit of claim 20 wherein the nonlinear elements are dual gate field effect transistors and the input electrode are the gate electrodes of said transistors.

22. The circuit of claim 18 wherein said input means comprises a distributed transmission line connected to the input terminal and a plurality of capacitors each one disposed between the input electrode of each respective nonlinear element and a portion of the distributed transmission line to tailor the amplitude and phase characteristics to the input signals fed to the first input electrodes,

said means for coupling the second signal comprises a distributed transmission line fed by the second signal, which feeds each of said second input signal portions to said respective nonlinear elements having successively increasing electrical pathlengths to tailor the phase characteristics to the signals fed to the second input electrodes,

said output means comprises a distributed transmission line connected between the pair of output terminals and a plurality of capacitors connected between said line and reference potential to increase the electrical pathlength of said distributed transmission line in accordance with the frequency of said signals coupled to the output means.

23. The circuit of claim 22 wherein a first nonlinear element is fed the first input signal with successively nonlinear elements to a last one of said nonlinear elements having progressively increasing phase shifts and the last nonlinear element is fed the second input signal with successively nonlinear elements to said first nonlinear element having progressively increasing phase shifts.

24. The circuit of claim 23 wherein the nonlinear elements are dual gate field effect transistors and the input electrodes are the gate electrodes of said transistors.

25. A radio frequency mixer circuit having a first input terminal fed by an input radio frequency signal having a variable frequency f.sub.O, a second input terminal fed by a local oscillator signal having a known frequency f.sub.LO and first and second output terminals comprising;

means for providing at a first one of said first and second output terminals a first output signal related to a first one of the sum of the frequencies and difference between the frequencies of said input signal and local oscillator signal, when the frequency of the local oscillator signal is greater than the frequency of the input signal and a second one of said first and second output terminals a first one of the sum of the frequency and difference between the frequencies of said input signal and local oscillator signals when the signal of the local oscillator signal is less than the frequency of the input signal said means, further comprising:

a plurality of nonlinear devices, each device having first and second input electrodes and an output electrode;

input means, fed by the input signal comprised of at least one two port phase shifting element connected to the first input terminal of the circuit, and comprising means for coupling portions of said input signal to a first input electrode each one of the nonlinear devices, for providing a successively increasing phase shift difference between said portions of said signal coupled to successively ones of said input nonlinear devices,

means having a first end connected to the second input terminal of the circuit for feeding the local oscillator signal to the second input electrodes of each nonlinear device; and

output means connected between the pair of output terminals of the circuit for successively, electrically connecting the output electrode of each nonlinear element.

26. The mixer circuit as recited in claim 25 wherein the means for providing first and second output signals further comprises means for selectively tailoring as a function of frequency the amplitude and phase characteristics of each of the output signals propagating along the output coupling means.

27. The radio frequency mixer circuit of claim 26 wherein each of the plurality of nonlinear devices includes a dual-gate field effect transistor.

28. The radio frequency mixer circuit of claim 27 wherein the means for selectively tailoring comprises a plurality of capacitors coupled between the input coupling means and a first input electrode of each one of the field effect transistors, with each of said capacitors having a capacitance selected to provide amplitude and phase characteristics to input signals coupled to said field effect transistors, to provide in response at the first one of the output terminals and the second one of the output terminals output signals in accordance with the frequency of the input signal relative to the frequency of the local oscillator signal.