Feed reflected Doherty amplifier and method for driving Doherty amplifiers

Abstract

The feed reflected Doherty amplifier utilizes the output characteristics of the carrier amplifier to control the input signal of the peaking amplifier to improve the gain, linearity and efficiency of a Doherty amplifier. The feed reflected Doherty amplifier comprises an input power splitter, a carrier amplifier branch and a peaking amplifier branch combined into a common load, an output directional coupler and an input directional coupler connected via a phase shift element.

Claims

1. A feed reflected Doherty amplifier comprising; at least one input power splitter which divides the input signal to produce a first split signal and a second split signal, at least one carrier amplifier amplifying the first split signal to generate the carrier output signal and having a reflection coefficient at its output which remains constant below a threshold point and decreases depending on the load current driven by the peaking amplifier above the said threshold point, at least one impedance inverting network coupled to the output of the carrier amplifier, at least one output directional coupler coupled at the output path of the carrier amplifier to have a reflected coupled output signal of the carrier amplifier at its isolated port, which increases in accordance with the increasing carrier amplifier output signal below the said threshold point and decreases with the decreasing output reflection coefficient of the carrier amplifier above the said threshold point, at least one phase shift element connected to the isolated port of the output coupler to adjust the phase of the said reflected coupled output signal, at least one input directional coupler coupled between the input power splitter and the input path of the peak amplifier, and its isolated port is connected to the phase shift element to combine the said second split signal of the input power splitter and the said reflected coupled output signal of the carrier amplifier to drive the peaking amplifier with a dynamically controlled signal which remains in a substantially lowered amplitude below the said threshold point and increases with the decreasing output reflection coefficient of the carrier amplifier above the said threshold point, at least one peaking amplifier operating in class B which amplifies the combined signal at the output of the input directional coupler to generate a peaking output signal, at least one load.

2. The Feed-reflected Doherty amplifier according to claim 1 wherein the input port of the output directional coupler is connected to the carrier amplifier output path, the output port is connected to the load, and the isolated port is connected to the isolated port of the input coupler via the phase shift element.

3. The Feed-reflected Doherty amplifier according to claim 1 wherein the input port of the input directional coupler is connected to one output path of the input power splitter, the output port is connected to the input of the peaking amplifier, and the isolated port is connected to the isolated port of the output coupler via the phase shift element.

4. The Feed-reflected Doherty amplifier according to claim 1 wherein the output directional coupler produces a sampled reflected signal at the isolated port, which increases to a threshold level proportional to the said carrier amplifier output signal until the carrier amplifier reaches saturation, and which starts to decrease from the threshold level as the said peaking output signal increases.

5. The Feed-reflected Doherty amplifier according to claim 1 wherein the input directional coupler receives the said second split signal from the input power splitter and the said reflected coupled output signal from the output directional coupler via the phase shift element to generate a low level signal below the threshold point, and an increasing signal above the threshold point.

6. The Feed-reflected Doherty amplifier according to claim 1 wherein the coupling ratio of the output coupler and the phase value of the phase shift element are determined correspondingly to apply the said reflected coupled output signal with comparable magnitude and negative phase to the isolated port of the input coupler, wherein the combination of the said reflected coupled output signal with the said second split signal results in a substantial decrease in the input signal level of the peaking amplifier until the main amplifier reaches saturation.

7. The Feed-reflected Doherty amplifier according to claim 1 wherein the peaking amplifier configured to operate at class B is driven by a dynamically controlled input signal which level substantially lowered by the contribution of the said sampled reflected signal below a threshold level, and which level monotonically increases above the threshold level.

8. A method of driving a Doherty amplifier comprising the steps of, Dividing the input signal to produce a first split signal and a second split signal, Driving the carrier amplifier branch with the said first split signal to produce an output signal including forward and reflected output signals, Obtaining a divided part of the said reflected output signal, Adjusting the phase and magnitude of the obtained reflected output signal, Combining the second split signal and the adjusted reflected output signal to drive the peaking amplifier branch.

Description

BRIEF DESCRIPTION OF FIGURES

(1) A Feed Reflected Doherty amplifier and method for driving a Doherty amplifier realized to fulfill the objective of the present invention is illustrated in the accompanying figures, in which:

(2) FIG. 1 illustrates the block diagram of the conventional Doherty amplifier.

(3) FIG. 2 illustrates the block diagram of the bias-adapted Doherty amplifier.

(4) FIG. 3 illustrates the block diagram of the feed-forward linearization amplifier.

(5) FIG. 4 illustrates the block diagram of the feed reflected Doherty amplifier.

(6) FIG. 5 illustrates the schematic diagram of a directional coupler.

(7) FIG. 6 illustrates the voltage variation of the reflected signal at the output of the carrier amplifier 4 with respect to the voltage of the input signal.

(8) FIG. 7A and FIG. 7B illustrate the comparison of the conventional Doherty amplifier and the preferred embodiment in terms of the voltage waveforms of the carrier and peaking amplifiers.

(9) FIG. 8 is the flow chart of method for driving a Doherty amplifier

DETAILED DESCRIPTION OF THE INVENTION

(10) Elements shown in the figures are numbered as follows: 1. Feed reflected Doherty amplifier 2. Input power splitter 3. Input directional coupler 4. Carrier amplifier 5. Peaking amplifier 6. Output directional coupler 7. Impedance inverting network 8. Input offset transmission line 9. Phase shift element 10. Load

(11) Feed reflected Doherty amplifier 1 essentially comprises; at least one input power splitter 2 which divides an input signal to produce a first split signal and a second split signal, at least one carrier amplifier 4 for amplifying the first split signal to generate the carrier output signal, has a reflection coefficient at its output which remains constant below a threshold point and decreases depending on the load current driven by a peaking amplifier 5 above the said threshold point, at least one peaking amplifier 5 operating in class B which amplifies the combined signal at the output of the input directional coupler to generate the peaking output signal, at least one impedance inverting network 7 coupled to the output of the carrier amplifier 4, at least one output directional coupler 6 coupled at the output path of the carrier amplifier 4 to have a reflected coupled output signal of the carrier amplifier 4 at its isolated port, which increases in accordance with the increasing carrier amplifier 4 output signal below the said threshold point and decreases with the decreasing output reflection coefficient of the carrier amplifier 4 above the said threshold point, at least one phase shift element 9 connected to the isolated port of the output coupler 6 to adjust the phase of the said reflected coupled output signal, at least one input directional coupler 3 coupled between the input power splitter 2 and the input path of the peaking amplifier 5, and its isolated port is connected to the phase shift element 9 to combine the said second split signal of the input power splitter 2 and the said reflected coupled output signal of the carrier amplifier 4 to drive the peaking amplifier 5 with a dynamically controlled signal which remains in a substantially lowered amplitude below the said threshold point and increases with the decreasing output reflection coefficient of the carrier amplifier 4 above the said threshold point, at least one load 10,

(12) Before describing the details of the invention, it will be useful to give a brief explanation of the directional coupler. FIG. 5 shows a directional coupler with port numbers 1-4. When the input signal is applied to port 1; port 2 is the output port, port 3 is defined as the coupled port and port 4 is defined as the isolated port. C, coupling ratio is the amplitude ratio of the input signal and the coupled signal at the coupled port. Directivity is the amplitude ratio of the input signal to the reflected coupled signal at the isolated port. Coupled port has the coupled part of the forward signal with a coupling ratio (C), and the isolated port has the reflected coupled part of the reflected signal with the same coupling ratio (C).

(13) FIG. 4 illustrates the structure of the feed reflected Doherty amplifier 1 comprising a carrier amplifier 4 and a peaking amplifier 5, input power splitter 2, input directional coupler 3, output directional coupler 6, input offset transmission line 8, a phase shift element 9, and impedance inverting network 7. The carrier amplifier 4 branch and the peaking amplifier 5 branch are combined to a common load 10. The carrier amplifier 4 and the peaking amplifier 5 can be comprised of one or more stages. The input offset transmission line 8, the phase shift element 9, and the impedance inverting network 7 are phase shifting elements with a characteristic impedance of Z.sub.0. The input offset transmission line 8, the phase shift element 9, and the impedance inverting network 7 can be replaced by equivalent lumped-element components. The input directional coupler 3, output directional coupler 6 are the coupled micro-strip or strip-line transmission lines where the distance between the coupled lines and the length of the coupled lines determine the coupling ratio, C. The input directional coupler 3, output directional coupler 6 can also be replaced by equivalent lumped-element components.

(14) The input power splitter 2 receives an input RF (radio frequency) signal and divides it into a first split signal and a second input signal to drive the carrier amplifier 4 branch and the peaking amplifier 5 branch, respectively. The carrier amplifier 4 is fed directly from input power splitter 2; whereas the peaking amplifier 5 is connected via the input offset transmission line 3. The length of the input offset transmission line 3 is adjusted for phase equalization of the amplifier branches. The output of the peaking amplifier 5 is directly connected to the load 10. The output of the carrier amplifier 4 is connected to the impedance inverting network 7. The output directional coupler 6 can be connected either between the impedance inverting network 7 and the load 10, or between the carrier amplifier 4 and the impedance inverting network 7. The input directional coupler 3 is connected between the input offset transmission line 8 and the input of the peaking amplifier 5. The isolated ports of the input and the output directional couplers 3 and 6 are connected to each other via the phase shift element 9.

(15) The input signal is divided by the input power splitter 2 into a first split signal and a second split signal. The carrier amplifier 4 amplifies the first split signal to generate the carrier output signal. The carrier amplifier 4 has a reflection coefficient at its output as the load impedance seen by the carrier amplifier 4 is different from Z.sub.0, where Z.sub.0 is the matched impedance required for maximum output power and efficiency. The reflection coefficient is defined as the ratio of the amplitude of the reflected signal to the amplitude of the incident signal.

(16) The load 10 terminated at the combining node has an impedance value of R.sub.L=Z.sub.0/(1+) where is the periphery ratio of the carrier amplifier 4 and the peaking amplifier 5. For a symmetrical configuration where the carrier amplifier 4 and the peaking amplifier 5 have the same periphery, the load 10 impedance value is R.sub.L=Z.sub.0/2. And Z.sub.C the load impedance seen by the carrier amplifier 4 is expressed as Z.sub.C=R.sub.L(1+I.sub.P/I.sub.C), where I.sub.C is the output current of the carrier amplifier 4, and I.sub.P is the output current of the peaking amplifier 5. The reflection coefficient at the output of the carrier amplifier 4 is given by; .sub.C=(Z.sub.CZ.sub.0)/(Z.sub.C+Z.sub.0). For a symmetrical Doherty configuration where Z.sub.C=Z.sub.0/2(1+I.sub.P/I.sub.C), the reflection coefficient at the output of the carrier amplifier 4 can be written as;
.sub.C=(I.sub.PI.sub.C)/(I.sub.P+3I.sub.C)

(17) Below the threshold point wherein the peaking amplifier 5 is non-conductive and I.sub.P is zero; the reflection coefficient at the output of the carrier amplifier 4 has a constant value which is equal to for a symmetrical Doherty configuration. The threshold point is commonly preferred as the first peak efficiency point where the carrier amplifier 4 reaches saturation and the peaking amplifier 5 starts to conduct. Above the threshold point wherein the peaking amplifier 5 starts to conduct, the reflection coefficient at the output of the carrier amplifier 4 starts to decrease with the increasing peaking amplifier current I.sub.P and eventually converges to zero as h, is equal to I.sub.C at the maximum output power level. So, the carrier amplifier 4 has a reflection coefficient at its output which remains constant below a threshold point and decreases depending on the load current driven by the peaking amplifier 5 above the said threshold point. As the reflection coefficient is equal to the amplitude ratio of the reflected signal the incident signal, V.sub.R the reflected signal voltage at the output of the carrier amplifier 4 is expressed as V.sub.R=.sub.CV.sub.C where V.sub.C is the output voltage of the carrier amplifier 4. Below the threshold point wherein the reflection coefficient .sub.C is constant, the reflected signal at the output of the carrier amplifier 4 increases with the increasing carrier amplifier 4 output signal and decreases with the decreasing output reflection coefficient of the carrier amplifier .sub.C above the said threshold point.

(18) FIG. 6 illustrates the voltage variation of the reflected signal at the output of the carrier amplifier 4 with respect to the voltage of the input signal. The reflected signal at the output of the carrier amplifier 4 reaches a saturation level at the threshold point and starts to decrease above the threshold point. At the maximum input level, the reflected signal V.sub.R converges to zero as the reflection coefficient .sub.C decreased to zero in ideal conditions.

(19) The input port of the output directional coupler 6 is connected to the carrier amplifier 4 output path, the output port is connected to the load 10, and the isolated port is connected to the isolated port of the input directional coupler 3 via the phase shift element 9. The output directional coupler 6 samples the incident and the reflected signals at the output path of the carrier amplifier 4 with a coupling ratio of C.sub.O. Therefore, the reflected coupled signal at the isolated port of the output directional coupler 6 is a divided part of the reflected signal at the output of the carrier amplifier 4, which increases in accordance with the increasing carrier amplifier 4 output signal below the said threshold point and decreases with the decreasing output reflection coefficient of the carrier amplifier 4 above the said threshold point. The amplitude of the reflected coupled signal at the isolated port of the output directional coupler 6 is equal to C.sub.OV.sub.R. The phase shift element 9 coupled in series between the isolated ports of the input and the output directional couplers adjusts the phase of the reflected coupled signal to apply to the isolated port of the input directional coupler 3. The phase shift element 9 can be implemented by a micro-strip, strip-line transmission line or lumped-element equivalent. The electrical length of the transmission line determines the phase shift value of the phase shift element 9.

(20) The input port of the input directional coupler 3 is connected to one output path of the input power splitter 2, the output port is connected to the input of the peaking amplifier 5, and the isolated port is connected to the isolated port of the output directional coupler 6 via the phase shift element 9. The input directional coupler 3 combines the said second split signal of the input power splitter 2 with the said reflected coupled signal to generate a driving signal for the peaking amplifier 5. In the case of an equal input power division, the said first and second split signals can be considered as the half of the input signal of the Doherty amplifier. So, the said driving signal at the output of the input directional coupler 3 V.sub.IP can be expressed as;
V.sub.IP=V.sub.in/2+C.sub.IC.sub.OV.sub.R
where V.sub.in/2 is the second split signal and the C.sub.I is the coupling ratio of the input directional coupler 3. The reflected signal voltage at the output of the carrier amplifier 4 is also written as V.sub.R=.sub.CV.sub.C=.sub.CG.sub.C(V.sub.in/2) where G.sub.C is the gain of the carrier amplifier 4 and V.sub.in/2 is the first split signal. By substituting the resultant V.sub.R equation into the V.sub.IP equation, the output signal of the input directional coupler 3 V.sub.IP can be rewritten as;
V.sub.IP=(1+C.sub.IC.sub.O.sub.CG.sub.C)(V.sub.in/2)
Below the threshold point wherein the reflection coefficient .sub.C is equal to , the coupling ratios of the input and output couplers is adjusted to cancel the said second split signal and the said reflected coupled signal which results in a zero amplitude signal at the output of the input directional coupler 3. And above the threshold point, the output signal of the input directional coupler 3 V.sub.IP increases with the decreasing output reflection coefficient of the carrier amplifier 4 and converges to the maximum input level of the said second split signal. At the threshold point wherein the carrier amplifier 4 reaches saturation with gain compression; the output signal of the input directional coupler 3 V.sub.IP starts to increase with the decreasing gain of the carrier amplifier 4 G.sub.C, and activates the peaking amplifier 5 to conduct. So, the peaking amplifier 5 is self-activated by the saturation and gain compression of the carrier amplifier 4 which is commonly preferred for an optimum Doherty amplifier operation.

(21) The peaking amplifier 5 amplifies the combined signal at the output of the input directional coupler 3 V.sub.IP to generate peaking output signal. As the output signal of the input directional coupler 3 V.sub.IP is substantially lowered below the threshold point, the peaking amplifier 5 is allowed to operate at class B or AB bias conditions which improve the total gain and linearity of the Doherty amplifier. The output signals of the peaking amplifier 5 and the carrier amplifier 4 are combined to the common load 10 to generate the final output signal of the Doherty amplifier 1. The input signal of the peaking amplifier which is the output signal of the input directional coupler 3 and expressed as above equation;
V.sub.IP=(1+C.sub.IC.sub.O.sub.CG.sub.C)(V.sub.in/2),
is proportional to the gain and output reflection characteristics of the carrier amplifier. By adjusting the initial parameters such as the coupling ratios of the input and output directional couplers C.sub.I,C.sub.O, and the phase of the phase shift element 9, the preferred embodiment provides a flexibility to synthesize an inverse gain compression characteristics at the peaking amplifier 5 branch which improves the linearity of the Doherty amplifier 1 while maintaining the efficiency. FIG. 7 illustrates the comparison of the conventional Doherty amplifier and the preferred embodiment in terms of the voltage waveforms of the carrier and peaking amplifiers. In FIG. 7, the output voltage of the carrier amplifier V.sub.OC and the output voltage of the peaking amplifier V.sub.OP are illustrated with dashed lines, the combined output voltage of the Doherty amplifier V.sub.O is illustrated with solid line. FIG. 7.b illustrates the preferred embodiment with an improved gain and linearity performance.

(22) A method 100 of driving a Doherty amplifier said method 100 comprising the steps of; Dividing the input signal to produce a first split signal and a second split signal (101), Driving the carrier amplifier 4 branch with the said first split signal to produce an output signal including forward and reflected output signals (102), Obtaining a divided part of the said reflected output signal (103), Adjusting the phase and magnitude of the obtained reflected output signal (104), Combining the second split signal and the adjusted reflected output signal to drive the peaking amplifier branch (105).

(23) In the method 100, the input signal is divided into two; a first split signal and a second split signal in step 101. In step 102, the first signal is used for driving the carrier amplifier 4 branch of the Doherty amplifier. Being driven by the first signal, the carrier amplifier 4 produces a forward output signal and a reflected output signal. In step 103, the reflected signal at the output of the carrier amplifier 4 is sampled and a divided part of the said reflected output signal is obtained. In step 104, the phase and magnitude of the obtained reflected output signal are adjusted. In step 105, the adjusted reflected output signal and second split signal are combined for driving the peaking amplifier 5.