Doherty功率放大器英文.docx

1、Doherty功率放大器英文The Doherty Power AmplifierThe current wireless communication systems have made significant progress toward increased bandwidth and number of carriers for high-data-rate applications. Memory effects, however, make it very difficult to design a high-power amplifier with a wide instantan

2、eous bandwidth. In addition to bandwidth concerns, the instantaneous transmit powers of the wireless communication systems, such as CDMA-2000, wide-band code division multiple access (WCDMA), orthogonal frequency division multiplexity (OFDM) and so on, vary widely and rapidly, carrying high peak-to-

3、average ratio (PAR) signals. The base station power amplifiers for the systems require a high linearity to amplify the high PAR signal source without distortion. To satisfy linearity requirements, the power amplifiers are usually biased at class A or AB mode and must operate at a large amount of bac

4、k-off from the peak output power. Another requirement of the base station power amplifier for the modern wireless communication systems is high efficiency. As the communication systems are reduced in both size and cost, the cooling system should be simple and small, requiring a power amplifier with

5、high efficiency. Because the base station power amplifiers have a low efficiency due to the back-off operation, efficiency enhancement techniques become very important. The design technique of the base station power amplifiers with high efficiency and linearity across a wide instantaneous bandwidth

6、has become a hot issue.In this article, we show that the Doherty amplifier is capable of delivering the stringent requirements of the base station power amplifiers. We explain the operation principles, including both linearity and efficiency improvements, and the basic circuit configuration of the a

7、mplifier. Advanced design methods to operate across wide bandwidth and improve the linearity are also described. For verification, the Doherty amplifier is implemented using laterally diffused metal oxide semiconductor (LDMOS) transistors and measured using a WCDMA 4FA signal. These results show tha

8、t the Doherty amplifier is a promising candidate for base station power amplifiers with wide bandwidth, high efficiency, and linearity.Doherty Amplifier OperationFigure1. (a) Operational diagram of the Doherty amplifier. (b) Fundamental currents. (c) Load impedances.The Doherty amplifier was first p

9、roposed by W.H. Doherty in 1936. The original Doherty amplifier consisted of two tube amplifiers and an impedance inverting network. The efficiency of an RF power amplifier is increased using the RF Doherty amplifier technique, as described in detail in . This amplifier consisted of a carrier amplif

10、ier and a peaking amplifier. The output load is connected to the carrier amplifier through an impedance inverter (a quarter-wave transmission line) and directly to the peaking amplifier. Figure 1(a) shows an operational diagram to analyze the Doherty amplifier circuit. Two current sources represent

11、the amplifiers. It is assumed that each current source is linearly proportional to the input voltage signal, operating as a class AB or class B amplifier with harmonic short circuits after it is turned on, and the efficiency analysis can be carried out using the fundamental and dc components only. A

12、s shown in Figure 1(b), the peaking amplifier turns on at one-half the maximum input voltage.The Doherty amplifier technique is based on the load impedance change of each amplifier, referred to as load modulation, according to the input power level. Figure 1(b) shows the fundamental currents from th

13、e amplifiers. The load impedances of two amplifiers are given bywhere ZL is the load impedance of the Doherty amplifier; IC and IP represent the fundamental currents of the carrier and peaking amplifiers, respectively; and ZC and ZP are the output load impedances of the carrier and peaking amplifier

14、s, respectively, and are depicted in Figure 1(c).In the low-power region (0 Vin, max/2), the peaking amplifier remains in the cut-off state, and the load impedance of the carrier amplifier is two times larger than that of the conventional amplifier. Thus, the carrier amplifier reaches the saturation

15、 state at the input voltage(Vin, max)/2 since the maximum fundamental current swing is half and the maximum voltage swing reaches Vdc. As a result, the maximum power level is half of the carrier amplifiers allowable power level (a quarter of the total maximum power or 6 dB down from the total maximu

16、m power), and the efficiency of the amplifier is equal to the maximum efficiency of the carrier amplifier as shown in Figure 2.In the high-power region (Vin, max/2 Vin, max), where the peaking amplifier is conducting, the current level of the peaking amplifier plays an important role in determining

17、the load modulation of the Doherty amplifier see (1) and (2). Assuming that gm of the peaking amplifier is twice as large as that of the carrier amplifier, the current and voltage swings of the peaking amplifier increase in proportion to the input voltage level and the voltage swing reaches the maxi

18、mum voltage swing of Vdc only at the maximum input voltage. The load impedance of the carrier amplifier varies from 2Zopt to Zopt, and the peaking amplifier varies from to Zopt according to the input voltage level as shown in Figure 1(c). The efficiency of the Doherty amplifier at the maximum input

19、voltage is equal to the maximum efficiency of the amplifiers. When the peaking amplifier is the same size as the carrier amplifier, which is normally the case, gm of the two amplifiers are identical and the peaking amplifier can not be fully turned on, so the power performance is degraded 4. From th

20、e basic operation principle, we have explored the Doherty amplifier, which provides higher efficiency over whole power ranges compared to the conventional class AB amplifiers. The resulting Doherty amplifier can solve the problem of maintaining a high efficiency for a large PAR signal.Linearity of t

21、he Doherty AmplifierThe linearity of the Doherty amplifier is more complicated than that of a class AB amplifier. The class AB biased carrier amplifier has a load impedance at the low power level that is twice as large and the high impedance of the carrier amplifier compensates the low gain characte

22、ristic due to the input power division. At high power levels, the two amplifiers generate full power using normal load impedances, equalizing the power gain. Additionally, in the low-power region, the linearity of the amplifier is entirely determined by the carrier amplifier. Therefore, the carrier

23、amplifier should be highly linear even though the load impedance is highAt a high power level, linearity of the amplifier is improved by the harmonic cancellation from the two amplifiers using appropriate gate biases. Figure 3 shows the third-order harmonic generation coefficient gm3 of an LDMOS tra

24、nsistor and the bias points of the two amplifiers. In terms of gain characteristics of each amplifier, a late gain expansion of the class C biased peaking amplifier compensates the gain compression of the class AB carrier amplifier. Thus, the Doherty amplifier, which is based on the load modulation

25、technique, is capable of delivering more linear output power than a conventional class AB power amplifier. The third-order intermodulation (IM3) level from the carrier amplifier increases and the phase of the IM3 decreases because the gain of the carrier amplifier is compressed. In contrast, when th

26、e gain of the peaking amplifier is expanded, both the IM3 level and phase increase. To cancel out the IM3s from the two amplifiers, the components must be 180 out of phase with the same amplitudes. Therefore, the peaking amplifier should be designed appropriately to cancel the harmonics of the carri

27、er amplifier.The Circuit Configuration of Doherty AmplifierFigure 4 shows a schematic diagram of the fully matched microwave Doherty amplifier with offset transmission lines at the output circuits 5. The carrier and peaking amplifiers have input/output matching circuits, which transform from the inp

28、ut impedances of the devices to 50 and from the optimum load impedance Zopt of the devices to 50 , respectively. The additional offset transmission lines with characteristic impedance of 50 are connected after the matching circuits of the carrier and peaking amplifiers. In the low-power region, the

29、phase adjustments of the offset lines cause the peaking amplifier to be open-circuited and the characteristic load impedance of the carrier amplifier is doubled to 2Ro by a quarterwave impedance transformer. This is illustrated in Figure 5(a) and (b). The offset line of the carrier amplifier varies

30、from Zopt to 2Zopt for the proper load modulation as shown in Figure 5(a). Figure 5(b) illustrates that the offset line of the peaking amplifier adjusts to the high impedance so that it prevents power leakage. Figure 5(c) shows the appropriate transformations on a Smith chart to determine the offset

31、 line length of each amplifier. The lines do not affect the overall matching condition and load modulation because they are matched to the characteristic impedance of 50 . The Doherty output combining circuit consists of a quarter-wave transmission line with the characteristic impedance of 50 and a

32、quarter-wave transmission line that transforms from 50 to 25 to determine the load impedance of the output combining circuit. A phase delay line is needed at the input of the peaking amplifier to adjust the same delay between the carrier and peaking amplifiers 6.The Doherty amplifier consists of a c

33、lass AB biased carrier amplifier and a class C biased peaking amplifier. Due to the different biasing, the RF current from the amplifiers are different depending on the input drive level. The asymmetric powers are combined by the Doherty operation through a quarter-wave impedance converter.Advanced Design Methods for the Doherty AmplifierThe