Mitigation of Bandwidth Limitation in Wireless Doherty ... - IEEE Xplore

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 9, SEPTEMBER 2012

2875

Mitigation of Bandwidth Limitation in Wireless Doherty Amplifiers With Substantial Bandwidth Enhancement Using Digital Techniques Ramzi Darraji, Student Member, IEEE, Fadhel M. Ghannouchi, Fellow, IEEE, and Mohamed Helaoui, Member, IEEE

Abstract—This paper proposes a new method for extending the bandwidth of Doherty power amplifiers (PAs) in the digital domain. The bandwidth enhancement is achieved through a frequency-selective pre-compensation mechanism that is derived to prevent the efficiency degradation that naturally occurs as the frequency of operation deviates from the center frequency. A methodical analysis of the frequency response of the conventional Doherty PA and that of the proposed Doherty PA is carried out to point out the limitations of the former and demonstrate the capability of the latter in recovering the bandwidth. Over the frequency range spanning from 1.96 to 2.46 GHz, the measured drain efficiency at 6–7-dB output power back-off is higher than 40% for the proposed Doherty PA. Such efficiency performance is achievable only from 2.04 to 2.22 GHz using the conventional Doherty PA. Hence, the bandwidth is enhanced from 180 to 500 MHz, which corresponds to an increase by a factor of 2.8 (i.e., almost triple). By applying the proposed methodology, a Doherty PA that is originally designed at the center frequency of 2.14 GHz for downlink wideband code division multiple access became operative at 1.98-GHz uplink wideband code division multiple access (UL-WCDMA), 2.22-GHz long-term evolution (LTE), and 2.34-GHz worldwide interoperability for microwave access (WiMAX) bands. The average drain efficiencies for UL-WCDMA, LTE, and WiMAX applications, were 40.1%, 44.2%, and 41.4%, respectively, using the proposed Doherty PA, and 37%, 37.3%, and 35.2%, respectively, using the conventional Doherty PA. Index Terms—Bandwidth, digital Doherty power amplifier (PA), efficiency, frequency, load modulation, wideband operation.

I. INTRODUCTION

T

HE CONTINUOUS proliferation of wireless communication standards coupled with the incessant quest for higher data throughput has given rise to stringent challenges in the design of wireless transmitters. Indeed, emerging communication standards, such as wideband code division multiple access (WCDMA), worldwide interoperability for microwave access

Manuscript received December 13, 2011; revised April 14, 2012; accepted April 17, 2012. Date of publication August 02, 2012; date of current version August 28, 2012. This work was supported by Alberta Innovates Technology Futures (AITF), the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Space Agency (CSA), Focus Microwaves, Nanowave Technologies, and the Canada Research Chairs (CRC) Program. The authors are with the Intelligent RF Radio Technology Laboratory (iRadio Lab), Department of Electrical and Computer Engineering, Schulich School of Engineering, University of Calgary, Calgary, AB, Canada T2N 1N4 (e-mail: [email protected]; [email protected]; [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TMTT.2012.2207910

(WiMAX), and long-term evolution (LTE), employ spectrum efficient modulation techniques with more compact constellations to enhance the transmission rate and optimize the usage of the congested RF spectrum. To avoid the distortion of the transmitted information and preserve its integrity, the power amplifier (PA) of the wireless transmitter is forced to operate at large back-off from its saturation point, which degrades the power efficiency of the RF transmitter. Therefore, power amplification architecture with efficiency enhancement at back-off operation is essential [1]. Among others, the Doherty PA is currently considered as the favorite solution for efficiency enhancement in field-deployed wireless transmitters [2]–[21]. Simple circuit configuration, high back-off efficiency, and proven linearizibility are the main advantages of the Doherty technique [2]–[21]. A well-known weakness of the Doherty PA, however, is its narrow bandwidth, which compromises its convenience for multisandard/multiband applications [11]–[19]. In this context, several research efforts were focused in designing broadband Doherty PAs with improved efficiency over a wide frequency range [11]–[19]. This intend was achieved in an analog way through a variety of solutions, such as a varactor-based impedance transformer [15], quasi-lumped transmission-line impedance inverter [16], ladder-type multisection output network [17], stepped-impedance inverter [18], and Doherty PA design via a real frequency technique [19]. This paper proposes a new technique for enhancing the bandwidth of Doherty PAs in a digital way. The Doherty PA proposed in this research allows surmounting the bandwidth restrictions imposed by the quarter-wavelength transformers of the output combining network and the frequency-dependent behavior of the active devices. This is achieved by digitally controlling the input power distribution and the phase variation between the carrier and peaking PAs. At first, the bandwidth extension capability of the proposed Doherty PA is demonstrated from a theoretical perspective. In particular, a set of generic equations is derived to illustrate its operation and assess its bandwidth. The experimental validation is then carried out based on the dual-input digital Doherty architecture [20], [21], which is adopted in this study to enable the bandwidth enhancement in the digital domain. Henceforth, the terms Doherty PA and analog Doherty PA will be used interchangeably to designate a conventional Doherty PA that employs an analog input splitter, identical devices for the carrier and peaking cells, and an output Doherty comequal bining network with an impedance inversion factor

0018-9480/$31.00 © 2012 IEEE

2876


to 0.5. The impedance inversion factor is the design parameter that controls the load seen by the carrier amplifier at low-input power drive, which is equal to , where is the optimal load of the carrier and peaking cells. The term digital Doherty PA, on the other hand, indicates a Doherty PA in which the input signal splitting is performed in digital domain (i.e., the carrier and peaking amplifiers are driven separately). This paper is organized as follows. In Section II, the frequency response of the digital Doherty PA is analyzed, and its bandwidth performance benchmarked against that of the analog Doherty PA. In Section III, the bandwidth enhancement due to the proposed digital Doherty PA is experimentally verified. Conclusions are drawn in Section IV. II. BANDWIDTH EXTENSION ANALYSIS OF THE DIGITAL DOHERTY PA To overcome the problem of bandwidth restrictions imposed by the quarter-wavelength transformers of the Doherty combining network and the frequency-dependent load modulation behavior, the digital Doherty concept [20], [21] is adopted in this research. Precisely, the proposed digital Doherty PA employs a pre-compensation mechanism that acts on the input power distribution and the phase variation between the carrier and peaking amplifiers to compensate for the frequency-selective behavior of the Doherty PA. The pre-compensation function will be defined based on the enhanced current factors and , which are used herein to model the variation of the output current of the carrier cell and that of the peaking cell, respectively, in response to a variation in the injected input power, at a given frequency . In what follows, the expressions of and versus are derived based on the operation of the Doherty PA with frequency-dependent components, and the effectiveness of the proposed digital Doherty PA in recovering the bandwidth performance is theoretically demonstrated. The analysis is carried out considering the flowing assumptions. 1) The carrier and peaking amplifiers are biased at class-B and class-C modes, respectively. 2) Each current source is linearly proportional to the input drive voltage , as shown in Fig. 1, and terminated with perfect harmonic short circuits so that the efficiency and the output power can be assessed using only the fundamental and dc components. 3) The carrier and peaking cells exhibit an ideal bandwidth performance, i.e., the current sources generate the same fundamental and dc components at any , and the turn-on point of the peaking cell remains equal to at all frequencies ( is the maximal input drive voltage). A. Frequency Response Analysis at the Low-Power Region , only the At the low-power region carrier cell is active. As such, the equivalent circuit diagram of the Doherty amplifier reduces to the carrier source operating into the impedance , as shown in Fig. 2. is resulting from the impedance transformation of the output load by means of two transmission lines having an electrical length of 90 at the

Fig. 1. Fundamental currents of carrier and peaking cells of the Doherty PA.

Fig. 2. Doherty PA operational circuit diagram at low-power region.

center frequency and characteristic impedances of and , respectively. Based on Fig. 2, the output impedance of the carrier cell at the low-power region is given by (1) where (2) (3) and is the impedance seen at the input of the transmission line that is connected to . Fig. 3(a) shows the variation of the real part of versus . It can be seen that decreases rapidly from the optimal value of (at as the shift from increases. The delivered RF power of the Doherty PA at the lowpower region is related to by (4) is the fundamental RF current of the carrier cell at the where low-power region.

DARRAJI et al.: MITIGATION OF BANDWIDTH LIMITATION IN WIRELESS DOHERTY AMPLIFIERS

2877

and expressing the fundamental RF current of the carrier cell of the digital Doherty PA as (6) then using (5) and (6), the delivered RF power of the digital Doherty PA is given by

(7) Fig. 3. Real part of

Therefore, in theory, the digital Doherty PA with nonideal quarter-wavelength transformers can mimic the operation of a Doherty PA with an ideal output combining network as it can deliver for . Referring to [21, eq. (15)], such a performance is achieved provided that the available input power of the Doherty PA is distributed as follows:

versus frequency.

(8) (9)

Fig. 4. Normalized output power of the analog Doherty PA at low-power region versus frequency.

Referring to (4), it is clear that the drop of induces a significant loss of output power, and consequently, restricted efficiency improvement at back-off for . For instance, Fig. 4 shows that the output power degrades by 14%, 26%, and 38% (with respect to the maximal level at ) when deviates by , and , respectively. The normalized output power reported in Fig. 4 is obtained based on the ratio , which is evaluated using (1)–(4) and found to be equal to . Ideally, a perfect Doherty combining network transforms to at all frequencies. In this case, is identically equal to . Introducing the enhanced current factor of the digital Doherty PA as (5)

and denote the input powers injected into the where carrier and peaking branches, respectively. Fig. 5(a) depicts the variation of versus . Note that increases from 1 at (i.e., ) to at (i.e., ). Likewise, Fig. 5(b) shows that increases from at (i.e., ) to at (i.e., ). The uneven drive splitting with higher power into the carrier branch allows to compensate for the loss of RF output power due to the drop of , which results in improving the efficiency of the Doherty PA for . By using (4), the efficiency of the Doherty PA at the back-off region can be calculated as (10) is the dissipated dc power of the Doherty PA at where the low-power region, which is given by the product of the dc current of the carrier cell and the drain supply voltage . Similarly, by using (7) and (10), the efficiency of the digital Doherty PA at the back-off region can be expressed as

(11)

2878


Fig. 6. Normalized efficiency of the analog and digital Doherty PAs at back-off versus frequency.

Fig. 7. Doherty PA operational circuit diagram at high-power region.

Fig. 5. (a) Variation of versus frequency. (b) Input power distribution between carrier and peaking cells versus frequency.

where is the dissipated dc power of the digital Doherty PA at the low-power region, which is given by the product of and . From (11), the normalized efficiency of the digital Doherty PA at back-off can be written as

Fig. 6. depicts the variations of and versus , which were computed using (12) and (13), respectively. In theory, the analog Doherty PA can achieve a bandwidth of 16% at the cost of 10% efficiency degradation. The digital Doherty PA, on the other hand, achieves a 24% bandwidth for the same efficiency degradation, which corresponds to a relative improvement of 50%. B. Frequency Response Analysis at the High-Power Region At the high-power region , the carrier and peaking cells are both active. As depicted in Fig. 7, the equivalent circuit diagram of the Doherty PA consists of the carrier and peaking sources, which are operating into the common load . The fundamental RF currents of the carrier and peaking cells of the Doherty PA at peak power are given by (14) (15)

(12) is recognized as the normalized efficiency of the Dowhere herty PA, which can be evaluated from (1)–(3) and (10) as (13)

and designate the RF where currents of the carrier and peaking cells at peak power, respectively (see Fig. 1). The lower fundamental RF current of the peaking cell together with the frequency-selective behavior of the Doherty combining network impairs the load modulation mechanism and degrades the efficiency. Introducing the enhanced current


factor required to ensure the ideal load modulation behavior in the digital Doherty PA as

2879

Thus, using (14), (15), and (24), the enhanced current factor of the digital Doherty PA is explicitly obtained from (23) as

(16) where is the fundamental RF current of the peaking cell of the digital Doherty PAs, then the purpose of the following is to derive an explicit expression of versus in order to evaluate the performance of the digital Doherty PA and compare it with that of the analog Doherty PA. Starting from the -matrix of the quarter-wavelength transformer with the characteristic impedance and referring to Fig. 7, one can write

(17)

(18) Also, referring to Fig. 7, the following expressions arise: (19) (20) Substituting

(25) is recognized as the enhanced current ratio where of the uneven and the asymmetrical Doherty PAs [8], [9], which is given by the ratio of the peak power RF current of the carrier cell to the peak power RF current of the peaking cell at . It is noteworthy that design parameter gives a good indication on the bias point of the peaking cell. Indeed, the higher , the deeper the class-C bias of the peaking cell. Conversely, biasing the peaking cell at class-B results in . Fig. 8 shows the variation of the magnitude and phase of versus . It can be noted that decreases from at to at . Therefore, the uneven drive power into the peaking path is relaxed as deviates from . In addition, the phase response of the peaking cell should be adjusted for in order to ensure that (23) is strictly fulfilled at any . The real RF power of the digital Doherty PA at full drive is given by

in (19) with its expression (20) results in (21)

is given by (2) where Rearranging (18) for an expression of for (21) gives

(26)

while accounting By using (24) and (25),

can be written as (27)

(22) where To guarantee the ideal load modulation behavior in the digital Doherty PA, the peak power impedance seen by the carrier cell and that seen by the peaking cell then has to be equal to at all frequencies. This intent is achieved by satisfying the necessary and sufficient condition given by

(28) From (27) and (28), the normalized output power of the digital Doherty PA can be explicitly expressed as

(23) where

is deduced from (22) as

(24)

(29)

2880


Fig. 9. Normalized peak output power of the analog and digital Doherty PAs.

Fig. 8. Variation of

Fig. 10. Normalized efficiency of the analog and digital Doherty PAs at peak power versus frequency.

versus frequency. (a) Magnitude. (b) Phase.

On the other hand, the real RF power of the Doherty PA at peak power is given by

From (32) and (33), the normalized output power of the Doherty PA can be explicitly expressed as

(30) Considering that and (25) as

and

can be related by using (14), (15),

(31) (34) then based on (22) (30) and (31),

can be expressed as (32)

where (33)

and versus , which Fig. 9 depicts the variations of were calculated based on (29) and (34), respectively. The results confirm that the digital Doherty PA permits significant boost of output RF power for different from compared to the analog Doherty PA. In addition, it can be noted that the output power degradation in the analog Doherty PA is more severe for deep class-C operation of its peaking cell.


2881

Fig. 11. Block diagram of the proposed digital Doherty PA.

By using (27), the efficiency of the digital Doherty PA at peak power can be written as

(35) where and are the peak power dc currents of the carrier and peaking cells, respectively. is the dissipated dc power of the digital Doherty PA at peak power, which is given by the product . Assuming that at peak power can be approximated by (36) then by using (25), (35), and (36),

is expressed as (37)

From (28) and (37), the normalized peak power efficiency of the digital Doherty PA is explicitly given by

(38)

Fig. 12. Variation of the enhanced current factors and digital Doherty PA.

for the implemented

the digital Doherty PA is capable of maintaining the maximal efficiency for . This result is expected because the digital amplifier is designed to ensure that the carrier and peaking cells are operating into at all frequencies. III. EXPERIMENTAL VALIDATION

Besides, from (32), the efficiency of the Doherty PA peak power can be expressed as

at

(39) is the dissipated dc power of the Doherty PA at peak where power, which is given by the product . By using (33) and (39), the normalized peak power efficiency of the Doherty PA is then explicitly expressed as (40) and versus , which Fig. 10 shows the variations of were obtained using (38) and (40), respectively. The results show that, regardless of the bias point of the peaking device,

In Section II, it was theoretically demonstrated that the digital Doherty PA allows for significantly improved bandwidth performance compared to the analog Doherty PA. In this section, the bandwidth-extension ability of the digital Doherty PA is experimentally established. The experiments are carried out based on a relatively narrowband Doherty PA prototype that is nominally designed to operate at 2.14 GHz for down-link WCDMA band. The analog Doherty PA is implemented using a 10-W packaged gallium–nitride (GaN) device (CGH40010) from Cree Inc., Durham, NC. The carrier and peaking cells are biased in class-AB (quiescent current, mA, drain voltage V) and class-C ( mA, V) conditions, respectively. The implementation procedure of the dual-input digital Doherty PA based on the single-input analog PA is performed following by

2882


Fig. 13. Measured efficiencies of the analog and digital Doherty PAs at 6–7-dB output power back-off and at peak power.

Fig. 15. Measured gain and drain efficiency of the analog and digital Doherty PAs at 1.96, 2.22, and 2.34 GHz.

analog Doherty PA prototype exhibits acceptable sub-optimal performance in terms of efficiency, output power, and gain. A. Continuous-Wave (CW) Measurement Results

Fig. 14. Measured SSG and

of the analog and digital Doherty PAs.

eliminating input analog splitter and isolating the input paths of the dual-input PA. The block diagram of the proof-of concept experimental setup used in this research is reported in Fig. 11. The baseband streams and are tailored at the digital domain according to the carrier frequency of the input signal and then downloaded into two synchronized arbitrary waveform generators (AWGs) (ESG-4438C from Agilent Technologies, Palo Alto, CA). The experimental validation is conducted over a frequency range of 500 MHz (i.e., GHz GHz) where the

Fig. 12 depicts the variations of the enhanced current factors and of the implemented digital Doherty PA. It can be noted that follows the theoretical tendency that is reported in Fig. 5(a). Indeed, the overdrive power into the carrier branch at back-off is increased gradually when the deviation from the design frequency 2.14 GHz increases in order to compensate for the RF power loss due to the frequency response of the carrier branch. On the other hand, due to the nonideal bandwidth behavior of the carrier and peaking cells, the analytically derived function response of cannot be applied as it is. The phase disparity [related to the phase of through (25)] between the carrier and peaking paths and the overdrive power into the peaking branch at high-power region [related to the magnitude of through (25)] were derived experimentally at each frequency using CW measurements to maximize the peak power performance. For instance, referring to Fig. 12, the optimal value of at 2.04 GHz is found to be equal to . This means that the best performance at the high-power region was achieved when first, the input signal of the peaking PA is lagging behind that of the carrier PA by


2883

TABLE I MEASUREMENT SUMMARY OF THE ANALOG AND DIGITAL DOHERTY PAs

Fig. 16. Output spectra of the analog and digital Doherty PAs. (a) Using WCDMA at 1.98 GHz. (b) Using LTE at 2.22 GHz. (c) Using WiMAX at 2.34 GHz.

[as inferred from (16)], and second, the peaking PA is driven with dB more input power compared to the carrier PA (as inferred from [21, eq. (19)]). Fig. 13 shows the measured efficiencies of the analog and digital Doherty PAs at 6–7-dB back-off from the maximal output power and at peak drive. It is clear that the digital Doherty amplifier enables important bandwidth improvement. Indeed, the proposed PA maintains higher that 40% efficiency at back-off across the frequency range from 1.96 to 2.46 GHz, whereas the efficiency is greater than 40% only between 2.04–2.22 GHz for the analog Doherty PA. As such, the digital Doherty improves the bandwidth to 500 MHz compared to 180 MHz for the analog Doherty, which is an increase of approximately 180% in bandwidth. The performance in terms of small-signal gain (SSG) and saturated output power of the analog and digital Doherty PAs is reported in Fig. 14. The digital Doherty PA has higher

SSG and due to the digital control of the power distribution and the phase variation between the carrier and peaking branches. It is noteworthy that the slightly fluctuating frequency response observed for peak power performance of the analog Doherty PA is mainly attributed to the dynamic phase and amplitude variations due to the nonideal bandwidth behavior of the carrier and peaking devices, as well as the frequency response of the input and output matching networks. To further explore the performance of the Doherty PAs, Fig. 15 depicts the measured gain and efficiency characteristics of the analog and digital Doherty PAs at several frequencies within the operation band. B. Modulated Signal Measurement Results To verify the appropriateness of the proposed digital Doherty amplifier for multiband/multistandard applications and evaluate its linearity performance, WCDMA, LTE, and WiMAX signals

2884


were employed. The peak-to-average power ratios (PAPRs) of the signals are 7.04, 7.18, and 7.31 dB, respectively. When driven with a 1.98-GHz WCDMA signal, the proposed digital Doherty PA achieved a drain efficiency of 40.1% at 7-dB back-off compared to 37% for the analog Doherty PA. Besides, using a 2.22-GHz LTE signal, a drain efficiency of 44.2% is measured for the digital Doherty PA against 37.3% for the analog Doherty PA. At 2.34 GHz, the digital Doherty PA exhibited an efficiency of 41.4% at 7-dB back-off using a WiMAX signal. The efficiency of the analog Doherty PA, on the other hand, is limited to 35.2% under identical drive conditions. The measurement results for the modulated signal tests are summarized in Table I. The measured output spectra of analog and digital Doherty PAs for the modulated signal tests at 1.98, 2.22, and 2.34 GHz are reported in Fig. 16. It can be deduced that the digital Doherty PA has an acceptable linearity, which can be enhanced to meet the spectrum mask specifications in each band by using the digital predistortion (DPD) technique [22]–[24] based on the procedure reported in [21]. Since the generation of the specific input streams for the carrier and peaking branches according to the proposed technique is, in principle, not resource-intensive compared to the DPD algorithm, the proposed bandwidth enhancement mechanism does not result in a significant increase in the computational load, and thus, it can be implemented along with the DPD algorithm in the digital signal processing (DSP) block without compromising the performance of the linearizer. IV. CONCLUSION In this paper, a digital technique has been proposed to enhance the bandwidth of Doherty PAs though the mitigation of the limiting effects of the quarter-wavelength transformers of the Doherty combiner and the frequency-dependent behavior of the carrier and peaking amplifiers. Applied to a 2.14-GHz Doherty PA with an initial bandwidth of 180 MHz, the proposed technique permitted to raise the bandwidth to 500 MHz. Besides, the average efficiency is enhanced from 37%, 37.3%, and 35.2% to 40.1%, 44.2%, and 41.4% for 1.98-GHz WCDMA, 2.22-GHz LTE, and 2.34-GHz WiMAX signals, respectively. These results confirm the convenience of the proposed digital Doherty PA for multistandard applications within the frequency band of wireless communication standards. It is worth stating that better bandwidth performance can be obtained if the proposed digital technique is associated with a well-designed broadband Doherty PA in which the matching networks are broadband, the quasi-open-circuit requirement on the output impedance of the peaking amplifier is fulfilled over a large frequency band, and the fluctuations in the frequency response of the analog Doherty PA are less prominent. All these benefits allow applying the proposed digital techniques over a larger frequency range. The processing speed requirements for wider bandwidth (0.5–1 GHz), however, are challenging in both DPD and the proposed bandwidth-extension techniques. Sub-band processing techniques [24] can be adopted to alleviate the processing speed requirement in such conditions. In view of that, the complexity increase due to the additional transmitting path and the supplementary power consumption of the DSP block of the digital Doherty transmitter can be totally justified

because a single broadband digital Doherty PA would be able to cope with several communication standards. This will permit to avoid assigning a specific hardware to a given standard, and thus, reduce the running costs of base stations and optimize their operation. ACKNOWLEDGMENT The authors would like to acknowledge M. Akbarpour and A. Kwan, both with the University of Calgary, Calgary, AB, Canada, for their input during the measurements. The authors would also like to acknowledge the support of Agilent Technologies, Palo Alto, CA. REFERENCES [1] F. H. Raab, P. Asbeck, S. Cripps, P. B. Kenington, Z. B. Popovic, N. Pothecary, J. F. Sevic, and N. O. Sokal, “Power amplifiers and transmitters for RF and microwave,” IEEE Trans. Microw. Theory Tech., vol. 53, no. 3, pp. 814–826, Mar. 2002. [2] M. Iwamoto, A. Williams, P.-F. Chen, A. G. Metzger, L. E. Larson, and P. M. Asbeck, “An extended Doherty amplifier with high efficiency over a wide power range,” IEEE Trans. Microw. Theory Tech., vol. 49, no. 12, pp. 2472–2479, Dec. 2000. [3] M. Nick and A. Mortazawi, “Adaptive input-power distribution in Doherty power amplifiers for linearity and efficiency enhancement,” IEEE Trans. Microw. Theory Tech., vol. 58, no. 11, pp. 2764–2771, Nov. 2010. [4] C. Steinbeiser, T. Landon, C. Suckling, J. Nelson, J. Delaney, J. Hitt, L. Witkowski, G. Burgin, R. Hajji, and O. Krutko, “250 W HVHBT Doherty with 57% WCDMA efficiency linearized to 55 dBc for 2c11 6.5 dB PAR,” IEEE J. Solid-State Circuits, vol. 43, no. 10, pp. 2218–2228, Oct. 2008. [5] P. Colantonio, F. Giannini, R. Giofrè, and L. Piazzon, “Theory and experimental results of a class F AB-C Doherty power amplifier,” IEEE Trans. Microw. Theory Tech., vol. 57, no. 8, pp. 1936–1947, Aug. 2009. [6] J. Moon, J. Kim, J. Kim, I. Kim, and B. Kim, “Efficiency enhancement of Doherty amplifier through mitigation of the knee voltage effect,” IEEE Trans. Microw. Theory Tech., vol. 59, no. 1, pp. 143–152, Jan. 2011. [7] W. C. E. Neo, J. Qureshi, M. J. Pelk, J. R. Gajadharsing, and L. C. N. de Vreede, “A mixed-signal approach towards linear and efficient -way Doherty amplifiers,” IEEE Trans. Microw. Theory Tech., vol. 55, no. 5, pp. 866–879, May 2007. [8] J. Kim, J. Cha, I. Kim, and B. Kim, “Optimum operation of asymmetrical cells-based linear Doherty power amplifier uneven power drive and power matching,” IEEE Trans. Microw. Theory Tech., vol. 53, no. 5, pp. 1802–1809, May 2005. [9] J. Kim, B. Fehri, S. Boumaiza, and J. Wood, “Power efficiency and linearity enhancement using optimized asymmetrical Doherty power amplifiers,” IEEE Trans. Microw. Theory Tech., vol. 59, no. 1, pp. 425–434, Jan. 2011. [10] S. C. Jung, O. Hammi, and F. M. Ghannouchi, “Design optimization and DPD linearization of GaN-based unsymmetrical Doherty power amplifiers for 3 G multicarrier applications,” IEEE Trans. Microw. Theory Tech., vol. 57, no. 9, pp. 2105–2113, Sep. 2009. [11] J. Moon, J. Kim, I. Kim, J. Kim, and B. Kim, “A wideband envelope tracking Doherty amplifier for WiMAX systems,” IEEE Microw. Wireless Compon. Lett., vol. 18, no. 1, pp. 49–51, Jan. 2008. [12] D. Kang, D. Kim, J. Moon, and B. Kim, “Broadband HBT Doherty power amplifiers for handset applications,” IEEE Trans. Microw. Theory Tech., vol. 58, no. 12, pp. 4031–4039, Dec. 2010. [13] D. Kang, D. Kim, and B. Kim, “Design of bandwidth-enhanced Doherty power amplifiers for handset applications,” IEEE Trans. Microw. Theory Tech., vol. 59, no. 12, pp. 3474–3438, Dec. 2011. [14] K. Bathich, A. Z. Markos, and G. Boeck, “A wideband GaN Doherty amplifier with 35% fractional bandwidth,” in Proc. 40th Eur. Microw. Conf., Paris, France, Sep. 2010, pp. 1006–1009. [15] M. Sarkeshi, O. B. Leong, and A. van Roermund, “A novel Doherty power amplifier for enhanced load modulation and higher bandwidth,” in IEEE MTT-S Int. Microw. Symp. Dig., 2008, pp. 733–766. [16] J. Qureshi, N. Li, W. Neo, F. van Rijs, I. Blednov, and L. de Vreede, “A wideband 20 W LMOS Doherty power amplifier,” in IEEE MTT-S Int. Microw. Symp. Dig., May 2010, pp. 1504–1507.


[17] K. Bathich, A. Z. Markos, and G. Boeck, “Frequency response analysis and bandwidth extension of the Doherty amplifier,” IEEE Trans. Microw. Theory Tech., vol. 59, no. 4, pp. 934–944, Apr. 2011. [18] L. Cen, T. Liu, Y. Ye, G. Xu, and Y. Zhao, “Optimization design of wideband asymmetric Doherty power amplifiers,” in WiCON, Wuhan, China, Sep. 2011, pp. 1–4. [19] G. Sun and R. H. Jansen, “Broadband Doherty power amplifier via real frequency technique,” IEEE Trans. Microw. Theory Tech., vol. 60, no. 1, pp. 99–111, Jan. 2012. [20] R. Darraji, F. M. Ghannouchi, and O. Hammi, “A dual-input digitally driven Doherty amplifier architecture for performance enhancement of Doherty transmitters,” IEEE Trans. Microw. Theory Tech., vol. 59, no. 5, pp. 1284–1293, May 2011. [21] R. Darraji and F. M. Ghannouchi, “Digital Doherty amplifier with enhanced efficiency and extended range,” IEEE Trans. Microw. Theory Tech., vol. 59, no. 11, pp. 2898–2909, Nov. 2011. [22] O. Hammi, S. Carichner, B. Vassilakis, and F. M. Ghannouchi, “Synergetic crest factor reduction and baseband digital predistortion for adaptive 3G Doherty power amplifier linearizer design,” IEEE Trans. Microw. Theory Tech., vol. 56, no. 11, pp. 2602–2608, Nov. 2008. [23] F. M. Ghannouchi and O. Hammi, “Behavioral modeling and predistortion,” IEEE Microw. Mag., vol. 10, no. 7, pp. 52–64, Dec. 2009. [24] O. Hammi, S. Boumaiza, M. Jaidane-Saidane, and F. M. Ghannouchi, “Digital subband filtering predistorter architecture for wireless transmitters,” IEEE Trans. Microw. Theory Tech., vol. 53, no. 5, pp. 1643–1652, May 2005. Ramzi Darraji (S’10) received the B.Eng. and M.Sc. degrees in communications engineering from the École Supérieure des Communications de Tunis, Ariana, Tunisia, in 2007 and 2008, respectively, and is currently working toward the Ph.D. degree at the University of Calgary, Calgary, AB, Canada. He is currently with the Intelligent RF Radio Technology Laboratory (iRadio Lab), University of Calgary. He has one patent pending. His research interests include DSP for wireless transmitters, high-efficiency broadband RF PAs and intelligent digital transmitters for ultra-wideband and multistandard wireless communications. Mr. Darraji was the recipient of the Alberta Innovates Technology Futures Doctoral Scholarship in information and communications technologies. He was the recipient of the IEEE Microwave Theory and Techniques Society (IEEE MTT-S) High Achievement Award of the 2010 Student High Efficiency Power Amplifier Design Competition. He was also the recipient of the First Place and Best Design Awards of the 2010 Wireless Innovations Forum’s Smart Radio Challenge.

2885

Fadhel M. Ghannouchi (S’84–M’88–SM’93–F’07) is currently a Professor, Alberta Innovates Technology Futures/Canada Research Chair, and Director of the Intelligent RF Radio Technology Laboratory (iRadio Lab), Electrical and Computer Engineering Department, Schulich School of Engineering, University of Calgary, Calgary, AB, Canada. He has held several invited positions with several academic and research institutions in Europe, North America, and Japan. He has provided consulting services to a number of microwave and wireless communications companies. He has authored or coauthored over 500 publications. He has authored and coauthored three books. He holds 12 U.S. patents with five pending. His research interests are in the areas of microwave instrumentation and measurements, nonlinear modeling of microwave devices and communications systems, design of power- and spectrum-efficient microwave amplification systems, and design of intelligent RF transceivers for wireless and satellite communications. Prof. Ghannouchi is a Fellow of the Institution of Engineering and Technology (IET). He is a Distinguished Microwave Lecturer of the IEEE Microwave Theory and Techniques Society (IEEE MTT-S).

Mohamed Helaoui (S’06–M’09) received the M.Sc. degree in communications and information technology from the École Supérieure des Communications de Tunis, Tunis, Tunisia, in 2003, and the Ph.D. degree in electrical engineering from the University of Calgary, Calgary, AB, Canada, in 2008. He is currently an Assistant Professor with the Department of Electrical and Computer Engineering, University of Calgary. He has authored or coauthored over 60 publications. He has seven patents pending. His current research interests include DSP, power efficiency enhancement for wireless transmitters, switching-mode PAs, and advanced transceiver design for software-defined radio and millimeter-wave applications. Dr. Helaoui is a member of the COMMTTAP Chapter, IEEE Southern Alberta Section.