Owing to the high impedance transformation ratio, the Doherty power amplifier (DPA) with a large output power back-off generally has bandwidth limitations. This study proposes an asymmetric DPA with an impedance distribution and control circuit (IDCC) at the post-matching network to improve the bandwidth. The IDCC, based on a resonance circuit and a series transmission line, distributes and controls the load impedance according to the frequency so that the bandwidth of the DPA can be extended. To verify the proposed IDCC, an asymmetric DPA was designed using GaN HEMT with a power capacity of 6 W and 10 W for the carrier and peaking amplifiers, respectively. The implemented DPA was evaluated for the broad frequency band between 3.3 GHz and 3.8 GHz using a 5G new radio (NR) signal with a bandwidth of 100 MHz and a peak-to-average power ratio of 7.8 dB. A drain efficiency between 43.2% and 50.7% and an adjacent channel leakage power ratio between −23.4 dBc and −27.3 dBc were achieved at an average power level that ranged between 33.5 dBm and 34.3 dBm.

Keywords: Doherty Power Amplifier; GaN HEMT; Impedance Distribution and Control Circuit; 5G New Radio

I. Introduction

The modulation schemes of wireless communication systems are required to have high spectral efficiency for high data rate. As a result, the modulated signals essentially have a wide signal bandwidth and high peak-to-average power ratio (PAPR), which degrade the efficiency of the power amplifier due to the operation with a large output power back-off (OBO). Doherty power amplifiers (DPAs) have been widely adopted in base transceiver systems because of their simple structure and their high efficiency at an OBO of about 6 dB [1].

With the increasing PAPR of modulated signals, DPAs must have a larger OBO of even more than 6 dB. Some techniques have been reported to have an extended OBO level greater than 6 dB [2–8]. However, the bandwidth of DPAs with an extended OBO level is generally reduced, mainly due to the increased impedance transformation ratio. Furthermore, as the operating frequency increases, the increased influence of the drain-to-source capacitance C_ds limits the bandwidth. Only a few studies have thus obtained an OBO range greater than 6 dB with extended bandwidth, especially at the sub-6 5G frequency band of above 3 GHz [2–23].

Multi-way DPAs [7, 8] and multi-stage DPAs [9–11] have been proposed for extended OBO range. However, they have degraded power gain and increased circuit complexity. On the other hand, asymmetric DPAs have a simple two-way structure and an extended OBO range and use the higher peak output power of the peaking amplifier [4–6]. However, due to the increased impedance transformation ratio, compared to the conventional, it becomes more difficult to secure the required bandwidth. Some previous studies utilized impedance distribution over the frequency using the additional resonant circuits at the load network of the peaking amplifier or at the post-matching network for the DPAs to gain increased bandwidth [3, 12–17]. The broadband post-matching structure was adopted to extend the bandwidth using a multi-section matching network [12–14]. Kang et al. [3] optimized the bandwidth by adjusting the distribution of peaking impedance at a low power level using a multiple resonance circuit. Xia et al. [15] also utilized the reactance compensation of the peaking impedance for bandwidth enhancement, while not affecting the load modulation of the DPA. Fang et al. [16] used a multi-section post-matching network, which was designed to appropriately distribute impedance according to frequency. However, these techniques increased the circuit complexity by using additional multiple resonant circuits or multi-section matching networks.

In this study, to extend the bandwidth of an asymmetric DPA and gain an extended OBO range, an impedance distribution and control circuit (IDCC), composed of a shunt resonance circuit and a series transmission line, was proposed. The IDCC was applied to the post-matching network of the DPA and used to appropriately distribute the load impedance at the combining node according to the frequency.

A shunt resonance circuit distributes the load impedance according to the frequency; then, the series transmission line rotates the load impedance on the Smith chart. The component values of the resonance circuit and the electrical length of the series transmission line can be adjusted to optimize the bandwidth. Compared to previous methods [3–16], the proposed IDCC is simple and straightforward with two independent parameters to distribute the impedance using the resonance circuit. Moreover, the distributed impedance over frequency can be rotated using a series transmission line for broadband impedance matching. This method can be independently applied after designing other circuits to further improve bandwidth. The proposed DPA was designed and implemented for the frequency band between 3.3 GHz and 3.8 GHz. For verification, the measured results using continuous-wave (CW) and 5G NR signals are presented and compared to those without IDCC.

II. Impedance Distribution and Control Circuit

Fig. 1 shows an example of the matching network that matches from Z_L to R_opt including the proposed IDCC. The IDCC can be applied after a certain matching network to simply make the overall circuit achieve a broader bandwidth. A shunt or series resonance circuit is located in the middle of a quarter-wave transmission line (QWTL). Z_L represents the load impedance, while the primed load impedance, ZL′, is distributed by the resonance circuit and rotated by the offset line before the resonance circuit using the proposed IDCC. The characteristic impedance of the transmission line Z₀ and the resonance frequency ω₀ are defined as follows:

Fig. 1

A circuit diagram of the matching network including the proposed IDCC.

(1) Z0=RLRL′,

(2) ω0=1LC,

where, Z_L and ZL′ are R_L and RL′ at the resonance frequency, respectively. The combination of an inductor and a capacitor in the shunt resonance circuit can be replaced by a combination of open- and short-circuited stubs. The shunt resonance circuit could be replaced by a series circuit using an inductor and a capacitor as well. There are infinite combinations of the values of L and C to obtain a resonance at a certain frequency. The optimal impedance distribution level can be obtained by selecting the best combination of the values of L and C for the application. The electrical length of the transmission line θ₀ can be optimized to appropriately rotate the impedance on the Smith chart as well. The electrical length of the rest of the transmission line can be automatically determined, as the sum of the total electrical length of the two transmission lines at the left and the right of the resonance circuit should be 90° for impedance transformation from a real value to a real value. If R_L and RL′ are the same—i.e., there is no impedance transformation required at the resonance frequency—the remainder of the transmission line with an electrical length of 90°–θ₀ at the right side of Fig. 1 can be removed.

Fig. 2 shows the impedance trajectories over frequency for the circuit presented in Fig. 1. To match from Z_L to R_opt, a simple L-section matching network can be designed for the center frequency of ω₀. If the circuit is simple, we could have a large frequency variation for Z_out (as shown in Fig. 2(a)), which results in a large variation for Z_opt (as shown in Fig. 2(b)) over the frequency. Using the resonance circuit in the IDCC, a frequency distribution of as much as Z_out for the impedance ZL′ can be obtained. Then, using the offset line before the resonance circuit, the complex conjugate of the impedance ZL′ can be rotated to closely match to Z_out (see Fig. 2(a) for this operation). Since Zout* and ZL′ are matched in the broadband, Z_opt should be matched to R_opt in the broadband as well. As shown in Fig. 2(b), the trajectory of Z_opt with IDCC can have a knot for broadband impedance match, while Z_opt without IDCC has a monotonic spread as the matching network gives. As illustrated in the example, when using IDCC to appropriately control the impedance over the frequency range at the output impedance of an arbitrary matching network, it is possible to enhance matching within the target bandwidth.

Fig. 2

Impedance trajectories over frequency on the Smith chart: (a) Zout* and ZL′, (b) Z_opt with and without IDCC.

III. Design and Bandwidth Extension for the DPA

The load impedance of the DPA can be appropriately distributed to broaden the bandwidth [3, 12]. The IDCC was applied to the post-matching network of the DPA to simply broaden the bandwidth. An asymmetric DPA was designed for experimental verification using Cree’s CGH40006P (6 W GaN HEMT) for the carrier amplifier and CGH400010F (10 W GaN HEMT) for the peaking amplifier.

Fig. 3 shows the schematic of the load network using the IDCC in the post-matching network of the DPA. The optimum impedances at the internal plane of the transistors are presented as Z_C_,_low (135 Ω for this design), which is thrice the R_opt for an OBO of 9.5 dB for the low power level of the carrier amplifier, Z_C_,_peak (45 Ω for this design), which is R_opt for the peak power level of the carrier amplifier, and Z_P_,_peak (20 Ω for this design) for the peak power level of the peaking amplifier. Since ZC′ and ZP′ were selected as 150 Ω and 75 Ω for the peak power level to make the load impedance at the combining load as 50 Ω, the characteristic impedance of the series transmission line in the IDCC should be 50 Ω. The ratios of ZC′ and ZP′ should be inversely proportional to the power capacities of the carrier and peaking amplifiers, respectively.

Fig. 3

The simplified schematic of the proposed load network.

The carrier amplifier was designed using a quasi-lumped QWTL including the internal passive network and external matching network to transform the ZC,peak′ of 150 Ω to the Z_C_,_peak of 45 Ω at the center frequency and the peak power level. The peaking amplifier was matched to attain an equivalent electrical length of 180° including the internal network, a π-type matching network, and an offset line at the center frequency. The post-matching network included the IDCC, as shown. Between the combining node and the resonance circuit, a series transmission line with an angle of θ₀ was applied. After the resonance circuit, the remainder of the QWTL was removed, since there was no impedance transformation required for the post-matching network in this design. In other words, the characteristic impedance of the QWTL at the post-matching network was selected as 50 Ω. Table 1 presents four representative combinations of the shunt inductor and capacitor in the parallel resonance circuit that were selected to optimize the bandwidth. The selected four cases are examples among the possible combinations including the optimal case (Case 3), which can be found by conducting an elaborated search using simulations or experiments.

Table 1

Four combinations of the values of the shunt inductor and capacitor in the resonance circuit

Fig. 4 shows the simulated fractional bandwidths (FBWs) for the four cases according to the electrical length of the series transmission line. For each case, the FBWs were simulated for Z_C_,_low, Z_C_,_peak, and Z_P_,_peak using the 1.5 dB theoretical power contour as a reference of power levels at the center frequency. The dashed line represents an FBW before applying the IDCC. Among the infinite number of combinations in the resonance circuit and electrical lengths of the transmission line, Case 3, with an electrical length of 60°, was found to be the best case while considering all the simulated FBWs for Z_C_,_low, Z_C_,_peak, and Z_P_,_peak simultaneously. Compared to the cases with slightly different component values, the optimized impedances of Case 3 were plotted on the Smith chart with the 1.5 dB theoretical power contour (Fig. 5). For Case 3, the simulated FBWs were obtained as 17.2% for Z_C_,_low, 22.1% for Z_C_,_peak, and 17.6% for Z_P_,_peak, while they were 15.1%, 13.2%, and 12.4% without IDCC for Z_C_,_low, Z_C_,_peak, and Z_P_,_peak, respectively. Moreover, a slight performance degradation for Case 2 and Case 4 was found.

Fig. 4

The simulated FBWs for the four cases of the resonance circuit according to the electrical length of the series transmission line: (a) Z_C,low, (b) Z_C,peak, and (c) Z_P,peak.

Fig. 5

Optimized impedances: (a) Z_C,low, (b) Z_C,peak, and (c) Z_P,peak.

Using the optimized load network including the IDCC, an asymmetric DPA was additionally designed using the input-matching networks, biasing circuits, and input power splitter. Fig. 6 shows the full schematic of the designed DPA. For the input-matching networks, one section-matching network using an open stub and a series transmission line was adopted. An offset line to compensate for the phase difference between the carrier and peaking amplifiers was employed at the input of the carrier amplifier. A 90° hybrid coupler was used to split the input power.

Fig. 6

Schematic of the designed DPA.

IV. Implementation and Experimental Results

The designed DPA was implemented as a hybrid integrated circuit on a printed circuit board (PCB) based on Roger’s RO4350B, which has a dielectric constant of 3.66 and a thickness of 20 mil. Its size is about 5.9 cm × 6.7 cm. Fig. 7 shows the photograph of the implemented DPA. The carrier amplifier operates in the Class-AB condition with V_gs_,_carrier of −2.9 V and I_Q_,_carrier of 19 mA, while the peaking amplifier operates in the Class-C condition with V_gs_,_peaking of −6 V. Fig. 7 shows that the resonance circuit was realized using a chip capacitor of 1.1 pF and a chip inductor of 0.9 nH. To compensate for the nonideal parasitic effect, the values of the capacitor and inductor were slightly tuned from the ideal values optimized in the previous section. The optimum length of the series transmission line was also tuned to be about 70°. The implemented DPA was measured with and without IDCC. For the case with IDCC, the lumped inductor and capacitor for the resonance circuit were mounted by soldering them on the PCB (Fig. 7). To build a circuit without IDCC, the inductor and capacitor can be simply removed.

Fig. 7

Photograph of the fabricated DPA.

Fig. 8 shows the measured results of the implemented DPA using a CW signal. Fig. 8(a) and 8(b) represent the power gain and drain efficiency (DE) without the IDCC, while Fig. 8(c) and 8(d) represent the power gain and DE with the IDCC. The OBO was observed to be between 9 dB and 10 dB. For the frequency band between 3.3 GHz and 3.8 GHz, a DE between 29.2% and 56.1% at the average power region and 42.8% and 64.7% at the peak power region were achieved without IDCC, while a DE between 40.1% and 62.1% at the average power region and between 60.3% and 68.8% at the peak power region were achieved with IDCC. The DEs at both the low power and high power regions were considerably improved throughout the frequency band.

Fig. 8

Measured performances using a CW signal: (a) gain without IDCC, (b) DE without IDCC, (c) gain with IDCC, and (d) DE with IDCC.

Fig. 9 shows the measured results of the implemented DPA using a 5G new radio (NR) signal with a signal bandwidth of 100 MHz and a PAPR of 7.86 dB. Fig. 10 shows the simulated and measured performances with and without IDCC. Fig. 10(a) and 10(b) show the simulated and measured output power and DE using the CW signal, respectively. The overall performances over the frequency band were improved, especially for the peak output power and back-off power. Fig. 10(c) shows the measured DE and adjacent channel leakage power ratio (ACLR) at an output power of 34 dBm in the band. The figures show that the implemented DPA exhibited a DE between 32.1% and 45.1% without the IDCC and a DE between 43.2% and 50.7% with the IDCC at an output power of 34 dBm in the carrier frequency range between 3.35 GHz and 3.75 GHz. A significant improvement of DE can be thus observed. Fig. 11 shows the measured power spectral density (PSD) at an output power of 34 dBm before and after applying the DPD. Table 2 summarizes the measured performances compared to previous studies [11, 17–21]. The present study had a relatively broad bandwidth, with an FBW of 14% with a large OBO range between 9 dB and 10 dB. It used an extremely simple circuit compared to previous studies, except for [21], which employed a complex circuit and design scheme.

Fig. 9

Measured performances using a 5G NR signal with signal bandwidth of 100 MHz and a PAPR of 7.86 dB: (a) gain and DE without IDCC, (b) gain and DE with IDCC, (c) ACLR without IDCC, and (d) ACLR with IDCC.

Fig. 10

The simulated and measured performances with and without IDCC: (a) the simulated output power and DE using CW signal, (b) the measured output power and DE using CW signal, and (c) the measured output power, DE, and ACLR using the 5G NR signal.

Fig. 11

The measured PSD using the 5G NR signal before and after DPD at 3.55 GHz.

Table 2

Performance summary compared to previous studies

V. Conclusion

This study proposed an IDCC using a series transmission line and a resonance circuit for the post-matching network of the DPA. This IDCC simply distributed the impedances using the resonance circuit and rotated the impedances on the Smith chart to use them as a bandwidth extension method for the matching networks. For the DPA, the impedances looking at the post-matching network can be distributed and controlled for the quasi-lumped QWTL at the carrier amplifier to work in a broader frequency band.

For experimental verification, an asymmetric DPA was designed and implemented using GaN HEMT’s on a PCB. The implemented DPA exhibited significantly improved performances compared to the DPA without using the IDCC: power gain between 9.4 dB and 13.7 dB and DE between 43.2% and 50.8% with an FBW of 14%. This relatively large FBW compared to that found by past studies and a significantly improved DE compared to the DPA without the IDCC prove that the proposed method is effective in broadening the DPA bandwidth.

References

1. Doherty W. H.. A new high efficiency power amplifier for modulated waves. Proceedings of the Institute of Radio Engineers 24(9):1163–1182. 1936; https://doi.org/10.1109/JRPROC.1936.228468.

2. Kang H., Lee W., Oh S., Oh H., Choi W., Lee H., et al. Optimized broadband load network for Doherty power amplifier based on bandwidth balancing. IEEE Microwave and Wireless Components Letters 31(3):280–283. 2021; https://doi.org/10.1109/LMWC.2020.3043770.

3. Kang H., Lee H., Lee W., Oh H., Lim W., Koo H., et al. Octave bandwidth Doherty power amplifier using multiple resonance circuit for the peaking amplifier. IEEE Transactions on Circuits and Systems I: Regular Papers 66(2):583–593. 2019; https://doi.org/10.1109/TCSI.2018.2869905.

4. Oh H., Kang H., Lee H., Koo H., Kim M., Lee W., et al. Doherty power amplifier based on the fundamental current ratio for asymmetric cells. IEEE Transactions on Microwave Theory and Techniques 65(11):4190–4197. 2017; https://doi.org/10.1109/TMTT.2017.2701376.

5. Jee S., Lee J., Son J., Kim S., Kim C. H., Moon J., et al. Asymmetric broadband Doherty power amplifier using GaN MMIC for femto-cell base-station. IEEE Transactions on Microwave Theory and Techniques 63(9):2802–2810. 2015; https://doi.org/10.1109/TMTT.2015.2442973.

6. Kim J., Cha J., Kim I., Kim B.. Optimum operation of asymmetrical-cells-based linear Doherty power amplifiers-uneven power drive and power matching. IEEE Transactions on Microwave Theory and Techniques 53(5):1802–1809. 2005; https://doi.org/10.1109/TMTT.2005.847073.

7. Yang Y., Cha J., Shin B., Kim B.. A fully matched N-way Doherty amplifier with optimized linearity. IEEE Transactions on Microwave Theory and Techniques 51(3):986–993. 2003; https://doi.org/10.1109/TMTT.2003.808713.

8. Kang H., Lee H., Oh H., Lee W., Park C. S., Hwang K. C., et al. Symmetric three-way Doherty power amplifier for high efficiency and linearity. IEEE Transactions on Circuits and Systems II: Express Briefs 64(8):862–866. 2017; https://doi.org/10.1109/TCSII.2016.2609460.

9. Xia J., Chen W., Meng F., Yu C., Zhu X.. Improved three-stage Doherty amplifier design with impedance compensation in load combiner for broadband applications. IEEE Transactions on Microwave Theory and Techniques 67(2):778–786. 2019; https://doi.org/10.1109/TMTT.2018.2884404.

10. Srirattana N., Raghavan A., Heo D., Allen P. E., Laskar J.. Analysis and design of a high-efficiency multistage Doherty power amplifier for wireless communications. IEEE Transactions on Microwave Theory and Techniques 53(3):852–860. 2005; https://doi.org/10.1109/TMTT.2004.842505.

11. Maroldt S., Ercoli M.. 3.5-GHz ultra-compact GaN class-E integrated Doherty MMIC PA for 5G massive-MIMO base station applications. In : Proceedings of 2017 12th European Microwave Integrated Circuits Conference (EuMIC). Nuremberg, Germany; 2017; p. 196–199. https://doi.org/10.23919/EuMIC.2017.8230693.

12. Pang J., He S., Huang C., Dai Z., Peng J., You F.. A post-matching Doherty power amplifier employing low-order impedance inverters for broadband applications. IEEE Transactions on Microwave Theory and Techniques 63(12):4061–4071. 2015; https://doi.org/10.1109/TMTT.2015.2495201.

13. Chen X., Chen W.. A novel broadband Doherty power amplifier with post-matching structure. In : Proceedings of 2012 Asia Pacific Microwave Conference. Kaohsiung, Taiwan; 2012; p. 370–372. https://doi.org/10.1109/APMC.2012.6421601.

14. Pang J., He S., Dai Z., Huang C., Peng J., You F.. Design of a post-matching asymmetric Doherty power amplifier for broadband applications. IEEE Microwave and Wireless Components Letters 26(1):52–54. 2016; https://doi.org/10.1109/LMWC.2015.2505651.

15. Xia J., Yang M., Guo Y., Zhu A.. A broadband high-efficiency Doherty power amplifier with integrated compensating reactance. IEEE Transactions on Microwave Theory and Techniques 64(7):2014–2024. 2016; https://doi.org/10.1109/TMTT.2016.2574861.

16. Fang X., Cheng K. K. M.. Broadband, wide efficiency range, Doherty amplifier design using frequency-varying complex combining load. In : Proceedings of 2015 IEEE MTT-S International Microwave Symposium. Phoenix, AZ, USA; 2015; p. 1–4. https://doi.org/10.1109/MWSYM.2015.7166785.

17. Zhou J., Chen W., Chen L., Feng Z.. 3.5–0Hz high-efficiency broadband asymmetric Doherty power amplifier for 5G applications. In : Proceedings of 2018 International Conference on Microwave and Millimeter Wave Technology (ICMMT). Chengdu, China; 2018; p. 1–3. https://doi.org/10.1109/ICMMT.2018.8563718.

18. Choi Y. C., Choi W., Oh H., Chen Y., Shin J., Jeon H., et al. Doherty power amplifier with extended high-efficiency range based on the utilization of multiple output power back-off parameters. IEEE Transactions on Microwave Theory and Techniques 70(4):2258–2270. 2022; https://doi.org/10.1109/TMTT.2022.3144422.

19. Shen C., He S., Yao T., Peng J., Xiao Z., You F.. A high-gain Doherty power amplifier with harmonic tuning. IEEE Microwave and Wireless Components Letters 32(4):320–323. 2022; https://doi.org/10.1109/LMWC.2021.3129179.

20. Kwon K., Choi W., Shin J., Chen Y., Choi Y. C., Bae S., et al. Compact load network having a controlled electrical length for Doherty power amplifier. IEEE Access 10:70440–70446. 2022; https://doi.org/10.1109/ACCESS.2022.3187087.

21. Li M., Pang J., Li Y., Zhu A.. Bandwidth enhancement of Doherty power amplifier using modified load modulation network. IEEE Transactions on Circuits and Systems I: Regular Papers 67(6):1824–1834. 2020; https://doi.org/10.1109/TCSI.2020.2972163.

22. Mochumbe S. E., Yang Y.. Design of a load modulated balanced amplifier with a two-stage control power amplifier. Journal of Electromagnetic Engineering and Science 23(3):294–301. 2023; https://doi.org/10.26866/jees.2023.3.r.170.

23. Kim J., Choi W., Choi Y., Oh H., Yang Y.. 3.5 GHz high-efficiency asymmetric Doherty power amplifier design using a complex combining load. The Journal of Korean Institute of Electromagnetic Engineering and Science 32(8):708–716. 2021; https://doi.org/10.5515/KJKIEES.2021.32.8.708.

Biography

Seungmin Woo, https://orcid.org/0000-0003-0386-5731 was born in Daegu, South Korea, in 1996. He received his B.S. degree from the Department of Electronic and Electrical Engineering, Sungkyunkwan University, Suwon, South Korea, in 2021, where he is currently pursuing his M.S. degree with the Department of Electrical and Computer Engineering.

Woojin Choi, https://orcid.org/0000-0003-4365-1519 was born in Siheung, South Korea, in 1993. He received his B.S. degree from the Department of Electronic and Electrical Engineering, Sungkyunkwan University, Suwon, South Korea, in 2018, where he is currently pursuing his Ph.D. with the Department of Electrical and Computer Engineering. His current research interests include the design of RF power amplifiers for base stations, broadband techniques, and MMICs.

Jaekyung Shin, https://orcid.org/0000-0003-4790-0156 was born in Seoul, South Korea, in 1993. He received his B.S. degree from the Department of Electronic and Electrical Engineering, Korea Aerospace University, Goyang, South Korea, in 2018. He is currently pursuing his Ph.D. with the Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South Korea.

Yifei Chen, https://orcid.org/0000-0003-2030-3351 was born in Hebei, China, in 1994. He received his B.S. degree from the Department of Electronic and Electrical Engineering, Korea University, Seoul, South Korea, in 2018. He is currently pursuing his Ph.D. with the Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South Korea.

Youngchan Choi, https://orcid.org/0000-0002-4510-4685 was born in Seoul, South Korea, in 1996. He received his B.S. degree from the Department of Electronic and Electrical Engineering, Sungkyunkwan University, Suwon, South Korea, in 2020, where he is currently pursuing his Ph.D. with the Department of Electrical and Computer Engineering.

Sooncheol Bae, https://orcid.org/0000-0001-7905-3196 was born in Daegu, South Korea, in 1995. He received his B.S. and M.S. degrees from the Department of Electronic and Electrical Engineering, Sungkyunkwan University, Suwon, South Korea, in 2019, where he is currently pursuing his Ph.D. with the Department of Electrical and Computer Engineering.

Hyeongjin Jeon, https://orcid.org/0000-0003-1223-3701 was born in Mokpo, South Korea, in 1994. He received his B.S. degree from the Department of Electronic and Electrical Engineering, Sungkyunkwan University, Suwon, South Korea, in 2020, where he is currently pursuing his Ph.D. with the Department of Electrical and Computer Engineering.

Young Yun Woo, https://orcid.org/0009-0006-5890-5621 received his Ph.D. in electrical engineering from the Pohang University of Science and Technology (POSTECH), Pohang, Korea, in 2007. He joined Samsung Electronics in 2007 and has been working on the H/W R&D Group. His current research interests include 5G RF PA design, DPD linearization techniques, and 5G RF advanced techniques.

Youngoo Yang, https://orcid.org/0000-0003-3463-0687 was born in Hamyang, South Korea, in 1969. He received his Ph.D. in Electrical and Electronic Engineering from the Pohang University of Science and Technology, Pohang, South Korea, in 2002. From 2002 to 2005, he was with Skyworks Solutions Inc., Newbury Park, CA, USA, where he designed power amplifiers for various cellular handsets. Since 2005, he has been with the School of Information and Communication Engineering, Sungkyunkwan University, Suwon, South Korea. His research area is RF power amplifiers.

Article information Continued

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

	L (nH)	C (pF)
Case 1	0.5	3.9
Case 2	1.0	1.9
Case 3	1.5	1.3
Case 4	2.0	1.0

Study	Freq. (GHz)	FBW (%)	Gain (dB)	P_sat (dBm)	P_avg (dBm)	DE at P_avg (%)	ACLR^b) (dBc)	OBO (dB)	PAPR (dB)	Signal BW (MHz)	Signal	Device
Zhou et al. [17]	3.3–3.55	7	11–15^c)	47.5	40	50.6^a)	−26/−46.7	7.5–8	7.5	20	LTE	CGH40025 CGH40035
Choi et al. [18]	3.45–3.75	8	9.5–11.8	41.8–43.5	34.6–36.8	38.5–50.2	−24.6/NA	8.5	8	100	5G NR	CGH40006P CG2H40010F
Maroldt and Ercoli [11]	3.4–3.6	6	28.0–29.0	43	35	43^a)	−24/−50	8	7.2	20	LTE	GaN MMIC
Shen et al. [19]	3.1–3.6	15	13.0–15.1	41.8–42.8	35.8	38.5	−29.6/−47.9	6	6.5	100	64QAMCGHV27015S
Kwon et al. [20]	3.4–3.8	11	10.5–12.5^c)	43.6–44.4	35.8–36.6	43–51^c)	−23/−42	6	7.8	100	5G NR	CGH40010F
Li et al. [21]	2.8–3.55	23	8.3–9.1	43–45	36.5–38.5	49.1–57.2	−30.3/−53.5	6	6.5	40	OFDM	CGH40010F
This work	3.3–3.8	14	9.4–13.7	42.9–43.4	33.2–34.3	43.2–50.8	−25.8/−44.8	9–10	7.86	100	5G NR	CGH40006P CGH40010F