• 沒有找到結果。

A Design of CMOS Class-E Power Amplifier with Phase Correction for Envelope Elimination and Restoration (EER)/Polar Systems

N/A
N/A
Protected

Academic year: 2021

Share "A Design of CMOS Class-E Power Amplifier with Phase Correction for Envelope Elimination and Restoration (EER)/Polar Systems"

Copied!
4
0
0

加載中.... (立即查看全文)

全文

(1)IEICE TRANS. ELECTRON., VOL.E93–C, NO.1 JANUARY 2010. 128. BRIEF PAPER. A Design of CMOS Class-E Power Amplifier with Phase Correction for Envelope Elimination and Restoration (EER)/Polar Systems Wen-An TSOU†a) , Student Member, Wen-Shen WUEN† , and Kuei-Ann WEN† , Nonmembers. SUMMARY A circuit technique to correct Vdd /PM distortion and improve efficiency as supply modulation of cascode class-E PAs has been proposed. The experimental result shows that the phase distortion can be improved from 20 degrees to 5 degrees. Moreover, a system co-simulation result demonstrated that the EVM can be improved from −17 dB to −19 dB. key words: envelope elimination and restoration (EER), polar system, class-E power amplifier, orthogonal frequency division multiplexing (OFDM). 1.. Introduction. The class-E power amplifier achieves 100% drain efficiency theoretically and is a good candidate for envelope elimination and restoration (EER) or polar transmitters. Considering reliability issues of PAs in CMOS process technology, a cascode topology of class-E power amplifier has been presented and shows the significant performance [1]. In EER or polar systems, the modulated signal is decomposed into the envelope signal and the phase signal. The phase signal is used to drive the class-E PA to be operated as a switch while the envelope signal modulates the supply voltage of PA. In theory, the class-E PA can adequately recombine both envelope and phase signals. However, a cascode class-E PA introduces undesired Vdd /AM and Vdd /PM distortions due to supply modulation. As mentioned in [2], Vdd /AM distortion is small enough to be neglected but, on the other hand, Vdd /PM distortion results in significant nonlinear phase shift in the supply voltage range. This nonlinearity will be damaging to the modulation accuracy of signal transmission. Since this effect is due to the variation of drain to source capacitance and drain to source conductance [2], a circuit technique to attain the constant drain to source impedance is proposed which can correct the Vdd /PM distortion to meet the system requirement. Detailed description of this technique is elaborated in Sect. 2. Section 3 shows the comparison of the simulation and measurement results. Moreover, a system co-simulation with orthogonal frequency division multiplexing (OFDM)-based signal has been established to demonstrate that this circuit technique for Vdd /PM distortion correction can effectively improve system error vector magnitude (EVM) performance. Finally, Sect. 4 draws conclusions. Manuscript received January 20, 2009. Manuscript revised August 13, 2009. † The authors are with National Chiao Tung University Electronics Engineering, Trans. Wireless Technology Laboratory, Hsinchu 300, Taiwan. a) E-mail: [email protected] DOI: 10.1587/transele.E93.C.128. 2.. Circuit Principle and Design Methodology. 2.1 Circuit Principle Figure 1 shows the schematic of the class-E PA. The gate voltage of the transistor M2 is generally biased at supply voltage Vdd or a fixed voltage Vdc and the inductor L3 and capacitor C3 are added to improve the efficiency [1]. In practice, due to supply modulation the variant condition of the transistor M2 leads to a non-constant drain to source voltage function of the transistor M2 and, then, varies the drain to source impedance (ZDS2 = |VDS2 /IDS2 |) of the transistor M2 , where VDS2 , IDS2 means drain to source voltage and drain to source current of the transistor M2 , respectively. Furthermore, the variable impedance results in the Vdd /PM distortion as supply modulation. During modulating the supply voltage of PA, the operating conditions of the transistor M2 are inferred as follows: 1. When Vdd < (Vdc − Vth )/(Vd2,max /Vdd,max ), the transistor M2 induces the strong inversion channel layer for current conducting but only occupies extremely small drain to source voltage headroom, where Vdd,max =1.8 V and Vd2,max is the peak drain voltage of the transistor M2 . 2. Until Vdd ≥ (Vdc − Vth )/(Vd2,max /Vdd,max ), it would be large enough to completely turn on the transistor M1 and, hence, the transistor M2 could have the ability of current driving and get the significant voltage headroom. The relationship of impedance and voltage of the transistor M2 can be found in Fig. 2. This simulation result shows that the non-constant voltage VDS2 would apparently vary drain to source impedance ZDS2 . Furthermore, the voltage waveform at carrier frequency on the drain node of. Fig. 1. A cascode class-E power amplifier.. c 2010 The Institute of Electronics, Information and Communication Engineers Copyright .

(2) BRIEF PAPER. 129. Fig. 2 The simulation result of ZDS2 and VDS2 as supply voltage variations.. Fig. 3. Fig. 4. The DC current (IDC ) and efficiency of the PA.. Fig. 5 The simulation result of ZDS2 and VDS2 of proposed PA as supply voltage variations.. Fig. 6 The real part and imaginary part of drain to source impedance of the transistor M2 .. The equivalent model of proposed cascode class-E PA.. the transistor M2 would suffer from the phase shift due to the non-constant impedance ZDS2 . Then, the output signal would accompany the phase shift from drain voltage of the transistor M2 . This phase shift is called Vdd /PM distortion. In addition, the other impact on the PA due to the nonconstant voltage is the drain efficiency reduction. Figure 3 shows the simulation results of DC current and drain efficiency. In small supply voltage range, DC current increases rapidly and, hence, results in a serious reduction in efficiency. Unfortunately, this efficiency property is not suitable for polar applications. 2.2 Design Methodology A design methodology which can correct the Vdd /PM distortion and improve the efficiency is proposed in this paper. Since the variable impedance induces the nonlinear phase shift, the constant impedance ZDS2 is expected to cancel this distortion effect. Figure 4 shows the equivalent model of. Fig. 7. The IDC and improved efficiency of proposed cascode class-E PA.. proposed cascode class-E PA, in which the transistor M1 is operated as a switch and the transistor M2 is degenerated into a resistance. The simulation result of ZDS2 and VDS2 as shown in Fig. 5 shows that the voltage VDS2 , indeed, develops the almost constant impedance ZDS2 . The voltage VDS2 and constant impedance ZDS2 would be beneficial for supply modulation of the PA. For more detail of this analysis, the simulation result of the real part and imaginary part of drain to source impedance of the transistor M2 is shown in Fig. 6. It demonstrated that the expected impedance, the almost constant real part and imaginary part of the drain to source impedance, can be achieved when the transistor M2 has been degenerated into a resistance. The more gain of the degenerative transistor M2 is to pull up PA drain efficiency in small voltage range of Vdd and,.

(3) IEICE TRANS. ELECTRON., VOL.E93–C, NO.1 JANUARY 2010. 130. therefore, can extend the operating voltage range as supply modulation. This obtained property of drain efficiency is further serviceable for EER/polar systems. The simulation result of improved efficiency can be found in Fig. 7. As shown on Eq. (1), when the PA with or without Vdd /PM distortion correction technique operates on the identical condition of Pout and Vdd , the efficiency would have an inverse proportion of IDC . Efficiency =. Pout Pout 1 × 100% ∝ ∝ PDC Vdd IDC IDC. (1). Comparing the result of Fig. 3 and Fig. 7, in small supply voltage range the PA with Vdd /PM correction can have the lower IDC value and, hence, the efficiency of PA can be improved by 15%. The proposed technique flattens the drain efficiency curve and, therefore, is beneficial for EER/Polar systems. 3.. area is 1.25 × 1.3 mm2 . The measured result as shown in Fig. 10 reveals that the PA with Vdd /PM distortion correction technique can improve the output signal phase shift from 30 degrees to 6 degrees in 0.4 V to 1.8 V of supply voltage Vdd . As Vdd below 0.5 V, the serious output signal phase shift due to the transistor M1 switching incompletely, may have severe degradations in system performances and, hence, the operating voltage above 0.5 V is recommended for EER/polar applications [2]. In Fig. 10, the result, meanwhile, represented that the proposed Vdd /PM distortion correction technique indeed decreases the output signal phase shift and, on the other hand, it demonstrated that the measured result is close to the simulation result. Furthermore, a system co-simulation platform of EER architecture has been established to manifest the influence of Vdd /PM effect of cascode class-E PA on system per-. Simulation and Measurement Results. The simulation result of Vdd /PM response in Fig. 8 shows that the output signal phase shift of PA can be improved from 20 degrees to less than 5 degrees in 0.3 V to 1.8 V of supply voltage Vdd . It exhibits that the proposed technique could evidently reduce the output signal phase shift. Therefore, the small phase shift variations can extend the supply voltage range of PA in EER/Polar systems. The cascode class-E PA has been implemented using 0.18µm CMOS process technology as shown in Fig. 9 and its. Fig. 10 The measured result of Vdd /PM response of cascode class-E PA without and with correction technique.. Fig. 11 Fig. 8. Histogram of the envelope voltage.. The simulation results of Vdd /PM response of cascode class-E PA.. Fig. 9. Die micrograph.. Fig. 12 Received constellation (a) without correction, (b) with correction..

(4) BRIEF PAPER. 131. formance. The signal source, IEEE 802.11a-like broadband OFDM transmission having 20 MHz bandwidth and operating at 2.6 GHz frequency band, is generated by ADS Ptolemy simulator. The histogram of envelope signal shown in Fig. 11 has 93% in the range from 0.5 to 1.8 V and the mean voltage is 1 V while the requested time delay is 2.8 ns [3]. With a Vdd /PM distortion correction technique of cascode class-E PA, the EVM can be improved from −17 dB to −19 dB and received constellation is shown in Fig. 12. 4.. Conclusions. The technique, degenerating M2 into a resistance, which can effectively correct the Vdd /PM distortion has been proposed. Moreover, the technique can improve the drain efficiency of the PA and extend the supply voltage range of the PA for EER/Polar systems. The measured result of Vdd /PM response has a good agreement with the simulation result. And, the system co-simulation result demonstrated that the proposed technique improves system EVM performance effectively.. Acknowledgments This work was conducted by the Trans-Wireless Technology Laboratory (TWT Lab.) and sponsored jointly by the Ministry of Education and the National Science Council, Taiwan under the contract: NSC 97-2220-E-0090-007. The authors would like to thank Dr. K.W. Huang of National NanoDevice Laboratory (NDL), Taiwan, for chip testing. References [1] A. Mazzanti, L. Larcher, R. Brama, and F. Svelto, “Analysis of reliability and power efficiency in cascode class-E PAs,” IEEE J. SolidState Circuits, vol.41, no.5, pp.1222–1229, May 2006. [2] A. Diet, M. Villegas, and G. Baudoin, “EER architecture specifications for OFDM transmitter using a class E amplifier,” IEEE Microw. Wirel. Compon. Lett., vol.14, no.8, pp.389–391, Aug. 2004. [3] F.H. Raab and D.J. Rupp, “Class-S high-efficiency amplitude modulator,” RF Design, vol.17, no.5, pp.70–74, May 1994..

(5)

參考文獻

相關文件

Understanding and inferring information, ideas, feelings and opinions in a range of texts with some degree of complexity, using and integrating a small range of reading

 Promote project learning, mathematical modeling, and problem-based learning to strengthen the ability to integrate and apply knowledge and skills, and make. calculated

Then they work in groups of four to design a questionnaire on diets and eating habits based on the information they have collected from the internet and in Part A, and with

利用 determinant 我 們可以判斷一個 square matrix 是否為 invertible, 也可幫助我們找到一個 invertible matrix 的 inverse, 甚至將聯立方成組的解寫下.

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

For example, Liu, Zhang and Wang [5] extended a class of merit functions proposed in [6] to the SCCP, Kong, Tuncel and Xiu [7] studied the extension of the implicit Lagrangian

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

A convenient way to implement a Boolean function with NAND gates is to obtain the simplified Boolean function in terms of Boolean operators and then convert the function to