Chapter 4 Design of Low-Power High-Linearity Inductorless Mixer for UWB
4.4.1 Measured considerations
The proposed fully integrated mixer is fabricated in TSMC 0.18 µm CMOS technology and for on-wafer measurement. Fig. 4.7 and Fig. 4.8 show the chip layout and micrograph with a chip size of 0.84 x 0.72 mm2 including pads. According to the testing rules of the measured environment, pad-to-pad must have limited minimum distance to prevent a knock of the probes. Therefore, the layout area of the designed mixer with pads can’t be reduced. For actual application, the UWB mixer can achieve smaller size. The layout uses bottom ground metal under important signal path to shield off noise from substrate. The circuit operates at supply voltage of 1.8 V, and the core mixer consumes 4.68 mW power. The on-chip bypass capacitors is included between each voltage source and ground in the circuit.
Fig. 4.7. Layout of the proposed mixer.
Fig. 4.8. Micrograph of the proposed mixer.
The proposed mixer for ultra-wideband application is designed for on-wafer measurement. The layout follows the CIC probe station testing rules. Fig. 4.9 expresses the arrangements of the probes. The proposed mixer is designed to require three RF GSGSG probes for differential signals and a 6-pin DC probe. All probes are pitch 100 µm for saving area. The RF port, LO port, and IF port all need external balun and DC block capacitors for measured purpose.
Fig. 4.9. On-wafer measurement of UWB LNA test diagram.
The measurement setup for RF port and IF port return loss, conversion gain, P1dB, IIP3, and noise figure is shown in Fig. 4.10(a-d). The required equipments include a network analyzer, a spectrum analyzer, a noise analyzer, three signal generators, and a DC power supply.
(a)
Balun Balun
Balun
Spectrum Analyzer Signal
Generator
Signal Generator
(b)
(c)
Balun Balun
(d)
Fig. 4.10. Measurement setup of the proposed mixer for (a) RF port and IF port return loss. (b) conversion gain and P1dB. (c) IIP3. (d) noise figure.
4.4.2 Measured Results and discussion
Fig. 4.11(a-c) shows the conversion gain versus the LO power to find what LO power could lead to maximum conversion gain. The measured result shows the LO power of -2.5 dBm can achieve maximum conversion gain in UWB band. The measured and simulated conversion gain of the proposed mixer versus RF frequency is shown in Fig. 4.12 with a fixed IF frequency of 100 MHz, RF power of -30 dBm, and LO power of -2.5 dBm. Fig. 4.13 shows the input RF port return loss, which the measured result achieves better than -11 dB over 3.1 to 10.6 GHz. It reveals that the proposed transconductance stage is suitable for matching of broad bandwidth. The measured output IF port return loss is -26 dB at 100 MHz due to the usage of the buffers for measured requirement. Fig. 4.14(a-c) shows the measured and simulated P1dB at RF frequency of 3.1 GHz, 6.6 GHz, and 10.6 GHz. The measured IIP3 at RF frequency are illustrated in Fig. 4.15. Fig. 4.16 shows the measured and simulated IIP3 versus RF frequency with a fixed IF frequency of 100 MHz and LO power of -2.5 dBm. The Measured and simulated double-sideband noise figure (DSB NF) from 3.1 to 10.6 GHz as shown in Fig. 4.17. The measured DSB NF is worst compared with simulated result about 4 dB. The mixer isolation are shown in Fig 4.18 and Fig.
4.19. The measured LO-to-RF isolation is better than 59 dB over the operation frequency due to the excellent isolation of CG configuration. The RF-to-IF isolation also has better than 31 dB performance. The performance summary is listed in table 4.1. The proposed core mixer is suitable for low power operation, which only consumes 4.68 mW power from 1.8 supply voltage.
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6
RF frequency of 3.1 GHz
(a)
RF frequency of 6.6 GHz
(b)
RF frequency of 10.6 GHz
(c)
Fig. 4.11. Measured conversion gain versus LO power at (a) RF frequency of 3.1 GHz.
(b) RF frequency of 6.6 GHz. (c) RF frequency of 10.6 GHz.
2 3 4 5 6 7 8 9 10 11
Fig. 4.12. Measured and Simulated Conversion Gain of the proposed mixer.
2 3 4 5 6 7 8 9 10 11 12
RF Port Return Loss (dB) Measurement Simulation
Fig. 4.13. Measured and simulated RF port return loss.
-30 -25 -20 -15 -10 -5 0 RF Input Power (dBm)
-4
RF Input Power (dBm) -5
RF Input Power (dBm) -8
Fig. 4.14. Measured and simulated P1dB at (a) RF frequency of 3.1 GHz. (b) RF frequency of 6.6 GHz. (c) RF frequency of 10.6 GHz.
-30 -25 -20 -15 -10 -5 0 5 10 RF Input Power (dBm)
-80
IF Output Power (dBm)
RF frequency of 3.1 GHz
-30 -25 -20 -15 -10 -5 0 5 10
RF Input Power (dBm) -80
IF Output Power (dBm)
RF frequency of 4.9 GHz
(a) (b)
-30 -25 -20 -15 -10 -5 0 5 10
RF Input Power (dBm) -80
IF Output Power (dBm)
RF frequency of 6.6 GHz
-30 -25 -20 -15 -10 -5 0 5 10
RF Input Power (dBm) -80
IF Output Power (dBm)
RF frequency of 8.6 GHz
(c) (d)
-30 -25 -20 -15 -10 -5 0 5 10
RF Input Power (dBm) -80
IF Output Power (dBm)
RF frequency of 10.6 GHz
(e)
Fig. 4.15. Measured IIP3 at (a) RF frequency of 3.1 GHz. (b) RF frequency of 4.9 GHz.
(c) RF frequency of 6.6 GHz. (d) RF frequency of 8.6 GHz. (e) RF frequency of 10.6 GHz.
2 3 4 5 6 7 8 9 10 11
Fig. 4.16. Measured and simulated IIP3 with a fixed IF frequency of 100 MHz and LO power of -2.5 dBm.
Fig. 4.17. Measured and simulated double sideband noise figure.
2 3 4 5 6 7 8 9 10 11
Fig. 4.18. RF-to-LO isolation.
2 3 4 5 6 7 8 9 10 11
Fig. 4.19. LO-to-RF isolation.
Table 4.1
Performance summary of the proposed mixer
Specification Measurement Post Simulation
RF Return Loss (dB) -11.4~-10.8 -11.4~-10.6
IF Return Loss (dB) -26 -20.3
Conversion Gain (dB) -0.7± 2.6 2.2 ± 0.9
DSB Noise Figure (dB) 14.3~19.6 13.3~15
P1dB (dBm) -6.5~-2.5 -5~-3
IIP3 (dBm) 3~8 5.6~7.8
RF-to-IF Isolation (dB) 31.2~53.5 47.1~48
LO-to-RF Isolation (dB) 59.1~62.5 66~76
VDD (V) 1.8 V 1.8 V
Core Mixer Power (mW) 4.68 4.3
4.4.3 Comparison with other literatures
The performances of the proposed UWB mixer are listed and compared with other works as shown in Table 4.2. The conversion gain of this work is worse than simulation and other works perhaps owning to transistor variation, thus the DSB NF also degrades about 4 dB at high frequency. The propose UWB mixer achieves better IIP3 than other works. It requires the smallest power of 4.68 mW, which is suitable for UWB low power application.
Table 4.2
Comparison of the UWB mixer
Ref. Process
Chapter 5
Conclusion and Future Work
5.1 Conclusion
This thesis contains three works: the low voltage low-noise amplifier with image-rejection function for WLAN system, the low-noise amplifier for UWB system, and the low-power high-linearity inductorless mixer for UWB system. These three circuits are fabricated in TSMC 0.18 um CMOS process supported by CIC. The design concepts and research results will be described as follow.
5.1.1 Design of LNA with IR function for WLAN system
This work presents a new design of the LNA with IR function by adding a simple parallel LC tank into the inter-stage of the current-reused structure to obtain satisfying performance and IR function. Comparing with other researches, no complex circuit is required in this design, thus it is easier to achieve for designers.
Due to the use of passive filter, there are no additional power and noise contribution than active filter. In general, on-chip inductor is usually considered as low quality factor, thus filter with on-chip inductors is difficult to achieve better performance.
Thus the Q enhancement technique is also adopted in the proposed LNA to successfully achieve better IR ability than [3] with active filter. In addition, all reported researches exploit cascode configuration and improve it by adding IR filter, while this work employs current-reused technique, which is suitable for low voltage, low power operation and obtains high gain. The proposed LNA only uses 1 V supply voltage to lower power consumption.
5.1.2 Design of current-reused LNA for UWB system
The input impedance matching of wideband LNA for matching 50 Ω to external antenna is a challenge. One of wideband impedance matching approach is resistive shunt feedback, which can reduce input resistance throughout resistor feedback due to Miller effect, but the drawback is degradation of gain. This work improves resistive shunt feedback by stacking PMOS and NMOS to enhance the transconductance, thus the gain can be increased. By adding gate inductor and source inductor, the high frequency input return loss can be improved. The measured results shows the proposed UWB LNA can achieve average gain of 14.1 dB while consumes 14.4 mW power.
5.1.3 Design of low-power high-linarity inductorless mixer for UWB system.
For conventional Gilbert-cell mixer, the use of common-source structure as transconductance stage requires input matching network for wideband application.
The matching network usually contains inductors, which occupies large die area. The proposed mixer exploits common-gate structure with source resistor as transconductance stage to achieve inductorless wideband input matching. Except for saving area, this design has wider bandwidth to achieve better linearity. The dynamic current-injection technique is also adopted to improve voltage headroom and no contribution of noise compared to traditional current-injection method. The measured results show the good IIP3 of 3 dBm to 8 dBm. The proposed mixer only consumes 4.68 mW power from 1.8 V supply voltage.
5.2 Future work
In this thesis, there are some directions that can be improved for the future. For the proposed LNA with IR function, the image frequency can be designed closer to the RF frequency, and the NF has potential to achieve better performance, For the proposed current-reused LNA, we can retune the component value in the LNA to flatten the gain. For the inductorless mixer, an object is to increase conversion gain and maintain high linearity at the same time. Because both LNA and mixer are application for UWB system, we can combine these two circuits to design a RF front-end.
References
[1] J. Macedo et al., “A 1.9 GHz silicon receiver with monolithic image reject filtering,” IEEE J. Solid-State Circuits, vol. 33, no. 3, pp. 378–386, Mar. 1998.
[2] H. Samavati et al., “A 5-GHz CMOS wireless LNA receiver front-end,” IEEE J.
Solid-State Circuits, vol. 35, no. 5, pp. 765–772, May 2000.
[3] Trung-Kien Nguyen, Nam-Jin Oh, Choong-Yul Cha, Yong-Hun Oh, Gook-Ju Ihm, and Sang-Gug Lee., “Image-Rejection CMOS Low-Noise Amplifier Design Optimization Techniques,” IEEE Trans. Microwave Theory and Techniques, vol.
53, no. 2, February 2005.
[4] Ming-Chang Sun, Shing Tenqchen, Ying-Haw Shu, and Wu-Shiung Feng., “A 2.4 GHz CMOS image-reject low noise amplifier,” Circuits and Systems, 2003.
ISCAS '03. Proceedings of the 2003 International Symposium, vol. 1, pp.
I-329–I-332, 2003.
[5] Moneim Youssef A. A., Sharaf K., Ragaie H.F., and Marzouk Ibrahim M., “VLSI design of CMOS image-reject LNA for 950-MHz wireless receivers,” Circuits and Systems for Communications, 2002. Proceedings. ICCSC '02. 1st IEEE International Conference, pp. 330-333, 2002.
[6] R.-C. Liu, K.-L. Deng, and H. Wang, “A 0.6-22 GHz broadband CMOS distributed amplifier,” Radio Frequency Integrated Circuits Symp. Dig. Papers, 2002, pp. 103-106.
[7] Robert Hu, ”Wide-Band Matched LNA Design using transistor’s intrinsic Gate-Drain Capacitor,” Microwave Theory and Techniques, vol. 54, No. 3.
[8] C.W. Kim, M.S. Jung and S.G. Lee, “Ultra-Wideband CMOS low noise amplifier,” electronics letters, vol. 41, no. 7, pp.384-385.
[9] Liao, C.F. and Liu, S.I., “A broadband noise-canceling CMOS LNA for 3.1-10.6-GHz UWB receiver,” Proc. IEEE Custom Integrated Circuits Conf., pp.
161–164. 2005.
[10] Kuan-Hung Chen and Chorng-Kuang Wang, “A 3.1~10.6 GHz CMOS Cascaded Two-stage Distributed Amplifier for Ultra-Wideband Application,” IEEE Asia-Pacific Conference, pp. 296 – 299, Aug. 2004.
[11] J.-H. Lee, C.-C. Chen, and Y.-S. Lin., “0.18um 3.1–10.6 GHz CMOS UWB LNA with 11.4± 0.4 dB gain and 100.7 ± 17.4 ps group delay,” ELECTRONICS LETTERS, vol.43, no. 24, 22 Nov. 2007.
[12] H.-Y. Yang, Y.-S. Lin, and C.-C. Chen, “ 2.5 dB NF 3.1–10.6 GHz CMOS UWB LNA with small group-delay variation,” ELECTRONICS LETTERS, vol.44, no.8, 10 Apr. 2008.
[13] Li, Q., Zhang, Y.P., and Chang, J.S., “An inductorless low-noise amplifier with noise cancellation for UWB receiver front-end,” Proc. IEEE Asian Solid-State Circuits Conf. pp. 267–270, 2006.
[14] Safarian, A. Yazdi, and P. Heydari, “Design and Analysis of an Ultra wide-Band Distributed CMOS Mixer,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 13, no. 5, pp. 618-629, May, 2005.
[15] M. Tsai and H. Wang, “A 0.3-25 GHz Ultra-Wideband Mixer using Commercial 0.18um CMOS Technology,” IEEE Microwave and Wireless Components Letters, vol. 14, no. 11, pp. 522-524, Nov. 2004.
[16] Leonard A. MacEachern and Tajinder Manku, “A Charge-Injection Method for Gilbert Cell Biasing,” Electrical and Computer Engineering, pp. 365-368, 1998.
[17] Hooman Darabi and Janice Chiu, “A Noise Cancellation Technique in Active RF-CMOS Mixers,” IEEE Journal of solid-state circuits, vol. 40, no. 12, Dec.
2005.
[18] W. L. Chen, G. W. Huang, and S. F. Chang, ” BROADBAND CMOS MIXER CELL FOR UWB APPLICATIONS,” Microwave and Optical Technology Letters, vol. 49, Issue 1, pp.127-128, Jan. 2007.
[19] Jeong-Bae Seo, Kun-Man Park, Jong-Ha Kim, Jin-Hong Park, Young-Sop Lee, Jeong-Hyun Ham, and Tae-Yeoul Yun, “A Low-Noise UWB CMOS Mixer Using Switched Biasing Technique,” Radio-Frequency Integration Technology, 2007.
RFIT 007, pp. 42-45, 9-11 Dec. 2007.
[20] Satyanarayana Reddy, K., Annamalai Arasu, M., King Wah Wong, Yuanjin Zheng, and Fujiang Lin, “Low-Power UWB LNA and Mixer using 0.18-um CMOS Technology,” ESSCIRC 2006 Solid-State Circuits Conference, pp.
259-262, Sept. 2006.
[21] Goo-Young Jung, Jae-Hoon Shin, and Tae-Yeoul Yun, “A low-noise UWB CMOS mixer using current bleeding and resonant inductor techniques,”
Miceowave and Optical Technology Letters, vol. 49, issue 4, pp. 1595-1597, July 2007.
[22] Chang, F.-C., Huang, P.-C., Chao, S.-F., Wang, H., “A Low Power Folded Mixer for UWB System Applications in 0.18-um CMOS Technology,” IEEE Miceowave and Wireless Compoenets Letters, vol. 17, issue 5, pp. 367-369, May 2007.
Vita
姓 名: 魏廉昇 學 歷:
國立中壢高級中學 ( 88 年 9 月 ~ 91 年 6 月)
國立彰化師範大學電機工程學系 ( 91 年 9 月 ~ 95 年 6 月) 國立交通大學電信工程所碩士班 ( 95 年 9 月 ~ 97 年 6 月)
Publication Remarks
1. Lien-Sheng Wei and Christina F. Jou, “An Image Rejection Low Noise Amplifier for WLAN System,” Progress In Electromagnetics Research Symposium (PIERS) 2008, March 24–28, 2008, Hangzhou, China.
2. Lien-Sheng Wei and Christina F. Jou, “Design of low voltage CMOS low-noise amplifier with image-rejection function,” ELECTRONICS LETTERS.