A K-Band Gilbert-Cell Mixer with Transformer-Enhanced Transconductance,45nm Band-Pass Filtered Switching, and low-Pass Load Conversion in 22nm FDSOI CMOS

Abstract

This paper presents \"A K-Band Gilbert-Cell Mixer with Transformer-Enhanced Transconductance,45nmBand-Pass Filtered Switching, and low-Pass Load Conversion in 22nm FDSOI CMOS\" presents a low-power K-band down-conversion mixer designed using 22nm FDSOI CMOS technology. The mixer adopts a modified Gilbert-cell architecture, replacing the active transconductance stage with a passive transformer to overcome voltage headroom limitations in advanced CMOS processes. Additionally, an extension band pass filter is incorporated into the switching stage to suppress harmonics, noise, and spurious signals, ensuring a cleaner intermediate frequency (IF) output. Further, a filter in the load conversion stage refines the final IF signal by minimizing unwanted frequency components and improving signal integrity. The design achieves a conversion gain of 9.03dB, an input 1dB compression point of -7dBm, while consuming just 0.799mW from a 0.8V supply. Its compact size of 0.54×0.45mm², combined with high linearity and low power consumption, makes it ideal for applications like 5G transceivers and short-range radar systems. This project highlights the advantages of 22nm FDSOI technology in developing energy-efficient, high-frequency RF circuits.

Introduction

The study presents a novel down-conversion mixer design for RF front-end systems, crucial in applications like 5G, automotive radar, and satellite communications. Traditional Gilbert-cell mixers with active transconductance stages face challenges in advanced CMOS technologies (e.g., 22nm FDSOI) due to limited voltage headroom and high power consumption, which degrade performance and linearity. To address this, the proposed design replaces the active transconductance stage with a passive transformer-based approach, improving voltage headroom, reducing power use, and enhancing efficiency by directly converting RF voltage to current.

An integrated extension filter within the switching stage suppresses unwanted harmonics and noise, enhancing spectral purity of the intermediate frequency (IF) output. Implemented in 22nm FDSOI CMOS, the K-band mixer achieves low power consumption (0.799mW at 0.8V), a conversion gain of 9.03dB, and robust linearity with a 1dB compression point of -7dBm, all in a compact footprint.

The design methodology includes RF input matching, transformer-based transconductance, LO switching with integrated filtering, and load conversion. The design was realized and optimized via Tanner EDA tools with simulations confirming excellent performance in gain, noise figure (4.7 dB), linearity, and energy efficiency.

Future work suggests extending this architecture for higher frequencies (above 100 GHz) relevant to 6G and beyond, integrating tunable on-chip filters for multi-band and cognitive radios, and leveraging 3D IC and SoC technologies for further miniaturization and efficiency. Applications include automotive radar, satellite communications, IoT devices, and wireless sensor networks. AI-driven tuning may further enhance spectral purity and performance, making this mixer design promising for next-generation wireless systems.

Conclusion

This paper presents a low-power, high-linearity K-band Modified Gilbert-Cell Mixer, designed using 22nm FDSOI CMOS technology with a transformer-based transconductance stage and an integrated switching-stage filter. The proposed architecture successfully overcomes the challenges of limited voltage headroom and increased power dissipation in deep-submicron CMOS processes by eliminating the active transconductance stage and employing a passive transformer for efficient RF-to-current conversion. The inclusion of a 45nm integrated filter in the switching stage further enhances spectral purity by suppressing unwanted harmonics and spurious signals, ensuring a clean intermediate frequency (IF) output. The mixer achieves an impressive conversion gain of 9.03 dB, while consuming only 0.799 mW from a 0.8V supply. These results highlight the effectiveness of 22nm FDSOI CMOS technology in achieving low-power and high-performance RF circuits, making this design ideal for next-generation 5G transceivers, short-range radar systems, and satellite communication applications. Future advancements in passive transformer designs and adaptive filtering can further improve performance, making this mixer a strong candidate for emerging RF and mmWave communication systems.

References

[1] V. Issakov, H. Knapp, M. Tiebout, A. Thiede, W. Simburger and L. Maurer, ”Comparison of 24 GHz low-noise mixers in CMOS and SiGe:C technologies,”2009 European Microwave Integrated Circuits Conference (EuMIC), Rome, Italy, 2009, pp. 184-187. [2] Y. Peng et al., ”A K -Band High-Gain and Low-Noise Folded CMOS Mixer Using Current-Reuse and Cross-Coupled Techniques,” in IEEE Access, vol. 7, pp. 133218-133226, 2019, doi: 10.1109/ACCESS.2019.2941048. [3] B. Bae and J. Han, ”24–40 GHz Gain-Boosted Wideband CMOS Down-Conversion Mixer Employing Body-Effect Control for 5G NR Applications,” in IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 69, no. 3, pp. 1034-1038,March2022, doi: 10.1109/TCSII.2021.3119995. [4] R. Ciocoveanu, R. Weigel, A. Hage Lauer and VIssakov, ”Bias-switched down-conversion mixer for flicker noise reduction in 28-nm CMOS,” 2018 Texas Symposium on Wireless and Microwave Circuits and Systems (WMCS), Waco, TX,USA, 2018,pp. 1-4, doi: 10.1109/WMCaS.2018.8400634. [5] R. Ciocoveanu, R. Weigel, A. Hage Lauer and V. Issakov,” Modified Gilbert-Cell Mixer with an LO Waveform Shaper and Switched Gate Biasing for 1/f Noise Reduction in 28-nm CMOS,” in IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 66, no. 10, pp. 16881692, Oct. 2019, doi: 10.1109/TCSII.2019.2923595. [6] P. V. Testa, L. Szilagyi, X. Xu, C. Carta and F. Ellinger,” A Low-Power Low-Voltage Down-Conversion Mixer for 5G Applications at 28 GHz in 22-nm FD-SOI CMOS Technology,” 2020 IEEE Asia-Pacific Microwave Conference (APMC), Hong Kong, Hong Kong, 2020, pp. 911-913, doi:10.1109/APMC47863.2020.9331695.

Copyright

Copyright © 2025 T. Jyothi, M. Sai Vasavi, R. Kavitha, P. Sai Kushal, P. Mahesh . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.