With the rapid growth of data traffic, there is an increasing demand for larger bandwidths in fixed networks. The transmitter bandwidth extension is usually achieved by combining analog multiplexers (AMUX) with digital-to-analog converters (DAC) designed in CMOS. Additionally, feedforward equalization (FFE) is incorporated at the high-speed front-end. However, this adds complexity and significant power consumption. In this work, we describe an alternative FFE approach that reduces the complexity of power-hungry high-speed front-ends, thereby improving power efficiency. This approach uses a 28 nm CMOS chip that sends two outputs with FFE predistorted data at 40/80 Gbps NRZ/PAM4. When combined with an AMUX, it generates equalized eye diagrams at 80/160 Gbps NRZ/PAM4 after a channel model with an insertion loss of 8.4 dB at 40 GHz, with a voltage swing of 100mVpp,diff/190mVpp,diff, while dissipating 0.889W of power (5.56 pJ/bit).
Gorzka, T, Ingels, M, Wang, X, Van Kerrebrouck, J, Singh, N, Torfs, G, Bauwelinck, J, Craninckx, J & Wambacq, P 2025, 'A 2-way interleaving 3-tap analog feedforward equalization for high-speed analog multiplexing', Analog Integrated Circuits and Signal Processing, vol. 123, no. 3, 51, pp. 1-13. https://doi.org/10.1007/s10470-025-02387-z
Gorzka, T., Ingels, M., Wang, X., Van Kerrebrouck, J., Singh, N., Torfs, G., Bauwelinck, J., Craninckx, J., & Wambacq, P. (2025). A 2-way interleaving 3-tap analog feedforward equalization for high-speed analog multiplexing. Analog Integrated Circuits and Signal Processing, 123(3), 1-13. Article 51. https://doi.org/10.1007/s10470-025-02387-z
@article{5894a2a65d324c7f9bfb7f0f3d4d21a9,
title = "A 2-way interleaving 3-tap analog feedforward equalization for high-speed analog multiplexing",
abstract = "With the rapid growth of data traffic, there is an increasing demand for larger bandwidths in fixed networks. The transmitter bandwidth extension is usually achieved by combining analog multiplexers (AMUX) with digital-to-analog converters (DAC) designed in CMOS. Additionally, feedforward equalization (FFE) is incorporated at the high-speed front-end. However, this adds complexity and significant power consumption. In this work, we describe an alternative FFE approach that reduces the complexity of power-hungry high-speed front-ends, thereby improving power efficiency. This approach uses a 28 nm CMOS chip that sends two outputs with FFE predistorted data at 40/80 Gbps NRZ/PAM4. When combined with an AMUX, it generates equalized eye diagrams at 80/160 Gbps NRZ/PAM4 after a channel model with an insertion loss of 8.4 dB at 40 GHz, with a voltage swing of 100mVpp,diff/190mVpp,diff, while dissipating 0.889W of power (5.56 pJ/bit).",
author = "Thomas Gorzka and Mark Ingels and Xin Wang and {Van Kerrebrouck}, Joris and Nishant Singh and Guy Torfs and Johan Bauwelinck and Jan Craninckx and Piet Wambacq",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.",
year = "2025",
month = jun,
doi = "10.1007/s10470-025-02387-z",
language = "English",
volume = "123",
pages = "1--13",
journal = "Analog Integrated Circuits and Signal Processing",
issn = "0925-1030",
publisher = "Springer Netherlands",
number = "3",
}