SVM based method for multi-equalizer optimization

Optimization of equalization parameters in SerDes systems is one of the most challenging problems in modern communication systems. The difficulty of properly tuning equalizer parameters for minimum bit error rate (MBER) increases as channel loss becomes harder to overcome, driving design into more complex equalizers. The complexity of equalizer and forward error correction (FEC) systems are directly determined by the equalizer’s ability to correct intersymbol interference (ISI) from the channel. In this work, the author presents a novel method of tuning equalizer parameters that specifically focuses on MBER and eye dimensions over traditional minimum mean squared error (MMSE) based methods.

[1]  C.-Y. Chen,et al.  A novel decision feedback equalizer design based on generalized space translation , 2004, Conference Record of the Thirty-Eighth Asilomar Conference on Signals, Systems and Computers, 2004..

[2]  Sheng Chen,et al.  Design of the optimal separating hyperplane for the decision feedback equalizer using support vector machines , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[3]  Weisheng Hu,et al.  Low-overhead Equalization by SVM Classification Enabled 25-Gb/s NRZ per Wavelength Transmission with 10-G Class Optics , 2018, 2018 Asia Communications and Photonics Conference (ACP).