Low-complexity non-integer fractionally spaced feed-forward equalization with half-symbol-spaced kernel estimation for 100-GBaud/λ PAM-4 transmission
A low-complexity non-integer fractionally spaced feed-forward equalizer (FFE) with half-symbol-spaced kernel estimation (HSSKE) is proposed for high-speed intensity modulation and direct detection (IM/DD) transmission systems. By employing HSSKE, the proposed scheme enables accurate and aliasing-free kernel estimation at 2 samples per symbol (sps) while achieving efficient equalization at fractional sampling rates below 2 sps. Experimental results on a 100-GBaud/λ PAM-4 system over 1-km and 2-km standard single-mode fibers (SSMFs) demonstrate that the proposed 1.2-sps FFE with HSSKE not only outperforms 1.2-sps FFE with symbol-spaced estimation (SSKE) but also achieves equalization performance comparable to that of the conventional 2-sps FFE, while reducing computational complexity by more than 27%. Furthermore, when integrated with a noise whitening filter (NWF) and maximum likelihood sequence estimation (MLSE), the proposed scheme achieves a 0.9-dB improvement in receiver sensitivity compared to the 1.2-sps FFE with SSKE under the KP4-forward error correction (FEC) threshold of 2.4 × 10 −4 over a 1-km link, and reduces the bit error rate (BER) below the 7% hard-decision FEC (HD-FEC) threshold of 3.8 × 10 −3 after 2-km transmission.
Authors
- Qiang Bin
- Alan Pak Tao Lau
- Zhaohui Li
- Chao Lu (ORCID: https://orcid.org/0000-0001-7585-9408)
- Junwei Zhang
Institutions
- Hong Kong Polytechnic University (HK)
Publication Details
- Journal
- Optics Letters
- Published
- 2026-04-21
- DOI
- https://doi.org/10.1364/ol.597634
- Primary Topic
- Advanced Wireless Communication Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Natural Science Foundation of Guangdong Province