Modified by special feature engineering, a powerful and low-order equalizer based on K-nearest neighbors (KNN) classifier is applied to improve performance of high-speed system with bandwidth-limited optics. The feature construction and feature weighting are specially designed to conduct an appropriate a feature engineering-based KNN (FE-KNN) scheme, which contains more data characteristics to enhance the equalization performance. Experimental comparisons of KNN classifier with/without feature engineering, decision feedback equalizer (DFE) and feed-forward equalizer (FFE) are implemented to prove the feasibility of our scheme in both 25-Gb/s NRZ and 50-Gb/s PAM-4 transmission experiments with 10-G optics system. The corresponding results show that, without the feature engineering, the performance achieved by the common KNN is not improved even in the case of hard decision (HD). In contrast, compared to the common 11-taps DFE, the performance achieved by FE-KNN with only 5 taps is improved by 1-dB at KP4-FEC threshold (BER=2.2E-4) for 25-Gb/s NRZ transmission. While, for 50-Gb/s PAM-4 case, 0.5-dB sensitivity improvement is achieved by our scheme compared to the common 11-taps DFE under the HD-FEC limit (BER=3.8E-3).
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1364/OE.27.025802 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
RetinaRegNet: A zero-shot approach for retinal image registration.
Comput Biol Med
January 2025
Department of Electrical and Computer Engineering, University of Florida, Gainesville, FL, 32610, United States; Department of Medicine, University of Florida, Gainesville, FL, 32610, United States; Department of Health Outcomes and Biomedical Informatics, University of Florida, Gainesville, FL, 32610, United States; Intelligent Clinical Care Center, University of Florida, Gainesville, FL, 32610, United States. Electronic address:
Retinal image registration is essential for monitoring eye diseases and planning treatments, yet it remains challenging due to large deformations, minimal overlap, and varying image quality. To address these challenges, we propose RetinaRegNet, a multi-stage image registration model with zero-shot generalizability across multiple retinal imaging modalities. RetinaRegNet begins by extracting image features using a pretrained latent diffusion model.
View Article and Find Full Text PDFVcaNet: Vision Transformer with fusion channel and spatial attention module for 3D brain tumor segmentation.
Comput Biol Med
January 2025
College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua, 321004, China; Zhejiang Institute of Optoelectronics, Jinhua, 321004, China. Electronic address:
Accurate segmentation of brain tumors from MRI scans is a critical task in medical image analysis, yet it remains challenging due to the complex and variable nature of tumor shapes and sizes. Traditional convolutional neural networks (CNNs), while effective for local feature extraction, struggle to capture long-range dependencies crucial for 3D medical image analysis. To address these limitations, this paper presents VcaNet, a novel architecture that integrates a Vision Transformer (ViT) with a fusion channel and spatial attention module (CBAM), aimed at enhancing 3D brain tumor segmentation.
View Article and Find Full Text PDFA DIY Bioreactor for in Situ Metabolic Tracking in 3D Cell Models via Hyperpolarized C NMR Spectroscopy.
Anal Chem
January 2025
Institute for Bioengineering of Catalonia (IBEC), Barcelona Institute of Science and Technology (BIST), 08028 Barcelona, Spain.
Nuclear magnetic resonance (NMR) spectroscopy is a valuable diagnostic tool limited by low sensitivity due to low nuclear spin polarization. Hyperpolarization techniques, such as dissolution dynamic nuclear polarization, significantly enhance sensitivity, enabling real-time tracking of cellular metabolism. However, traditional high-field NMR systems and bioreactor platforms pose challenges, including the need for specialized equipment and fixed sample volumes.
View Article and Find Full Text PDFStable Air Plastron Prolongs Biofluid Repellency of Submerged Superhydrophobic Surfaces.
Langmuir
January 2025
School of Chemical Engineering, Department of Chemistry and Materials Science, Aalto University, Tietotie 3 Espoo 02150, Finland.
Superhydrophobic surfaces find applications in numerous biomedical scenarios, requiring the repellence of biofluids and biomolecules. Plastron, the trapped air between a superhydrophobic surface and a wetting liquid, plays a pivotal role in biofluid repellency. A key challenge, however, is the often short-lived plastron stability in biofluids and the lack of knowledge surrounding it.
View Article and Find Full Text PDFThCTi@(6)-C: Th═C Double Bond in a Mixed Actinide-Transition Metal Cluster.
J Am Chem Soc
January 2025
College of Chemistry, Chemical Engineering and Materials Science, and State Key Laboratory of Radiation Medicine and Protection, Soochow University, Suzhou, Jiangsu 215123, P. R. China.
A thorium-carbon double bond that corresponds to the sum of theoretical covalent double bond radii has long been sought after in the study of actinide-ligand multiple bonding as a synthetic target. However, the stabilization of this chemical bond remains a great challenge to date, in part because of a relatively poor energetic matching between 5f-/6d- orbitals of thorium and the 2s-/2p- frontier orbitals of carbon. Herein, we report the successful synthesis of a thorium-carbon double bond in a carbon-bridged actinide-transition metal cluster, i.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!