DOAJ Open Access 2021

A Fiber Nonlinearity Compensation Scheme With Complex-Valued Dimension-Reduced Neural Network

Pinjing He Feilong Wu Meng Yang Aiying Yang Peng Guo +2 lainnya

Abstrak

A fiber nonlinearity compensation scheme based on a complex-valued dimension-reduced neural network is proposed. The proposed scheme performs all calculations in complex values and employs a dimension-reduced triplet feature vector to reduce the size of the input layer. Simulation and experiment results show that the proposed neural network needed only 20% of computational complexity to reach the saturated performance gain of the real-valued triplet-input neural network, and had a similar saturated gain to the one-step-per-span digital backpropagation. In addition, the proposed scheme was 1.7 dB more robust to the noise from training data and required less bit precision for quantizing trained weights, compared with the real-valued triplet-input neural network.

Topik & Kata Kunci

Applied optics. Photonics Optics. Light

Penulis (7)

Pinjing He

Feilong Wu

Meng Yang

Aiying Yang

Peng Guo

Yaojun Qiao

Xiangjun Xin

Format Sitasi

APA MLA BibTeX

He, P., Wu, F., Yang, M., Yang, A., Guo, P., Qiao, Y. et al. (2021). A Fiber Nonlinearity Compensation Scheme With Complex-Valued Dimension-Reduced Neural Network. https://doi.org/10.1109/JPHOT.2021.3123624

Akses Cepat

PDF tidak tersedia langsung

Cek di sumber asli →

Lihat di Sumber doi.org/10.1109/JPHOT.2021.3123624

Informasi Jurnal

Tahun Terbit: 2021
Sumber Database: DOAJ
DOI: 10.1109/JPHOT.2021.3123624
Akses: Open Access ✓