Dispersion Compensation FIR Filter With Improved Robustness to Coefficient Quantization Errors

Sheikh, Alireza; Fougstedt, Christoffer; Amat, Alexandre Graell i; Johannisson, Pontus; Larsson-Edefors, Per; Karlsson, Magnus

doi:10.1109/jlt.2016.2599276

Cited by 24 publications

(21 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In FD the equalization exploits the computational efficiency of fast Fourier transform (FFT) in conjunction with overlap-save/add algorithms [3]. Aiming to reduce implementation the complexity, several optimized CDE algorithms have been proposed, including the use of sub-band processing [4] and coefficient quantization [5,6]. Exploiting the high multiplicity of the quantized TD-CDE coefficients, we have recently proposed a distributive FIR-CDE (D-FIR-CDE) algorithm for 100G PM-QPSK transmission systems, which was shown to yield significant complexity and latency gains [6].…”

Section: Introductionmentioning

confidence: 99%

Low-Complexity Chromatic Dispersion Equalizer for 400G Transmission Systems

Martins¹,

Amado²,

Rossi³

et al. 2017

Optical Fiber Communication Conference

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

Low-Complexity Chromatic Dispersion Equalizer for 400G Transmission Systems

Martins¹,

Amado²,

Rossi³

et al. 2017

Optical Fiber Communication Conference

View full text Add to dashboard Cite

show abstract

“…As a reference, we show the performance of linear equalization (red) and DBP with 1 StPS using frequency-domain filtering (blue). The linear equalizer uses LS-optimal coefficients with constrained out-of-band gain (LS-CO) [23]. LDBP achieves a peak SNR of 21.9 dB using 13 · 4 + 12 · 2 + 1 = 77 total taps.…”

Section: Resultsmentioning

confidence: 99%

“…For example, since the inverse Fourier transform of H(ω) can be computed analytically, filter coefficients may be obtained through direct sampling and truncation [19]. Other approaches include frequency-domain sampling (FDS) [5], wavelets [21], and least-squares (LS) [22], [23].…”

Section: Filter Design For Chromatic Dispersionmentioning

confidence: 99%

Deep Learning of the Nonlinear Schrödinger Equation in Fiber-Optic Communications

Häger

Pfister

2018

2018 IEEE International Symposium on Information Theory (ISIT)

View full text Add to dashboard Cite

An important problem in fiber-optic communications is to invert the nonlinear Schrödinger equation in real time to reverse the deterministic effects of the channel. Interestingly, the popular split-step Fourier method (SSFM) leads to a computation graph that is reminiscent of a deep neural network. This observation allows one to leverage tools from machine learning to reduce complexity. In particular, the main disadvantage of the SSFM is that its complexity using M steps is at least M times larger than a linear equalizer. This is because the linear SSFM operator is a dense matrix. In previous work, truncation methods such as frequency sampling, wavelets, or least-squares have been used to obtain "cheaper" operators that can be implemented using filters. However, a large number of filter taps are typically required to limit truncation errors. For example, Ip and Kahn showed that for a 10 Gbaud signal and 2000 km optical link, a truncated SSFM with 25 steps would require 70-tap filters in each step and 100 times more operations than linear equalization. We find that, by jointly optimizing all filters with deep learning, the complexity can be reduced significantly for similar accuracy. Using optimized 5-tap and 3-tap filters in an alternating fashion, one requires only around 2-6 times the complexity of linear equalization, depending on the implementation.

show abstract

“…Here, we use a one-step-per-span (1-StPS) TD-DBP algorithm. EachD uses the least-squares constrained-optimization (LS-CO) filter [11] to compensate for accumulated dispersion corresponding to the step length, with in-band response optimized with respect to the pulse-shaped spectrum.…”

Section: A System Contextmentioning

confidence: 99%

ASIC Implementation of Time-Domain Digital Back Propagation for Coherent Receivers

Fougstedt

Svensson

Mazur

et al. 2018

IEEE Photon. Technol. Lett.

Self Cite

View full text Add to dashboard Cite

Digital back propagation (DBP) is often proposed and implemented offline for mitigation of nonlinear impairments in long-haul fiber communications. However, complexity in terms of chip area and power consumption in a realistic ASIC implementation is yet to be determined. Here, we implement Time-Domain DBP (TD-DBP) in a 28-nm FD-SOI process technology, considering digital implementation aspects such as limited-resolution arithmetic and finite-length filters. We choose as example a coherent optical transmission system, viz. a singlechannel, single-polarization, 20-GBd 16-QAM system, for which DBP is known to perform well. For the considered system, we find that TD-DBP can enable a reach increase from 3 400 to 5 300 km, at a power dissipation of < 20 W (or, conversely, an energy dissipation of < 230 pJ/bit), at a pre-FEC BER of 10 −2 .

show abstract

Dispersion Compensation FIR Filter With Improved Robustness to Coefficient Quantization Errors

Cited by 24 publications

References 14 publications

Low-Complexity Chromatic Dispersion Equalizer for 400G Transmission Systems

Low-Complexity Chromatic Dispersion Equalizer for 400G Transmission Systems

Deep Learning of the Nonlinear Schrödinger Equation in Fiber-Optic Communications

ASIC Implementation of Time-Domain Digital Back Propagation for Coherent Receivers

Contact Info

Product

Resources

About