arXiv subjects
Publications and source records attributed to Antonio Napoli.
We analyze polarization data streamed from a commercial coherent transponder, used in MedNautilus link, during 2023 Turkey earthquakes. We show how earthquake features and cable geometry fundamentally govern the optical polarization response, taking the first steps for detailed interpretation in designing fiber-based seismic sensing systems.
State-of-Polarization sensing with coherent transponders enables wide-area geophysical monitoring over existing submarine cables, but its performance is limited by polarization noise from both the fiber and terminal hardware. This work investigates how terminal noise affects polarization rotation estimates derived from receiver equalizer coefficients and how it can obscure stochastic polarization drift used for sensing. We analyze Jones-matrix time series from two deployed receivers in the Sparkle Mediterranean link MedNautilus (approximately 2000 km and 450 km) and compare them with a laboratory back-to-back reference. Power Spectral Density (PSD) analysis reveals a low-frequency random-walk regime and a high-frequency white-noise floor, separated by a link-dependent corner frequency. The rotation innovation variance increases with link length, while the longest field link also shows elevated white noise consistent with accumulated amplifier and terminal contributions. Additionally, harmonic spectral components are observed, indicating a transponder-related artifact that should be considered in practical sensing applications.
The convergence of metro and access networks into unified optical infrastructures requires cost-effective alternatives to conventional solid-core fiber (SCF). This paper examines using hollow-core fiber (HCF) with digital subcarrier multiplexing (DSCM) transceivers in horseshoe-and-spur filterless optical architectures. Leveraging HCF's ultra-low nonlinearity, we optimize amplifier placement to maximize power budgets under realistic constraints. Our results show that HCF shifts the main limitation from nonlinearity to amplifier output power, enabling up to a 20 dB spur power-budget gain over SCF. Considering balanced and unbalanced couplers, we find that increasing amplifier density boosts reach only up to a saturation point (about 11-13 units in a 5-transit-node network). A techno-economic break-even analysis of hybrid SCF-HCF deployments shows that targeted HCF use provides intermediate performance gains and can fully recover its fiber cost premium through amplifier reductions.
We propose a hybrid active-online learning framework for label-efficient concept drift adaptation in optical network failure detection. Using margin-based selective labeling, our method achieves nearceiling accuracy and AUC scores while querying only 3.4% of streaming samples, with negligible latency overhead compared to static inference.
We present a novel online learning-based approach for concept drift adaptation in optical network failure detection, achieving up to a 70% improvement in performance over conventional static models while maintaining low latency.
Machine learning-based failure management in optical networks has gained significant attention in recent years. However, severe class imbalance, where normal instances vastly outnumber failure cases, remains a considerable challenge. While pre- and in-processing techniques have been widely studied, post-processing methods are largely unexplored. In this work, we present a direct comparison of pre-, in-, and post-processing approaches for class imbalance mitigation in failure detection and identification using an experimental dataset. For failure detection, post-processing methods-particularly Threshold Adjustment-achieve the highest F1 score improvement (up to 15.3%), while Random Under-Sampling provides the fastest inference. In failure identification, GenAI methods deliver the most substantial performance gains (up to 24.2%), whereas post-processing shows limited impact in multi-class settings. When class overlap is present and latency is critical, over-sampling methods such as the SMOTE are most effective; without latency constraints, Meta-Learning yields the best results. In low-overlap scenarios, Generative AI approaches provide the highest performance with minimal inference time.
We compare pre-, in-, and post-processing techniques for class imbalance mitigation in optical network failure detection. Threshold Adjustment achieves the highest F1 gain (15.3%), while Random Under-sampling (RUS) offers the fastest inference, highlighting a key performance-complexity trade-off.
The extension of elastic optical networks (EON) to multi-band transmission (MB-EON) shows promise in enhancing spectral efficiency, throughput, and long-term cost-effectiveness for telecom operators. However, designing MB-EON networks introduces complex challenges, notably the optimization of physical parameters like optical power and quality of transmission (QoT). Frequency-dependent characteristics of fiber, such as loss, dispersion, and nonlinear effects, alongside inter-channel stimulated Raman scattering, pose significant hurdles when extending beyond the L+C (LC) band to a continuous spectrum over 100 nm. In this study, we propose a span-by-span methodology for optimal power allocation, introducing two hyper-accelerated power optimization (HPO) strategies: flat launch power (FLP) and flat received power (FRP). These approaches significantly expedite network power optimization while preserving the stability of running services. Our comparative analysis of FLP and FRP models reveals that while FRP has a minimal effect on capacity (increasing less than 10 Tbps for an L+C+S (LCS) system over 100 km), it improves flatness and GSNR/OSNR metrics in the S-band by approximately 2/0 dB and 2.5/6 dB, respectively. A network-wide analysis across various topologies shows that the FRP technique enhances minimum GSNR, contributing to a throughput increase of 12% to 75%, depending on network scale, at a 1% bandwidth blocking rate. Lastly, our application of HPO in MB-EON for both local and global power optimization demonstrates that while both approaches offer comparable performance, global optimization is simpler and more cost-effective for large-scale networks.
In this paper, we provide a systematic approach for assessing and comparing the computational complexity of neural network layers in digital signal processing. We provide and link four software-to-hardware complexity measures, defining how the different complexity metrics relate to the layers' hyper-parameters. This paper explains how to compute these four metrics for feed-forward and recurrent layers, and defines in which case we ought to use a particular metric depending on whether we characterize a more soft- or hardware-oriented application. One of the four metrics, called `the number of additions and bit shifts (NABS)', is newly introduced for heterogeneous quantization. NABS characterizes the impact of not only the bitwidth used in the operation but also the type of quantization used in the arithmetical operations. We intend this work to serve as a baseline for the different levels (purposes) of complexity estimation related to the neural networks' application in real-time digital signal processing, aiming at unifying the computational complexity estimation.
For the first time, multi-task learning is proposed to improve the flexibility of NN-based equalizers in coherent systems. A "single" NN-based equalizer improves Q-factor by up to 4 dB compared to CDC, without re-training, even with variations in launch power, symbol rate, or transmission distance.
The quantization of neural networks for the mitigation of the nonlinear and components' distortions in dual-polarization optical fiber transmission is studied. Two low-complexity neural network equalizers are applied in three 16-QAM 34.4 GBaud transmission experiments with different representative fibers. A number of post-training quantization and quantization-aware training algorithms are compared for casting the weights and activations of the neural network in few bits, combined with the uniform, additive power-of-two, and companding quantization. For quantization in the large bit-width regime of $\geq 5$ bits, the quantization-aware training with the straight-through estimation incurs a Q-factor penalty of less than 0.5 dB compared to the unquantized neural network. For quantization in the low bit-width regime, an algorithm dubbed companding successive alpha-blending quantization is suggested. This method compensates for the quantization error aggressively by successive grouping and retraining of the parameters, as well as an incremental transition from the floating-point representations to the quantized values within each group. The activations can be quantized at 8 bits and the weights on average at 1.75 bits, with a penalty of $\leq 0.5$~dB. If the activations are quantized at 6 bits, the weights can be quantized at 3.75 bits with minimal penalty. The computational complexity and required storage of the neural networks are drastically reduced, typically by over 90\%. The results indicate that low-complexity neural networks can mitigate nonlinearities in optical fiber transmission.
In this paper, we investigate the use of the learned digital back-propagation (LDBP) for equalizing dual-polarization fiber-optic transmission in dispersion-managed (DM) links. LDBP is a deep neural network that optimizes the parameters of DBP using the stochastic gradient descent. We evaluate DBP and LDBP in a simulated WDM dual-polarization fiber transmission system operating at the bitrate of 256 Gbit/s per channel, with a dispersion map designed for a 2016 km link with 15% residual dispersion. Our results show that in single-channel transmission, LDBP achieves an effective signal-to-noise ratio improvement of 6.3 dB and 2.5 dB, respectively, over linear equalization and DBP. In WDM transmission, the corresponding $Q$-factor gains are 1.1 dB and 0.4 dB, respectively. Additionally, we conduct a complexity analysis, which reveals that a frequency-domain implementation of LDBP and DBP is more favorable in terms of complexity than the time-domain implementation. These findings demonstrate the effectiveness of LDBP in mitigating the nonlinear effects in DM fiber-optic transmission systems.
To reduce the complexity of the hardware implementation of neural network-based optical channel equalizers, we demonstrate that the performance of the biLSTM equalizer with approximated activation functions is close to that of the original model.
In this work, we demonstrate the offline FPGA realization of both recurrent and feedforward neural network (NN)-based equalizers for nonlinearity compensation in coherent optical transmission systems. First, we present a realization pipeline showing the conversion of the models from Python libraries to the FPGA chip synthesis and implementation. Then, we review the main alternatives for the hardware implementation of nonlinear activation functions. The main results are divided into three parts: a performance comparison, an analysis of how activation functions are implemented, and a report on the complexity of the hardware. The performance in Q-factor is presented for the cases of bidirectional long-short-term memory coupled with convolutional NN (biLSTM + CNN) equalizer, CNN equalizer, and standard 1-StpS digital back-propagation (DBP) for the simulation and experiment propagation of a single channel dual-polarization (SC-DP) 16QAM at 34 GBd along 17x70km of LEAF. The biLSTM+CNN equalizer provides a similar result to DBP and a 1.7 dB Q-factor gain compared with the chromatic dispersion compensation baseline in the experimental dataset. After that, we assess the Q-factor and the impact of hardware utilization when approximating the activation functions of NN using Taylor series, piecewise linear, and look-up table (LUT) approximations. We also show how to mitigate the approximation errors with extra training and provide some insights into possible gradient problems in the LUT approximation. Finally, to evaluate the complexity of hardware implementation to achieve 200G and 400G throughput, fixed-point NN-based equalizers with approximated activation functions are developed and implemented in an FPGA.
To circumvent the non-parallelizability of recurrent neural network-based equalizers, we propose knowledge distillation to recast the RNN into a parallelizable feedforward structure. The latter shows 38\% latency decrease, while impacting the Q-factor by only 0.5dB.
In this paper, a new methodology is proposed that allows for the low-complexity development of neural network (NN) based equalizers for the mitigation of impairments in high-speed coherent optical transmission systems. In this work, we provide a comprehensive description and comparison of various deep model compression approaches that have been applied to feed-forward and recurrent NN designs. Additionally, we evaluate the influence these strategies have on the performance of each NN equalizer. Quantization, weight clustering, pruning, and other cutting-edge strategies for model compression are taken into consideration. In this work, we propose and evaluate a Bayesian optimization-assisted compression, in which the hyperparameters of the compression are chosen to simultaneously reduce complexity and improve performance. In conclusion, the trade-off between the complexity of each compression approach and its performance is evaluated by utilizing both simulated and experimental data in order to complete the analysis. By utilizing optimal compression approaches, we show that it is possible to design an NN-based equalizer that is simpler to implement and has better performance than the conventional digital back-propagation (DBP) equalizer with only one step per span. This is accomplished by reducing the number of multipliers used in the NN equalizer after applying the weighted clustering and pruning algorithms. Furthermore, we demonstrate that an equalizer based on NN can also achieve superior performance while still maintaining the same degree of complexity as the full electronic chromatic dispersion compensation block. We conclude our analysis by highlighting open questions and existing challenges, as well as possible future research directions.
A convolutional neural network is proposed to mitigate fiber transmission effects, achieving a five-fold reduction in trainable parameters compared to alternative equalizers, and 3.5 dB improvement in MSE compared to DBP with comparable complexity.
For the first time, recurrent and feedforward neural network-based equalizers for nonlinearity compensation are implemented in an FPGA, with a level of complexity comparable to that of a dispersion equalizer. We demonstrate that the NN-based equalizers can outperform a 1 step-per-span DBP.