OptiComm-GPT: a GPT-based versatile research assistant for optical fiber communication systems.

With the increasing capacity and complexity of optical fiber communication systems, both academic and industrial requirements for the essential tasks of transmission systems simulation, digital signal processing (DSP) algorithms verification, system performance evaluation, and quality of transmission (QoT) optimization are becoming significantly important. However, due to the intricate and nonlinear nature of optical fiber communication systems, these tasks are generally implemented in a divide-and-conquer manner, which necessitates a profound level of expertise and proficiency in software programming from researchers or engineers. To lower this threshold and facilitate professional research easy-to-start, a GPT-based versatile research assistant named OptiComm-GPT is proposed for optical fiber communication systems, which flexibly and automatically performs system simulation, DSP algorithms verification, performance evaluation, and QoT optimization with only natural language. To enhance OptiComm-GPT's abilities for complex tasks in optical fiber communications and improve the accuracy of generated results, a domain information base containing rich domain knowledge, tools, and data as well as the comprehensive prompt engineering with well-crafted prompt elements, techniques, and examples is established and performs under a LangChain-based framework. The performance of OptiComm-GPT is evaluated in multiple simulation, verification, evaluation, and optimization tasks, and the generated results show that OptiComm-GPT can effectively comprehend the user's intent, accurately extract system parameters from the user's request, and intelligently invoke domain resources to solve these complex tasks simultaneously. Moreover, the statistical results, typical errors, and running time of OptiComm-GPT are also investigated to illustrate its practical reliability, potential limitations, and further improvements.

Real-time demonstration of two-aperture coherent digital combining free-space optical transmission with a real-valued MIMO adaptive equalizer.

Compared with the single-aperture system, the multi-aperture coherent digital combining system has the technical advantage of the effective mitigation of deep fading under strong turbulence, ease of scalability, and potential higher collected optical power. However, the tricky problem of a multi-aperture system is to efficiently combine multiple branch signals with a static skew mismatch and with time-varying characteristics of received power scintillation. In this Letter, a real-valued massive array multiple-input multiple-output (MIMO) adaptive equalizer is proposed for the first time to our knowledge to realize multi-aperture channel equalization and combining, simultaneously. In the proof-of-principle system, the feasibility of the combining technique is verified based on a MIMO 4 × 2 equalizer in a 2.5-GBaud data rate QPSK modulation FPGA-based two-aperture coherent receiver with a dynamic turbulence simulator. The results show that no reduction in combining efficiency is observed under static turbulence conditions at the hard-decision forward error correction (HD-FEC) limit of 3.8 × 0-3, and combining efficiencies of 95% and 88% are obtained for the dynamic moderate and strong turbulence.

Blind skew compensation and digital combining with widely linear equalizer for multi-aperture coherent FSO communication.

Coherent digital combining technology using multiple small apertures has a lot of advantages over doing so with a single large aperture, including the effective mitigation of deep fading under strong turbulence, ease of scalability, and potential higher collected optical power. However, the in-phase/quadrature (I/Q) imbalance and I/Q skew induced by manufacturing imperfections of the coherent receiver front end, and the time mismatch caused by the unequal length of multi-aperture branches will induce a high OSNR penalty and reduce the digital combining efficiency, especially when the system scales to a larger number of apertures, such as massive aperture system. In this work, a complex-valued multiple-input multiple-output (MIMO) 4N×2 widely linear (WL) equalizer is designed to combine multi-aperture signals. Using WL complex analysis, a general analytical model is derived and it is indicated that multi-aperture channel equalization and combining operations can be achieved simultaneously using a MIMO equalizer as long as appropriate tap coefficients are selected. Moreover, the feasibility of the proposed WL equalizer is verified by a 10-Gbps PM-QPSK modulation and a 20-Gbps PM-16QAM modulation four-aperture offline simulated turbulence experiment. The four-aperture combining efficiency of PM-QPSK exceeds 96% even at a single-aperture extremely low OSNR of -6 dB, and 80% for PM-16QAM at a single-aperture OSNR of 0 dB.

Simple amplifier configuration algorithm for dynamic C + L-band systems in the presence of stimulated Raman scattering.

We propose a simple two-step amplifier configuration algorithm based on signal power across different channels to improve the generalized signal-to-noise ratio (GSNR) performance of dynamic C + L-band links in the presence of amplifier spontaneous emission (ASE) noise, Kerr nonlinearity, and stimulated Raman scattering (SRS) using erbium-doped fiber amplifiers (EDFA). In step 1, ASE noise and Kerr nonlinearity are taken into account to derive sub-optimal signal power profiles at the beginning of each span using the local optimization global optimization (LOGO) strategy. The effect of SRS is compensated through amplifier gain pre-tilt in step 2. Simulations for links with homogeneous/heterogeneous spans, static full-channel loading, and dynamic loading due to gradual channel additions for C + L-band upgrades show that the proposed algorithm can achieve similar GSNR performance, but requires much less execution time, compared to other iterative methods that target for improving the GSNR across the C + L band, thus making it a fast and efficient GSNR management strategy for future dynamic C + L-band networks.

AI-assisted intent-based traffic grooming in a dynamically shared 5g optical fronthaul network.

The extremely high number of services with large bandwidth requirements and the increasingly dynamic traffic patterns of cell sites pose major challenges to optical fronthaul networks, rendering them incapable of coping with the extensive, uneven, and real-time traffic that will be generated in the future. In this paper, we first present the design of an adaptive graph convolutional network with gated recurrent unit (AGCN-GRU) network to learn the temporal and spatial dependencies of traffic patterns of cell sites to provide accurate traffic predictions, in which the AGCN model can capture potential spatial relations according to the similarity of network traffic patterns in different areas. Then, we innovatively consider how to deal with the unpredicted burst traffic and propose an AI-assisted intent-based traffic grooming scheme to realise automatic and intelligent cell sites clustering and traffic grooming. Finally, a software-defined testbed for 5G optical fronthaul network was established, on which the proposed schemes were deployed and evaluated by considering traffic datasets of existing optical networks. The experimental results showed that the proposed scheme can optimize network resource allocation, increase the average efficient resource utilization and reduce the average delay and the rejection ratio.

Cause-aware failure detection using an interpretable XGBoost for optical networks.

Failure detection is an important part of failure management, and network operators encounter serious consequences when operating under failure conditions. Machine learning (ML) is widely applied in the failure management of optical networks, where neural networks (NNs) have particularly attracted considerable attention and become the most extensively applied algorithm among all MLs. However, the black-box nature of NN makes it difficult to interpret or analyze why and how NNs work during execution. In this paper, we propose a cause-aware failure detection scheme for optical transport network (OTN) boards, adopting the interpretable extreme gradient boosting (XGBoost) algorithm. According to the feature importance ranking by XGBoost, the high-relevance features with the equipment failure are found. Then, SHapley Additive exPlanations (SHAP) is applied to solve the inconsistency of feature attribution under three common global feature importance measurement parameters of XGBoost, and can obtain a consistent feature attribution by calculating the contribution (SHAP value) of each input feature to detection result of XGBoost. Based on the feature importance ranking of SHAP values, the features most related to two types of OTN board failures are confirmed, enabling the identification of failure causes. Moreover, we evaluate the failure detection performance for two types of OTN boards, in which the practical data are balanced and unbalanced respectively. Experimental results show that the F1 score of the two types of OTN boards based on the proposed scheme is higher than 98%, and the most relevant features of the two types of board failures are confirmed based on SHAP value, which are the average and maximum values of the environment temperature, respectively.

Topology design of digital metamaterials for ultra-compact integrated photonic devices based on mode manipulation.

Precise manipulation of mode order in silicon waveguides plays a fundamental role in the on-chip all-optical interconnections and is still a tough task in design when the functional region is confined to a subwavelength footprint. In this paper, digital metamaterials consisting of silicon and air pixels are topologically designed by an efficient method combining 2D finite element method for optical simulations, density method for material description and method of moving asymptotes for optimization. Only around 150 iterations are required for searching satisfactory solutions. Six high-quality and efficient conversions between four TE-polarized modes are achieved in a functional region with footprint 0.645λ 2 (center wavelength λ = 1550 nm). Based on asymmetric mode conversion, a reciprocal optical diode with high contrast ratio is further obtained with the optimization starting from TE0-to-TE1 mode converter. Moreover, we successfully design a 1 × 2 demultiplexer with footprint 1.0λ 2 and demonstrate a simple mode division multiplexing system with satisfactory performances. Finally, by changing the refractive index to an equivalent value, quasi-3D designs are obtained and the functionalities are validated in 3D simulations for both free-standing and SOI configurations.

Digital twin-enabled self-evolved optical transceiver using deep reinforcement learning.

Due to their high flexibility, programmable optical transceivers (POT) are regarded as one of the key optical components in optical fiber communications, where diverse transceiver freedom degrees can be controlled according to real-time network states. However, the adaptivity of classic POT modeling and controlling is limited to the prior-knowledge-dependent quality of the transmission estimation model or uncomprehensive training dataset, which has great difficulties in enabling adaptive POT modeling and controlling to evolve with time-varied network states. Here, a powerful dynamic modeling technique called digital twin (DT), enabled by the deep reinforcement learning (DRL), is first proposed for the adaptive POT modeling and controlling, to the best of our knowledge. The experimental and simulation results show that the lowest spectrum consumption and minimum latency are both available in the proposed POT, compared with the classic POTs based on neural networks and maximum capability provisioning. We believe that the proposed DT will open a new avenue for the adaptive optical component modeling and controlling for dynamic optical networks.

Deep learning based adaptive sequential data augmentation technique for the optical network traffic synthesis.

The lack of the sufficient and diverse training data is one of the main challenges limiting performances of the machine learning enabled applications in optical networks. Here, we propose a deep learning based sequential data augmentation technique for the aggregate traffic data augmentation for diverse optical network scenarios. A generative adversarial network (GAN) model is trained with the experimental traffic data to automatically extract the substantial characteristics of the experimental traffic data through the zero-sum game theory and then augment the traffic data adaptively. The statistical evaluation parameters of the augmented traffic are mean, variance and Hurst exponent. To add comparisons, two other classical generative models including the statistical parameter configuration (SPC) model and the variational autoencoder (VAE) model are also adopted to generate the traffic data that are similar to the actual traffic data. The comprehensive comparisons among the proposed GAN, the SPC and VAE show that the performances of the GAN exceed those of the SPC and the VAE obviously. The mean and the variance of the augmented traffic data from the GAN are almost equal to those of the experimental traffic data, where the average deviations are both within 2%. The Hurst exponent of the augmented traffic data from the GAN is respectively near 90% and 96% of those of the experimental traffic data in the access network and the core network. To estimate the similarity intuitively, the well-known k-mean algorithm is used to cluster the augmented traffic data according to the centroids determined by the corresponding experimental traffic data and the clustering accuracies are all higher than 95% for 6 kinds of typical traffic types in the optical networks. These results demonstrate that the proposed GAN is able to effectively generate the traffic data that is very close to the experimental traffic data and is difficult to be distinguished for diverse traffic types. Moreover, a relatively small dataset with a few hundred pieces of experimental traffic data is required and the amount of the augmented traffic data from the GAN is unlimited in theory, which can be augmented as much as we need. The proposed traffic data augmentation technique also has the potential to be utilized in other sequential data augmentation applications for the optical networks.

Cost-effective and data size-adaptive OPM at intermediated node using convolutional neural network-based image processor.

A cost-effective and data size-adaptive optical performance monitoring (OPM) scheme is proposed, which is based on asynchronous delay-tap plot (ADTP) using convolutional neural network (CNN) from the perspective of image processing. First, we design an OPM framework, based on the electrical domain-processing technique for the future optical networks. These networks include coherent detection-based end-to-end channel monitoring at destination node and direct detection-based transmission link monitoring at intermediate node. Aiming at the link monitoring, CNN is applied to recognize and analyze ADTP images that are converted from two-dimension (2D) digital vectors, so that adaptive to the stable algorithm structure. In simulation system, three high-order modulation formats, 16 quadrature amplitude modulation (QAM), 32QAM, 64QAM, are investigated for optical signal-to-noise ratio (OSNR) estimation and modulation format identification (MFI). The 100% accuracies under different chromatic dispersions (CDs) at different iteration epochs are obtained. Compared with asynchronous amplitude histograms (AAH)-based method, the better accuracy and faster convergence rate are achieved, especially in terms of strong CDs. Additionally, the experimental system is also conducted of 16QAM and 64QAM signals. Based on the partially-trained CNN model from simulation, the OSNR estimation accuracies of 16QAM and 64QAM are 97.81% and 96.56%, respectively. The maximum standard deviation is less than 0.45 dB and the MFI accuracies is 99.84%, presenting the satisfactory results and proving the feasibility of ADTP-based image processor for link monitoring at intermediate nodes.

Joint atmospheric turbulence detection and adaptive demodulation technique using the CNN for the OAM-FSO communication.

A novel joint atmospheric turbulence (AT) detection and adaptive demodulation technique based on convolutional neural network (CNN) are proposed for the OAM-based free-space optical (FSO) communication. The AT detecting accuracy (ATDA) and the adaptive demodulating accuracy (ADA) of the 4-OAM, 8-OAM, 16-OAM FSO communication systems over computer-simulated 1000-m turbulent channels with 4, 6, 10 kinds of classic ATs are investigated, respectively. Compared to previous approaches using the self-organizing mapping (SOM), deep neural network (DNN) and other CNNs, the proposed CNN achieves the highest ATDA and ADA due to the advanced multi-layer representation learning without feature extractors designed carefully by numerous experts. For the AT detection, the ATDA of CNN is near 95.2% for 6 kinds of typical ATs, in cases of both weak and strong ATs. For the adaptive demodulation of optical vortices (OV) carrying OAM modes, the ADA of CNN is about 99.8% for the 8-OAM system over the computer-simulated 1000-m free-space strong turbulent link. In addition, the effects of image resolution, iteration number, activation functions and the structure of the CNN are also studied comprehensively. The proposed technique has the potential to be embedded in charge-coupled device (CCD) cameras deployed at the receiver to improve the reliability and flexibility for the OAM-FSO communication.

Multifunctional all-optical signal processing scheme for wavelength-division-multiplexing multicast, wavelength conversion, format conversion, and all-optical encryption using hybrid modulation format exclusive-OR gates based on four-wave mixing in highly nonlinear fiber.

A multifunctional optical signal processing scheme for 10-Gbaud quadrature phase shift keying (QPSK) and binary phase shift keying (BPSK) signals based on four-wave mixing (FWM) in highly nonlinear fiber (HNLF) is proposed. Wavelength-division-multiplexing (WDM) multicast, wavelength conversion, modulation format conversion, and hybrid modulation format exclusive-OR (HMF-XOR) logic gates are realized simultaneously by simulation. One-to-three WDM multicast of 20-Gbps QPSK signals are achieved paying the optical signal-to-noise ratio (OSNR) penalties <0.8 dB at the bit error rate (BER) of 10-3. The converted BPSK signals with high performance are obtained; they are generated from wavelength conversion of 10-Gbps BPSK and format conversion from 20-Gpbs QPSK. In the case of HFM-XOR logic gates, we discuss the extension application in all-optical encryption. Therefore, both theoretical analysis and simulation results are conducted to analyze the feasibility of the dual-channel all-optical encryption for QPSK and BPSK signals. The OSNR penalties of decrypted QPSK and BPSK signals at the BER of 10-3 are 1 dB and 0.7 dB, respectively. The concentrated constellations and clear eye diagrams indicate the high performance of decrypted signals.

Failure prediction using machine learning and time series in optical network.

In this paper, we propose a performance monitoring and failure prediction method in optical networks based on machine learning. The primary algorithms of this method are the support vector machine (SVM) and double exponential smoothing (DES). With a focus on risk-aware models in optical networks, the proposed protection plan primarily investigates how to predict the risk of an equipment failure. To the best of our knowledge, this important problem has not yet been fully considered. Experimental results showed that the average prediction accuracy of our method was 95% when predicting the optical equipment failure state. This finding means that our method can forecast an equipment failure risk with high accuracy. Therefore, our proposed DES-SVM method can effectively improve traditional risk-aware models to protect services from possible failures and enhance the optical network stability.

Bit-based support vector machine nonlinear detector for millimeter-wave radio-over-fiber mobile fronthaul systems.

An effective bit-based support vector machine (SVM) is proposed as a non-parameter nonlinear mitigation approach in the millimeter-wave radio-over-fiber (RoF) mobile fronthaul (MFH) system for various modulation formats. First, we analyze the impairments originated from nonlinearities in the millimeter-wave RoF system. Then we introduce the operation principle of the bit-based SVM detector. As a classifier, the SVM can create nonlinear decision boundaries by kernel function to mitigate the distortions caused by both linear and nonlinear noise. In our design, SVM can learn and capture the link characteristics from only a few training data without requiring the prior estimation of the system link. The bit-based SVM only needs log2M SVMs to detect the signal of M-order modulation format. Experimental results have been obtained to verify the feasibility of the proposed method. The sensitivities are improved by 1.2-dB for 16-QAM, 1.3-dB for 64-QAM, 1.8-dB for 16-APSK and 1.3-dB for 32-APSK at BER = 1E-3 with SVM detector, respectively. The proposed bit-based SVM gains a large improvement in the nonlinear system tolerance and outperforms the system employing k-means algorithm.

Intelligent constellation diagram analyzer using convolutional neural network-based deep learning.

An intelligent constellation diagram analyzer is proposed to implement both modulation format recognition (MFR) and optical signal-to-noise rate (OSNR) estimation by using convolution neural network (CNN)-based deep learning technique. With the ability of feature extraction and self-learning, CNN can process constellation diagram in its raw data form (i.e., pixel points of an image) from the perspective of image processing, without manual intervention nor data statistics. The constellation diagram images of six widely-used modulation formats over a wide OSNR range (15~30 dB and 20~35 dB) are obtained from a constellation diagram generation module in oscilloscope. Both simulation and experiment are conducted. Compared with other 4 traditional machine learning algorithms, CNN achieves the better accuracies and is obviously superior to other methods at the cost of O(n) computation complexity and less than 0.5 s testing time. For OSNR estimation, the high accuracies are obtained at epochs of 200 (95% for 64QAM, and over 99% for other five formats); for MFR, 100% accuracies are achieved even with less training data at lower epochs. The experimental results show that the OSNR estimation errors for all the signals are less than 0.7 dB. Additionally, the effects of multiple factors on CNN performance are comprehensively investigated, including the training data size, image resolution, and network structure. The proposed technique has the potential to be embedded in the test instrument to perform intelligent signal analysis or applied for optical performance monitoring.

Multifunctional switching unit for add/drop, wavelength conversion, format conversion, and WDM multicast based on bidirectional LCoS and SOA-loop architecture.

We propose a multifunctional optical switching unit based on the bidirectional liquid crystal on silicon (LCoS) and semiconductor optical amplifier (SOA) architecture. Add/drop, wavelength conversion, format conversion, and WDM multicast are experimentally demonstrated. Due to the bidirectional characteristic, the LCoS device cannot only multiplex the input signals, but also de-multiplex the converted signals. Dual-channel wavelength conversion and format conversion from 2 × 25Gbps differential quadrature phase-shift-keying (DQPSK) to 2 × 12.5Gbps differential phase-shift-keying (DPSK) based on four-wave mixing (FWM) in SOA is obtained with only one pump. One-to-six WDM multicast of 25Gbps DQPSK signals with two pumps is also achieved. All of the multicast channels are with a power penalty less than 1.1 dB at FEC threshold of 3.8 × 10⁻³.

Simultaneous multichannel wavelength multicasting and XOR logic gate multicasting for three DPSK signals based on four-wave mixing in quantum-dot semiconductor optical amplifier.

In this paper, we experimentally demonstrate simultaneous multichannel wavelength multicasting (MWM) and exclusive-OR logic gate multicasting (XOR-LGM) for three 10Gbps non-return-to-zero differential phase-shift-keying (NRZ-DPSK) signals in quantum-dot semiconductor optical amplifier (QD-SOA) by exploiting the four-wave mixing (FWM) process. No additional pump is needed in the scheme. Through the interaction of the input three 10Gbps DPSK signal lights in QD-SOA, each channel is successfully multicasted to three wavelengths (1-to-3 for each), totally 3-to-9 MWM, and at the same time, three-output XOR-LGM is obtained at three different wavelengths. All the new generated channels are with a power penalty less than 1.2dB at a BER of 10(-9). Degenerate and non-degenerate FWM components are fully used in the experiment for data and logic multicasting.