Nonlinear Impairments Research Articles

Deep Learning-Based SC-FDMA Channel Equalization

It is very challenging to design an effective wireless communication system. That’s because of numerous factors affecting the performance of a typical wireless communication system, such as nonlinear channel distortions and impairments. single carrier frequency division multiple access (SC-FDMA) is a multiple access scheme that is an important part of the long-term evolution (LTE) standard for uplink transmission. An advanced mobile radio system’s multiple access schemes should indeed meet stringent requirements, such as a low bit error rate (BER). In this article, we investigate the equalization problem for nonlinear channel distortions and impairments using deep neural networks (NN). We introduce a novel combined deep neural network channel equalization and symbol detection scheme based on a deep learning (DL) recurrent feedback (RF) long short-term memory (LSTM) neural network to achieve blind equalization and decoding for SC-FDMA systems without knowing the channel state information (CSI). To train the model efficiently, the training data is gathered by simulation, with channel effects and noise treated as a complete black box. CSI and constellation demapping are learned by a deep neural network (DNN) model. Then, the frequency-domain sequences that have been corrupted are implicitly equalized to get the broadcasted signal back. Our specified SC-FDMA system, which uses a quadrature phase-shift keying (QPSK) modulation method and the suggested Deep Learning-based model channel equalizer, performs better than the existing equalizers by an average of 1 to 4 dB at moderate signal-to-noise (SNR) ratios, according to simulation data. A complexity comparison between the proposed and the conventional equalizers was conducted in terms of training time, execution time, and number of operations. On combined channel equalization and symbol detection, the suggested system delivers state-of-the-art performance.

Physics-Informed Scattering Transform Network for Modulation Recognition in 5G Industrial Cognitive Communications Considering Nonlinear Impairments in Active Phased Arrays

Physics-Informed Scattering Transform Network for Modulation Recognition in 5G Industrial Cognitive Communications Considering Nonlinear Impairments in Active Phased Arrays

Deep photonic reservoir computing recurrent network

Deep neural networks usually process information through multiple hidden layers. However, most hardware reservoir computing recurrent networks only have one hidden reservoir layer, which significantly limits the capability of solving practical complex tasks. Here we show a deep photonic reservoir computing (PRC) architecture, which is constructed by cascading injection-locked semiconductor lasers. In particular, the connection between successive hidden layers is all optical, without any optical-electrical conversion or analog-digital conversion. The proof of concept PRC consisting of 4 hidden layers and a total of 320 interconnected neurons (80 neurons per layer) is demonstrated in experiment. The deep PRC is applied in solving the real-world problem of signal equalization in an optical fiber communication system. It is found that the deep PRC exhibits strong capability in compensating for the nonlinear impairment of optical fibers.

Deep Learning Equalizer Connected with Viterbi-Viterbi Algorithm for PAM D-Band Radio over Fiber Link.

D-band (110-170 GHz) has been regarded as a potential candidate for the future 6G wireless network because of its large available bandwidth. At present, the lack of electrical amplifiers operating in the high frequency band and the strong nonlinear effect, i.e., the D-band, are still important problems. Therefore, effective methods to mitigate the nonlinear issue resulting from the ROF link are indispensable, among of which machine learning is considered the most effective paradigm to model the nonlinear behavior due to its nonlinear active function and structure. In order to reduce the computation amount and burden, a novel deep learning neural network equalizer connected with typical mathematical frequency offset estimation (FOE) and carrier phase recovery (CPR) algorithms is proposed. We implement D-band 45 Gbaud PAM-4 and 20 Gbaud PAM-8 ROF transmission simulations, and the simulation results show that the real value neural network (RVNN) equalizer connected with the Viterbi-Viterbi algorithm exhibits better compensation ability for nonlinear impairment, especially when dealing with serious inter-symbol interference and nonlinear effects. In our experiment, we employ coherent detection to further improve the receiver sensitivity, so a complex baseband signal after down conversion at the receiver is inherently produced. In this scenario, the complex value neural network (CVNN) and RVNN equalizer connected with the Viterbi-Viterbi algorithm have better BER performance with an error rate lower than the HD-FEC threshold of 3.8 × 10-3.

Neural network-aided receivers for soliton communication impaired by solitonic interaction.

In this paper, different neural network-based methods are proposed to improve the achievable information rate in amplitude-modulated soliton communication systems. The proposed methods use simulated data to learn effective soliton detection by suppressing nonlinear impairments beyond amplifier noise, including intrinsic inter-soliton interaction, Gordon-Haus effect-induced timing jitter, and their combined impact. We first present a comprehensive study of these nonlinear impairments based on numerical simulations. Then, two neural network designs are developed based on a regression network and a classifier. We estimate the achievable information rates of the proposed learning-based soliton detection schemes as well as two model-based benchmark schemes, including the nonlinear Fourier transform eigenvalue estimation and continuous spectrum-aided eigenvalue estimation schemes. Our results demonstrate that both learning-based designs lead to substantial performance gains when compared to the benchmark schemes. Importantly, we highlight that exploiting the channel memory, introduced by solitonic interactions, can yield additional gains in the achievable information rate. Through a comparative analysis of the two neural network designs, we establish that the classifier design exhibits superior adaptability to interaction impairment and is more suitable for symbol detection tasks in the context of the investigated scenarios.

Bayesian generative adversarial network emulator based end-to-end learning strategy of the probabilistic shaping for OAM mode division multiplexing IM/DD transmission.

Orbital angular momentum (OAM) mode division multiplexing (MDM) has emerged as a new multiplexing technology that can significantly increase transmission capacity. In addition, probabilistic shaping (PS) is a well-established technique that can increase the transmission capacity of an optical fiber to close to the Shannon limit. However, both the mode coupling and the nonlinear impairment lead to a considerable gap between the OAM-MDM channel and the conventional additive white Gaussian noise (AWGN) channel, meaning that existing PS technology is not suitable for an OAM-MDM intensity-modulation direct-detection (IM-DD) system. In this paper, we propose a Bayesian generative adversarial network (BGAN) emulator based on an end-to-end (E2E) learning strategy with probabilistic shaping (PS) for an OAM-MDM IM/DD transmission with two modes. The weights and biases of the BGAN emulator are treated as a probability distribution, which can be accurately matched to the stochastic nonlinear model of OAM-MDM. Furthermore, a BGAN emulator based on an E2E learning strategy is proposed to find the optimal probability distribution of PS for an OAM-MDM IM/DD system. An experiment was conducted on a 200 Gbit/s two OAM modes carrierless amplitude phase-32(CAP-32) signal over a 5 km ring-core fiber transmission, and the results showed that the proposed BGAN emulator outperformed a conventional CGAN emulator, with improvements in modelling accuracy of 29.3% and 26.3% for the two OAM modes, respectively. Moreover, the generalized mutual information (GMI) of the proposed E2E learning strategy outperformed the conventional MB distribution and the CGAN emulator by 0.31 and 0.33 bits/symbol and 0.16 and 0.2 bits/symbol for the two OAM modes, respectively. Our experimental results demonstrate that the proposed E2E learning strategy with the BGAN emulator is a promising candidate for OAM-MDM IM/DD optic fiber communication.

23.1-Gb/s 135-GHz Wireless Transmission Over 4.6-Km and Effect of Rain Attenuation

In this article, we have experimentally demonstrated a fiber-wireless-integration <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$D$</tex-math> </inline-formula> -band (110–170 GHz) transmission system. To extend the wireless transmission distance while maintaining the wireless bit rate as high as possible, we designed high-gain lens horn antenna modules consisting of a pair of dielectric plano-convex lenses and horn antennas for long-haul wireless links. In addition, we employ the Volterra nonlinear equalization (VNE) algorithm based on the multiple-input–multiple-output (MIMO) structure to precisely compensate for the in-phase/quadrature (I/Q) imbalance and nonlinear impairment of quadrature amplitude modulation (QAM) signals mainly caused by photoelectric (O-E) conversion and electric-photo (E-O) conversion. Meanwhile, one of the methods adopted in our experiment to decrease the influence of nonlinearity on high-order QAM signals is coupling the high-order modulation format with the probabilistic shaping (PS) technology. As a result of outdoor field experimental verification, the demonstration at 135 GHz carrier with a net rate of 23.1 Gb/s over 10 km optic-fiber and 4.6 km wireless transmission distance has been successfully carried out. It is, as we know, the first time that a record-breaking product of net bit rate and wireless transmission distance, i.e., 23.1 Gb/s <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times$</tex-math> </inline-formula> 4.6 km <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$=$</tex-math> </inline-formula> 106.3 Gb/s <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\cdot$</tex-math> </inline-formula> km, has been accomplished based on a fiber-wireless-integration system. Also, for the first time, we have measured the effect of rain attenuation for the 135 GHz millimeter-wave (mm-wave) wireless link and established more accurate rainfall models for the Shanghai region of China by comparing them with multiple rainfall models. This work can be used as a complement to the International Telecommunication Union-Radiocommunication sector (ITU-R) recommendations.

Design of Ultra Dense WDM Optical Communication System to Reduce Signal Impairments Using DCF and Repeater Techniques

Linear and non-linear impairments are the major problems that affected the dense wavelength division multiplexing (DWDM) system performance. In this simulation designs, two different compensation techniques have been proposed to eliminate signal impairments. The repeater technique was proposed to reduce non-linear effects like four-wave mixing (FWM) which cause transfer power from DWDM channels to another new channels. The dispersion compensation fiber (DCF) technique was used to reduce dispersion effects which cause signal pulse broadening over long transmission distances. Also, three different modulation formats and four frequency separations were used to determine which modulation scheme is more compatible with the compensation techniques. The results is calculated in term of quality factor against power and noetic that the carrier suppressed return to zero (CSRZ) format offers the best performance with the DCF technique with 32.1804 Q-factor at 15dBm threshold power then, the differential phase shift keying (DPSK) format is compatible with the repeater technique with 27.7959 quality factor and 14dBm threshold power. Index Terms— Dense Wavelength Division Multiplexing (DWDM), Dispersion Compensation Fiber DCF, Repeater, and Non-linear Effects.

Low-Complexity Conditional Generative Adversarial Network (c-GAN) Based Nonlinear Equalizer for Coherent Data-Center Interconnections

Nonlinear impairments emerging from both fiber optical links and optoelectrical devices are the major bottleneck for enhancing the throughput of photonic data-center interconnection (DCI). Here, we propose a data-driven nonlinear equalizer based on a conditional generative adversarial network (c-GAN) for high-speed and large-capacity photonic DCI. Its performance is experimentally evaluated by transmitting C-band 40-channel 60-GBaud dual polarization-16 quadrature amplitude modulation (DP-16QAM) dense wavelength division multiplexing (DWDM) signals over 20-km standard single-mode fiber (SSMF). Our experimental results verify that, the proposed c-GAN equalizer outperforms both traditional nonlinear equalizers and other data-driven counterparts, in terms of both the bit-error ratio (BER) and computational complexity. Only the proposed c-GAN equalizer can guarantee the BER performance of all 40 channels to reach the threshold of 20% soft decision forward error correction (SD-FEC) at BER=0.027, leading to a net transmission capacity of 16 Tb/s. Meanwhile, the computational complexity of the proposed c-GAN equalizer can be significantly reduced by 31.8%, 80.7%, and 98.8%, respectively, in comparison with Volterra filter equalizer (VFE), multilayer perceptron (MLP), and long short-term memory neural network (LSTM-NN) schemes.

KRLS post-distorter with adaptive kernel width for visible light communications.

As a viable supplement to the fifth generation wireless communication, visible light communications (VLC) with affluent spectrum resources can cater to the ever-increasing high speed data transmission demand. However, the nonlinear characteristics of light emitting diode (LED) can distort the transmitted signal in the VLC link, which damages the communication quality. To mitigate the nonlinear impairments, a reproducing kernel Hilbert space post-distortion scheme is proposed in this paper, which is based on kernel recursive least squares (KRLS) with adaptive kernel width. In this kernel based method, the kernel width will affect the approximation ability of the model. Therefore, in the recursive process of KRLS, Gauss-Newton (GN) algorithm is adopted to update the kernel width. In addition, combined with the enhanced novelty criterion (ENC), the KRLS-GN post-distorter learns the sparse dictionary adaptively according to the input data, which is beneficial to complete the linearization under the limited memory budget constraints. The performance of the proposed KRLS-GN-ENC scheme is verified by simulations, and the results show that KRLS-GN-ENC can achieve a significant improvement over KRLS-ENC. Compared with the schemes based on classical polynomial filtering, KRLS-GN-ENC exhibits better nonlinear compensation performance and faster convergence speed.

Hidden conditional random field-based equalizer for the 3D-CAP-64 transmission of OAM mode-division multiplexed ring-core fiber communication.

As a key technique for achieving ultra-high capacity optical fiber communications, orbital angular momentum (OAM) mode-division multiplexing (MDM) is affected by severe nonlinear impairments, including modulation related nonlinearities, square-law nonlinearity and mode-coupling-induced nonlinearity. In this paper, an equalizer based on a hidden conditional random field (HCRF) is proposed for the nonlinear mitigation of OAM-MDM optical fiber communication systems with 20 GBaud three-dimensional carrierless amplitude and phase modulation-64 (3D-CAP-64) signals. The HCRF equalizer extracts the stochastic nonlinear feature of the OAM-MDM 3D-CAP-64 signals by estimating the conditional probabilities of the hidden variables, thereby enabling the signals to be classified into subclasses of constellation points. The nonlinear impairment can then be mitigated based on the statistical probability distribution of the hidden variables of the OAM-MDM transmission channel in the HCRF equalizer. Our experimental results show that compared with a convolutional neural network (CNN)-based equalizer, the proposed HCRF equalizer improves the receiver sensitivity by 2 dB and 1 dB for the two OAM modes used here, with l = + 2 and l = + 3, respectively, at the 7% forward error correction (FEC) threshold. When compared with a Volterra nonlinear equalizer (VNE) and CNN-based equalizer, the computational complexity of the proposed HCRF equalizer was found to be reduced by 30% and 41%, respectively. The bit error ratio (BER) performance and reduction in computational complexity indicate that the proposed HCRF equalizer has great potential to mitigate nonlinear distortions in high-speed OAM-MDM fiber communication systems.

Towards a photonic integrated all-optical phase regenerator.

All-optical phase regeneration aims at restoring the phase information of coherently encoded data signals directly in the optical domain so as to compensate for phase distortions caused by transceiver imperfections and nonlinear impairments along the transmission link. Although it was proposed two decades ago, all-optical phase regeneration has not been seen in realistic networks to date, mainly because this technique entails complex bulk modules and relies on high-precision phase sensitive nonlinear dynamics, both of which are adverse to field deployment. Here, we demonstrate a new, to the best of our knowledge, architecture to implement all-optical phase regeneration using integrated photonic devices. In particular, we realize quadrature phase quantization by exploring the phase-sensitive parametric wave mixing within on-chip silicon waveguides, while multiple coherent pump laser tones are provided by a chip-scale micro-cavity Kerr frequency comb. Multi-channel all-optical phase regeneration is experimentally demonstrated for 40 Gbps QPSK data, achieving the best SNR improvement of more than 6 dB. Our study showcases a promising avenue to enable the practical implementation of all-optical phase regeneration in realistic long-distance fiber transmission networks.

Low-Complexity Pruned Convolutional Neural Network Based Nonlinear Equalizer in Coherent Optical Communication Systems

Nonlinear impairments caused by devices and fiber transmission links in a coherent optical communication system can severely limit its transmission distance and achievable capacity. In this paper, we propose a low-complexity pruned-convolutional-neural-network-(CNN)-based nonlinear equalizer, to compensate nonlinear signal impairments for coherent optical communication systems. By increasing the size of the effective receptive field with an 11 × 11 large convolutional kernel, the performance of feature extraction for CNNs is enhanced and the structure of the CNN is simplified. And by performing the channel-level pruning algorithm, to prune the insignificant channels, the complexity of the CNN model is dramatically reduced. These operations could save the important component of the CNN model and reduce the model width and computation amount. The performance of the proposed CNN-based nonlinear equalizer was experimentally evaluated in a 120 Gbit/s 64-quadrature-amplitude-modulation (64-QAM) coherent optical communication system over 375 km of standard single-mode fiber (SSMF). The experimental results showed that, compared to a CNN-based nonlinear equalizer with a 6 × 6 normal convolutional kernel, the proposed CNN-based nonlinear equalizer with an 11 × 11 large convolutional kernel, after channel-level pruning, saved approximately 15.5% space complexity and 43.1% time complexity, without degrading the equalization performance. The proposed low-complexity pruned-CNN-based nonlinear equalizer has great potential for application in realistic devices and holds promising prospects for coherent optical communication systems.

Suppressed Noise Enhancement Equalization Combined With Nonlinear-MLSE Techniques for Bandwidth-Limitation IM/DD System

Bandwidth limitation introduced inter-symbol interference (ISI) is the major impediment for high baud rate intensity modulation and direct detection (IM/DD) system with commercial devices. Least mean square (LMS) criterion based feedforward equalizer (FFE) is the most common method to eliminate ISI in single-carrier system. However, there is a tradeoff between the residual ISI and enhanced noise when the system bandwidth is severely limited. Although a post filter (PF) followed by maximum likelihood sequence estimation (MLSE) can suppress the enhanced noise effectively, the residual ISI left by post-FFE will degrade the system performance as the subsequent MLSE only handles ISI introduced by PF. In this paper, an adaptive partial response equalization realized by a channel-shortening filter (CSF) is investigated, which is combined with MLSE. As the target of CSF is a partial response signal with desired low-pass feature, the noise will not be enhanced in this scheme. And the residual ISI with determined response can be fully solved by MLSE. The performance of CSF based MLSE (CSF-MLSE) is evaluated and compared to PF based MLSE (PF-MLSE) in 60/70/75Gbaud four-level pulse amplitude modulation (PAM-4) system with 10-dB bandwidth of 16.2GHz. The experimental results show that CSF-MLSE has a better capability of ISI elimination compared to PF-MLSE. The bit error rate (BER) can be reduced from 7.4×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> to 4×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> and from 3×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> to 1×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> for 70Gbaud and 75Gbaud in optical back-to-back (OBTB) transmission. To further compensate the system nonlinear impairments, several nonlinear MLSE (NL-MLSE) techniques assisted by look-up table (LUT) or Volterra filter are proposed and investigated in this paper. The BER of 75Gbuad PAM-4 can be below 3×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> after using NL-MLSE schemes.

400 Gbit/s 4 mode transmission for IM/DD OAM mode division multiplexing optical fiber communication with a few-shot learning-based AffinityNet nonlinear equalizer.

Nonlinear impairment in a high-speed orbital angular momentum (OAM) mode-division multiplexing (MDM) optical fiber communication system presents high complexity and strong stochasticity due to the massive optoelectronic devices. In this paper, we propose an Affinity Network (AffinityNet) nonlinear equalizer for an OAM-MDM intensity-modulation direct-detection (IM/DD) transmission with four OAM modes. The labeled training and testing signals from the OAM-MDM system can be regarded as "small sample" and "large target", respectively. AffinityNet can be used to build an accurate nonlinear model using "small sample" based on few-shot learning and can predict the stochastic characteristic nonlinearity of OAM-MDM with a high level of generalization. As a result, the AffinityNet nonlinear equalizer can effectively compensate the stochastic nonlinearity in the OAM-MDM system, despite the large difference between the training and testing signals due to the stochastic nonlinear impairment. An experiment was conducted on a 400 Gbit/s transmission with four OAM modes using a pulse amplitude modulation-8 (PAM-8) signal over a 2 km ring-core fiber (RCF). Our experimental results show that the proposed nonlinear equalizer outperformed the conventional Volterra equalizer with improvements in receiver sensitivity of 1.7, 1.8, 3, and 3.3 dB for the four OAM modes at the 15% forward error correction (FEC) threshold, respectively. In addition, the proposed equalizer outperformed a convolutional neural network (CNN) equalizer with improvements in receiver sensitivity of 0.8, 0.5, 0.9, and 1.4 dB for the four OAM modes at the 15% FEC threshold. In the experiment, a complexity reduction of 37% and 83% of the AffinityNet equalizer is taken compared to the conventional Volterra equalizer and CNN equalizer, respectively. The proposed equalizer is a promising candidate for a high-speed OAM-MDM optical fiber communication system.

Optical cable deployment versus fiber leasing: an operator’s perspective on CapEx savings for capacity upgrade in an elastic optical core network

Multiband (MB) and multifiber (MF) technologies get significant attention along with their individual pros and cons to solve the optical fiber capacity crunch problem. The MB system is more susceptible to nonlinear transmission impairments such as interchannel stimulated Raman scattering. On the contrary, MF solutions are costly due to the requirement of parallel fiber deployment. Therefore, it is necessary to find a hybrid solution using MB and MF technologies. This paper performs a techno-economic comparison between MF and MB technologies while taking a robust cost model. We propose, for the first time to our knowledge, a domain-knowledge-assisted, cost-effective network upgrade algorithm to enhance the overall network capacity while including minimum capital expenditure (CapEx). The proposed algorithm prioritizes the targeted deployment of cost-effective upgrade solutions in different links of the network to minimize the overall cost of the upgrade. The comparison between optical cable deployment and fiber leasing is captured to upgrade the overall network capacity with minimum cost-per-bit. Reported results show that a single fiber C+L band system can provide approximately 66.67% gain in terms of traffic admissibility compared to MF C band systems in smaller networks such as BT-UK, while considering fiber leasing. If operator owned dark fibers are not available in any of the links, numerical results show that the MF C+L band system in the BT-UK network can still provide CapEx savings of approximately up to 40.5% compared to the MF C band systems while minimizing the cost-per-bit of the network by approximately 21.9% to achieve the targeted network capacity of 150 Tbps. The effect of longer link length on the network upgrade cost is also studied in this paper. Reported result shows that, for larger geography such as the pan-Europe network, the channel launch power needs to be tuned to achieve the benefit of the MF C+L band compared to the MF C band system.

Adaptive Fuzzy Logic Deep-Learning Equalizer for Mitigating Linear and Nonlinear Distortions in Underwater Visible Light Communication Systems

Underwater visible light communication (UVLC) has recently come to light as a viable wireless carrier for signal transmission in risky, uncharted, and delicate aquatic environments like seas. Despite the potential of UVLC as a green, clean, and safe alternative to conventional communication methods, it is challenged by significant signal attenuation and turbulent channel conditions compared to long-distance terrestrial communication. To address linear and nonlinear impairments in UVLC systems, this paper presents an adaptive fuzzy logic deep-learning equalizer (AFL-DLE) for 64 Quadrature Amplitude Modulation-Component minimal Amplitude Phase shift (QAM-CAP)-modulated UVLC systems. The proposed AFL-DLE is dependent on complex-valued neural networks and constellation partitioning schemes and utilizes the Enhanced Chaotic Sparrow Search Optimization Algorithm (ECSSOA) to improve overall system performance. Experimental outcomes demonstrate that the suggested equalizer achieves significant reductions in bit error rate (55%), distortion rate (45%), computational complexity (48%), and computation cost (75%) while maintaining a high transmission rate (99%). This approach enables the development of high-speed UVLC systems capable of processing data online, thereby advancing state-of-the-art underwater communication.

Key Technologies for High-Speed Si-Substrate LED Based Visible Light Communication

Nowadays, the demand for high-speed information transmission is increasing rapidly. Visible light communication (VLC) based on light-emitting diodes (LEDs) is supposed as a potential candidate for future 6G communication networks. However, as an emerging field of optical communications, the data rate of VLC systems is restricted to the extremely limited modulation bandwidth and nonlinear characteristics of optoelectronic devices. In this paper, we outline the state of art and challenges of VLC, and introduce the key technologies for high-speed Silicon (Si) -substrate LED-based VLC systems. Based on an optimized integrated 4×4 8-wavelength LED array and advanced digital signal processing (DSP) methods, we demonstrate an ultrahigh-speed VLC system. Digital Zobel network (DZN) pre-equalization and transfer learning bidirectional gate recurrent unit (TL-Bi-GRU) based post-equalization are proposed to mitigate linear and nonlinear impairment. The yellow and green LEDs are especially enhanced in high-efficacy, and achieve the impressive data rates of 3.45Gbit/s, and 3.83Gbit/s respectively, which is a remarkable improvement. The achievable overall data rate of the device is 28.93 Gbit/s. As far as we know, it's the highest data rate achieved by a single-chip Si-substrate LED array.

High Dynamic Range 100G PON Enabled by SOA Preamplifier and Recurrent Neural Networks

In recent years the PON research community has focused on future systems targeting 100 Gb/s/ <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\lambda$</tex-math></inline-formula> and beyond, with digital signal processing seen as a key enabling technology. Spectrally efficient 4-level pulse amplitude modulation (PAM4) is seen as a cost-effective solution that exploits the ready availability of cheaper, low-bandwidth devices, and Semiconductor Optical Amplifiers (SOA) are being investigated as receiver preamplifiers to compensate PAM4's high signal-to-noise ratio requirements and meet the demanding 29 dB PON loss budget. However, SOA gain saturation-induced patterning distortion is a concern in the context of PON burst-mode signalling, and the 19.5 dB loud-soft packet dynamic range expected by the most recent ITU-T 50G standards. In this paper we propose a recurrent neural network equalisation technique based on gated recurrent units (GRU-RNN) to not only mitigate SOA patterning affecting loud packet bursts, but to also exploit their remarkable effectiveness at compensating non-linear impairments to unlock the SOA gain saturated regime. Using such an equaliser we demonstrate <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ &gt; 28$</tex-math></inline-formula> dB system dynamic range in 100 Gb/s PAM4 system by using SOA gain compression in conjunction with GRU-RNN equalisation. We find that our proposed GRU-RNN has similar equalisation capabilities as non-linear Volterra, fully connected neural network, and long short-term memory based equalisers, but observe that feedback-based RNN equalisers are more suited to the varying levels of impairment inherent to PON burst-mode signalling due to their low input tap requirements. Recognising issues surrounding hardware implementation of RNNs, we investigate a multi-symbol equalisation scheme to lower the feedback latency requirements of our proposed GRU-RNN. Finally, we compare equaliser complexities and performances according to trainable parameters and real valued multiplication operations, finding that the proposed GRU-RNN equaliser is more efficient than those based on Volterra, fully connected neural networks or long short-term memory units proposed elsewhere.

Unequal channel spacing 8×5Gbps optical system utilizing different signal representations.

Data representations are important to communication networks; they convert the data bits into a signal form, which affects the system capacity, maximum bit rate, transmission distance, and different linear and nonlinear impairments. In this paper, we propose non-return-to-zero (NRZ), chirped NRZ, duobinary, and duobinary return-to-zero (DRZ) data representations with eight dense wavelength division multiplexing channels for transmitting a 5Gbps data rate through a 250km fiber length. The results of the simulation design are calculated at different equal and unequal channel spacings, and the quality factor is measured over a wide range of optical power. For equal channel spacing, the DRZ has a preferable performance with a 28.40 quality factor at 18dBm threshold power, and the chirped NRZ has a preferable performance with a 26.06 quality factor at 12dBm threshold power. For unequal channel spacing, the DRZ has a 25.76 quality factor at 17dBm threshold power, and the NRZ has a 25.06 quality factor at 10dBm threshold power.