A5025311119- Optical Network Technologies
- Advanced Photonic Communication Systems
- Photonic and Optical Devices
- Advanced Fiber Laser Technologies
- Semiconductor Lasers and Optical Devices
- Neural Networks and Reservoir Computing
- Advanced Optical Network Technologies
- Advanced Fiber Optic Sensors
- Spectroscopy Techniques in Biomedical and Chemical Research
- Spectroscopy and Laser Applications
- Photonic Crystal and Fiber Optics
- Optical Wireless Communication Technologies
- Advanced Wireless Communication Techniques
- Quantum Information and Cryptography
- Mechanical and Optical Resonators
- graph theory and CDMA systems
- Quantum optics and atomic interactions
- PAPR reduction in OFDM
- Wireless Signal Modulation Classification
- Advanced Optical Sensing Technologies
- Acoustic Wave Resonator Technologies
- Advanced Memory and Neural Computing
- Blind Source Separation Techniques
- Photonic Crystals and Applications
- Millimeter-Wave Propagation and Modeling
Technical University of Denmark
2016-2025
Ørsted (Denmark)
2011-2024
Photonic Systems (United States)
2019-2021
Polytechnic University of Turin
2021
Friedrich-Alexander-Universität Erlangen-Nürnberg
2017-2019
Swinburne University of Technology
2017
Xi'an Institute of Optics and Precision Mechanics
2017
University of Electronic Science and Technology of China
2017
City University of Hong Kong
2017
University of Hong Kong
2017
Today's telecommunication networks have become sources of enormous amounts widely heterogeneous data. This information can be retrieved from network traffic traces, alarms, signal quality indicators, users' behavioral data, etc. Advanced mathematical tools are required to extract meaningful these data and take decisions pertaining the proper functioning network-generated Among tools, machine learning (ML) is regarded as one most promising methodological approaches perform network-data...
Machine learning techniques relevant for nonlinearity mitigation, carrier recovery, and nanoscale device characterization are reviewed employed. Markov Chain Monte Carlo in combination with Bayesian filtering is employed within the nonlinear state-space framework demonstrated parameter estimation. It shown that time-varying effects of cross-phase modulation (XPM) induced polarization scattering phase noise can be formulated model (SSM). This allows tracking compensation XPM impairments by...
Accurate channel models are essential to evaluate mobile communication system performance and optimize coverage for existing deployments. The introduction of various transmission frequencies 5G imposes new challenges accurate radio prediction. This paper compares traditional a model obtained using Deep Learning (DL)-techniques utilizing satellite images aided by simple path loss model. Experimental measurements gathered compose the training test set. considers modelling techniques offered...
We experimentally demonstrate an 100 Gbit/s hybrid optical fiber-wireless link by employing photonic heterodyning up-conversion of 12.5 Gbaud polarization multiplexed 16-QAM baseband signal with two free running lasers.Bit-error-rate performance below the FEC limit is successfully achieved for air transmission distances up to 120 cm.
Linear signal processing algorithms are effective in dealing with linear transmission channel and detection, whereas the nonlinear algorithms, from machine learning community, detection. In this paper, a brief overview of various methods their application optical communication is presented discussed. Moreover, supervised methods, such as neural networks support vector machine, experimentally demonstrated for in-band to noise ratio estimation modulation format classification, respectively....
New services and applications are causing an exponential increase in internet traffic.In a few years, current fiber optic communication system infrastructure will not be able to meet this demand because nonlinearity dramatically limits the information transmission rate.Eigenvalue could potentially overcome these limitations.It relies on mathematical technique called "nonlinear Fourier transform (NFT)" exploit "hidden" linearity of nonlinear Schrödinger equation as master model for signal...
To accommodate the demand of exponentially increased global wireless data traffic, prospective rates for communication in market place will soon reach 100 Gb/s and beyond. In lab environment, transmission throughput has been elevated to level over attributed development photonic-assisted millimeter wave terahertz (THz) technologies. However, most recent demonstrations with are based on spatial or frequency division multiplexing techniques, resulting system's complexity energy consumption....
In this letter, the fiber-optic communication channel with a quadrature amplitude modulation (QAM) input constellation is treated. Using probabilistic shaping, we show that high-order QAM constellations can achieve and slightly exceed lower bound on capacity, set by ring constellations. We then propose mapping function for turbo-coded bit-interleaved coded based optimization of mutual information between output. mapping, spectral efficiency as high 6.5 bits/s/Hz/polarization achieved...
In this paper, probabilistic shaping is numerically and experimentally investigated for increasing the transmission reach of wavelength division multiplexed (WDM) optical communication system employing quadrature amplitude modulation (QAM). An optimized probability mass function (PMF) QAM symbols first found from a modified Blahut-Arimoto algorithm channel. A turbo coded bit interleaved then applied, which relies on many-to-one labeling to achieve desired PMF, thereby achieving gain. Pilot...
In this paper, an alternative approach to design linear equalization algorithms for optical coherent receivers is introduced. Using widely (WL) complex analysis, a general analytical model it shown, where In-phase/quadrature (IQ) imbalances and IQ skew at the receiver front-end are naturally included in channel problem. Next, problem of chromatic dispersion (CD) compensation after imbalanced skewed detection analyzed. Based on models obtained demonstrated that, under presence such...
A wide range of highly-relevant problems in programmable and integrated photonics, optical amplification, communication deal with inverse system design. Typically, a desired output (usually gain profile, noise transfer function or similar continuous function) is given the goal to determine corresponding set input parameters voltages, currents, powers, wavelengths). We present novel method for design using machine learning apply it Raman amplifier Inverse amplifiers consists selecting pump...
A new geometric shaping method is proposed, leveraging unsupervised machine learning to optimize the constellation design. The learned mitigates nonlinear effects with gains up 0.13 bit/4D when trained a simplified fiber channel model.
We present a radio-frequency (RF) and bit-rate scalable technique for multigigabit wireless signal generation based on all-optical orthogonal frequency-division multiplexing (OFDM) photonic up-conversion. Coherent detection supported by digital processing is used demodulation data recovery. In order to demonstrate the RF frequency scalability transparency, system tested at 60 GHz in 75- 110-GHz band baud rates of 5 10 Gbaud. terms bit rate, proposed experimentally up 40 Gb/s demodulation....
In this paper, we show numerically and experimentally that expectation maximization (EM) algorithm is a powerful tool in combating system impairments such as fibre nonlinearities, inphase quadrature (I/Q) modulator imperfections laser linewidth.The EM an iterative can be used to compensate for the which have imprint on signal constellation, i.e. rotation distortion of constellation points.The especially effective non-linear phase noise (NLPN).It because NLPN severely distorts tracked by...
We present a technique for modulation format recognition heterogeneous reconfigurable optical networks. The method is based on Stokes space signal representation and uses variational Bayesian expectation maximization machine learning algorithm. Differentiation between diverse common coherent formats successfully demonstrated numerically experimentally. proposed does not require training or constellation diagram to operate, insensitive polarization mixing frequency offset can be implemented...
In this paper, we demonstrate a photonic up-converted 25 Gbit/s fiber-wireless quadrature phase shift-keying (QPSK) data transmission link at the W-band (75-110 GHz). By launching two free-running lasers spaced 87.5 GHz into standard single-mode fiber (SSMF) central office, radio-over-fiber (RoF) signal is generated and distributed to remote antenna unit (RAU). One laser carries 12.5 Gbaud optical baseband QPSK data, other acts as carrier frequency generating laser. The signals are...
We report on a 116 Gb/s on-off keying (OOK), four pulse amplitude modulation (PAM) and 105-Gb/s 8-PAM optical transmitter using an InP-based integrated packaged externally modulated laser for high-speed interconnects with up to 30 dB static extinction ratio over 100-GHz 3-dB bandwidth 2 ripple. In addition, we study the tradeoff between power penalty equalizer length foresee transmission distances standard single mode fiber.
A single-transmitter/single-receiver THz link (0.3-0.5 THz) with a record net data rate of 260 Gbit/s is experimentally demonstrated. Spectrally efficient multi-channel signal transmission enabled by novel frequency-band-allocation scheme pre-and-post-digital equalization.
Photonic integrated circuits are facilitating the development of optical neural networks, which have potential to be both faster and more energy efficient than their electronic counterparts since signals especially well-suited for implementing matrix multiplications. However, accurate programming photonic chips multiplication remains a difficult challenge. Here, we describe simple analytical models data-driven offline training multipliers. We train evaluate using experimental data obtained...
End-to-end learning has become a popular method to optimize constellation shape of communication system. When the channel model is differentiable, end-to-end can be applied with conventional backpropagation algorithm for optimization shape. A variety algorithms have also been developed over non-differentiable model. In this paper, we compare gradient-free based on cubature Kalman filter, model-free and fiber-optic modeled by split-step Fourier method. The results indicate that provide decent...
The nonlinear Fourier transform is a new approach for addressing the capacity limiting Kerr nonlinearities in optical communication systems. It exploits property of integrability lossless Schrödinger equation and thus incorporates as an element transmission. However, practical links employing erbium-doped fiber amplifiers include losses/gains introduce noise which breaks equation. Although path average approximation proposes integrable model, its imprecision still leads to unintended...
Autoencoder-based geometric shaping is proposed that includes optimizing bit mappings.Up to 0.2 bits/QAM symbol gain in GMI achieved for a variety of data rates and the presence transceiver impairments.The gains can be harvested with standard binary FEC at no cost w.r.t.conventional BICM.
We experimentally demonstrate an optical network compatible high-speed optoelectronics THz wireless transmission system operating at 400-GHz band. In the experiment, Nyquist quadrature phase-shift keying signals in a 12.5-GHz ultradense wavelength-division multiplexing grid is converted to radiation by photomixing antenna integrated unitravelling photodiode. The transparent modulation formats. also experiment scalability of our applying single four channels, as well mixed three channels....