This PDF file contains the front matter associated with SPIE Proceedings Volume 10947, including the Title Page, Copyright information, Table of Contents, Introduction, and Author and Conference Committee lists

We report on the time-resolved measurement of the full transmission matrix (TM) of a short length of specialty annularcore few-mode fiber which guides 10 vector modes. We show how our method can isolate the fiber TM from "misalignment" contributions from free space optics upstream and downstream of the fiber. From measurements spanning two days, we extract the drift of the fiber TM. We show that drifts in the TM elements are mostly described as correlated phase variations rather than amplitude variations. We show that an empirical model of the fiber TM parametrized in one parameter can successfully account for the drift.

This paper reviews high-capacity space-division multiplexed (SDM) optical transport system technologies that support the evolution of broadband networks. Technologies that offer new spatial degrees of freedom such as core multiplexing and mode multiplexing can be introduced to optical transmission systems, optical switching systems, and optical fibers; they overcome the physical limits of today’s transport systems based on single-mode fibers. The expected features of future SDM-based optical transport networks are achieving both high transmission capacity over 1 Pbit/s and node throughputs above 10 Pbit/s, while reducing the power consumption and the size of transport node equipment.

The orbital angular momentum (OAM) mode is considered as a new degree of freedom for mode division multiplexing (MDM) to overcome the increasing data capacity because of its orthogonality and theoretically unlimited number of OAM mode. Although OAM modes have been successfully demonstrated in free-space-based and conventional fibers, these have limited applications because of the gradual enlargement of these modes with propagation and small number of modes. In this paper, we propose a photonic quasi-crystal fiber (PQF) for supporting up to 38 orbital OAM mode with a flat dispersion characteristic over the C+L bands. We numerically investigated the eigenmodes in the proposed PQF and these effective indices, electric field intensity distributions, dispersion, and confinement loss. The designed PQF which consists of a large hollow center and quasi structural small air holes in the clad region exhibits low confinement losses and many number of OAM modes, and satisfies a radially single mode condition and a large effective index separation (>10^-4) between the same order of the OAM modes. This proposed fiber could potentially be exploited for mode division multiplexing and other OAM mode applications in fibers.

The use of a rectangular core fiber is suggested for mode division multiplexed optical communication, and as an alternative fiber geometry having advantageous transmission and component integration attributes. The rectangular core is constrained to support single transverse modes in one direction and multiple modes in the horizontal transverse direction. The supported modes are thus polarization degenerate only (i.e., TE1x and TM1x), with well separated momenta and favorable mode profiles for device coupling and wavelength multiplexing and manipulation. A fiber prototype is experimentally characterized for its modal delays and field profiles by time-gated interferogram analysis.

Turbulence can distort the signal wavefront in free space optical (FSO) communications, leading to errors. The state-of-the-art method for correcting distortions is adaptive optics (AO). We show that improvements in turbulent FSO communication link performance can be obtained using a few-mode optical pre-amplified receiver, without AO.We compare pre-amplified few-mode and single-mode receivers for both OOK and DPSK modulation formats.

The information rate (IR) of a digital coherent transceiver is constrained by the inherent practical signal-to-noise ratio (SNR) limit. Coded modulation, which is the combination of multi-level modulation and forward error correction, aims to maximize the IR within this SNR envelope. While probabilistic constellation shaping has enhanced this methodology by providing an increase in IR over conventionally employed square quadrature amplitude modulation (QAM) formats, it is the ability to eloquently tune the per wavelength IR by varying the symbol probabilities that has gained this scheme significant traction within optical communications in recent years. As commercial line cards continue their evolution towards 100 GBd and to modulation formats beyond 64QAM, we discuss the merits of probabilistic shaping for high symbol rate digital coherent transceivers in the presence of a practical SNR limit.

In data centers, 100Gigabit Ethernet is being deployed and 200GbE/400GbE is expected to come in a couple of years. For optics, conventional OOK/PAM4 modulation with direct detection (DD) can only provide limited bandwidth and reach. Recently a few novel techniques such as Stokes vector receiver (SVR) and Kramers-Kronig receiver (KKR) has been proposed to further improve the system SE and/or reach. In this talk, we present a comprehensive review on various aspects of the two distinct detection schemes. We make comparison between the two schemes in terms of transceiver architecture, spectral efficiency, receiver bandwidth, and performance.

Advanced modulation formats, e.g. multi-dimensional modulation, probabilistic shaping, and geometric shaping, are promising candidates to realize beyond 400 Gbps serial long haul transmission. Multi-dimensional modulation encoded with a short block length extends the minimum Euclidean distance between multi-dimensional symbols, resulting in improving noise tolerance. We have recently proposed eight-dimensional (8D)-16QAM with low-complexity iterative demodulation scheme and have demonstrated 600 Gbps/wavelength 120-GBaud 8D-16QAM WDM transmission over 3,900 km. In this paper, we discuss a design method of multi-dimensional modulation formats and demodulation techniques.

Spatial division multiplexing (SDM) is crucial for ultra-high capacity optical fiber transmission. In addition to SDM using multi-core fiber, multi-modal multiplexing using multi-mode fiber is another option. MDM and OAM are intensively investigated. For any SDM transmission, typical ways are based on optical fan-in and -out devices for multiplexing and demultiplexing tributaries. In this paper, spatial coherent matched detection is investigated for showing the capability of multi-modal coherent detection, where such SDM signals are received without using optical fan-out devices. Numerical proof focusing on reception of high-speed SDM coherent multilevel signals is provided.

In this paper, a Mode division multiplexing (MDM) optical transmission system that uses deep learning neural network (DLNN) for Multiple Input Multiple Output (MIMO) detection is presented. Two channels operating at 250Mbps are QPSK modulated and transmitted at different mode through a Multi-Mode fiber and successfully detected. For MIMO detection, a supervised DLNN, which is designed, trained and evaluated using a Keras library and TensorFlow, is implemented in this MDM optical transmission system. The performance of our DLNN for MIMO detection is compared with Zero Forcing and Semi-Definite Relaxation Row-by-Row detectors. Our DLNN outruns the performance of these MIMO detectors.

Surface-normal electro-absorptive modulators based on III-V quantum well superlattices are of interest for a large number of applications, including on-chip and free-space photonic links, detection-and-ranging and optical tagging. In recent years, novel designs of the quantum well layers stack, such as asymmetric stepped quantum well, have allowed to reach energy efficient, wide spectral bandwidth modulation performance. Nevertheless, the design of these structures is still based on intuition rather than on a quantitative assessment of the device and system performance. Moreover, the increasing number of applications has highlighted the need for a comprehensive design approach, that incorporates the performance metrics of the specific system into the design considerations.

We present a novel approach for the systematic optimization of the design of electro-absorptive modulators, based on a combination of analytical modeling and supervised machine learning. Fully-validated analytical modeling of the electronic transitions and optical propagation in the semiconductor compound is used for the training of an evolutionary algorithm, which drives the global search for optimal design.

The approach was tested for the optimization of the superlattice design of the electro-absorptive modulator for two different applications: time-of-flight 3D ranging camera, and remote sensing of electro-chemical signal via optical tagging. In both cases, a system-specific figure-of-merit is proposed and employed for the evaluation and optimization of the performance, yielding two novel optimized designs which allow for considerable performance improvement of the respective systems.

This paper presents a patent pending approach to store digital data using high bandwidth laser light in motion. Data will be stored in a loop. The higher the data rate, and the longer the storage loop, the more data can be stored. We plan to use : (i) highly multiplexed optical communications to maximize data rate, and (ii) a long light path, to maximize length of time it takes light to traverse the full storage loop. Lyteloop is developing data storage in motion in 4 domains : fiber, space, and two forms of near vacuum cavities. Three of those domains are in a free space near vacuum. Multiplexing will be done across very wide wavelength ranges, and many spatial modes will be used. OEM modes will be used in free space at the shorter wavelengths, and spatial division multiplexing (SDM) in fibers. Multiple cores will be used in fibers A 1.5 Gigabyte demonstation data storage unit has been built in fiber. This démonstration unit, and its performance, is described. Lyteloop will build on, and advance, both free space, and fiber, high data rate communications technology as a method of storing data in motion. We will also describe two patent pending methods to extend path length in free space applications. One approach we call angle multiplexing, and the other uses highly reflective interior surfaces. Less complex, and lower power, signal regeneration will be highly beneficial to achieve Lytelops’s goals.

A safe smart vehicle lighting system that combines the functions of illumination, signaling, communications, and positioning is presented. The communication between the infrastructures and the vehicles (I2V), between vehicles (V2V) and from the vehicles to the infrastructures (V2I) is performed through Visible Light Communication (VLC) using the street lamps and the traffic signaling LEDs to broadcast the information. Vehicle headlamps are used to transmit data to other vehicles or infrastructures (traffic lights) allowing digital safety and data privacy. As receivers and decoders, pin/pin SiC Wavelength Division Multiplex (WDM) photodetectors, with light filtering properties, are being used. We propose the use of white polychromatic-LEDs to implement the WDM. This allows modulating separate data streams on four colors which together multiplex to white light. When a probe vehicle enters the infrastructure’s capture range, the receivers respond to light signal and the infrastructure ID and traffic message are assigned. They perform simultaneously the V2V distance measurement and data transmission functions and, using the headlamps, resend the data to the other vehicles or to the traffic signals (V2I). A I2V2V2I traffic scenario is stablished. A vulnerable road user case that covers pedestrians, cyclists and wheelchairs is also considered. A phasing traffic flow is developed as a Proof of Concept (PoC). The experimental results confirm that the cooperative vehicular VLC architecture is a promising approach concerning communications between road infrastructures and cars, fulfilling data privacy.

White LEDS revolutionized the field of illumination technology mainly due to the energy saving effects. Besides lighting purposes LEDs can also be used in wireless communication systems when integrated in Visible Light Communication (VLC) systems. Indoor positioning for navigation in large buildings is currently under research in order to overcome the difficulties associated with the use of GPS in such environments. The motivation for this application is also supported by the possibility of taking advantage of an existing lighting and WiFi infrastructure. In this work it is proposed a bidirectional communication system between a static infrastructure and a mobile picking robot. The LED luminaires at the warehouse ceiling are used to lighten the warehouse space, and to transmit information for positioning and about available racks. The picking robots communicate with the ceiling luminaires to send information about the rack that is being removed and carried to the packaging station. Different codification schemes are proposed to establish both uplink and downlink communication between LED luminaires and mobile robots. Tri-chromatic white LEDs with the red and blue chips modulated at different frequencies and a pinpin photodetector with selective spectral sensitivity are used at the emitter and receiver modules.

This invited paper discusses the potential of the Intermediate Frequency-over-Fiber (IFoF) technology for future mobile fronthaul links. First, we explain the fundamental of IFoF and its advantages over the conventional digital-based schemes such as Common Public Radio Interface (CPRI) from the point view of centralized radio access network (C-RAN) architecture. Then, our recent results including the experimental demonstration of the 1-Tb/s CPRI-equivalent-rate transmission with a transmitter composed of parallel intensity/phase (IM/PM) modulators are presented.

According to the sampling theorem, bandwidth limited signals can be seen as superposition of time shifted sinc pulses weighted with the sampling values. Since sinc pulses are orthogonal to each other, bandlimited signals can be perfectly sampled by an integration over the product between them and a sinc pulse with the correct time shift and bandwidth. Because sinc pulses have an infinite time length, they cannot be realized experimentally. Instead, generating a periodical sinc pulse sequence is straightforward. For a low duty cycle the pulses in such a sequence come close to single sinc pulses and thus the sampling might come closer to ideal sampling. In the frequency domain, this nearly ideal sampling is represented by a convolution between the signal spectrum and a rectangular frequency comb with many lines. The bandwidth of the comb corresponds to the sampling rate, while a bigger number of comb lines reduces the duty cycle and might enhance the sampling quality. We present the generation of a flat frequency comb with up to 33 lines in the optical domain as well as how to convolve it with an optical input spectrum for optical sampling. Already with one Mach- Zehnder modulator driven with m equidistant radio frequencies, the sampling with a comb consisting of 2m+1 lines can be realized. Additionally, with a second Mach-Zehnder modulator driven with n equidistant radio frequencies, the comb line number can be enhanced to (2m+1)(2n+1).

We have developed an external cavity self-injection feedback locking (SIFL) system to simultaneously reduce optical linewidth of each individual channel of an InAs/InP qantum dot (QD) 34.46-GHz coherent comb laser (CCL). Optical linewidths are reduced from a few MHz down to less than 300 kHz over 47 filtered individual channels, varying from 5.3% to 9.1% of the original linewidth, between 1531.60 nm to 1544.20 nm. By using this ultra-narrow linewidth QD CCL we have demonstrated 12.032 Tbit/s (16QAM 47x32 GBaud PDM) back to back coherent data bandwidth transmission capacity.

We report optical Nyquist pulse train generation by non-auxiliary wavelength selective switch (WSS) in communication band. Nyquist pulses have the attractive feature of tolerance to inter-symbol interference (ISI), which means that densely arranged pulse trains are possible. The typical approach for optical Nyquist pulse train generation uses an auxiliary optical circuit for time division multiplexing as well as a WSS for a single Nyquist pulse generation. The auxiliary use of the optical circuit gives rise to optical losses and inflexibility of pulse-train parameters. The optical loss is estimated as 10 dB even if an ideal optical circuit is used in the case of 10-multiplexing. To resolve these problems, we have recently proposed a new approach for optical Nyquist pulse train generation by non-auxiliary WSS in the near-infrared band, where the WSS combines a single Nyquist pulse generation and time division multiplexing. The key point of the approach is how to design a filter function to minimize the optical loss. We have established the method to design the low-loss filter function, which takes advantage of the ISI-free property of Nyquist pulses. We have experimentally demonstrated Nyquist pulse train generation with the proposed approach in the near-infrared band. In this report, we widen its application range to the optical communication band, and experimental results show that the optical loss for 10- multiplexing is successfully reduced to 1.36 dB. The new approach without an auxiliary optical circuit realizes low-loss, highly flexible and compact optical Nyquist pulse train generator in the optical communication band.

We report a coherent driver modulator sub-assembly for 100-Gbaud class transmitter. The sub-assembly employs a codesigned high-speed driver IC and optical modulator, where the designs for the characteristic impedance between the driver IC and IQ modulator and for the RF pads are optimized. The EO bandwidth is over 67 GHz. This response is sufficient for 100-Gbaud-class operation even if we consider RF losses caused by a package, PCB, and DAC. We also demonstrated QPSK and 16-QAM at the speed of 128 and 112 Gbaud, respectively.

The coherent optical transponder is ubiquitous in long-haul optical network. The power imbalance among four tributaries limits the transmission performance of coherent transponder. This impairment is particularly severe for the 400 gigabit per second (Gb/s) signal, which utilizes the advanced modulation format like quadrature amplitude modulation (QAM) and operates at the high baud rate. To investigate the impairment, we constructed 400Gb/s coherent optical transponder with high speed DAC, IQ modulator, and high speed ADC. The power imbalance is introduced in digital domain. We experimentally investigated the impact of power imbalance in various scenarios. For the different baud rates like 45Gbd, 64Gbd and 86Gbd, the higher the baud rate is, the larger the penalty from the power imbalance is. For the different modulation formats like QPSK, 16-QAM, and 64-QAM, the higher the order of the modulation format is, the larger the penalty from the power imbalance is. Two types of power imbalance exist: IQ imbalance between in-phase and quadrature tributaries within one polarization, and XY imbalance between two polarizations. With high optical-to-signal noise ratio (OSNR), the penalty from the IQ power imbalance is larger; with low OSNR, the penalty from the XY power imbalance is larger. Furthermore, we demonstrated a technique to detect and compensate the power imbalance during the initial power-up of the coherent IQ transmitter. The technique is implemented through the adjustment of gain scaling factor of finite impulse filter (FIR). It works over multiple wavelength and the accuracy of internal photo diode is sufficient.

We experimentally demonstrate three-fold wavelength multicasting of a 64-quadrature-amplitude-modulation (QAM), 120-Gbit/s data channel using a microresonator Kerr frequency comb and nonlinear wave mixing. The multicasting is achieved with a data signal and four comb lines serving as the pump lasers in a periodically poled lithium niobate (PPLN) waveguide. Minimal extra phase noise from the pumps is introduced into the multicast copies due to the mutual coherence between the Kerr comb lines. All three multicast copies achieve a bit-error rate (BER) <= 3.5E-3, which is below the forward-error-correction threshold. Both the error vector magnitude (EVM) and BER performances show <0.5-dB optical signal-to-noise ratio (OSNR) penalty for the multicast copies compared to the original data signal.