This PDF file contains the front matter associated with SPIE Proceedings Volume 11309, including the Title Page, Copyright information, Table of Contents, Author, and Conference Committee lists.

Flexible dielectric optical fibers guiding light in a hollow core were conceptually imagined at the end of the 19th century, but first demonstrated in practice about 2 decades ago. Since then, many geometric variants have been described and implemented, and theoretical models developed and finessed. Despite this, for a fairly long time the key metric by which their performance was judged – attenuation – has remained quite considerably higher than standard all-glass fibers. In this paper, we describe the recent breakthroughs in hollow core fiber technology. We trace the story of this breakthrough from the theoretical exploration of a new design of hollow core fiber, through early implementations, up to the staggering results achieved over the last 18 months. The progress reported concerns not only a reduction in the fiber attenuation level, but also a considerable improvement in modal quality of the fibers, which have led to excellent data transmission performance. These fabricated fibers tell a story of improvements in all aspects of the technology, including preform preparation, performance modelling, fiber draw dynamics and coatings.

We will review physical and economic benefits of space division multiplexing (SDM) in submarine transmission. While multicore and multimode technology is still in research stage, SDM principles could be applied today using a multiplicity of single mode fibers. When power is limited, SDM approach allows one to achieve a significantly higher cable capacity compared to the traditional approach of maximizing per-fiber capacity. When looking at systems economy, SDM allows one to obtain the best cost per bit for the whole cable. This is achieved through optimization of number of spatial paths and sharing the pump power between the fiber paths.

In the space division multiplexing system, differential mode delay (DMD) is an essential parameter for transmission quality and signal processing. In this paper, we proposed the collective measurement method for the DMD of each core in fewmode multi-core fiber (FM-MCF) using low-coherence digital holography (LCDH). In the conventional method, given that the DMD of each core needs to be measured individually, the number of measurements increases in proportion to the number of cores, which causes fluctuation of the measurement conditions. In contrast, the proposed method reduces the measurement time and realizes the accurate measurement under the same conditions for all cores. We experimentally demonstrate the proposed method by using 6-mode 19-core fiber as a fiber under test. As a result, DMD values were successfully measured for all 19 cores.

Increasing optical fiber communication data speeds with mode division multiplexing, i.e., data transmission over the modes of few mode optical fibers, has received recent interest. As mitigation of mode crosstalk remains a prominent problem, over few kilometer lengths, the canonical solution via coherent detection based multiple-input-multiple-output digital signal processing may be prohibitive. In this presentation, we review the use of few mode elliptical core optical fibers for mode division multiplexing. Due to core ellipticity, an intrinsic reduction of mode crosstalk obviates its mitigation, thereby enabling direct detection based transmission over each mode. Dependence on transmitted mode alignment, optical fiber bending, and polarization mode dispersion, are discussed.

Photonic lanterns are novel photonic devices initially developed for an astro-photonics application, but later found immense potential in the field of space-division multiplexing and novel optical communication interconnects. They are adiabatically tapered structures, thus having large device lengths which scale quadratically with the number of cores. Adiabaticity is exploited to design a mode-selective photonic lantern. The Shortcuts to adiabaticity (STA) protocol is used to design an optimum quasi-adiabatic taper profile corresponding to a certain device length and a measure of adiabaticity. By homogenizing adiabaticity, a framework was developed to obtain an optimum taper profile for any required device length, which will have a corresponding measure of quasi-adiabaticity. This measure relates to the coupling losses suffered by the system digressing from absolute adibaticity. We have shown that the optimum taper profile significantly reduces the device length without compromising on adiabaticity. Photonic lanterns have also been examined for the generation of Orbital angular momentum modes in fibers. Here results for a three core photonic lantern structure for the generation of OAM₁ mode have been discussed.

In this research, we propose a mode selective switch (MSS) using volume holograms. MSS is a device that distributes the mode-division multiplexed (MDM) signal to different output ports for each spatial mode component contained in the signal. Using MSS, the function of reconfigurable optical add / drop multiplexer (ROADM) can be implemented, which can manipulate arbitrary spatial mode signal at any position in the next generation MDM network. In our proposed MSS, the incident signal is separated into spatial mode components by the volume hologram on the input side. The signal destination will be selected for each mode by diffracting independently. The spatial mode components are separated in direction of the plurality of volume holograms on the output side using a spatial light modulator (SLM). In the volume hologram on the output side, the multiple spatial mode components are recombined and emitted once again as an MDM signal. This method has the advantageous that one volume hologram can multiplex or de-multiplex multiple modes, allowing it to cope easily with the increase in the number of modes and ports to be multiplexed. In the experiment, the optical signal lights of three-mode are separated for each spatial mode by the volume hologram on the input side, and the switching to either of the two ports on the output side is successfully performed for each mode using the SLM.

In mode-division multiplexing (MDM) systems, transmission quality is restricted by differential mode delay (DMD). We have proposed and developed a spatial-mode exchange technique using volume holograms (VHET) as a leading technology to reduce DMD. VHET can equilibrate the transmission time of each spatial mode which has different transmission speeds in a few mode fiber using a volume hologram. This technology enables low signal distortion and high spectral efficiency, which are indispensable to the long-haul transmission for the MDM system. However, in this technology, modal cross-talks (MXTs), which are caused by inter-page cross-talks in the volume hologram, seriously degrade the performance of VHET. In this study, we proposed a method of combining a volume hologram and a random optical diffuser to reduce the MXTs. In our method, the intensity distribution of the input spatial mode is diffused uniformly by a random optical diffuser. The high exchange performance will be attained because the non-targeted holograms included in the multiplexed holograms do not affect most of the spatial mode. Moreover, our method can be applied to the communication wavelength bands using the dual wavelength method. We confirmed the basic operation of the proposed scheme using a linearly polarized mode group comprising LP0,1, LP1,1, and LP2,1. Compared with the conventional VHET, the simulation results show that the maximum MXTs are significantly suppressed from 0.4 to -15.8 dB.

Exponential growth of Internet traffic demands data center interconnect (DCI) systems to provide 400 Gb/s and higher per wavelength capacity under tight power consumption limitations for optical transceivers. We investigate the potential advantages of applying ultra-low loss and low dispersion fibers in DCI systems. Link optical signal to noise ratio (OSNR) and capacity analysis shows that ultra-low loss fiber (0.16 dB/km) provides significantly higher data capacity as compared with regular single-mode fiber (0.2 dB/km) for 80 km long DCI links. Also, the lower fiber attenuation reduces the required transceiver output power by 10 dB to achieve the same data capacity for 100 km DCI links. This implies substantial simplification in optical transceiver design. Digital chromatic dispersion compensation (CDC) is one of the major power consumers in optical transceivers. Our analysis shows that low dispersion fiber (4 ps/(nm·km)) reduces CDC computational complexity by 20% to 71% for different DCI link lengths versus regular single-mode fiber, indicating significant reduction in power consumption. Moreover, employing the CDC capability of the built-in adaptive filter in coherent receiver digital signal processing (DSP), the digital CDC unit could be completely removed using low dispersion fibers in DCI systems. Finally, we performed Monte-Carlo simulations of DCI links with different fiber types and confirmed the benefits of ultra-low loss and low dispersion fibers.

We developed a family of silica-based BDFAs operating over telecom O-band (1260-1360 nm). We demonstrated that 80 meters long single pump single stage amplifier can provide up to 19 dB gain, 20 dBm output power with 5 dB noise figure and 20% power conversion efficiency over 80 nm bandwidth (6-dB). The amplifier gain peak can be flexibly centered over 1305-1325 nm by pump wavelength selection. We designed simple BDFA operating over IEEE standardized part of the O-band (1272-1310 nm) and demonstrated that it can extend 425 Gb/s 400GBASE-LR8 transmission (eight 26.6 Gbaud/s PAM-4 channels) beyond 50 km of G.652 fiber.

In order to keep pace with the increasing data traffic, the next generation of optical server connections in a data center requires data rates of 400 Gbps at 100 Gbps per channel. Data center applications require highly integrated, low-cost solutions with very low power consumption and a very small package size. This can be achieved by co-integration of photonic and electronic function blocks. Optical signal processing with Nyquist pulses makes it possible to achieve very high data rates even at relatively low optical and electrical bandwidths. Nyquist pulses have the property of no inter-symbol interference and exhibit a rectangular spectrum, enabling transmission at the maximum possible symbol rate for a given bandwidth. Additionally, with one or two coupled modulators, pulse sequences with three or four times the RF-bandwidth of the single modulator can be achieved. This method neither requires a mode locked source, nor any other complicated equipment. Thus, the generation of Nyquist pulse sequences by integrated modulators is a very promising candidate for the integration on a silicon photonics platform. Within the proposed system modulators, receivers and electronics have only a bandwidth of 12.1 GHz, leading to Nyquist pulses with a bandwidth of 36.3 GHz that are modulated by PAM4 to reach a data rate of 72.6 Gbit/s. By interconnecting several such systems on one chip, the targeted 400 Gbit/s can be achieved in a single common transceiver module. Preliminary simulation results show the data transmission with low bit error rates.

Toward an ultra-compact optical coherent transceiver solution, recently, carrier-less full-field detection techniques based on phase-retrieval (PR) have been proposed. In these PR-based receivers, optical phase information is recovered digitally from multiple intensity-only measurements by using the PR algorithms, such as the Gerchberg-Saxton algorithm and the Wirtinger flow algorithm. Any local lights or reference lights such as in the Kramers-Kronig coherent receiver are not needed. In this talk, the recent advances in the PR receiver are introduced including the experimental demonstration of PR-based carrier-less coherent detection of polarization-multiplexed QPSK signals only by using a 2-D photodetector array.

In the context of short-reach data-center interconnect, such as OIF 400ZR, we study the system impact of bandwidth loss of Silicon Photonics Mach-Zehnder modulators in high-speed coherent modems. As well, we propose a minimum-mean square gradient-descent based method to optimize the compensation of such loss at both transmitter and receiver signal processing finite-impulse response filters. The method improves required-optical signal to noise ratio (ROSNR), under realistic hardware restrictions such as implementation noise and clock jitters at data converters, by 0.5 dB compared to fully pre-compensating the frequency response. Other advantages, such as lower power consumption, are highlighted as well.

We report on an integrated-optic spectrum synthesis circuit that can individually control amplitude and phase of 40 GHzspaced and 64 frequency components. This synthesis circuit is fabricated with silica waveguide technology, and is composed of two arrayed-waveguide gratings (AWGs), and an array of variable optical attenuators (VOAs) and phase shifters. The AWGs function as frequency multi/demultiplexers, and the VOA and phase shifter are used to adjust amplitude and phase of each spectral component, respectively. The size of the circuit is 80 mm x 80 mm, and the manipulation accuracy relating to the VOA and phase shifter is in the order of 0.1 dB and 0.01, respectively. The VOA extinction ratio, namely, the dynamic range is about 30 dB.

Here a novel modulator design containing a graphene double layer based on the Fabry-Perot interferometer is designed and fabricated. This proposed device bypasses the power consumption problem by enhancing the optical absorption in graphene through enlarging the interaction of light with graphene layers by > 40-fold compared with conventional designs. Due to the direct dependency of applied voltage with respect to the absorption in graphene, sub-volt operation with an adequate modulation depth (>5 dB) is expected from this device. Only few fJ/bits power consumption is required to ambulate the Fermi-level from Dirac to the desired point in Pauli Blocking mechanism.

The explosive increase in data traffic over the last few decades has been driving the demand for high-capacity optical transmission systems. State-of-the-art spectrally efficient technologies such as high-order modulation formats, probabilistic constellation shaping (PCS) and high-performance forward error correction (FEC) have realized capacities very close to the theoretical limit of the wavelength bands used. The widely-used erbium-doped fiber amplifier (EDFA) is unable to support another wavelength-division multiplexing (WDM) channel because few amplification bands remain unused. Ultra-wideband (UWB) transmission with extra bandwidths, e.g., S, C and L bands, is a promising candidate for expanding transmission capacity. UWB systems can enlarge capacity without replacing any of the existing fiber infrastructure, which offers dramatic efficiencies in the cost and delay for system deployment. Recently, 100-Tb/s-class UWB transmission has been experimentally demonstrated, and the S-band region is regarded as the next candidate beyond conventional C and/or L band WDM systems. Designing such systems demands an understanding of how interchannel stimulated Raman scattering (ISRS) impacts WDM-system performance; because the S and L bands are separated by around 100 nm, ISRS is a significant issue.

In this paper, we investigate the effects of ISRS on signal quality of UWB transmission systems with experiments on S- (35 channel) and L- (40 channel) band WDM transmission using DP-16QAM signals. The results show that ISRS causes only signal power transfer, whereas the nonlinear cross-talk generated by ISRS has only minimal effect on signal quality. We prove the concept of UWB transmission by a DP-128QAM 150.3-Tb/s transmission experiment over a 40- km single-mode fiber in S, C, and L bands with WDM bandwidth of 13.6 THz; success is due to our proposed signal power optimization scheme which considers ISRS-induced power transfer between S and L bands.

We introduce a new technique for obtaining the frequency offset introduced at the receiver due to the heterodyning of the transmitter laser and the local oscillator. This technique is needed for multi-subcarrier systems, as the offset must be removed without knowledge of the modulation format, making well-known algorithms that correct using it impractical. By detecting the spectral edges for the transmitted and received signals, it is possible to get a coarse estimate of the intermediate frequency by finding the difference between these two values. We then use this approximation as a starting point to fine search for the true value, by taking a window of possible values around the approximation. By taking the cross-correlation between the received data and a subset of the transmitted data for the subcarriers, we can verify whether the correct value is chosen, and if not, move on to the next estimate.

We discuss various applications of machine learning techniques in different aspects of optical communications and networking including optical performance monitoring, fiber nonlinearity compensation, cognitive network failure prediction, dynamic planning and cross-layer optimization of software-defined networks, quality of transmission estimation, and physical layer design of optical communication systems. Recent works employing deep learning technologies are also discussed.

Optical transceiver imperfection including device bandwidth limitation, various tributary imbalances and even the PCB induced long-memory ripple and tributary crosstalk is one of the biggest obstacles impeding the high-order QAM coherent systems towards 100GBaud and beyond. Generally, the static imperfections could be calibrated at the preservice stage. The dynamic imperfections should be handled at the in-service stage. This talk will firstly review the instrument-free calibration process for the pre-service stage. Then various digital monitors for diagnosis and fault allocation either at pre-service stage or in-service stage, relying on the inherent monitoring function of receiver-side digital equalizers and the machine learning tool, will also be reviewed.

In order to serve the changing needs of road traffic control, the road space and road structure surrounding an intersection have evolved into complex forms. Using a new concept of request/response in a two-way-to-way traffic light controlled crossroad, the redesign of the trajectories can be accomplished by the application of methods for navigation, guidance and combination of expert knowledge of vehicle road traffic control. In this work, 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 to broadcast the information. Vehicle headlamps are used to transmit data to other vehicles or infrastructures, allowing digital safety and data privacy. Data is encoded, modulated and converted into light signals emitted by the transmitters. Tetra-chromatic white sources are used, providing a different data channel for each chip. As receivers and decoders, SiC Wavelength Division Multiplexer (WDM) devices, with light filtering properties, are considered. The primary objective is to control the arrival of vehicles to an intersection and schedule them to cross at times that minimize traffic delay. A further objective is to allocate delays between left-turns and forward movements, moderating the speed and slot between vehicles travelling in these directions, maintaining a safe distance from one to another. Pedestrians and bicycles are also incorporated. An I2V2V2I traffic scenario is proposed, and bidirectional communication between the infrastructure and the vehicles is tested, using the VLC request/response concept. A phasing traffic flow is developed as a proof of concept. The experimental results confirm the cooperative VLC architecture, showing that communication between connected cars and infrastructures can be optimized using the mentioned request/response concept. A significant increase in traffic throughput with the least dependency on infrastructure is achieved.

We present optimized structure of waveguide photodetector (WGPD) having large responsivity in spite that its absorption layer is thin. Through beam-propagation method simulations, spot-size converter integrated WGPD is found to be able to have larger responsivity with thinner absorbing layer. Calculated responsivity and polarization dependent loss of the optimized design are 0.83 A/W and 0.13 dB with 150-nm-thick and 25-μm-long absorber.

A novel low cost millimetre-wave over fiber (MMWoF) network based on Filter-bank multicarrier (FBMC) waveform with optical heterodyning technique is presented over 30-km standard single mode fiber (SSMF) as well as 10-m radio frequency (RF) wireless link. In the proposed scheme, five sub-bands FBMC signal of aggregated bit rate of 12 Gb/s to meet the demand for high data rate wireless connectivity and coherent heterodyning technique by beating two coherent optical tones without use of costly electronic components have been proposed for low phase noise 59.2 GHz millimeter wave (MMW) signal generation respectively. FBMC method is a promising option for 5G, which not only reduces out - of-band emissions but also overcomes the synchronization requirement of OFDM. A passively mode locked laser diode is used as a cost-efficient source since it exhibits a wide spectrum of equally spaced optical phase-locked modes and produce correlation of phase noise between the optical lines. High receiver sensitivity, small power penalty and very good spectra, low error vector magnitude (EVM) value, and excellent constellation diagrams in our proposed system make more authentic and stable with acceptable performance. Therefore, proposed MMWoF system could be the feasible solution for next -generation ubiquitous wireless network.

We investigated and compared the performances of three different types of optical twin-single sideband (twin-SSB) modulation methods. Twin-SSB can transmit two different signals on lower side-band (LSB) and upper side-band (USB), and has been studied as a technique to improve spectral efficiency. Twin-SSB arranges the spectra side-by-side without any spectral spacing. Dispersion tolerance of twin-SSB is as high as that of conventional SSB schemes. The twin-SSB modulation can be performed by using a dual-polarization quadrature phase shift keying (DP-QPSK) modulator, a QPSK modulator, or a single Mach-Zehnder modulator (MZM). The twin-SSB with the three modulation methods need to be modulated at a limited linear domain of the optical field output function of the modulator. Therefore, if we apply a large electrical modulation signal to the modulators to decrease the optical loss, the nonlinear characteristics of the modulators degrade the performance of the twin-SSB signals. We investigated the modulation performance by changing the amplitude of the electrical modulation signal for the modulators, and investigated the optimal conditions of the twin-SSB modulation methods. We also performed numerical simulations of 50-km standard single mode fiber (SSMF) transmission of the twin-SSB signals with the optimal modulation conditions, and evaluated the error vector magnitude (EVM) performances of the transmitted signals. The results revealed that the EVM performances of the twin-SSB signals using a DP-QPSK modulator and a QPSK modulator were superior to that of a single MZM.

Using the vertical integration of the Synopsys environment, we analyze a 2 2 integrated optical switch obtaining a layer-0 abstraction used to analyze the impact of the design options on transmission performances of a PM-64QAM 600G channel in multi-hop routing in meshed optical networks. The optical switch is designed targeting the Analog Photonics Process Design Kit. The QoT degradation depending on the design option and on the choice for the transmission technique is assessed, taking into account the number of traversed switches. In addition, different routing techniques for the integrated optical waveguides of the 2x2 switches are investigated in terms of system performances. The reported analysis is an example of comprehensive investigation carried out by abstracting the network elements starting from the component design up to the networking management. This approach is today mandatory to enable the maximum capacity in state-of-the art optical networks. To face this challenging problem, Synopsys proposes a vertically integrated software environment for the design of optical communication systems with photonic integrated circuits: it is the integration of OptSim c -optical communication system, OptSim Circuit -schematic-driven photonic circuit, OptoDesigner c -mask layout, and RSoft component design tools. These tools have proven to be reliable aids to virtually designing and estimating the performance of optical transmission systems and photonic chips.