This PDF file contains the front matter associated with SPIE Proceedings Volume 10130, including the Title Page, Copyright information, Table of Contents, and Conference Committee listing.

In this talk, ,we present classical key distribution (CKD) in optical communication.including steganography/bounded-storage CKD and 2) random channel cryptography/bounded-observability CKD.

This talk will review our studies on the mode coupling mechanism and differential modal group delay characteristics of the coupled multi-core fiber, and the fabrication results of 125-μm-cladding coupled four-core fibers which realized the record-low spatial mode dispersion (SMD) of 3.14±0.17 ps/√km over C-band and the ultra-low attenuation of 0.158 dB/km at 1550 nm, both of which are the lowest ever reported among optical fibers for the space-division multiplexed transmission. By assuming the 3.14-ps/√km SMD accumulation, the tap count of the multiple-input-multiple-output digital signal processing for the crosstalk compensation is estimated to be only 63 taps for covering 99.99% power of the impulse response after 10,000-km propagation when the 25-GBaud (50-GHz sampling) is assumed. The present results demonstrate the strong and practical applicability of the coupled multi-core fiber to the ultra-long-haul transmission system.

Space division multiplexing (SDM) utilizing few-mode fibers or multicore fibers supporting multiple spatial channels, is currently under intense investigation as an efficient approach to overcome the current capacity limit of high-speed long-haul transmission systems based on single mode optical fibers. In order to realize the potential energy and cost savings offered by SDM systems, the individual spatial channels should be simultaneously multiplexed, transmitted, amplified and switched with associated SDM components and subsystems. In this paper, recent progress on the implementation of various SDM amplifiers and its related SDM components is presented.

Integrated space-division multiplexed (SDM) erbium-doped fiber amplifiers (EDFAs) are not only inevitable for SDM systems, but can be an alternative solution to nowadays EDFA array for parallel amplification. SDM EDFAs are expected to provide substantial complexity and cost savings through spatial-integration compared to duplicating single-mode fiber amplifiers. High output power and low noise figure can be achieved by cladding-pumped SDM EDFAs. In this paper, different cladding pumping solutions, cladding-pumped single-mode and multimode multi-core EDFAs will be discussed.

Self-confined beams and spatial solitons were always investigated for a purely academic point of view, describing their formation and cross-interaction. We propose a novel paradigm for integrated photonics circuits based on self-confined interconnections. We consider that circuits are not designed since beginning; a network of writing lasers provide the circuit configuration inside which information at a different wavelength travels. we propose new designs for interconnections and both digital and analog switching gates somehow inspired by Nature, following analog decision routes used in biological networks like brain synapsis or animal path finding.

We investigate the application of space division multiplexing (SDM) technology to a long haul undersea transmission using multi-core fiber (MCF). We discuss two experimental demonstrations conducted with a 12-core fiber. The first study shows that SDM can help to achieve simultaneous improvement of power efficiency and capacity of long-haul transmission. Using SDM, power efficient optical system design and modulation format with high receiver sensitivity we demonstrate 105.1 Tb/s transmission over 14,350 km with pump power per 12 EDFAs equivalent to a single 800 mW 980 nm pump laser. In the second study, we investigated capacity limits of transoceanic transmission achievable with MCF and extended C and L bandwidth of optical amplification. The study shows 520 Tb/s potential transmission capacity over 8,830 km using 9 THz of optical bandwidth. To achieve this capacity, we also used new coded modulation with 1.0 dB sensitivity improvement compared to 8QAM with the same SE of 4.86 b/s/Hz.

The Asymmetric Directional Coupler (ADC) based on SOI (Silicon-on-Insulator) technology converts and couples the fundamental mode to the first higher order mode. The ADC is designed to achieve phase-matching condition, which is accomplished when both propagation constants are equal in each waveguide arm. Devices are fabricated in a SOI wafer with a 220 nm thick silicon layer. The refractive indexes of Si and SiO₂ are n_Si=3.47 and n_Si02=1.46 respectively. The access waveguides (W₁=0.45 μm) have been designed to propagate just the fundamental mode, TE₀. The optimum width for the second waveguide was chosen to achieve the phase-matching condition for the TE₁ mode, which corresponds to W₂=0.962 μm. The coupling to the input and output waveguides is achieved through grating couplers. The input grating coupler will need to couple the LP₀₁ mode from the SSMF (Standard Single-Mode Fiber) to the TE₀ mode in the SOI waveguide; thus a typical design for a SOI coupler can be used. However, the output coupler must simultaneously couple the TE₀ and TE₁ modes in the SOI wide waveguide to the LP₀₁ and LP₁₁ modes in the FMF (Few-Mode Fiber). Input gratings are designed to have an area of 12x12 μm² and a period of Λ=610 nm in order to maximize the optical power coupled between the fiber and the waveguide for an incident angle of 10 degrees. Output gratings are designed with the same period but distinct area (12.5x12.5 μm²) to correctly couple the LP₀₁ and LP₁₁ modes in the FMF.

All-fiber mode selective coupler (MSC) is comprised of a few mode fiber (FMF) and a single mode fiber (SMF), coupling the LP₀₁ mode of the SMF to a specific higher-order mode (HOM) of the FMF. In order to achieve high coupling ratio and low insertion loss, phase-matching condition between the LP₀₁ mode of SMF arm and the HOM of FMF arm should be satisfied. A polished-type MSC is made by getting their cores into intimate contact. Prism coupling with a polished coupler block can measure the effective refractive index of the mode accurately. We propose and demonstrate three kinds of allfiber mode multiplexer that is composed of consecutive MSCs. 4-mode multiplexer can multiplex 4 modes of LP₀₁, LP₁₁, LP₂₁, and LP₀₂ by cascading LP₁₁, LP₂₁, and LP₀₂ MSCs. It is used for MDM transmission of three modes with 120 Gb/s DP-QPSK signals. In order to enhance the signal transmission performance by receiving degenerate LP modes simultaneously, a mode multiplexer to utilize two-fold degenerate LP₁₁ modes is proposed. It is composed of two consecutive LP₁₁ MSCs that allows the multiplexing of LP01 mode and two orthogonal LP₁₁ modes. We demonstrates WDM transmission of 30 wavelength channels with 33.3 GHz spacing, each carrying 3 modes, over 560 km of FMF. 6- mode multiplexer can multiplex 6 modes of LP₀₁, LP_11a, LP_11b, LP_21a, LP_21b, LP₀₂ modes. We demonstrated WDM-MDM transmission with the all-fiber 6-mode multiplexer. In this paper, the manufacturing method and the recent advancements of the all-fiber mode multiplexer based on the MSCs are reviewed. Long-distance mode division multiplexing (MDM) optical signal transmissions with the all-fiber mode multiplexer are experimentally demonstrated.

Recently, mode-division multiplexing (MDM) has been investigated in transmission systems and optical access networks for capacity enhancement. In this paper, recent progress in MDM optical passive networks (PON) enabled by low-modal crosstalk few-mode fibers (FMFs) and optical components is reviewed. All-fiber mode multiplexer/demultiplexers (MUX/DeMUX) composed of mode-selective couplers (MSCs) are introduced to simultaneously multiplex or demultiplex multiple modes. Few-mode circulators (FMCs) are employed to realize bidirectional MDM-PON transmission. Direct detection without complex multiple-input-multiple-output (MIMO) digital signal processing (DSP) is applied. Moreover, multidimensional PONs by cascading MDM optical distribution network (ODN) with conventional time-division multiplexing (TDM) and wavelength-division multiplexing (WDM) ODNs are proposed and experimentally demonstrated.

We present research into the capacity and spectral efficiency of various spatial and mode division multiplexed transmission schemes in multimode optical fibres, including MIMO schemes based linearly polarized (LP) modes and orbital angular momentum (OAM) modes. We estimate their spectral efficiency by calculating the transmission matrix and considering mode coupling under different schemes. We also consider the signal-processing prefix for LP-MIMO and OAM based technologies to calculate the effective spectral efficiency. Simulation results show LP-MIMO using step-index fibre can support higher spectral efficiency than both LP-MIMO with graded-index fibre and OAM, while LP-MIMO with gradedindex fibre has similar levels of spectral efficiency as OAM. We further compare the computational resources required to support the signal processing algorithms in order to achieve certain amount of capacity or spectral efficiency and reveal the scaling of computational resources with capacity in different MDM schemes. Considering the implementation of MIMO algorithms and any processing required to support spatial light modulator (SLM) based optical crosstalk equalization schemes, we demonstrate that OAM-based MDM has the lowest computational resource requirement and offers a much higher spectral efficiency with limited computational resource in both processing power and memory size requirements resulting from intermodal group velocity dispersion in multimode fibres. We also show that OAM based schemes could save significant signal processing energy overhead due to its low computational resource requirement.

We present the novel 7-core and 19-core hole-assisted fibers designed to satisfy the most demanding requirements of the ITU-T G.657.B3 recommendation for bend-insensitive fibers. The fibers are compatible with standard single-mode fibers with regard to modal properties, dispersion characteristics, and transmission loss. The fibers presented exhibit no crosstalk and it is possible to use them together with other multiplexing methods like CWDM or DWDM. Dedicated fanin/ fan-outs have been created in order to make immediate use in industry possible. The hole-assisted 19-core fiber with single-mode cores is being presented for the very first time.

Bandwidth and capacity demand in metro, regional, and long-haul networks is increasing at several tens of percent per year, driven by video streaming, cloud computing, social media and mobile applications. To sustain this traffic growth, an upgrade of the widely deployed 100-Gbit/s long-haul optical systems, based on polarization multiplexed quadrature phase-shift keying (PM-QPSK) modulation format associated with coherent detection and digital signal processing (DSP), is mandatory. In fact, optical transport techniques enabling a per-channel bit rate beyond 100 Gbit/s have recently been the object of intensive R and D activities, aimed at both improving the spectral efficiency and lowering the cost per bit in fiber transmission systems. In this invited contribution, we review the different available options to scale the per-channel bit-rate to 400 Gbit/s and beyond, i.e. symbol-rate increase, use of higher-order quadrature amplitude modulation (QAM) modulation formats and use of super-channels with DSP-enabled spectral shaping and advanced multiplexing technologies. In this analysis, trade-offs of system reach, spectral efficiency and transceiver complexity are addressed. Besides scalability, next generation optical networks will require a high degree of flexibility in the transponders, which should be able to dynamically adapt the transmission rate and bandwidth occupancy to the light path characteristics. In order to increase the flexibility of these transponders (often referred to as “flexponders"), several advanced modulation techniques have recently been proposed, among which sub-carrier multiplexing, hybrid formats (over time, frequency and polarization), and constellation shaping. We review these techniques, highlighting their limits and potential in terms of performance, complexity and flexibility.

Recent research in multidimensional modulation has shown great promise in long reach applications. In this work, we will investigate the origins of this gain, the different approaches to multidimensional constellation design, and different performance metrics for coded modulation. We will also discuss the reason that such coded modulation schemes seem to have limited application at shorter distances, and the potential for other coded modulation schemes in future transmission systems.

Intradyne coherent receiver (ICR) is an essential component for modern coherent systems. Critical parameters of ICR include frequency response, bandwidth, and gain imbalance. Conventionally one creates heterodyne beating between tunable laser source (TLS) and local oscillator, and sweeps frequency of TLS to measure those parameters. A complicated control loop aligns state of polarization (SoP) of TLS to 45/135 degree of principle axis of polarization beam splitter. Otherwise unequal amount of power will launch into two polarizations, leading to inaccurate result. To overcome this complexity, we present a novel scheme to characterize ICR using polarization-multiplexed laser source generated using internal components of coherent transmitter. The optical signals on two polarizations are modulated with sinusoidal signal at different frequencies. If frequency difference is larger than laser line width, the output at the coherent transmitter is a polarization-multiplexed laser source. This method allows on-board measurement of analog front-end of coherent receiver by connecting coherent transmitter with coherent receiver. Influence of automatic bias control and radio frequency amplifier is also discussed. Another method is to combine outputs of two separate TLSs with orthogonal polarization through a polarization beam combiner. Both methods lead to robust performance without active control of SoP. The measurement accuracy using polarization-multiplexed laser source is the same as that of the conventional method. The intrinsic resiliency to polarization change leads to simple setup and enables in-field characterization of ICR.

To cope with the exponential growth of the Internet traffic, optical communications has advanced by leaps and bounds. For several decades, Intensity modulation with direct detection (IM-DD) dominates the commercial short-reach optical communications. However, when upgrading the data-rate distance product to 1000 Gb/s·km per wavelength and beyond, IM-DD faces severe performance barrier. Aiming to improve the electrical SE and extend the transmission distance, advanced DD modulation formats have been proposed through a so-called self-coherent (SCOH) approach, where a carrier is transmitted together with the signal to achieve a linear mapping between the electrical baseband signal and the optical field. In that way, the impact of the CD can be removed from the received signal, greatly extending the transmission distance of the DD system. Particularly, Stokes-vector direct detection (SV-DD) has been proposed to realize linear complex optical channels as well as enhance the electrical spectral efficiency and transmission reach. In this talk, we present the principle and discuss the performance of SV-DD systems.

Spatial-division multiplexing (SDM) techniques have been purposed to increase the capacity of optical fiber transmission links by utilizing multicore fibers or few-mode fibers (FMF). The most challenging impairments of SDMbased long-haul optical links mainly include modal dispersion and mode-dependent loss (MDL), whereas MDL arises from inline component imperfections, and breaks modal orthogonality thus degrading the capacity of multiple-inputmultiple- output (MIMO) receivers. To reduce MDL, optical approaches include mode scramblers and specialty fiber designs, yet these methods were burdened with high cost, yet cannot completely remove the accumulated MDL in the link. Besides, space-time trellis codes (STTC) were purposed to lessen MDL, but suffered from high complexity. In this work, we investigated the performance of space-time block-coding (STBC) scheme to mitigate MDL in SDM-based optical communication by exploiting space and delay diversity, whereas weight matrices of frequency-domain equalization (FDE) were updated heuristically using decision-directed recursive-least-squares (RLS) algorithm for convergence and channel estimation. The STBC was evaluated in a six-mode multiplexed system over 30-km FMF via 6×6 MIMO FDE, with modal gain offset 3 dB, core refractive index 1.49, numerical aperture 0.5. Results show that optical-signal-to-noise ratio (OSNR) tolerance can be improved via STBC by approximately 3.1, 4.9, 7.8 dB for QPSK, 16- and 64-QAM with respective bit-error-rates (BER) and minimum-mean-square-error (MMSE). Besides, we also evaluate the complexity optimization of STBC decoding scheme with zero-forcing decision feedback (ZFDF) equalizer by shortening the coding slot length, which is robust to frequency-selective fading channels, and can be scaled up for SDM systems with more dynamic channels.

This invited talk reviews recent technology advances in our three generations of low-power CMOS coherent digital signal processor (DSP) implemented with 40, 20, and 16-nm CMOS technologies, with highlights on its functional integration, adaptation, and design optimization for power-efficiency CMOS DSP implementation. The latest 16-nm third-generation (Gen3) DSP implementation achieves sub-10-watt per 100 Gb/s coherent transmission in both 100G DP-QPSK and 200G DP-16QAM transport modes for the first time, and experimentally confirms its trade-off between transmission performance and power dissipations.

We investigate the individual and combined performance of correlated digital back propagation (CDBP) and extended Kalman filtering (EKF) in mitigating inter and intra-channel non-linearities in wavelength division multiplexed (WDM) systems. The afore-mentioned algorithms are verified through numerical simulations on 28 Gbaud polarization multiplexed (PM) 16-quadrature amplitude modulation (16-QAM) 9-channel WDM system with 50 GHz spacing. A single channel CDBP with one-step-per-span based on asymmetric split step Fourier method (A-SSFM) with optimized non-linear coefficient has been employed. We also study an amplitude dependent optimization (AO) of the non-linear coefficient for CDBP which shows an improvement of ≈ 0.8 dB compared to the conventional optimized CDBP, in the non-linear regime. Moreover, our proposed carrier phase and amplitude noise estimation (CPANE) algorithm based on EKF outperforms AO-CDBP in both linear and non-linear regimes with an enhanced performance besides significantly reduced complexity. We further investigate the combined performance of AO-CDBP and EKF which results in an enhanced non-linear tolerance at the expense of increased computational cost trading off to the number of required CDBP steps per span. Furthermore, we also analyze the impact of cross phase modulation (XPM) on the combined performance of AO-CDBP and EKF by varying the number of WDM channels. Numerical results show that the obtained gain from employing AO-CDBP prior to EKF reduces with increasing effects of XPM. Additionally, we also discuss the computational complexity of the aforementioned algorithms.

Different approaches to design modulation formats have been reviewed to improve granularity and receiver sensitivity. Set-partitioning (SP) quadrature amplitude modulation (QAM) is found to be better than time-hybrid QAM at the same spectral efficiency. Multi-dimensional QAM optimized by pairwise optimization can outperform SP-QAM by ~ 0.5 dB. The results also suggest that modulation formats have to be designed at a reasonable system margin rather than comparing forward error correction (FEC) limit.

DP-QAM is one of the most promising paths towards 400-Gb/s and 1-Tb/s commercial optical communications systems. For DP-QAM transmitter, different tributary channel powers lead to IQ or XY power imbalance. Large uncompensated IQ or XY power imbalance can significantly degrade the performance in the coherent optical communications system. In this work, we propose and experimentally demonstrate a technique to detect and compensate DP-QAM transmitter power imbalances for tributary channels. By reconfigurably interfering de-skewed identical BPSK channels, the optical powers of any two tributaries can be balanced by minimizing the output power from their optical interference.

In today’s fiber-optic communication systems, the bandwidth of the photonic components, i.e. modulators and photo diodes, is way greater than that of their electrical counterparts, i.e. digital-to-analog converters (DACs) and analog-to-digital converters (ADCs). In order to increase the transmission capacity, the bandwidth limitations need to be overcome. We review the progress and the recent results in the field of high-speed DACs, which are desirable for software-defined transmitters. Furthermore, we evaluate interleaving concepts regarding their ability to overcome the above mentioned limitations and demonstrate recent experimental results for a bandwidth interleaved DAC with 40 GHz analog electrical bandwidth.

In this invited paper, we summarize the current developments in linear optical field modulators (LOFMs) for coherent multilevel optical transmitters. Our focus is the presentation of a new, novel LOFM design that provides beneficial and necessary features such as lowest hardware component counts, lowered insertion loss, smaller RF power consumption, smaller footprint, simple structure, and lowered cost. We refer to this modulator as called Double-Pass LOFM (DP-LOFM) that becomes the building block for high-performance, linear Dual-Polarization, In-Phase- Quadrature-Phase (DP-IQ) modulator. We analyze its performance in term of slope linearity, and present one of its unique feature --- a built-in compensation functionality that no other linear modulators possessed till now.

In this paper, we propose a linear polarization coding scheme (LPC) combined with the phase conjugated twin signals (PCTS) technique, referred to as LPC-PCTS, for fiber nonlinearity mitigation in coherent optical orthogonal frequency division multiplexing (CO-OFDM) systems. The LPC linearly combines the data symbols on the adjacent subcarriers of the OFDM symbol, one at full amplitude and the other at half amplitude. The linearly coded data is then transmitted as phase conjugate pairs on the same subcarriers of the two OFDM symbols on the two orthogonal polarizations. The nonlinear distortions added to these subcarriers are essentially anti-correlated, since they carry phase conjugate pairs of data. At the receiver, the coherent superposition of the information symbols received on these pairs of subcarriers eventually leads to the cancellation of the nonlinear distortions. We conducted numerical simulation of a single channel 200 Gb/s CO-OFDM system employing the LPC-PCTS technique. The results show that a Q-factor improvement of 2.3 dB and 1.7 dB with and without the dispersion symmetry, respectively, when compared to the recently proposed phase conjugated subcarrier coding (PCSC) technique, at an average launch power of 3 dBm. In addition, our proposed LPCPCTS technique shows a significant performance improvement when compared to the 16-quadrature amplitude modulation (QAM) with phase conjugated twin waves (PCTW) scheme, at the same spectral efficiency, for an uncompensated transmission distance of 2800 km.

In this paper, an effective design of all-optical logic gates like XOR gate and AND gate is presented. The structure of these two logic gates is based on T-shape waveguide with optimized silica dielectric rod. Along with the two input ports which are essential for the required logical operation, an extra reference input port is used. These two logic gates can be used to construct for various combinational logic circuits, data bit comparison circuits, pattern matching, data encoding/decoding and different switching functions etc.

The trend of increasing data traffic in conventional communication systems demands utilizing new methods for data transmission, which in combination with traditional techniques, enable overcoming the predicted capacity limit. Mode division multiplexing (MDM), where higher-order modes (HOMs) in a few-mode fiber (FMF) are used as multiple spatial communication channels, comes in this context as a viable approach to enable the optimization of high-capacity links. From this perspective, it becomes highly necessary to possess a diagnostic tool for the precise modal characterization of FMFs. Among existing approaches for modal content analysis, several methods as S², C² in time and frequency domain are available. In this contribution we will present an improved time-domain cross-correlated (C²) imaging technique for the experimental evaluation of modal properties in HOM fibers over a broad range of wavelengths. Our modified setup makes it possible to adjust the time resolution of the system according to the needs of the required fiber measurement. We show that by tuning the spectral shape of the source (SuperK EXTREME filtered by SuperK Select), we enhance the time resolution of the system, which allows us to distinguishing differential time delays between HOMs in the picosecond timescale. Broad wavelength scanning in combination with spectral shaping, allows us to estimate the modal behavior of FMF without prior knowledge of the fiber parameters. We performed our demonstration at wavelengths from 850nm to 1100nm which can be easily extended to other wavelengths of interest just by replacing components with the appropriate coating. The method presented here aims to serve as flexible diagnostic tool that can be implemented in MDM systems for judicious evaluation of modal dispersion in FMFs.