This PDF file contains the front matter associated with SPIE Proceedings Volume 12028, including the Title Page, Copyright information, Table of Contents, and Conference Committee listings.

.

Cladding pumping in optical amplifiers has recently seen rising interest in the context of multicore amplifiers. The use of multicore erbium-doped fibers (EDFs) can decrease the complexity of amplifier subsystems since it allows pumping with a single high-power multimode laser diode for multiple parallel amplification channels in the same fiber cladding. Due to the lower pump intensity in cladding-pumped amplifiers, high population inversion is harder to achieve compared to core-pumped amplifiers, which favors L-Band applications over C-Band applications. The various designs of cladding-pumped fiber amplifiers that have been proposed to amplify the L-Band tend to use either homogeneous erbium doping or homogeneous erbium/ytterbium co-doping in the core. In this paper, we demonstrate that a carefully engineered combination of erbium (Er) doped layers and erbium-ytterbium (Er/Yb) co-doped layers in a single core (concentric layers with heterogeneous doping) can improve the power conversion efficiency (PCE) compared to homogeneous doping designs

The main idea behind this investigation is to use this additional degree of freedom in the design to better control the erbium ion population inversion along the fiber, which helps to minimize the amplified spontaneous emission and thus increases the PCE. Through numerical simulations, we compare the PCE and noise figure (NF) of four types of homogeneous doping geometries in the core (annular Er, annular Er/Yb, central Er and central Er/Yb) with a design based on concentric layers with heterogeneous doping. For all the scenarios, the doping geometry and fiber length are carefully optimized to maximize the PCE. The input signal power, input pump power, pump propagating direction, refractive index profile and other not-doping-related parameters are kept constant for all the scenarios.

In this work, two experimental methods are proposed to quantify the contributions of the different light scattering mechanisms to the total attenuation of each guided-mode of two weakly-coupled FMF presenting different index profiles. The first approach is based on the analysis of the angular distribution of the light scattered by each mode of the two fibers whereas the second technique is based on bidirectional OTDR measurements also performed for each mode of the two fibers at different wavelengths. Some preliminary examples are reported and discussed for the fundamental mode and one higher order mode guided in a step-index and a trapezoidal index profile few-mode fiber.

With the number of modes in photonic lantern (PL) increasing, a longer adiabatic taper is required, which is not feasible for fabrication. A cascaded-taper-based PL is proposed as a type of efficient and compact multiplexer/demultiplexer. It has many advantages to design and draw the PL through multi-taper cascade. For the double-taper PL with its longitudinal profile optimized, the coupling efficiency of mode group LP01 and LP11 reaches its optimal value of 90.8% and 90.0% when the length is 5.15 cm, while the corresponding efficiency of the single-taper PL is only 74.6% and 85.6%, respectively. Besides, the triple-taper PL optimized with the same material has a length of 4.65 cm. The optimal coupling efficiency of mode group LP01 and LP11 is 89.5% and 89.1%, while the single-taper PL with the same length is only 72.6% and 84.4%, respectively. Without loss of generality, the optimal adiabatic taper profiles of the two-modegroup PLs are obtained by beam propagation method. As a comparison, only when the length of the single-taper twomode- group PLs is 9 cm, the coupling efficiency of mode group LP01 and LP11 reaches the optimal values of 92.4% and 94.1% respectively. Although the optimal coupling efficiency of the double-taper and triple-taper PLs is slightly lower than that of the optimal single-taper ones, they will exhibit better performance when the length must be shorter.

In order to construct 5G and beyond 5G networks effectively, it becomes important to prepare a sufficient number of interfaces with optical fiber at arbitrary locations. An optical wavelength- and power-tapping device is expected to provide a flexible junction point in optical networking, and a fiber-based optical device has great advantages in terms of low-loss and high compatibility with existing single-mode fibers. In this paper, we propose a wavelength- and powertunable in-fiber optical tap inscribed in a two-mode fiber (2MF) using a femtosecond laser. This optical tap is composed of a long-period fiber grating (LPG) and an asymmetric tap waveguide that is inserted transversely across the core to the cladding. In the LPG, transmitted LP01-mode light is converted into LP11 mode, and the converted LP11 mode is coupled to the tap waveguide while the LP01 mode passes through the fiber core. The operation wavelength and branching-power ratio can be controlled by the LPG design. We have revealed the optimum design for the tap waveguide. Here, the relationship among diameters of core and tap, the relative refractive indices of core and tap, and the angle between fiber core and tap was considered. We also experimentally confirmed the coupling efficiency to the core and tap when LP01 and LP11 modes were separately launched by using a tap waveguide inscribed in a 2MF. The factors that degraded the coupling efficiency were discussed by comparison with the calculation results.

Breakthroughs of high-capacity transmission technologies have been demanded. Obviously, one strong candidate is space division multiplexing (SDM) using multi-core fibers and/or few mode fibers. Recent progress in this area enabled >10Pb/s transmissions using multi(few)-mode multi-core fibers. However, using modes still increase the complexity of fiber properties measurements as well as the system configurations. One relatively straightforward way to increase space efficiency is increasing number of cores but this direction is limited by the cladding diameter. Here, we will report recent research trials and progresses to solve these trade-offs.

Small coherent optical transceivers have been demanded to provide a high data rate density in next-generation metro area networks and data-center interconnects. Therefore, the mechanical sizes of optical modules and sub-assemblies have been shrinking, and accordingly, highly integrated optical modules have been demanded. We present two types of highly integrated optical modules for ≥400-Gb/s coherent optical links. Our wavelength-tunable laser module consists of an ultra-compact DBR/Ring laser chip and a PLC-based wavelength locker. The laser module covers the wavelength range of full C-band with a narrow linewidth of >100 kHz and a fiber output power of 17 dBm. Furthermore, we demonstrate a tiny transmitter-receiver optical sub-assembly that integrates all required optical components into a single gold-box of 26 × 14 × 4 mm in compliant with the Optical Internetworking Forum (OIF) specifications. The electrical-optical and optical-electrical bandwidths measure >40 GHz, which is wide enough to realize ≥400-Gb/s per lambda operations.

For increasing the data rates in digital communication networks, high-speed signal generation is required. To generate these high-speed signals, electronics-based arbitrary waveform generators (AWGs) are the key components. However, most of the commercially available high-speed electronic AWGs are subject to linearity and resolution limitations. Photonics-based AWG, instead, might offer high bandwidth with better resolution and phase noise. Several photonic techniques have been proposed in recent years but with increased system complexity and limited dynamic range. We have recently proposed a photonics based architecture for high-speed arbitrary waveform generation using low-speed electronics, which is based on optical Nyquist pulse sequences and time-domain interleaving to obtain high-quality waveforms. Within this system, a single laser source is split into N branches. A Nyquist pulse sequence is generated by an integrated modulator driven by a single electrical sinusoidal frequency in each branch. Subsequently, they are modulated and multiplexed to obtain the targeted waveforms. The time delay between the pulse sequences is realized by a simple electrical phase shift of the sinusoidal driving signal. Here, a theoretical validation for the N channel system is presented along with simulation and experimental results for a three-branch photonic AWG. Using an integrated silicon Mach-Zehnder modulator saw-tooths, sinusoidal and some bandwidth-limited analog waveforms are generated. With available 100 GHz integrated modulators, the maximum possible sampling rate of 300 GS/s can be achieved. The mathematical proof validates that this simple concept can generate bandwidth limited user-defined waveforms with very high precision.

The bandwidth and resolution of the electronic digital-to-analog converters (DAC) and analog-to-digital converters (ADC) of modern-day communication systems defines the link capacity to a large extent. For high analog bandwidths, the performance of state-of-the-art DACs is limited in terms of the effective number of bits (ENOB). A drastic improvement in ENOB might be realizable with photonic based DAC by employing integrated Mach- Zehnder modulators (MZM) and time-domain interleaving. Especially, the optical signal processing of Nyquist pulses with MZM might provide a possible solution to achieve high analog bandwidths with relatively low required electronic and photonic bandwidth. By using optical time interleaving and pulses synthesized by an MZM with a bandwidth of 100 GHz for the modulator and the electronics, sampling rates of 300 GS/s can be achieved. Thus, with standard silicon components available on the market, a compact and low-cost integrated photonic DAC module can easily be realized. The ENOB of such a system is limited by the quality of the Nyquist pulses, which in turn is affected by the jitter of the used signal generator (SG) and MZM nonlinearities. Here we present analytically that an ENOB of more than 8 can be achieved for analog bandwidths greater than 100 GHz by using a low phase noise SG. With experimental validation, we analyze the upper operation limit of such photonic DACs and their dependence on non-idealities of the Nyquist pulses.

Adaptive multi-level symbol detection technique in free-space optical communication (FSOC) is presented in this paper. Generally, the on-off keying (OOK) format is used in intensity modulation/direct detection (IM/DD) based free-space optical (FSO) system. For OOK symbol detection in FSOC, the maximum likelihood (ML) based optimal detection scheme is adopted. The ML detection techniques need the accurate marginal distribution of the turbulence-induced fading. However, it is non-attainable in practical system, and the distribution model is also not fixed in the FSO link. The response of intensity fluctuation in turbulence is relatively slow compared with that of the data rate with a cutoff frequency of 1 kHz. For this reason, there is a variation in time about the optimal threshold, causing significant burst errors when a fixed threshold is used. Applying a DC block or AC coupling of PD can stabilize symbol detection in OOK symbol. However, when it comes to multi-level pulse amplitude modulation (PAM) format in FSO system, the symbol detection problem cannot be solved with previous detection techniques. In this paper, the channel state information (CSI) prediction technique for adaptive multi-level optical symbol detection is proposed. Considering the slow varying characteristic of scintillation channel, we used an adaptive linear prediction (ALP) filter to predict the CSI of the upcoming frame. The proof-of-concept experiments were conducted with a Mach-Zehnder modulator (MZM) based channel emulator. The validity of the proposed symbol detection technique was experimentally demonstrated by comparing with the fixed threshold decision (FTD) and adaptive threshold decision (ATD) scheme.

Raytheon Intelligence & Space (RIS) in El Segundo, CA is currently developing photonic integrated circuit (PIC) devices for the improved efficiency of photonics-based optical phased arrays (OPAs). The goal is to functionalize PIC OPAs as coherent transmit and receive terminals for high data rate, high bandwidth, free-space laser communication. RIS has built a testbed to simultaneously probe the near and far field of the PIC under test with plug and play installation of our different designs. Here we investigate a 1x4 silicon-on-insulator (SOI) OPA with microring resonator phasing elements. A tunable infrared laser couples the transverse electric mode (TE0) into the rectangular PIC waveguide via a polarization-maintaining tapered fiber. The waveguide is then power-split by low loss splitters and coupled to the microring resonators (one for each array antenna) via a pulley scheme with a gap of 170 nm. In this paper, we compare experimental transmission spectra of identical 1x4 OPAs with waveguide widths of either 340 nm or 360 nm. Combining measured spectra with Lumerical FDE mode analysis, we show that the 360 nm system does not support coupling conditions between the waveguide and resonator to satisfy both bandwidth and power consumption requirements. On the other hand, the 340 nm system supports one potentially strongly-overcoupled resonance at 1503 nm with minimal amplitude distortion (~ 2 dB), ~42 GHz-wide bandwidth and 𝑄~103 that could be used for coherent beam combination. This effort will help refine resonator designs and coupling geometries of future larger-scale OPAs.

Silicon photomultipliers (SiPMs) are photon-counting detectors based on single-photon avalanche diodes which have the potential to detect single photons in the visible spectrum. In this paper, we present an embedded real-time communication system based on the Xilinx Zynq-7000 FPGA platform with a 1mm-sq SiPM demonstrating the full potential of low power single-photon communication. The experimental results show that the implemented real-time system archives a Bit Error Rate (BER) 10-3 with less than 5pW of incident optical power at 100 kbps. The analysis of the implemented system performance in photon counting and BER enables the potential future adoption of low power FSO communication links in IoT, remote sensors, and energy harvesting devices.

The underwater wireless optical communication (UWOC) technology is vastly developing due to its advantages of high bandwidth, large capacity, and low latency. However, the complex underwater channel characteristics and strict requirements on pointing, acquisition, and tracking (PAT) systems hinder the performance and augmentation of UWOC. A large-area scintillating-fiber-based UWOC system is proposed to solve the PAT issue while offering high-speed, omnidirectional data detection over turbulent underwater channels. In this work, we utilized 120-cm² coverage area scintillating fibers as a photoreceiver. The large area scintillating fibers realize omnidirectional signal detection by absorbing an incident optical radiation, re-emitting it at a longer wavelength, and then guided to the end of the fibers connected with an avalanche photodetector. The UWOC system offers a 3-dB bandwidth of 66.62 MHz, and a 250 Mbit/s data rate is achieved using non-return-to-zero on-off keying (NRZ-OOK) modulation. The system was tested over a 1.5-m underwater channel under turbulences of air bubbles, temperature, salinity, and turbidity. We generated bubbles by blowing 0.20, 0.63, and 1.98 mL/s speeds of Nitrogen gas flow. A temperature gradient of 1.33 and 2.67 Celsius/m was introduced by circulating warm and cold water at the two tank ends, respectively. Salinity concentrations at 35 and 40 ppt were introduced to emulate the salinity in the Red Sea. Lastly, different volumes of Maalox^TM were added into pure water to emulate pure sea, coastal ocean, and turbid harbor water. The fiber-based UWOC system operates under those turbulence conditions with error-free communication and 0% outage probability.

With the growing number of underwater vehicles and devices used for marine environmental monitoring, there is an urgent need for real-time and high-speed underwater wireless communication technologies to transmit huge amounts of data. This poses great challenges to conventional underwater acoustic communication technology due to its low bandwidth and high latency. Therefore, underwater wireless optical communication with high bandwidth and low latency has become a promising technology. To this end, we develop the first underwater optical wireless sensor network prototype in this work. It consists of two sensor nodes and an optical hub. There is a transceiver circuit, a pH sensor, and an integrated temperature, salinity, and conductivity sensor in the sensor nodes enabling real-time underwater environmental monitoring. There are four transceivers facing four sides in the optical hub to implement bi-directional optical wireless communication with the sensor nodes. In a laboratory testbed and a field trial conducted in an outdoor diving pool, 100% packet success rates are achieved between the optical hub and the sensor nodes at a transmission distance of 60 cm. In the field trial, one of the sensor nodes is placed 60 cm away from the optical hub for real-time underwater environmental monitoring. The other sensor node is mounted on a remotely operated vehicle to collect underwater environmental information. This prototype shows great potential in future underwater mobile sensor networks and the underwater Internet of Things.

Relaxing the alignment in underwater wireless optical communication systems is highly favorable for practical use. Employing wavelength-division-multiplexing (WDM) adds to the requirement of alignment since multiple filters are used at the receiver side to separate the incoming wavelengths. We report the use of scintillating fibers in WDM systems as signal detectors that offer valuable advantages such as large-area detection, widefield- of-view and high data rates. We demonstrate the optimal selection of wavelengths based on the fibers’ characteristics, and realise an aggregated data rate of 400-Mb/s using on-off keying modulation format with zero-forcing equalization and maximum ratio combining in an outdoor diving pool in a maximum separation distance of 10-m.

Modern submarine cable systems have been evolved from harvesting ultra-high capacity within available limited bandwidth of single fiber pair into more power efficient space division multiplexing (SDM) system by employing pump sharing techniques cross multiple fiber pairs. Applying ITU recommended open cable design standard (G.977.1) for undersea cable system design will be discussed in more details. A few challenging issues including cost consideration vs maximizing capacity and connectivity optimization will be investigated. Representative key undersea products to support an integrated turn-key solution of modern undersea cable system will be briefly introduced.

The ever-increasing demand for enhanced data transmission capacity has led to the exploration of new transmission windows, beyond the C and L band typical in medium to long range optical fiber communications. Indeed, recent demonstrations have shown that although the fiber loss in the S-band is typically higher than the C-band, adoption of new amplifiers can enable a broad new transmission band that may be exploited both to extend the life of deployed fibers and to increase throughput in new space-division multiplexing fibers. In particular lower-wavelength bands are of interest to low-core count multi-core fibers where the strong wavelength dependence of inter-core crosstalk can place a limit on core-density for C/L-band transmission. Here, we describe a series of experiments, exploring S-, C and L-band transmission in single-mode and multi-core fibers covering bandwidths <150 nm and distances over 3000 km.

Beneš networks represent an excellent solution for the routing of optical telecom signals in integrated, fully reconfigurable networks because of their limited number of elementary 2x2 crossbar switches and their nonblocking properties. Various solutions have been proposed to determine a proper Control State (CS) providing the required permutation of the input channels; since for a particular permutation, the choice is not unique, the number of cross-points has often been used to estimate the cost of the routing operation. This work presents an advanced version of this approach: we deterministically estimate all (or a reasonably large number of) the CSs corresponding to the permutation requested by the user. After this, the retrieved CSs are exploited by a data-driven framework to predict the Optical Signal to Noise Ratio (OSNR) penalty for each CS at each output port, finally selecting the CS providing minimum OSNR penalty. Moreover, three different data-driven techniques are proposed, and their prediction performance is analyzed and compared.

The proposed approach is demonstrated using 8x8 Beneš architecture with 20 ring resonator-based crossbar switches. The dataset of 1000 OSNRs realizations is generated synthetically for random combinations of the CSs using Synopsys® Optsim™ simulator. The computational cost of the proposed scheme enables its real-time operation in the field.

We demonstrate a record gain of 11.4-dB for 300-nm broadband single-mode Cr-doped crystalline core fibers (SMCDCCFs) employing a novel growth of smaller core. The gain-per-unit-length efficiency of the SMCDCCF is 38-dB/m, which is much higher than current Er and Bi-doped fibers of 0.3 dB/m. The record gain achieved is mainly due to constantly maintain conical molten-zone shape in growth process to fabricate a smaller core of 15-µm and a longer fiber length of 30-cm.

By employing distributed fiber optic sensing (DFOS) technologies, field deployed fiber cables can be utilized as not only communication media for data transmissions but also sensing media for continuously monitoring of the physical phenomenon along the entire route. The fiber can be used to monitor ambient environment along the route covering a wide geographic area. With help of artificial intelligence and machine learning (AI/ML) technologies on information processing, many applications can be developed over telecom networks. We review the recent field results and demonstrate how DFOS can work with existing communication channels and provide holistic view of road traffic monitoring included vehicle counts and average vehicle speeds. A long-term wide-area road traffic monitoring system is an efficient way of gathering seasonal vehicle activities which can be applied in future smart city applications. Additionally, DFOS also offers cable cut prevention functions such as cable self-protection and cable cut threat assessment. Detection and localization of abnormal events and evaluating the threat to the cable are realized to protect telecom facilities.