Low crosstalk monolithic wavelength demultiplexing switch arrays

We describe a new method of sensing the linear polarization of light using resonant cavity enhanced (RCE) photodetectors. The RCE detectors are constructed by integrating a thin absorption region into an asymmetric Fabry-Perot cavity. The top reflector is formed by the semiconductor air interface while the bottom mirror is a distributed Bragg reflector (DBR). Quantum efficiency of these RCE devices can be controlled by tuning the cavity length by recessing the top surface of the detector for off-normal incidence of light the reflectivity of the semiconductor-air interface can be significantly different for TE(s) and TM(p) polarizations. A pair of monolithically integrated RCE photodetectors with cavity lengths tuned for resonance and antiresonance provide a large contrast in response to TE and TM polarizations. An alternative polarization sensor can be formed by vertically integrating a conventional and a RCE photodetector. We show that a large contrast in the TE/TM responsivities of the vertical cavity polarization detectors (VCPD) can be achieved, thus combining detection and polarization sensing in a single mesa semiconductor device. These devices alleviate the problems associated with the bulkiness and critical alignment constraints of the conventional sensors based on polarizing filters or splitters and have potential for fabrication of monolithic smart pixels and imaging arrays.

A trend of high speed computing and communication has demonstrated incredible performance increase, 1000x in every 10yrs. Such enhancements have been realized partly due to Moore’s law based and parallelism based progresses [1]. To keep up such a trend for future, we have to integrate photonics on a chip, i.e., Si photonics, especially on-chip WDM (wavelength division multiplexing) architecture [2]. In the history of the fiber optics, WDM has brought significant increase in bandwidth of optical communication and currently the channel spacing in WDM is as small as 0.8 nm and will soon be 0.4 nm. However, Si photonics has used coarse WDM where the channel spacing is as large as 10nm. The reason for the large spacing is a large TO (thermooptics) coefficient of Si. Due to chip temperature fluctuation on a chip, wavelength is changed widely, malfunctioning WDM. This is the fundamental limit in shrinking the wavelength spacing. In other words, the key challenge to implement on-chip WDM is to lock its channel wavelength within a given channel spacing. In this talk we will focus on silicon nitride based Si photonics, i.e., medium index contrast optics, MiDex Si photonics to lock the channel wavelength within the current dense WDM (DWDM) in optical fiber communication, i.e., 0.8nm spacing [3]. Fig. 1 shows bending radii of various waveguides vs. index difference between the core and cladding of the waveguide. Generally, Si waveguide and SiO2 cladding have been used in Si photonics, where index difference is 2.0. The materials set has a high index contrast, i.e., here we will refer it to as HiDex (high index contrast). The advantage is a sharp bending of the waveguide, which reduces device footfprints and allows high density integration of optical devices on a chip. Althougt the disadvantage is large light propagation loss, Si process technology allows to minimize the roughness of waveguide to reduce the loss to ~1dB/cm. It is known however that Si has a large TO coefficient, i.e., temperature dependence of refractive index. It has been reported that chip temperature has changed about 50˚C during its operation. Thus, on-chip temperature fluctuation changes wavelength, malfunctioning of WDM. To avoid it, a large channel spacing, i.e., 10nm has been chosen. We have been focusing a slightly smaller index contrast system in the following two reasons: When we choose SiNx waveguide and SiO2 cladding, the minimum bending radius is 10 µm which is still small enough and the TO coefficient of SiNx is ~1/10 of that of Si. This indicates this materials system should be robust against temperature fluctuation on a chip. Of course Planar Lightwave Circuit based on silica fiber technology has smaller TO coefficient, here we refer it to as LoDex (low index contrast), but the bending radii should be ~1cm. Thus, we have proposed MiDex Si photonics instead of current HiDex Si photonics. SiNx (0.6-0.8 µm thick) can be deposited on SiO2 (2µm thick) using a physical vapor deposition (PVD) technique. Chemical vapor deposition techniques have widely been used, but CVD SiNx contains N-H bonds that absorb 1520nm, that is in optical communication wavelength range of 1550nm. To dissociate H from SiNx, CVD SiNx should be annealed at high temperature such as 1100˚C. To avoid such high temperature annealing, we have chosen PVD SiNx, which should be free from N-H bonds. We have demonstrated SiNx without such absorption. The present report has verified the advantage of PVD SiNx over CVD SiNx, temperature stability of filter characteristics of ring based resonators, and athermalization using polymer materials.

Recently there have been many advances in vertical-cavity surface-emitting (VC-SE) lasers. VC-SE lasers are promising because they facilitate simple fabrication of high-density, low-threshold, and cost-effective two dimensional (2-D) laser arrays. Monolithic 2-D multiple-wavelength laser arrays are potentially useful for applications in optical communications and interconnects using wavelength-division multiplexing (WDM).1

Deep neural networks (DNNs) have been applied to recover signals in optical communication systems and have shown competence of mitigating linear and nonlinear distortions. However, as the data throughput increases, the heavy computational cost of DNNs impedes them from rapid and power-efficient processing. In this paper, we propose an optical communication signal recovery technology based on a photonic convolutional processor, which is realized by dispersion delay unit and wavelength division multiplexing. Based on the photonic convolutional processor, we implement an optoelectronic convolutional neural network (OECNN) for signal post-equalization and experimentally demonstrate on 16QAM and 32QAM of an optical wireless communication system. With system parameters optimization, we verify that the OECNN can achieve accurate signal recovery where the bit error ratio (BER) is below the 7% forward error correction threshold of 3.8×10-3 at 2Gbps. With adding the OECNN-based nonlinear compensation, compared with only linear compensation, we improve the quality (Q) factor by 3.35 dB at 16QAM and 3.30 dB at 32QAM, which is comparable to that of an electronic neural network. This work proves that the photonic implementation of DNN is promising to provide a fast and power-efficient solution for optical communication signal processing.

Optical communications and sensing systems have required optical fibers to meet the demanding applications day by day, such as high bit rates, dense wavelength division multiplexing (DWDM) and long distance, so that there could be always renewed interest in fiber designs to lift fundamental limitations of silica optical fibers. One kind of fiber, which is different from the conventional fiber, is silica photonic crystal fiber (PCF) that has an array of microscopic air-holes running along its length. For PCF, there are two distinct classes as the guidance; one is using a type of total internal refection which is referred to the terms microstructured fiber, holey fiber, effective-index fiber, and hole-assisted fiber, and the other is employing photonic band-gap which is referred to photonic band-gap, Bragg fiber, and omni-directional waveguide. In this paper, we will address recent advances in PCFs. While the potential applications are also addressed, the emphasis will be on PCFs for applications in optical communications and sensing systems.

Wavelength multiplexing and tuning capabilities are highly important for applications in optical interconnects and optical communications using wavelength division multiplexing (WDM). We review novel methods and experimental results to achieve these functions in vertical cavity surface-emitting lasers (VCSEL), we believe for the first time. By implementing spatially chirped layers in the laser cavities and obliquely aligning the array axes to the thickness chirp direction, we obtained 7 × 20 VCSEL array emitting 140 unique, non-redundant, controllable, uniformly separated single-mode wavelengths. A large total wavelength span of 43 nm is obtained without compromising the performance of the lasers. Moreover, our techniques are generic and can be readily extended to other wavelength regimes. In addition, we describe our work on wavelength-tunable VCSELs with a 1.8-nm (~540 GHz) continuous tuning and no mode hopping. The laser output power remains nearly constant throughout the tuning range. The wavelength tuning speed is ~400 ns. The multiple wavelength tunable laser arrays are promising for achieving ultrahigh bandwidth (hundreds Gbit/sec to Tbit/sec), high redundancy and reliability, reconfigurable optical networks.

Phase-shifted gratings are expected to play a significant role in wavelength division multiplexing, lasers, and optical signal processing due to their exceptional performance, including an ultra-narrow bandwidth, a high extinction ratio, and a sharp line shape. The key to introducing an accurate phase shift is to implement high-precision lithography, which is currently not available. However, this bottleneck can be addressed by employing the method of equivalent phase shift. Compared to implementing a true phase shift, the requirement for an equivalent phase shift can be reduced by three orders of magnitude. Currently, most phase-shifted gratings are fabricated in silicon-on-insulator (SOI) thanks to their compatibility with CMOS technology. Since the grating's periodic sides affect the phase shift, they also impact the full width at half maximum of the transmission peak. Ultra-low-loss waveguides based on silicon nitride are a favorable platform for researching microwave photonics and their applications in a wide variety of fields at present. Nevertheless, to date, a survey of phase-shifted gratings reveals a paucity of those fabricated on silicon nitride platforms. In this paper, we propose an optical filter based on a silicon nitride platform utilizing equivalent phase shift technology. This component under consideration comprises sampled gratings and true phase shift, with a length of 5 mm on a 350-nmthick Si₃N₄ platform. This component has a pitch of 480 nm and a sampling length of 100 μm, utilizing side-wall modulation. For this device, a narrow and sharp passband window of approximately 0.75 nm was observed over a stopband region at the target wavelength of 1566.35 nm. The reflection spectrum also shows good repeatability of the passband response for both the 1595.43 nm and 1598.12 nm wavelengths. Moreover, this scheme has broad potential applications in optical frequency combs, optical communication, and optical sensing.

Resonant cavity enhanced (RCE) photodetectors are promising candidates for applications where high-speed high-efficiency photodetection is desirable. In RCE devices, the active device structure is placed inside a Fabry-Perot resonant microcavity that causes a large enhancement of the resonant optical field. The increased optical field allows the design of thinner photodetectors while simultaneously increasing the quantum efficiency at the resonant wavelengths. We present measurements of quantum efficiency of GaAs/AlGaAs RCE p-i-n photodiodes designed for near-unity quantum efficiency. The bandwidth-efficiency product is in excess of 50 GHz. We consider two applications, namely optical communications and quantum optical measurements employing pulsed lasers which would benefit from a wider peak response than that is readily accomplished using RCE detection.

Wavelength Division Multiplexing (WDM) networks with reconfigurable optical add/drop multiplexers (ROADMs) realize automatic scheduling and restoration at the wavelength level and are therefore part of the solution to meet increasing bandwidth and routing flexibility requirements for transport networks. The wavelength selective switch (WSS) is the key component in all commercial ROADM systems. Liquid crystal on silicon (LCoS) technology dominates current WSS installations as LCoS-based WSS can support Flexgrid technology enabled networks, and high port counts with excellent performance. WSS designs are becoming increasingly specialized for a wide variety of applications. For single WSS used in broadcast and select (B&S) architecture, superior port isolation is key for the desired signal to be filtered from the whole spectrum. While for twin WSS deployed in route and select (R&S) nodes, many features are desired for various network applications. These include multiple common ports, premium ports with superior port isolation, and high port counts. Future WSS may be asked to address a number of different end user applications, and as such there is a demand for combinations of custom spectrum range, compact form factors, increased WSS port count, low cost, thermal and mechanical ruggedization. The first commercially available quad WSS is part of this increasing trend of utilization of WSS throughout network architectures.

Harnessing the orbital angular momentum (OAM) of light is an appealing approach to developing photonic technologies for future applications in optical communications and high-dimensional quantum key distribution (QKD) systems. An outstanding challenge to the widespread uptake of the OAM resource is its efficient generation. In this work we design a new device that can directly emit an OAM-carrying light beam from a low-cost semiconductor laser. By fabricating micro-scale spiral phase plates within the aperture of a vertical-cavity surface-emitting laser (VCSEL), the linearly polarized Gaussian beam emitted by the VCSEL is converted into a beam carrying specific OAM modes and their superposition states, with high efficiency and high beam quality. This new approach to OAM generation may be particularly useful in the field of OAM-based optical and quantum communications, especially for short-reach data interconnects and QKD. (C) 2015 Optical Society of America

Learned digital back-propagation (LDBP) is emerging as a promising solution to mitigate the nonlinear fiber distortions that limit the capacity of optical communications. However, the computational complexity of LDBP, which is affected by dispersion compensation, will increase significantly as optical communications move toward higher baud rates and longer transmission distances. Herein, we propose what we believe to be a novel method called exponential pruning LDBP (EP-LDBP), which is achieved by pruning the redundant LDBP taps with adjustable parameters. Experimental results show that EP-LDBP achieves a 61.5% reduction in computational complexity compared to LDBP in frequency domain without sacrificing compensation performance in a 21-channel wavelength division multiplexing (WDM) transmission over 1600 km fiber using 60 Gbaud 16-ary quadrature amplitude modulation (16QAM). Furthermore, our analysis of EP-LDBP under varying baud rates (30-60 G) and transmission distances (400-1600 km) demonstrates its superior potential in reducing complexity, thus aligning more effectively with the evolving landscape of optical fiber communications.

Optical isolation enables nonreciprocal manipulations of light with broad applications in optical communications. Optical isolation by rotating structures has drawn considerable attention due to its magnetic-free nature and unprecedented performance. Conventional rotation-based optical isolation relies on the use of bulky cavities hindering applications in subwavelength photonics. Here, we propose a mechanism of optical isolation by integrating the unique dispersion of a hyperbolic metamaterial with the transverse spin-orbit interaction of evanescent waves. We show that rotation of a subwavelength hyperbolic nanoparticle breaks the time-reversal symmetry and yields two resonant chiral modes that selectively couple to the transverse spin of waveguide modes. Remarkably, the transverse spin-orbit interaction can give rise to unidirectional coupling and more than $95%$ isolation of infrared light at an experimentally feasible rotation speed. Our work fuses the two important fields of optical isolation and photonic spin-orbit interactions, leading to magnetic-free yet compact nonreciprocal devices for applications in optical communications, chiral quantum optics, and topological photonics.

Enhanced single-mode lasers of all-inorganic perovskite nanocube by localized surface plasmonic effect from Au nanoparticles

Nanophotonic devices leverage unique interactions between photons and materials at the nanoscale, enabling applications in optical communication, biosensing, and quantum computing. These devices' properties are susceptible to material composition and structural design. Nanofabrication techniques, such as optical lithography, e‐beam lithography, two‐photon polymerization, and direct laser writing, have been widely applied to fabricate nanophotonic devices. Notably, rewritable fabrication stands out due to its low cost, flexibility, efficiency, and multi‐functionality. In this paper, a novel rewritable nanofabrication technique is proposed, which combines electrochemical reactions with direct laser writing, to fabricate nanophotonic devices on low‐cost indium tin oxide (ITO) films. The experimental results have demonstrated that high‐quality and erasable photonic structures such as diffraction gratings and holography masks can be directly fabricated using our technique. Hence, it is believed that this method can be applied in diverse fields such as nanophotonics, optoelectronic devices, biosensors, micro‐electromechanical systems, and nonlinear optics.

Packet-switched unidirectional and bidirectional ring wavelength division multiplexing (WDM) networks with destination stripping provide an increased capacity due to spatial wavelength reuse. Besides unicast traffic, future destination stripping ring WDM networks also need to support multicast traffic efficiently. In this paper, we provide a probabilistic analysis of the mean hop distances traveled by multicast packet copies on the wavelength channels, and based on the mean hop distances analyze the nominal transmission capacity, reception capacity, and multicast capacity of both unidirectional and bidirectional ring WDM networks with destination stripping. The developed analytical methodology accommodates not only multicast traffic with arbitrary multicast fanout but also unicast and broadcast traffic. In our numerical investigations we examine the impact of number of ring nodes and multicast fanout on the transmission, reception, and multicast capacity of both types of ring networks for different unicast, multicast, and broadcast traffic scenarios and different mixes of unicast and multicast traffic. Our analytical methodology provides a foundation for extended analyses of the multicast capacity of WDM ring networks and enables the evaluation and comparison of future multicast-capable medium access control (MAC) protocols for unidirectional and bidirectional ring WDM networks in terms of transmitter, receiver, and multicast throughput efficiency.