This PDF file contains the front matter associated with SPIE Proceedings Volume 11828, including the Title Page, Copyright Information, and Table of Contents.

With a rising trend to use optical fiber in both short-reach and long-haul network applications, it has become necessary to detect faults with high spatial resolution, sensitivity, and dynamic range in industry. Unfortunately, the most widely used diagnostic technique, optical time-domain reflectometry (OTDR), has an inherent trade-off between the above mentioned figures of merit. Consequently, traditional OTDR systems can either be used in short-reach applications with high spatial resolution or long-haul applications with high dynamic range. Both OTDR and optical frequency-domain reflectometry systems have been proposed in the past which can arguably close the gap between the two extremes, but they also have a trade-off with the cost, form-factor, and complexity. Recently, researchers have demonstrated an input/output (I/O) interface integrated OTDR (iOTDR), which uses digital I/O ports to perform OTDR measurements with high spatial and voltage resolution. The iOTDR eliminates the necessity of using high-end analog-to-digital converters and demonstrates its potential to be fully integrated in an optical switch with minimum hardware modification. Additionally, the iOTDR is also reconfigurable and software-defined, making it a power- and resource-efficient solution. Thus, it is especially attractive for short-distance communication links, such as those in a datacenter, where computational resources are limited. This manuscript expands upon the advantages brought upon by the iOTDR to propose and demonstrate a new versatile iOTDR that can achieve high spatial resolution, sensitivity, and dynamic range for both short-reach and long-haul networks thanks to its highly reconfigurable design.

Three-dimensional (3D) plasmonic sensors have been developed using the reversal nanoimprint technology. Using this technology, plasmonic sensors with additional levels of metals and asymmetrical profiles were optimized to achieve high sensitivity for biomolecule detection. Combining the unique designs of these 3D nanostructures, the plasmonic sensors have high performance as the devices combined the hybrid coupling effect of localized surface plasmon resonance, Fano resonance, and Fabry-Perot cavity modes to achieve sharp resonance peaks with large resonance peak shifts. Applications of these high performance nanoplasmonic sensors to biosensing will be presented.

This article presents a novel automated measurement prototype for 3D geometry of mobile and large-scale conical workpiece, manipulated through two independent robot platforms that placed on its two sides with laser scanner and motorized linear stage. First, with point cloud that covers end point provided by laser scanner. Then, modeling and identification of end point of workpiece is established based on height variations in its nearest neighborhood with respect to virtual measurement datum plane, which is step-by-step derivatively generated according to initial datum point in an online virtual inspection environment. Next, the current geometry-relations between neighboring end points can be subsequently used to guide the laser scanner for high precision sampling surface area incorporating an automatic simple module. Moreover, both orientation and position geometrical relationships of the corresponding features on the fitting circles are analyzed too. Details preliminary experimental tests were performed to verify the measuring accuracy of this method.

Most of waveguide implementation for HUD or augmented reality combiner are flat pieces of glass because the image propagation does not suffer from any aberration when traveling along their length. However, this type of combiner does integrate seamlessly in front of the viewer eyes and a curved optics would be much more appealing. Using holographic optical elements, we demonstrated that it is possible to correct the aberrations induced by the curved surfaces of the waveguide, and display a aberration-free image to the viewer. This correction applies for different waveguide geometries (1D or 2D curvature) as well as different pupil expansions (1D or 2D expansion). A Zemax model is presented along a curved waveguide demonstrator.

Over the past years, light detection and ranging (lidar) technologies have been investigated and commercialized for various applications such as autonomous vehicles, terrestrial mappings, and precision measurements. Currently, the frequently used ranging methods are the pulsed time of flight (PToF) and frequency modulated continuous wave (FMCW) lidars that relies on frequency sweeping to capture range and velocity information. We have previously developed and demonstrated the multi-tone continuous wave (MTCW) that operates by employing amplitude modulation via multiple radio frequencies (RF) and coherent detection. Here, we present a theoretical and experimental study on phase-based MTCW lidar that can detect the range and velocity of objects with arbitrary velocities. The experiments demonstrate that the phase and frequency of the Doppler-shifted fixed RF tones can be used to extract the range and velocity information in a single shot measurement. We show that a <±1cm resolution in the ranging, limited by the temporal resolution of the detection system, and a 0.5cm/s speed resolution is limited by the frequency resolution of the detection system are achievable. Moreover, the proposed approach has the potential to mitigate the requirement for a narrow linewidth laser for coherent detection.

The optical implementation of neural networks is proposed to have advantages over electronic implementations with lower power consumption and higher computation speed. However, most optical neural networks (ONNs) utilize conventional real-valued frameworks that are designed for digital computers, forfeiting many advantages of optical computing such as efficient complex-valued operations. Complex-valued neural networks are advantageous to their real-valued counterparts by offering rich representation space, fast convergence, and strong generalizations. We propose and demonstrate an ONN that implements truly complex-valued neural networks, achieving high accuracy and strong learning capability in many benchmark tasks.¹ On the other hand, efficiently training ONNs remains a formidable challenge, due to the difficulty in obtaining gradient information from a physical device. We propose an efficient on-chip training protocol for ONNs and demonstrate it by several practical tasks.² The protocol is gradient-free and physical agnostic, and is applicable for various types of chip structures, especially those that cannot be analytically decomposed and characterized. The protocol is robust to experimental perturbations like imperfect phase detection and photodetection noise. Our results present a promising avenue towards deep complex networks with smaller chip size, stronger performance, and flexible reconfiguration to realistic applications (e.g., facial recognition, natural language processing, and autonomous vehicles).

A concept of Texas Instrument (TI)-Phase-only Spatial Light Modulator (PLM) is reported which utilizing the dynamic piston motion of PLM pixels to form a discretized blazed grating. By fully manipulating the piston motion of pixels and increasing the available discretization level of the quasi-blazed grating, the Diffraction Efficiency (DE) can achieve close to 99%, which qualifies PLM an ideal candidate for beam steering functionality combining with its MEMS based high refresh rate and large aperture. The DE of the discretized blazed grating is proven to have 86% with 633 nm monochromatic light incident at 25° with 16 discretization levels and 2𝜋 round-trip phase modulation by RCWA algorithm. Furthermore, additional factors which lead to the degradation of diffraction efficiency is also analyzed.

Nanoscale optical writing enables high-density optical data storage. However, current techniques usually require high laser beam intensity with high energy consumption and short device lifetime. Upconversion nanoparticles (UCNPs) have shown great potential for high-density optical data storage due to their exceptional luminescence emissions. In addition, UCNPs have enabled low-power STED microscopy. We show that UCNPs can induce the reduction of graphene oxide (GO) at the nanoscale. Dual-beam super-resolution irradiation was used to write features in UCNP-conjugated GO with lateral feature size at the nanoscale and inhibition intensity of <15 MW/cm^2. This approach might offers a convenient and energy-efficient solution for the storage demands in the Data Age.

Phase retrieval is the key technique in phase-modulated holographic storage. In this paper, a deep convolutional neural network is proposed to directly retrieve phase data. Compared with the traditional non-interferometric phase retrieval method, this method has the advantages of fast retrieval speed and high reconstruction accuracy. In this paper, the influence of intensity image noise on retrieval results under different retrieved conditions is researched and analyzed. By establishing a simulation system that is in strict agreement with real experiments, the lensless spatial diffraction images are generated. By adding different proportions of random noise into the intensity images we get the training dataset. The convolutional neural network is trained by a training dataset and tested by a new noisy test dataset. Experimental results show that the phase retrieval method based on deep learning has a high tolerance for systematic errors and strong anti-noise performance.

To increase the capacity of optical disc systems, various techniques are available such as improvements of optical readout channels, signal processing method and recording media. Although the recorded information is digital but rather analog like readout signal is adopted for current optical disc systems. The partial response readout has been used in Bluray disc systems. As an alternative technique, we have reported an application of orthogonal frequency division multiplexing to optical disc systems. In such technologies the readout sampling intervals are longer than the information bit rate. Multi-level signal is employed instead to achieve high information density so that the signal to noise ratio is critical to determine the recording density. The kinds of noise in optical disc signal can be classified to thermal (amplifier), shot and medium. The first two are basically random. Those noise signals vary at different readout events so that they can be suppressed by averaging plural readout from the same position. But this technique is helpless for medium noise because it is created in manufacturing or recording processes. However the medium noise may have a different character from the recorded information because of the difference between the processes to create them. For example, the reflectance change of the information marks is different from the surface roughness of the disc noise. Actually several readout channels such as divided photo detectors, multi-wavelength illumination and optical filters enable us to derive various characteristics of signals from readout media. Some methods and calculated results to suppress the medium noise by using such multimodal readout will be discussed in this paper.

Heat-assisted magnetic recording (HAMR) is a promising technology for achieving more than 10 Tb/inch² recording density. A near-field transducer (NFT), which forms a small light spot on a recording medium, is necessary in HAMR. The authors’ group has proposed a novel device, in which a metal nano-antenna as an NFT is attached to a semiconductor ring resonator as a light source. There are multiple eigenmodes in this device. If they are excited simultaneously, the device becomes unstable because of the mode competition. Moreover, a near-field light is generated at the tip of nano-antenna for some eigenmodes but not generated for other eigenmodes. Therefore, in this study, how to excite a desired eigenmode selectively among the multiple eigenmodes was investigated through a numerical simulation. The eigenmodes were classified into four types: modes in the radial direction (characterized by the order of mode l), modes in the tangential direction (characterized by the order of mode m), even and odd modes, and TE and TM modes. The mode with l = 1 could be excited by increasing the inner diameter of the ring resonator. The mode with specific m and the even mode could be excited by forming the slits in the ring resonator and using the frequency dependence of gain. The TE mode could be excited automatically because of its strong light confinement in the active layer. By combining these methods, the device could be made stable and the near-field light could be generated at the tip of nanoantenna.

Flat optics with micro-nano structures fabricated on a flat substrate is promising for integrated optics for its compactness and compatibility for large volume manufacturing. In this talk, I will introduce our works on photon nano sieves with holy structures for polarization independent broadband high diffraction efficiency and large angle-of-view hologram, specially designed Fresnel flat lens to break the diffraction limit for higher resolution focusing and imaging, phase change materials and emerging 2D materials for reconfigurable and ultra-thin flat lens demonstration, and applications of these flat optics in lithography, medical imaging and optical data storage.

Based on polarization holography theory, the plane bifocal vector lens is studied. In previous studies, the bifocal vector lens were limited to cross-angle π/2 and bulk materials. However, when the two waves are orthogonal circularly polarized state, the plane bifocal vector lens can be realized, and the limitation of cross-angle π/2 and bulk materials can be broken. The lens produces corresponding focus output through the reading wave with different polarization states, which can be used for large-area optical element research.

A new scheme of advanced driving beam (ADB) module employing ultra-reliable Ce: YAG-based single crystal phosphor (SCP) for use in autonomous vehicles is demonstrated. The Ce: YAG-based SCP layers fabricated by high-temperature of 1500°C exhibits excellent thermal stability. The ADB module consists of a Nichia blue LED with silicone-based phosphor, a digit mirror device (DMD), a projection lens, and two Nichia laser diodes with a Ce: YAG-based SCP layer. The ADB pattern is measured to be 88,436 luminous intensity at 0°, 69,393 cd at ± 2.5°, and 42,942 cd at ± 5°, which are well satisfied the ECE R112 class B regulation. The proposed high-performance ADB module with ultra-reliable Ce: YAG-based SCP layer is favorable as one of the promising ADB module candidates for use in the next-generation automobile headlight applications.