For circumventing the alignment requirement of line-of-sight (LOS) underwater wireless optical communication (UWOC), we demonstrated a non-line-of-sight (NLOS) UWOC link adequately enhanced using ultraviolet (UV) 375-nm laser. Path loss was chosen as a figure-of-merit for link performance in this investigation, which considers the effects of geometries, water turbidity, and transmission wavelength. The experiments suggest that path loss decreases with smaller azimuth angles, higher water turbidity, and shorter wavelength due in part to enhanced scattering utilizing 375-nm radiation. We highlighted that it is feasible to extend the current findings for long distance NLOS UWOC link in turbid water, such as harbor water.

A transverse-magnetic-pass (TM-pass) optical polarizer based on Cu complementary metal-oxide-semiconductor technology platform is proposed and analyzed using the 2-D method-of-lines numerical model. In designing the optimum configuration for the polarizer, it was found that the metal-insulator-metal (MIM) polarizer structure is superior compared to the insulator-metal-insulator polarizer structure due to its higher polarization extinction ratio (PER) and low insertion loss. An optimized MIM TM-pass polarizer exhibits simulated long wavelength pass filter characteristics of >;~1.2 μ.m, with fundamental TM₀ and TE₀ mode transmissivity of >;70% and <;5%, respectively, and with PER ~11.5 dB in the wavelength range of 1.2-1.6 μ.m. The subwavelength and submicrometer features of this TM-polarizer are potentially suitable for compact and low power photonics integrated circuit implementation on silicon-based substrates.

Multimode fiber (MMF) specklegram sensors have recently drawn significant attention due to the incorporation of machine learning (ML) algorithms in detecting different sensing parameters. Deep learning (DL) techniques provide an efficient way of extracting information from fiber specklegrams, which can be utilized in different sensing applications. Convolutional neural networks (CNN) have proven to be extremely successful in imaging technologies over the last decade. The breakthrough from CNN has instigated new frontiers for applications in different domains. CNNs can automatically learn the variations in MMF specklegrams under different conditions. Besides detecting slight variations in the fiber, CNNs are also insusceptible to environmental noise and fluctuations, thus, making them superior in terms of performance accuracy. They provide a low-cost and simple alternative to extracting information from fiber specklegrams which has piqued the interest of many researchers. In the last few years, there has been a growing number of research articles studying the applicability of DL frameworks in various sensing technologies. In this paper, we present a comprehensive review of such articles that explore the use of ML in different MMF sensing applications, such as bending sensors, endoscopes, tactile or position sensors, and others. The principle of specklegram in MMF, the data generation process, an overview of different DL approaches as well as open challenges and future research directions have also been discussed in this article.

We report on self-injection locking in InGaN/GaN (blue/green) and InGaP/AlGaInP (red) visible-light laser diodes. The free-space optical feedback path was accomplished via an external mirror. The effect of injection current, optical power injection ratio, and external cavity length on the spectral linewidth and modulation bandwidth of the lasers is investigated. Our results show that the laser performance was substantially improved. In particular, we achieved a significant increase of ~57% (1.53-2.41 GHz) and ~31% (1.72-2.26 GHz) in the modulation bandwidth, and ~9 (1.0-0.11 nm) and ~9 (0.63-0.07 nm) times reduction in spectral linewidth of the green and blue lasers, respectively. Consequently, side-mode-suppression-ratio was considerably increased in all the cases, reaching as high as ~20 dB in self-injection locked blue laser diode, thus enabling a close to single-mode operation. This paper paves the way for attaining high-speed optical wireless communications by overcoming the challenges of limited modulation bandwidth and multimode operation of visible laser diodes with this simple scheme.

We report on the demonstration of 50 nm (full-width at half-maximum) broadband stimulated emission from a chirped AlGaInAs barrier thickness multi-stack InAs/InP quantum dash (Qdash) laser. The 2 μm wide uncoated Fabry-Perot (FP) ridge-waveguide laser exhibits a total power of 0.18 W, corresponding to an average spectral power density of 3.5 mW/nm, under pulsed current conditions. Intentional extended inhomogeneity across the Qdash stacks have been attributed to the enhancement of broadband emission.