IEEE PhotonicsGlobal[at]Singapore

For GaN-based multiple quantum well (MQW) light-emitting diodes (LEDs), embedding a potential well in the traditional-width electron blocking layer (EBL) enhances the device's light output power. This simulation-based study predicts that increasing the EBL width enables effective improvement of performance through introduction of a barrier layer in the wide EBL. However, applying the same potential-well insertion strategy, originally designed for the traditional-width EBL, to the wide EBL instead reduces performance. This study investigates the electrical and optical performance of LEDs with three distinct 75 nm-wide EBL architectures. The reliability of the simulation is calibrated by fitting the experimental light output power (LOP) and external quantum efficiency (EQE) data of the reference LED1, providing a robust basis for subsequent analysis. The results reveal that the proposed EBL structure with an intralayer potential barrier (Al composition x = 0.29) exhibits the best performance. At an injection current density of 40 A/cm², the LOP and EQE of the optimized device are remarkably enhanced by 102.8% and 68.3% compared to the traditional EBL design. The enhanced performance arises from an elevated effective conduction-band barrier that suppresses electron overflow and from mitigated polarization-induced band bending that enhances hole injection into the active region. Consequently, carrier distribution becomes more uniform and the radiative recombination rate increases. Additional sensitivity and structural-parameter analyses further confirm the robustness of our proposed LEDs under different polarization conditions and clarify the influence of EBL width, as well as the Al composition and thickness of the intralayer potential barrier on optical performance. This work provides a promising band-engineering strategy for realizing high-efficiency GaN-based LEDs.

Integrated semiconductor lasers are important for optical communications, gas sensing, biomedicine, and related fields. Conventional devices are commonly fabricated by dry etching, where sidewall roughness and surface defects may affect scattering and waveguide losses. In this work, we propose and fabricate an InP-based semiconductor laser with a fully coupled straight-ridge/racetrack waveguide structure using wet chemical etching. Under p-side-up continuous-wave (CW) operation and without facet coatings, the device exhibits a threshold current of 230 mA and a single-facet output power above 17 mW. At an injection current of 230 mA, single-longitudinal-mode lasing is obtained at a center wavelength of 1655.2 nm with a 3-dB linewidth of 0.12 nm. When the injection current is increased to 1520 mA, the spectrum shows comb-like characteristics within a broadband envelope of approximately 1670–1697 nm, with an average adjacent peak spacing of 0.22 nm in the pronounced multi-peak region. To the best of our knowledge, this is the first report of a derived satellite-peak spectral feature in an InP-based fully coupled straight-ridge/racetrack integrated semiconductor laser fabricated by wet chemical etching. These results show that the fully coupled structure enables current-dependent spectral evolution and produces equally spaced comb-like spectral features, providing experimental evidence and a useful reference for future studies of integrated semiconductor light sources.

In Mueller matrix polarimetry, the accuracy of polarimeter calibration directly determines the reliability of measurement results. As a general calibration method, the eigenvalue calibration method (ECM) is widely adopted since it does not require exact prior knowledge of the optical responses of system components. When over-measurement is employed to enhance data redundancy, it is necessary to extend the conventional ECM to handle high-dimensional intensity matrices. However, existing extended methods typically solve the instrument matrices of the polarization state generator (PSG) and the polarization state analyzer (PSA) independently, neglecting the inherent coupling between the two instrument matrices, which leads to additional noise amplification and systematic errors. To address this, this paper proposes a Joint Optimized extended eigenvalue calibration method (JO-ECM). The method formulates the matrix extension as a joint optimization problem, solved via an alternating projection algorithm by alternately solving for the extension coefficient matrices to achieve consistent estimation of both PSG and PSA extended states. Theoretical analysis shows that the proposed method retains the advantages of the existing extended ECM while enhancing numerical stability and noise suppression. Simulations demonstrate that under 8-state over-measurement, JO-ECM improves the air Mueller matrix measurement accuracy by approximately 16.7% on average compared with the original extended ECM, with robust performance across various noise conditions. Experimental validation further confirms the effectiveness of the proposed method, demonstrating a 13.91% improvement in the air Mueller matrix measurement accuracy, with error reductions of 3.54%, 14.09%, and 11.27% for the 0° polarizer, 90° polarizer, and 30° retarder, respectively.

The combination of modulating retro-reflector (MRR) architecture and coherent detection constitutes a high sensitivity, low power asymmetric link solution for underwater wireless optical communication (UWOC). Because the transmitter and receiver share the same optical source and the signal undergoes round-trip propagation, coherent MRR-UWOC systems give rise to a unique phase noise accumulation mechanism, rendering the system markedly more sensitive to laser-noise–turbulence coupling. In this paper, an analytical expression for the average symbol error probability (SEP) of a balanced detection coherent MRR-UWOC system using binary phase-shift keying (BPSK) is derived, based on the generalized oceanic turbulence optical power spectrum (OTOPS) and the Wiener model for laser phase noise. Based on this expression, the performance of the MRR coherent communication system under turbulent conditions is analyzed for linewidths ranging from 100 kHz to 4.9 MHz. The results show that system performance is sensitive to both transmission distance and laser linewidth: at 10 m, a 5 MHz linewidth is acceptable, whereas at 40 m, the linewidth must be below 2 MHz to maintain an acceptable SEP. Water temperature amplifies linewidth sensitivity much more than salinity does. The rate of dissipation of mean-squared temperature markedly worsens linewidth-induced performance degradation and is the most sensitive turbulence parameter. A higher kinetic energy dissipation rate corresponds to a lower SEP floor and greater linewidth tolerance. For linewidths in the MHz regime, increasing oceanic turbulence not only significantly raises the baseline SEP of the system, but also leads to a systematic increase in its sensitivity to linewidth variations. In addition, the system linewidth tolerance improves with increasing wavelength, indicating stronger robustness to linewidth at longer wavelengths. This research offers critical insights and practical guidelines for laser selection and engineering design, facilitating the optimization of coherent MRR-UWOC systems in varying turbulence regimes.

The coexistence of optical fiber sensing technologies with densely deployed telecommunication infrastructures is a topic of high interest in the context of integrated sensing and communication (ISAC). In recent years, distributed acoustic sensing (DAS), based on the $\phi$OTDR principle, has gained significant attention for its high sensitivity and localization capabilities. However, it relies on the transmission of high-peak-power optical pulses (up to $\sim$23 dBm), which may interfere with coexisting communication channels. In this work, we experimentally investigate the coexistence of chirped-pulse DAS operating in the C-band with telecommunication channels across the C+L bands. By analyzing post-FEC errors, we show that Kerr-induced nonlinearities from high-peak-power pulses can distort neighboring C-band channels, with measurable impairments at 100 GHz spacing and pulse peak powers as low as 16.5 dBm. These effects can be partially mitigated by increasing channel spacing, adopting longer FEC overhead, or using lower-order modulation formats (e.g., QPSK instead of 16QAM). In contrast, L-band channels experience weaker nonlinear effects, allowing higher tolerated pulse powers (up to $\sim$23 dBm for sufficiently spaced channels). Our findings highlighted Raman impairments on the L-band, reducing with increasing frequency distance, and mitigated by the chromatic dispersion walk-off. Additionally, our findings show that counter-propagation between DAS and telecom signals does not introduce measurable penalties. Finally, we did not record any worsening of the DAS sensing performances in any of the coexistence (co- and counter-propagating) scenarios tested.

Accurate real-time detection of atmospheric methane is crucial for environmental monitoring and industrial safety, and 1650nm band lasers are ideal light sources for tunable diode laser absorption spectroscopy (TDLAS). This paper proposes a novel DFB laser design structure with dual electrodes. The device employs a dual-functional waveguide structure consisting of a constant-current section and a tuning section, enabling optoelectronic performance regulation through independent current injection into the two electrodes. Experimental results show that the laser achieves a tuning range of over 3.2nm within the injection current range of 1–35mA, with a side-mode suppression ratio (SMSR) exceeding 50dB at room temperature, enabling precise matching of the methane characteristic absorption peak at 1653.7nm. The device structure eliminates the need for butt-joint integration, significantly reducing fabrication complexity and cost, thereby providing a low-cost core light source solution for high-performance methane detection systems.

We demonstrate a dispersion-tuned, actively mode-locked bismuth-doped fiber (BDF) laser operating in the 1700 nm spectral region. The laser employs a ring cavity incorporating 60 m of high-germanosilicate BDF, which simultaneously provides optical gain and large chromatic dispersion. Wavelength tuning is achieved through two complementary approaches based on dispersion-induced cavity synchronization. In the first approach, the lasing wavelength is controlled by varying the modulation frequency of an intracavity electro-optic modulator, enabling all-electronic wavelength sweeping over a range of 1670–1764 nm (94 nm). In the second approach, wavelength tuning is realized by adjusting the cavity length using an intracavity optical delay line, resulting in a tuning range of 1670–1758 nm (88 nm). In both cases, stable active mode-locking is achieved with a linear dependence of wavelength on the tuning parameter. A comparison at the 530th and 1870th harmonic orders show that higher-order operation improves the tuning linearity but reduces the RF SNR from 38–39 dB to 30–32 dB and increases the normalized pulse-period jitter from 0.75–0.91% to 5.3–6.2%. These results establish a compact and versatile platform for wideband wavelength-tunable sources in the 1700 nm band.

Traditional infrared imaging-system performance metrics are primarily developed for human-observer assessment, whereas their applicability to automatic object detection remains insufficiently understood. To examine how variations in front-end imaging-system parameters affect detection performance, this study investigates the influence of the optical–sampling parameter Fλ/d on infrared object detection. A controlled image-domain simulation framework is constructed by varying focal length, aperture diameter, and detector pixel size to generate different imaging conditions. Scene-reference infrared images are processed using MTF-based filtering, detector sampling, and NETD-related noise injection. The corresponding system characteristics are evaluated using pre-sampling MTF, sampling response, NETD, and Targeting Task Performance (TTP), while detection performance is assessed using six object detection networks, mAP, and TIDE-based error decomposition. The results show that detection performance is generally highest in the transition region of Fλ/d, where optical transfer, sampling behavior, and sensitivity are jointly balanced. The image-domain implementation is further evaluated using measured infrared target images, and the observed trend is examined across different infrared datasets. These results provide a physically interpretable basis for analyzing the influence of front-end imaging conditions on infrared object detection.

In this work, we demonstrate compact paper-clip spiral silicon photonic waveguides with ultra-low delay loss on a 300 mm-wafer foundry platform. We characterize the optical loss and group delay of single-mode and multi-mode silicon waveguides across the telecom O-, S-, C-, and L-bands. For spiral devices with 2.0-μm-wide waveguides, we measure propagation losses of 0.11 and 0.06 dB/cm at 1310 and 1550 nm, representing 10- and 20-times improvements, respectively, compared to the single-mode waveguides. Additionally, we observe a group delay of 1163 ps for a 9.5 cm-long waveguide with a device footprint of (0.3 × 3.0) mm2, yielding a delay loss of 0.5 dB/ns. These results establish a new benchmark for ultra-low-delay-loss silicon waveguides and open pathways for large-scale silicon photonic integration, delay lines, and on-chip programmable systems.

This paper proposes a novel heterogeneous 7-core 5- Linear Polarization (LP)-mode space-division multiplexing (SDM) fiber based on low-index ring and differentiated air-hole array (LRDA). The core innovation of this structure lies in the synergistic design of low-index ring and air-hole array, which achieves high-density multiplexing while simultaneously ensuring low loss and ultra-low crosstalk characteristics. Numerical simulation results indicate that the fiber's bending loss meets the requirements for stable operation of five LP modes within the C+L bands. At 1550 nm, the effective refractive index difference (Δneff) between any two modes of the fiber is greater than 1.6 × 10-3a. The heterogeneous inter-core crosstalk (HE-ICXT) between the central core and any adjacent outer core is less than -80 dB/100 km, and the homogeneous inter-core crosstalk (HOICXT) between adjacent outer cores is less than -70 dB/100 km. In addition, the relative core multiplexing factor (RCMF) reaches as high as 99.54. The manufacturing process for the designed fiber with LRDA structure is presented to verify its engineering feasibility. This paper provides an effective approach for achieving high-density independent transmission across multiple channels in SDM system.

Optical imaging has long been dominated by conventional lens-based systems, which, despite their success, remain constrained by size, weight, and cost. Lensless imaging seeks to overcome these limitations by replacing lenses with thinner, lighter, and cheaper optical modulators and reconstructing images computationally, while facing trade-offs in image quality, artifacts, and flexibility inherent in traditional static modulation. Here, we review recent progress in lensless imaging and high light a recent advance based on a programmable Fresnel zone aperture (FZA), termed LIP, which addresses these challenges through optimized aperture design and reconstruction strategies. By establishing an optimal parameter matching criterion and introducing an offset-FZA parallel merging strategy, LIP achieves high-quality reconstruction with reduced artifacts. Experimen tally, LIP achieves a 2.5-fold resolution improvement and a 3 dB signal-to-noise-ratio gain in the static mode, while maintaining an interactive frame rate of 15 fps in the dynamic mode. We further discuss lens-based HoloFZA as an extension of programmable FZA modulation and discuss the broader significance, current limitations, and future prospects of lensless imaging toward compact, high-quality, and multifunctional visual sensing.

Single-shot structured-light techniques are essential for dynamic 3D measurement. However, speckle-assisted correspondence matching is often affected by correlation mismatches, which severely degrade disparity estimation and reconstruction accuracy. In this paper, a robust single-shot 3D measurement method based on color fringe-speckle pattern is proposed to address correlation-induced mismatches in speckle assisted correspondence. To suppress unreliable correspondences, a region-based disparity optimization framework is introduced. Specifically, reliable disparity candidates are first identified via density-based clustering within continuous phase regions, effectively eliminating severe mismatches caused by false correlation peaks. Subsequently, a neighborhood-assisted diffusion compensation strategy is applied by exploiting local spatial consistency to recover unreliable or missing disparities, thereby improving disparity completeness. Experimental results on isolated and complex static objects demonstrate that the proposed method significantly improves reconstruction quality. Dynamic experiments involving depth-direction translation and rotational motion further verify that the proposed approach enables stable single-shot 3D reconstruction without noticeable motion-induced artifacts. Quantitative evaluation using a standard sphere shows that the proposed method achieves an RMS accuracy of 0.1 mm. The proposed method enhances the reliability of speckle-assisted single-shot structured-light measurement without requiring additional projected patterns, making it well suited for dynamic 3D sensing applications.

In industrial process monitoring, short-wave infrared (SWIR) imaging often encounters extreme luminance variations that exceed the dynamic range of conventional SWIR cameras, requiring simultaneous preservation of dark-region details and bright-region information. For dual-gain SWIR cameras, the high conversion gain (HCG) mode enhances dark-region visibility, whereas the low conversion gain (LCG) mode preserves bright-region details. Neither single-gain mode can adequately retain both dark-region and bright-region details. To address this issue, a dual-gain SWIR high dynamic range (HDR) imaging method based on irradiance normalization and constrained continuous logarithmic mapping is proposed. Experimental results from laboratory tests and field validations verify that the proposed method achieves smoother grayscale transitions, improved dark-region visibility, and effective suppression of highlight overexposure. Local profiles, ablation results, and regional metrics further validate the proposed method. Quantitative analysis of outdoor building scenes yields performance improvements of 12.13%, 6.11%, and 20.13% in information entropy, standard deviation, and average gradient, respectively, compared with the linear hard splicing method.

Quantum cascade lasers (QCLs) operating in the mid infrared (Mid-IR) spectral range are indispensable for high sensitivity sensing and spectroscopy; however, their continuous wave performance is often hindered by significant thermal accumulation and thermally activated carrier leakage. In this work, we demonstrate a high-performance InGaAs/InAlAs QCL emitting at 8.8 μm that incorporates an innovative hollow active region design realized through local compositional modulation. The proposed structure utilizes a simplified active-region scheme with only two alloy compositions, facilitating precise growth control via metal-organic chemical vapor deposition (MOCVD) while maintaining high structural integrity, as confirmed by high resolution X-ray diffraction and transmission electron microscopy. By increasing the energy separation between the upper lasing level and higher-lying leakage states to more than 60 meV, the fabricated devices exhibited exceptional thermal stability. Experimental results show a threshold current density of 2.66 kA/cm2 at 293 K and a peak optical output power of approximately 1.5 W. Notably, the device exhibited characteristic temperatures of T0 = 222 K and T1 = 200 K, which are among the higher values reported for MOCVD-grown QCLs operating in this wavelength range. These findings suggest that the compositionally modulated hollow structure provides a practical and effective pathway for developing thermally robust, high-power Mid-IR QCLs suitable for advanced industrial and medical applications.

Beyond serving as high-capacity communication links, fiber-optic transmission systems can function as kernel platforms. Crucially, the intrinsic optical nonlinearity facilitates the projection of input data into high-dimensional feature spaces, thereby enhancing the linear separability of complex patterns required for effective classification. This capability naturally unifies intelligent computation with optical communication. This paper investigates the feasibility of achieving integrated computation and communication (ICAC) by exploiting the intrinsic and complex dynamical properties of optical fibers through three machine learning tasks: digit recognition, voice recognition, and sentiment analysis on movie review. Numerical analysis results demonstrate that by leveraging the passive fiber-optic kernel, the proposed system achieves recognition accuracies of 95.7% for MNIST digit recognition, 90.8% for voice recognition, and 84.5% for IMDB sentiment analysis. Communication recovery is also evaluated using mean square error (MSE), bit error rate (BER), and error vector magnitude (EVM). Additional fixed-kernel, scalability, and ablation analyses verify the adaptability and physical mechanism of the fiber-optic kernel.

This paper investigates geometric shaping for multiple-input multiple-output optical wireless communications with intensity modulation and direct detection. We characterize the optimal shaping region for arbitrary-dimensional geometrically-shaped spatial-temporal constellations. An efficient method based on asymptotic analysis is also proposed to facilitate the evaluation of error performance and the choice of the temporal dimensionality. Simulation results show that the geometric shaping across spatial-temporal dimensions can attain significant performance gains over the benchmark, in line with our previous demonstrations. This paper contributes an efficient method for computing the optimal shaping region to enable such gains. The approximate and the theoretical results on the shaping gain match well even in low temporal dimensions.

This paper presents a reconfigurable front-end circuit for Single-Photon Avalanche Diodes (SPADs), capable of seamless switching between synchronous time-gated detection and asynchronous free-running photon counting. To minimize silicon area and reduce parasitic capacitance, the proposed architecture employs a shared high-speed active-quenching circuit for both operation modes. In the free-running mode, a bias-controlled current-starved delay path is implemented to provide a programmable dead time, enabling an adaptive trade-off between afterpulsing suppression and the maximum photon count rate. Experimental results demonstrate that the circuit achieves a timing histogram Full Width at Half Maximum (FWHM) of 70 ps in time-gated mode. Furthermore, the free-running mode exhibits a continuously tunable dead time ranging from 2.73 ns to over 100 μs. These results demonstrate the suitability of the proposed circuit for scalable SPAD array readout systems requiring high temporal resolution and adaptive operating modes.

Driven by the booming global digital economy and explosive growth of intercontinental data traffic, transoceanic submarine communication systems face an urgent demand for exponential capacity expansion. However, the capacity of conventional single-mode fiber system has gradually approached the nonlinear Shannon limit, resulting in insurmountable bottlenecks for traditional capacity-scaling approaches. Space division multiplexing (SDM) has emerged as a transformative technology to overcome this limitation by introducing multiple spatial channels, while reducing the cost per bit for submarine systems. Nevertheless, the stringent constraints of deep-sea environments, including limited spatial enclosure, remote power feeding and long-term reliability requirements, present critical challenges for SDM technology. This paper provides a thorough review of the technical evolution, current progress and future development trends of SDM for submarine optical communications. We systematically present the core technical approaches of SDM transmission fibers, SDM-optical amplifiers, and spatial division multiplexers. Furthermore, we summarize key experimental and system-level advances in SDM communication systems targeting submarine applications. Finally, we conclude that SDM has formed a complete technical ecosystem for submarine applications, and discuss some future directions towards high integration, energy efficiency, hybrid core/mode division architectures, and industrial standardization. This review aims to provide a systematic reference for the development of next generation submarine optical communication systems.

We propose and numerically analyze a silica-based birefringent photonic crystal fiber (PCF) featuring a honeycomb lattice with a squeezed geometry and a germanium doped core. The structure is designed to exhibit all-normal dispersion (ANDi) up to 2.3 $\mu$m with maximum of dispersion located around 1.55 $\mu$m. The honeycomb arrangement, combined with controlled lattice anisotropy and auxiliary sub-wavelength air holes, allows precise tailoring of dispersion and polarization properties. The addition of a germanium-doped core region enhances the refractive index contrast and supports a compact mode area favorable for efficient nonlinear interactions. Additionally, we analyze the influence of the design parameters on the chromatic dispersion and on the generated supercontinua characteristics.

We report fluctuation-based image scanning microscopy based on multiple signal classification for threedimensional super-resolution imaging (FBISM-MUSICAL). The method combines ISM-based optical sectioning, MUSICALenabled resolution enhancement and robust reconstruction, and DMD-enabled parallel scanning to improve the throughput of three-dimensional super-resolution imaging. In the proposed architecture, sparse multifocal excitation patterns are generated by a DMD to enable parallelized image scanning. Compared with cumulant-based fluctuation image scanning microscopy, the proposed system reduces the frame burden while preserving the optical-sectioning capability of image scanning microscopy. Under the same 100-frame low-frame-budget condition, FBISMMUSICAL provides cleaner and more stable reconstructions than higher-order SOFISM. Experiments on isolated quantum dots demonstrate an approximately threefold narrowing of the effective lateral point response and an approximately twofold axial response compression using 100 frames per scan position. In biological specimens, FBISM-MUSICAL resolves QD525- labeled microtubules with sub-100 nm resolution. These results show that the proposed DMD-enabled computational photonic imaging system provides a hardware-compatible route toward volumetric fluorescence super-resolution microscopy.