We report the development of low power high-speed VCSEL-based optical interconnects for focal plane arrays (FPAs). The VCSEL epitaxial structure is designed with a 1/2 λ cavity length, with multiple quantum-wells inside, and with top and bottom semiconductor DBRs. The devices were fabricated with a 4.5 μm oxide aperture. The oxide-VCSEL exhibits a 3 dB modulation bandwidth exceeding 50 GHz at low bias current (< 4 mA) for 77-120 K. Optical PAM-4 data transmission is demonstrated up to 138 Gb/s per lane with TDECQ values of 3.42 dB (77 K) and 4.15 dB (120 K). These results demonstrate VCSEL optical link as a plausible and cost-effective solution for FPAs.

Integrating high-speed visible light communication (VLC) with natural human–machine interaction is important for smart lighting and secure indoor Internet of Things (IoT) systems, yet most gesture interfaces rely on cameras or extra sensors, raising privacy concerns and system complexity. This paper presents a waveform-sharing, dual-task VLC receiver that jointly demodulates data and recognizes gestures from the same photoelectric waveform without additional sensing hardware, enabling camera-free, privacy-preserving interactive VLC. Machine learning (ML) is employed for end-to-end compensation and classification: a convolutional neural network (CNN)-based post-equalizer mitigates time-varying distortions from lightemitting diode (LED) nonlinearity, multipath propagation, and gesture occlusion, boosting the raw rate of a 1 m indoor blue-LED link from 30 Mb/s to 46 Mb/s in the communication-only setting, while spectro-temporal features with a lightweight classifier achieve ~98% accuracy over eight dynamic gestures. With gesture inference enabled, the net payload rate remains ~26 Mb/s with negligible added latency, demonstrating that interaction can be overlaid with limited throughput loss and reduced power, cost, and privacy risk.

Conventional vibration demodulation typically utilizes nearby sample points in conjunction with pulses having relatively wide width for demodulation, which has drawbacks, including inconsistent spatial positions corresponding to adjacent sampling points and limited spatial resolution. This research proposes a novel quantitative vibration demodulation technique based on direct detection Φ-OTDR to address this issue. At a set repetition frequency, this technique alternately emits optical pulses with comparable but different pulse lengths. To precisely demodulate the vibration phase, it obtains two optical intensity signals at the same fiber position, which have different initial phases but virtually similar vibration-induced phase shifts. Compared with the conventional demodulation method, the proposed technique achieves a signal-to-noise ratio (SNR) improvement of 9.4 dB. By not relying on the data acquisition card's (DAQ) sampling rate, this technique reduces the acquisition system's cost and performance requirements while avoiding the spatial resolution deterioration associated with conventional demodulation using nearby sampling points.

Thin-film lithium niobate (TFLN) electro-optic modulators have been widely explored at 1550 nm, achieving ultra-low driving voltages, broad bandwidths, and compact footprints. Yet, development at 1064 nm has remained comparatively limited, despite its growing importance for applications that demand high power and cost-effective integration with CMOS-compatible silicon photodetectors such as free-space optics, frequency combs and biophotonics. In this work, we demonstrate a high-efficiency TFLN modulator operating at 1064 nm contributing to fulfilling this gap. The device employs a 1 cm-long interaction region with capacitively loaded traveling-wave electrodes designed to minimize the velocity mismatch between RF and optical modes, achieving an index mismatch as low as 0.05. This careful optimization enables a high frequency half-wave voltage of 1.45 V and a 40 GHz 3-dB bandwidth, representing the best performance reported to date for 1064 nm operation. These results highlight the potential of TFLN modulators to extend beyond traditional telecom wavelengths leveraging lower Vπ, high quality lasers and easy integration with Si photodetectors.

We fabricated an optimized double nested anti-resonant nodeless fiber (DNANF) for the visualization of poled domain in periodically poled lithium niobate (PPLN). Through optimized structural and fabrication processes, the optimized DNANF achieved a low transmission loss below 3×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> dB/m at 920-1040 nm. The femtosecond laser with a central wavelength of 1034 nm and a repetition rate of 34 MHz was coupled with the optimized DNANF in second-harmonic generation (SHG) imaging system. It can be used to achieve a high-resolution three-dimensional (3D) imaging of poled domains in PPLN, with a field of view (FOV) of 102.4 μm × 102.4 μm and a lateral resolution of 0.815 μm. In future, this DNANF will provide a stable, high-efficiency, and simple method of femtosecond laser propagation for the 3D precision characterization of poled domains in materials such as ferroelectric crystals and liquid crystals, as well as two-photon neuroimaging.

In this letter, we present a physics-assisted design strategy for tapered suspended slot waveguides operating at 2 μm. The proposed methodology enables a precise control of the optical field, specifically tailored for laser-driven dielectric accelerating structures. By using calibration curves that link the silicon waveguide strips width to both the effective refractive index and the accelerating field amplitude, we ensure local phase synchronism along the propagation axis. The design was performed for an input energy of 79 keV, and validated by HFSS simulations as far as the optical field computation is concerned, and by ASTRA-code for the single-particle beam dynamics. The results show an output energy of 102 keV with the optimized structure, corresponding to 20% increase compared to the previous constant-field approach.

Doxycycline (DOX) is one of the most widely used veterinary antibiotics in animal husbandry, and its residues may affect the health of humans and animals due to the in-creased bacterial resistance. Here we developed a trace DOX detection means based on molecular imprinting technology with tilted fiber Bragg grating surface plasmon resonance (TFBG-SPR). Sputtering a 50 nm thick gold film on the surface of TFBG to excite surface plasmon resonance, and coating a layer of molecularly imprinted polymers (MIPs) on the surface of the gold film to capture DOX. Experimental results show that the proposed sensor shows a detection sensitivity of 1.89 dB/(μg/mL) for DOX in the concentration range of 0.1-1 μg/mL, with a detection time of about 5 minutes and a detection limit (LOD) of about 0.0323 μg/mL (0.0672 μM). The obtained LOD is half of that of the current similar DOX sensor. The proposed sensor is expected to provide technical support for real-time trace monitoring of DOX concentration in the environment and food.

Computational spectrometers have strong potentials in miniaturization of spectrometer, and have attracted broad interests. But bottleneck problems still exist in their practical performance, stability, cost and mass production. This work aims on a competitive type of computational spectrometer based on Fabry–Perot (F–P) microcavity array, and proposes methods in facile fabrication of the F–P microcavity array with different thicknesses of cavity media and in–situ characterization of it installed in a spectrometer module to extract preset calibration data for reconstructive measurement of optical spectrum. Fabrication of the F–P microcavity array involves only film deposition and UV photolithography techniques in an elaborately designed multiple–step processing flow. In characterization of the microcavity array, responses of the assembled spectrometer module to monochromatic light at different wavelengths are recorded by an inside image sensor. As–obtained calibration data is supposed to include various system errors and uncertain factors for spectral reconstruction.