We present a novel photonic-native compact modeling method for silicon photonics, enabling true bidirectionality and multi-mode/multi-channel simulation. This framework is applied to next-generation high-speed transceivers, offering a solution for high-density, energy-efficient optical interconnects for AI/HPC.

The rising demand for AI training and inference, as well as scientific computing, combined with stringent latency and energy budgets, is driving the adoption of integrated photonics for computing, sensing, and communications. As active photonic integrated circuits (PICs) scale in device count and functional heterogeneity, physical implementation by manual scripting and ad-hoc edits is no longer tenable. This creates an immediate need for an electronic-photonic design automation (EPDA) stack in which physical design automation is a core capability. However, there is currently no end-to-end fully automated routing flow that coordinates photonic waveguides and on-chip metal interconnect. Critically, available digital VLSI and analog/custom routers are not directly applicable to PIC metal routing due to a lack of customization to handle constraints induced by photonic devices and waveguides. We present, to our knowledge, the first end-to-end routing framework LiDAR 3.0 for large-scale active PICs that addresses waveguides and metal wires within a unified flow. We introduce a physically-aware global planner that generates congestion- and crossing-aware routing guides while explicitly accounting for the region of photonic components and waveguides. We further propose a sequence-consistent track assignment and a soft guidance-assisted detailed routing to speed up the routing process with significantly optimized routability and via usage. Evaluated on various large PIC designs, our router delivers fast, high-quality active PIC routing solutions with fewer vias, lower congestion, and competitive runtime relative to manual and existing VLSI router baselines; on average it reduce via count by similar to 99%, user-specified design rule violation by similar to 98%, and runtime by 17x, establishing a practical foundation for EPDA at system scale.

A fiber laser-type current sensing system based on a pre-biased yttrium iron garnet (YIG) probe is proposed and demonstrated. The fiber laser is composed of an erbium-doped fiber (EDF), a fiber Bragg grating (FBG), and a plane mirror. To transduce the external current variations, a YIG probe is strategically employed within the resonant cavity, proximal to the planar mirror, where the magnetic-optical phase shift caused by current variations modifies the phase difference between two orthogonal light waves, resulting in a shift in the polarization mode beat frequency (PMBF) produced by orthogonally polarized light, which exhibits a direct and measurable correlation to the variations in the current. Especially, a biased magnetic field to the YIG probe resolves the nonlinearity problem. The experimental results demonstrate that when a pre-biased magnetic field of 15 mT is applied to the YIG, the system exhibits sensitivities of 405.36 kHz/A at 11.125 MHz and 392.86 kHz/A at 796.3 MHz, with corresponding R-squared values of 0.99963 and 0.99802, respectively. Furthermore, the maximum sensitivity is increased to 661.1 kHz/A by lowering intrinsic polarization to 3.44 MHz. It provides a reliable new option for the upgrading of current detection technology in related fields.

Germanium photodetectors (Ge-PDs) are key components in silicon photonics (SiPh) to enable high-speed optical communications for various applications. Several works have been reported to improve the bandwidth, quantum efficiency, and manufacturing process of Ge-PDs. However, a significant trade-off between bandwidth and responsivity remains a major challenge in developing high-performance Ge-PDs for large-scale silicon photonic integrated circuits. Here, by co-optimizing the chemical mechanical polishing (CMP) process and device structure, ultra-thin Ge-PDs with high 3 dB bandwidth have been achieved. The dimensions of the Ge region and doping areas have also been optimized to attain a high-bandwidth, high-quantum-efficiency product. Consequently, we demonstrate lateral-PIN Ge-PDs with a high bandwidth of up to 100 GHz, 1.05 A/W responsivity at 1550 nm and <20nA dark current on 300 mm CMOS silicon-photonic process. Our work demonstrates that high-performance Ge-PDs with a simple fabrication process are feasible, which paves the way for developing next-generation 200 Gbaud intensity-modulated direct-detection (IMDD) and coherent optical systems.

A fiber Bragg grating (FBG) sensing system based on the beat frequency Vernier effect of a compound cavity laser is proposed and experimentally demonstrated. The FBG in the linear cavity fiber laser serves as the sensing head, while the chirped FBG (CFBG) is utilized for dispersion compensation. By introducing a fiber ring resonator to form a compound cavity and generate the Vernier effect, high-sensitivity sensing is achieved. Temperature sensing experiments show that when the equivalent cavity length difference is 0.24 m, a high sensitivity of 7.05582 +/- 0.11891 MHz/degrees C is exhibited by the sensing system, providing critical technical support for high-sensitivity sensing applications in relevant fields.

Objective Fiber optic nonlinear effects, primarily encompassing self-phase modulation, cross-phase modulation, and four-wave mixing, represent a fundamental technical bottleneck that constrains performance enhancements in contemporary high-speed, high-capacity optical communication systems. Driven by emerging technologies, global data traffic is experiencing exponential growth, creating an urgent demand for higher transmission rates and greater spectral efficiency. This demand invariably pushes optical systems to operate in regimes where nonlinear effects are more pronounced and detrimental. Deep learning has been identified as a potent tool for mitigating such impairments, owing to its powerful nonlinear mapping capabilities. However, a major challenge is the prohibitively high computational complexity associated with advanced neural network-based equalizers, which poses a significant obstacle to their practical implementation and commercial viability despite their superior performance. Achieving an optimal trade-off between compensation performance and computational overhead is therefore a central challenge in this field. To address this issue, this paper proposes and rigorously validates a novel, low-complexity nonlinear compensation scheme. The core of this research is a hybrid neural network architecture that synergistically integrates a convolutional neural network (CNN), a bidirectional gated recurrent unit (BiGRU), and an attention mechanism, complemented by a stepwise bit estimation (SBE) output strategy. The principal objective of this study is to demonstrate that this multi-component, collaborative design can effectively counteract fiber nonlinear impairments while substantially reducing computational complexity relative to other state-of-the-art neural network models, thereby paving a viable path for the practical application of deep learning in next-generation optical networks. Methods The proposed compensation framework consists of a structured deep learning pipeline. The input signal sequence, distorted by fiber transmission, is first processed by a CNN layer that extracts local and abstract features using multiple convolutional filters. These features capture essential nonlinear patterns and prepare the data for temporal modeling. Subsequently, the extracted features are input into a BiGRU layer, which processes the sequence bidirectionally. The GRU structure includes reset and update gates that control information flow and update hidden states, effectively learning long-range dependencies without suffering from vanishing gradients. The BiGRU layer leverages both past and future context to capture the dynamic characteristics of symbol-level distortions caused by dispersion and nonlinear interaction. Following the BiGRU layer, an attention mechanism is employed to dynamically weigh the relevance of each hidden state in the sequence. This selective focus allows the model to emphasize the most informative parts of the signal while suppressing irrelevant or noisy segments. The attention scores are normalized via softmax and used to generate a context vector by weighted summation of hidden states. This context vector is then passed through fully connected layers to synthesize a comprehensive representation for final prediction.At the output, the SBE strategy is adopted. Instead of treating symbol recovery as a multi-class classification task, the model decomposes the prediction of high-order modulation symbols into several binary classification sub-tasks, each corresponding to one bit. Each bit is predicted using a sigmoid neuron and thresholded independently. This output design aligns directly with the bit error rate (BER), the key performance metric in communication systems. Results and Discussions To comprehensively validate the efficacy of the proposed scheme, we conduct systematic evaluations on a coherent optical communication simulation platform. The platform models the transmission of a 32 GBaud polarization-division multiplexed 16-ary quadrature amplitude modulation (PDM-16-QAM) signal over a 1200 km standard single-mode fiber link. Our analysis encompasses both single-channel and five-channel wavelength-division multiplexing (WDM) scenarios, and the simulations incorporate realistic hardware impairments, including laser phase noise and frequency offset. Initially, to verify the effectiveness of the model's core components, we compare the performance of the integrated CNN-BiGRU network against a baseline BiGRU network. The results unequivocally demonstrate that the inclusion of the CNN layer yields a substantial performance improvement. In the single-channel system operating at a launch power of 0 dBm, the SBE-based CNN-BiGRU network achieves a quality (Q) factor improvement of 0.77 dB over its counterpart lacking the CNN, which confirms the CNN's proficiency in extracting nonlinearity-relevant features. The central contribution of this research stems from the integration and analysis of the Attention mechanism. An investigation into the attention weight distribution of the trained network reveals that the model autonomously learns to focus its attention predominantly on the central symbol and its immediate neighbors, while paying significantly less attention to more distant symbols. This observation aligns with the physical nature of fiber nonlinear effects and provides a strong rationale for our subsequent model optimization efforts. Informed by this finding, we implement a model pruning strategy, wherein the input sequences and hidden states processed by the BiGRU are truncated to include only the temporal regions identified as important by the attention mechanism. We then compare the performance of this pruned, low-complexity CNN-BiGRU-Attention network against the unpruned CNN-BiGRU network. The results show that, across both single-channel and WDM systems, the attention-based model achieves a significant reduction in computational complexity at the cost of only a minor Q-factor penalty ranging from 0.1 to 0.4 dB. This outcome highlights the model's excellent balance between performance and efficiency. Finally, we perform a quantitative analysis of the models' computational complexity. The analysis shows that our proposed CNN-BiGRU-Attention architecture exhibits substantially lower complexity than the comparative models across all tested output strategies. When compared to the corresponding CNN-BiGRU network, the introduction of the attention mechanism reduces the complexity by 43%. Moreover, even when compared to the baseline BiGRU network, our overall optimized architecture achieves a 36% reduction in complexity. These figures provide robust evidence of the attention mechanism's exceptional effectiveness in managing the computational load of the neural network. Conclusions This paper proposes and systematically validates a deep learning-based compensation method for fiber nonlinear impairments, built upon a CNN-BiGRU-Attention architecture with a stepwise bit estimation strategy. The method leverages the powerful local feature extraction of CNNs, the precise long-range dependency modeling of BiGRUs, and the intelligent, dynamic resource allocation of an attention mechanism to create an efficient model for nonlinear impairment compensation. The use of the SBE strategy further aligns the model's training objective with practical communication system requirements. Comprehensive simulation results robustly demonstrate that the proposed method not only delivers excellent nonlinear compensation performance but also achieves a significant 43% reduction in computational complexity compared to a similar network without the attention mechanism. This research strikes an effective balance between high performance and manageable computational cost, presenting a viable and promising technical path for the deployment of advanced deep learning technologies in future high-speed, high-capacity optical communication systems.

This paper proposes and validates a highly sensitive temperature detection system based on polarization mode beat frequency signal (PMBFS) demodulation. The system incorporates high birefringence fiber to enhance the stability of the PMBFS, while the low birefringence single-mode fiber (SMF) resonator is optimized for temperature measurement. Theoretical research demonstrates that the difference in the refractive index between the fast and slow axes of PANDA polarization-maintaining fiber (PMF) exhibits a strong linear response to external temperature, resulting in PMBFS drift. By monitoring this drift, we characterize the external temperature. In this study, we introduce Panda PMF with lengths of 10 cm, 14 cm, and 20 cm into the resonant cavity as the temperature-sensing structure. The experimental results confirm that as the length of the panda polarizationmaintaining optical fiber increases, the sensitivity of the sensor to temperature also improves. When PMF is 20 cm long, the maximum sensitivity reaches 3.53409 MHz/degrees C, and the resolution reaches 0.01 degrees C. This method utilizes only PMF for temperature sensing, featuring adjustable sensitivity and resolution. It offers significant advantages such as high sensitivity, high precision, and a simple structure during the temperature monitoring process. Given its excellent performance, the sensor has great potential for applications in biochemical reactions.

With the growing demand for data center interconnects and high-performance computing, silicon photonics has emerged as a key technology for next-generation optical communication and interconnect systems. However, efficient coupling between silicon photonic chips and standard single-mode fibers remains a critical challenge limiting large-scale deployment. In this work, we demonstrate a sub-decibel loss, polarization-insensitive O-band edge coupler fabricated on a 300 mm silicon photonics active platform with two-layer silicon nitrides. The edge coupler employs a cantilever-based structure with a structurally optimized silica waveguide facet, significantly improving mode field overlap and alignment tolerance with standard single-mode fibers. Direct coupling of the optical field into a single-mode silicon nitride waveguide achieves excellent performance, with TE/TM mode insertion losses below 0.87 dB/1.0 dB and a polarization-dependent loss (PDL) below 0.14 dB across the 1260-1360 nm wavelength range. Furthermore, efficient interlayer coupling enables adiabatic optical field conversion from the silicon nitride waveguide to a single-mode silicon waveguide, maintaining high performance over a 100 nm bandwidth with TE/TM mode coupling losses below 0.85 dB/1.2 dB and a PDL below 0.37 dB. The device also exhibits excellent alignment tolerances in all three dimensions (horizontal, vertical, and coupling gap). This research provides a scalable and cost-effective solution for realizing high-performance optical input/output interfaces on advanced silicon photonics platforms, paving the way for large-scale integration in high-density optical interconnects and computing systems.

Vanadium-based aqueous zinc-ion batteries (V-AZIBs) face significant challenges in terms of capacity degradation, zinc dendrite growth, and electrode corrosion; they are affected by the pH fluctuations of the electrolyte during cycling. In this study, a balloon-like bent fiber-optic sensor (BBFOS) was embedded into the electrode/electrolyte interface of a pouch cell, and light signals were collected during charging/discharging and further converted to pH values. Finally, a smart and nondestructive operando device was designed for operando real-time monitoring of the pH evolution of the electrolytes in V-AZIBs. The operando testing results showed a strong correlation between pH evolution and capacity decay in pouch cells and clarified the relationship between pH evolution mechanisms and capacity degeneration in V-AZIBs. Furthermore, the pH stability and the reversible layered chemical mechanism during the charging/discharging process were revealed by using operando BBFOS and operando X-ray powder diffraction analysis, respectively. The highly accurate and nondestructive operando device can promote the progress of AZIBs and other battery systems.

The exponential growth of data traffic in modern data centers has driven the demand for ultra-high-bandwidth, low-power optical interconnects. While 200 Gbps per wavelength has become standardized, next-generation systems aim to exceed this benchmark without relying on power-hungry digital signal processors (DSPs). Leveraging the CMOS compatibility and scalability of germanium photodetectors (Ge-PDs) over traditional III-V-based counterparts, we report, for the first time, to the best of our knowledge, a 240-Gbps O-band optical interconnect over 2 km of standard fiber without any DSP and amplification in both optical and electrical domains. This demonstration employs a >70 GHz thin-film lithium niobate (TFLN) modulator and a 91 GHz Ge PD, establishing a viable path toward compact, energy-efficient, and high-speed heterogeneous integration for future optical transceivers.

Fiber Bragg grating (FBG) sensors are extensively employed for structural health and condition monitoring. However, the high cost and large size of conventional FBG demodulation methods have hindered the widespread adoption of FBG-based sensing technologies. To overcome these challenges, this study proposes what we believe to be a novel demodulation system leveraging a tunable laser. The system utilizes a gated recurrent unit (GRU) peak-finding algorithm, implemented on an STM32 microcontroller with edge AI capabilities, all integrated onto a compact 100×100×10 mm circuit board. This approach enables demodulation across the entire C-band, achieving a monitoring range of 70 km, a demodulation frequency of 100 Hz, and a mean absolute error (MAE) of 9.6 pm.

We propose fiber microstructures based on the dispersion-compensated fiber (DCF) for industrial high- temperature detection (1000 degrees C). The single mode fiber - dispersion compensated fiber - single mode fiber (SDS) sensor achieves a temperature sensitivity of 47.35 pm/degrees C in the range of 30-110 degrees C, which is four times greater than that of a typical fiber Bragg grating (10 pm/degrees C). The single mode fiber - dispersion compensated fiber (SD) sensor based on the Michelson interference (MI) principle is fabricated by splicing a DCF at the end of a single mode fiber (SMF), which effectively suppresses multimode interference and attenuates interference peaks using Fresnel reflection. The sensitivities of the SD fiber sensors with DCF lengths of 7 mm, 8 mm and 9 mm are 62.38 pm/degrees C, 53.45 pm/degrees C and 51.25 pm/degrees C, respectively, in the temperature range of 30-200 degrees C, and the free spectral range (FSR) of the interference spectra decreases with increasing DCF length. After high-temperature annealing of the fiber microstructures, the internal stress of the DCF can be effectively released, which improves the dispersion compensation performance and reduces the transmission loss. We select SD fiber sensors with a DCF length of 7 mm for high-temperature annealing and then repeat the high-temperature experiments over a wide range of 30-800 degrees C, and the interference spectra all show consistent redshifts. The temperature sensitivity in the range of 500-800 degrees C is as high as 106 pm/degrees C. The refractive index difference between the DCF core and cladding changes as the temperature increases from 800-1000 degrees C, and the redshift trend of the interference spectrum with increasing temperature is reversed to a blueshift at 920 degrees C. This sensor, characterized by a wide temperature range, adjustable sensitivity and good repeatability and stability in high-temperature environments, has significant application potential in industrial production.

In this work, we propose a fiber-optic temperature sensor based on single-mode fiber (SMF) -multi-mode fiber (MMF)-etched dispersion compensating fiber (DCF)-single-mode fiber SMF (SMEDS) fiber structure coated polydimethylsiloxane (PDMS) Thermo-gel film with experimental validation. A single-mode fiber is gradually fused with a section of MMF and DCF in this sensor structure. The central region of the DCF was then etched with hydrofluoric acid (HF) to create a strong evanescent field. Finally, the temperature sensitivity of the sensor was significantly improved by coating the sensor area with PDMS thermos-gel material. The PDMS thermos-gel film material serves as a simple and efficient encapsulation layer that not only protects the sensor's microstructure from damage but also greatly improves the reliability and durability of the sensor structure. The experimental results show that the sensor has a temperature sensitivity of 0.113 nm/degrees C over a wide temperature range of 25-95 degrees C. When compared to fiber optic sensors without PDMS coating, the temperature sensitivity is 2.9 times higher. The small structural design of the SMEDS sensor, along with its easy production method, high-cost efficiency, and stable and reliable mechanical strength, suggests that it has a wide range of significant applications in the temperature sensing industry.

We propose a macroscopic loss-based olive-shaped single-mode fiber (OSSMF) for displacement sensing in the fiber loop ring-down, which validates the feasibility of displacement sensing. In this work, two systems consisting of single-point and multi-point displacement sensing are built, and the ring-down curves are demodulated using low-cost microcontroller unit and self-developed optical time domain reflectometer (OTDR), respectively. The long-cavity system is initially utilized for single-point displacement sensing to confirm the possibility of multi-point displacement detection with short cavities. The relationship between cavity length and sampling frequency is also discussed, enabling low-frequency, high-precision sampling of the ring-down curve. In the short-cavity multi-point experimental system, different sizes of olive-shaped sensing heads are compared in a 65 m cavity length scenario. It is concluded that the OSSMF with a short axis of 2.1 cm and a long axis of 3.5 cm exhibited the highest sensitivity. The attenuation is -2.11 dB/ (km mm) in the 8-12 mm range and -7.24 dB/ (km mm) in the 12-24 mm range. The displacements are detected in 65 and 125 m fiber loop ring-down (FLRD) cavities, and the ring-down curves are processed using Fast Fourier Transform, this method addresses the issue in multi-point measurements where an extension cord is needed for time-domain differentiation. Different cavity length designs enable distributed measurements of physical quantities, the system is of great significance in the fields of structural engineering monitoring and geological disaster monitoring.

Breathing is an important physiological indicator of human health. In this article, a wearable compact single-mode-no core-single-mode fiber (SNCS) optical microvibration sensor-based breathing monitoring system is proposed. First, piezoelectric vibration experiments were conducted, focusing on low-frequency signals to verify the feasibility of optical fiber microvibration sensor (FMVS) for breathing monitoring and measured frequencies reached to 0.1 Hz. The piezoelectric vibration frequency is fitted to the measured frequency with a slope of 0.99772 and correlation of 0.99879. Multipoint wearable breathing monitoring experiments on the nose and abdomen were conducted. It is demonstrated that breathing states could be identified by details, such as respiratory rate and exhale and inhale time between three normal and four irregular breathing states. Real-time and precise breath detection are ensured by the quick response time (0.11 s) and recovery time (0.14 s) of the sensor. The proposed sensor is repeatable, responsive, comfortable to wear, recognizes different breathing states, and can be used for the prediction of obstructive sleep apnea (OSA) syndrome and posttreatment rehabilitation monitoring, which has a promising medical application.

A fiber temperature sensing scheme based on beat frequency and Vernier effect of fiber ring resonator (FRR) with composite cavity is proposed and experimentally demonstrated. The optical FRR as a sensor is put into the linear cavity fiber laser, and Vernier effect is generated when the resonance spectrum of the optical FRR and the longitudinal mode of the laser are superimposed. The beat frequency envelope spectrum corresponding to the longitudinal mode Vernier spectrum is used to realize fiber ring sensing in electrical domain, and the sensing information is obtained by tracing the envelope of beat frequency signal, which has the advantages of stable system, simple structure, low cost and high sensitivity.

A fiber optic refractive index (RI) sensor based on an etched multimode fiber (MMF) with a double peanut-shaped structure is proposed and experimentally demonstrated. The sensor consists of two peanut-shaped and a section of etched MMF tapered fiber structure. The excitation of the fundamental mode to higher-order modes is facilitated by using the beam splitting/coupling effect of the double peanut-shaped and etched taper structures, and the higher-order modes can be excited into an evanescent field. In the sensing medium, the stronger the evanescent field, the stronger the energy shock between the fiber and the sensing environment. Experimental results showed that the sensitivity was 326.52 nm/RUI and 823.91 nm/RUI when the etched waist taper diameter was 51.92 μ m and the glycerol solution index ranged from 1.3395 to 1.3945 and 1.3945 to 1.4200, respectively. Compared to the MMF sensor structure without etching, the RI sensitivity is improved by about 2 times. In addition, the temperature characteristics of the sensor were investigated over a range of 30 °C–100 °C, and the results showed a maximum temperature sensitivity of only 30.24 pm °m −1 . The sensor structure has a low-temperature sensitivity and the temperature effect on the RI measurement results is negligible within the allowable error range. The sensor has the advantages of simple fabrication, wide measurement range, good stability, low cost, and compact structure, which has potential application value in the field of RI detection.

Underwater wireless optical communication (UWOC) plays key role in the underwater wireless sensor networks (UWSNs), which have been widely employed for both scientific and commercial applications. UWOC offers high transmission data rates, high security, and low latency communication between nodes in UWSNs. However, significant absorption and scattering loss in underwater channels, due to ocean water conditions, can introduce highly non-linear distortion in the received signals, which can severely deteriorate communication quality. Consequently, addressing the challenge of processing UWOC signals with low optical signal-to-noise ratios (OSNRs) is critical for UWOC systems. Increasing the transmitting optical power and investigating more advanced signal processing technologies to recover transmitted symbols are two primary approaches to improve system tolerance in noisy UWOC signal channels. In this paper, we propose and demonstrate the application of deep echo state networks (DeepESNs) for channel equalization in high-speed UWOC systems to enhance system performance with both PAM and QPSK-OFDM modulations. Our experimental results demonstrate the effectiveness of DeepESNs in UWOC systems, achieving error-free underwater transmission over 40.5 m with data rates up to 167 Mbps. Moreover, we compare the performance of DeepESNs to conventional echo state networks and provide suggestions on the configuration of a DeepESN for UWOC signals.

In this paper, we propose a method of multipoint displacement measurement using fiber-loop ring-down (FLRD) technology. We demonstrated the feasibility of multipoint displacement measurement using three cascaded FLRDs. Single-mode optical fiber wrapped around the cylinder airbag surface (CA-SMF) is used as the displacement sensor. Due to the low insertion loss of CA-SMF and the high-power modulated signal output of OTDR, it is not necessary to add any device for compensating the pulse signal power in FLRD. Finally, we also consider the errors that arise in the engineering manufacturing process. We have experimentally demonstrated that the length and intrinsic losses of FLRD do not affect the sensing performance of displacement sensors in a reasonable range, which is extremely helpful for practical engineering applications.