We experimentally demonstrate a multi-wavelength fiber laser based on dual-Sagnac comb filter with LP11 modes output. The filter is used to generate comb spectrum, which is consisted of three optical couplers, two polarization controllers and two segments of polarization maintaining fibers. The proposed fiber laser can stably output up to quintuple-wavelength with the side-mode-suppression ratio higher than 40 dB. Sextuple-wavelength lasing is also obtained, but due to the strong mode competition, the stability should be further improved. A mode selective coupler (MSC) is utilized as mode converter in the cavity. Thus, the multi-wavelength LP11 modes is feasible to generate, owing to the combination of MSC and the filter. The mode purity of the output LP11 modes is higher than 95%. This work has potential applications in optical communication and optical sensing system.

A novel approach for multimode fiber (MMF) curvature sensing based on the texture feature of the specklegram from the fiber facet is proposed, developed and demonstrated. Different from previously reported methods, the texture feature vector of the fiber specklegram is used as the descriptor of the curvature of the fiber. After the specklegrams from two kinds of MMF under different curvatures are recorded, the texture feature vector of each specklegram is first extracted using the uniform local binary pattern algorithm. Then, the texture feature vector is mapped into the target curvature in a nonlinear manner using a one-dimensional convolutional neural network (1D CNN). In the experiment, the model trained by the texture features of fiber specklegrams with a core diameter of 50 $\boldsymbol {\mu }\text{m}$ exhibits high prediction accuracy and good generalization ability in the curvature range of 1.55-6.93 $\text{m}^{-1}$ . In the first experiment, good recognition results are provided by the texture feature of the MMF specklegram, for which the average curvature recognition accuracy is 100%. In the second experiment, the prediction error for 91.4% of the samples in the test set is within a range of $\boldsymbol {\pm }{\mathbf {{0}.{1}\,\,m}}^{\mathbf {-{1}}}$ . In addition, this approach has significantly better prediction ability and robustness than the traditional approach, which processes the specklegram image directly using a two-dimensional convolutional neural network (2D CNN) for sensing. The experimental results demonstrate the fiber curvature sensing capability of the proposed method based on the analysis of texture feature using a 1D CNN.

The increasing development of information technology has led to a higher demand for communication capacity. Moreover, the mode division multiplexing (MDM) is considered one of the important technologies to expand the optical fiber transmission capacity, and the few-mode erbium-doped fiber amplifier (FM-EDFA) is a main device applied for optical fiber loss compensation under long-haul communication transmission systems. This paper reports the design and characterization of a six-mode erbium-doped fiber amplifier (6M-EDFA) for MDM systems. A center-depressed optical fiber with a trench-assisted structure, is designed and the adjustment of the relevant parameters is applied to reduce the splice loss between the six-mode erbium-doped fiber (6M-EDF) and the transmission fiber. The results show the lowest splice loss is theoretically 0.293 dB. By analyzing the mode field distribution, two erbium ion doping regions are initially identified, the ring region is further layered, and the doping concentration of the three layers is optimized to achieve high gain and low differential modal gain (DMG) using a genetic algorithm (GA). In the case of core forward pumping of the LP ₀₁ mode at 1480 nm, the simulation results show that the average gain of 25.7 dB and the DMG of 0.277 dB are obtained at 1550 nm considering the mode coupling. Moreover, a 6-mode MDM transmission system is built to fully verify the performance of the designed amplifier to meet the requirements of the MDM communication system.

To evaluate variations in coagulation, fibrinolysis and endothelial marker expression in cirrhotic patients and to explore their clinical value and predictive performance in cirrhotic patients with or without portal vein thrombosis (PVT), we performed a case-control study with 175 cirrhotic patients and 50 healthy individuals. 99 patients had PVT and another 76 patients did not. All participants were evaluated for plasma levels of conventional hemostatic markers. Thrombin-antithrombin complex (TAT), plasmin-α2-plasmin inhibitor complex (PIC), thrombomodulin (TM), tissue plasminogen activator inhibitor complex (t-PAIC), von Willebrand factor antigen (vWF: Ag) and coagulation factor Ⅷ (FⅧ: c) were also assessed and the ratio of TAT/t-PAIC was calculated. We analyzed differences in these biomarkers among the three groups and constructed receiver operating characteristic (ROC) curves. Patients with PVT exhibited significantly higher TAT and TAT/t-PAIC than cirrhotic patients without PVT (both P < 0.001). Areas under the curve (AUC) of ROC analyses for TAT and TAT/t-PAIC were 0.68 and 0.66, the cut-off levels were 1.55 ng/ml and 0.46, with sensitivities and specificities of 78.79% and 51.32% regarding TAT, 39.8% and 90.79% regarding TAT/t-PAIC. Levels of FⅧ: c and vWF: Ag in patients with PVT were significantly lower than those without PVT (p = 0.026 and p = 0.027, respectively). The AUC^ROC, cut-off level, sensitivity and specificity of FⅧ: c were 0.64, 111.1%, 66.67% and 60%, respectively. For vWF: Ag they were 0.61, 429%, 89.66% and 38.71%, respectively. Cirrhotic patients have disorders of coagulation, fibrinolysis and the endothelial system. TAT, TAT/t-PAIC, FⅧ: c and vWF: Ag can be used as potential biomarkers for predicting PVT in cirrhotic patients.

A reflective tactile sensing scheme based on fiber specklegram analysis is proposed in this study. The specially designed sensor probe with a reflector inside was 3D printed from elastic material. A section of single mode fiber (SMF) and a section of multimode fiber (MMF) were embedded under the reflector. The light emitted from the SMF is reflected by the reflector and the reflected light illuminates the MMF. At the output facet of the MMF, specklegram is detected by a camera. The reflector consists of two soft mirrors that are perpendicular to each other originally. The contact force on the sensor probe causes the variation of the angle between the two mirrors. The transmission path of the light from SMF to MMF changes accordingly, which results in a changed specklegram. In this way, the zero-mean normalized cross-correlation coefficients of the fiber specklegrams decrease linearly with the increase of contact force. The sensitivity and measurement range of contact force are examined with MMFs of different diameters under two operating wavelengths. The negligible effect of fiber bending on the force sensing was also experimentally confirmed. In addition, the convolutional neural network is used to classify the specklegrams of three contact forces applied at nine positions of the sensor probe, and the average accuracy of classification is 100%, which shows its capability to identify the contact position on the probe. On the whole, the proposed tactile sensor is confirmed with a high sensitivity of force sensing in a proper measurement range and a capability to localize the contact position. The reflective structure also facilitates the integration with robot arm and other intelligent structures.

Neural networks (NNs), random forests (RF), and extreme gradient boosting (XGBoost) are applied separately to inversely design the ring-core few-mode fiber (FMF) with a desired weakly coupled performance. We demonstrate the procedure of inverse designing of FMF via machine learning (ML) algorithms and evaluate the prediction accuracy of the above ML algorithms. Compared with RF and XGBoost, the NN performs the highest prediction accuracy. For the NN, RF, and XGBoost, the correlation coefficients of the mode effective index difference are 0.99993, 0.99857, and 0.99937, respectively. Subsequently, by utilizing the method of permuting feature importance ranking, we obtain the high-correlation fiber structural features with the mode effective index difference. Moreover, we analyze the effect of the minimum index difference between two adjacent modes ( Δ n e f f , m i n ) on the structural parameters and get consistent feature attribution via permuting feature importance ranking above. Finally, we design a weakly coupled ring-core fiber that supports four modes ( H E 11 , T E 01 , H E 21 , T M 01 ) based on the NN algorithm, which could be successfully applied in vector mode division multiplexing.