We propose and experimentally demonstrate a multiwavelength erbium-doped fiber laser (EDFL) based on a compound filter. The compound filter is composed of polarization controllers (PCs), polarizers and a piece of home-made polarization-maintaining photonic crystal fiber (PM-PCF). Due to the high birefringence and few mode properties of the PM-PCF, Lyot filter and multimode interference filter can be realized simultaneously in the compact structure. Hence, the EDFL can achieve flexible lasing output. By adjusting PCs, the lasing can obtain single-, dual-, triple-and quad-wavelength output, respectively. The tuning range of single-wavelength lasing output can exceed 20 nm with side-mode suppression ratio (SMSR) above 40 dB. For dual-wavelength lasing output, the tuning range with fixed interval can reach 16 nm. Besides, the wavelength interval can be adjusted from 2.5 nm to 17.4 nm. The great interval adjustability is also observed in triple-wavelength output. The wavelength interval can be adjusted from 8.2 nm to 14.7 nm. Compact structure, flexible lasing output, and good stability make our proposed EDFL more convenient and practical to the applications such as sensing and wavelength division multiplexing.

A denoising method based on singular value decomposition (SVD) with particle swarm optimization (PSO) is proposed to improve the signal-to-noise ratio (SNR) of far-end disturbances in distributed sensor system based on phase-sensitive optical time-domain reflectometry ( $\Phi $ -OTDR). Also, an improved clustering algorithm is introduced to locate the position of disturbance. The effective sensing distance of the $\Phi $ -OTDR system is 25.05 km and four kinds of disturbance events, including watering, knocking, climbing and pressing, are applied on the sensing fiber respectively. A series of experiments of single-point far-end disturbances and five-point disturbances are carried out. Experimental results demonstrate that the SNR of far-end disturbance can be effectively improved to over 12dB, the processing time is less than 3 seconds, and the average location accuracy rate is more than 96%. Compared with the commonly used denoising methods, such as empirical mode decomposition (EMD), wavelet-1D and wavelet-2D, the SVD denoising with PSO method has better performance in SNR enhancement and real-time, which is beneficial for accurate positioning.
© 2001-2012 IEEE.

A novel hybrid hollow-core polarization-maintaining fiber is proposed by combining the photonic bandgap mechanism and anti-resonant effect. High birefringence can be achieved by introducing a pair of negative curvature anti-resonant layers into the air core of a 7-cell HC-PBGF to obtain two-fold rational symmetry. Within the wavelength range from 1.45 mu m to 1.65 mu m, the birefringence reaches up to the magnitude order of 10(-3) and high-order mode suppression ratio, which is defined as the ratio of the lowest loss of high-order modes to the highest loss of fundamental modes, is greater than 100. Hence, the proposed fiber can realize high birefringence and single mode simultaneously in the 200 nm wide wavelength range. Besides, an extended structure by changing the position of the anti-resonant layers also proves the advantages of hybrid structure in achieving high birefringence and single mode operation. The proposed hybrid structure owns great potential for polarization-sensitive applications and provides a new idea to design hollow-core polarization-maintaining fibers with high birefringence and single mode.

An antiresonant hollow core effective single-mode terahertz fiber is proposed and successfully fabricated with a photosensitive resin (SomosEvoLVe 128) by a 3D printer. The single mode characteristics of the fiber are found to be related to the area of the semielliptical tubes in the cladding. By optimizing the cladding structure parameters, a Higher Order Mode Extinction Ratio (HOMER) of above 6 can be achieved. A 30 cm long sample with optimized structural parameters is obtained by cascading two 15 cm long fiber samples fabricated by a 3D printer. The transmission loss and the electric field distribution are measured by a terahertz time domain spectroscopy (TDS) system. The electric field distribution measured at the output end of the fiber sample has the Gaussian profile, which accords with single-mode field distribution. Experimental results show that the average loss is 0.048 cm-1 within the frequency range from 0.2 to 1 THz, and the minimum loss is obtained as low as 0.009 cm-1 at the frequency of 0.82 THz.
© 2013 IEEE.

We propose a temporal Fresnel filtering based repetition rate multiplication (RRM) scheme suitable for multi-wavelength optical pulse trains. Residual chirp terms inside and outside the temporal Fresnel integral contribute to preserving of the multi-wavelength feature. Implementation of our scheme is very simple and compact, and only one linearly chirped Bragg fiber grating (LCBG) and one temporal grating (TG) are needed. Optical signals before and after the TG share the same LCBG by forward and backward passing to implement temporal original and inverse Fresnel transforms, respectively, avoiding possible mismatch between these two transforms and minimizing the influence of the third and higher order dispersion coefficients. Electrical programming of the TG contributes to the tunability of multiplication factor m. In numerical simulations, our scheme succeeds to preserve the multi-wavelength feature of an input pulse train, while the current temporal Fourier filtering and temporal Talbot effect based schemes fail, verifying the novelty of our scheme. Our work is especially useful for tuning the scanning resolution of steering angles of microwave beams, formed by photonic-assisted phased array antennas, in the application of five-generation (5G) wireless communications.
© 2020

A temporal phase-truncated double random phase encoding cryptosystem is proposed to be improved by replacing the temporal Fourier transformers (TFTs) of the dispersion-time lens-dispersion structure with temporal skew Fourier transformers (TSFTs). A TSFT is similar to a TFT except for the time lens (TL) replaced by a skew time lens (STL). A STL is a temporal analog of a spatial cylindrical lens with the in-plane rotational asymmetry mapping to a front–back asymmetry. The front–back asymmetry is achieved by adding the chirp factor C of a TL by +ϵ⋅C and −ϵ⋅C respectively along the positive and negative half axis of the local time coordinate. The skew coefficients ϵ₁ and ϵ₂ of the two TSFTs involved in the encryption stage are dimensionless and ideally unbounded, serving as additional secret keys. By mathematical analysis, STLs are proven to have complex-value modulations on low frequency components of input signals of TSFTs, which cannot be removed by phase truncations (PTs) in the legal decryption stage. The plain/cipher-text-independent and non-removability features of STLs make phase retrieval attacks unable to exempt eavesdroppers from brute-force trials of STLs. A STL is implemented by an electrical arbitrary waveform generator (AWG) driven electro-optic phase modulator (EOPM). By numerical simulations, the using of STLs is proven to increase the difficulty of phase retrieval attacks in terms of the time consuming required for convergence. Even if phase retrieval attacks converge, the plaintext signal cannot be recovered without the knowledge of ϵ_1,2 in the encryption stage. The total key space of ϵ_1,2 is also evaluated as 2¹⁷ at AWG bandwidth B_m=20 GHz, and as 2³⁷ at B_m=100 GHz. The fidelity robustness of our optical cryptosystem against possible noise and occlusion penalties on the communication channels and networks of optical cipher-text signals is also proven.
© 2020 Elsevier B.V.

To reduce the nuisance alarm rate (NAR) in phase-sensitive OTDR sensing system, a novel event identification model based on principal component analysis (PCA) and probabilistic neural network (PNN) is proposed. By training a PCA-PNN model, five kinds of disturbance events including four kinds of real disturbance and one kind of false disturbance can be effectively identified. Experimental results indicate that the average identification rate of five kinds of events reach 97.74%, with an average response time of 0.93 s. Multiple events identification with a high identification rate and fast response makes the proposed method more adaptable in practical application.