Under the paraxial approximation and based on the generalized Huygens-Fresnel diffraction integral formula, the propagation expressions of three-dimensional intensity of sinh-squared-Gaussian beams in free space, in a left-handed material slab and after propagation through the left-handed material slab are derived, and the numerical calculation and analysis are performed with the expressions to explore propagation properties of the sinh-squared-Gaussian beams through the left-handed material slab. The results show that the influence of the left-handed materials on the axial and the transversal intensity distribution of sinh-squared-Gaussian beams makes the intensity exhibit symmetrical distribution. The negative refractive index of the left-handed material slab changes the position of the maximum intensity. The axial and the transversal intensity changes with different propagation distances and decentered parameters, which shows that the sinh-squared-Gaussian beams cannot remain the same during propagation. In practical applications, it is feasible to shape beams in the manner of adjusting the decentered parameter at different locations.

The electric vector feature of fractional high-order Bessel vortex beam (FBV) is studied based on the vector wave analysis. Under the tightly focused and non-tightly focused conditions, the changes of three electric-field components of Ex, Ey and Ez are studied with the topological charges (TCs) from 2.1 to 3 in the increment of 0.1. Moreover, the Ex component of the FBV beam with integer and half integer TCs are comparatively analyzed during the process of the imaging scheme from tightly focused to non-tightly focused conditions. Numerical simulation results show that there are obvious difference within the three electric-field components and the circular symmetry of the bright rings are all broken. Under the non-tightly focused conditions, the circular symmetry of Ex component intensity increases. However, the distribution of Ey and Ez components demonstrate same with that under the tightly focused conditions. The circular symmetry of the Ex component of integer TCs bright rings gradually increased during the process of the imaging scheme from tightly focused to non-tightly focused conditions. Different with those, the formation of the half-integer TCs bright rings patterns remain unchanged and only has a magnification velationship.

The aim of this study is to optimize the robustness and controllability of the 3D printed hydrogel scaffolds by iteratively reducing the mismatch between the designed and the as-printed. A feedback loop approach based on optical coherence tomography (OCT) in vivo online quantitative evaluation was performed for twice. The experimental results show that OCT has quantitatively characterized the morphological parameters, and the difference correlation analysis based on the characterization of OCT feedback controls the 3D printing process and enables decrease of the mismatch. The mismatch of the averaged pore size of the hydrogel scaffolds has decreased from 30% to 2%. It concludes that OCT can further expand its applications in the field of tissue engineering, and may be a key tool for scaffold design and characterization, 3D bio-printing process control, and so on.

Phase retrieval and quantitative phase imaging are central subjects in optical measuring and imaging technologies. The most well-established method for obtaining quantitative phase is through interferometry. However, this class of methods relies heavily on the superposition of two beams with a high degree of coherence, and complex interferometric device, stringent requirement on the environmental stability, and associated laser speckle noise greatly limit its applications in the field of microscopic imaging. On a different note, as one of the typical phase retrieval approaches, the transport of intensity equation (TIE) provides a new non-interferometric way to access the quantitative phase information. In recent years, it has been extensively studied and remarkable advancements have been made in the fields of adaptive optics, X-ray diffraction imaging, electron microscopy, and optical microscopy. In this work, we will review the basic principles and some recent advances in TIE phase retrieval, including its solutions, axial intensity derivative estimation, partially coherent imaging and light field imaging, with emphasis on its applications in the field of quantitative phase microscopy. The challenging problems as well as future research directions will also be discussed.

Structured illumination formed by modulating the phase and the amplitude of input beams is used to improve the resolution of digital holographic microscopy. According to the properties of structured illumination, bright-field and dark-field recording systems are built. In the bright-field system, an amplitude type sinusoidal grating and scattering elements are used to modulate the high frequency information which overpasses the diffraction limit into the frequency lower than the cut-off frequency, which can be recorded by an imaging system. By shifting the high frequency information to the correct position and combining the low frequency information, the resolution of the reconstructed image can be improved in the process of digital representation. In the dark-field system, both the amplitude and phase of the input beams are modulated by loading different images on a spatial light modulator to generate Laguerre-Gaussian vortex beam, Airy beam and Airy beam with vortex phase. A darkfield condenser used in the system focuses these beams on the sample respectively. It is found that not only the resolution but also the contrast is enhanced.

Imaging model and parameter calibration method on light field camera are the theoretical basis and key points on computer vision applications, and drawing a great number of attentions in computational photography. Different from the principle of traditional camera, light field camera corresponds the light in the scene with imaging sensor pixel of novel optical systems according to a given relationship, both the position and angular information of ray are sampled and recorded and it is able to recover the three dimensional structure of real scene by computational photography. The model and the calibration of light field camera are analyzed with micro-lens array, and the effects of different physical parameters on the results of light field sampling are compared. Then major problems of calibration on light field camera are analyzed, including decoding, calibration model, error analysis and evaluation. The parameter calibration method of light field camera is induced and summarized.

The prolate spheroidal wave function is a set of complete orthogonal functions set in both the finite and infinite space domain, which is suitable to analyze the practical optical systems with finite aperture size .The linear canonical transform is a kind of important time-frequency analysis tool, and the Fresnel transform is a special case of linear canonical transform. So, the linear canonical transform can model optical system. The offset linear canonical transform of prolate spheroidal wave functions in finite space domain and frequency domain is instituted, and the offset linear canonical transform models a two-dimension optical system. Using the spheroidal wave functions as signal functions, the signal′s energy loss after passing through the system is analyzed. Based on prolate spheroidal wave functions′ eigenvalue property, the eigenvalue reflects the energy-preservation ratio. The numerical calculation result suggests that the signal function′s energy ratio after passing through the system is consistent with the energy ratio′s estimation value gained by the function′s eigenvalue, which indicates the validity of this method.

Based to the Lagrange invariant law, the lateral and axial imaging resolution of self-interference incoherent digital holography (SIDH) imaging system is discussed in detail, comparing with the resolution of classical optical imaging system. The mathematical expression for the lateral, axial imaging resolution and the resolution capability criterion are presented in theory, according to the system magnification and the full-width at half-maximum of the point spread function. It indicates that the SIDH can improve the lateral resolution of one time when the hologram is recorded at the position, and the two spherical waves from the object point are perfectly overlapped, comparing with the classical coherent optical imaging systems. The compression sensing numerical reconstruction algorithm is introduced to improve the axial resolution of system, and the corresponding numerical simulation and experimental results are obtained. The analysis results have important guiding significance for SIDH in imaging, measurement and optical system design.

To solve the problem that the memory space is large and the calculation speed is slow in the novel look up table (N-LUT) method when calculating the color hologram three-dimensional scene. The depth and wavelength compensating N-LUT method (DWC-N-LUT) is proposed to generating color hologram three-dimensional scene fast. The point cloud, which compose three-dimensional scene, are sliced as a series of two-dimensional images with defined depth. The principle fringe pattern (PFP）for red (R) channel corresponding to one depth is obtained according to the pre-stored depth compensating factors. The complex amplitude of object point for R channel is acquired by shifting the PFP. The complex amplitude of object point for G and B channel is directly generated through multiplying that for R channel by wavelength compensating factors. Numerical simulation experiments with an airplane model and a car model indicate that the color and depth of three-dimensional scene can be precisely provided with DWC-N-LUT method. The coherent noise in the reconstruction images can be obviously eliminated by adding random phase to the complex amplitude of object point. The memory space of DWC-N-LUT method is 10 times less than that of N-LUT method, and the computational speed of the proposed method is approximately improved 22% comparing with N-LUT method.

After the CMOS detector in imaging system is damaged by laser, the corresponding cat eye echo will change. The changes of the cat eye echo are researched experimentally. With the increase of laser power for damage, the microlens of the CMOS detector decomposes gradually, and disappeares at last. The surface of the detector turns into a two-dimensional grating composed of aluminum film and sensitive area in the end. The array spots in the cat eye echo-wave pattern goes through messy and then turns back arrays. The surface morphology of the damaged microlens is modeled by a Gaussian random surface in simulation, and the changes of the echo-wave pattern are calculated with the damage of the microlens. The theoretical results are consistent with the experimental phenomena.

An acousto-optically Q-switched vortex Nd:YAG laser by annular pumping of circular Dammann grating (CDG) shaping is investigated. The cavity of the laser is composed of a laser crystal, an acousto-optical modulator and a planar output coupling mirror. The pump source is a fiber coupling semiconductor laser of 808 nm. The annular intensity distribution of pump light is formed by the first-order diffraction of a CDG with high diffraction efficiency, which is used to end pumped Nd:YAG laser crystal. The octive Q-swith pulse laser is obtained with high optical quality, linear polarization and spiral phase. The obtained power of laser pulse is 470 mW, the peak power of laser pulse reaches 588 W and pulse width is 160 ns, when the absorbed pump power is 5.6 W and the frequency of acousto-optical modulator is 5 kHz.

Based on the model of ytterbium-doped fiber amplifier and the effect of the central wavelength of the amplifier and the length of gain fiber on amplified spontaneous emission (ASE) is analyzed. According to the calculation results, the 1030 nm narrow linewidth fiber laser amplifier is optimized. Adopting the structure of master oscillator power amplifier (MOPA), a continuous wave (CW) laser with central wavelength of 1030 nm, 3 dB linewidth of 0.072 nm and maximum power of 1.01 kW is achieved with narrow band seed and one stage amplifier, in which 25 μm/400 μm (diameters of core/cladding) ytterbium-doped fiber is used as the gain fiber. The optical-to-optical conversion efficiency is 81%, and 1030 nm output power is more than 99% of total output power. The calculated and experimental results show that by reasonable optimization design, 1030 nm narrow linewidth CW fiber amplifier with high power, high efficiency and high signal noise ratio can be achieved.

A laser diode dual-end-pumped, single-frequency Q-switched Nd:YAG laser is implemented. Single-frequency seed injection is realized by utilizing an RbTiOPO4 (RTP) crystal as the intracavity phase modulator to modulate the length of the cavity, which leads to the single longitudinal mode output. Adopting another pair of RTP crystals as a Q-switcher, the laser is capable of generating 7.5 mJ pulse-energy with 14.6 ns pulse duration, and 45 MHz linewidth at 900 Hz repetition rate. The beam qualities M2 of 1.30 in the horizontal direction and 1.36 in the vertical direction are obtained. The frequency stability is 1.1 MHz and the linewidth stability is 0.52 MHz over 2 min laser output. In conductively cooled structure, the output of single-frequency pulsed laser with narrow linewidth, high frequency stability and high linewidth stability is obtained.

An acousto-optic Q-switched fiber laser experimental system has been established, and the multi-peak phenomenon of the output Q-switched pulse has been analyzed. In order to find the Q-switched pulse with less multi-peak number and better shape, we adjust the controllable parameters like the pump power of system, the repetition rate of acousto-optic Q-switch, opening time and rise time. At last, when the pump power is 8 W, repetition frequency of Q-switch is 10 kHz, opening time is 260 ns, and rise time is 150 ns, we have got a Q-switched output pulse with a slightly protuberance in the front, the conversion rate of laser is about 23.6% at the moment. The multi-peak number of output pulse is reduced successfully in the experiment, and the output waveform of Q-switched pulse is optimized. The analysis of the experiment result provides methods for outputting smooth pulse of acousto-optic Q-switched fiber laser.

Amplified spontaneous emission (ASE) is a key issue which affects the gain of large-aperture Ndglass slab amplifiers. Gain measurement experiments under different pump conditions are carried out to study the effect of ASE on gain coefficients and gain distributions. The gain coefficient derating ratio caused by ASE is calculated based on the experimental data. The results show that the gain turning-down and the decrease of average gain becomes notable as the pump energy increases. The study provides experimental support for further optimization of home-made large-aperture slab amplifiers.

Using the rubidium oxygen titanium phosphate (RTP) electro-optic deflector as a Q-switch, we realize a continuous wave LD-pumped Q-switched Nd:YVO4 laser and study the output characteristics in experiments at different output coupling rates and different repetition rates. An output pulse train of 1.22 W average output power with pulse width of 1.0 ns and peak power of 224 kW is obtained at the repetition rate of 5 kHz and an output coupler with transmittance of 60%. Another out pulse of 2.67 W average output power with pulse width of 2.2 ns and peak power of 60.7 kW is obtained when the repetition rate is 20 kHz, the slope efficiency is 37%, and the beam quality factors are M2x=1.226, M2y=1.229. Further, by extra-cavity double-frequency based on a potassium titanium oxide phosphate (KTP) crystal, a green laser at 532 nm with repetition rate of 20 kHz and average output power of 1.33 W is obtained, and the corresponding frequency-doubling efficiency is 50%.

By using a fiber grating as a reflector, a small part of the dual-polarization fiber laser is delayed by the fiber delay line, and then fed back to the laser cavity, which can reduce the phase noise of beat frequency. Results show that the longer the feedback time is, the greater the noise reduction magnitude is. With a 50 m single mode fiber as the delay line, more than 20 dB phase noise reduction is achieved for a dual-polarization fiber grating laser, and a phase noise of -92 dBc/Hz at 10 kHz offset is achieved. The polarization state of optical feedback has influences on noise suppression capability of the external optical feedback system, and it has weak tuning capability to beat frequency of the dual-polarization fiber.

Laser underwater cutting of 1 mm thick 304 stainless steel is demonstrated by a fiber-transmitted laser beam aided by high pressure argon gas. In order to evaluate cutting efficiency and quality via average kerf width, laser underwater cutting is carried out by changing the parameters of laser power, cutting speed, thickness of the water layer, salinity and temperature of the water. On a macro level, lower laser power, faster cutting speed and thicker water layer decrease laser cutting efficiency and quality. Meanwhile, high salinity and low temperature of water decrease laser cutting efficiency as well. On a micro level, metallographic structure and microhardness of the fusion zone, heat affected zone (HAZ) and substrate are different. Nucleation is detected on the surface of fusion zone. High laser power density promotes crystal grain growth. The microstructure in HAZ is coarse. The microhardness of HAZ is lower than that of the substrate and fusion zone. Microhardness of the outer surface of the fusion zone is 242.8 HV with some oxidized zone showing a hardness of 963 HV, 4.3 times of the substrate. While those of the middle part of the fusion zone, HAZ and substrate are 165.1, 124.6 and 223.4 HV, respectively.

Laser shock peening is a novel material surface modification technique, which is superior to conventional shot peening technology. The experiments and finite element analysis method are combined, the residual stress distribution of the fastener hole after laser shock peening is discussed under certain impact order. The results show that the combination of several small spots with diameter of 2.6 mm can lead to a large circular processed area of nearly 6 mm in diameter, which can replace a single large spot for shock peening. The amplitude of the surface residual stress is increased from 134 MPa after the first impact to 254 MPa after multiple laser shock peenings, and the maximum depth of shocked region is also gradually enhanced to 26.6 μm. After hole drilling in the shocked region, the maximum residual stress around the fastener hole edge decreases sharply. The simulation value agrees well with the experimental value.

Stress distribution and failure behavior of laser transmission welding parts are studied by means of simulation and experiment during the tensile process. The welding parts of PA66 are chosen as the research object and the numerical simulation model of the tensile experiment is established. The tensile deformation and the corresponding load-displacement curves of welding parts are simulated and compared with the results of tensile experiments. Analysis of the shear and Von Mises stress distribution of welding parts during the process of tensile experiment is carried out. The failure behavior of tensile specimen can be studied from the aspect of shear and tensile failure. The numerical simulation model can well predict the tensile deformation of the welding parts. The maximum shear stress occurs in the vicinity of the four corners of a rectangular weld area near the weld interface, which are also the starting positions of the shear failure. Because the maximum shear stress is much less than the shear strength of PA66, the tensile shear failure is less likely to occur. The predicted tensile failure and the failure position of welding parts by numerical simulation are consistent with the experimental results. The accuracy of numerical simulation method for analyzing and predicting the tensile failure behavior of the welding parts and the reliability of the numerical simulation model of tensile process are verified.

A7N01 aluminum alloy with the thickness of 4 mm is welded by using fiber laser-variable polarity TIG hybrid welding with filler wire. The effects of both welding current and welding speed on weld formation and porisities are discussed. The microstructure and mechanical properties of the joint obtained at optimal welding parameters are investigated. Experimental results show that the top weld width increases and the bottom weld width is unchanged with the increase of the welding current. With the welding speed decreasing, both the top and the bottom width increase, and the internal porosities decreases. Under the condition of laser power of 6 kW, welding current of 180 A, welding speed of 4 m/min and wire feeding rate of 4 m/min, the joint with good formation and few welding defects is obtained. Weld cross section is characterized by the wide top weld width and narrow bottom weld width. The microstructure of the joint is composed of equiaxed grain zone, columnar zone and equiaxed dendrite zone from the fusion line to the center of weld. The sizes of the grains and equiaxed dendrites from the top to the bottom are gradually reduced along the weld centerline. The fusion zone hardness is lower than that of the base metal, which indicates that the soften zone is occurred in the joint. The average tensile strength of joints is 369.8 MPa, which is approximately 83% of the base metal. And the extension rate is about 4%. The tensile fracture occurs in the weld zone, and the fracture morphology is distinctly ductile characteristics.

Laser double-wire MIG arc hybrid welding has the characteristics of large weld penetration, high metal deposition rate and good welding quality, however, the underlying interaction is so complicated that it brings some difficulties in the practical applications. A coupling mechanism between two arcs and laser-induced plasma is proposed based on the electromagnetic field theory. When the difference between the Lorentz force and electric field force generated by the two arcs acting on the electrons in the middle part of the laser-induced plasma is significant, the electron distributions at two ends of plasma are not uniform, which causes one arc bend, the driving force to be reduced along the wire axis, and the droplet transfer become more difficult. When the two force are balanced, the electron distributions at two ends are relatively uniform, and the plasma can attract and stabilize the two arcs, which results in a smooth droplet transfer. Laser double-wire MIG arc hybrid welding process is experimentally studied by using the high speed photography system and signal acquisition system, and the results show that when the welding process parameters are not suitable, the arc severely deviates from the wire axis so that unstable globular transfer and short circuiting transfer occur. In contrast, when the welding process parameters are suitable, the roots of the two arcs are fixed to the bottom of laser-induced plasma and a smooth and directional spray metal transfer with low spatter loss is achieved.

The micro-dent pattern with different center distances was fabricated on nodular cast iron QT600 specimens by laser peening, and the effect of laser texturing on its surface friction and wear properties was investigated. The results show that the beneficial compressive residual stress was introduced on nodular cast iron QT600 surface by laser peening, and the wear mass loss is significantly reduced by 7.9% under dry friction conditions. Compared with that of the untextured specimens, the friction coefficient of the textured surface increases by 12.9% under dry conditions, and decreases by 11.8% under lubrication conditions. Meanwhile, the surface wear of the textured specimens is obviously weaker than that of the untextured ones, and the oxide content also decreases obviously.

One method of dielectric microsphere assisted laser irradiation followed by chemical etching for fabrication of sub-micro structure arrays on silicon surface with high processing controllability is presented. Three dimensional micro-structure arrays with high quality of periodicity and homogeneity can be fabricated on silicon surface. The formation rules of microstructure depending on microsphere size, etching time, and laser fluence are analyzed systematically. Based on the finite-difference time-domain simulation and the basic theories of silicon crystallization and chemical etching, the microstructure formation mechanism of the microstructures is investigated. By means of testing reflectivity of fabricated silicon samples, the regulation effect of the microstructure arrays fabricated under different processing conditions on the optical performance of the silicon surfaces is verified.

Gas metal arc welding (GMAW) and laser beam welding (LBW) were used to weld 2.8-mm-thick low-alloy high-strength Q345 steel butt joints. The out-of-plane deformation in the welded joints was measured experimentally. Meanwhile, based on the Abaqus software, a thermal-elastic-plastic property-based finite element method considering both material and geometrical nonlinearity is developed to calculate the welding temperature field, welding deformation and residual stress in the LBW and GMAW processes. In the simulation process of temperature field, an ellipsoid volumetric heat source model with uniform density flux was used to simulate the heat input induced by the GMAW process, while a conical heat source model with Gaussian distribution and a combined source model consisting of half ellipsoid volumetric heat source and conical heat source were used to model the heat input induced by the LBW process. The simulated and experimental results show that the out-of-plane deformation induced by the GMAW process is far larger than that produced by LBW. In addition, the area with high longitudinal tensile stress in the LBW joint is much smaller than that in the GMAW joint. Both longitudinal and transverse residual stress distributions in the LBW joint are notably different from those in the GMAW joint. In the simulation of LBW, welding deformation and residual stress distribution in the conical heat source model with Gaussian distribution and in the combined heat source model are similar, although the heat flux distribution in the thickness direction is different. The simulation results are not sensitive to the used heat source models.

HfO2 thin films are deposited on the crystal and the laser ceramic by electron beam evaporation technology respectively. Grazing incidence X-ray diffraction (GIXRD) and nano-scratch tester are used to investigate the crystalline structures and mechanical properties of the films. The experimental results show that crystalline orientations of the HfO2 films on single crystal substrate and polycrystalline ceramic substrate are all polycrystalline structures, and have (020) preferred growth, while the films on ceramic structures have stronger preferred orientation. The crystal-HfO2 films with a worse adhesive force have more diffraction peaks, while the ceramic-HfO2 films with a better adhesive force have less diffraction peaks. Compared the X-ray diffraction (XRD) results of two substrates and the films coated on them, the difference between single crystal structure of crystalline substrate and polycrystalline structure state of film is big, which contributes to a bigger residual stress, and a worse adhesive force between the crystal and the film. The weak binding force leads to small constraint effect of substrate on films. So, the films on the substrate of single crystal have more diffraction peaks. The difference between polycrystalline structure of ceramic substrate and the polycrystalline structure of films is small. This similar crystalline state makes the preferred orientation of the films more obvious and weakens the residual stress. So the adhesive force between the ceramic substrate and the HfO2 film is relatively better than that of crystal-HfO2 system. The strong binding force limits the growth of crystal orientation of HfO2 films, and the diffraction peaks of the films are less.

In order to compensate the shortness of the pseudorandom ranging system performance research, pseudorandom generator of 1 GHz and 2.5 GHz is realized by field-programmable gate array. Using 50 m as tested range, two channel photons arriving times are measured and calculated by fast Fourier transformation to obtain cross-correlation wave. Based on that, the cross-correlation mathematic model is set up and the parameter evaluation method is used, Markov technique is utilized to optimize parameter iteration. Finally, the cross-correlation wave is reconstructed and the range evaluation value is obtained. The ratio of pseudorandom pattern of 1 is 1% and the detected photon counting rate is 2×106 s-1. The research result manifests that in low background condition, with the increased integration time, the signal to noise ratio increases, which agrees with the theory result. In high background condition, with the increased integration time, the deviation amplifying speed with direct peak value is greater than the speed with the proposed method. Compared to the peak value method, the whole resolution is improved. With the integration time of 0.001 s, the resolution is increased from 7.3 cm to 4.62 cm. In addition, the resolution of 1 GHz is much lower than the resolution of 2.5 GHz. At the same bit speed, the effect of background noise on resolution is weakened.

The all-optical wavelength conversion (AOWC) based on orthogonal dual-pump for polarization multiplexing signal in semiconductor optical amplifier (SOA) is investigated. The influence of laser line-width on the performance of all-optical wavelength conversion is theoretically investigated, and the transmission performance of polarization-multiplexing-orthogonal-frequency-division-multiplexing (PM-OFDM) signal in the system is simulated. The simulation results show that when the optical signal-to-noise ratio is 10 dB, the polarization insensitive signal can be received as two light waves without crosstalk as a result of four-wave mixing effect for 10 Gbit/s PM-OFDM signal propagating in the SOA, and the bit error rates are 1.16×10-4 in x-direction and 1.06×10-5 in y-direction, respectively. Meanwhile, the simulation results indicate that laser line-width, number of sub-carriers of OFDM signal, channel power, and modulation format, obviously influence the AOWC conversion efficiency. The theoretical derivation is well consistent with the simulation result.

Models of the Y branch optical modulator′s half-wave voltage with wavelength and temperature are established. The half-wave voltage increases with wavelength increasing and temperature decreasing. A half-wave voltage automatic test system based on Sagnac interferometer modulated with four steps waveform is designed and it can measure half-wave voltage continuously on line. Relative measuring accuracy reaches 8.5×10-6. By changing working wavelength and temperature, the actual measurement of Y branch′s half-wave voltages in 1310 nm and 1550 nm band is realized by using a tunable laser as the light source. The results show that the half-wave voltage is linear with the temperature and the wavelength in the two test bands. The wavelength correlation coefficients are 3.61×10-3 V/nm and 9.54×10-3 V/nm, respectively, and the temperature correlation coefficients are -1.76×10-3 V/℃ and -2.00×10-3 V/℃, respectively. The experimental results verify the correctness of theoretical model.

A method of beam alignment detection is proposed for the acquisition, pointing and tracking system in atmospheric laser communication based on extension of beam propagating in atmospheric turbulence, back projection imaging principle and Cassegrain telescoping system. The method is analyzed theoretically and verified experimentally. Based on the shape and centroid of the beam spot, the transverse deviation and the axial deflection of the beams are calculated comprehensively. As a result, the problems with the general method are solved, one of which is that the beam spot fluctuation is caused by transverse displacement or axial displacement, and another is that the transverse displacement can not be calculated by the general method. The experimental results show that the method can detect effectively the beam alignment state, and realize finally resolution ratio with 2.2 μrad in axial deflection and 0.25 mm in transverse deviation.

A kind of three modes tapered fiber mode multiplexer/demultiplexer is developed based on mode coupling theory for realizing modes multiplexing and demultiplexing between a few-mode fiber (FMFs) and three single-mode fibers (SMF). It′s significant to improve the transmission capacity of optical transmission system. When the fundamental mode lights are input into three SMFs with the diameters of 10 \mm, 8.4 \mm and 7.2 \mm, the lights will couple in the tapered area and convert into higher order modes. The similarity between optical field at output port and standard mode field is further calculated by overlap integral. As a consequence, the fundamental modes can be converted into LP01 mode and two degenerate LP11 modes at the output port, i.e. realization of modes multiplexing. When the LP01 mode and two LP11 modes are separately input into the FMF, the output light power of corresponding port is the maximum, i.e. realization of modes demultiplexing. Finally, the multiplexer's tolerance is calculated with different parameters. Calculation results show that 3.88 cm and 3.8931 cm are the tapered structure′s length of the multiplexer for best effect of mode converting and demultiplexing.

In order to deal with the nonlinearity, non-stationarity and intermittence of intrusive vibration signal in the optical fiber perimeter system, a method combining time and frequency domains characteristics is proposed to distinguish and locate intrusive vibration signal. The minimum frame length is determined by computing embedded dimension to better reserve the dynamic characteristic of the time series signal. The two stages judging and recognizing methods of intrusive vibration signal are proposed. Firstly, short-term energy and zero-crossing measurements are used for determining whether the vibration signal is generated, and then the intrusive signal is recognized according to the characteristic of the energy distribution of each layer′s wavelet coefficients. This method effectively reduces the efficiencies of recognizing error and loss for optical fiber perimeter system. In order to improve the accuracy of locating intrusive signal, Bayesian adaptive threshold estimation in wavelet domain is applied to reduce the noise of the signal, and the reconstruction signal is finally transformed to frequency domain to find the intrusive point. The experiments result shows that the proposed algorithm is effective.