The output laser beam from a pulsed diodes side pumped single longitudinal mode laser is amplified by single passing through two pre-amplifiers and two main amplifiers, by which single longitudinal mode laser with pulse energy of 36.5 mJ, pulse width of 24 ns at 400 Hz repetition rate is obtained, with beam quality factor being M2x=1.78, M2y=2.13. After applying a large aperture tapered fused silica fiber as phase conjugating mirror, which shows more than 50% stimulated Brillouin scattering reflectivity, the input laser pulse width can be reduced to 6 ns and peak power increased significantly. The returned laser backtracks because of the phase conjugation characteristics and passes through two main amplifiers again, not only compensating the phase aberration of the laser amplifier crystal but also distilling more energy and amplifying the laser effectively. Finally, single longitudinal mode laser with pulse energy of 101.25 mJ, pulse width of 6 ns at 400 Hz repetition rate is achieved, corresponding to a peak power of 16.8 MW. Its beam quality factor is M2x=1.74, M2y=1.93. Obviously, the beam quality is improved instead of deteriorated after applying the phase conjugation mirror. However, the beam quality deteriorates to M2≈4 by using plane high reflective (HR) mirror, which validates the effectiveness of the phase conjugation mirror.

With CrYAG as the saturable absorber, passively Q-switched microchip laser is obtained by laser diode (LD) end-pumped NdGdVO4 crystal. The NdGdVO4 crystal is a-cut with the size of 3 mm×3 mm×1 mm, and the atom fraction of Nd3+ ions is 0.5%. The pump end face of NdGdVO4 crystal is anti-reflective coated for 808 nm and high-reflective coated for 1.06 μm, which serves as the input mirror of laser cavity. The size of Cr4+YAG is 9.5 mm×1.1 mm, the initial transmission is 77%, and its outer end face is partially reflective coated for 1.06 μm (T=15%), which serves as the output mirror of laser cavity. The total length of laser cavity is 5~6 mm. The pump threshold of the laser is 4.62 W. When the pump power reaches 13.86 W, a maximum average output power of 0.98 W is obtained, corresponding to optical conversion efficiency of 7%, and slope efficiency of 9.5%. The highest repetition rate, largest pulse energy and shortest pulse width are measured to be 23.4 kHz, 44.6 μJ, and 2.9 ns respectively, corresponding to peak power of 15.4 kW.

A long-cavity mode-locked Yb-doped fiber laser is reported. The laser consists of only single mode fiber (SMF) components except the nonlinear polarization rotation (NPR) parts which are utilized to achieve self-starting mode-locking, and operates at the all-normal-dispersion (ANDi) region since there are no dispersion compensation elements inside the cavity. Dissipative solitons were obtained via the introduction of only a narrow-bandwidth spectral filter into the pulse shaping process. Two long-cavity conditions were implemented by lengthening the SMF, where pulses with 6.66 MHz repetition rate and 12 nJ single pulse energy and those with 5.05 MHz repetition rate and 20 nJ single pulse energy were obtained respectively, both showing a noise suppression ratio of 75 dB on the radio frequency (RF) spectra. Compared to the former ANDi mode-locked fiber lasers, this laser operates at a lower repetition rate with higher single pulse energies and better stability, which should be useful in the construction of a simplified fiber chirped pulse amplification (CPA) system.

A novel fiber laser with two output ports is designed to realize high power and high quality fiber laser coherent beam combining output. The two fiber lasers associated with each other are connected by a fiber bragg grating (FBG) (R=0.85 at 1085 nm). Based on a corner-cube, the energy is mutual injected with each other. And by a polarizer, coherent beam combining output is realized. Through experiments, steady interference stripes with high contrast ratio (about 0.92) are observed and 8.6 W output power is achieved which indicate that the power combining efficiency is approximately 90%. This combining method can be scaled to a large number of lasers. Successful attempts of phase-locked show that this method may provide a promising way in the field of fiber laser coherent combining.

Using a pump laser coupling the hyperfine transition of D2 line in a cesium cell, the optical pumping spectra are demonstrated by detecting the absorption of the co-propagational probe beam with the identified frequency. The peaks or dips are related to the optical pumping where the pump laser and probe laser coupling different hyperfine transitions with a frequency difference. Employing a locking loop and the optical pumping spectrum, a scheme for frequency offset locking of two diode lasers with the reference frequency based on the hyperfine splitting is presented. The frequency offsets can be the intervals between the hyperfine levels of the 62P3/2 excited states or the 62S1/2 ground states from hundreds megahertz to 9 GHz.

In order to evaluate the performance of high-accuracy ring laser gyroscope (RLG) more precisely, a new method of analyzing the collecting data of high-accuracy dithered RLG using dynamic Allan variance (DAVAR) is proposed. The theory of DAVAR is introduced in detail and applied in the study of stochastic errors in dithered RLG. The result shows that the analysis method of DAVAR can not only refine and identify RLG′s stochastic error terms, while confirming the ratio of each error term to the overall error, but also analyze the stability of RLG′s stochastic errors. Compared with Allan variance, DAVAR can represent the characterization of stochastic errors in RLG more completely.

Temperature is one of the main factors that affect the accuracy of fiber optic gyroscope (FOG), and static temperature characters of FOG are studied through theoretical analysis and experiments. Temperature modeling of zero bias based on typical wavelet network is proposed, which is compared to polynomial fitting, and the accuracy is improved greatly. Further, modification of typical wavelet network is studied, including initialization of parameters and a new momentum adaptive gradient descent method for parameter updating. Through experiments, it is proved that the modified wavelet network can further improve convergence speed and fitting accuracy, which is better in describing the temperature characters of FOG. In order to verify the university of wavelet network algorithms, temperature modeling and accuracy analysis based on testing data of different inertial instruments are carried out, and results indicate that the proposed methods can be used for static temperature modeling of inertial instruments sensitive to temperature.

In order to test and evaluate the performance of the composition of laser radar systems, modules and components, and the co-working efficiency of modules and components, a heterodyne laser radar imaging of electronic tracking devices loop simulation platform is established. The electronic tracking device is composed of simulative target scene library module, signal simulator module, intermediate frequency amplifier, data acquisition logistic control module, imaging and data analyze module. The main features of electronic tracking devices of heterodyne imaging ladar system include: testing the various components of the laser radar system performance indices; efficiency analysis of heterodyne laser radar systems, and providing advices to improve the efficiency of proposed system. The results show that electronic tracking device loop simulation platform can reproduce the working process of imaging laser radar, analyze the co-working efficiency of the various modules; assess the performance of laser radar devices, and provide practical analysis results based on specific application requirements which provides a reliable analytical tods to eveluate the performance of the modules and device of laser radar system.

The principle of Fresnel quasi-phase matching technology used to wide tuning terahertz wave detection is analyzed. We discuss the resonant and non-resonant Fresnel phase matching condition and the effect of Goos Hanchen delay to Fresnel phase matching. The effective nonlinear efficiency during sum frequency generation, the accept angle width of internal total reflective angle inside of the crystal and the absorption of GaAs crystal for three interaction waves also are investigated in more detail. Finally, in order to achieve wide tunable terahertz detection, we present an experiment scheme and experimental setup with Fresnel phase matching by angle tuning.

Research on the polarization patterns is of great importance to reduce polarization disturbance errors, improve measurement accuracy and optimize system design. Scattering angle (SA) and degree of polarization (DOP) of skylight are measured by vector polarization measurement system under different sky conditions. The experiments demonstrate that all relationship curves between SA and DOP can be described as parabolas regardless of sky conditions. DOP changes with sizes of atmospheric scattering particles. When the sizes of the particles increase, DOP decreases and SA drifts. A single light scattering polarization model based on Mie theory is proposed to analyze the experiments. Simulation results are in good agreement with the experimental results.

ArF immersion interference lithography system can produce high resolution periodic pattern as a powerful tool for the study of hyper numerical aperture ArF immersion lithography. The coherence requirements on exposure source can be alleviated and high resolution imaging can be realized by the use of achromatic ArF immersion interference lithography system. A novel double-fluid-layer achromatic immersion interference lithography system is developed. The system′s achromatic principles and imaging mechanism are studied by using diffraction theory and principle of geometrical optics. The imaging performance is analyzed and the advantages are summarized compared with traditional single-fluid-layer system. The results indicate that this system enables a large depth of focus (DOF) and width of resolved imaging field which are independent of the bandwidth of laser. The system′s performance is insensitive to environment，so that the imaging stability is improved. The system employs simple symmetry structure to overcome the problem of poor coherence properties for ArF excimer lasers, and ensures stability of imaging performance in interference lithography.

We report the design and analysis of a rod-type photonic crystal fiber with Er-Yb co-doped for the high power 1.5-μm band amplifier. The fiber structure is designed to be the 120-μm extreme large core diameter, 300-μm inner cladding diameter, and 1.5-mm outer cladding diameter that ensure the single mode output during high power amplification. Both the continuous wave (CW) and pulsed amplification characteristics are analyzed based on the exact modeling and simulation under the designed geometry. The 4-mJ pulse energy and 400-kW peak power are obtained in theory, so the 1.5-μm band amplifier that achieves milojoule level pulse energy meanwhile keeping single mode is firstly designed.

In backward pumped fiber Raman amplifier (FRA), cascaded stimulated Brillouin scattering (SBS) is observed. In the experiment, a three-wavelength Raman fiber laser whose maximum output is 3000 mW was used as Raman pump, and 25 km single mode fiber (SMF) is used as both Brillouin and Raman gain media, an external cavity laser (ECL) whose power is 3 dBm is used as Brillouin pump. The backward cascaded SBS of 89 orders and forward cascaded SBS of 156 orders are observed. The frequency shifts of the backward and forward SBS between adjacent order are 22 GHz and 11 GHz respectively, namely their wavelength intervals are 0.166 nm and 0.085 nm respectively. The spectral bandwidths of the backward and forward SBS are 14.812 nm and 13.580 nm respectively. The flatness of backward spectral lines is better than that of forward. Compared with single-wavelength FRA, the three-wavelength Raman fiber laser used as Raman pump is easier to generate cascaded SBS.

It is an effective approach to obtain high power laser beam output by spectral beam combining. The model of spectral beam combining using phase-shifted reflective volume Bragg gratings is established, and the diffraction characteristic is analyzed by transfer matrix method. The diffraction characteristics of reflective volume Bragg grating and phase-shifted reflective volume Bragg grating are compared. Taking the spectral beam combining of two beams by phased-shifted reflective volume Bragg gratings as an example, the effects of Gaussian monochromatic beam′s spectrum width on combining efficiency are analysed. The results show that the side-lobe of diffraction efficiency can be eliminated and the channel number can be increased by phase-shifted reflective volume Bragg gratings, and the combining efficiency can be more than 97% when the spectral beam combines of two beams with a spectral separation of 30 pm.

The beam divergence angle of lidar returned signals which is directly influenced by propagation properties of Gaussian beam affects the properties of Fabry-Perot etalon (FPE). In this paper, the three-dimensional interference rings of Gaussian beam are investigated and manifested to some trend. Gaussian beam with small divergence may be propitious to obtain higher energy of FPE zero-order or first-order interference ring. For a lidar system, the effects of beam quality factor M2 and expander coefficient on returned signals are discussed. The numerical results indicate that FPE transmission is changed clearly with the height in the near field and leveled off in the far field, and at a given height, it decreases with the enhance of the value of M2 and increases with the expander. It is supposed that laser with lower beam quality factor M2 and the transmitter with higher expander coefficient are chosen and developed to improve FPE transmission and in the near field decrease the measurement error in lidar using an FPE as spectroscope, which is expected to supply numerical foundation for the lidar system.

The discrete equivalent spherical scatterers aggregate model is developed to investigate the photon′s transport characteristics in highly scattering media. Based on Monte-Carlo method, we propose a new ellipse algorithm to determine the scattering point distributions. According to Mie theory and the radiative transfer equation, combined with the obtained scattering point distributions, path-length dependent scattering order distribution and scattering intensity in different scattering orders with different particle sizes are numerically calculated. The results show that the photon's transport in highly media is essentially a multiple scattering process. The numerical algorithm can be used effectively to analyze dynamic scattering properties of scattered light from single-scattering to multiple-scattering regime.

Laser holds a higher output power when working in high-order-mode than working in fundamental mode, coherent combining of high-order mode lasers is a promising way for higher output power. For the inside 0-π phase difference of the wave front, the beam quantity of the high-order mode is not as good as that of the fundamental mode, therefore the brightness of the coherent combining is limited. Coherent combining of high-order mode lasers with wave front compensation is proposed. Inside 0-π phase differences of high-order mode laser are compensated using beam cleanup technology firstly, and then coherent combining is implementing with the wave front compensated lasers. Results of simulation and experiment of inside wave front compensation using optimal algorithm based beam cleanup technology are given. Numerical simulation of coherent combining of high-order Hermite-Guass mode lasers with and without inside wave front compensation is done. Result shows that the power in bucket of one time diffraction limitation of the combining beams with inside wave front compensation is at least 10 times or even more than 1000 times of that value of the combining beams without inside wave front compensation.

In order to get a well reliability parameter, a new daughter wavelet for phase reconstruction of optical carrier-fringe pattern was proposed. The phase of the fringe pattern can be reconstructed from the ridge of the wavelet transform by employing this new daughter wavelet. The modulus of the new wavelet transform coefficients include both the modulation information and fringe density information, which can be treated as a well reliability parameter representing the phase reconstruction. Rigorous mathematical demonstration of this method was given. Theoretical demonstration shows that the modulus can be treated as a reliability parameter of the phase reconstruction. Computer simulation and experiment verify the feasibility of the proposed method.

A method of iterative phase retrieval of complex optical field was studied. This method uses a series of diffraction patterns to solve for both amplitude and phase of the object field by using hybrid input-output algorithm. The method requires a phase plate that need transversely rotated to one or more positions. The method needs no a priori knowledge about the object field. The convergence rate was rapid. The simulation results showed this algorithm could fast retrieve a complex optical field in high accuracy. The retrieval accuracy of complex optical field was greatly enhanced. The method does not suffer any stagnation or ambiguity problem, and it has high stability towards additive noise.

For the prediction and control of mechanical property of laser tailor-welded blanks, combining it′s microstructure-property rule microstructual qualitative analysis of joint microstructure of 0.8~1.5 mm St12 steel and galvanized steel sheet is constructed. The predominant influencing factor is martensite content and the optimal control process is obtained. On the basis of the research, prediction models of mechanical properties of tailor-welded blanks trained with welding processes and sheet thickness as input variables are put forward based on partial least-squares regression. According to the testing results, the forecasting accuracy of tensile strength and elongation is above 91%. Therefore, the proposed models are reasonable and applicable to examine deep-drawing steel for demonstrative purposes.

Through laser irradiation on the nonmetal material adulterated with metal complex, the irradiated area could be chemically plated on copper and nickel after the sensitizing treatment, so the electricity conducting channel is formed and the quality of metal conductor appears. The non-irradiated area still remains the original nonmetal quality. The NdYAG laser beam, after going through dynamic focusing system, was reflected by galvanometers and then focused by f-theta lens on the surface of metal complex. Through the 1, 3, 10 and 20 kHz laser modulation comparative experiments, the copper-plate thickness exceeded 11 μm, and the nickel-plate thickness exceeded 2 μm. In this way, the quality of metal and nonmetal coexist on the surface of the same material.

In order to solve the problems of poor utilization of powder and nozzle blocked by rebounded powder by improving the powder gathering characteristic and increasing the powder focal length, the influence of the cone ring gap and the cone angle of the coaxial nozzle on the powder convergence is studied. The results show that the powder gathering characteristic is better when the cone ring gap is smaller and the nozzle focal length is longer when the cone angle is bigger with other process parameters invariable. The conclusion is valuable for the coaxial nozzle designing and performance improving．The simulation results are verified by experiments. The effects of process parameters on the height and width of the single-trace are studied. The thin wall parts are manufactured by optimizing the process parameters.

Low-power pulsed laser weld the 0.2 mm thickness of TiNi shape memory alloy /stainless steel welding dissimilar materials, the crack of the weld joint morphology and elemental distribution are studied, the reasons for crack formation are analyzed, and crack control measures are proposed. The results show that, TiNi alloy /stainless steel butt welding of dissimilar materials has large crack sensitivity. Only by adjusting the laser welding process parameters can′t avoid cracks. The main reason of cracks is the weld elements Fe, and Ti elements in the stainless steel base metal in the formation of brittle intermetallic compounds, resulting in welds brittle, cracking under sress in the weld. TiNi alloy and stainless steel by pure Ni wire between the pre-filling materials can be diluted as a weld in Fe, Ti element content, on the formation of brittle cracks organizations barrier to avoid cracks. The tensile strength of the welded joint with pure Ni wire as filler material is 580 MPa.

A CO2 laser-metal active gas welding (MAG) hybrid welding technology is used to weld the high strength steel of 7.0 mm thick. The effect of the arc energy, laser energy and distance between laser and MAG arc on the droplet transfer character, process stable and weld geometry are investigated. Results indicate that inputting of laser decreases the resistance of plasma channels of locate laser keyhole. In this way, arc is attracted and constricted, which stabilizes the arc cathode spots. The arc energy determines the mode of droplet transfer, and the laser energy affects the frequency of droplet transfer. When the laser energy is less than 4 kW, the transfer mode is short circuit transfer, or globular transfer or transfer mode between short circuit and globular. When laser energy exceeds 4.68 kW, the transfer mode is spray transfer. The droplet transfer mode is very important for achieving a stable and repeatable process: projected/spray transfer should be preferred to short/globular transfer. Distance between laser and arc sources must be at a 2～4 mm in order to avoid turbulence in the weld pool and disturbance of the keyhole formation and to achieve synergy between the processes.

Temperature field is the key factor affecting manufacturing quality in laser welding. For a nail head shape of weld cross-section in the partial penetration laser lap welding, a hybrid heat source model combined with a circular disk source with a Gaussian distribution on the top surface and a linear increasing column source along the depth of weld is proposed, by analyzing the absorption mechanism of the welding material. The effect of the contact thermal resistance between the upper sheet and bottom sheet is considered in the model. The comparison of experimental and simulated results shows that the geometry of weld cross sections is in good agreement. In addition, the relationship is investigated between the thermal cycles and microstructure as well as the micro-hardness. Besides the heating and cooling rate, the peak temperature also has an important influence on mechanical properties of the weld. The performance of the weld is similar when thermal cycles are consistent. The numerical model can help to study the temperature field of the laser deep penetration welding and choose appropriate laser parameters.

The gear teeth were shocked by high intensity laser shock waves. The transmission error of sample after shocking were tested. The residual stress and microstructure phase transformation were measured by X-ray diffraction (XRD) technique, and the fatigue damage experiment was done to test the fatigue resistant property. The results indicated the tooth profile error after laser processing still kept lower levels and the laser shock had no effecting on transmission accuracy. That the residual compressive stress of gear enhanced greatly after laser processing, and the residual stress of gear was increased by 107% in the normal direction and 85% in the tangent direction. And that the volume fraction of austenite phase decreased after shocked process. The area of defect spot on teeth flank un-shocked was 2.1 times of that shocked, which indicates that the laser shock processing can extend the gear fatigue life.

This paper mainly simulates the fatigue life of specimens treated by laser shock processing(LSP), and studies the effects of LSP on fatigue life. Two groups of 7050-T7451 aluminum alloy specimens pull-pull experiment are performed. One group of specimens is untreated, while another is treated. The results of experiment indicate that under the stress of 101.3 MPa, two groups of average fatigue life are 26024 and 108336 cycles respectively, raised to 416.3%. The results of numerical simulation indicate that two groups of average fatigue life are 25400 and 102870 cycles respectively, raised to 405%. The result of experiment agrees with that of numerical simulation quit well. The study shows that the residual stress is generated by LSP on the surface of specimen, thus the fatigue life increases obviously.

When thin film resistors are directly written by laser, auxiliary material can cause impurity pollution. Thickness and roughness of the film cannot be reduced easily. To reduce impurity pollution, film thickness, roughness and improve the performance of microcircuit directly written by laser, the research of metallic film resistors directly written by transmissive laser is carried out. The characteristic relation of laser spot energy distribution and film line width is analyzed. The conductive film of line width less than spot dimension can be obtained. The experimental results show that the increase of scanning speed or the decrease of defocusing amount can lead to the decline of overlapping cofficient of spot and line width decline. Meanwhile, excessively high laser power can lead to abliative damage to thin film; excessively low laser power cannot melt the powder on the glass substrate. With different powders in the process, it is known that Zn powder can form a uniform thin film while Al /Zn powder cannot form a uniform thin film.

Properties of reflection phase in one-dimensional photonic crystals constituted by alternating layers of negative- and positive-index materials were investigated by transfer matrix method. The results show that the reflection phase difference between TE and TM waves in the zero- gap increases with the incident angle, and remains almost unchanged in a broad frequency band (the relative spectral bandwidth Δω/ω>19%). Based on properties, phase compensators with some advantages such as working effectively in a broad frequency range and continuously tunable compensatory phase etc. The influences of the changes of the scale factor, the thicknesses ratio, the refractive index and the period number on the phase compensators were also investigated.

In order to improve the shrinkage properties and the diffraction efficiency of photopolymer material, a photopolymer doped with TiO2 nanoparticle is fabricated. Aparting from using methylene blue trihydrate as photosensitizer, this material is exposed at a wavelength of 647 nm. The experimental results show that the incorporation of TiO2 nanoparticle leads to a significant decrease of the shrinkage and the improvement of diffraction efficiency: the shrinkage is decreased from 3.2% to 2.4% and the diffraction efficiency is increased from 68% to 70.8% in the photopolymer contained with 36 nm TiO2 nanoparticle. However, when the photopolymer doped TiO2 with the size of 10 nm, there are only 0.8% of the shrinkage and the diffraction efficiency amount to 78.1%, which indicates that doped TiO2 nanoparticle with smaller size has more advantages in improving the holographic characteristics than that with the bigger size.

Er3+CaMoO4 crystal is grown by CZ (Czochralski) method from a flux of Na2MoO4. The optical homogeneity, absorption and emission spectra of Er3+CaMoO4 crystal are investigated. The absorption cross-section is 0.6×10-20 cm2 at 981 nm, and the emission cross-section is 0.91×10-20 cm2 at 1538 nm. 532 nm and 553 nm green up-conversion emission and 670 nm red up-conversion emission are observed at the 980 nm excitation. Er3+CaMoO4 crystal may be a potential 1.5 μm laser gain material and up-conversion luminescent material.

Effect of fabrication error on the absolute bandgap of two-dimensional 12-fold photonic quasicrystals (PQCs) is numerically investigated by using the finite difference time domain (FDTD) method. Results indicate that, i) the effect of the size error is bigger than that of the position error; ii) the absolute bandgap is more sensitive to the bandgap of TE mode than its TM mode; iii)the bigger the fabrication error, the narrower the absolute bandgap of 12-fold PQCs is. In the current fabrication conditions, the effect of the fabrication error can be almost ignored for the 12-fold PQCs which lattice constant is bigger than 500 nm. However, if lattice constant of 12 PQCs is less than 200 nm, its absolute bandgap even disappear completely and cannot be used as absolute bandgap materials.

The Si-OH group on particle in the silica colloidal suspension is modified to Si-O-Si(CH3)3 non-polar hydrophobic group using hexa-methyl-disilazane (HMDS). The organic modified silica antireflective (AR) coatings that coated on potassium di-hydrogen phosphate (KDP) crystal and silica glass substrates by spinning method, has good moisture-resistant property. The transmission of coated substrate is above 99%. The optical roughness root-mean-square value is 0.94 nm and the water contact angle of coating is about 140°. The HMDS modifies SiO2 sol in non-polar solvents which need no treatment can be used in KDP crystals. In the condition of the room temperature and high humidity, the transmittance decrease of KDP crystals with the moisture-resistant(MR) AR coating is equal to the MR/AR two-layer coating. The reflection losses at three wavelengths from two surface of the tripling crystal can be minimized by spin coating.

In order to investigate how the laser pulsed energy influences the microstructure, topography and infrared property of the diamond-like carbon (DLC) film in pulsed laser deposition (PLD), the DLC films are deposited on Si substrates by 120 mJ and 150 mJ KrF excimer pulsed laser in the chamber with 10-1 Pa N2 respectively while other experimental conditions are kept unchangeable. The microstructure and composition of the DLC films are detected by the visible Raman spectroscopy and the X-ray photoelectron spectroscopy (XPS). The topography of the DLC films is detected by the atomic force microscopy (AFM). The infrared transmission spectra of the DLC films and Si substrate are detected by the Fourier transform infrared (FTIR) spectroscopy. The experimental result demonstrates that when the laser pulsed energy is enhanced from 120 mJ to 150 mJ, the fraction of N atoms and sp3 bonded carbon atoms increase in the DLC films; the fraction of CO, C=O and O decrease in the films; the size and number of graphitic crystallites decrease in the films; the roughness of the DLC films decreases remarkably. At the same time, the infrared antireflection rate of the DLC films decreases and the infrared antireflection range of the DLC films becomes narrow when pulsed energy is enhanced.

Using pulse laser deposition(PLD) technology, different component x(x=0.03, 0.30, 0.53, 0.80, 0.97) and doped M (atom fraction 3％, M=Na, Sr, Bi, Ce) PZT ferroelectric thin films were successfully prepared on SrTiO3 monocrystalline substrates. The test results of XRD showed that all the thin films were single-phase and highly oriented. But the changes of components and doped elements could affect the crystallinity of thin films in various degrees. Furthermore, laser-induced thermoelectric voltage (LITV) signals were measured clearly in PZT ferroelectric thin films grown on 10° vicinal-cut SrTiO3 monocrystalline substrates. Through researching the influence of components and doped elements on the size of the signal, the maximun LITV signal was found under UV light irradiation of 0.16 J/cm2. For different component PZT ferroelectric thin films, the maximum peak voltage was 60 mV if and only if the component was x=0.03. For different doped Pb(Zr0.53 Ti0.47)O3 ferroelectric thin films, the maximun peak voltage was 61 mV if and only if the doped element was Na, which increased nearly 50％ compared with the maximum peak voltage of no doped Pb(Zr0.53Ti0.47)O3 ferroelectric thin films.

90° or 270° phase retarder is introduced to change the polarization of light from linear to circular. The combination of metal and dielectric is used in this design. The silver thin film with excellent optical quality is used as the metal layer. 270°±1°phase shift between p- and s-polarization component as well as broadband high reflection mutilayer coatings is designed by optimizing dielectric layers in the wavelength range 1000 ~ 1100 nm at 45°. The errors produced in manufacturing and using process are also analysed. Under the manufacturing precision nowadays, more attention ought to be paid to ensure that the phase shift error δ<7.2°and reflectivity over 99.9% are achieved in the 1000~1100 nm spectral region for use at 45°。

H4 thin films (which is a compound of titanium oxide and lanthanum oxide，chemical formula is LaTiO3) have been prepared by the ion assisted electron beam depositing technology. The effect of the substrate temperature and oxygen gas stress influencing on the optical characteristics of H4 thin films have been investigated.It is found that the refractive index n dramatically increases with the substrate temperature building-up，the value of n808 nm is 2.14 at 100 ℃ of the substrate temperature; the extinction coefficient k has changed little with the oxygen pressure reduction, the value of k400 nm is 2×10-4 at 2.67×10-2 Pa of the oxygen pressure. The optimized parameters are applied to the preparation of high reflection mirror of 808 nm lasers, compared with the lasers using titanium oxide as HR films, have obtained a little better laser output characteristics. Therefore, it is a kind of new method of completely feasible that H4 thin films are used as high reflectivity facet coatings of laser diodes.

A double metal-cladding structure which can achieve high contrast approaching unity and high-reflectivity is proposed. This double metal-cladding structure with sub-millimeter scale can accommodate a series of resonant modes. Using the Fresnel reflection formula of the four layers, the relationship between coupling efficiency and thickness of upper metal layer is analyzed and the optimum value of thickness of upper metal layer can also be obtained. It is also found that such optimum value is closely related to the intrinsic and radiative damping. The interference fringes of reflected beam and coupling efficiency on double metal-cladding structure were observed experimentally. Experimental results showed that high contrast approaching unity can be observed from interference fringes and coupling efficiency can reach nearly 100% on double metal-cladding structure. It is useful for the investigation of high contrast high reflectivity films and the sensors based on double metal-cladding structure.

Temperature can affect the accuracy of tunable diode laser absorption spectroscopy (TDLAS) system. A system that can monitor temperature and methane in the same time is introduced. Two parameters are regulated by each other in order to increase accuracy not only in gas monitoring but also in temperature monitoring. The system uses 1653.7 nm as center wavelength of distributed feedback (DFB) laser, by taking the sawtooth signal to modulate the light frequency of the DFB laser. The scanning can obtain wavelength of fiber Bragg grating (FBG) and methane absorbing strength simultaneously. By wavelength scanning, the system can calculate temperature value and methane concentration. Temperature monitoring can regulate the deviation of methane detection caused by temperature changing, meanwhile, methane′s absorption line can locate the wavelength of FBG in order to make temperature monitoring more accurate. This temperature and methane monitoring system is more stable and reliable. It possesses wide applications in coal mine and methane extraction.

The loss of a high-finesse Fabry-Perot cavity was measured using frequency scan and cavity ring-down techniques. Different transverse modes were examined by sweeping the frequency of the laser across the corresponding resonances. In addition, cavity ring-down technique was used to measure the loss of the Fabry-Perot cavity and a simple model was employed to remove the influence of finite response time of the optical switch and the detection circuit.

To improve measurement accuracy, the coherent noise of interferometric system should be reduced effectively. A theoretical model of signal-to-noise ratio (SNR) in interferometric system with ring source is proposed by means of statistical optics. The relations among parameters of ring source, coherent area of optical field and SNR of system are analyzed. By maximizing the objective function of SNR, optimization is employed to ring source and then the optimized parameters are controlled by computer generated holography. According to the theoretical model, the SNR of a given system depends on area of ring source, which is inverse proportion to coherent area of optical field. The parameters of ring source can be controlled by spectral width of conical wavefront, distance between rotating diffuser and focal plane, and spatial carrier frequency of computer generated hologram (CGH). Experimental comparison between point source and ring source shows that the average SNR of system increases from 5.2 dB to 8.6 dB by utilizing ring source.

To measure output parameters of a laser amplifier, with the repetition rate of several tens of minutes one pulse and single pulse energy of 5 J, a special designed measurement system is put forward for high-energy, large cross-section, low frequency laser system. This system uses the principle of Fourier image transfer to realize real-time and clear monitoring to the output characteristics of the laser beam. And this system can not noly detect laser wavelength, spectral width, spot size, single pulse energy, repetition rate, and polarization direction of the laser beam simultaneously, but also statistically calculate the beam modulation, divergence angle, directivity, beam quality factor, energy stability, average power, power stability, pulse width, pulse stability, and degree of polarization.

During the dynamic process of sports mechanics, ballistics, etc., the three-dimensional (3D) surface distribution of the transient moving target is closely related to its structure, deformation, stress, erosion, etc. But the non-contact, multi-view and fast measurement for it is still not realized. In response to this problem, a hexagonal configuration of the measurement system which is based on structured light technology is designed and structured. By using three projectors-camera subsystems, the sinusoidal grating is projected onto the surface of the transient moving target instantaneously and simultaneously. The deformation patterns catched by high-speed CCD are then processed by the wavelet transform. Finally, the 3D surface distributions can be obtained. The experimental results indicate that the system is effective to achieve the non-contact, multi-view and fast measurement for the 3D surface distribution of the transient moving target.

The asymmetrically clipped optical orthogonal frequency division multiplexing (OFDM) modulation technique is investigated. It is a new technique for using OFDM in optical system. Asymmetrically clipped OFDM is derived from a bipolar OFDM waveform by setting the negative values to zero. The asymmetrically clipped OFDM technique is applied to atmospheric communication system in this paper, and the performance of atmospheric communication system using asymmetrically clipped OFDM technique through atmospheric turbulence channels is analyzed. On this basis, the low density parity check (LDPC) codes and belief propagation (BP) iterative decoding algorithm are applied into the atmospheric communication system, and it is simulated in the atmospheric turbulence channel with combination of asymmetrically clipped optical OFDM intensity modulation. The simulation results show that LDPC codes have excellent error correction capabilities and access to a larger coding gain,and the above scheme can satisfy the need of atmospheric laser communication system.

Light trail (LT), a novel optical network, promises small switching granularity and good bandwidth utilization. There is little research on LT node until now, especially on its performance simulation and experimental demonstration. A novel LT node architecture that supports 4 wavelengths for duplex communication is introduced. And a simulation platform is built to test the performance of the LT node, including insertion loss as well as bit error rate (BER) and eye diagram in LT networking. The simulation shows that its insertion loss is about 16 dB, which means that it should work at intervals of less than 20 km if without help of semiconductor optical amplifiers (SOAs). Further simulation shows that a single SOA can be employed to improve eye diagram, increase work distance of 5 km, and decline BER about 105 at 15 km distance or 102 at 20 km. Finally the conclusion can be drawn that the reported LT node has the advantages of simple structure, low cost and low loss, and is suitable for application in metropolitan and access optical networks.