Routine estimates of the turbulent fluxes of momentum, heat and trace gas flux have important significance in many meteorological, hydrological and other related studies. Combining with large-aperture scintillometer (LAS), temperature-gradient measurement system and tunable diode laser absorption spectroscopy (TDLAS), sensible heat flux and carbon dioxide (CO2) flux can be derived by iterative over a long optical path. This method has realized fast detection of heat flux in large natural areas and CO2 flux. The measured fluxes of sensible heat and carbon dioxide are compared with the results measured by a eddy-covariance (EC) instrument. The results show correlation coefficient of sensible heat flux is 0.91 and that of CO2 flux is 0.74, respectively. Experimental results show that optical measurement overcomes those limitations of traditional method, so this method is an inspiring method in heat and gas flux monitoring.

Study on the bore-sight error (BSE) induced by high-speed flow-field is important for the hit precision of a hit-to-kill interceptor with infrared ray (IR) terminal homing system. Based on the flow-field by solving the thermal-chemistry non-equilibrium Navier-Stokes (N-S) equation, the ray-tracing method is used to calculate the BSE of ray transmission through the flow field outside the optical window. The impact factors of BSE are considered, including incidence angle, Mach number, attack angle, altitude and the IR seeker shape.

A prototype of Raman lidar first developed by Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, for monitoring lower-tropospheric CO2 profiles in China is introduced. An ultraviolet laser with wavelength at 355 nm is selected as light source and a photon-counting card is used to acquire atmospheric N2 and CO2 Raman backscattering signals. Methods of the Raman lidar system calibration are analyzed in detail and Li7500 H2O/CO2 analyzer is employed to contrast with our lidar. The results show good agreement in the trend of changing and verifying high detection sensitivity and accuracy of the lidar system. By linear fitting, the calibration error is smaller than 0.2% and 1.5% in horizontal and vertical direction, respectively. Then profiles of atmospheric CO2 can be calculated in accordance with calibration equations and typical measurement results in the western suburbs of Hefei are given. The detection range can reach 2.0 km and 2.5 km in horizontal and vertical direction, respectively.

The reflectance of cirrus atmosphere from 0.4 μm to 2.5 μm is simulated by a combined atmospheric radiative transfer (CART) software. The effects of wavelength, optical thickness, effective size, cirrus altitude and surface type on the reflectance of cirrus atmosphere are analyzed. The relationships between reflectance of cirrus atmosphere at 0.55, 1.38 and 2.75 μm bands are simulated. Results show that the reflectance increases with optical thickness from visible with near infrared wavelength regions. Variation of the reflectance with effective size of cirrus clouds is small at the visible band, and the reflectance decreases with effective size at near IR wavelength band. The reflectance increases with cirrus altitude at IR absorption band of water vapor. Variation of the reflectance with different surface types is distinct in the atmospheric window region. The cirrus optical thickness and effective size can be retrieved with the relationship between reflectance of cirrus atmosphere at 0.55 μm and 2.75 μm, 1.38 μm and 2.75 μm bands.

Performance of adaptive optics (AO) system is limited by the delay errors caused by the servo system and photoelectron noise at the wavefront sensor. A multi-model univariate prediction model is proposed, which is based on the two-layer back propagation neural network with Levenberg-Marquardt learning algorithm. Using the multi-core processors, a novel predictive controller with parallel processing capabilities is designed that is able to predict the control voltage in the closed-loop AO system and eliminate the delay errors. Through numerical simulation, the prediction performance and parallel efficiency are studied. The control voltages of the AO system and the Strehl ratio are calculated and compared for the multi-model univariate prediction algorithm and proportional integral (PI) control algorithm. The results show that the residual error caused by servo delay in the system and Strehl ratio are improved effectively by using the predictive controller than by using the PI control algorithm. The prediction time is reduced by using multi-model univariate prediction algorithm.

A new approach to measure the temperature of large-scaled cold atomic cloud is reported. In this approach, a pair of anti-Helmholtz coils is used to generate a quadrupole magnetic field, the spatial distributions of cold atoms in different moments are measured by applying a vertical probe beam which can detect the density of cold atomic cloud point by point, and then the temperature of cold atoms is fitted from the experimental results. The temperature of cold atoms in integrating sphere is demonstrated to be 73±12 μK by using this approach. The result is also compared with time-of-flight (TOF) method.

In extreme-ultraviolet (EUV) lithography with off-axis illumination, the mask shadowing effects occur such as the orientation dependent pattern shift and critical dimension (CD) variation. Based on a simplified mask diffraction model, a theoretical analysis of the mask shadowing effects is proposed. The calculation formulas for the best mask (object space) focal-plane position and the correct value of mask pattern size are derived and used to compensate the mask shadowing effects. When focal plane of the mask is positioned on the equivalent plane of the multilayer, the pattern shift amount is reduced. When the mask pattern size is corrected using the derived formula, taking lines with the target CD of 22 nm as an example, the imaging CD bias is below 0.3 nm. However, with the decrease of the target CD, the formula error increases. It is necessary to consider the compensation of energy loss within the whole pupil that is caused by the off-axis illumination.

With a fully vectorial effective index method, the total dispersion coefficient of photonic crystal fiber is calculated, and it is deeply analyzed that how the total dispersion coefficient is affected by the structural parameters. It is found that the dispersion coefficient has two-extreme characteristics with the changes of structural parameters. 1) When Λ is unchanged, with the reduction of d/Λ, the zero-dispersion wavelength (ZDW) moves to the long wavelength. After reaching the maximum ZDW (MZDW), instead, the ZDW moves to the short wavelength. For example, when Λ=2.3 μm, the MZDW is about 1728.9 nm, as d/Λ=0.24, and the smaller Λ has the longer MZDW than the larger. 2) When d/Λ is unchanged, with the reduction of Λ, the ZDW moves to the short wavelength. After reaching the minimum ZDW (M-ZDW), the ZDW moves to the long wavelength. For example, when d/Λ=0.9, the M-ZDW is about 564.29 nm, as Λ=0.6 μm, and the larger d/Λ has the shorter M-ZDW than the smaller. These discoveries could be used to optimize design of the special photonic crystal fiber.

Fiber optic ring resonator (FRR) is the core unit of resonator fiber optic gyro (RFOG). A novel type of ring resonator of hollow core photonic bandgap fiber (HC-PBF) based on micro-optics structure is demonstrated. The normalized transfer function of such kind of FRR is built. Then simulations are done to formulate the relations of the resonator performance and the shot-noise-limited sensitivity of RFOG caused by optical parameters of FRR. The laboratory sample of HC-PBF ring resonator based on micro-optics structure is made and its performance is tested. An optimum design is put forward based on the test data. The research results provide the basis for further research of HC-PBF RFOG based on micro-optics structure.

Using the finite element method (FEM), the mode area, confinement loss and dispersion property of a novel microstructured optical fiber (MOF) with double-cladding structure are numerically simulated, analyzed and compared with those of traditional MOF. Results show that the fiber not only has a novel structure but also has better dispersion performance than the traditional photonic crystal fiber. Through rational optimization, several flat dispersion MOFs with large mode area in the 500 nm wavelength range are designed. The novel fiber proposed will have some reference value to the subsequent theoretical study and manufacture.

Using the exponential growth of idler gain and the properties of gain saturation of the four-wave mixing effect in a dispersion-shifted fiber (DSF), a 2×40 Gb/s two-channel all-optical 3R (re-amplification, re-shaping, re-time) regeneration system based on degraded signal pump is proposed. The theoretical analysis and experimental validation of the regenerative principle are conducted, and the regeneration experiment of different degraded signals in two wavelengths of 1550.92 nm and 1557.36 nm is completed. The receiver sensitivity (power penalty) of the degraded signals is improved from -20.3 dBm to -27.3 dBm and from -20.4 dBm to -25.6 dBm. The amount of improvement is 7.0 dB and 5.2 dB. System experimental results verify the theoretical analysis. A viable solution is proposed to solve the problem of simultaneous multi-channel signal regeneration in wavelength division multiplex (WDM) system.

The study about the compressed sensing technology indicates that selecting sampling model according to the sparse base has great influence on the reconstruction results. In the Fourier space, there are huge differences between the reconstructions of the stellate sampling of polar coordinates and the random sampling. The spectral distribution of the Fourier space is analyzed by Fourier optical theory and the phenomena is explained theoretically. It is proposed that the matching degree between the sparse base and the sampling model may affect the reconstruction effect. In the wavelet space, a comparison experiment of reconstruction is taken with the uniform sampling and the random sampling, and the latter is better. Therefore, a method is provided for the practical application to reduce the sampling rate and improve the reconstruction effect by choosing suitable sampling models according to the sparse bases.

The quality of the diffracted reference spherical wave in point diffraction interferometer (PDI) depends on the aberrations in the illumination objective lens and the quality and status of the pinhole. The pinhole diffraction of the visible light focused by the aberrated illumination objective lens is calculated based on the vector diffraction theory, the finite thickness and real conductivity of the pinhole are considered. The effect of illumination objective lens aberration on the quality of far-field diffracted wave front is analyzed. The best pinhole diameter of the PDI used to test the extreme ultraviolet lithography (EUVL) system and its element are determined. The calculation and analysis show that, when a system with NA of 0.3 is tested by PDI, the appropriate pinhole diameter is 800 nm, the RMS deviation of the diffracted wave front is 6.51×10-5λ, intensity uniformity is 0.812, when an optical element with NA of 0.3 is tested by PDI, the appropriate pinhole diameter is 500 nm, the RMS deviation of the diffracted wave front is 8.40×10-5λ, intensity uniformity is 0.664.

A new null Ronchi test method is presented to measure the shape of a large-aperture aspheric mirror. Null sinusoidal gratings are designed by using ray tracing and phase information of sinusoidal fringes. A transmission type liquid crystal display (T-LCD) screen generates null sinusoidal gratings and controls phase shift. An off-axis point source illuminates the whole mirror surface and a camera captures the phase-shifted fringe patterns containing the information of deviations of the mirror. By analyzing these fringe patterns, the actual transverse aberration and the corresponding ideal one are obtained. Then, the slopes of the deviations of the mirror are determined based on the geometric principle of the Ronchi test. The shape of the mirror is reconstructed after restoring the deviations of the mirror by integrating. Compared with the traditional null Ronchi test, the method eliminates the jagged edge of bands on the null gratings, and acquires the information of enough tested point. Simulations and a preliminary experiment validate the feasibility of this method.

In order to measure the fraction of laser-entrance energy, which is carried out mainly through the measurement of the fraction of backscattering light, a full aperture backscattering station measurement system is developed on the Shenguang-Ⅲ prototype laser facility. The system collects backscattering light from the laser-plasma interactions by the focusing final lens and decreases the beam caliber by the off-axial parabolic mirror. Stimulated Brillouin and Raman backscattering light are separated by single-edge dichroic beam splitter. The backscattering light′s energy, history and spectra can be detected by the system. The energy fraction of backscattering light is measured with and without beam smoothing. The results indicate that the fraction of laser-entrance energy will increase and the energy fraction of backscattering light will decrease with beam smoothing.

The dispersion formula of nitrogen gas at wavelengths from 0.145 μm to 2.058 μm in standard condition (temperature at 293.15 K, pressure at 101325 Pa) is deduced based on the theory of gas dispersion. A least square method is used for fitting this formula to measured data on the basis of two-order Sellmeier formula. The uncertainty of this formula in the overall spectral range is about 2.1×10-7, which is almost the same as the accuracy of the measured data. Comparing with the current dispersion formulas of nitrogen gas, the new formula presented here covers a wider spectral range, and provides more information of refractive index of nitrogen gas in 0.27～0.47 μm. The validity of the new formula has been approved by measuring the refractive index of nitrogen gas at 633 nm directly.

Photonic Doppler velocimetry (PDV) is a new type of laser velocimetry system. It is widely used in the velocity measurement of the shock wave, detonation wave and other short-time high-velocity moving objects. The velocities of different points of the target can be obtained using the multi-point measurement so as to find out the deformation of the target surface. It is proposed to use bare fiber bundle as probe of PDV system to improve the spatial resolution of measurement. Two-point velocity measurement with spatial resolution of 375 μm is experimentally realized. Because of smaller probe spacing, the measuring results obtained from a probe could be influenced by the reflected light from another probe. Theoretical and experimental results show that the measuring results are not affected by the stray light when the points on the target surface have the same velocity, and the measuring errors are related not only with the velocity difference of two points but also with the intensities and phases of sensing and stray light when the points on the target surface have different velocities.

The absolute testing method is an effective way to remove system errors of interferometer and improve the accuracy of the measurements of surface deviation. The method of single rotation based on least-squares fitting of Zernike polynomials requires rotating the spherical surface once and taking two measurements to calculate the rotationally asymmetric surface deviation. The formulas are derived theoretically in detail and the analysis of the impact of angle of single rotation on algorithm accuracy is presented. To prove the effectiveness of the method, the experimental result obtained from this method is compared with that of the multi-rotation method, and they meet well.

The coupled nonlinear differential equations are utilized to solve the loss spectrum of saturable absorber grating (SAG). Without laser cavity, the influences of the parameters of SAG on the filtering behavior of the gratings such as grating bandwidth, loss of the central frequency, sidemode suppression ratio (SMSR), etc. are investigated. Furthermore, the longitudinal mode SMSR in the presence of resonator cavity and the parameter regions for the improvement of the SMSR are discussed. The flat sidemode loss spectrum can be achieved when the length of the saturable absorber is almost one third of the fiber length in the cavity, and SMSR is 0.3~0.6 dB higher than that without a cavity when the length of the SAG is under a certain value due to the power of writing beam. Finally, the condition for the flat sidemode loss spectrum is validated elementally by a frequency-sweep method.

A high repetition rate femtosecond laser is used to irradiate the surface of dysprosium molybdate glass, and then two kinds of crystals, the β′-Dy2(MoO4)3 crystal with MoO4 structure cell and the α-MoO3 crystal with MoO6 structure cell, could be identified in the modified regions through micro-Raman spectra analysis. Energy dispersive spectra (EDS) further show that concentration of Mo element would be reduced in the center of the irradiated region as well as the α-MoO3 crystallization region, which indicates that laser induced microexplosion causes a rarefaction center under high temperature and pressure shockwave. With the irradiation time increasing, a phase transformation from β′-Dy2(MoO4)3 to α-MoO3 occurs. It implies that the concentraion reducing of Mo element due to femtosecond laser continuously irradiating could effectively promote the transformation of Mo-O structure from MoO4 to MoO6, and further for the α-MoO3 crystallization in lower Mo elements region.

A photo-thermo-refractive (PTR) glass based on SiO2-Al2O3-ZnO-Na2O (F, Br) glass system is prepared with a two-step melting process. The crystallization mechanism of PTR glass is studied by the transmittance spectrum, X-ray diffraction (XRD) and differential scanning calorimetry. It is demonstrated that the photosensitive spectrum of PTR glass is from 280 to 350 nm, the working spectrum is from 400 to 2700 nm, the optimum nucleation temperature is above 490 ℃, the crystallization temperature is near 595 ℃ and the crystalline phase is NaF. Volume Bragg grating (VBG) with the spatial frequency of 1000 mm-1 is recorded in PTR glass with direct laser interference patterning and “two-step” heat treatment process. The prepared VBG has a relative diffractive efficiency up to 91% and the angular filtering ability.

The fabrication of silicon nanowire arrays by post-treatmant in alkali solution is proposed. First, vertically aligned silicon nanowire (SiNW) arrays are fabricated over large areas using an electroless etching (EE) method. Then the SiNW is dipped in the mixture of NaOH and IPA for 10, 30, 50, 60, 90 s, respectively. Each SiNW is separated from the bunched SiNW. Compared to the bunched SiNWs, the dispersed SiNW arrays could drastically suppress the optical reflection(<4%) over a wide range of 400~1000 nm spectral bandwidth. In additon, the mechanism of nanowires scattered by the alkaline solution is tentatively analysed which provides the reference for the optimization of the preparation and application of silicon nanowires arrays.

For optimizing the optical system with liquid lenses, a new concept is proposed. Because liquid lenses have multiple discrete zoom locations, it is inevitable to select an initial solution for the damped least square algorithm (DLS) optimization process. The selection of the initial solution mostly depends on the experience of the designer. Thus different choices not only can bring the influence for different design processes, but also can bring difference in nature for the design results. To obtain a dynamic solution during the optimization process, an improved tabu search (ITS) algorithm with the DLS is proposed. The results show that automatic selection of zoom positions makes imaging effect better.

A design method of Wasserman-Wolf surfaces is proposed, and it can be used for non-rotationally symmetric catadioptric system. The method is used to design an elliptical dome optical system with compact structure and small volume. Least square method is used to make the initial structure of the system concern the aberration-corrected need of each scanning field. It is beneficial to overcoming the disadvantage that the system evaluation function is difficult to convergence when the entire scanning field is expanded from zero field of regard with respect to calculating and correcting the initial parameters of the components. In addition, the aberration properties of plane-symmetry mirrors with the vector aberration theory of plane symmetry are analyzed and principle by using tilt mirrors to compensate the dynamical aberration in different scanning fields is expounded. At last, a design example of mid-wave infrared elliptical dome optical system is demonstrated, and the results show that the modulation transfer function (MTF) of the optical system approaches the diffraction limit across the entire scanning field.

An optical system of mid-wavelength infrared for high resolution cooled wide-angle imaging system is designed. The F-number of the system is 2 and full field of view (FOV) 2ω=111.2°. The system chooses a double imaging configuration. By introducing an diffractive optical element, setting aspheric surfaces and symmetrically distributing three types of material Si, Ge and ZnSe with 6 elements, the system is athermalized across a wide temperature range from -55 ℃ to +80 ℃ and the optical aberration is well balanced. In the athermalized temperature range, the modulation-transfer function (MTF) is higher than 0.4 at 33 lp/mm, and the diffractive radial energy is more than 70% in a circle of 15 μm. Based on the principle of f-θ lens, the wide-angle system gets the same angular resolution for different fields. By intently introducing stop aberration and specifically controlling the FOV in imaging space, the focal plane gets a very good illumination uniformity, the relative illumination of off-axis field is more 90.9% and the efficiency of the cold stop is almost 100%. Moreover, the narcissus of the system is controlled very well. This system is suitable for high resolution cold 640 pixel×512 pixel detector with pixel size of 15 μm.

According to the problem of bad uniformities of expanding beams for excimer laser of projected lithography which holds 90 nm exposure linewidth, a kind of reflective beam expanding unit is modified based on the principle that the beam edges of Gaussian type can be superposed to enhance uniformities. It is shown that there is a contradiction between the sizes of superposing beams and the transmission rate on some areas of the parallel-reflective beam expander, in that case the improved capacities for the beam uniformities are restricted. Based on the above analysis, a non-parallel reflective beams expanding unit is proposed. The relational expressions about output beam sizes, number of superposed sub-beam units and wedge angle between two mirrors are deduced and the span of wedge angle and the structure of beam expanding unit are determined based on the deduced equations. The proposed beam expanding unit is proved to be useful to achieve beam expanding, as well as to reduce the interfering speckle effects. And the uniformities of output beams is increase.

Based on micro-electromechanical systems (MEMS) technology, a new MEMS Fabry-Perot filter structure combining with grating and Fabry-Perot cavity is proposed. This filter can not only ensure a wide free spectral range, but a narrow full width at half maximum (FWHM). The implementation mechanism, the parameters selection and the structure of the Fabry-Perot filter are analyzed deeply, and a simulation to calculate the mechanical and electrical properties of the micro-bridge deck is made. By choosing different thicknesses of the micro-bridge deck (0.5, 0.6, 0.7, 0.8, 0.9, 1 μm), and comparing with the flatness of micro-bridge within electrostatic interactions, it is found that when thickness is 1 μm and voltage is 5 V, the tilt displacement of the upper-mirror is 9.11 nm and the maximum cavity length change is 242 nm, which can maintain top and bottom cavity surface reflecting light parallel, and ensure filter′s filtering efficiency and selected light utilization. This novel filter is able to solve the contradiction that the free spectral range and the FWHM restrict from each other in traditional micro-filters. Thereby, the performance of the micro-filter has improved.

The Lyman-alpha line at 121.6 nm emitted by hydrogen, the most abundant element of the sun, plays a fundamental role in solar and coronal physics. As an intense chromospheres line, it is used to analyze many chromospheres features. To observe the Lyman alpha radiations in the field of 1.1R⊙ to 2.5R⊙ (R⊙ is solar radius), a reflective Lyman-alpha coronagraph with an aperture 40 mm, focal length 590 mm is designed. Two aspheric surfaces and one spherical mirror are adopted, and one piece of reflective mirror is used to fold the light path. The system′s transfer function is close to the diffraction limit, and the main stray light could be suppressed under 10-6 of solar photosphere radiation, which is in an acceptable level of the Lyman-alpha corona measurement.

Near infrared spectroscopy (NIRS) has advantages such as high speed, no destroying the samples, and online analysis. Therefore, NIRS is one of the most prospective ways to be applied in noninvasive biochemical analysis of blood. However, the absorption spectrum of blood is weak, and in near infrared noninvasive biochemical analysis, light energy utilization efficiency of conventional spectral instruments is lower. Thus, it is a little difficult in clinical noninvasive analysis with NIRS. In order to solve this problem, ellipsoidal reflector is used which has the focusing characteristic. The distributions of image points are analyzed according to geometric optics, and initial structure of ellipsoidal reflector is optimized by the method of ray tracing. In this way, the detector receives diffused light of finger which is collected with high efficiency. Through the simulation, light energy utilization efficiency of this ellipsoidal reflector raises nearly five times than that of non-optical collection device. It is beneficial to test the absorption spectra of blood and improve the signal-to-noise ratio(SNR).

The photometric, electrical, thermal and life features of LEDs are highly dependent on each other. All these factors should be considered together in order to optimize the operating point of LEDs. A photo-electro-thermal-life theory about LEDs is presented. This theory shows the inner links among photometric, electrical, thermal and life features of a LED. Using this theory, one can build a life prediction model and find out the relationship between output luminous flux and lifetime of LED. Based on this relationship, one can predict the lifetime of a LED and find the proper operating point, at which the LED will generate the maximum amount of luminous flux in its total life cycle.

In order to research the integration surface plasmon amplification by stimulated emission of radiation (SPASER) amplifier for surface plasmon (SP) amplification, the basic components of amplifier which includes metal-insulator-metal (MIM) structure embeded by the saturation absorber are designed and discussed. According to the basic principle of SPASER, stimulated emission conditions are analyzed. Meanwhile, the fabrication process of amplifier and pump pulse are designed, and the performance index is given. The results show that under the conditions of the incident light wavelength of 566 nm, pump light wavelength of 532 nm and the length of amplification region of 1~1.5 μm, the pulse response time can reach 100 fs, the bandwidth of 1.5~2 THz and gain of SP-ranges from 30 to 60. SPASER amplifier research will provide theoretical and technology foundation for large-scale integrated photonic chip, which will be widely used for high-speed communications system in next generation.

The relation between bending loss and curvature radius of the directly curved polysiloxane multimode optical waveguide for optical interconnection is analyzed. The bending loss is computed by using the Marcuse straight waveguide approximation method. The theoretical analysis indicates that the bending loss increases with the order of the mode and decrease-with the radius; the total bending loss transmitting in waveguide exhibits step changes. When the radius is larger than 4 mm, the bending loss is less than 1 dB/cm. The transmitting optical field is simulated by using the BeamPROP software under three different radius 5 mm, 10 mm and 20 mm, respectively. The bending loss is also measured by using digital scattering method. Experimental results indicate that the optical bending loss is between 0.55~0.8 dB/cm when curvature radius is between 5~6 mm, which agree with the theory in view of inherent propagation loss of the prepared polysiloxane straight waveguide.

Rugby-shape optical micro-resonators with diameters of 10~300 μm are fabricated by drawing poly (trimethylene terephthalate) (PTT) microwires immersed in biocompatible polyvinylpyrrolidone (PVP) polymer aqueous solution. Resonating characteristics are observed by coupling 532 nm green light and 671 nm red light respectively into the rugby-shape optical micro-resonators with diameters of 94.7 μm and 68.3 μm. By coupling 1549~1552 nm light into the rugby-shape optical micro-resonator with a diameter of 94.7 μm, absorption spectra are obtained and whispering gallery mode-based absorption peaks are observed, with a measured quality factor of 1.05×105.

To solve the problem of the low accuracy of the fiber optic gyroscope (FOG), which is influenced by the bad temperature stability of FOG that the parameters are changed, a new static synthetically compensation method is proposed. Thinking of the property of the temperature, the scale factor nonlinearity and the null-bias drift of FOG, a novel model is researched, which is related to factors of operating time, temperature and input angular rate. The model orders are identified using the classification fitting method. Based on a great deal of tests of temperatures and input angular rates, error-compensation iterative algorithm is proposed. The test results show that the compensation algorithm is effective at full temperatures and full rotate rates, which eliminates the influence of temperature and nonlinearity of FOG.

Blue GaN-based power LEDs are biased by negative human body mode electrostatic-discharge (ESD) before and during the aging time, and the characteristics of LEDs are analyzed before and after aging experiment and ESD stress. It is found that the generation of defects in active region and cladding layers after ESD stress could finally lead to degradation of LEDs. The defects induced by ESD could be restored partly in the early aging. However, these defects can cause parameters drift more seriously. Moreover, LEDs could be more sensitive to ESD stress and have lower antistatic ability during aging.

In order to realize double-band properties of frequency selective surface (FSS), the composite element FSS, composed of rectangle grid and three-ring loop element, is presented. The spectral domain solution method is described in detail for FSS. The frequency response properties of FSS are analyzed for electromagnetic waves with different incident angles and polarizations based on the spectral domain method. The result shows that the designed FSS can maintain stable double-band, flat top transmission and sharp skirt properties for electromagnetic waves with different incidence angles and polarizations. The double-band properties show stop-band from 1.8 GHz to 5.4 GHz and pass-band from 5.4 GHz to 20.0 GHz. The resonant frequency of stop-band is situated at about 3.1 GHz, and the bandwidth of flat top transmission is over 14.3 GHz at -4 dB. The sharp skirt properties show that the S-band signal is reflected intensively and other band signals are transmitted, and multi-band communication is achieved. The composite FSS can be used in satellite communications, radomes and other fields.

From the viewpoint of classical electrodynamics. The role of spin-to-orbital angular momentum conservation in spin Hall effect (SHE) of light is identified. The qualitative influence of polarization state, refractive-index gradient, incident angle upon the transverse shift of beam centroid is investigated: If the angle of incidence is the same, the absolute value of the horizontal polarization transverse shift is larger than that of the vertical polarization when the refractive index gradient is positive, but when the refractive index gradient is negative, the situation is exactly opposite. The transverse shift can be modulated by altering the refractive index gradient, the SHE of light can be enhanced obviously with the increase of the incident angle; For a certain circularly polarized component, the transverse shifts of left-handed and right-handed circularly polarized light have same magnitudes but the opposite signs. These findings provide a pathway for modulating the SHE of light.

A new optical component named detachable combined axicon for generating size adjustable bottle beam is proposed. This new type of optical element is opened a circular hole in the central part of the traditional axicon along the axis. Then a sencond axicon which is maded of common glass materials is embed in the circular hole. Single bottle beams or periodic bottle beams of different size can be generated by the way of replacing the second axicon (or the first axicon) of different base angle in this new optical element. Based on the geometrical optics, the principle of the generation of bottle beam is analysised and the related paramaters of the bottle beam are calculated. According to the Fresnel diffraction integral theory, the intensity distribution of the beam passing through this new optical element is simulated. The results of experiments coincide exactly with the therory discussed above.

A five-level 87Rb atomic system which is coherently driven by two standing-wave fields and a microwave field is investigated, and tunable triple photonic band-gaps at different resonance frequencies are obtained. The density-matrix equations are used to describe the interaction between fields and multi-level atoms, and the transfer-matrix method is used to describe the propagation of light in periodic media. The reflection spectra and transmission spectra of the coherent atomic system are deduced in the steady condition. The numerical simulation show that the positions and widths of the photonic band-gaps can be controlled by the detuning of the coherent fields and the Rabi frequency of the microwave field, and the type of triple photonic band-gaps can be applied to simultaneously manipulate the propagation of light pulses with three different central frequencies.

A system of angle-Doppler resolved reflective tomography laser imaging radar and its algorithm are given. The issue between transverse range resolution and the sampling time of single angle is solved. The condition of far-field diffraction transmission in the laboratory is designed, and the angle-Doppler reflective projections of the target are collected. Filtered back-projection algorithm is used to reconstruct the cross section of the target image. Because of the utilization of coherent detection of coaxial beams, both the imaging signal to noise ratio and the receiving sensitivity are improved. Due to the simplification in configuration and operations without involving signal phase processing, this technique has a great potential for applications in extensive laser radar imaging fields.

The global sea surface wind speeds are retrieved from the Version 3.01 CALIPSO lidar level 1 profile products and level 2 aerosol optical depth (AOD) products. The corresponding AOD data are used for correction of the two-way atmospheric transmittance, which allows wind speeds retrieval in the presence of significant amount of aerosols. Four months (January, April, July and October, 2007) of the CALIPSO lidar 532 nm data are selected. The AMSR-E sea surface wind speed data are collocated with the CALIPSO data and used for comparison. Effects from whitecaps and subsurface backscattering signals are corrected empirically using the lidar depolarization ratio measurement. The standard deviations of wind speeds retrieved by the CALIPSO lidar single-shot data and those by AMSR-E are 1.24, 1.24, 1.24, 1.20 m/s, respectively, for January, April, July and October, 2007. The standard deviations are reduced to 0.98, 1.02, 0.98, 0.94 m/s, respectively, when the lidar data is smoothed with a 5 km running window. The results show that the method works well and the data utilizing rate increases.

Multi-angle and polarized remote sensing can solve the problems such as atmospheric sounding and target recognition which cannot be solved by traditional methods. To implement the multi-angle detection of aerosol polarizing information, a new type of multi-angle polarimeter is proposed. This polarimeter can obtain polarizing spectral signals of the atmosphere on different angles, which provides a measured data source for the precise retrieval of the optical and microphysical characteristics of the atmosphere. As the principle prototype of a spaceborne instrument, the opto-mechanical system of the multi-angle polarimeter is composed of multi-channel parallel telescopic optical modules to realize the multi-angle detection of polarizing information within a view-field range of 110° and a sampling interval of 0.5°. With the design of dual controlling units and high-speed communication interfaces of the sampling control and data transporting system, the synchro-sampling and real-time transportation of the polarizing signals on multi-spectral channels has been fulfilled even on a high-speed platform, which eliminates the measurement errors brought in by platform movement and asynchronous acquisition of multi-channel signals. The on-flight experimental results show that this instrument can realize the high-precision measurement of aerosol polarizing information.

The hyperspectral image classification algorithm of relevance vector machine (RVM) is a supervised machine learning algorithm based on Bayes probability model, whose classification accuracy is good and the test time is short. However, the traditional RVM has some shortcomings that the training time will be very long and the effectiveness of the algorithm might decrease if the size of training samples is big or the dimensionality of the data is high. To solve these problems, a hyperspectral image classification algorithm of variational relevance vector machine (VRVM) is proposed. A new distribution is imported into the traditional probability model, which can replace complicated convolution operation with simple logarithm addition operation. Experimental results show that, in the classification of hyperspectral image, the overall classification accuracy and the number of relevance vectors of VRVM are nearly the same with RVM. However, with the increase of the sample, the training time has obviously reduced.

Since blurred image caused by defocus is of low clarity, image restoration is necessary. For traditional methods, circular disk or Gaussian function is adopted to approximate point spread function (PSF) caused by defocus, and the restored image is not ideal. Therefore, a method using Hartmann-Shack wavefront sensor to detect the wavefront distorted by defocus, calculating the PSF of the optical system with the obtained wavefront, and adopting the Richardson-Lucy algorithm to deblur the degraded image is proposed. An experimental system is built, from which both the blurred image and the corresponding wavefront information are collected. Clear restored images are obtained. Objective image quality assessment methods are used to evaluate the degraded and restored images. In addition, the restored images are also compared with the results obtained by traditional methods. Experimental results show that the proposed method can accurately reconstruct the PSF, and the quality of the image is efficiently improved.

In optical wavelength scale, any targets can be hardly regarded as stationary. Therefore, target vibration may affect synthetic aperture ladar (SAL) imaging severely. To address this problem, the imaging mathematical equation of strip-map mode SAL, in which quasi-monochromatic chirp laser is used as the illumination source in the situation of target vibrating, is set up. Using this equation, the effection of the vibration parameters is analyzed and discussed, and some simulations for this are performed. The result shows that SAL imaging is extremely sensitive to the target vibration. Tiny target vibration in range may bring a series of false images in azimuth to the final SAL image and seriously affect the realization of a high resolution SAL imaging.

Light scattering properties of textile fibers are essential in microscopic structure, optical properties and non-destructive examination for textile material. A published spectral approach is improved by utilizing anisotropic structural characteristics of textile fibers. The scattered electric fields of the oblique plane wave illuminating textile fibers and the Mueller matrix characterizing the scattering events are obtained, sequentially. Monte Carlo algorithm is then applied to simulate the multiple scattering of polarized light illuminating obliquely the parallel cotton fiber assemblies, and its results are verified experimentally. The scattering pattern calculated theoretically is in good agreement with the one measured experimentally on shape features and intensity distribution, which indicates the validity of Monte Carlo method for simulating the textile fibers scattering in this paper and indicates that this method has important guiding significance in the theoretical study of light propagation characteristics of textile material.

The Bessel beams are often used for investigating the scattering properties of the non-diffracting beams, but the Bessel beams cannot be realized physically. As the pseudo-nondiffracting beams and the exact solution to the paraxial Helmholz equation, the Bessel-Gauss beams can be generated directly from the laser resonator and possess finite spatial width. The dimensionless scattering function is derived for the Bessel-Gauss beams scattered by a sphere by means of Fourier-transform, the plane wave expansion and the vector spherical wave expansion. By numerical simulation and comparison with that of the Bessel beam and the Gauss beam, it can be found that only the scattering intensity is influenced as the spherical scatterer is shifted off the beam axis, the directions where the scattering extreme points exist almost keep no changes. The scattering is dominant in the direction of the conical angle for the Bessel-Gauss beam and the Bessel beam, but for the Gauss beam the forward scattering is always dominant.

The selection of modeling method is one of the main factors influencing the quantitative accuracy with visible-near infrared (Vis-NIR) spectroscopy. We compare the performance of three calibrations methods, i.e., principal component regression (PCR), partial least squares regression (PLSR), and back propagation neural network (BPNN) based on Vis-NIR reflectance spectra of soil total nitrogen (TN) quantitative forecast results. Covered in the 470~1000 nm wavelength range, spectroscopy of 48 soil samples selected from 12 profiles are air-dried, screened and mushed, then processed by the first order derivative and Savizky-Golay smoothing methods. Leave-one-out cross validation is also adopted to determine the optimal factor numbers. The results indicate that PCR and PLSR linear models are able to meet general prediction and with little difference, where coefficients of determination (R2) are 0.74 and 0.8, respectively, and residual predictive deviation (RPDs) are 2.23 and 2.22. The two nonlinear models built by BPNN in combination with PCR and PLSR, respectively, are superior to the linear models of PCR and PLSR in the precision of prediction. BPNN, principal components (PCs) whose input is the PCs resulted from the PCR, while the BPNN latent variables (LVs) whose input is the first 4 LV results obtained from PLSR has the best performance (R2=0.9, RPD is3.11). It is recommended to adopt BPNN-LV model to rapidly predict the vertical spatial and temporal distribution of TN with Vis-NIR spectroscopy.

TiO2 films deposited on well clean K9 glass substrates are investigated for their thickness and optical constants by transmittance and X-ray reflectivity (XRR) measurements. XRR provides highly accurate results on film electron density, thickness and interface roughness. The film thickness from XRR is used as an anitial value to accelerate the evaluation of the transmittance data. With the help of Forouhi-Bloomer dispersion model, the calculated transmittance curves fit the experimental results well. For the same sample, the film thickness values from the two methods are very close and the maximum difference is 4.9 nm, which indicates that the combination of XRR and transmittance spectra can improve the reliability of the optical characterization.

Considering the quantum confinement effect in silicon nanoparticles, the effect of the mean silicon nanoparticle size on optical band gap, optical transition oscillator strength, and the temperature dependence of the band gap and optical radiation, a model is introduced to analyze the photoluminescence (PL) of silicon-nanostructure thin film with a certain size distribution of silicon nanoparticles. Where Gaussian function as well as log-normal function is considered for size distribution of silicon nanoparticles. The results show that there will be blue shifts of the PL peak energy decreasing with mean size and size dispersion. As the temperature increases, the PL peak exists red shifts accompanied by decreasing of light intensity. The comparative analysis of the simulation results and experimental data of PL implies shows that our model can well explain the PL in silicon nanostructure thin film in different temperatures.