The characteristics of media in communication channel are analyzed briefly and the reasonable optical parameters of media are adopted. With certain communication system parameters the temporal and spatial distributions of the received signal from submerged platform are simulated using Monte Carlo method. The upper limit of the ratio of Monte Carlo estimated error to averaged value is about 0.3%. From the simulated results, the optimized sampling timing of receiver and field of view of telescope are obtained. Also the signal-to-noise ratio of the receiver is calculated. Based on this, the error probability of the communication system is deduced from laser pulse position modulation and maximum likelihood detection. The results show that under severe environment robust laser communication from a satellite to a submerged platform can be achieved.

In order to achieve the accurate pointing during the course of deep space optical communication link setting up, a scheme is proposed to acquire the extendbeacon for deep space optical communication system, which uses visual earth image as beacon, and stores a beacon image in aircraft as reference image. CCD is adopted to image the pointing area based on antenna scan pattern. The correlation coefficient of the detected image and reference image is calculated based on pixel scan pattern. The location where the correlation coefficient is maximum is found out after antenna scan, which can be regard as the location where the beacon image is acquired. Because of the rotation and scale invariance of FourierMellin transform, the effect from rotation and scale of image will be eliminated based on FourierMellin transform. 1000 random fields of view are created using MonteCarlo method, to simulate the progress of acquisition, and the result shows that the successful acquisition probability is 99.6% within 3σ, validating its feasibility.

A tunable wavelength conversion between picosecond pulses is experimentally demonstrated by using the cascaded sum and difference-frequency generation (cSFG/DFG) in a periodically poled LiNbO3 (PPLN) waveguide. The pulsed signal with 40 GHz repetition rate and 1.57 ps pulse width is adopted. When the input signal and the first control wavelengths are kept at 1554.2 nm and 1532.5 nm, respectively, the output signal wavelength can be tuned from 1536.0 nm to 1545.2 nm as the second control wavelength is varied from 1550.5 nm to 1541.0 nm. By varying the first control wavelength to satisfy the quasiphase matching (QPM) condition for the sumfrequency generation (SFG), and simultaneously adjusting the second control wavelength, the tunable output signal wavelength can be also obtained when the input signal wavelength is changed. In the experiment, the amplified spontaneous emission (ASE) noise from the erbium-doped fiber amplifier (EDFA) is effectively suppressed by employing two narrow band tunable filters. Therefore, the wavelength down and upconversions are simultaneously observed.

Separate detection of the degenerate mode groups (DMGs) of multimode fiber was realized by using a multimode tilted fiber Bragg grating (TFBG). With multimode TFBG, the DMGs were coupled into the radiation ray that was focused separately on the focal plane of a lens. The radiation ray corresponding to the nondegenerate mode group is separated on the focal plane, so that the radiation ray of DMGs was selected and detected. The experiment of DMG separate detection was conducted with a standard communication multimode fiber with a core diameter of 62.5 μm. The multimode fiber was excited by a single mode fiber, and optical power of DMGs was measured while changing the lateral misalignment of exciting fiber. The measured results of the lowerorder DMGs were in consistence with the numerical results, but there appeared a few deviations for the higherorder DMGs. Further experiments needed to be conducted to verify the idea of exciting a single DMG in a multimode fiber using a TFBG and then measuring the coupling coefficient under this condition.

A method is proposed to realize millimeterwave subcarrier optical signal generation by using a doublewavelength optical spectrum of a certain distance with fiber grating filter. The feasibility is analyzed by designing an apodized moiré fiber grating in which Fourier transformation reverseengineering method is used, the simulation result is also presented, and the related fiber grating filter is fabricated experimentally. Moreover, based on the designed fiber grating and the technology of wavelength division multiplexing, a frequency shift keying radiooverfiber communication system is suggested, and the proposed scheme may be taken as one of the candidates for the next generation highspeed and largecapability radiooverfiber system.

Alloptical buffer can store the packet in optical domain without the opticalelectronicoptical conversion and becomes the important element in alloptical network. Its performance greatly influences the loss packet probability of network. But in semiconductor optical amplifier (SOA) cross phase modulation and cross gain modulation are coexisting. So in recircular alloptical buffer based on SOA, the output power between stored packet and packet without being stored is unbalanced and bit-error ratio of network increases. After analyzing the working principle of alloptical buffer, a new method for power equilibrium in recircular alloptical buffer based on SOA is presented. By changing the usual positive control pulse to negative and using an electronic variable optical attenuator, the buffer gets an uniform output power and its performance is improved markedly. This method can be applied in all recircular alloptical buffer based on SOA.

Autofocusing is an important technology in the measurement of images and computer vision. After careful experiments, it is found that some conventional autofocusing algorithms are valid without the influence of the noise. However, they will be invalid in the influence of noise. So, an auto-adaptive algorithm to autofocusing is proposed. At first, edge pixels are separated from nonedge pixels by an adaptive threshold in small local area. The global threshold is adapted to eliminate the influence of noise further. Finally, images are computed with some autofocusing method carefully, which are calculated by the adaptive method. At the same time, the focusing curves are plotted. Experiments show that the method not only owns the characteristics of good unbias, clear single apex and high sensitivity and so on, but also has strong performance of antinoise.

Based on the duality theory of mathematical morphology and the relation between the mathematical morphological theory and linear shift invariant system, polarizationencoding method is introduced and a scheme is proposed to realize hitmiss transform by onechannel and zerothreshold value incoherent optical correlator. With polarization encoding, the scheme combines the foreground and background of an image into an encoded image, so the hitmiss transform can be completed by an erosion operator, onechannel realization. The experimental setup is simplified greatly, more compact, and the demand for adjusting light path is reduced. Because of the zero threshold, the error for the threshold of output image resulted from the ununiformity of optical intensity distribution and the imperfect linearity of CCD optoelectronic response is avoided. The experimental result proves the convenience, accuracy and validity of the scheme.

An objective lens composed of a spherical Fresnel zone plate mirror and aspheric mirrors is designed. The Fresnel zone plate with spherical shape is analyzed, and the method to replace aspherical mirror with spherical Fresnel zone plate approximately is deduced. The objective lens is designed with a single spherical Fresnel zone plate mirror and three aspherical mirrors. Under the condition of magnification 100, Fresnel number 2.5 and field angle 120°, the modulation transfer function reaches above 40% at 0.6 line pairs/mm on the magnification side, and the distortion is less than 2.2%. This method can provide reference for application of Fresnel zone plate in visible light imaging, and possesses a bright future with the continuous development of fabrication technique of Fresnel devices.

For the sake of the requirement for attitude determination at arc second order in spacetoearth platform, the imaging model of star sensor in inertial frame is constructed. It is pointed out that the main error for attitude determination of star sensor is the positional error of star in focal plane. Meanwhile, based on the defocused model of ideal optical system, the quantitative relationships among the main indexes of star sensor are analyzed and the minimum projection position error is also given. It is shown from the simulation results that, the blur area's diameter between 2～3 pixels is the best performance scheme using the slightdefocusing of star sensor's focal plane array and the positional error is only 0.2 pixel, which meets the requirement of star sensor about 4～5 arc second in inertial frame. The analyzed and simulated results can be references for the following indexes determination and designing of star sensor.

The transmission fluctuation spectrometry (TFS), a new method can measure the particle size distribution (PSD) and particle concentration simultaneously in twophase fluid system. With simple optical arrangement and easy operation, the method can be applied to realtime, online measurements. The transmission signal with fluctuations are analyzed by using a second order low pass filter (LPF), and the experimental data of transmission fluctuation spectra in the frequency domain are obtained. The particle size distributions and concentrations are extracted from the experimental data with the modified Chahine iterations. Effects on the spectra of transmission fluctuations of the high concentration are discussed. The high concentration correction includes frequency response to characteristic function and step height of the transition function. Parameters for the high concentration correction are achieved and further employed in the inversion algorithm, and reasonable measurement results are obtained. Analysis to the measurement data shows that, with the high-concentration correction, the particle size distribution and particle concentration can be recovered within a broad dynamic range and the largest measurable particle volume concentration depends on the particle size.

Moiré tomography is applied to measure the complex density distribution in supersonic wind tunnel. Multidirectional moiré fringe patterns of flow field in supersonic wind tunnel are obtained by using high precision moiré deflectometer and discrete angle rotating mode. In the tomography, a new moiré tomographic algorithm with deflection angle revision iteration is employed to realize the limited angle sampling and incomplete data reconstruction containing covering. In the iterative process higher reconstructing precision is achieved by using inner boundary smoothing filtering. In experiments, 9 tomograms with different view angles in supersonic wind tunnel with Mach number 2.52 are obtained, the distribution on the cross section of asymmetric flow field in the swelling area of supersonic wind tunnel is reconstructed after 50 iterative computations, and the measured results and errors are analyzed and discussed. Computing fluid dynamics is used to simulate and compute the density field, verifies the validity of the reconstruction results of moiré tomography, and proves the important value of the technology in measuring complex fluid field.

Based upon the dispersion principle of cylindrically bent crystal, a transmissioin crystal spectrometer has been presented primarily to characterize the hotelectron energy distribution of laser generated plasma by registering the hard Xray Bremsstrahlung spectrum emitted by hot plasma in a larger energy range. Actually according to the different experimental requirements and variety of the distance between source and crystal, Johann adjustable Xray transmission spectrometer is designed. The focal slit of the spectrometer is planned to be movable, and its place can be changed when the distance between the source and crystal is changed. By the theoretical calculation the relation is acquired about the position of the focal slit and the distance from source to crystal, and the consequence can also be reached as long as Bragg angle is less than 0.2 rad that the incident light will approximately focus on one point. Furthermore, the primary theoretical analysis and conclusion on the questions of designing the device are illustrated.In conclusion, the adjustable focal slit breaks the limit of fixed distance from source to crystal; moreover, the frontal replaceable crystal makes the spectrometer more flexible.

In the indirectdrive target compression process, filamentation instability should be considered seriously, especially light intensity fluctuation of size 10～100 μm on the surface of target, and a proper beam smoothing method must be adopted. A new method, using fiber system to stack chirped pulses, treating the composite pulses with angular spectral dispersing, depressing the high frequency fluctuation in the focal spot with the beam smoothing mechanism of focal spot at different time determining different spectral components, is proposed. Using Fourier transform method, the expression for the composite pulses dispersed by a grating is deduced. Simulation results show that the size of the focal spot increases with the number of color cycles and the basic pulse bandwidth in the dispersing direction. Through calculating the flux contrast including 95% focal spot energy with time, it is found that the optimal smoothing effect is achieved with one color cycle and 0.3～0.5 nm basic pulse bandwidth. This new kind of temporal beam smoothing technique has the advantages of simple configuration and adjustable focal spot performance, and is of important application value in improving irradiation uniformity on target surface.

The SmithPurcell radiation characteristic is analyzed from a train of line charges moving parallel to a sinusoidal metal reflection grating using an integral equation model. The numerical computations have been performed for short grating period with low-energy line charges and long grating period with high-energy line charges. The maximal radiated intensity is obtained at harmonics of the line charges modulating frequency. The bandwidth of the radiation becomes narrow, energy peaks are intensified. The normalized energy distribution produced by the high energy charges will have a narrow angular range emission peaks in the forward direction. The coherent radiation of frequency-locked will be generated gained from a train of line charges. The results show that the THz radiation will be produced for the reasonable parameters of line charge energy and grating.

A new scheme to realize laser cooling of blocky Tm3+doped glass ZBLANP (ZrF4BaF2LaF3NaFPbF2) is proposed. In this scheme, the absorption of pumping laser power is greatly enhanced by multiple reflections between two sets of mirrors with high reflectance, and with the incident angle of the pumping laser being Brewster angle, the reflection losses on the two endfaces of the sample are reduced to nearly zero. The cooling temperature of the bulk sample and its cooling time constant in this scheme are calculated in detail, and the relationship between cooling power and number of passes with 1 W incident laser power and between sample temperature and number of passes under the different incident laser power is studied . It is found that the sample is cooled from ambient temperature (300 K) to 275.7 K, 267.4 K and 259.1 K when the incident laser power are 3 W, 4 W and 5 W respectively. Especially, when the incident laser power is 4.5 W, the sample temperature is reduced to 263.2 K, namely, the sample temperature change is 36.8 K, which is 1.53 times of that froma conventional scheme, and the corresponding cooling time constant is about 22.7 min.

The organic-inorganic hybridized polymer system is studied for its photorefractive, which is sensitized by CdS nanoparticles, and combines PVK polymer to transmit carriers, DO3 nonlinear chromophores and ECZ as plasticizer. It is confirmed that the CdS colloidal particle has a nanoscale size and quantum confinement effect by adopting transmission electron microscopy and UVVisible absorption spectroscopy. The photoconductive properties of the PVK-CdS thin films is studied. Significant enhancement in photoconductivity induced by charge transform between CdS nanoparticles and PVK has also been demonstrated. An asymmetric twobeam coupling gain of 78.4 cm-1 is obtained without the external electric field, confirming the photorefractive effect of the polymer. The experimental results showed that the polymer film possesses strong photoinduced orientationenhanced photorefractive effect. When irradiated by two coherent beams, the sample can set up refractive index grating with the diffraction efficiency of 4.4%.

Monochromatic aberrations of postLASIK eyes, which are categorized into starburst group, doublevision group, moisture group and reference group according to their visual symptoms, are measured. The rootmeansquare (RMS) value of the third and the fourth-order aberrations from the starburst group and moisture group are larger than the counterparts from the reference group. The modulation transfer function (MTF) of the first three groups is obviously lower than that of the reference group. It is found that in daytime, the fourth-order aberrations, especially the symmetrical Zernike spherical aberration, have more effects on the modulation transfer function, while the effects of the third-order aberrations, especially the asymmetrical Zernike coma, on MTF are more important at night. Higher-order aberrations above the fourthorder still affect the eyes suffered from visual symptoms, whereas no evident effect on eyes of the reference group is found. RMS and MTF are effective ways to study the visual symptoms after surgery.

Nowadays, all of the iris identification methods are utilizing the strategy that extracts the feature points in the iris image firstly and then encodes these points into a certain length feature data for matching. But this strategy can make the iris identification system fragile to be attack. In order to improve the security and speed of iris identification systems, an iris identification method based on gray surface matching is proposed. This method abandons the traditional operations including the feature extracting and encoding, but uses the theory of gray surface matching based on feature analysis. Firstly, it computes the gray difference of two pixels from two different images and gets the gray difference surface. Secondly, it calculates the variance of the gray difference surface and considers this variance as the distance between two feature surfaces. At last, it decides whether these two irises are in the same eye or not according to the variance. When the distance threshold equals to 40, the accuracy is 96.89% and the speed is 53.2 ms. Experimental results demonstrate that this method has high accuracy and speed.

The static fluorescence polarization spectrum of human whole blood whose concentrations are from 0.3% to 10% at the band 350～800 nm induced by 408 nm polarized light of LED is reported. The spectral characteristics and redshifted phenomenon for different concentrations of the sample solutions are discussed, and the mechanism of the red shift is also given. The polarization degrees of two key fluorescence emission bands (around 490 nm and 610 nm) in polarization spectrum are measured. The result shows that the polarization degrees of two key flourescence emission bands are 0.4796 and 0.4344 when the concentration of blood is 1.0%, and the polarization degrees are 0.2064 and 0.0538 when the concentration is 8.0%. The variation rule of the polarization degree with the concentrations of whole blood is studied. The different energy transfer mechanisms as well as the variation of polarization degree in different fluorescence areas for different concentrations of whole blood is studied by the theory of energy transfer. The research will gives reference to lightinduced autofluorescence diagnostic techniques of organic tissue.

The condition of the gap soliton production in fiber Bragg grating and the parameters conditionality are advanced. The propagation equation of the solitons in fiber Bragg grating is described by the nonlinear coupledmode equation,then the differential equations of parameters are obtained by applying the perturbation method, and the approximate results for parameters are calculated. Based on the conditions of the gap soliton in periodical nonlinear optical medium, two inequalities for relation of parameters are obtained. These parameters conditionality and physical significance are proved by numerical calculation. It is theoretically showed that proper parameters choice is needed for gap solitons production in fiber Bragg grating. The study lays the theoretical fundation for solitons production experiments and further engineering application in fiber Bragg grating.

Compared with other nonlinear optical materials, KTP crystal has prominent advantages. But it also has high conductivity and it becomes difficult to efficiently control domain-reversion with conventional method, due to the existence of ionic current. So to conquer the difficulty, it is necessary to monitor the polarization-reversal process of KTP crystal in realtime. Realtime monitoring method into the fabrication of PPKTP, short for KTiOPO4 is carried out by utilizing electrooptical effect. The principle is analyzed theoretically and the result demonstrates the validity of the method experimentally. Compared with the results without the monitoring method, the conversion efficiency of PPKTP crystal increases by many times. It is proved that this method can be used to enhance the quality and repeatability of PPKTP crystal fabrication, and is also effective to examine the quality of PPKTP crystal.

The selfdeflection characteristic of dark photovoltaic spatial solitons in onedimensional stable closed circuit is investigated by taking diffusion effects into account by use of numerical method. The results show that dark photovoltaic spatial soliton possesses a selfdeflection process in the direction of the crystal’s axis during propagation and the center of the dark solitary beam moves on a parabolic trajectory. The selfbending distance of the dark solitary beam increases monotonously with the photovoltaic field and the solition input intensity. The self-deflection of dark photovoltaic solitons is very different from that of bright ones.

The refractive index of a medium will change periodically when ultrasonic wave propagates in it, and the medium can serve as a grating with light propagating perpendicularly to the propagating direction of ultrasonic wave. When the light and ultrasonic wave propagate in the same direction through the medium, the medium can be considered as one-dimensional photonic crystal, one-dimensional ultrasonic photonic crystal (1DUSPC). With plane wave method, it is proved that the 1DUSPC shares the same bandgap characteristic of conventional multilayered one-dimensional photonic crystal, and the bandgap can be modulated by adjusting the wavelength and amplitude of ultrasonic wave. New method is provided to control light by the study.

A new method, laserassisted wet twostep etching, which can wipe off the crystal tropism influence in laser induced wetchemical etching is proposed. The essential of this method is to lengthen the etching time in noncrystaltropism direction and assure the same etching degree as that in the crystaltropism direction. Theoretical analysis and experimental result show that the crystal tropism influences the etched image greatly; and the twostep method can wipe off the crystal tropism influence effectively; and also, compared with the adopted normally surface mask film and laser intensity distributing accommodation, the twostep etching method can deal with the inside crystal tropism influence, operate easily and the requirement for equipment is low. The laserassisted wet twostep etching can overcome the disadvantages of conventional ones, and wipe off the crystal tropism influence and is useful in the fabrication of specialstructured optoelectronic devices and optoelectronic integration.

A novel photonic crystal fiber structure with high nonlinearity and flattened dispersion is proposed and analyzed. A parameter δ is introduced to measure the nonlinearity and dispersion flattening. The effect of hole size on nonlinearity and dispersion of the photonic crystal fiber (PCF) is analyzed by the plane wave expansion method. The new structure introduces six additional smaller holes between the holes of the first ring to reduce the effective area, and increase the nonlinearity of the fiber. By optimizing the diameters of the first, third, and the additional rings of holes, a PCF with high nonlinearity and flattened dispersion is achieved. The dispersion parameter is within ±0.5 ps/(km·nm) in a 330 nm wavelength range, within ±0.1 ps/(km·nm) in a 230 nm wavelength range, and within ±0.05 ps/(km·nm) in a 200 nm wavelength range. The effective area is 2.26 μm2. The nonlinearitydispersion flattening parameter δ=11.8 ps·W/μm2.

The squeezing properties of twomode atom laser coupled output from the system of Ξtype threelevel atomic BoseEinstein condensate (BEC) interacting with single-mode squeezed coherent light field are studied, and the influence of interaction among atoms in BoseEinstein condensate on the squeezing properties of atom laser is discussed emphatically. The relation between squeezing of atom laser and initial squeezing factor of light field is also considered. The results show that the atom appears periodically squeezing, and the squeezing properties of atom laser depend closely on both interaction among atoms in BEC and initial squeezing factor of light. The interaction among atoms affects the oscillation frequency of squeezing of atom laser but doesn't squeezing depth. The initial squeezing factor of light modulates the squeezing depth of atom laser, and the greater the initial squeezing factor, the shorter the squeezing time of atom laser is.

The four characteristic parameters of the fluorescence spectrum of ethanolwater solution have been studied. Their timedomain and frequencydomain characteristics have been obtained too. The emission and excitation spectra have indicated that there are three luminescent substances with different molecular structures in the ethanol-water solution. The fluorescence peaks center at 290 nm, 305 nm and 330 nm with the corresponding optimal absorptive peaks located at 265 nm, 280 nm and 236 nm, respectively. The variation of the fluorescent intensity with the volume fraction of ethanol also attests to the existence of the three different luminescent substances. The decay process of fluorescent intensity is detected at the center of the emission peaks, and it is fitted by an exponential function and the analyzed data are obtained by deconvolution. The fluorescence lifetimes are 8 ns, 12 ns, and 25 ns, respectively. From the four fluorescence characteristic parameters of ethanolwater solution, it can be concluded that when ethanol and water molecules are gathered together the clusters with new structures are formed which can emit fluorescence photons with different energy. The research contributes to the study of the molecular structure characteristic of ethanolwater cluster.

ZnO∶H thin films were prepared on cAl2O3 substrate by adding a little of H2 gas in the reaction gas when depositing ZnO films using atmospheric pressure metal organic chemical vapor deposition (APMOCVD) with carrier gas of nitrogen. Zn(C2H5)2 (DEZn) and H2O were used as Zn and O precursors, respectively. The crystallizing performance and optical properties were studied with doublecrystal Xray diffraction and photoluminescence. The full widths at half maximum of the (002) and (102) omega rocking curves of ZnO∶H films were 46.1 mrad and 81.4 mrad, respectively, indicating the high crystal quality of the films. The room temperature photoluminescence (PL) spectra showed that there was a strong ultraviolet emission at 380 nm. Meanwhile, a strong hydrogen-related bound exciton peak (I4) at 3.363 eV and its two electron satellite at 3.331 eV were observed in the low temperature PL spectra at 10 K. In addition, the thermal stability of hydrogen in ZnO∶H films were investigated by observing the intensity change of I4 peak during the annealing process. The results showed that the intensity of I4 peak decreased with the annealing temperature increasing, indicating the hydrogen would evolve out from ZnO∶H films annealed at high temperature.

GeC/GaP doublelayer thin film is used as longwave infrared (8～11.5 μm waveband) antireflective and protective thin film system for ZnS substrates. Gallium phosphide thin film has been deposited by RF magnetron sputtering with high-purity Ar as the working gas and single crystal GaP disc as the target, and germanium carbon thin film has been deposited by reactive RF magnetron sputtering in high-purity Ar and CH4 mixture with single crystal Ge disc as the target as well. The thicknesses as well as the optical parameters of both GaP film and GeC film, such as refractivity and absorption coefficient are determined by least square fit of the films' IR transmission spectra, with the refractive indexes expressed in Cauchy formula and absorption coefficients in Urbach formula. The refractive index of GaP film is about 2.9 at 10 μm wavelength and is close to that of the bulk GaP; and that of the GeC film is about 1.78 at 10 μm wavelength. With these refractive indexes, GeC/GaP double-layer antireflective and protective thin film system is designed by computer and deposited on ZnS substrate. The GeC/GaP film system shows relatively large absorption when the GaP film is relatively thick, which makes the antireflective effect poor, whereas the antireflective effect is good when the GaP film is relatively thin.

The frozen deuterium-tritium (DT) foil compressed by shaped laser pulses is explored, and analytical results show that the foil plasma target with relatively high density and low temperature can be obtained after compression by stepped lasergenerated multi-shock waves. The choice of initial laser intensity influences the plasma density after compression and the number of neutron to be produced. The optimized result is obtained by adjusting the initial laser intensity to keep a high neutron high productivity in the compressed DT target. When the original scaled laser amplitude is 0.5 and the final 32, the foil's density reaches 18416 times critical density, the temperature reaches 16 keV, and 109 /J neutrons are produced after compression, which is four orders of magnitude larger than that from other ways.

The relative diffraction performance of soft Xray phase condenser zone plate and amplitude zone plate is obtained by measuring the first order diffraction intensity. The experimental setup is simple and relatively easy to adjust. The size of the pinhole is selected to obtain sufficient signalto-noise ratio. The thickness of the polyimide substrate is normalized. The first order diffraction intensity of three Ni and three Ge zone plates is compared with that of an Au zone plate. The experimental average first order efficiencies of the Ni and Ge zone plate are 1.60 and 1.26 times that of the Au amplitude zone plate at 3.2 nm, which are close to the theoretical results: 1.77 and 1.34. The measurement result indicates that the phase zone plates fabricated has high accuracy.