To effectively reduce the packet loss probability (PLP) and guarantee quality of service (QoS) of different priority bursts, a deflection routing mechanism is proposed based on priority and burst segmentation in optical burst switching networks. In the core node, contention is resolved through incorporating prioritized burst segmentation with deflection routing scheme. The burst segmentation scheme allows the head of contending bursts or the tail of original bursts to be segmented. The segmented burst is scheduled on the optimum deflection path by the parameter-tunable deflection routing scheme. An analytical model is proposed to evaluate the contention resolution scheme through calculating PLP and the normalized end-to-end delay. Results show that high-priority bursts have significantly lower PLP and the delay than low-priority. So the deflection routing mechanism based on priority and burst segmentation can effectively resolve the issue of the burst contention, and improve the performance of OBS networks.

Based on the non-linear Schrdinger equation which suits the optical soliton propagation，by using the symmetry split-step Fourier transformation， the transmission evolution of super-Gaussin pulse in the zero dispersion region was simulated and the effect of attenuation, high order chromatic dispersion, high order non-linearity and initial chirp on propagation was analyzed. Some conclusions were obtained: attenuation has a great effect on the pulse amplitude but little on the shape of the pulse. The amplitude and pulse width of basic soliton pulse are always at the same standard when the pulse transmitted a certain distance, which is the ideal media for optical communication. As for the high order soliton pulse, it is changing all the time. When all the influence factors come to work together, it has a great effect on the pulse propagation. However, selecting a reasonable combination of these factors, it is possible to get a high order soliton pulse to transmit information. This result has certain amount of valuable reference significance on the study of improving the relay distance between two optical repeaters by strengthening the input power of pulse.

The relationship between reflected polarization properties of apodized low-birefringence fiber Bragg gratings and temperature effect are analyzed theoretically. Numerical value simulates reflection spectrum of the apodized low-birefringence fiber Bragg gratings, polarization dependent loss and differential group delay curves with wavelength.The reflected center wavelength, polarization dependent loss and differential group delay which evolve with wavelength are measured under different temperature and the curves are drawn accordingly. The evolution of polarization dependent loss and differential group delay are analyzed according to the results of the experiment. In the experiment, reflected polarization dependent loss presents two peaks which shift to an equal extent as temperature changes and it indicates that polarization dependent loss has no significant difference. The maximum of difference group delay with temperature increase linearly shifts to long wavelengths and it is proved that the loss of orthogonal modes has changed equally. The synthetical theory analysis and experimental results show that the evolution of reflected polarization properties of apodized low-birefringence fiber Bragg gratings changes obviously with temperature changing and the changing of orthogonal modes presents proportional characteristics.

Existing color separation diffraction grating spectroscopic methods are generally discussed on parallel incident light,and there is little discussion of the case of non-parallel incident light.Based on the Fresnel diffraction theory and the angular spectrum theory,the spherical wave propagation field and plane wave diffraction wave field are linked together,and angle function parameters are imported.Combined with fractional Talbot effect theory,the dual-wavelength wide-angle incident wave field function is derived.The numerical simulation of the function is presented to calculate the diffraction distribution in different locations,which are compared to standard parallel incident wavefield distribution and calculated horizontal broadening and offsets.By adjusting the width of grating steps，the diffraction field broadening and offset is compensated.Each single-wavelength diffraction width is less than half of the grating period,thereby reducing the dual-wavelength optical diffraction aliasing.In this paper,the relation between the wave field width and grating step width is presented.The step width curve is fitted based on multi calculate points on the phase grating and the grating parameter can be used to achieve well spatial separation of dual-wavelength wide angle incident light.The result is available for a variety of wide angle incident situation,such as dual-wavelength imaging,LCD display,LCD projection,etc.

An innovative method that refractive index distribution of single-mode ion exchange is reckoned using an experimental way was proposed and implemented.The key conditions are that a diffusion coefficient must be constant.Then two-dimensional diffusion equation with ion-exchange waveguide was deduced,the presumption of the ion exchange waveguide refractive index distribution of article was fitted.The Lorentz relation solved the optical dispersion problem in sensor test. Experimental results showed that characterization guided mode propagation imaginary part constants of absorption loss by the KK transform.On this basis,the trial of fiber-article waveguide - fiber optic integrated sensor was designed.The glucose concentration was measured and the detection limit was 0.1 μM.The experiments confirmed the validity of the method for a low concentration of trace tests.

The band structures of triangular hole lattice photonic crystal are studied via plane wave expansion method, with radiation loss considered. The spectrum transmission characters of the designed structure are investigated through simulation of finite difference time-domain method (FDTD) for hexagonal ring resonant, and the parameters of lattice period and radius of the lattice hole are optimized. The drop efficiency rate of 98.8% and FWHM of 32.9 nm are reached through optimization of the designing in C-band. The samples are tentative fabricated through photolithography and reactive ion etching and the parameters are given. The transmission spectrum is measured and the reasons of differences from the theoretical result are analyzed. The schemes of slab based on ring resonant holes show a more stable character than the rod ones, and have potential future for filters, WDM devices and micro gyroscope.

The sub-wavelength micro/nano structure surfaces have an excellent anti-reflective property. Based on application foundation of silicon-based solar cells response spectrum of 300～1 200 nm, reflection characteristics of two-dimension micro/nano structure surface were calculated influenced by the surface face forms,structure parameters with duty ratio,height and period,and incident angle of light waves,using the finite difference time domain method.Further analysis was done by effective medium theory.The results show that the reflectivity is large for the grating structure of uniform section,and the influence of structure parameters is small; anti-reflective performance of a taped gradient section structure is much better,and the reflectivity is observably reduced with duty ratio and height increased; and it also showed that the reflectivity is a little smaller at the incident angle range of ±40°.The simulation and analysis on reflective property of sub-wavelength micro/nano grating structure,provide the basis on the design and fabrication of anti-reflection surfaces.

Metal-enhanced fluorescence from CdTe nanocrystals placed in close proximity to novel synthesized Ag/TiO2 core/shell nanoparticles films is studied. The Ag/TiO2 core-shell nanoparticles are synthesized in the aqueous solutions by colloid chemistry method. Transmission electron microscopy images show that mostly Ag/TiO2 core/shell nanoparticles are clearly core/shell structure and ball-shaped. The deep black cores are Ag nanoparticles and the light-colored shell are titania. In addition, the UV-Vis absorption spectra of Ag nanoparticles have a surface Plasmon absorption at 409 nm arising from the particles. A similar surface plasmon absorption due to Ag nanoparticles is observed for the titania-coated Ag nanoparticles at a wavelength longer than for the Ag nanoparticles. This absorption shift is caused by a high refractive index of the titania shell and strongly suggests a complete coverage of Ag nanoparticles with titania. The as-prepared Ag/TiO2 core-shell nanoparticles are assembled with the CdTe nanocrystals and spin-coated on the quartz glass. Significantly enhanced fluorescence emissions are observed, implying that enhanced fluorescence from Ag/TiO2 core/shell nanoparticles films is mostly due to an electric field enhancement effect. The current work is available for the applications in light emitting device, fluorescence imaging, and biological detection.

Silicon monoxide (SiO) thin films is one of the most commonly used optical film in short-wave and medium-wave infrared. High packing density for enhancing the spectra stability and the quality of the optical thin film elements is extremely important. Used 99.99% purity of the SiO bulk materials, SiO thin films were prepared by molybdenum boat thermal evaporation in 5×10－4 Pa vacuum pressure at different deposited temperature, while deposition rate was monitored and demonstrated at 1.2~1.5 nm/s by quartz crystal oscillation controller. The thickness of the thin films on silicon substrate is about 2.2~2.4 μm. Used the Fourier transform infrared spectrometer to test the spectral characteristics of SiO thin films before and after the thin film suck tide, and according to the wavelength deviation dispersion theory, calculated the packing density. The results show that as the deposition temperature increased, the packing density increased, from 0.91 at room temperature to 0.99 at 250 ℃.

In order to simulate the equivalent radiation environment of target and interference in the IR guidance process, and compare the effect of differences in target thermometry between monochromatic thermometry and chromatic thermometry under complex environment, the target and interference on space 6 km is simulated using IR dual color composite simulation system. Through the IR thermal imager calibrated accurately, the target and interference at different temperatures is tested using two methods of temperature measurement-monochromatic thermometry and chromatic thermometry. Blackbody temperature is measured by calibrated MW thermal imager and calibrated LW thermal imager, in the range from 25 ℃ to 40 ℃ of Blackbody temperature, the absolute error limit of LW thermal imager is 0.5 ℃, and in the range from 50 ℃ to 120 ℃ of Blackbody temperature, the absolute error limit of MW thermal imager is 0.2 ℃. When the target temperature is 500 ℃ and the interference temperature is 1000℃, the temperatures of target measured with LWIR, MWIR and chromatic thermometry are 28.5 ℃, 148.3 ℃ and 322.4 ℃, and the temperatures of interference are 56.7 ℃, 223.2 ℃ and 660.1 ℃ respectively. The experiment result indicates that the characteristics of target temperature can be better reflected using chromatic thermometry than monochromatic thermometry under complex environment.

A novel algorithm is presented to compute the plane equation of line structured light, which used in a machine vision measurement system. Image subtraction is introduced to obtain stripe image, Steger algorithm is applied to extract sub-pixel level centric points of stripe in the image, and orthogonal line fitting is performed to gain line equation of stripe in image coordinate system. The approach of calculating world coordinates of points on the plane of line structured light from feature points in calibration board is introduced, which are based on the principle of cross-ratio invariance. Finally, orthogonal least-squares plane is proposed fitting to get light plane equation. Some specific optimizations in algorithm are put forward to improve total accuracy, and the scheme design of calibration system and detailed experimental procedure are given. Experimental results demonstrate that the method is of robustness and has high accuracy.

Motion precision of scanning mirror is vital for motion compensation results of imaging spectrometer. So a set of scanning mirror testing equipment was developped to test scanning mirror′s motion precision. Firstly, the motion compensation principle was analyzed and the theory motion compensation curve was given. Then according to 15 m GSD and 2% degradation of system MTF, the scanning mirror′s position and speed precision target was confirmed. At the same time the relationship between motion compensation times and relative rotation error was outlined. Based on 1/3 principle, the guide line of scanning mirror′s motion precision testing equipment was proposed. In succession a set of testing equipment found on 0.8″ absolute optoelectronic encoder was established. The testing precision of testing equipment was analyzed carefully from the testing method error and the encoder′s error aspects. Finally, scanning mirror′s motion precision testing experimentation was carried out by the testing equipment. The testing results indicate that the scanning mirror′s position precision is better than 3″±1.2″, speed precision is 5.6%±1.29% when 4 times compensation and 3.85% ±1.29%when 6 times compensation. These precision results meet the requirements of imaging spectrometer for scanning mirror. Favorable imaging property of the imaging spectrometer proves that the proposed testing method and equipment is suitable for scanning mirror′s check and accept.

For the further development and exploring the instantaneous laser velocimetry, the laser beat frequency velocity interferometry is analyzed and its applications and future development are discussed. Based on the key technology of it, the concept of time-divided laser velocity interferometry is proposed. It uses the time delay τ in the laser beat frequency velocity interferometry, in which the interferometer is keyed to measure, respectively, the displacement and velocity in the time range 0~τ and above τ, to determine the velocity of the first front edge in the shock process (i.e., solve the uncertainty of fringe losing). In addition, it provides a higher upper bound of velocity measuring with lower hardware requirement (i.e., with lower requirement for the bandwidth and sampling of the oscilloscope). Combining with the consideration of the property of shocking and explosion process, the parameters are selected properly, the fiber time-divided laser velocity interferometer is designed, its primary applications are investigated, and the application results show its effectiveness.

When the ship target is far away from infrared/visible compound seeker, the target information and features are very weak which can be extracted from the infrared and visible images, and conventional image registration algorithms are hard to take effect on the registration of infrared and visible image with small target. To solve the problem, an infrared and visible image registration algorithm is presented based on sensor parameters. Firstly, according to the imaging model of compound seeker, the image registration is decomposed to two independent steps (field of view registration and shift registration). Secondly, the field of view for different sensors are registered using the sensor parameters. Finally, the shift registration is performed by matching point located based on horizontal high pass energy image and sea-sky line. The experimental results prove that the method is feasible and effective, and can be used in the actual registration of infrared and visible image with small ship target.

In the process of enhancing the image compression, the traditional image enhancement method has some difficulties in the details recovery and generates blocky effect easily. A new method of JPEG2000 compressed image enhancement is proposed by applying the thory of Retinex into JPEG2000 compact frame. The dynamic range of scene illumination is adjusted through two low-frequency coefficients non-linear mapping with the low frequency coefficients as the wavelet transform and the high-frequency coefficients as the reflected light. By adjusting the high frequency coefficient, image contrast is enhanced,so that the image detail is trended overall. Then illumination quantity table is modifyed and more details is retained and block effect is controled through judging activities of each sub-block and adaptive quantization.The experimental results show that this algorithm performed well in the reinforcing effect and the compressed quality,etc.

An algorithm for line segments detection in digital image was proposed to improve the computation and the robustness of the traditional algorithm. The image gradient is meaningful to detect the edge of structure in the image; therefore its module and direction are computed in the first step. According to the pseudo-sort result of the gradient module, the region-growing method is implemented to enlarge the pixels with the same direction, where the connected pixel region is considered as a candidate region of line segments. Finally, enclosing rectangle is used to describe and judge the candidate region by long axis and short axis of the rectangle, belong which the long axis that meets the requirement is the desired line segment. Based on MATLAB program, an experimental simulation was performed. Experimental results depicted that 108 line segments after 8.87s of computation were detected by the proposed algorithm; which saved 17% of computation time and also detected 16% line segments more than the traditional LSD algorithm. This algorithm can accurately and quickly detect line segments in many complex environments in the image.

A robust zero-watermarking algorithm resisting JPEG compression and geometric attacks is proposed, since digital images always face with JPEG compression and geometric attacks through transmission. An original image is split into non-overlapping blocks, and each block is conducted with singular value decomposition.Then, each block′s singular value matrix is transformed with Harr wavelet transformation. The zero-watermark sequence is produced by judging the numerical relation between the mean of diagonal elements of singular value matrix′s low frequency band from two adjacent blocks. The mathematical theoretical analysis shows that the zero-watermark sequence is essentially produced by judging the numerical relation between the mean of singular value matrix′s singular values from two adjacent blocks. The proposed algorithm has perfect visibility due to no alteration made to the original image. The experimental results show that the proposed algorithm has strong robustness to resist JPEG compression and several geometric attacks such as rotation, scaling, random row & column removal, row & column shifting, print-and-scan.

Based on making holographic lampbox backlights need,the possibility of using LEDs as the illumination sources for reconstruction of color holograms is studied on chrominance,compared with tungsten halogen spotlights which are common practiced currently.And the reproduced images those are displayed by a LED and a tungsten halogen spotlight are evaluated.In experiments,the RGB proportions in spectral power distributions of a LED and a tungsten halogen spotlight are calculated.Color differences,in CIE1976 uniform color space,between an original picture and the reproduced images are also calculated and analyzed separately.Results indicate that it is feasible to use LEDs to illuminate color holograms,and feasible to use LEDs as holographic lampbox backlights.

For most of the auto-focus function′s problems of focus curve losing desirable characteristics ,falling into local minum value and even losing single peaks,which are caused by changes in contrast,Zernike orthogonal moments was introduced and the focus curves were analyzed.A new auto-focusing algorithm was presented which based on Zernike orthogonal moments and the algorithm′s formula was defined.Compared with the focusing effect of several commen auto-focus funtions under different contrast,the algorithm can adapt to different contrast under the focusing process by adjusting the weighting coefficients to change the formula.The experiment compared the algorithm focus curves′ difference between completely symmetric objects and non-symmetric objects,proved that the algorithm is more effective for the completely symmetric objects such as the round or spherical objects.

The micro pore array with the feature of aspect ratio of more than 50 was fabricated on a 5-inch silicon wafer by means of photo-assisted electrochemical etching technique. After analyzing the factors that determine the morphology of pores, the following experiments were proceeded to study the relation of pore morphology to the etching voltage and electric current. When choosing a current by i-V curve, several different etching voltages were applied respectively and corresponding wafer was fabricated by adjusting experimental parameters according to the Lehmann′s model. It was found that the better pores could be produced under a voltage of 2 V. The results show that photo-assisted electrochemical etching technique can be more adopted to make the large area micro-pore array of high aspect-ratio on silicon wafer with low cost, being compared to the state of the art of micro-manufacturing technology.

Photo response non uniformity (PRNU) noise sets a fundamental limit on image sensor performance, especially under low illumination in remote sensing system. After introducing a complete noise model of time delay and integration charge coupled device (TDICCD), a complete model of TDICCD noise is proposed. Then PRNU noise model for all selectable integration stage is developed, which is linearly related to illumination. Moreover, exposure stages are taken into consideration due to their inherent averaging effect, and relationship between nonuniformity parameter and integration stage is developed. Finally a techinque to identify and measure the PRNU noise in TDICCD is presented by analysis of the output images, and the experimental results are illustrated from test system, measured PRNU noise at different illumination and integration stages are also analyzed.

The surface texturing of multi-crystalline silicon is a technology that needs to be broken through. For the improvement of pit-trap shape on mc-Si surface, a new method is presented that the passiviation of Si surface is done to reduce the acid reaction velocity. A series of experiments are carried out in which mc-Si is textured in the acid solution containing different content of additives under the same technology parameters. Samples′ structure is observed using scanning electron microscope, and the reflectance spectrum is measured by spectrophotometer. The SEM of experimental sample′s surface illustrates that the textured surface is full of a lot of pit trap with the good light trapping effect. Moreover, the deep ditch like valley, which increases the leakage current of mc-Si solar cells and reduces the conversion efficiency, is not found in the surface. This method is valuable for the improvement of solar cell efficiency.

Based on the frequency response model,high frequency characteristic analysis for the TO packaged avalanche photodetectors is presented.The influence on frequency performance of TO packaging,including chip,bondwire,transimpedance amplifer,and TO header elements are investigated.By changing inductance parameters induced by different bandwires,the different frequency responses are observed.Finally,by considering the engineering conditions,a －3 dB bandwidth of 10GHz is optimally obtained for TO packaging.

Block matching electronic image stabilization algorithm is a stable algorithm with high precision.Block matching algorithm reiteratively calculates the difference of the two image blocks,from the starting point to the matched point in the interesting area.Huge computering work is the main problem preventing the application of the block matching algorithm.Time series prediction method is used to estimate the optimal starting point,in order to reduce the search time.Proper time series model is selected according to the inner statistical properties of the global motion vector,then the model parameter is estimated and the model is checked by residual analysis.The time series model and historical data are used to predict the current global motion vector,which is set as the starting point.The experiments show that this method can reduce the hunting range effectively,making the matching process more quick,and it can be adopted by the other electronic image stabilization algorithms.

980 nm vertical cavity surface emitting laser (VCSEL) ohmic contact technology is investigated. In order to enhance the output power of VCSEL, and extend its reliability, the contact resistance has to be reduced. Ti/Pt/Au alloy is chosen as the metal contact system for P type high doped GaAs, and Ge/Au/Ni/Au alloy is chosen as the metal contact system for N type GaAs. By chosing the optimum alloying temperature of 440 ℃ , the lowest ohmic contact resistance of 0.04Ω is obtained. Comparing output power and conversion efficiency of the device for 440 ℃ and the device for 450℃, the test results show that, the ohmic contact resistance is 0.04Ω, the peak wavelength is 980.1 nm, the FWHM of spectral is 0.8 nm, the lateral divergence angle θ‖ is as low as 15.2°, the vertical divergence angle θ⊥ is as low as 13.5°, the output power is 1.4 W, the maximum conversion efficiency is 14.4% for the device of 440 ℃ , while the ohmic contact resistance is 0.049 Ω, the output power is 1.3 W, the conversion efficiency is 12.8% for the device of 450 ℃ . By optimizing the alloy temperature can effectively reduce ohmic contact resistance of VCSEL 980 nm.