The growth conditions and optical properties of InGaN/GaN dual-wavelength light-emitting diodes structures were studied. The optimized dual-wavelength quantum well parameters were obtained by designing several groups of epitaxy structures with different parameters. The optical properties of the dual-wavelength light-emitting diodes devices were also studied. The results show that light emitting of quantum well depends on localized exciton emitting caused by In clusters, which also leads to abnormal optical phenomenon under small current. Mechanism of the dual-wavelength peaks drift was explained through numerical calculation of internal polarization field in InGaN/GaN quantum-well. Luminous efficiency under different drive current of dual-wavelength light-emitting diodes was investigated and the possible reason of "droop" effect was given.

Signal lights, navigation lights and searchlights used in transportation, ships, airport and other applications are vulnerable to be influenced by weather factors. The penetration properties of different color LED light sources in water fog and sea fog were compared by a series of experiments, and the relationship between the transmittance T of LED light sources and distance in water fog was researched. Experimental results show that the transmittance T of all kinds of color LEDs declines exponentially with the increasing of transmission distance in fog; both in water fog and sea fog, the yellow LED has the maximum transmittance compared to other color LED sources. This research work will provide a basis for the design of signal lights, searchlights and other functional illuminating fields.

In order to realize the high-power operation of the Tm doped silica fiber laser，there are two issues to be resolved. One is how to couple the pump light with low beam quality into the inner cladding of gain fiber efficiently，the other is on thermal management of gain fiber. For an efficient coupling，firstly the output beam of high power LD modules were collimated to reduce their beam diverge in the plane parallel to the array by use of a telescope system consisting of cylindrical aspheric lenses, and then the collimated beams were cut with splitters. The beam parameter in the plane parallel to the array was optimized to 84 mm·mrad(BPPx)after the reshaping, and a coupling efficiency of 70% was achieved by this means. A series connection structures consisting of two sections of Tm fiber with identical length were fabricated to increase available pump power and reduce the heat generation along the fiber ends, which achieved a total available pump power up to 580 W. With regard to the thermal management, each end section(250 mm long) of fibers was embedded in a V groove in a water cooled aluminum heat sink. Based on this configuration a continuous-wave output power of up to 280 W operating at 2 015 nm was obtained for 6.4m-long gain fiber, corresponding to a slope efficiency of 55.6% with respect to the launched pump power. The experimental results indicate that the special reshaping and laser configuration can improve pump coupling efficiency, reduce thermal loading and give more even axial temperature distribution along the gain fibers.

An efficient tunable, continuous-wave, mid-infrared, intra-cavity singly resonant Nd∶YVO4/MgO∶PPLN optical parametric oscillator is presented. A laser-diode-pumped 1 064 nm laser is taken as the incident light and a MgO∶PPLN crystal is used for frequency conversion. The variation of the mid-infrared output wavelength with the grating period and temperature of nonliear crystal is investigated. The grating period of multi-grating MgO∶PPLN chip is from 29.52 to 31.59 μm by a step of 0.5 mm. The tuning wavelength of the signal light is from 1.48 to 1.63 μm and that of the idle light is 3.0 to 3.8 μm. A maximum signal power of 760 mW and idler power of 360 mW are obtained under a pump power of 8.06 W at 808 nm, with corresponding to a total optical efficiency being 13.9% when the grating period is 30.5 μm.

Based on the theory of quasi-three-level, a model of threshold pump power of Ho∶YAG laser was built and the formula of threshold pump power of Ho laser was derived. The relationship between the threshold pump power and the absorption efficiency of gain medium or the transmission of output coupler was analyzed. The mW-level threshold pump power of Ho laser can be obtained by optimizing parameters of Ho laser, which was pumped at 1 907 nm or wing-pumped at 1 930 nm. Furthermore, center wavelength lasing at 2 092 nm or 2 124 nm should be achievable by choosing an output coupler of suitable transmission.

Slab laser has characteristics of high output power and high beam quality. However, the far field facula of its primitive outgoing beam presents as an oval, and its far field divergence angle is slightly big which makes it inconvenient to transmit in very long distance. To solve this problem, a cylindrical telescope system with optimized parameter was used to expand the x-scale outgoing beam (wide divergence angle). The goal is to make the x-scale divergence angle approximate to the y-scale divergence angle which is a small angle, and constrain the beam waist position of x-scale and y-scale to the same place in case of out of focus. The general sphere beam expander was used to compress far field divergence angle of the reshaped beam. At last, the laser beam with a far field divergence angle less than 100 μrad and ratio of x-scale to y-scale less than 1.1 in out-field experiment was obtained.

Owing to based on the nonlinear effect, the all-optical encryption scheme can only be applied to encrypt signal wave optical signal，in order to solve the problem that recent all-optical scheme can not be applied to encrypt multi-wave signal simulatanously, an all-optical encryption system scheme which can simultaneously encrypt multi-wave optical signals is proposed. This scheme is based on pockels effect of LiNbO3 waveguide and achieves signals encrypting operation by interference of coherent optical signals and keys. Because the difference of modulated phase shift of different wavelength through the same external electric field in LiNbO3 waveguide is very small, so, this scheme can be used to encrypt multi-wave signals simultaneously. The simulation of encrypting operation of 8×10 Gbps multe-wave signals whose interval of wavelength is 0.8 nm has been realized, the code type is NRZ with rise and fall time of 10ps, the BER of the encrypted output is smaller than 4.30×10－86. And the Q factor of the encrypted output is larger than 21.51. It is verified that this all-optical encryption scheme can achieve encrypting operation for multi-wave input signals within 400 nm by theoretic analyzing and simulative experiment. Moreover, when the wavelegth range of multi-wave signal is small than 78 nm, the operation of all-optical enctyption can improve the extinction ratio of input signals. Theoretic analyzing and simulative experiment comfirm the feasibility and effectiveness of this multi-wave all-optical encryption system.

The spectral reflectivities of Na2KSb film both with and without the activation layer of Cs-Sb were measured. The measurement results show that the spectral reflectivity of these two films within the wavelength range of 200~700 nm are substantially the same. So it can be inferred that the refractive index of both films are also substantially the same, and both films are substantially no change in its internal constituent. Because the spectral reflectivity is not changed, and the optical path difference of the two films are substantially the same, it can be inferred that the Cs-Sb layer was coated on the surface of Na2KSb, so the activation process of Na2KSb surface is a surface effect. In addition, Na2KSb (Cs) multi-alkali cathode film was analysis by XPS, the results showed that three elements was existed mainly in addition to Na, K, and Sb, but also the presence of CO2 and a small amount of Cs elements. The reason of the presence of C and O on the surface of Na2KSb (Cs) is oxidation in the unsealed process of sample. After 10 second time of argon ion etching, Na2KSb (Cs) multi-alkali cathode film was analyzed by XPS, the results show that the energy spectrum of Cs atoms has disappeared. When continued for 600 seconds time of the argon ion etching, on Na2KSb (Cs) multi-alkali cathode film surface, the spectrum of the Si atoms in XPS spectroscopy was appeared. This showed that argon ion has already etched to the glass window surface. According to the etching time analysis, the thickness of the multi-alkali cathode film surface Na2KSb (Cs) of Cs atomic layers approximately 3nm, and the Cs atomic layer exists only in the surface of the multi-alkali cathode layer, No deep into the multi-alkali cathode film inside. The fluorescence spectrum of multi-alkali cathode with the Cs-Sb activation and without Cs-Sb activation has been measured. The measurement results showed that the peak wavelength of fluorescence spectrum of sample with Cs activation has shifted toward to short directions, while peak intensity of samples with the Cs-Sb has reduced compared with sample without Cs activation. This phenomenon lustrated that not only quantity of transition electron of Na2KSb multi-alkali cathode with Cs activation has increased, but also the energy level of transition electron has increased. This phenomenon can be defined as "volume effect" during the activation process. So during activation process of Na2KSb multi-alkali cathode, the reason of sensitivity increasing is a kind of surface effect and a kind of “volume effect”. Yet this “volume effect” refers to the structure change inside the multi-alkali cathode layer material, rather than surface Cs atoms diffused inside film.

In order to optimize the performance of optical devices fabricated with side-polished fibers, online polishing and measurements were conducted to study sensing characteristics of side-polished fiber effected by polishing length and remaining cladding thickness. The results show that, the sensing characteristics for the side-polished fiber used here can be divided into three regions according to its sensitivity. In the first region with refractive index ranging from 1.300 to 1.450, low sensitivity and high linearity were achieved. In the second region with refractive index ranging from 1.450 to 1.500, high sensitivity can be obtained, while very low stability. Compared to the first two regions, both relatively good sensitivity, linearity and stability can be spontaneously achieved in the third region with refractive index higher than 1.50. Comparisons of fibers with polishing length of 10 mm and 20 mm show that the increase of polishing length do not result in improvement of sensitivity, while a loss in stability. The remaining cladding thickness plays an important role in determinating the sensitivity and the insert loss of side-polished fiber, therefore it should be optimized to achieve the desired sensors.

A new type of optical fiber refractive index sensor was reported. The partial multimode fiber of Singlemode-Multimode Singlemode (SMS) fiber structures were side polished, and then the refractive index sensing properties were analyzed. The results show that multiple loss peaks can be observed in the spectrum of the SMS fiber structures because of the mutual interference by the high order modes of multimode fiber. The spectrum of the SMS fiber structures just drift a few nanometres when the SMS fiber structures are polished to close to the core; but when being side polished to the core of the fiber, the spectrum changes dramatically. After being side polished, the refractive index resolution of the SMS fiber structures will be higher. When the polishing depth is 10 μm and the polishing length is 10mm, the refractive index sensitivity of the SMS fiber structure is 703.2 nm/RIU. When the polishing depth is 15 μm, the refractive index sensitivity of the SMS fiber structure is 798 nm/RIU. When the polishing depth is 20 μm, the refractive index sensitivity of the SMS fiber structure is up to 1 190 nm/URI. A new type of optical fiber refractive index sensor can be realized with a single side polished SMS fiber structure, and the sensitivity of the sensor can be improved by increasing the depth of polishing area.

In optical burst switching networks, burst retransmission can reduce data loss which is arising from the burst contention in optical burst switching core nodes, but uncontrolled burst retransmission may increase network load and data loss probability. A quality-of-service -supported controlled retransmission scheme was developed in optical burst switching networks. In this scheme, different values to retransmission probability were set at each contention and the total transmission times were controlled referred to the network load condition. In order to reach a balanced network load, the conflict bursts would no longer be retransmitted but be discarded when the number of burst retransmission reached our times. The path blocking probability and the burst loss probability for different priority bursts were calculated to evaluate the performance of the proposed scheme. From the simulation, with the retransmission probability and the number of retransmission increasing, it can be seen that burst loss probability is reducing while the path block probability is reducing. So it is necessary to control the retransmission probability and the number of retransmission. This result has some theoretical significance for the designing of an optical burst switching network test bed.

Four types of two-stage reflection systems were compared, of which the secondary mirror was an ellipsoidal, a hyperboloidal, a planar or a paraboloidal type, respectively. A paraboloidal dish with a diameter of 1 000 mm and a rim angle of 60° was employed as a primary mirror in each system. The final concentrated spots were on the vertexes of the dishes in ellipsoidal and hyperboloidal systems. As to the ellipsoidal system, the flux density on the receiver increases when numerical aperture (NA)＜0.4 and then decreases. The spot size decreases with NA increasing. The maximum flux concentration ratio is about 5 000× at NA=0.4. As to the hyperboloidal system, both the flux density and spot size decrease with NA increasing. The maximum flux concentration ratio is about 8 990× at NA=0.5. The ellipsoidal and hyperboloidal systems are applicable to cases which need low concentrated spots, and the latter has a higher flux concentration ratio with a more compact structure and a smaller secondary mirror. As to the planar system, with the relative location of the concentrated spot increasing, the flux density increases and spot size changes less obviously. The maximum flux concentration ratio is about 21 000×. A planar secondary mirror is suitable for a system which needs a high concentrated spot. As to the paraboloidal system, with relative location of the secondary mirror increasing, the flux density increases and spot size decrease. The maximum flux concentration ratio is 80×, which is the lowest among systems mentioned above. A paraboloidal secondary mirror is suitable for a system that requires low concentration ratio.

Film diffractive optical elements have the characteristics of low surface mass density and loose surface shape tolerance. Large aperture diffractive telescope system with film diffractive optical element as the primary lens, as the optical payload with super large aperture, light quality and high etendue, has much more ability to observe earth real time and undisturbed. In this paper, the fabrication tolerance of the diffractive element and relative position tolerance of the two parts of the diffractive telescope system were analyzed. A measuring method of the relative position of the two parts of the diffractive telescope system was proposed and the measuring accuracy was analyzed. This method can form two spots on the CMOS active detectors with laser source using two computer-generated holograms (CGH) elements of different structures. The center deviation and tilt can be measured exactly by analyzing the relative position of the two spots real time. The measuring accuracy can achieve 50 μm and 0.003° respectively. These critical technologies lay the foundation for the implementation of the large aperture diffractive telescope system in the future.

According to phase shift method of binary optics, the phase delay expression and the diffraction efficiency of Harmonic Diffractive Optical Element (HDOE) at oblique incidence were derived. And the relationship between the diffraction efficiency and the beam incident angle was qualitatively analyzed. The results show that in the visible spectral range, resonant wavelength of HDOE becomes longer with incidence angle increasing, otherwise in the short-wave part wave crest drift more violently; with incident angle of monochromatic light increasing, the diffraction order sequence appears; with the incidence angle increasing, the wave crest of ordinary binary element drift much faster than HDOE; beam incidence angle has a greater impact on diffraction efficiency of short wave.

In order to evaluate the space camera image degradation induced by micro-vibration, an integrated modeling was utilized. During the CCD integrated time, the image motion of dynamical imaging system equaled to static image spot blurring, dynamical imaging evaluation criterion was proposed and dynamical imaging was analyzed. All kinds of disturbance sources were introduced, the finite element model (FEM) of space camera was founded, and the jitter values of optical elements were acquired by dynamic analysis with finite element software. The analysis results were imported to optical design software CODE V, the system spot diagram and MTF were attained, and the image motion was calculated quickly. The integrated modeling which shortens the computational quantity effectively suits to any type of micro-vibration. This method was utilized to simulate the space camera image degradation induced by micro-vibration, and the dynamical imaging evaluation criterion was used to estimate the effects on imaging system. The relationship between micro-vibration and image degradation was achieved, that provides important references for system optimum design and vibration isolation.

A novel miniature double-beam spectrum measurement system called double-beam miniature spectrometer is proposed based on twin-core fiber and 2-D CCD detector Czerny-Turner spectrometer. The advantage of the miniature spectrometer with small size and fast measurement and the superiority of the double-path spectrophotometer with double light paths are both presented. In this system, the light intensity of the light source is divided equally into two parts by the first two-core fiber. One part is considered as the reference light path, and the other is considered as the sample light path. Then the reference light and the sample light are transferred to the slit of the spectrometer by the second two-core fiber. The two cores of the second two-core fiber are distributed along the direction of the length of the slit (the direction of the non-dispersion). In order to avoid the overlapping and the crosstalk between the sample spectrum and the reference spectrum, astigmatism corrected cylindrical lens is added into the symmetrical crossed Czerny-Turner optical bench. The optical design software is used to optimize the system aberration. The ray tracing result shows that the imaging spectra of the reference path and the sample path can be clearly distinguished in the wavelength range of 300~800 nm. Eventually, the experimental prototype of the double-beam miniature spectrometer is manufactured and aligned. The sample spectrum and the reference spectrum can be displayed and measured simultaneously with the system. The design method of the double-beam miniature spectrometer is verified by the experiment. The light path of double beam, the characteristic of fiber connector and flexible configuration are realized in the double-beam miniature spectrometer.

In order to overcome the problem caused by similar spectrum of the adjacent pixels, the clustering algorithms were introduced into the hyperspectral imagery abnormal target detection. A new algorithm using principal component quantization and density estimation on EM clustering was proposed in this paper. Using EM algorithm to cluster hyperspectral spectrum vectors in the high dimension space, the relations between adjacent pixels in spatial space were be represented by the relations inside or between classes. According to the theory that the abnormal target pixels would spread around the edge of the classes, abnormal target was detected in the unit of class to effectively avoid abnormal point information flooded. And this algorithm achieved good detection effect. For the requirement of EM algorithm initialization is sensitive, in related with the first principal component information of the imagery dataset, EM clustering algorithm was initialized by vector quantization and density estimation method. This can reduce the problems caused by initialization of EM clustering algorithm, and improve the detection effect of the algorithm and computation efficiency. With simulated and real AVIRIS hyperspectral dataset used in simulation experiment, the results show that the proposed anomaly detection algorithm is obviously superior to the traditional detection algorithm.

The problem of shape target detection was formulated as a binary classification problem of each pixel under Markov Random Field (MRF)theoretical framework and the adaptive neighborhood system of MRF was introduced. Firstly, the factors that cause the changing of target shapes were analyzed and classic target shapes presented on obtained infrared images were concluded. Secondly, the classic shapes were used as templates while establishing the new neighborhood system of MRF. Thirdly, to achieve the optimal detection performance, a criterion function for adaptively selecting the proper neighborhood for each pixel was proposed and at last a new potential function using finite difference operator was proposed for the classification of target and background at each pixel. For the usage of adaptive neighborhood system, the proposed algorithm has following advantages: further reduction of the threshold crossing rate of target detection in single image frame while maintaining the target detection rate and better preservation of target shape details than algorithms using classic neighborhood system of MRF. By simulations and experiments, the results show that the proposed algorithm can optimally detect targets under various image Signal-to-clutter ratios, and perfectly protect target shape details.

Due to the limitation of the laser modulation technique and the interference of the atmospheric turbulence, inverse synthetic aperture ladar (ISAL) has the problems of azimuth Doppler ambiguity and echo noncoherence when imaging a spinning target. Meanwhile, the large relative rotation angle of the spinning target results in a poor correlation between adjacent echoes that makes the present phase error compensation methods fail to work. Hence, the traditional coherent integration algorithms are incapable to obtain a well-focused 2-D image. In order to solve these problems, the envelope of the echo is considered for imaging. The ISAL imaging geometry of a spinning target is established, and the characteristic of the echo is analyzed. An ISAL imaging algorithm based on the generalized radon transform (GRT) is proposed for spinning targets. Firstly, the spinning velocity of the target is estimated by autocorrelation with the range-compressed envelope in the range-slow-time domain. Then, noncoherent integration with the range-compressed envelope is realized by using the GRT, and hence the 2-D image of the spinning target with high resolution is obtained. Since the phase is not used in imaging, the phase errors are avoided. Simulation results show that, the traditional range-Doppler algorithm fails to work, while the proposed algorithm successfully obtained a well-focused ISAL image in the case of low signal-to-noise ratio, Doppler ambiguity, and echo noncoherence.

In order to solve the problem that infrared polarization image cannot be suitable for classifying targets, by analyzing the infrared polarization image at each polarization angle of redundancy and complementarity, a pattern recognition algorithm is proposed that is based on Fuzzy C-means clustering algorithm and edge extraction algorithm. Fuzzy C-means clustering algorithm and edge detection algorithm are respectively used for image processing, and then weighting method for image fusion is used to obtain the final result. The full accuracy, production accuracy and user accuracy of the obtained image are calculated and compared with the proposed algorithm and Fuzzy C-means clustering algorithm. The results show that the proposed algorithm is able to identify man-made targets clearly and classify them, which makes computers and observers recognize and classify the objects in the picture. Infrared polarization image pattern recognition makes infrared polarization imaging widely used in the area of detection, and it can detect man-made object under the background of mixed natural and obtain highly accurate results effectively and quickly.

The effects of the atomic motion and the field-mode structure, on the evolution of the linear entropy of the atom were examined in the Jaynes-Cummings model by means of linear entropy when the field was initially prepared in Schrdinger cat state and squeezed coherent state. It was found that the degree of entanglement between the atom and the field is very sensitive to the squeezing parameter. It was shown that the atomic motion leads to the periodic evolution of the linear entropy and an increase in field-mode structure parameter results in shortening of the evolution period of the linear entropy and decreasing in the amplitude of the linear entropy when the field is initially in Yurke-Stoler state. Furthermore, the increase of field-mode structure parameter leads to the shortening of the evolution periodicity of the linear entropy, but has no effect on the amplitude of the linear entropy when the field is initially in even coherent state.

A scheme of quantum dialogue was proposed based on four-particle omega state. First, Bob, a communication party, prepared a four-particle omega state, encoded classical secret information by performing an approprite unitary transformation on two particles of the omega state and sent them to Alice. Then, Alice also performed an approprite unitary transformation on the two particles to encode secret information and sent back to Bob. Finally, Bob made joint measurement to final omega state, and announced the results. Alice and Bob, two parties of communication, could decode classical information the opposite one transfers, according to the announced measurement results and itself unitary operator, making the bidirectional communication come true. The result shows that the communication scheme only transfers two particles and makes unitary operations twice, and two communication parties get the opposite four-bit classcical information. It realizes both two-way communication of classical information with quantum channel and dense encoding to improve the communication capacity. The Eve′s intercept-resend attack can be found in time by inserting decoy photons in the system to raise the security of communication system.