A method of compensating dispersion precisely in Fourier domain optical coherence tomography was proposed, based on the combination of the measurements of the values of the dispersion with the rapid scanning optical delay line. Firstly, a plane mirror was placed in the sample arm, and the values of differences of the group delay dispersion were measured while moving the grating in rapid scanning optical delay line to introduce differentdispersion. When the measured value was zero, the dispersion was matched. The new method was compared with the one based on the full width at half maxmum of point spread function. It was found that the measuring error of full width at half maxmum was 4.43% while proposed method was 0.76% and the sensitivity of full width at half the maxmum decreased from 92.105 3 fs2 to 1.344 7×103 fs2 while it remained 165.789 5 fs2 in the proposed method. The results show that this method is more accurate and sensitive than the full width at half the maximum. In addition, the direction of grating movement can be determined by the sign of the measured value of dispersion difference, which may be used to compensate dispersion. Finally, the experimental data show that the axial resolution after compensating dispersion nearly approaches the theoretical value.

The nonlacalized interference in multiple-beam interferometer was studied. The reflex light intensity distribution function was obtained. The numerical analysis shows that as the reflection coefficient of specular surface increase, the stripe of the nonlocalized interference becomes sharp, resolution ratio is improved, while the half-value width decreases; narrow bright stripe nearby the dark stripe exists at low interference order with higher reflection coefficient of specular surface; the amplitude of the bright stripe gradually decreases to the background intensity; the oscillation becomes more obviously at lower interference order; the first reflex light provides an uniformity background, and the second reflex light makes a double beam interferes with the first reflex light; with the introduce of more reflex lights, the dark stripe becomes sharp and many low intensity interference stripes appear between neighbour dark stripes; the low intensity interference stripes between the dark stripes disappear and narrow bright stripes nearby the dark stripes arise while with sufficient reflection lights. At last, the filter based on the multiple beam nonlocalized interference theory was investigated. The filtering bandwidth of the nonlocalized interferometer can be reduced notably by increasing the reflection coefficient of specular surface. While changing the mirror spacing, the filtering wavelength and the interval can be adjusted flexibly.

The joint scheduling problem of tasks and network communication is a non-deterministic polynomial hard(NP-hard) problem in the optical grid. In order to further extend the list scheduling, an iterative scheduling algorithm was proposed to optimize the scheduling length. Firstly, an initial task list was sorted according to the priority called bottom level of each task. Then the scheduling order was modified according to the updated time-weight of tasks by estimating the actual communication cost between tasks. The simulation results show that the proposed iterative scheduling algorithm can effectively improve the scheduling length in the majority of the cases, especially for data-intensive applications in task scheduling.

Network coding was introduced to optimize the selection of multi-core nodes and transmission of multicast information. A method was proposed to construct optical multicast routing based on multi-core node shared tree and network coding to decrease the wavelength consumption and to improve the performance of network load balance. The network coding candicate core node set was produced by deleting the nodes which lead to the source node bypass loop. The network coding core nodes of multi-source shared trees were determined by utilizing the heuristic matrix operation to achieve the minimal core nodes covering the maxmial source nodes. Then, the network coding was applied to transmit the information between the core nodes and destination nodes to reduce the wavelength consumption. And, the wavelength assignment and edge disjoint path were discussed for the multi-core node shared tree. Comparing with the multicast routing with single-core shared trees and singel-core shared trees based on the network coding, the simulation results show that the multicast routing with multi-core node shared trees based on network coding can get the least required wavelenght and achieve the best performance of network load balance.

An improved fundamental frequency mixing phase generated carrier demodulation algorithm was proposed. By introducing a direct current filter, the improved algorithm could overcome the limitation that the traditional fundamental frequency mixing phase generated carrier demodulation algorithm was only suitable for small-signal. The improved algorithm retained the useful components of the interference signal and reduced the sampling frequency of the system by introducing an anti-aliasing filter. The up-limit amplitude of the demodulated signal was maximized when the sampling frequency was fixed. And the hardware implementation scheme to realize the improved algorithm was presented. Simulations were carried on the Matlab platform and the digital signal processor simulation platform code composer studio prove the feasibility of the improved algorithm.

For space-ground optical communication, computational formulas for maximum acquisition time, mean acquisition time and acquisition probability were presented which were used by the multiplex raster spiral method. The simulation model of acqusition were established, the relation between the acquisition time and acquisition probability were analyzed, and effects which caused by the relative velocity of air communication platform to acqusition simulation were also disscussed. The simulation results show that when acqusition area of uncertainty of communication terminal is 50 mard, scanning overlap factor is 0.12 and signal-to-noise ratio of acqusition detector is more than 6, the acquisition probability of space-ground optical communication system is better than 95%, and the maximum acquisition time is about 36 s and the mean acquisiton time is about 12 s.

A distributed feedback (DFB) fiber laser with a ratio of backward to forward output power of 100∶1 was composed by a 45mm length asymmetrical phase-shifted fiber grating fabricated on erbium-doped photosensitive fiber. Forward output laser was amplified and population inversion was got by using a certain length of Nufern EDFL980-Hp erbium-doped fiber to absorb surplus pump power after the active phase-shifted fiber grating. Using OptiSystem software, the best fiber length of the EDFL to get the highest gain was simulated. In order to keep the amplified laser with narrow line-width and low noise, a narrow-band light filter consisted of a fiber grating with the same Bragg wavelength as the laser and an optical circulator was used to filter the ASE noise of the out-cavity erbium-doped fiber. The designed laser structure sufficiently utilized the pump power, and a fiber laser of 32.5 mW output power, 11.5 kHz line width, and －87 dB/Hz relative intensity noise (RIN) at 300 mW of 980 nm pump power was gave.

The direct modulation performance of a reconstruction-equivalent-chirp distributed feedback laser diode was experimentally investigated. The frequency response and the dynamic range were measured and analyzed. The influence on the direct modulation performance caused by the bias current of the laser diode and the RF power of the modulated signal was also investigated. Error-free transmission of a 300 Mb/s baseband signal together with a 1.562 5 Gb/s ultra-wideband signal in a 10 km single mode fiber using the distributed feedback laser diode was achieved. The power penalties of the baseband signal and the ultra-wideband signal were 1.98 dB and 0.92 dB, respectively. The distributed feedback laser diode can find applications in the wavelength-division-multiplexed passive optical network based hybrid networks.

A method for measuring liquid volatilization by using fiber loop ring-down technique incorporating an etched fiber was analyzed and experimentally demonstrated. The transmissivity of the single mode fiber were deduced according to the coupled-mode theory; the energy distribution over the cross section of the fiber was numerically analyzed by using the finite element method. The results showed that, thinner the diameter of the single mode fiber, larger would be the extinction coefficient, thus leading to more energy loss in the fiber core. The etched fiber with length of 1 cm and diameter of 26 μm was chosen through analyzing. Through the relationship between ring-down time and concentration of ethanol, the changement of ethanol′s concentration in ethanol-glycerol solution was obtained during volatilization at 30℃. The experimental results show that the concentration of ethanol decreases exponentially with the time which is in good agreement with the theoretical derivation. Since the refractive index of liquid which varies linearly with the concentration increases with the volatilization of ethanol, leading to stronger evanescent wave absorption, as a result, the ring-down time decreases. In addition, this method has very low temperature cross-sensitivity.

In order to overcome the low recognition rate of “cat-eye” effect target and ineffective distinction of car lights and telescopic sights, a recognition method of “cat-eye” effect target based on texture character was proposed. Firstly, car lights were removed by filtering the difference images with the column-mean -subtraction, EHPF and median filter. Secondly, the discrimination was carried out in terms of the features of the shape of the “cat-eye” effect target. Finally, according to the texture features of the “cat-eye” effect target, membership vectors were constructed, and the fuzzy comprehensive evaluation was used to recognize the “cat-eye” effect target under the maximum membership degree principle. Experimental results indicate that the proposed method reduces false alarm rate and improves recognition rate compared with the shape-frequency dual criterions (SFDC) method.

In order to overcome the effect of the atmosphere turbulence and reconstruct the high resolution image of the extended object, an improved speckle imaging algorithm based on bispectrum truncation method was proposed. There are two problems in phase recovery using the conventional algorithms: one is the enormous amount of the bispectrum data, and the other is the complicated computation. The improved algorithm combines phase recovery with the bispectrum computation by the Hermite symmetry of image and the look-up table technique. Only the bispectrum in neighbored domain of a spatial frequency below the cut-off frequency are calculated and some constraint is added to the bispectrum coordinates, therefor the improved algorithm makes a dramatic reduction in the bispectrum data. The coordinate look-up table, which is applicable to the two neighbored quadrants in Fourier domain, describes the computation sequence of the bispectrum and the phase recovery. The look-up table makes the phase recovery become simple, for the bispectrum and the object phase can be obtained just in order of the look-up table. The results on simulated demonstrate that the improved algorithm is able to recover the object phase correctly with a 24% reduction in the bispectrum data. The phase spectrum obtained by the improved algorithm displays the profile and structure of the object after inverse Fourier transform. The results on real world astronomical data demonstrate the improved algorithm can restore the image with higher resolution than the original image and it obtain almost the same final image as bispectrum truncation method with less computation time.

A method based on the median filter and gradient technique to detect the small target in infrared image is presented. It firstly uses median filtering to smooth the infrared image, followed by making use of the background subtraction technique to process the original image and the median filtering image. Based upon that, a gradient sharpening method is applied to strengthen the marginal information of the residual image. Finally, a binary technique is utilized to highlight the target. With the complementary effects of the median filter and gradient technique, the infrared dim target can be detected effectively. The results of the simulation have shown the effectivity of the presenting method.

High speed optical switches are critical components in the realization of light transmission path switching technology for optical networks. Current Mach-Zehnder Interferometer type optical switches either occupy larger footprint or have a longer one-dimensional scale. A Silicon-based rectangular Mach-Zehnder 2×2 thermo-optical switch unit is proposed for the highly integrated applications. The novel optical switch consists of trench-based nanophotonic frustrated total internal reflection couplers and total internal reflection mirror-based 90° waveguide bends. The switching operation is achieved through thermo-optical effect of Silicon. As a design example, the ~84 μm long L-shape phase-shifting arm is determined for π phase shift when its temperature of the arm is increased up to 50℃. A switching voltage of 3.5 V on a NiCr metal electrode heater is obtained by using finite element method. The voltage may lead to the increase of 50℃ for an optical ridge waveguide in the L-shape phase-shifting arm. The time response of the device is simulated as a function of different upper dielectric thickness of SiO2, and the response time including raise time of 35 μs and fall time of 48 μs is optimized for the device with 0.5 μm thickness of upper SiO2. The switching function of the thermo-optical switch is simulated and verified by Rsoft FullWAVE software, a finite difference time domain method. The chip size of 56×38 μm2 is designed for the Silicon-based rectangular Mach-Zehnder 2×2 optical switch. The device is ultra-compact, and its configuration is beneficial to extend at two-dimensional directions, making them attractive for Silicon-based high dense photonic integrated circuits and on-chip optical interconnect system.

A type of surface plasmonic waveguide applied in the optoelectronic integrated circuit was designed. It was simulated by using three dimensional full-vectorial finite-difference time-domain method. The dependence of distributions of electromagnetic field, effective index and propagation length of the fundamental mode supported by this waveguide on geometrical parameters were presented. The coupling length, maximum transfer power and crosstalk between two waveguide structures were studied. The simulation results showed: the light field is highly confined to the core layer area, and the energy restriction factor of light field is higher when the metal structure apex angle is 135 degrees; at a given metal, the effective index will decrease as the core width of ridge increases, meanwhile the propagation length of the fundamental mode will increase as radius of ridge increases; as the waveguide distance increases under given core width, the coupling length between two waveguide structures will increase and the maximum transfer power and crosstalk will decrease.

A silicon-based microring modulator and a narrow-band optical Gaussian filter were employed to generate optical duobinary at 10 Gbit/s. Performance of the optical duobinary signal was analyzed in terms of laser linewidth, laser operating wavelength, fiber length, and filter bandwidth using Optisystem and Matlab. Simulation results show the optical duobinary signal is sensitive to laser linewidth and wavelength. At the BER of 10－9, the maximum transmission distance of the optical duobinary format can be up to 60 km with the optimal Gaussian filter bandwidth of 8 GHz.

To realize the electromagnetically induced transparency-like effect on the silicon waveguide, a double ring resonators structure based on silicon-on-insulator was designed and fabricated. By post processing of thermal oxidation annealing under the temperature from 300 ℃ to 1 200 ℃, wet chemical etching with buffered oxide etch and N2 annealing under the temperature of 1 000 ℃, the electromagnetically induced transparency-like curve via the vertical grating coupling method was observed. The effect of different distance between the two resonators to the electromagnetically induced transparency-like curve was analyzed. The Q factor of a single ring in this fabrication was 5.1×104. The measured spectral width of the transparency window was as narrow as 500 MHz.

A novel neodymium pentafluoropropionate complex, Nd(C2F5COO)3·Dmbp (Dmbp: 4,4′-dimethyl-2,2′-bipyridine), was synthesized and characterized by elemental analysis, Fourier transform infrared spectra. The optical porperties of the liquid medium which was abtained by dissolving Nd(C2F5COO)3·Dmbp in N,N-dimethylformamide was investigated. Based on absorption and luminescence spectra of Nd(C2F5COO)3·Dmbp in DMF, as well as Judd-Ofelt theory, the three intensity parameters Ωt(t=2,4,6) and emission cross-section were calculated and analyzed The emission cross-section of 4F3/2→4I11/2 fluorescence transition of Nd3+ion was 5.2×10－20cm2, and comparable with some laser glasses, which indicated good radiative properties of this neodymium complex in liquid matrix. According to the photoluminescence spectra, Nd(C2F5COO)3·Dmbp complex will be a promising material for optical gain material in the future.

The transmission matrix method theory was applied to study the localized electric field inside the one-dimensional photonic crystal quantum well of double barrier (AB)k(CD)m(DCD)n(DC)m(BA)k. The result shows that a strong electric field, which will become stronger as approaching the well center, is distributed inside the photonic crystal quantum well of double barrier. When barrier layers of the well thicken, especially the inner and outer barrier layers thicken simultaneously, the localized electric field rapidly strengthens, and the react is most obviously in the center of the well. When the well layers widen, the closer to the well center, the stronger the electric field is. But when the well layers widen by even multiple, the localized electric field in the well center remain steadily in maximum value. While when the well layers widen by odd multiple, the intensity of the inner localized electric field tend to be the maximum produced by even multiple widening. The distribution characteristic above provides instructions for quantization effect mechanism and the intrinsic formation factors of discrete transmission spectra of photonic crystal quantum potential well, and for applied designing of quantum optics devices.

A high sensitive single photon counting system with a new photomultiplier tube as a sensor was developed to detect the ultra-weak biophoton emission from wheat, which fulfilled the absolute quantity detection of photon emitted from wheat in the storage condition. An illuminating system for the delayed bioluminescence was designed with a proper LED array. The relation of the illuminatingintensity and the intensity of photons was established through calculation of equivalent single photon energy and lumen efficiency for white LED, and the quantity comparison of illumination intensity and delayed bioluminescence intensity was realized. There existed different illuminating saturation time at different illumination. The accurate calibration of time for testing data of the delayed bioluminescence was fulfilled by accurate control of the illuminating time and data acquisition time, and also setting the delayed time parameter in fitting function. The ultra-weak delayed bioluminescence from wheat was detected and analyzed systematically, and the results show that the intensity curves of delayed luminescence from wheat follows a hyperbolic attenuation trend, and the duration of the luminescence lasts for several hours. Subsequently, the hyperbolic function fitting of testing data was carried out and the changing regularity of fitting parameters was discussed.

The best dark adaptation time of Chlorella pyrenoidosa, Snedesmus obliquus and Microcystis aeruginosa stressed by Cu2+ was investigated by chlorophyll fluorescence measurements. Photosynthetic fluorescence parameters were measured under light and dark conditions. Measurement time in light condition were 30 s, 1 min, 3 min, 5 min, 10 min and 20 min as well as in dark condition. Significant difference of dark adaptation time was analyzed mainly based on the photochemical quenching parameter and t-test method. The results show that compared to light, the potential maximum quantum efficiency values of three algaes increase slightly under dark adaptation and the yield value remaines the same; the photochemical quenching parameter value and non photochemical quenching parameter value increase significantly as time goes on; the maximum photochemical quenching parameter value of Microcystis aeruginosa achieves when the algae is treated under light in 1 min, and Microcystis aeruginosa needs no dark adaptation may because blue-green algae state conditions shifts under dark adaptation; the dark adaptation time of algaes are not better by long time treated, and the best dark adaptation time of Chlorella pyrenoidosa and Snedesmus obliquus are 5 min and 10 min respectively. All of these will provide a reliable basis for further studies on the inhibiting effect of toxicant on algaes.

Spontaneous fluorescence, as a kind of life information, can reflect cellular metabolism and DNA synthesis, so it plays an important role in evaluation of drought resistance during germination of crop seeds. To establish the method of drought resistance evaluation based on spontaneous fluorescence in germinating stage of maize, change of spontaneous fluorescence was studied in Wanrui No. 168 and Yandan No. 8 under water stress induced by PEG-6000 solution with the osmotic potential of －0.1 MPa and －0.3 MPa. The results show that spontaneous fluorescence of the two cultivars increase with the germination process, while is inhibited by water stress. The inhibition effect of spontaneous fluorescence to Wanrui No. 168 is more obvious than that to Yandan No. 8 under the same intensity of water stress, and the higher the intensity is, the greater the interspecific difference will be. It is also found that relative change rate of spontaneous fluorescence named RSL during germination of seeds can reflect the dynamic process sensitively of response, adaptation and damage to water stress in life activities of seeds, and it is nondestructive and sensitive using RSL to evaluate drought resistance in germination period of maize.

A scheme of star simulator spectrum simulation based on mixed light source of xenon lamp and halogen tungsten lamp for foundation was proposed. The wide spectrum light source array was composed of xenon lamp and halogen lamp. After color temperature simulation and module controlling, the light entered into the integrating sphere to provide ideal light source for star simulator, and spectral curves were stimulated at specific color temperatures of nearly 3 900 K, 4 800 K, 6 500 K. Through the simulation of the blackbody spectrum curve of standard, the optimal transmission rate was calculated for each band filter, and the transmission rate of interval in tiny brand was combined for getting the transmission rate which satisfied the design index. The simulation results showed that, by using the proposed scheme, the small band matching simulation error would be less than 10%, with the relative area method verification matching error is within 4%, and this scheme provides an effective method for the star simulator to simulate spectral curves of the specific color temperatures.

The push-broom infrared remote sensing imaging technology plays an important role in the high-resolution earth observation, but the internal development of infrared imaging technology is baffled by the technical level of long-linear array infrared detector. To achieve high-resolution push-broom infrared imaging, a method by the use of line-plane-switching infrared fiber bundle was proposed. In this method, the linear array end of the imaging fiber bundle was used as a long-linear array infrared detector and the plane array end of the bundle was coupled by a mature small infrared focal plane array with the single fiber corresponding to the pixel of the focal plane. The main problem of the design of the objective lens and coupling lens in the infrared fiber image transmission system was analyzed. And a system with an objective lens and an coupling lens was designed for a kind of fiber bundle which has 4 000×6 pixels on its entrance end and has 160×150 pixels on its exit end. The imaging performance of the objective lens and coupling lens were close to the diffraction limits while meeting special requirements of infrared fiber image transmission system. The presented design is a better reference for this kind of fiber image transmission system. The mean transfer function was introduced for evaluating the MTF of the overall optical system, and the simulation results show a good image quality.

To overcome disadvantages of modeling application and uncertainties of the reliabilities of the existing LED light source model and others, the reliability of light source modeling was studied for distribution design on compact LED. By optimizing different models with different lens and LED diameters ratio, the difference of three modeling methods including point light source model, surface light source model and ray-file light source model in the system was compared and analyzed. Taking the ray-file light model as the standard, an optimal design and analysis computation of the optical element were carried out when the beam angle was ±15°, ±10°, ±5° for different lens and LED diameter ratio, by choosing the conventional compact total-reflection lens as the second optical element and using three modeling methods including point light source model, surface light source model and ray-file light source model. Experimental results indicate that surface light model can take place of the ray-file light model when the deviation of utilization of energy is within ±2%, the beam angle is ±15° meanwhile lens and LED diameter ratio is more than 36∶2. It happens too when the beam angle is ±10° while the lens and LED diameter is 44∶2. In the case that the beam angle is ±5°, and the lens and LED diameter ratio is 82∶2, the ray-file light source model can also be taked place by the surface light source model. However, the ray-file light source model cannot be replaced by the point light source model in any case when the deviation range falls into ±2%. The results provide basis for the system design and solve the problem of time wasting of assisting optimal design by using ray light source model, which will be of great benefit to the future research.

Clear display of high-resolution pictures on LED displays is one of key problems to be solved in displaying quality improvement areas. Aiming at the features of LED displays, such as low-resolutions, high production costs and flexible in pixels arrangements,applications of sub-pixel sampling technique were analyzed. Firstly, the mathematical models of LED displays were set up from the information processing perspective. Then under theluminance and chrominance separated space such as YUV color space, by frequency analysis,it was demonstrated that aliasing in display images can be transferred from the luminance to the chrominance which was insensitive to human eyes.The Nyquist limits of them in all directions were effectively expanded. The simulation results on the real LEDdisplay further show the improvement of displaying quality by sub-pixel sampling.

A novel optical phased array based on Liquid crystal was studied, and its structure scheme feature is a planar wavegude with liquid crystal cladding. Propagation, output diffraction characteristics and other peformances of liquid crystal-optical phased array were all studied and analyzed according to Frank-Oseen continuum elastic theory and grating diffraction theory. The research results show that the electrical control phase delay of the device is quantitatively analyzed to obtain larger optical path difference; the period number of electrodes, elcetrode width, spacing of adjacent electrodes of the device are of great effect on the intensity distribution and the full width at half maximum of output diffraction beam, and the feasibility of beam scanning is also demonstrated; this novel strcture could improve the response time for an order of magnitude as well as wavelength dispertion of the device. These theoretical basis and technical design basis are favour of developing this novel liquid crystal-optical phased array in the future.

A relationship between the focal length of the double-liquid lens and other parameters such as the applied voltage, the dielectric constant of the double-layer dielectric film and the film thickness etc. was analyzed based on the cylindrical double-liquid lens model with double-layer dielectric film. And the tantalum pentoxide film with a high relative dielectric constant as an inter-layer dielectric layer and a relatively low dielectric constant for the outer dielectric layer which is a waterproof layer, were used to reduce the driving voltage of double-liquid variable-focus lens. The effects of the thickness of two dielectric films and the thickness matching relationship between them were described on the variable-focus range and the driving voltage of the double-liquid lens. The optimum film parameters were obtained on the condition of minimizing the drive voltage of the lens at a certain variable-focus range. Simulation results show that double-liquid lens can be driven by relatively low voltage when the waterproof film thickness is much thinner than that of the high dielectric constant thin film. Within a certain variable-focus range, the driving voltage of the double-liquid variable-focus lens can drop to 10 volts. But double-liquid lens will not make efficient use of the high dielectric property of the tantalum pentoxide film, if the thickness of waterproof film is equivalent to or higher than that of the high dielectric constant thin film.

A novel flow sensor based on double fiber Bragg grating was designed and made to overcome the temperature cross-sensitivity problems of traditional single fiber Bragg grating sensor. The elastic umbrella structures were employed as conversion elements and silicone resin as encapsulating material, and an effective solution double-grating structure was taken. The temperature natural compensation and its measuring sensitivity were improved. Finite element was used to optimize the structure of the flow sensor. Its structure and the fluid field around it were simulated and analyzed by ANSYS software within the velocity of 1~20 m/s and the sensitivity of the sensor was calculated. The advantages of silicone resin as encapsulating material and its temperature sensitivity were verified through experiments. The performances of the sensor packaging before and after were tested and the reflection spectrum signal characteristic values of the fiber grating were extracted with good linearity, high response sensitivity of 1.71 nm/kg and 0.103 nm/kg, respectively. The experiment results show that the flow sensor owns simple structure, good linearity and high sensitivity.

The fluorescent spectrometry is a common method to detect the concentration of SO2 in the atmospheric monitoring. The detection system adopting double light paths can eliminate the noise jamming from the light source and light path. However, background noise produced by photoelectric converting device under the laser irradiation will also affect the accuracy of quantitative analysis. Empirical Mode Decomposition (EMD) filtering algorithm was used to reduce various kinds of noise existing in the detection, which could retain the useful original signal and reduce the noise effectively. The simulation results show that for the sulfur dioxide concentration detection system, using EMD de-noising, the Signal Noise Ratio (SNR) increases to 204.273 6, and the Mmean Squared Error (MSE) is 0.007 0. Compared with the wavelet de-noising method, the effect of EMD detection is much better. Finally, the signal processed with the two signal methods were applied to the gas detection system. From the experimental data of the linear relationship, it can be concluded that the EMD method applied to the proposed concentration detection system is feasible.

The model of near infrared spectroscopy of water content in oil is complex, nonlinear, and difficult to be specified with mathematical methods. Uninformative variables elimination was applied to the extraction of effective wavelengths, and fuzzy C-means clustering algorithm was applied to obtain the input space location and the clustering center. By identifying the consequent parameters with recursive least-squares method, Takagi-Sugeno, a fuzzy model of near infrared spectroscopy of water content in oil, was established. This identification algorithm was compared with Partial Least Squares model and tested by experimental data. The results indicate: the Takagi-Sugeno model, constructed by a total of 34 variables selected by uninformative variables elimination, can accurately reflect the relation between near infrared spectral data of oil and moisture content; the correlation coefficient the model predicted for the samples from validation set is 0.964 6 and the root of mean square error is 1.531 2×10－4, which are satisfactory. The experimental results verify that it is feasible to detect the water content in oil by means of near infrared spectroscopy, which also offers a new alternative approach for the on-line monitoring of other contamination content in oil.

Effects of photoinduced electron transfer on the Raman spectroscopy of Alizarin (Alz) dye-sensitized TiO2 nanoparticles were investigated by theoretical and experitmental methods. The results indicate that red-shift of absorption spectra and quenching of the fluorescence spectra are attributed to the photo-induced electron transfers from the excited state of the absorbed Alz dye molecules and the charge transfer complex (Alz/TiO2) to the conduction band manifold of the TiO2 nanoparticles. The resonance Raman spectroscopy of the Alz dye-sensitized TiO2 nanoparticles reveals that the ultrafast photoinduced interface electron transfer significantly enhances the C=C, C-O, and C=O bond stretching vibrations of the absorbed Alz dye molecule at the interface.