A new all-optical switching device, which is constructed by connecting an erbium-doped fiber with two symmetrical long-period fiber gratings (EDF-LPFG) pair was proposed. The transmission spectra of signal light for different pump powers in cross-phase modulation are simulated, and the transmissions of signal light as a function of the pump power under different effective index change and absorption coefficients are also discussed. The formula for threshold switching power of EDF-LPFG pair switch is derived. Compared with the single LPFG switch, the threshold switching power of EDF-LPFG pair switch is dramatically reduced by 5 orders, less than 25 mW. Such kind all-optical switch is hopeful to be used in future all-optical communication.

The residual intensity modulation (RIM) of a Y-waveguide modulator used in an interferometric fiber optic gyroscope (IFOG) directly impacts the stability of IFOG’s scale factor, RIM is tested, and based on above results, RIM’s characteristic formula is given. RIM deteriorates the scale factor stability of an open-loop IFOG when there is inner static voltage in the modulator. The deterioration is proportional to the inner static voltage and α2－α1, the difference of RIM’s coefficient between the two branches of the modulator. For a close-loop IFOG, RIM deteriorates its scale factor stability not only directly but also indirectly through 2π-replacement loop that determines the peak voltage of modulation signals. The deterioration is proportional to α2－α1, but inversely proportional to the modulator’s electro-optic coefficient. A demodulation method employing three time slots is suggested that can completely remove the direct and indirect deterioration.

In this paper, we discussed the theory of the double loop optical buffer (DLOB) based on collinear 3×3 coupler, and put forward 4 types of read-write methods in DLOB: positive optical pulse control, negative optical pulse control, positive electrical pulse control and negative electrical pulse control. We also introduced 0.89π-phase adjusting method. A double loop optical buffer experiment system had been built in which the loop was 63 meter length. The experiment had not only proved 4 types of read-write methods, but also demonstrated that the negative optical pulse control was the only doable method when the optical packets traversed in DLOB exceeding 20 round-trips. In order to suppress the noise, we can enhance the bias current of the control light, the “base” power of the control light was required from 500 μW to 1 mW by the way of enhancing the bias current of the control laser, or injecting a direct light whose wavelength was different from the signal light and control light. In the electrical pulse control, it also needed injecting a direct light to suppress the noise.

Using photonic crystal fiber (PCF) fabricated by ourselves, the changes of frequency spectrum and remarkable anti-Stokes phenomenon caused by a series of nonlinear effects have been observed through increasing the central wavelength λ0 of pump pulse to mainly make it located in the abnormal dispersion region. By adjusting the incidence angle of the coupling light, the first-higher-order mode is output from PCF. In the process of λ0 achieving and exceeding 820 nm which is the zero-dispersion wavelength of the first-higher-order mode, the working region of pump pulse is in abnormal dispersion region and the energy transform from pump pulse to anti-Stokes wave is increasing gradually. Especially when λ0 exceeds 860 nm, the intensity of anti-Stokes wave is five times more than the intensity of residual pump energy, and the transform-efficiency exceeds 80%.

Pressure phase sensitivity is one of the most important performance indexes for the fiber-optic hydrophone. Maximizing the pressure phase sensitivity is a basic principle in the fiber-optic hydrophone design. In this work, the orthogonal experimental method was applied to optimize the design of the pressure phase sensitivity of the fiber-optic air-backed mandrel hydrophone. A general three-dimensional (3D) theoretical model was created, and nine fiber-optic hydrophones were designed and made for orthogonal experiment. Several factors which affect the sensitivity were analyzed in the experiments, and the results validate the theory. The optimized designs are also gained by the analysis of the results to maximize the pressure phase sensitivity.

Starting from the extended nonlinear Schrdinger equation in which the quintic nonlinearity effect is included, the evolution and splitting process of continuous optical wave which is amplitude perturbed by the sine optical wave into ultra-short optical pulse trains in optical fibers is numerically simulated by adopting the split-step Fourier algorithm. The effects of the quintic nonlinearity and the modulation period of the sine optical wave on the generation and evolution of ultra-short optical pulse trains and the corresponding spectra are investigated. The results show that, in comparison with the case of cubic nonlinearity only, the positive quintic nonlinearity can shorten the optimal fiber length required to form the pulse trains and make every pulse shorter in width and higher in peak power, while the negative quintic nonlinearity takes the opposite. The sine modulation period may influence the pulse repetition rate and the optimal fiber length. With the increase of the propagation distance, every single pulse may split into two and even three sub-pulses. Moreover, some small pulses with weak peak powers may appear among the main pulses. In terms of the frequency spectra, the positive (negative) quintic nonlinearity can make the number of frequency components increase (decrease) and the spectral width wide (narrow). Depending on whether the main pulses split and the small pulses exist or not, the spectra take on different shapes.

Soliton pulse compression based on dispersion decreasing fiber (DDF) with increasing nonlinearity is proved according to the theoretical deduction. Some characteristics of compressed pulse such as compression factor and pedestal energy are analyzed through the symmetrized split-step Fourier method. It is found that the final compression ratio achieved by a nonlinearity increasing fiber is the highest while the ratio achieved by a dispersion decreasing fiber is actually the smallest. Through comparing with dispersion decreasing method, the results show that almost the same compression ratio and lower pedestal energy can be achieved with half of the original fiber length by nonlinearity increasing method.

Based on the principle of Fresnel reflection and the optical time-domain reflectometry (OTDR) techniques, a quasi-distributed optical fiber temperature sensor which can achieve accurate temperature measurement is presented. The sensor head is made of two face-to-face single-mode optical fibers whose end surfaces have been polished. The gap betwen two fibers is filled with a temperature-sensitivity material. The temperature variation causes the refractive index of the filling material to change, and then the intensity of Fresnel reflection will be changed. A sensing scheme with three sensor heads in series is investigated in experiment. The slight changes in Fresnel reflection intensity are detected by OTDR, with which the temperature sensing characteristics are studied. The experimental results show that the Fresnel reflectivity increases monotonically when the temperature varies from -30 ℃ to 80 ℃. The sensor heads have good repeatability and low insertion loss.

gratings; fiber Bragg grating; temperature compensations; packaging structure; fiber

A polarization splitter based on resonant tunneling phenomenon in three-core photonic crystal fibers (PCFs) is presented. The supermodes in three-core PCFs will be resonant if their effective refractive indices satisfy some conditions. Choosing the appropriate parameters of PCFs, one polarization light will be close to the resonant condition and the other will be far away from the resonant condition. In the end of PCFs, the light power of the two polarizations can be confined in different cores. Full-vector finite element method (FEM) is applied to calculate the effective refractive indices in three-core PCFs. Full-vector beam propagation method (BPM) is employed to analyze the performance of the splitter. Numerical simulations demonstrate that it is possible to obtain a 1.039-mm-long polarization splitter for 1.55-μm light, and the extinction ratio can reach -36.98 dB. In this case, the bandwidth with the extinction ratio better than -11 dB is 24 nm.

Multimode fiber (MMF) has become an ideal medium to realize the short distance communication of “the last kilometer”. But the severe mode dispersion of the MMF limits its transmission ability. Mode group diversity multiplexing (MGDM) is an optical multiple-input multiple-output (MIMO) technique that aims at creating independent communication channels over a MMF and using subsets of propagating modes. This technique enables to carry a number of signals and greatly enhances the transmission capacity of the MMF. A 2-input 2-output direct detection MGDM communication system over 75 m graded-index multimode fiber (GI-MMF) is designed. The excitation of different mode groups is realized by using selective mode excitation. The multiplexing and demultiplexing are realized by offset connectors and mode couplers. The transmission of an analog video signal and a non-return-to-zero (NRZ) digital signal is demonstrated with this system.

The 1+1 two-photon resonant enhanced multiphoton ionization spectra of chlorobenzene were obtained at the range of 262～272 nm on the laser mass spectrometer with heatable sample inlet system which was designed by ourselves. The result shows that the structure of the excitation spectra in this wavelength range represents the first excited state 1B2(S1) of the chlorobenzene. 14 spectrum lines were assigned. The characteristics of the photon ionization mass spectrum at 266 nm were analyzed. The non-symmetric broadening of the mass spectrometry of the 77C6H+5 was analyzed. The relationship between the signal intensity of chlorobenzene and the sample concentration was studied. It is shown that the detection limit of the chlorobenzene is about ng/L range.

A coherent ladar system for detecting vibration of a moving target was demonstrated. Micro-Doppler information was obtained to identify moving targets. Single frequency laser at 1.064 μm was input to a 4 km long single mode fiber to simulate the translation of long distance. A loudspeaker was used as the target with micro-motion. Back-scattered light from the target was collected by the receiving optics and coupled into a single mode fiber. A 3 dB fiber coupler was used to combine the received signal and the master oscillation, the output of the coupler was detected by a photodiode detector with 3.5 GHz bandwidth. The electrical signals of detector’s output were sampled by a high speed A/D card. The information of micro-motion can be extracted through the arithmetic of time-frequency analysis. With the delay of 4 km fiber, the lowest speed that the system can detect was 0.5 mm/s, the resolution of speed detection was in the order of mm/s, the resolution of frequency was in the order of kHz. Utilizing the technology of micro Doppler detection, the targets of different motion dynamics can be classified and identified.

Atmospheric refractivity has a significant impact on measurement precision in the field of photoelectric detecting. To improve the detecting accuracy of photoelectric fields, a pure rotational Raman lidar technique for detecting tropospheric refractive index is discussed in detail. Higher-lower isolating two portions of the pure rotational Raman backscattering spectrum of N2 and O2 are obtained by separating from other singles using the double grating monochromator. By caculating ratio of the two return signals, atmospheric temperature and pressure profiles are derived, as well as the measurements of air refractivity. Compared with the theoretical model, the accuracy for detecting atmospheric refractivity using the pure rotational Raman lidar is convincingly illustrated by the results of laboratory experiments. Several atmospheric refractive profiles are shown and the characteristic of refractivity is acquired at different times and on different days. The results show that the changes of the refractivity are little during the night and reach only 0.4% below 7.5 km. The fluctuation of the refractivity between different days is obvious and reaches 3.5% below 4.5 km.

The concept of the lasing lifetime of working substance in the solid dye laser is promoted to liquid dye laser, and a twin laser beam testing method to measure the lasing lifetime is presented. Nd:YAG laser with the wavelength of 532 nm is used as the pumping source to simulate the pump condition in the dye laser. Low-power continuous-wave (CW) laser with the same wavelength is used to monitor the inactivation velocity of the dye molecules under the effect of pumping laser. Semiconductor refrigeration device, PID temperature control device and the water-bath are used to realize the thermostatically control of the dye solution. Lasing lifetime of Rhodamine 6G alcohol solution is tested while laser power density varies from 6.3×103 W/m2 to 2.2×104 W/m2. The experimental result indicates that the lasing lifetime of Rhodamine 6G is inversely proportional to the pumping laser power density. The proportional coefficient can be explained as the irradiation energy accumulated on unit molecule absorption cross section, when half of the dye molecules are dissociated. It is more convenient to use this coefficient to characterize the lasing lifetime of dye.

Temperature tuning and domain period tuning optical parametric oscillator based on periodically poled MgO doped lithium niobate (PPMgLN) in singly-resonant was studied experimentally. The output spectrum of optical parametric oscillator was discussed. A laser diode (LD) end-pumped acousto-optically Q-switched Nd:YVO4 laser was used as the pump source. The double-concave cavity structure was employed as PPMgLN-optical parametric oscillator (OPO) resonance cavity. Under Q-switch repetition rate of 10 kHz and temperature of 25.4 ℃, the threshold pump power of 110 mW is gained. The signal output power of 84 mW is obtained with a pump power of 325 mW and optical-optical conversion efficiency is 25.8%. It generates signal wave-lengths tunable in the range of 1449.6～1635 nm by temperature tuning and domain period tuning of PPMgLN. The spectrum from frequency mixing between the pump and signal is obtained at 25.4 ℃. The experimental results show that the full-width at half-maximum (FWHM) of output spectrum is less than 0.5 nm.

The quasi-continuous-wave and passively Q-switched output performance of laser-diode-array(LDA)-side-pumped domestic Nd:YAG ceramic laser were studied. A uniformly side-around arranged compact LDA pumping structure with pumping length of 20 mm was used in this laser. The Nd:YAG ceramic rod was 3 mm×35 mm sized and 1% (atomic percent) doped. In a plano-plano concave resonator, with the working frequency of 1 kHz, and the output coupler transmission of 47.3%, a quasi-continuous-wave output with an average power of about 23 W at 1064 nm is obtained. The beam divergence angle is 4.5 mrad, and the slope efficiency is up to 12%. In the same condition, using Cr4+:YAG crystal as the passive Q-switcher, the 1 kHz Q-switched laser pulse is obtained successfully, when the original transmission ratio of Cr4+:YAG crystal is 60%, the pulse width (FWHM) is less than 14.5 ns ,and the dynamic to static ratio is about 40%.

Blue, top emissive organic light-emitting diodes (TEOLEDs) are hard to fabricate because of strong microcavity effects in these diodes. In this letter blue TEOLEDs based on the blue emissive material, 4,4′-bis(2,2′-diphenylvinyl)-1,1′-biphenyl, are acquired with the choice of a highly transparent cathode together with the utilization of a light outcoupling layer onto the cathode, in which microcavity effects, especially for multiple-beam interference, are restrained by decreasing the reflectivity of the cathode. It is worth mentioning that wide-angle interference is used in this letter to enhance the blue emission. The thickness of the light outcoupling layer is optimized and its effect on the electrical and optical characteristics of the TEOLEDs is also explored. As a result, the high performances such as high brightness, large luminous efficiency, and saturated chromaticity are obtained in the blue TEOLED, which are comparative to those in the bottom emissive OLEDs.

Based on the Sommerfeld diffraction integral, the anomalous spectral properties of an ultrashort pulsed Hermite-Gaussian beams propagating in dispersive media are studied , and its comparisons with anomalous spectral behaviors in free space are analyzed. It is found that the on-axis spectrum blueshift of ultrashort pulsed Hermite-Gaussian beams is relative to group velocity and pulse duration, but not to group velocity dispersion. The size of relative spectral shift in dispersive media is slightly smaller than that in free space, and the relative spectral shifts decrease with increasing the pulse duration. The location of spectral switching is relative to the transversal distance, transmission media,and the Hermite function indexes m and n, but not to pulse duration. The transversal distance of spectral switching in dispersive media is less than that in free space, the relative spectral shift slowly decreases as the order number of the spectral switching increases.

In order to discriminate and identify phytoplankton of different divisions and genuses, coiflet2 (coif2) wavelet function was utilized to extract the characteristics of the three-dimensional (3D) fluorescence spectra of 12 phytoplankton species belonging to 9 genuses of 4 divisions. The third scale vectors selected as the discriminating characteristic spectra, obviously express the distinguish characteristics of different genuses and divisions. The results of Bayes discriminant analysis showed that these characteristic spectra had average discriminating rates of 99.0% and 97.4% at the division and the genus level, respectively. Reference spectra were obtained from these characteristic spectra by cluster analysis. A fluormetric method was established by multiple linear regression resolved by the nonnegative least squares. These reference spectra identified the single species added with 10% and 20% ratios of random noise with the rates of more than 98.0% and 85.0%, respectively, at the division and the genus level. All the dominant species of the phytoplankton mixtures could be identified 100% at both the division and the genus level.

The process of thermal denaturation of the albumin was studied using dynamic speckle method based on wavelet entropy. At first, the dynamic speckle pattern sequences generated by albumin colloid during denaturing were acquired using charge-compled device (CCD) camera. Using the speckle sequences, the time history of speckle patterns (THSP) were generated. And then, with the wavelet entropy as a parameter, each row of THSP was separated into eight time windows which were decomposed by db4 orthogonal wavelet family into three levels . Thus a three-dimensional matrix of 256×256×8 of wavelet entropy was generated. Finally, this matrix was grayed into eight 256×256 patterns, which made the change of dynamic speckle signals visulized. According to the patterns, the movement properties of the protein molecule ensemble were analyzed during thermal denaturation of the albumin. The results show that this method is effective to analyze the process of movement and aggregation of protein molecules quantitatively. Experimental results prove that this method is a useful tool to investigate the motion of particles in solution.

To improve the stability during welding and extend the range of optimal technical parameter, joining of Ti/Al dissimilar alloys with V-shape and Y-shape groove respectively is performed by rectangle spot laser welding-brazing(LWB) with homogenize energy distribution. Interfacial microstructure and mechanical property of the joints with different line energies are obtained. The interfacial characteristics of the joints are investigated by scanning electron microscope (SEM) and metallographic microscopy. Results indicate that the thickness of the interfacial reaction layer, which depends on welding line energy, decreases from the top to the bottom of the joint. The unequal thickness of the reaction layer plays a great role in tensile strength of the joint. According to the results of tensile test, joints with Y-shape groove have mixed fracture and fractures in the seam and in the interface. Joints with V-shape groove have fractures in the seam and in the interface. Tensile strength of joint is 80% of the Al alloy matrix up to 290 MPa.

The phenomena in ablation process of Nd:YAG laser irradiating carbon fibers reinforced epoxy resins composites under different intensities were observed by high-speed camera. Below low intensity (50 W/cm2), composite material laser ablation is dominated by surface ablation mechanism rather than burning, and the surface slight decomposing leads to form carbon powder or flocs which slowly disperse into the air under long time irradiation; At middle intensity of 300 W/cm2 composite material laser ablation is dominated by volume ablation mechanism, inner decomposition gases spill into the air and ignite surface combustion; At high intensity 4500 W/cm2 the main mechanism of the mass ablation is surface ablation, surface layer is damaged and instantaneously burns in a very short time (0.001 s), jet of plasma forms in the center of laser spot, anisotropic heat conduction causes layered ablation and deforms the ablation zone appearance.

Experiments of high power laser welding of 1.8 mm thick galvanized steel B450LAD are performed by using a 15 kW CO2 laser. Four sets of welding process parameters are acquired through process optimization. The macrostructure and microstructure analysis, microhardness measurements and tensile tests have been done on these four welded joints. It is shown that sound welds can be achieved on 1.8 mm thick B450LAD steel at a laser power of 5.8～12.9 kW and a welding speed of 3～7 m/min. The weld section looks like an inverted trapezia and the heat affect zone (HAZ) is very narrow. Martensite structure is found in weld center and little martensite with more ferrite structure is observed in fine grained region . The hardness in weld center is 2 times as high as the base material. The welded joints show good tensile properties, and all tensile specimens are broken in the base material (BM). So these four groups of welding parameters can be used for guiding the actual high power laser welding of B450LAD at different applications.

355 nm LaF3/MgF2 antireflection coatings were prepared by thermal boat evaporation combining masking technology. Some of these coatings were annealed in vacuum. The reflectance and transmittance spectra of the anti-reflective coatings were measured by a spectrophotometer Lambda 900. Meanwhile, the spectrum stability is tested. Laser induce damage threshold (LIDT) was performed by a 355 nm laser system with 8 ns pulses. The results show that the prepared anti-reflective coatings have very low reflectance and good spectrum stability. The vacuum annealing has no effect upon the LIDT. The damage morphologies are shown as dispersive spots. Moreover, the deep analysis shows that the damage origins from the defect at the film-substrate interfaces. The JGS1 substrate has better surface condition and higher laser damage resistance, and the antireflection coatings upon it has lower surface electric-field intensity. These advantages make the antireflection coatings on these substrates have higher LIDT.

Tin-doped indium oxide (ITO) thin films were deposited on glass substrates by direct current (DC) magnetron sputtering,using an ITO target with a combination of 90％ In2O3 and 10% SnO2 in mass fraction. The effects of substrate temperature on the thin film’s transparency and resistivity were studied by spectrophotometer and four-point probe meter. The structural was analyzed by X-ray diffraction (XRD) diffractometer. The interplanar spacing and crystal grain size were calculated, the mechenical properties of films were studied.The results showed that the lowest resistivity could be got while the transparency maintained above 85% with the proper substrate temperature of 200 ℃ , that is to say there is an optimal substrate temperature to prepare the best performance thin film. It showed that the higher the substrate temperature became the better the crystallization was.

To enlarge the field-of-view of digital holographic surface measurement, so that digital holography can be applied in three-dimensional(3-D) surface measurement of large objects,based on Fresnel off axis digital holography, different part of the object is illuminated singly and their holograms are recorded respectively. The object-reference angle remains the same for each recording by precisely controlling the incident angle of the reference wave using an electronically controlled turntable. The position of the illuminated part can be obtained from the change of the incident angle of reference wave. The unwrapped phase maps are spliced to obtain the 3-D surface shape of the whole object. The 3-D surface shape of a 11 cm×19 cm plaster mouth model is measured using this method. The absolute splicing error is no more than 1.14 mm and the surface height error is about 0.5 mm. It is proved that this method is very effective in enlarging the field-of-view of digital holographic 3-D surface shape measurement and its splicing precision is as high as the lateral resolution of digital holography.

High ranging speed, wide measurement range, and high accuracy are the trends of modern satellite/lunar laser ranging (SLR/LLR) . High precision event timer is demanded as the timing unit in such a laser ranging system. The time interval timing principle of event timer is analyzed.A high-accuray event timer is realized and designed based on time-to-digital converter (TDC) and field programmable gate array (FPGA) technique, for which the TDC chip is used to measure fine time and the FPGA chip is used to measure coarse time and control the whole timing system. Signal period measurement experiment is performed,showing the standard deviation better than 50 ps, the systematic error less than 11 ps, the measurement range of 24 h,the temperature drift below 100 fs/℃, and temporal stability lower than±3 ps/h.

Two independent resonators have been set up by using wide band polarization maintaining fiber Bragg grating(PFBG), common active fiber and narrowband common fiber Bragg gratings(FBG), and the central wavelength of the narrowband common FBG aims at one of the reflective peaks of PFBG, which can output stable dual/single wavelength single polarized laser.The Dual-wavelength of single polarized fiber laser based on common active fiber is fabricated and the fiber laser can output dual/single wavelength at normal temperature. The optical noise-to signal ratio(ONSR) of dual-wavelength laser is 46.7dB, and 59.6dB for one wavelength laser. The degree of polarization(DOP) of fiber laser is 98.5% by HP 8509B. This technique can be used in generation of microwave/millimeter-wave for radio-over-fiber(ROF).

The performance of optical en/decoder is one of the key factors affecting the whole performance of optical code-division multiple-access (OCDMA) system. Wavelength deviation influences the performance of en/decoder. The wavelength-deviation-tolerance (WDT) of equivalent phase shift superstructure fiber Bragg grating (EPS-SSFBG) encoder/decoder is investigated with different optical codes, different code sizes, different encoding bandwidths, different encoding efficiencies and different refractive index modulation intensities, based on the analysis of EPS-SSFBG as OCDMA encoder/decoder. The result shows that the WDT between optical source and encoder/decoder is relatively moderate, the auto-correlation ratios of PA to the maximum wing lavel (P/W) increases when the center wavelength of source deviates from that of encoder/decoder with wide input pulse width. The WDT is influenced by different optical codes, different frequency efficiencies and different refractive index modulations.

Multiple-access interference (MAI) and beat noise (BN) in coherent time-spreading optical code-division multiple access (OCDMA) system are analyzed according to aperiodic cross-correlation function. The mean intensity of aperiodic cross-correlation depends on the transit delay of interfering users, and MAI and BN will change with the mean intensity of aperiodic cross-correlation accordingly. The relationship between the mean intensity of aperiodic cross-correlation and MAI and BN is deduced, and then the relationship between the bit error rate (BER) and the mean intensity of aperiodic cross-correlation is discussed for coherent time-spreading OCDMA. For Gold sequence with the length N=511, mean and upper bound of BER performances are derived according to the different transit delay of the interfering users in coherent time-spreading OCDMA system. In the case of mean BER performance, OCDMA system can support about 12 interfering users. However, for the worst case (upper bound of BER), OCDMA system can support no more than 4 interfering users.

In order to improve the bandwidth efficiency of the L-level pulse-position modulation (PPM) scheme used in wireless optical communication, a new modulation scheme, L-level pulse-position-width modulation (PPWM) scheme was proposed based on L-level PPM scheme and L-level pulse-width modulation (PWM) scheme . It increases the bandwidth efficiency through reducing the power efficiency properly. The PPWM signal’s power efficiency and bandwidth efficiency are analyzed and the packet error rate in the weak turbulence channel model is also derived and compared with that in other modulation schemes. The result shows that the L-level PPMPWM scheme has low bit error rate, high power efficiency, better bandwidth efficiency, and low packet error rate.

A fiber grating based spectral-phase optical code-division-multiple-access (OCDMA) en/decoder was designed and realized. According to equivalent chirp theory, different frequency components of a broad band optical source can be separated and recombined by using step-sampling-chirped method, and phase shifts which carry address code can be introduced into these frequency components by an equivalent method. En-decoders fabricated by this method have some merits such as simple fabrication, low precision requirement. Therefore this method is suitable for system application. Comparison between computer simulation and measured data shows that the en/decoder has good performance.

An optical secure communication system based on digital chaos and polarization shift keying (PolSK) technology is proposed. Several PolSK schemes are employed, in which states of polarization are partially or totally uniform. In different cycles of the system clock, different PolSK schemes are selected in the transmitter to modulate the information according to the real-time generated digital chaotic sequences, and corresponding PolSK schemes are used in the receiver to demodulate the information according to the synchronous chaotic sequences. The pseudo-random of the digital chaotic sequences makes the PolSK modulation schemes randomly vary. When the point number of constellation is not less than 6 and number of selected PolSK schemes is not less than 4, the optional PolSK schemes exceed 1060 kinds. Therefore, the secret key space of the system is huge. Simulation is carried out with OptiSystem and Matlab softwares, and the numerical analysis is also presented. Simulation and analytical results indicate that this system works well in transmitting information and has strong anti-attack ability.

Optical code division multiple access (OCDMA) networks and several typical encoder/decoder (E/D) are introduced. The tunable E/D based on optical fiber delay line (OFDL) and on fiber Bragg gratings (FBGs) have been emphatically investigated. Two schemes of tunable OFDL are presented and their performance is analyzed. The principle of the E/D based on FBGs and the selection of parameters with piezoelectric ceramic (PZT) to tune FBGs are introduced. A two-dimensional tunable structure of E/D based on FBGs arrays are presented, whose merits are analyzed. The construction method of binary phase shift keying (BPSK) and tunable quaternary phase shift keying (QPSK) E/D based on super-structured fiber Bragg grating (SSFBG) are introduced, and the development prospects for OCDMA E/D are analyzed.

According to the nonreciprocal characteristic of magnetooptic material and the filtering characteristic of waveguide grating, the nonreciprocal filtering characteristic and the application of magnetooptic waveguide grating are presented. YIG crystal doped by Ce (Ce:YIG) with Faraday rotation angle (ΘF) of 4800 °/cm, rib section of compensation wall of single mode and apodizer grating of cosine type are chosen to design the magnetooptic waveguide grating. The method of finite-difference and equivalent effective index are used to simulate the intensity of the nonreciprocal effect of the magnetooptic waveguide grating. Besides, coupled-mode theory and transfer matrix method are also used to analyze the nonreciprocal filtering characteristics of the magnetooptic waveguide grating. The result indicates that for transverse electric (TE) mode and the wave band of 1550 nm, the forward center-wavelength of the magnetooptic grating is shifted by 0.8 nm relative to the backward center-wavelength and the bandwidth is 0.4 nm (-20 dB). This kind of nonreciprocal filtering characteristic can be applied to integrated optical devices such as nonreciprocal isolator and optical add-drop multiplexer (OADM).

A radio over fiber (ROF) system to generate optical millimeter wave with four time frequency of the local oscillator frequency using two-cascaded intensity modulators based on optical carrier suppression scheme was investigated numerically and experimentally. In the central office, the upstream data were mixed with the radio frequency (RF) clock to generate a electrical signal, and then an external modulator was used to generate a optical millimeter with double freqnency of RF clock, after the second external modulator, the optical millimeter wave was generated with frequency quadruple. By this scheme, a 40 GHz millimeter wave was generated by the frequency mixing of the 10 GHz signal source and the 2.5 Gbit/s baseband data signal passing through the two external modulator. After transmission over single-mode fiber (SMF) 20 km, the power penality is less than 2 dB.

A new modulation scheme called dual duration pulse position modulation (DD-PPM) is proposed for optical wireless communication system to overcome the short comings of pulse position modulation (PPM) and digital pulse interval modulation (DPIM). After introducing symbol structure, we study the bandwidth requirement, average transmission power and capacity. Meanwhile, the packet error rate in weak turbulence channel model is derived, and compared with on-off keying (OOK), PPM and DPIM. Theoretical and simulation results show that DD-PPM requires less average optical power and gives better bit error performance compared with OOK, and offers high bandwidth efficiency and transmission capacity compared with PPM for the given parameters. Because the symbol length is fixed in DD-PPM, there is no problems such as buffer overflow or waiting in modem, and it is more realizable than DPIM. Hence, DD-PPM is superior in optical wireless communication system.

Progress of optical phased array (OPA) technology based on electro-optic material is summarized. The basic principle is introduced and the basic structure and key techniques of OPA are reviewed with different electro-optic materials, including LiNbO3, AlGaAs waveguide, liquid crystal and lead lanthanum zirconate titanate (PLZT) ceramic and so on. The development of OPA based on PLZT electro-optic ceramic material is reviewed emphatically, and the latest achievements in the areas, such as single-stage OPA, cascaded OPA, and OPAs with surface electrodes and up-down electrodes are listed. The applications of OPA in laser radar are also introduced briefly.