On the basis of the stationary phase method, a new diffraction-free carpet beam is generated. Specifically, the Fresnel diffraction of radial gratings after passing through the axicon is investigated by the stationary phase method, and under the approximate conditions of the stationary phase, a kind of carpet beams with the diffraction-free transmission characteristics is theoretically obtained. Then, the amplitude-based spatial light modulator (SLM) is adopted to load the radial gratings with different parameters, and the phase modulation element (axicon) is installed behind SLM. Upon the Fresnel diffraction, a new family of instances of diffraction-free carpet beams can be generated within a certain transmission range behind the axicon. Finally, it is found that the experimental results are in good agreement with the theoretical simulation results.

In this paper, a digital orthogonal phase grating with adjustable period and phase modulation depth is simulated by using a programmable spatial light modulator. On the basis of analyzing the structure of orthogonal phase grating, its transmission function, far-field diffraction patterns, and diffraction efficiency are derived and calculated, and the influence of grating period T and phase modulation depth ? on the diffraction characteristics of orthogonal phase grating is discussed. The simulation and experimental results show that as ? increases from 0 to π, the relative intensity of zero-order light decreases gradually, and the intensity of eight first-order lights enhances. When ? equals π, the relative intensity of zero-order light reaches the minimum value of 0.25, and the maximum diffraction efficiency of first-order light is 56.96%. As ? increases from π to 2π, the relative intensity of zero-order light increases gradually, and the intensity of eight first-order lights decreases. This research provides a theoretical basis for the application of orthogonal phase grating.

We propose and investigate a highly sensitive fiber optic temperature sensor based on a two-mode-single-mode microfiber Sagnac loop (TS-MSL). The TS-MSL is fabricated by fused biconical tapering after a section of two-mode optical fiber is spliced between two sections of single-mode optical fibers, and it is packaged by the polydimethylsiloxane (PDMS) material. The experimental results reveal that the temperature sensing performance of the proposed sensor is effectively improved due to the optical field interference enhancement of TS-MSL as well as the photothermal sensitivity of the PDMS material. The temperature sensitivity of the proposed fiber sensor is up to 8.13 nm/℃, which is about 120 times higher than that of the ordinary single-mode microfiber Sagnac loop. With high sensitivity, the proposed temperature sensor is of high practical significance for accurate temperature monitoring in special occasions such as toxic gases and corrosive environments.

A flexible fiber bundle shape sensor with a thin diameter and its glue injection packaging device are designed. By the distributed strain measuring technology of an optical frequency-domain reflectometer, we experimentally explore the influence of the glue material and ratio on the generation, accumulation, and distribution of residual stress in the packaging process of the sensor. The experimental results reveal that the cross-linking reaction of glue makes the residual strain gradually increase with the curing time. Moreover, for the three optical fibers in the shape sensor, the accumulation of residual strain with time and the spatial distribution of residual strain along the fiber axle are consistent. After about 16 h curing reaction, the residual strain of the three fibers reaches the maximum and remains stable in the range of 80-100 με. The research results minimize the residual strain during the packaging process of the optical fiber shape sensor and lay a foundation for high-precision shape measurement.

A cubature Kalman filter (CKF) phase noise compensation algorithm of the block sub-symbol second-order Lagrangian interpolation (BSSLI) is proposed to reduce the serious influence caused by phase noise in the coherent optical orthogonal frequency division multiplexing (CO-OFDM) system. The algorithm first uses the phase noise value of the time-domain OFDM symbol estimated by the pilot to perform the first second-order Lagrangian interpolation (SLI) and obtain a rough estimation of the phase noise, and then divides each OFDM symbol into several sub-symbols in the time domain. The estimated value of each sub-symbol is used as the estimated value of its intermediate sampling point to perform the second SLI, thus both the time resolution and the compensation accuracy are improved, and finally the CKF is used to compensate the residual phase noise. The simulation results show that: for 16 order orthogonal amplitude modulation (16QAM) and the laser linewidth of 800 kHz, the bit error rate (BER) of the algorithm can be close to 10-7 and the error floor is reduced; for 32 order orthogonal amplitude modulation (32QAM) and the laser linewidth of 800 kHz, the BER of the algorithm can reach below 10-3. Furthermore, compared with the linearly interpolated sub-symbol optical phase noise compensation scheme (LI-SCPEC) algorithm and the Lagrange interpolation based extended Kalman filter (LRI-EKF) algorithm, the proposed algorithm can better compensate the phase noise. Therefore, it can effectively improve the performance of the CO-OFDM system.

Data centers are facing dual challenges from bandwidth and energy consumption, for which optical switching is a promising solution owing to its advantages of high bandwidth, low power consumption, and transparent transmission. Considering the high energy consumption and long delay caused by the electronic buffers used in the Clos architecture, an all-optical switching architecture based on Ring-Clos is proposed. Routing between adjacent central modules is provided for optical packets by employing intra-stage connection and tunable wavelength converters, thereby solving the output port conflict problem at certain input or central modules. Meanwhile, the concurrent dispatching for Ring-Clos switch scheduling algorithm is applied to distribute routes for optical packets as it has low complexity and simple hardware implementation. The simulation results show that the packet loss rate of the proposed architecture is as low as 48.81% of that of the Clos architecture, which means the proposed architecture effectively improves the network performance.

Nonlinear wavelength crosstalk is one of the key factors which restrict transmission distance and detection performance of remotely interrogated optical fiber hydrophone (OFH) systems with dense wavelength division multiplexed (DWDM) and time division multiplexed (TDM) structures. In this paper, wavelength power conversion and nonlinear crosstalk problems caused by stimulated Raman scattering (SRS) under the condition of multi-wavelength and high peak power injection are analyzed theoretically and simulated. And a crosstalk suppression structure based on wavelength-staggered emission is proposed. Experimental results of the remotely interrogated 16DWDM×8TDM OFH system show that with this wavelength-staggered scheme, the wavelength optical power imbalance after 100 km transmission is reduced by 9.7 dB compared with the original scheme, and phase noises of the hydrophone channels associated with the wavelength 1 and wavelength 16, which are most affected by the crosstalk, are reduced by 20.0 dB and 10.5 dB, respectively, to the level of -97.0 dB and -96.5 dB. So the results verify the validity of the wavelength-staggered structure. The study can provide an important reference for the design of the optical emission and transmission links of remotely interrogated OFH systems.

A few-mode fiber for space division multiplexing is proposed. By adopting a rectangular-assistant ring-core refractive index profile structure, which is similar to an ancient coin pattern, the fiber can eliminate the spatial degeneracy between the mode groups LPmn,a (m=0,1,2,…;n=1,2,3,…) and LPmn,b in the conventional few-mode fiber, maintaining a low level birefringence characteristics. These characteristics can significantly reduce the application difficulty of this type of mode group in space division multiplexing. In the optimized structure proposed in this paper, the refractive index difference of LPmn,a and LPmn,b is in the range of (2.2-6.5)×10-4 at 1550 nm, which is at the same level as the refractive index difference of adjacent modes. In addition, the polarization separation level of each mode is 10-6 or below. The design goals of significant separation of spatial degeneracy and unseparated polarization degeneration are well achieved. The simulation analysis shows that the structure has good deformation tolerance for the machining process.

Based on the OOK (On-Off Keying) dimming mode in IEEE 802.15.7 standard, this paper proposes a dimming scheme based on punctured polar code, which can not only obtain better communication performance, but also support the user's dimming requirements and suppress the discomfort caused by flicker to human eyes. Using punctured polar code scheme based on column greedy deletion algorithm and CRC (Cyclic Redundancy Check) aided SCL (Successive Cancellation List) decoding method, this scheme can obtain superior bit error performance. The pseudo-random scrambling code is used to scramble the punctured polar codes to obtain a 01 balancing data sequence with limited run length, so that the light intensity changes within the maximum flickering time period (MFTP), so as to avoid the perception of uncomfortable light flicker to the human eyes. The average light intensity can be obtained by inserting the compensation symbol, which meets the user's requirement of dimming. The simulation results show that the scheme can support any dimming rate (0-1). Compared with other schemes using forward error correction (FEC) code for dimming, the scheme proposed in this paper improves the transmission efficiency by more than 1.25 times when the code rate is 0.5; the scheme can provide a gain of 0.5-1.7 dB at the dimming rate of 0.2 and bit error rate of 10-4 when using a polar with code rate of 0.5.

The conventional phase-generated carrier (PGC) technology is faced with a high sampling rate and long calculation time during the demodulation process as it adopts a scheme of mixing with a multi-frequency of the carrier. For this reason, improved signal mixing methods are proposed according to the spectral characteristics of speech signals. Signals collected and converted by a photodetector are mixed with the sine and cosine terms of the carrier respectively. After being processed by a band-pass filter, the signals are demodulated by arctangent and differential cross multiplication approaches. The demodulation results of a simple sinusoidal signal prove that compared with the traditional method, the improved demodulation methods require lower sampling rates and shorter calculation time. The differential cross multiplication approach has a faster signal processing speed, whereas the arctangent method has a better demodulation effect and higher noise resistance. The demodulation results of a real voice signal show that the arctangent method has a favorable demodulation effect.

Disparity estimation is an important technique in the field of computational imaging. A calculation framework for reconstructing scene depth information from the Fourier disparity layer (FDL) is provided. A new method of deconstructing the disparity layer (DL) of the scene for achieving disparity reconstruction in the transform domain of the light field on the basis of FDL representation is realized. FDL is reconstructed with light field data, and the corresponding DL is obtained by inverse Fourier transform of FDL. Normalized cross-correlation (NCC) is applied to measure the correlation of the pixels in DL images with those in central view images to achieve the accurate reconstruction of scene disparity. Experimental results of both simulation data and real data show that the proposed method can effectively perform accurate scene disparity reconstruction.

Polarization aberration will affect the accuracy of polarization measurement and the polarization imaging effect of an off-axis optical imaging system, and thus it is necessary to calibrate and compensate for it. On the basis of three-dimensional polarization ray tracing, this study analyzes the polarization aberration of the coded super-resolution off-axis optical imaging system with a digital micro-mirror device (DMD) in different fields of view and proposes a method of polarization compensation by adding a linear attenuator (LD) and a linear retarder (LR) in the optical path near DMD. The calculations indicate that the maximum diattenuation and the maximum phase retardance introduced by the DMD surface are 1.43×10-3 and 9.52×10-3 rad, respectively, while the maximum diattenuation and the maximum phase retardance introduced by the overall optical system are 2.32×10-3 and 1.55×10-2 rad, respectively. Hence, the polarization aberration introduced by DMD accounts for more than 60% of that introduced by the whole system. Then, this paper compares the polarization aberration distributions, Jones pupils, and polarization imaging simulations of the overall optical system before and after polarization compensation. The results reveal that after an appropriate weak polarizer is used for compensation, the diattenuation and phase retardance are reduced by about a half, with the Jones matrix close to the unit matrix and the crosstalk phenomenon in polarization imaging alleviated significantly. It can be concluded that DMD introduces severe polarization aberration, but the utilization of the LD and LR in the optical path near DMD can simply and effectively compensate for the polarization.

In terms of the problem that ablation defects in the buffer layer of in-service high-voltage cables mostly appear at the bottom of the cable, a local source-translation computed tomography (L-STCT) method is proposed to detect defects in the water-blocking buffer layer at the bottom of in-service high-voltage cables. During the operation of L-STCT, the flat panel detector is placed close to the tested cable, estimated defect approximately exists at the maximum projection coverage angle, and projection data is collected by X-ray STCT. This paper establishes an L-STCT imaging model, analyzes scanning parameters and the maximum projection coverage angle of each data point of the scanning system, and studies their influence on image reconstruction. In addition, an experimental platform is set up to carry out a preliminary study on image reconstruction by the simultaneous iterative reconstruction technique (SIRT) algorithm. The simulation and actually experimental results show that L-STCT can detect local defects in the water-blocking buffer layer of high-voltage cables, which provides an important reference for the application of in-service high-voltage cable detection.

In three-dimensional imaging of structured light, the high-reflective areas of scenes to be measured will be saturated on camera imaging arrays, which affects the phase accuracy of phase-shift structured light. Therefore, an adaptive phase-shift structured light measurement method is proposed. First, several uniformly-grayscaled images are projected onto an object, the high-reflective areas and their saturation degrees are detected, and the optimal projected light intensity is determined pixel by pixel. Then, according to the object properties, the projected light intensity is reduced to realize coordinate mapping, and a group of adaptive fringes with a high dynamic range is automatically generated. Finally, the phases over the highly saturated area are fused with the phases of coordinate mapping. The experimental results reveal that the proposed method can accurately adjust the projected light intensity. Compared with the traditional method, the proposed method reduces the maximum root mean square error (RMSE) and the RMSE of the saturated areas by 99.43% and 92.48%, respectively, and it effectively improves the morphological measurement accuracy of the high-reflective areas.

To improve the precision and robustness of angle measurement by existing visual marker methods, a high-precision visual pose measurement approach that combines microlens arrays and random micrographics arrays for encoding is proposed. First, in the calibration stage, dense and high-precision encoding is carried out by dynamic imaging modes of random micrographics arrays under microlens arrays, and a high-precision coordinate datum of spatial angles is constructed by the Hamming code. Then, in the measurement stage, observation angle encoding is completed on the basis of observed display images, the weighted fusion of which with the nearest angle datum is applied to obtain the precise angles of measuring positions. In this way, the triaxial angle measurement and pose estimation with high precision can be achieved. At the same time, extensive experiments are performed to verify the validity of the proposed method. The results reveal that the angle measurement precision is higher than 0.08°, and compared with the results of previous work, the average angle measurement error is reduced by 97% without sacrificing the displacement measurement precision. In addition, the robustness of the proposed algorithm ensures high-precision pose measurement under natural illumination. Finally, by sufficient quantitative comparison, the effects of factors on angle measurement precision, such as threshold segmentation and angle encoding parameters, are comprehensively analyzed.

The alpha matting algorithm can effectively separate the foreground and background information of two-dimensional images to realize object extraction, alpha image compositing, and other computational imaging processing. A light field alpha matting propagation model satisfying spatial-angular consistency is established through the sparse representation of light field data based on the spatial-angular interaction of four-dimensional light field data. The alpha closed form solution of the central sub-aperture image is propagated to other sub-aperture image plane by the propagation model, and the alpha images of light field with spatial-angular consistency can be obtained efficiently. In addition, a quantitative evaluation metric of spatial-angular consistency for light field alpha images based on epipolar plane image (EPI) is proposed. The numerical experiments are conducted on synthetic data sets and real light field data. The results show that compared with each sub-aperture image matting algorithm, the proposed algorithm can reduce redundant computation and quickly obtain high-quality light field alpha images with better spatial-angular consistency.

When an ordinary target is used for camera calibration, it is necessary to ensure the integrity of its imaging at all angles. As the small proportion of the target in the camera field of view leads to the low camera calibration accuracy, the calibration accuracy will be further reduced when the camera field of view changes in zoom measurement. In this paper, we theoretically analyze the characteristics of homography matrices and find that the targets in target images captured from different angles are different, but it does not affect the calculation of homography matrices and camera calibration. According to the simulation analysis, we design a directional layered target. In the first layer of the target, the directional marks are designed, and the size of the targets gradually increases toward the outer layer. In different fields of view, layers of different ranges are used, and the targets should always have a large proportion in the field of view to ensure the camera calibration accuracy. Moreover, the target matching method based on the combination of the collinear principle and projective transformation is proposed. As long as the directional feature points of the target appear in the image, the recognition and matching of targets can be realized, which improves the flexibility of the target and the efficiency of camera calibration. The experimental results indicate that the target recognition and matching method of the directional layered target is accurate and effective, which is not affected by the lack of some targets. Therefore, this target can improve the camera calibration accuracy, and the calibration accuracy is free from the influence of the change in the camera field of view.

To improve the sensitivity and quality factor in refractive index sensing, a grating-assisted slot microring resonator is proposed. On the basis of the slot microring, the grating structure is introduced to realize the electromagnetically induced transparency (EIT)-like effect at different wavelengths, and obtain higher sensitivity and higher quality factor. A numerical model for spectrum calculation and parameter optimization of the hole-embedded microring resonator is built by the transfer matrix method. The transmission spectrum of the proposed device structure and electric field distribution of the resonator are simulated using the finite-difference time-domain (FDTD) method, and the influence of device structure parameters is discussed. The results indicate that the proposed device achieves the EIT-like. In the application of refractive index sensing, the quality factor of the proposed device is 35495, and the sensitivity is as high as 360 nm/RIU (RIU is refractive index unit). Compared with the results of the conventional microring resonator with similar parameters, the sensitivity is improved by 270 nm/RIU, and the lower limit of sensitivity detection of the proposed device is 2.3×10-4 RIU.

A 1.3 μm high-speed directly modulated semiconductor laser based on AlGaInAs material is designed. The designed laser uses the ridge waveguide structure, the cavity with short length and eleven multi-quantum wells with thickness of 5 nm combined with the graded index separate confinement heterostructure (GRIN-SCH) with thickness of 30 nm to achieve laser output with low threshold, wide bandwidth and high power. A stable single longitudinal mode output is achieved by using uniform grating and asymmetric cavity surface coating. Based on the 1.3 μm high-speed directly modulated semiconductor laser at room temperature, the threshold current is about 7.5 mA, the 3 dB small signal modulation bandwidth can reach 25 GHz, the large signal back-to-back transmission rate can reach 40 Gb/s, the slope efficiency is 0.35 mW/mA, the maximum output power is about 39 mW, and the side mode rejection ratio can reach 40 dB.

By the cladding technology of inside-laser internal powder feeding, the crossing cross-forming technology of cross structure is studied using the strategy of local acceleration in the overlapping position. First, the cross-unequal height deposition process is proposed. According to the law of mass conservation, the mass of powder entering the molten pool and the forming mass of the cladding layer are calculated, and the influence laws of the first formed transverse melting channel on the height of the later formed longitudinal melting channel by the system of equations are obtained. Then, a mathematical model of the unequal height of the velocity is built. On the basis of the cross-unequal height deposition process, the scanning speed is calculated when the overlapping height of the crossed position is equal to the height of the melting channel in the uncrossed position. Finally, in view of the self-healing effect and volume flow hypothesis, a process is put forward to reduce the range of the acceleration section, which solves the problem of sags in the cross-linking part of the melting channels after the acceleration. The results reveal that the height error in the crossed position is within 1.76%. The overall microhardness is between 257.5 HV and 296.8 HV, and the surface of the final formed part is smooth and bright, without macroscopic cracks or inclusion defects.

An all-optical passive synchronous ultrashort pulse laser source based on an all-polarization-maintaining fiber is built by the master-slave injection locking method. By optimizing the pulse energy and pulse width of pulse from the master laser, the mismatch length of 0.879 cm is achieved. The designed synchronous laser source consists of an erbium-doped fiber mode-locked pulse laser (master laser) and a ytterbium-doped fiber mode-locked pulse laser (slave laser) based on nonlinear amplifying loop mirror. The central wavelength of the master laser is 1560 nm, and its spectral width is 5.01 nm. When the output single pulse energy is increased from 0.06 nJ to 3.26 nJ, the narrowest and widest pulse widths are 221 fs and 1.79 ps after the output pulse passes through the erbium-doped fiber amplifier. The central wavelength of the slave laser is 1064 nm, and its spectral width and pulse width is 0.22 nm and 9.5 ps respectively. By controlling the output parameters of the erbium-doped fiber amplifier, the tolerance range of cavity-length mismatch between the master and slave lasers approaches to centimeter level, which provides an efficient way to avoid using the time delay control components and fiber optical path coupling components in the synchronous laser source.

The multi-sensor information fusion method of the existing multi-object tracking algorithms for self-driving cannot give full play to synergy. To solve this problem, a three-dimensional multi-object tracking algorithm based on multi-modal feature fusion and learnable object similarity estimation is proposed. The multi-modal feature fusion module fuses the feature of images and point clouds on the basis of the channel attention mechanism to further improve the expressive ability of multi-modal features. The object similarity estimation module directly generates the similarity matrix through the network, and realizes the cross-modal joint reasoning between multiple objects in a learnable way, which avoids massive manual parameter setting. The proposed algorithm is verified and tested on the KITTI data set, and its higher-order tracking accuracy (HOTA) reaches 69.24% in the test set, which indicates that the algorithm is superior to other algorithms in accuracy and has good robustness.

Alloying/doping molybdenum disulfide (MoS2) is a new way of exploring the potential applications of two-dimensional materials in microelectronic devices. Chemical vapor deposition is applied, for which sodium chloride is used to assist growth. Moreover, the mass ratio of sulfur powder to selenium powder is adjusted, and six kinds of monolayer MoS2(1-x)Se2x alloys with different compositions are thereby obtained on SiO2/Si substrates. The photoluminescence peak position varied between 678 nm (~1.83 eV) and 813 nm (~1.53 eV). The transverse dimension of the continuously grown large-area monolayer MoS2(1-x)Se2x (x=0.25) alloys can reach 200 μm. For the investigation of the photoelectric properties of MoS2(1-x)Se2xalloys, a large-area monolayer MoS2(1-x)Se2x (x=0.25) alloy is used to prepare field-effect transistors. The photoelectric test results show that the response of the monolayer MoS2(1-x)Se2x(x=0.25) field-effect transistor irradiated by the 520 nm laser reaches 940 mA·W-1, with a detection rate of 5.32×1010 cm·Hz1/2·W-1 and a fast response time of 8 ms.

In terms of the problem that polystyrene (PS) nanoparticles are hard to be aggregated and detected in water, this paper studies the mixed solution containing gold (Au) nanoparticles and PS nanoparticles, captures and detects the photothermal effect of PS nanoparticles by using a self-built optical manipulation-micro Raman system, and analyzes the photothermal effect of PS nanoparticles and the signal enhancement effect of the surface-enhanced Raman scattering (SERS) in the mixed solution. The results show that the movement speed of PS nanoparticles (80 nm) can be affected by the size and concentration of Au nanoparticles, and Au-PS aggregates with a diameter of 30 μm are formed in the photothermal trap. In addition, the SERS signal intensity of PS nanoparticles in aggregates is 7 times higher than that in the solution containing only Au nanoparticles, and it increases firstly before decreasing as the radius of Au-PS aggregates increases. The proposed method can capture the photothermal effect and detect the SERS signal of PS nanoparticles, with the signal intensity significantly improved and the detection limit reduced. Therefore, it is proven to be applicable in a variety of fields such as environmental pollution detection and nanodevice self-assembly.

A novel butterfly-shaped beam based on diffraction catastrophe is proposed and experimentally generated. According to the catastrophe theory, the light field structures of the beams are defined by the potential function composed of the state and control variables. The caustics of the beams are theoretically manifested as hypersurfaces in four-dimensional space due to the high dimensionality of the control variables, and these beams display diverse light field structures when they are mapped into a low-dimensional space. Furthermore, different light field structures of the beams can be obtained by manipulating the control variables. It is found that the spectral amplitudes of the beams can be expressed as polynomials. The experimental results are in good agreement with the numerical simulation ones. These beams have excellent properties including curved propagation trajectories and various light field structures, which are likely to be applied to wavefront control, optical micromanipulation, and biomedicine.

Landau damping of collective excitation in a homogeneous Bose-Einstein condensate (BEC) is studied by the Hartree-Fock-Bogoliubov (HFB) mean-field theory. The rigorous derivation is carried out without using the two approximations of quasi-particle resonance transition and collective excitation energy. The Landau damping as a function of temperature is given in a wide parameter range, and two limits of absolute zero and critical temperature of phase transition are highlighted. The contribution of quasi-particle transitions with different energies to damping is analyzed by the error function. In addition, the above two approximations are used for derivation, and the applicable range of the approximations is analyzed by the results of the error function.

To solve the inaccurate calculation problem caused by the large peak positioning error when the peak-to-peak (P2P) method is employed to demodulate low-finesse fiber-optic Fabry-Perot (FP) sensors, this paper proposes a P2P and interference-order positioning joint demodulation algorithm. For this purpose, two peaks are positioned in the reflection spectrum of the FP sensor, and cavity length is estimated by the conventional P2P method. Then, valley interference orders are introduced after a valley is positioned to generate a sequence of possible cavity length values corresponding to different interference orders. Finally, cavity length demodulation is achieved by retrieving the value in the cavity length sequence closest to the result estimated by the P2P method. To demonstrate the feasibility and superiority of the algorithm, this study also conducts demodulation simulations and experimental verifications of low-fineness FP sensors made from single-mode fibers. The experimental demodulation accuracy, better than 2.3 nm, is much higher than that of the conventional P2P method. The proposed algorithm can be used to accurately demodulate low-finesse FP sensors with a cavity length of 55-135 μm.

In order to adjust and detect micro-nano metasurfaces accurately and efficiently, the problem of coupling scattering between a grid micro-nano metasurface and buried micro-defects is studied based on the multi-resolution time domain (MRTD) method. The concept of multi-resolution is introduced, and the coupling scattering model is established from the Maxwell equation. The scattering field is derived and compared with the finite difference time domain (FDTD) method. The correctness of MRTD is verified and its advantages are analyzed. According to the field distribution of the grid metasurface containing micro-defects, the necessity of studying the influence of various parameters of defects on the optical system of metamaterials is given. Then, the effects of the parameters such as defect size, buried depth, and relative orientation on the coupling scattering characteristics are analyzed numerically. The results provide technical supports for the fields and directions of functional surface design, ultrasensitive detection, scattering peak orientation, and frequency selection.