Search

Squeezed state light is a fundamental resource for many quantum optics experiments. Important applications like quantum-enhanced magnetometer and quantum key distribution, require the squeezing level as high as possible. In balanced homodyne detection, the detector output is directly proportional to the noise of the squeezed beam. The detected quadrature is dependent on the relative phase between the squeezed beam and local oscillator. Here, we present a balanced homodyne detector (BHD) with independent phase control and noise detection branches. The independence between the phase control and noise detection branches is experimentally verified, the two branches are independently optimized according to the practical requirements. With the BHD, a bright squeezed light at 1550 nm with quantum noise reduction of 10.3 dB is directly detected. After stabilizing the relative phase between the squeezed states and local oscillator by the phase locking branch of the BHD, the fluctuation of the noise suppression is ±0.2 dB by 1-h testing time.

Metal contamination control is one of the most critical performance for ion implantation process, and any metal contamination introduced to junction will induce leakage current and device performance degradation. With CMOS technology developing, more and more aggressive requirements for metal contamination control of ion implantation process has been brought up, especially for CMOS image sensor product. In this work, metal contamination performance was evaluated for 300mm high current ion implantation tool. Experiments were designed to locate the metal contamination source for wafer mechanical transfer and on-beam process separately. Dedicated hardware modification was proposed and performed according to the results analysis. And after tool modification, the metal contamination, particle and SIMS-matching were characterized to evaluate its influence on ion implantation process. From the measured data, conclusion can be made that the metal contamination performance was greatly improved without any process performance degradation after hardware modification.

We propose a meta network (MNet) for the problem of target-aspect missing in radar high-resolution range profile (HRRP)-based noncooperative target recognition, where a classifier must be generalized to new aspects not seen in the training set, given only a small number of HRRP data of each new aspect. The MNet is a time domain convolutional neural network (TCNN) that is built based upon recent progress in meta-learning. In effect, it learns a model that is easy and fast to fine-tune, allowing the adaptation to happen in the right space for fast learning. Besides, we construct a new controllable HRRP dataset suitable for the scenario of noncooperative target-aspect missing using electromagnetic simulation. Compared with the traditional methods, the MNet is more efficient and could achieve better performance. Extensive experiments on the simulated HRRP dataset are conducted to illustrate the effectiveness of the proposed method.

In this paper, we propose a Memory-Based Neural Network(MBNN) for Radar Automatic Target Recognition (RATR) based on High Resolution Range Profile (HRRP) in imbalanced case to learn how to find out the discriminative representations and generalize the ability to barely appeared target samples of some categories. Specifically, we utilize a Convolutional Neural Network (CNN) to explore discriminative features among HRRP samples and employ a memory module to record misclassified samples or samples that are correctly classified with low confidence into a external storage, we called it buffer. Then we leverage a Long Short Term Memory (LSTM) to merge the classified samples with some of the most similar ones in the buffer to make the final decision. It is worth noting that MBNN can be inserted as a plug-and-play module into any discriminative methods. Effectiveness and efficiency are evaluated on the measured data.

In this paper, we purpose a variational auto-encoder with learnable prior to learn the latent features of high-resolution range profile (HRRP) samples for downstream task, target recognition in this paper. The measured data based experiments illustrate that compared with the traditional and mostly recent HRRP recognition methods, the proposed one is not only noise-robust and data insensitive but also achieves state-of-the-art performance.

Due to the increasing integration of different disciplines, Model-based Systems Engineering (MBSE) is becoming more beneficial for the development of mechatronic systems. However, even though all relevant discipline-specific models can be integrated into one system model, how to maintain consistency between a discipline-specific model and the central system model during model updates still needs to be considered. In this paper, an approach for automatic synchronization of a discipline-specific model i.e. mechanical CAD model and the central SysML-based model i.e. a SysML4 Mechatronics plant model, is introduced. AutomationML(AML), a data exchange format which enables a lossless exchange of information between different discipline-specific engineering tools, is applied during the model synchronization. Mapping rules are proposed to facilitate a consistent round-trip engineering and transformation. Feasibility and applicability of this approach are demonstrated through a use case on a production plant with two applied commercial tools, PTC Integrity Modeler and Creo Parametric.

As an unsupervised task, hyperspectral image (HSI) clustering separates pixels into different groups. In this paper, an orthogonal graph-regularized non-negative matrix factorization (OGNMF) algorithm is proposed for HSI clustering. Because of large size of HSI, the HSI clustering problem is computational consuming. On the other hand, non-negative matrix factorization (NMF), which is a popular multivariate analysis method, has related to light computation budget. Furthermore, the orthogonal and graph regularization, which can improve the clustering performance and capture the local structure features, are applied to NMF for better performance in HSI. Moreover, both spatial and spectral information are extracted and utilized. Our experiments show that the performance of the proposed OGNMF is better than the NMF and ONMF algorithms in HSI clustering.

In this paper, parallel multi-view low-rank sparse subspace clustering (MLRSSC) is investigated for unsupervised classification of remotely sensed hyperspectral imagery. A 3-dimensional (3D) hyperspectral image contains abundant spectral and spatial information. Such diverse information can be considered as multiple views during the clustering process. In this paper, multiple spectral views are generated from correlated spectral band groups, a decorrelated and denoised view from principal components, and spatial views from morphological features. To make such a computational expensive clustering technique applicable to large-scale remote sensing images, parallel MLRSSC is implemented to non-overlapping 3D blocks. Experimental results demonstrate that the performance of the MLRSSC is better than other subspace clustering based algorithms.

Clustering is an unsupervised task, which groups samples into classes without training datasets. It is a challenging task, especially for hyperspectral image (HSI), which contains more information in spectral bands than regular images. In this paper, kernel spatial-spectral based multi-view low-rank sparse subspace clustering (k-SSMLC) algorithm is proposed for HSI clustering. In the generated multi-view dataset, spectral views are created by spectral partition, spatial view is created by morphological feature extraction, and PCA provides clean view with less noise. After the multi-view dataset is created, kernel multi-view low rank sparse subspace clustering is applied. Experiments results demonstrate that the clustering accuracy of the proposed k-SSMLC is better than other HSI clustering algorithms, such as sparse subspace clustering (SSC), low-rank sparse subspace clustering (LRSSC), spectral-spatial sparse subspace clustering (S4C), and regular SSMLC.

Hyperspectral image (HSI) Clustering is an unsupervised task, which segments pixels into different groups without using labeled samples. In this paper, spatial-spectral based multi-view low-rank sparse subspace clustering (SSMLC) algorithm is proposed. Due to significant number of spectra bands HSI contains much more information than a regular image. These spectral information can be considered as multiview. In this paper, the spectral partitioning is applied to generate spectral views which contain correlated bands. Morphological features of the original HSI are taken as another view which contains spatial features. Principal components construct another view, which eliminates the noise in the original dataset. After the multi-view dataset is formed, multi-view low-rank sparse subspace clustering is applied to segment HSI. Our experiments show that the performance of the proposed SSMLC is better than other that of clustering algorithms such as sparse subspace clustering and low-rank sparse subspace clustering.

In this paper, the tensor-based offset-sparsity decomposition (TOSD) method, or low-rank and sparse decomposition, is applied to hyperspectral imagery, where the low-rank tensor is considered to be enhanced or pruned data and used for classification. In the tensor form of dataset, all the information of the original 3D data cube, includes spatial and spectral information, can be better reserved. To make the low-rank assumption more possibly true, spatial and spectral segmentations are conducted in a preprocessing step for the TOSD. The experimental results demonstrate the TOSD offers better performance than the matrix-based one, and the spatial-spectral segmentation can further improve the performance.

The increase will boost the Chinese aftermarket dramatically, reflecting the fact that the focus in automotive industry will be shifted from vehicle sales to after service. Up to now automotive aftermarket of China has been enlarging, but the level of Chinese automotive aftermarket still fall behind compared with the developed countries. Active after-sale service means to provide timely communication and help. This paper expounds the connotation of auto after-sale service and analyzes the current situation and problems of auto after-sale service in China. Then we come up with the new application and significance of active service. On this basis, we propose a new service model combining active service with auto after-sale service based on customer analysis by using some algorithms.

The RFID systems are used more and more widely, and the low profile and low cost antennas are strongly needed for some base stations. Here a novel 3D miniaturized RFID antenna is proposed. It consists of two vertical printed triangular monopole antenna elements. And top-loading is used to reduce the antenna's profile. Four tuning poles are adopted for better matching and broader bandwidth. The measured return loss is −33dB and its −10dB bandwidth is about 20%. The measured maximum gain is 3.73dB. Compared with conventional RFID antenna, 50% size reduction is achieved.

Oscillating-teeth end-face harmonic gear is of an advanced drive wheel suitable for large power and high transmission ratio system. It has a broad prospect of application. The paper introduces transmission types of oscillating-teeth end-face harmonic gear, and analyses different transmission types of oscillating-teeth end-face harmonic gear's meshing process, summarizes evolution of total meshing area of oscillating-teeth end-face harmonic gear's work meshing pair. This research provides a theoretical basis for further discussion to bearing capacity of oscillating-teeth end-face harmonic gear's work meshing pair.