In datacenter networks, it is necessary to determine whether the path is congested according to the one-way delay of packets. The accurate measurement of one-way delay depends on the high-precision time synchronization of the source device and destination device. We have proposed a time synchronization method based on timestamp mapping, combined with in-band network telemetry technology to obtain the packet send timestamp and receive timestamp on devices. The results show that the maximum synchronization error is 19 ns, and the standard deviation is 7.8 ns with a 100 ms time synchronization period and offset adjustment strategy. The proposed time synchronization method achieves outstanding synchronization accuracy and stability.
With the rapid growth of Wide Area Networks (WANs) and advancements in 5G, cloud computing, and IoT, networks face higher demands for low cost, high capacity, reliability, and security. To ensure these requirements, network impairment emulators have become essential tools for testing and optimizing network performance under various conditions. This paper presents an FPGA-based Network Impairment Emulator (FNIE) that accurately emulates packet loss, reordering, and delay with high throughput and low cost. FNIE can achieve up to 100 Gbps throughput, with a configurable packet loss rate ranging from 0.001% to 100%, with a resolution as fine as 0.001%. It also supports up to 64 descriptor queues for reordering, a maximum reordering extent of 511, and emulates delays from 1 µs to 1 s.
With the increase in scale and complexity of integrated circuits, the amount of test data volume has also grown. Due to the simple structure, low computational cost, and capability to handle large-scale data of clustering algorithms, they offer a new solution for processing test data of integrated circuits. This paper defines two characteristics based on the features of test data: frequency_rate and co unt_of_ones, and incorporates them into the K-MEANS, DBSCAN, and OPTICS algorithms, respectively, to achieve segmentation of circuit test data with the silhouette coefficient as the evaluation criterion. Based on extensive data experiments, the results indicate that the OPTICS algorithm achieves the best clustering effect and is suitable for subsequent data processing.
In this paper, we propose an extension to the barrier model, i.e., the Multi-Barriers Model, which could characterize an area of interest with different types of obstacles. In the proposed model, the area of interest is divided into two or more areas, which include a general area of interest with sampling points and the rest of the area with different types of obstacles. Firstly, the correlation between the points in space is characterized by the obstruction degree of the obstacle. Secondly, multiple Gaussian random fields are constructed. Then, continuous Gaussian fields are expressed by using stochastic partial differential equations (SPDEs). Finally, the integrated nested Laplace approximation (INLA) method is employed to calculate the posterior mean of parameters and the posterior parameters to establish a spatial regression model. In this paper, the Multi-Barriers Model is also verified by using the geostatistical model and log-Gaussian Cox model. Furthermore, the stationary Gaussian model, the barrier model and the Multi-Barriers Model are investigated in the geostatistical data, respectively. Real data sets of burglaries in a certain area are used to compare the performance of the stationary Gaussian model, barrier model and Multi-Barriers Model. The comparison results suggest that the three models achieve similar performance in the posterior mean and posterior distribution of the parameters, as well as the deviance information criteria (DIC) value. However, the Multi-Barriers Model can better interpret the spatial model established based on the spatial data of the research areas with multiple types of obstacles, and it is closer to reality.
Electrocardiogram (ECG) signal is critical to the classification of cardiac arrhythmia using some machine learning methods. In practice, the ECG datasets are usually with multiple missing values due to faults or distortion. Unfortunately, many established algorithms for classification require a fully complete matrix as input. Thus it is necessary to impute the missing data to increase the effectiveness of classification for datasets with a few missing values. In this paper, we compare the main methods for estimating the missing values in electrocardiogram data, e.g., the “Zero method”, “Mean method”, “PCA-based method”, and “RPCA-based method” and then propose a novel KNN-based classification algorithm, i.e., a modified kernel Difference-Weighted KNN classifier (MKDF-WKNN), which is fit for the classification of imbalance datasets. The experimental results on the UCI database indicate that the “RPCA-based method” can successfully handle missing values in arrhythmia dataset no matter how many values in it are missing and our proposed classification algorithm, MKDF-WKNN, is superior to other state-of-the-art algorithms like KNN, DS-WKNN, DF-WKNN, and KDF-WKNN for uneven datasets which impacts the accuracy of classification.
The rapid development of Internet finance has certainly affected the operation of commercial banks. This paper investigates the impact of Internet finance on commercial banks. First, a theoretical influence mechanism of Internet finance on commercial banks is explored, and the Internet finance index and integrated performance index of commercial banks are constructed using factor analysis. Then, a static panel and a dynamic panel model are established to empirically examine the impact of Internet finance on the profitability, security, liquidity and growth as well as the comprehensive business performance of commercial banks. Finally, the heterogenous impacts of Internet finance on city commercial banks, joint-stock banks and state-owned commercial banks are discussed. The results show that the development of Internet finance has a positive impact on the profitability, security and growth of commercial banks, and has a negative impact on the liquidity of commercial banks. In addition, Internet finance has promoted the improvement of the comprehensive business performance of commercial banks. Moreover, the impact of Internet finance on different types of commercial banks is heterogeneous with the impact on state-owned commercial banks being the weakest and the impact on city commercial banks is the most significant.
With the popularity of mobile devices and the development of e-commerce, more and more people choose to buy items in the mobile terminal. Therefore the mobile terminal commodity recommendation services and commodity recommendation algorithms are more and more important. Aim at this problem, this paper conducts a study of predicting the user's purchase behavior based on the online distribution platform and the desensitization data sets provided by the Chinese largest electricity platform Alibaba. Based on the GBDT (Gradient Boosting Decision Tree) model, by using ODPS (Open Data Processing Service) and Python to simultaneously implement machine learning and training online and offline respectively, and combining with the user behavior sequence recorded over a period of time, the user purchase behavior at a later time will be properly predicted.
Visualization provides an interactive investigation of details of interest and improves understanding the implicit information. There is a strong need today for the acquisition of high quality visualization result for various fields, such as biomedical or other scientific field. Quality of biomedical volume data is often impacted by partial effect, noisy, and bias seriously due to the CT (Computed Tomography) or MRI (Magnetic Resonance Imaging) devices, which may give rise to an extremely difficult task of specifying transfer function and thus generate poor visualized image. In this paper, firstly a nonlinear neural network based denoising in the preprocessing stage is provided to improve the quality of 3D volume data. Based on the improved data, a novel region space with depth based 2D histogram construction method is then proposed to identify boundaries between materials, which is helpful for designing the proper semiautomated transfer function. Finally, the volume rendering pipeline with ray-casting algorithm is implemented to visualize several biomedical datasets. The noise in the volume data is suppressed effectively and the boundary between materials can be differentiated clearly by the transfer function designed via the modified 2D histogram.
Although heart researches and acquirement of clinical and experimental data are progressively open to public use, cardiac biophysical functions are still not well understood. Due to the complex and fine structures of the heart, cardiac electrophysiological features of interest may be occluded when there is a necessity to demonstrate cardiac electrophysiological behaviors. To investigate cardiac abnormal electrophysiological features under the pathological condition, in this paper, we implement a human cardiac ischemic model and acquire the electrophysiological data of excitation propagation. A visualization framework is then proposed which integrates a novel depth weighted optic attenuation model into the pathological electrophysiological model. The hidden feature of interest in pathological tissue can be revealed from sophisticated overlapping biophysical information. Experiment results verify the effectiveness of the proposed method for intuitively exploring and inspecting cardiac electrophysiological activities, which is fundamental in analyzing and explaining biophysical mechanisms of cardiac functions for doctors and medical staff.
Human motion recognition is a hot topic in the field of human-machine interface research, where human motion is often represented in time sequential sensor data. This paper investigates human motion recognition based on feature-selected sequential Kinect skeleton data. We extract features from the Cartesian coordinates of human body joints for machine learning and recognition. As there are errors associated with the sensor, in addition to other uncertain factors, human motion sequential sensor data usually includes some irrelative and error features. To improve the recognition rate, an effective method is to reduce the amount of irrelative and error features from original sequential data. Feature selection methods for static situations, such as photo images, are widely used. However, very few investigations in the literature discuss this with regards to sequential data models, such as HMM (Hidden Markov Model), CRF (Conditional Random Field), DBN (Dynamic Bayesian Network), and so on. Here, we propose a novel method which combines a Markov blanket with the wrapper method for sequential data feature selection. The proposed algorithm is assessed using four sets of human motion data and two types of learners (HMM and DBN), and the results show that it yields better recognition accuracy than traditional methods and non-feature selection models.
Image segmentation plays a crucial role in breast ultrasound (BUS) for breast cancer detection. However, due to the heavy speckle noise, low contrast and shadowing effects of BUS images, it's a challenging task to develop an accurate and robust segmentation algorithm. In this paper, we present a novel algorithm for breast ultrasound image segmentation which is based on hybrid of level set and graph cuts. Firstly, speckle reducing anisotropic diffusion. Subsequently, the initial contour is achieved by graph cut. To smooth the boundaries, a boundary item is added into the energy function of level set. At last, the final contour converges to the objective boundary quickly and accurately after finite steps of iteration of level set. Comparing with the traditional level set method, the requirement of the costly re-initialization procedure is completely eliminated and the local minima are effectively avoided in our method. The experiment results show that the processing time for level set based method to BUS image can be substantially reduced while getting promising segmentation results.
With the ascension of computer processing capacity, the problems of massive satellite images presentation encountered in the rendering frame rate, large data scheduling, external memory organization has been basically solved, Google earth put forward one multi-user solution dependent on the powerful cluster rendering technology and collaborative network transmission technology. However, in the limited resources of the stand-alone system, the display of multi-user collaborative data become another challenge. This paper proposed a collaborative scheduling scheme for multi-user scheduling problem of satellite image data in the stand-alone mode, in the fully enhance the system resource utilization condition, balanced user's average experience and data throughput condition. Finally, in the prototype system, we verify its validity and analyze its efficiency and interactivity though the user's experience way.