Deep Neural Networks (DNN) model training usually requires a large amount of data as the foundation, so that the model can learn effective features and rules. However, these data often contain sensitive information of...
详细信息
ISBN:
(数字)9789819996148
ISBN:
(纸本)9789819996131;9789819996148
Deep Neural Networks (DNN) model training usually requires a large amount of data as the foundation, so that the model can learn effective features and rules. However, these data often contain sensitive information of users. This paper designs a DNN Classification Model for Ciphertext data (DNN-CMED), which consists of three-party entities, including two servers and one client. The auxiliary server assists the model training server in completing the computation of the nonlinear layer of the DNN model, and the two servers interact with each other to complete the task of classifying the ciphertext data. The communication protocols of DNN-CMED are designed, including secure linear computation protocol and secure nonlinear computation protocol. The classification process of DNN-CMED is given based on the above protocol. Through safety analysis and experiments, it shows that the models have better security and practicality. The results show that this model has better results in terms of accuracy, time and communication overhead, and on the MNIST test set.
In this paper, 3D image data obtained by Autodesk Map 3D flatbed scanning system was used to automatically extract point cloud data and conduct point cloud modeling. Firstly, this paper proposes a contour rendering me...
详细信息
ISBN:
(数字)9798350393682
ISBN:
(纸本)9798350393699
In this paper, 3D image data obtained by Autodesk Map 3D flatbed scanning system was used to automatically extract point cloud data and conduct point cloud modeling. Firstly, this paper proposes a contour rendering method based on elevation cloud image of data points to extract control points from the whole cloud image region. Then this paper uses the OpenGL graphics library provides rich graphics program interface. According to the principle of minimum distance and maximum Angle, a triangulation irregular network (TIN) model is established. Then, the elevation of data points and color RGB values are corresponded one by one to generate the elevation cloud map of data points. The experimental simulation analysis shows that the application of this method expands the search range of control points and increases the number of control points. In this way, the linearity of the contour map is better and more consistent with the actual terrain.
This paper studies continuous-Time model-free adaptive control (MFAC) framework for solving set-point tracking problems of second-order nonlinear time-invariant plants. First, the dynamical linearization process for c...
详细信息
This paper presents algorithms for performing data-driven reachability analysis under temporal logic side information. In certain scenarios, the data-driven reachable sets of a robot can be prohibitively conservative ...
详细信息
ISBN:
(纸本)9781728196817
This paper presents algorithms for performing data-driven reachability analysis under temporal logic side information. In certain scenarios, the data-driven reachable sets of a robot can be prohibitively conservative due to the inherent noise in the robot's historical measurement data. In the same scenarios, we often have side information about the robot's expected motion (e.g., limits on how much a robot can move in a one-time step) that could be useful for further specifying the reachability analysis. In this work, we show that if we can model this side information using a signal temporal logic (STL) fragment, we can constrain the datadriven reachability analysis and safely limit the conservatism of the computed reachable sets. Moreover, we provide formal guarantees that, even after incorporating side information, the computed reachable sets still properly over-approximate the robot's future states. Lastly, we empirically validate the practicality of the over-approximation by computing constrained, data-driven reachable sets for the Small-Vehicles-for-Autonomy (SVEA) hardware platform in two driving scenarios.
The increasing demand for large-scale, high-throughput network simulations poses significant challenges for the mainstream simulation tool NS3, particularly in parallel execution. The default NS3 framework, optimized ...
详细信息
ISBN:
(数字)9798331530808
ISBN:
(纸本)9798331530815
The increasing demand for large-scale, high-throughput network simulations poses significant challenges for the mainstream simulation tool NS3, particularly in parallel execution. The default NS3 framework, optimized for single-process operation, results in excessive simulation times and lacks an efficient node-to-process mapping strategy. Despite the existence of MPI-based parallelization techniques, a systematic methodology for distributing computational workloads remains unexplored. This study investigates the performance of NS3 simulations augmented with MPI in a multi-node cluster environment. Through empirical evaluation, we optimize node-process allocation strategies and demonstrate that leveraging MPI accelerates large-scale congestion control simulations in NS3, achieving substantial performance improvements.
Accurate and spatially reliable temperature data are crucial for effective agricultural management and climate adaptation. This study presents an innovative methodology for temperature correction using topographic fac...
详细信息
ISBN:
(纸本)9798350378115
Accurate and spatially reliable temperature data are crucial for effective agricultural management and climate adaptation. This study presents an innovative methodology for temperature correction using topographic factors and adiabatic models. Initial temperature information, sourced from the automatic weather stations (AWS) of the Dirección General de Aeronáutica Civil (DGAC), often has limited spatial representation, covering approximately 30 km in radius, however, significant temperature variations within this area require refined data for precise agronomic applications. This research proposes a method to spatialize and correct temperature data to enhance its reliability for agronomic use. More accurate spatial temperature maps are achieved by utilizing Digital Elevation Models (DEM) and applying corrections restricted to locations within 100 meters of elevation difference from the AWS. These corrections include wet and dry adiabatic lapse rates, which are applied appropriately. The corrected temperature data are then used to create detailed spatial maps, essential for modeling agronomic variables in changing environments. These maps enable better decision-making for irrigation scheduling, crop stress monitoring, and other critical agricultural practices. The methodology was tested in four distinct sites within the central-southern macrozone of Chile, specifically in the regions of O'Higgins, Maule, Nuble, and Biobío. The results from these case studies provided valuable insights into the applicability and accuracy of this technique in real-world agricultural settings, demonstrating significant improvements in climate resilience and sustainability. By improving the spatial accuracy of temperature data, this methodology supports more effective resource management and enhances the sustainability of agricultural operations. The findings highlight the potential of advanced topographic and adiabatic corrections to transform temperature dataanalysis, providing a valua
Based on the requirements of civil aircraft, this paper makes a deep study on the topology model and change control method of aircraft design data to solve the problem of uncomprehensive impact analysis of requirement...
详细信息
ISBN:
(数字)9798350352252
ISBN:
(纸本)9798350352269
Based on the requirements of civil aircraft, this paper makes a deep study on the topology model and change control method of aircraft design data to solve the problem of uncomprehensive impact analysis of requirements change. Due to the high complexity, interdisciplinary, highly integrated, long development life-cycle and diverse customer needs of civil aircraft, effective management and control of requirements change is very important. With the progress of complex engineering projects, the cost of requirement change increases exponentially, so it is important to identify and control the change in the early stage. Requirements changes can be caused by a variety of factors and need to be evaluated, approved and implemented to minimize change costs and risks and meet customer expectations. This paper introduces the topological relationship between the civil aircraft itemized requirement model and the design data, and forms a requirement change impact analysis model to help the requirement engineer assess the impact of the requirement change in the whole life cycle process and the degree of impact. And it can reduce the risk of repeated changes, prolonged life-cycles, and increasing costs due to insufficient change impact analysis. Through this research, it is expected to improve the efficiency and quality of civil aircraft development and reduce project risks.
Communication-based train control (CBTC) has become the main technical format of urban rail transit signal system. All kinds of software and hardware control logic in CBTC system need to follow the mechanism of rail t...
详细信息
The growing demand for accurate control in varying and unknown environments has sparked a corresponding increase in the requirements for power supply components, including permanent magnet synchronous motors (PMSMs). ...
详细信息
During surgery, invasive systolic blood pressure is an important basis for doctors to judge the patient's life state, which will directly affect the security of the surgery. Accurately predict the changes of invas...
详细信息
ISBN:
(纸本)9781450395670
During surgery, invasive systolic blood pressure is an important basis for doctors to judge the patient's life state, which will directly affect the security of the surgery. Accurately predict the changes of invasive systolic blood pressure during general anesthesia help to reduce the risk of surgery. In order to cope with the increasing surgical risk by fluctuations of invasive systolic blood pressure, this paper optimized and combined the traditional machine learning algorithm, and put forward a new fusion algorithm to predict the invasive systolic blood pressure after general anesthesia. In the modelingprocess, the patients' basic physical conditions, disease status, and intraoperative data collected by monitoring instrument during the surgical preparation stage were used as characteristic variable. In this paper, Linear Regression, Support Vector Machine Regression, Lasso Regression, and Ridge Regression were used to establish the new fusion algorithm. When the absolute error within 15mmHg, the fusion algorithm's predicting accuracy of invasive systolic blood pressure after general anesthesia reached 91.5%. The accurate prediction of invasive systolic blood pressure after general anesthesia in the preparation stage provides sufficient time for doctors to respond and reduces the risk of surgery to a certain extent.
暂无评论