检索结果-内蒙古大学图书馆

EFFECT OF PREDICTION ACCURACY ON THE PERFORMANCE OF AN INSTRUCTION PIPELINE

INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE 1990年第5期21卷 977-984页

作者： BASU, A Centre for Development of Advanced Computing Bangalore 560025 India

A general weakness of instruction pipelines is ‘bubbles’ or ‘holes’ in the pipeline due to conditional branch instructions. Most of the solutions (Lee and Smith 1984) to the branch problem attempt to predict whether or not a branch will be taken. If the prediction is correct, then the correct sequence of instructions can be initiated without delay. The performance of an instruction pipeline is analyzed to give a measure of the efficiency or utilisation of the pipeline segments in terms of the probability of correct branch prediction (referred to as the prediction accuracy). The analysis is based on the space-time relationship. The performance of an instruction pipeline with both equal and unequal segment times is analysed incorporating the effect of two types of conditional branch instructions in the instruction sets of computers. The results of the analysis enable one to study the sensitivity of the performance of an instruction pipeline to the number of conditional branch instructions and the probability of correct branch prediction as well as to estimate the utility of the different branch prediction strategies that have been proposed recently (Lee and Smith 1984). Although it is extremely difficult to obtain the value of 1 for prediction accuracy, this study reveals that a value of 0.8 can give reasonably good values of utilization.

关键词： Computer Systems, Digital

来源：评论

学校读者我要写书评

暂无评论

Intelligent Malware Detection Using Sentence Transformer and Hybrid Deep Learning Architecture

Intelligent Malware Detection Using Sentence Transformer and...

引用

International Conference on Signal Processing and Communication (ICSC)

作者： Sanjeev Kumar Yudhishthira Sapru Sugandha Sapru B. Janet Cyber Security Technology Division Centre for Development of Advanced Computing Mohali India Indian Computer Emergency Response Team CERT-In Delhi India Department of Mathematics PGGC Panjab University Chandigarh India Department of Computer Application National Institute of Technology Tiruchirappalli India

ISBN: (数字)9798331511890

ISBN: (纸本)9798331511906

The complex evasion techniques adopted by the malware creator make signature-based malware detection approaches unable to detect unknown malware attacks. This study proposes a new malware detection and classification approach with significantly less dependence on manual feature engineering. The bidirectional encoder representations from transformers (BERT) are used to model the opcode sequences of the malware file as the semantic behavior of malware. After that, the vectorized feature vector is fed to the hybrid one-dimensional convolution neural network (1D-CNN) and the gated recurrent unit (GRU) model for feature reduction and normalization before training a Softmax classifier. A hybrid learning architecture is designed with a custom 1D-CNN and a GRU model to take the strengths of both types of networks. The proposed scheme is evaluated using the Microsoft BIG dataset, consisting of 1625 assembly files belonging to 9 distinct malware families. The proposed method produced 95.77% accuracy even with the smaller dataset, which we believe will boost performance by adding a more diverse dataset. It uses the next-generation BERT language model, which gives a promising future area for malware defenders. This research presents malware detection without much reliance on manual feature engineering or signature-based detection, making it suitable for detecting unknown malware attacks.

关键词： Training Semantics Manuals Bidirectional control Transformers Feature extraction Malware Encoding Vectors Next generation networking

来源：评论

学校读者我要写书评

暂无评论

Software-based fault tolerant amay

引用

IEEE Potentials 2006年第1期25卷 41-45页

作者： Saha, Goutam Kumar Centre for Development of Advanced Computing Kolkata

A low-cost software technique that detects and recovers multiple bit errors in a one-dimensional (1-D) array during the run time of an application is discussed. This technique is providing a practical tool for the system engineers for tolerating multiple transient bit errors in their valuable information. The purpose of fault tolerance is to increase the dependability of an application system. It uses an affordable redundancy with time and memory space. This technique does not require multiple versions of software based on design diversification. It is easy to implement in a scientific or engineering application and is useful for gaining better system safety. It is also useful for designing a dependable computer controlled system that refers much stored information in the form of arrays throughout its lifecycle. This approach can be also be used for multidimensional arrays with required modifications.

关键词： Fault tolerant computer systems

来源：评论

学校读者我要写书评

暂无评论

Identification of patients with de novo Parkinson's Disease from chemosensory EEG signals using ICEEMDAN domain Entropy Features

IEEE Sensors Letters

引用

IEEE Sensors Letters 2025年第6期9卷

作者： De, Sagnik Pavuluri, Sreenija Gupta, Anil Kumar University of Calcutta Institute of Radio Physics and Electronics West Bengal Kolkata India GVPCE Department of Computer Science and Engineering Andhra Pradesh Visakhapatnam India Centre for Development of Advanced Computing HPC I and E Group Maharashtra Pune India

Parkinson's disease (PD) is a progressive neurodegenerative disorder that impairs motor and sensory functions, with early symptoms often involving olfactory dysfunction. Given the importance of detecting these early biomarkers for timely intervention, this letter proposes the novel use of chemosensory EEG for the early detection of PD, as it captures the brain's responses to olfactory stimuli, one of the primary sensory modalities affected by the disease. The proposed method employs an Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) to decompose EEG signals into intrinsic mode functions (IMFs). Entropic features, including approximate entropy (AEn), sample entropy (SEn), and Rényi permutation entropy (RpEn), are extracted from these IMFs to identify distinguishing characteristics. These features are then evaluated using several machine learning classifiers. A comprehensive evaluation reveals that combining RpEn features with Least Squares Support Vector Machine (LS-SVM) classifier achieves optimal performance, with an accuracy of 96.47%, a precision of 96.14%, and a kappa score of 0.95. © 2017 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

A Student Centric Approach for Mobile Learning Video Content development and Instruction Design

A Student Centric Approach for Mobile Learning Video Content...

引用

15th International Conference on advanced Communications Technology (ICACT)

作者： Mandula, Kumar Meda, Srinivasa Rao Muralidharan, V Parupalli, Ramu Centre for Development of Advanced Computing India

ISBN: (纸本)9788996865018;9781467331487

With the recent advancement in mobile computing technologies, there has been a paradigm shift from electronic learning(e-Iearning) to distance learning(d-Iearning) to mobile learning(m-Iearning). M-Iearning not only provides educational oppurtunities through handheld devices, but also facilitates just in time information required at the right time and right place. Though m-Iearning provides mobility and instant access to education, there are some implementation challenges and issues like limited processing capability and small screen display that needs researchers attention. Keeping in view the latest trends in Information, Electronics and Communication Technologies(IECT) and the increased usage of mobile phones in India, this paper discusses mobile learning technologies, opportunities and challenges for successfull implementation of m-Iearning. It is observed that most of the students in India use low cost mobile phones that too with different hardware capabilities and features. Researchers and developers in education domain should consider these mobile capabilities and challenges before developing m-Iearning video content. Pilot experiment was conducted to study different mobile video presentation styles, file formats and their playback capabilities using students mobile phones. Students from various finishing schools were involved in carrying out this experiment. Based on pilot results, instructional design strategy and mobile video content development guidelines were formulated catering to the requirements of student community in India.

关键词： Mobile Learning Ubiquitous Learning Mobile Video Instruction Design Video Content

来源：评论

学校读者我要写书评

暂无评论

MULTICLASS QUEUING MODELS FOR PERFORMANCE ANALYSIS OF COMPUTER-SYSTEMS

引用

SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES 1990年第4期15卷 355-363页

作者： REGE, KM Centre for Development of Advanced Computing Poona University Campus Pune India

Queueing models, networks of queues in particular, have been found especially useful for estimating the performance of computer systems. Networks of queues with multiple customer classes provide a flexible framework for modelling computer systems, where a rich set of analytical results and techniques are available. When because of the complexity of the system being modelled the analytical results cannot be applied directly, they often point to fairly accurate approximation schemes. In this paper, we present a brief survey of some of the important results and techniques from the theory of multi-class queueing networks. We also present a case study to illustrate how these results and techniques are used in a real-life situation where many of the modelling constraints are violated.

关键词： QUEUING NETWORKS MULTICLASS QUEUES PERFORMANCE ANALYSIS OF COMPUTER SYSTEMS

来源：评论

学校读者我要写书评

暂无评论

Electronic authentication for e-Government services - A survey 10

Electronic authentication for e-Government services - A surv...

引用

10th IET System Safety and CyberSecurity Conference

作者： Dubey, Ankita Saquib, Zia Dwivedi, Surabhi Centre for Development of Advanced Computing Mumbai India Centre for Development of Advanced Computing Bangalore India

ISBN: (纸本)9781785610929

Electronic authentication (a.k.a e-Authentication) is the process of establishing confidence in a user's identity electronically presented to an information system. In today's world of exploding internet usage and adoption of e-Governance and e-Commerce initiatives in countries across the world, it is imperative that a user is identified with confidence. In the last 5 years or so, there have been many cases of fraud or misappropriation that could have been avoided had there been a proper e-Authentication system in place. In this paper we present a survey of different e-Authentication systems and standard guidelines being followed for Electronic Government in countries across the world. The aim of this paper is to do a comparative study of various national and international e-Authentication frameworks and to map them with the guidelines of National Institute of Standards and Technology (NIST) [13] for e-Authentication. We explored USDA e-Authentication [12], National e-Authentication Framework of Australian Government [5], Qatar's e-Authentication Framework [10], Organization for Economic Co-operation and development Recommendation on Electronic Authentication [9], e-Pramaan e- Authentication framework of India [7] and Norwegian FANR [8]. Recently, there has been an increasing emphasis on additional functionalities such as Single-Sign-On (SSO), two-way authentication to guard against phishing attacks;and federated electronic authentication. In this paper we also present a case study of a framework that takes into account federated architecture for authentication as well as SSO for e-Government and public services. e-Pramaan can also be integrated with mobile based signature service. © 2015, Institution of Engineering and Technology. All rights reserved.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Experience of porting and optimization of seismic modelling on multi and many cores of hybrid computing cluster 77

Experience of porting and optimization of seismic modelling ...

引用

77th EAGE Conference and Exhibition 2015: Earth Science for Energy and Environment

作者： Rastogi, R. Srivastava, A. Sirasala, K. Chavhan, H. Khonde, K. Centre for Development of Advanced Computing India

ISBN: (纸本)9781510806627

In this paper, we report our experience of porting and optimization of legacy seismic acoustic modelling application on multi and many cores of hybrid architecture of PARAM Yuva II. This application was developed using MPI and used domain decomposition as parallelization approach across parallel processors. The same application has been modified for domain decomposition at node level and the parallel performance was improved using OpenMP within the node. The resultant application was optimized using different optimization techniques for multi core architecture of Intel's Xeon, which further improved the performance of the application along with efficiency. The optimized application was then ported on many core architecture of Intel's Xeon Phi in native and symmetric modes. The details of porting, optimizations and execution on Intel's Xeon and on Xeon Phi in native and symmetric modes are given in the paper. Performance, scalability and efficiency of the application has been studied using multi and many cores and experimental results are presented.

关键词： Seismology

来源：评论

学校读者我要写书评

暂无评论

Invigilated Online Assessment: Various ways to Minimize Unauthorized Help

Invigilated Online Assessment: Various ways to Minimize Unau...

引用

IEEE Conference on Open Systems (ICOS) / IEEE Symposium on E-Learning, E-Management and E-Services (IS3e)

作者： Mothukuri, Uday Kumar Jain, Sandesh Muralidharan, V Centre for Development of Advanced Computing(C-DAC) India

ISBN: (纸本)9781467323895

In eLearning, the main advantage of using advanced technologies is to lower the burden on the instructor and increase the student's scope of learning. eLearning provides a wide spectrum of learning services from content delivery to online assessment. Besides offering various opportunities to the students in improving their learning, it is also making the ways easier for the students to perform malpractices during the assessment conducted in the absence of the instructor. This will lead the students to get high score without gaining the knowledge, deviating the education goal. This paper puts forward various proposed solutions of constructing reliable assessment models in eLearning environment. One of the main goals of this paper is to provide a secured online assessment that drives the use of one time and one place assessments to ubiquitous environment. This paper also discusses a proposed approach with the consideration of the existing solutions to minimize unauthorized access/help during invigilated online assessment.

关键词： elearning eAssessment invigilation authentication monitoring system cheating

来源：评论

学校读者我要写书评

暂无评论

Migration velocity analysis using multiple realisation of parallel RCGA

Migration velocity analysis using multiple realisation of pa...

引用

作者： Rastogi, R. Srivastava, A. Gholap, S. Majumder, S. Centre for Development of Advanced Computing India

ISBN: (纸本)9781617386671

Seismic imaging requires a good estimate of the velocity model. Velocity model estimation can be done using migration velocity analysis (MVA). Scientists have explored various iterative techniques for MVA. The problem exhibits inherent global nature where multiple parameters are to be optimized simultaneously which requires a powerful global optimization tool like genetic algorithm. It has been proven that GA can solve multi-parameter optimization problems in Geophysics successfully. GA can work even with poor or no initial guess and without derivatives. But even with GA it is difficult to prove that the solution we have arrived is a global solution. To address this problem, we have developed an automated algorithm to estimate interval depth velocity model using multiple realisation (run) of parallel real-coded genetic algorithm (RCGA). We have used multiple run of RCGA, to maximize the possibility of global solution. The developed algorithm is highly compute intensive, hence parallelized using Hybrid Island Model. The proposed algorithm has been tested for a synthetic prestack data set of graben model that demonstrates the effectiveness of our velocity analysis approach. The results and the performance of the algorithm are presented and discussed. © 2010, European Association of Geoscientists and Engineers.

关键词： Global optimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：