The throughput results for file transfers using file sizes ranging from 1 Kbytes through 1 Mbytes using both the standard TCP/IP and SCPS protocol stacks over a PPP link are reported. Channel properties were simulated...
详细信息
ISBN:
(纸本)0780365992
The throughput results for file transfers using file sizes ranging from 1 Kbytes through 1 Mbytes using both the standard TCP/IP and SCPS protocol stacks over a PPP link are reported. Channel properties were simulated using a space channel simulator with a range of balanced and unbalanced link speeds and channel error rates. The throughput results will show the effects of link configuration and channel error rate on file transfer time. The host computer configuration options for the protocols are factored into the comparison. The throughput reporting shows the effects of header compression and selection of congestion algorithm upon the results. The TCP/IP ftp and SCPS-FP using VJ congestion control algorithm results give similar results and better results than SCPS-FP with the Vegas congestion control algorithm in these experiments. No noticeable delay effects were noted with links delays corresponding to GEO orbits with file transfers of 1 Mbytes.
Existing data models and design principles for temporal databases are based on the assumption that a temporal data is associated with an interval as well as its sub-intervals. This type of temporal data is called homo...
详细信息
Existing data models and design principles for temporal databases are based on the assumption that a temporal data is associated with an interval as well as its sub-intervals. This type of temporal data is called homogeneous data. A temporal data which is associated with an interval, but not its sub-intervals, is called a non-homogeneous data. Existing data models cannot capture non-homogeneous data accurately, and the existing design principles are not applicable in temporal databases that contain non-homogeneous data. In this dissertation, the relational data model is extended to support both homogeneous and non-homogeneous data. A design principle which avoids inconsistency in temporal databases, that contain homogeneous data as well as non-homogeneous data, is studied. In the proposed extension of the relational data model, temporal relations are classified into two types; property relations and representative relations. A tuple in a property relation is associated with the valid time and its sub-intervals, while a tuple in a representative relation is associated with only the whole interval of its valid time. Thus, the valid time in a property relation is decomposable, but the valid time in a representative relation is not. Based on this characteristics, the valid time in a property relation and that in a representative relation cannot be used in the same manner. In the extension of relational algebra for temporal relations, the calculation of the valid time in a relation, created by relational operators, is determined by the types of the temporal relations. Thus, it guarantees proper use of the valid time. A type of inconsistency, called P-inconsistency, can occur in temporal databases with homogeneous and non-homogeneous data. A normal form, called P-consistency Normal Form (PCNF), which avoids P-inconsistency, is proposed in this dissertation. PCNF is based on types of attributes, functional dependencies, and property dependencies (P-dependencies), in tempor
The arrival of cyber-physical system era is changing data analysis in many ways. Driven by the advances in Internet and sensor techniques, the amount of multidimensional contents, such as images, trajectories, video c...
详细信息
The arrival of cyber-physical system era is changing data analysis in many ways. Driven by the advances in Internet and sensor techniques, the amount of multidimensional contents, such as images, trajectories, video clips, etc., has grown to an unprecedented level. Supporting multidimensional objects in large scale requires significant extensions from traditional databases. One critical issue is indexing and query processing. In this thesis, we discuss two important queries in massive multidimensional datasets: frequent path finding and high-dimensional similarity join. In the context of big data, these two queries are challenging due to two reasons: 1) they both contain expensive comparison operations, and 2) the complex structures of their interested data complicate the index design. To address these issues, we conduct theoretical analysis, advanced algorithm design, as well as extensive experiments on large-scale datasets. First, we address the problem of frequent path finding by proposing a new query named the time period-based most frequent path (TPMFP). Specifically, given a time period T, a source s, and a destination d, TPMFP query searches the most frequent path (MFP) from s to d during T. Though there exists several proposals on defining MFP, they only consider a fixed time period. Most importantly, we find that none of them can well reflect people’s common sense notion which can be described by three key properties, namely suffix-optimal (i.e., any suffix of an MFP is also an MFP), length-insensitive (i.e., MFP should not favor shorter or longer paths), and bottleneck-free (i.e., MFP should not contain infrequent edges). The TPMFP with the above properties will reveal not only common routing preferences of the past travelers, but also take the time effectiveness into consideration. Therefore, our first task is to give an TPMFP definition which satisfies the above three properties. Then, given the comprehensive TPMFP definition, our next task is to find TP
Domain novices learning about a new subject can struggle to find their way in large collections. Typical searching and browsing tools are better utilized if users know what to search for or browse to. In this disserta...
详细信息
Domain novices learning about a new subject can struggle to find their way in large collections. Typical searching and browsing tools are better utilized if users know what to search for or browse to. In this dissertation, we present Multiple Diagram Navigation (MDN) to assist domain novices by providing multiple overviews of the content matter using multiple diagrams. Rather than relying on specific types of visualizations, MDN superimposes any type of diagram or map over a collection of documents, allowing content providers to reveal interesting perspectives of their content. Domain novices can navigate through the content in an exploratory way using three types of queries (navigation): diagram to content (D2C), diagram to diagram (D2D), and content to diagram (C2D). To evaluate the MDN user interface, we conducted a user study, which showed that users found MDN useful and easy to use in exploratory-navigation scenarios. Encouraged by these positive results, we extended the functionality of MDN to provide a ranking of collection documents for D2C queries (expressed by a selected diagram concept). We studied different elements of the ranking process. As a case study, we targeted our research towards the Wikipedia collection. With the goal of studying ranking in different types of diagrams, we introduced two diagram models: the Items-and-Attributes model and the Universal model. We also studied two ranking algorithms: Personalized PageRank (PPR), an algorithm used in similar applications; and Greedy Energy Spreading (GES), an algorithm that we designed. We also studied different approaches to computing rankings for C2D queries. Our results show encouraging performance on the ranking of D2C and C2D queries. For example, in an experiment targeting diagrams conforming to the Items-and-Attributes model, results showed reasonably high similarity between a diagram concept selected by the user and the top-ten-ranked pages. We also found that diagrams had a strong influence
The problem of multi-agent formation control with H(infinity) robustness incorporated is addressed. A distributed control architecture is proposed for the formation system. The architecture consists of three layers wh...
详细信息
ISBN:
(纸本)9781424474271
The problem of multi-agent formation control with H(infinity) robustness incorporated is addressed. A distributed control architecture is proposed for the formation system. The architecture consists of three layers where the control algorithms are divided into two parts. One of them is the cooperative part with an information flow law that endows the overall system with cooperative behaviors. Studies on the information flow law focus on discrete consensus algorithms based on the tridiagonal toeplitz matrix. The other part is the stabilization part with control law that robots depend on to maintain the formation while moving. Upon the proposed structure, a multi-agent formation system is modeled as a spatially interconnected system and spatially distributed controllers that are robust to external disturbances are developed. The effectiveness of the proposed architecture, as well as the distributed controllers, is verified by a group of wheeled robots.
For anonymous or new users on shopping websites, when their historical interaction information is unknown, a model can be established based on the user's historical behavior sequence data to mine their long-term i...
详细信息
In the age of Big Data, efficient algorithms are in higher demand more than ever before. While Big Data takes us into the asymptotic world envisioned by our pioneers, it also challenges the classical notion of efficie...
详细信息
ISBN:
(数字)9781680831313
ISBN:
(纸本)9781680831306
In the age of Big Data, efficient algorithms are in higher demand more than ever before. While Big Data takes us into the asymptotic world envisioned by our pioneers, it also challenges the classical notion of efficient algorithms: algorithms that used to be considered efficient, according to polynomial-time characterization, may no longer be adequate for solving today"s problems. It is not just desirable but essential that efficient algorithms should be scalable. In other words, their complexity should be nearly linear or sub-linear with respect to the problem size. Thus, scalability, not just polynomial-time computability, should be elevated as the central complexity notion for characterizing efficient computation. Scalable algorithms for Data and Network Analysis surveys a family of algorithmic techniques for the design of scalable algorithms. These techniques include local network exploration, advanced sampling, sparsification, and geometric partitioning. They also include spectral graph-theoretical methods, such as are used for computing electrical flows and sampling from Gaussian Markov random fields. These methods exemplify the fusion of combinatorial, numerical, and statistical thinking in network analysis. Scalable algorithms for Data and Network Analysis illustrates the use of these techniques by a few basic problems that are fundamental in analyzing network data, particularly for the identification of significant nodes and coherent clusters/communities in social and information networks. It also discusses some frameworks beyond graph-theoretical models for studying conceptual questions that arise in network analysis and social influences.
A new fast supervisory predictive control algorithm (FSPC) is proposed in this paper. In this algorithm, off-line approximate computation of the future optimize variable and only calculating of the current control act...
详细信息
Data mining is the process of extracting hidden patterns from data. As more data is gathered, with the amount of data doubling every three years, data mining is becoming an increasingly important tool to transform thi...
详细信息
ISBN:
(数字)9781614707974
ISBN:
(纸本)9781607412892
Data mining is the process of extracting hidden patterns from data. As more data is gathered, with the amount of data doubling every three years, data mining is becoming an increasingly important tool to transform this data into information. It is commonly used in a wide range of profiling practices, such as marketing, surveillance, fraud detection and scientific discovery. Consequently, data management is the development, execution and supervision of plans, policies, programs and practices that control, protect, deliver and enhance the value of data and information assets. Data management comprises all the disciplines related to managing data as a valuable resource. This new and important book gathers the latest research from around the globe in these fields and relative topics such as:cognitive finance, data mining of the Indian mineral industry, managing building information models, a new co-training method for data mining, and others.
暂无评论