Xiaoxing Gao

A P2P Network Topology Optimized Algorithm Based on Minimum Maximum K-Means Principle

The most popular routing algorithm in p2p network is that every node keeps a route table which records a certain number of other nodes. Such algorithm selects neighbor nodes at random, and this reduces routing efficiency. In this paper a new algorithm of optimizing p2p overlay network topology is proposed, which is based on minimum maximum k-means principle. According to the communication history, the division presents a steady network topology with k clusters and k cluster center nodes. This optimized p2p network topology presents a high performance of routing efficiency. The minimum maximum principle makes the division more appropriate, so that there will be less probability of being unsteady for the p2p network. The conclusions show that this algorithm is an efficient p2p network topology optimized algorithm.

Research on Traffic Information Service Grid Based on GT4

With the rapid prevalence of grid technologies, more attentions have to be focused on grid applications. The paper presents a framework for traffic information service grid (TISG) in order to eliminate information islands and to implement resource sharing and collaboration in the intelligent transportation systems (ITS). Based on Red Had Linux, Globus Toolkit 4.0 (GT4), and other related tools, the paper proposes the implementation model of TISG and investigates the key techniques to build TISG. Therefore, an application platform of traffic grid is built, which conforms to the standard of OGSA in the aspects of common use, convenience, efficiency, and security. The research on traffic laws is conducted in TISG. Applying grid to ITS will promote the development and popularization of grid applications.

Frequent Items Computation over Uncertain Wireless Sensor Network

There is an increasing interest in uncertain and probabilistics databases arising in application domains such as sensor networks, information retrieval, mobile object data management, information extraction, and data integration. A range of different approaches have been proposed to find the frequent items in uncertain database. But there is little work on processing such query in distributed, in-network inference, such as sensor network. In sensor network, communication is the primary problem because of limited batteries. In this paper, a synopsis with minimum amount tuples is proposed, which sufficient for answering the top-k query. And this synopsis can be dynamic maintained with new tuples been added. A novel communication efficient algorithm is presented in taking advantage of this synopsis. The test results confirm the effectiveness and efficiency of our approaches.

Design and Implementation of Drill Hole of PCB Detection System

In this paper, digital image processing is used for detecting and analyzing the Printed Circuit Board’s drill hole and the results is given to determine if PCB(Printed Circuit Board) meets the industrial production requirement. High-resolution optical camera sensors and image processing algorithms are selected for high-precision detection of the position and the number of PCB’s drill hole. The system provides the automatic detection and monitoring on the drill hole of circuit board, with high stability and practicality. And it is able to meet the requirements of the practical applications of PCB manufacturer.

A New Routing Protocol of Structured Peer-to-Peer Overlay Networks

The routing protocol of structured peer-to-peer overlay networks influences the p2p network performances in route table maintenance, routing hops, and network churning. Current protocols we used usually maintain a O(logN) route table and O(logN) routing hops. To get better routing speed with less scale routing table, using O(1) node finger to get O(N) route path, a new protocol named CSSP was presented. A short routing table was made to record ldquoLrdquo cache nodes, one super node and one successor node. Algorithms for maintaining the route table was presented in this paper. A selection algorithm was presented to select the super node who could cache all of the nodes. And the cache-nodes, super-node and successor-node ensure the high performance of the CSSP. Compared to chord, pastry and etc., the CSSP increases the performance in routing table maintaining, routing hops, fault-tolerance and the amount of churning nodes while some nodes inject and eject the CSSP p2p system. The simulations and analysis shows that the CSSP is an efficient route algorithm, but also some problems waiting for further study.

A P2P Overlay Network Routing Algorithm Based on Group-Average Agglomerative Clustering Topology

The use of peer-to-peer (p2p) applications is growing dramatically. Most p2p applications use the routing algorithm that selecting the neighbor nodes at random. This routing algorithm increases the routing hops and reduces the routing speed. To get better routing speed with less routing hops, a new routing algorithm named RGAAC based on group-average agglomerative clustering was presented. Structured by k clusters topology, this p2p network has k cluster center nodes. The cluster center nodes ensure the high performance of the RGAAC. Compared to some other p2p overlay routing algorithms, the RGAAC increases the performance in routing efficiency. The conclusions show that the RGAAC is an efficient routing algorithm, but also has some problems waiting for further study

Query Optimization over Distributed Data Stream

Recent research efforts in the fields of data stream processing show the increasing importance of processing data streams, e.g., in the e-science domain. Together with the advent of peer-to-peer (P2P) networks and grid computing, this leads to the necessity of developing new techniques for distributing and processing continuous queries over data streams in such networks. These systems often have to process multiple similar but different continuous aggregation queries simultaneously. Since executing each query separately can lead to significant scalability and performance problems, it is vital to share resources by exploiting similarities in the queries. The challenge is to identify overlapping computations that may not be obvious in the queries themselves. In this paper, we propose a novel algorithmic solution for problem of finding the minimum number of queries in such a distributed-streams setting, in order to optimize the communicate cost across the network. The experiment result show that our approach gives us as much as magnitude performance improvement over the no-share settings.

The description of Serial ATA bus_ Protocol and the design of Serial ATA bus control chip HPT183

In a PC system, external storage interface is still a bottleneck in spite of its continuous improving performance, in contrast to the fast development of CPU, memory, graphic chips. The transfer rate in ATA protocol has been improved drastically from the beginning 3.3MB/s to current 133MB/s, but the plate electrode of a parallel interface is inevitably puzzled by clock skew, which limit the increasing of frequency and transfer rate can not be improved. Serial ATA protocol is compatible with Parallel ATA protocol in software layer. Its transfer rate is improved greatly due to serial interface with embedded clock. This paper will discuss the differences between Parallel ATA protocol and serial ATA protocol, and describe the hierarchical classification of serial ATA protocol model. Last a design for HPT183, a parallel/serial ATA bridge connection chip, will be put forward and the test performance index for this chip is also provided.

High Speed Serial Interface & Some Key Technology Research

In a PC system, external storage interface is still a bottleneck in spite of its continuous improving performance. This makes I/O interface a bottleneck of high performance systems, especially graphics and disk storage. Increase the bandwidth of an I/O bus may be achieved by either increase bus width (e.g.16b to 32b) or by increasing operating frequency. Parallel interface is difficult to be implemented because of its nature drawback, e.g., Technologies such as embedded clock, point-to-point linkage, low voltage differential signal transmission and data encoding, enables gigabit serial bus get reliable high transfer rate at practical long distance, and become the next generation of interconnection interfaces. This paper will discuss the differences between parallel ATA protocol and serial ATA protocol, and describe the hierarchical classification of serial ATA protocol model. Last a design for a parallel/serial ATA bridge connection chip will be put forward and the test performance index for this chip is also provided.