Multi-way stream joins with expensive join predicates lead to great challenge for real-time (or close to real-time) stream processing. Given the memory- and CPU-intensive nature of such stream join queries, scalable processing on a cluster must be employed. This paper proposes a novel scheme for distributed processing of generic multi-way joins with window constraints, called Pipelined State Partitioning (PSP). We target generic joins with arbitrarily join conditions, which are used in non-trivial stream applications such as image matching and biometric recognizing. The PSP scheme partitions the states into disjoint slices in the time domain, and then distributes the fine-grained states in the cluster, forming a virtual computation ring. Compared to replication-based distribution of non-equi-joins, PSP scheme is superior since: (1) zero state duplication and thus no repeated computations, (2) pipelined processing of every input tuple on multiple nodes to achieve low response time, and (3) cost-based adaptive workload distribution. We have implemented the proposed PSP schemes within the CAPE DSMS. Our experimental study demonstrates the significant performance improvements compared to the state-of-the-art generic distributed stream join algorithms.

Stream Processing

Join (topology)

10.1145/1516360.1516396

Cite

Citations (30)

An improved method for odometry estimation based on EKF and Temporal Convolutional Network

Physical Communication (2020)

Xiaokun Leng Songhao Piao Song Wang Lin Chang Zheng Zhu

Odometry

Odometer

Torso

Sensor Fusion

Shake

10.1016/j.phycom.2020.101178

Cite

Citations (3)

Distributed adaptive control of mobile robots with unknown parameters

2022 34th Chinese Control and Decision Conference (CCDC) (2023)

Xiaojun Cheng Jiangshuai Huang Tingting Gao Song Wang Meng Zhang

In this paper, formation control problem of second-order mobile robots with uncertain model and reference trajectory parameters is investigated. It is assumed that the multiagent systems have the same structure of each subsystem, but the the model parameters are different and unknown, and only some of the systems can obtain the desired trajectory information. For this system, the communication topology of the system is represented by a directed graph, and the local consistency error is introduced. Based on the backstepping method, parameter estimators are designed, which estimate the unknown parameters in the model and the desired trajectory based on limited information, and the adaptive controller is also designed which achieves asymptotically tracking of the robot for the desired trajectory. Finally, the formation control of the multi-agent systems on the desired trajectory is realized through simulation, and the feasibility and effectiveness of the control strategy are verified.

10.1109/ccdc58219.2023.10327357

Cite

Citations (0)

A dynamic inversion controller design for miniature Unmanned Aerial Vehicles

Song Wang Juan Li Sihong Wang

This article focuses on the nonlinear control issue of miniature Unmanned Aerial Vehicles(UAVs). A complete nonlinear model for miniature fixed-wing UAVs is structured and simplified as an affine nonlinear system aiming at the fast dynamic loop of key attitude control. We adopt the state feedback method to dynamic inversion controller design. To compensate the external disturbance and modeling error, an adaptive compensation algorithm is discussed and an online estimation way for the error coefficient and the external disturbance is presented, which guarantees the asymptotic stability. Simulation result shows that the controller is of short response time and small overshoot, and can achieve decoupling between channels, while the adaptive algorithm can converge quickly with automatic compensation for external disturbances, not well equipped aircraft, and parameters error and so on.

Overshoot (microwave communication)

Decoupling (probability)

10.1109/cecnet.2011.5769014

Cite

Citations (5)

Video In Sentences Out

arXiv (Cornell University) (2014)

Andrei Barbu Alexander Bridge Zachary Burchill Dan Coroian Sven Dickinson

We present a system that produces sentential descriptions of video: who did what to whom, and where and how they did it. Action class is rendered as a verb, participant objects as noun phrases, properties of those objects as adjectival modifiers in those noun phrases, spatial relations between those participants as prepositional phrases, and characteristics of the event as prepositional-phrase adjuncts and adverbial modifiers. Extracting the information needed to render these linguistic entities requires an approach to event recognition that recovers object tracks, the trackto-role assignments, and changing body posture.

Adverbial

Specifier

Verb phrase

Event structure

Endocentric and exocentric

10.48550/arxiv.1408.6418

Cite

Citations (0)

Multi-Dimensional Spatial Transformation Network Topology Layout

Journal of Computer-Aided Design & Computer Graphics (2022)

Ting Cai Song Wang Liang Liu Siheng Yi Yongguo Han

The network topology layout helps users intuitively understand and analyze the network structure. With the increase of network scale and network attributes, a multi-dimensional spatial transformation network topology layout method is proposed to guide users to conduct progressive exploration analysis and give consideration to the presentation of multivariate network structural features and attribute correlation analysis of local networks. Firstly, the method makes 3-dimensional layout based on the characteristics of the community and presents the overall structure characteristics of the network. Based on the 3-dimensional layout and network attributes, the network structure filtered by users is transformed into dimensional space, and the attributes and topological associations are presented in the 2.5-dimensional layout. Finally, 2-dimensional layout is carried out for the single-dimensional attribute local network filtered by users, and the topology of the local network is presented through a diversified layout. Based on this topology layout method, three data sets of social network, brain network and computer network are used for experimental analysis. The results demonstrate the effectiveness of proposed method in both structural feature presentation and attribute association analysis, as well as progressive exploration analysis.

10.3724/sp.j.1089.2022.19195

Cite

Citations (0)

Methanobacterium aridiramus sp. nov., a methanogen isolated from potential methane hydrate bearing area offshore southwestern Taiwan

INTERNATIONAL JOURNAL OF SYSTEMATIC AND EVOLUTIONARY MICROBIOLOGY (2025)

K.-C. Lee Mei‐Chin Lai Saulwood Lin Shu-Jung Lai Weiling Zhang

A novel mesophilic, hydrogenotrophic methanogen, strain CWC-01 T , was isolated from the sediment sample of the piston core collected at station 3 at Good Weather Ridge area offshore southwestern Taiwan from ORI-902A cruise in 2009. Cells of strain CWC-01 T are rod-shaped, measuring 0.4–0.5 µm in width and 1.0–2.0 µm in length. Strain CWC-01 T utilizes formate and H 2 /CO 2 to produce methane, but not secondary alcohol, methanol, ethanol, methylamines or acetate as catabolic substrates. The optimal growth conditions are 37 °C, 0.043 M NaCl and pH 6.3. Growth effectors tests indicate tryptone, but not acetate and yeast extract, could stimulate the growth of strain CWC-01. Phylogenetic analysis of 16S rRNA gene reveals that strain CWC-01 T is most closely related to Methanobacterium petrolearium Mic5c12 T , with 96.63% identity. Genome size of strain CWC-01 T is 1.98 Mb and it is the smallest genome size of genus. The genomic DNA G+C content obtained from the genome sequence of strain CWC-01 T is 44.15 mol%. Based on these phenotypic, phylogenetic and genomic analyses, we propose that strain CWC-01 T represents a novel species in genus Methanobacterium , for which the name Methanobacterium aridiramus sp. nov. is proposed. The type strain is CWC-01 T (=BCRC AR10053 T =NBRC 113991 T ).

Methanogen

Methanobacterium

Strain (injury)

10.1099/ijsem.0.006658

Cite

Citations (0)

Systemic Thinking on Services Science, Management and Engineering: Applications and Challenges in Services Systems Research

IEEE Systems Journal (2013)

Song Wang Ling Li James D. Jones

As the world increasingly becomes characterized by the provision of services to customers, service research faces many challenges and needs a new vision. Many of these challenges stem from the fact that technologies are continually experiencing dramatic change, and savvy customers are increasingly demanding more qualified services. Although research and experiential findings on services in many disciplines have achieved impressive results, systematic research on modern services systems is rarely considered. Aiming at services systems (not only services themselves), this paper discusses the contributions and the potentials for the field of systems theory for Services Science, Management and Engineering (SSME). Thus, SSME can provide suitable referential clues for discussing impacts of systems theory on services research, and systemic thinking can be applied to discuss key issues in SSME. Providing basic definitions of services and services systems, this paper elaborates on the research objects and methodologies of SSME from systems perspective, and proposes that systems theory can be applied as the foundations of services engineering. This paper further sheds light on systemic issues in services science and engineering, services operations, and services systems management. It also provides a summary of relevant theories and tools applied to services systems at the systems level. Finally, some potential prospects in SSME research are presented to meet research demands and practical challenges.

Systems Thinking

Services computing

Medical services

Health systems engineering

10.1109/jsyst.2013.2260622

Cite

Citations (15)

Research on Triangular Mesh Simplification Algorithm of Virtual Object Model

Applied Mechanics and Materials (2012)

Yang Ping Wang Jian Wu Dang Jin Yu Yang Song Wang

To improve the model quality of mesh simplification, reduce the model deformation and the lost of the detail characteristics with multiple folding, the paper represents a triangular mesh simplification algorithm based on edge-surface folding. The algorithm introduces curvature characteristics of model vertexes into the weight calculation of effective points, and introduces weight of triangular facets at the same time. The facets with the smallest weight adjacent the candidate edge is shrunk. The experiment results show that the algorithm has provided a higher quality of model simplification and effectively kept the graphic features.

Triangle mesh

Folding (DSP implementation)

T-vertices

10.4028/www.scientific.net/amm.271-272.1410

Cite

Citations (3)