2011 Technical Report

Code

TR-IIS-11-001

Subject / Author / Abstract

Cyber-Physical Elements of Disaster Prepared Smart Environment
J. W. S. Liu, C. S. Shih and E. T. H. Chu

Recent years have ushered in tremendous advances in information and communication technologies (ICT) and infrastructures for disaster management. Still, daily news on disasters has been telling us that even people in technologically advanced regions remain ill prepared. Smart and intelligent environments now offer us an increasingly broader spectrum of devices and services for comfort and convenience, safety from intruders, and social connectivity but little or nothing to help us to improve our readiness against killer tornados, major earthquakes, landslides, floods and so on. This paper first describes scenarios based on recent calamities to illustrate the need for enhancing our environments for disaster preparedness. It then describes examples of standard-based cyber-physical devices, systems and applications that can help to minimize loss of lives and damages to property. The paper concludes with opportunities and challenges to make such devices and systems dependable and affordable enough to be used pervasively as parts of future smart living environments.

View

Fulltext

Code

TR-IIS-11-002

Subject / Author / Abstract

Workflow Architecture for Model-Based Development of User-Centric Automation and Assistive Devices
T. Y. Chen, Y. C. Huang, C. S. Shih, T. W. Kuo and J. W. S. Liu

This paper describes an on-going effort in building user-centric automation and assistive devices from workflows. Advantages of this approach include that workflow-based devices and systems are naturally componentized and easily configurable. Models of a new device, user actions and users defined in terms of activities and workflows are executable. We can use them to simulate the device and its interactions with the user for requirement capture and design purposes. Later on in the development process, programs and other resources required by the device workflows become available. By using them, the workflow model of the device becomes its implementation. The workflows used to model user actions define the use scenarios and scripts for testing and evaluation.

View

Fulltext

Code

TR-IIS-11-003

Subject / Author / Abstract

Propagation-Based Image and Depth Reconstruction from A Space-Variant Defocused Image
Chih-Tsung, Shen, Wen-Lian Hwang, Yi-Ping Hung, and Soo-Chang Pei

In this paper, we present a propagation-based method to reconstruct image and depth from a space-variant defocused image. Our proposed method mainly contains three parts: blur identification, image deconvolution and artifact removal. For blur identification, we adopt belief-propagation to obtain the initial patchbased depth map. We divide the input images into patches and form a Markov network to assign each patch a node. To each node, we deblur the corresponding image patch with total variation regularizer scale by scale. Then we form the compatibilities of belief-propagation by calculating the reconstruction error term among these deblurred image patches, the curvature of these reconstruction error, and the difference between the overlapping deblurred regions. After several iterations, we can obtain an initial patch-based depth map. For image deconvolution, we propose a deblurring algorithm called TTV/L2 model with both benefits of Tikhonov-like regularizer and total variation (TV) regularizer. For the artifact removal, we deal with two traditional artifacts: boundary artifacts and ringing artifacts. To avoid the boundary artifacts, we deconvolute each patch with its neighboring patches and extract the deconvoluted part from the center patch. In the other hand, to avoid the ringing artifacts, we propose an edge-aware deringing algorithm to obtain a pixel-based depth map meanwhile reconstruct the all-in-focus result image. Although to reconstruct image and depth from a defocused image is spacevariantly ill-posed and extremely challenging, the experimental results of both natural images and medical images demonstrate our proposed method is promising.

View

Fulltext

Code

TR-IIS-11-004

Subject / Author / Abstract

An Integrated Network Mobility Management and Call Admission Control Scheme for Internet Access on High-speed Trains
Cheng-Wei Lee, Tsung-Yin Lee, Meng Chang Chen and Yeali S. Sun

Automotive telematics has become an important capability of high-speed rail systems, which are increasingly popular in the era of green technology. As train speeds increase, communications between devices on the train and devices outside the train encounter difficulties, and maintaining high quality communication is a major challenge. Moreover, handovers on high-speed trains occur more frequently, and have shorter permissible handling times. In this paper, the proposed 2MR scheme takes the advantage of the physical size of high-speed trains to deploy two mobile routers (MRs) in the first and final carriages. This scheme offers a protocol to allow the two MRs and wireless network infrastructure to cooperate in providing a seamless handover. The 2MR Call Admission Control (CAC) scheme gives integrated admission controls for applications of different priority classes. Our simulation results demonstrate that the 2MR scheme ensures QoS provisioning of admitted sessions, and reduces handover latency as well as packet loss for high-speed trains.

View

Fulltext

Code

TR-IIS-11-005

Subject / Author / Abstract

Compressive Image Sensing: Turbo Fast Recovery with Lower-Frequency Measurement Sampling
Chun-Shien Lu, Hung-Wei Chen, Soo-Chang Pei

In order to get better reconstruction quality from compressive sensing of images, exploitation of the dependency or correlation patterns among the transform coefficients has been popularly employed. Nevertheless, both recovery quality and recovery speed are not compromised well. In this paper, we study a new image sensing technique, called turbo fast compression image sensing, with computational complexity O(m²), where m denotes the length of a measurement vector y=ϕx that is sampled from the signal x of length n via the sampling matrix ϕ with dimensionality m⨯n In order to leverage between reconstruction quality and recovery speed, a new and novel sampling matrix is designed. Our method has the following characteristics: (i) recovery speed is extremely fast due to a closed-form solution is derived; (ii) certain reconstruction accuracy is preserved because significant components of x can be reconstructed with higher priority via an elaborately designed ϕ. Our method is particularly different from those presented in the literature in that we focus on the design of a sampling matrix without relying on exploiting certain sparsity patterns. Simulations and comparisons with state-of-the-art CS methodologies are provided and demonstrate the feasibility of the proposed method in terms of reconstruction quality and computational complexity.

View

Fulltext

Code

TR-IIS-11-006

Subject / Author / Abstract

Learning Boolean Functions Incrementally
Yu-Fang Chen and Bow-Yaw Wang

Classical learning algorithms for Boolean functions assume that unknown targets are Boolean functions over fixed variables. The assumption precludes scenarios where indefinitely many variables are needed. It also induces unnecessary queries when many variables are redundant. Based on a classical learning algorithm for Boolean functions, we develop two learning algorithms to infer Boolean functions over enlarging sets of ordered variables. We evaluate their performance in the learning-based loop invariant generation framework.

View

Fulltext

Code

IM-IIS-11-001

Subject / Author / Abstract

SQLMR : A Scalable Database Management System for Cloud Computing
Meng-Ju Hsieh, Chao-Rui Chang, Li-Yung Ho, Jan-Jan Wu, Pangfeng Liu

View

Not for public 原文請洽圖書室

Code

IM-IIS-11-002

Subject / Author / Abstract

Inter-Thread Dataflow: from Testing to Fault Localization
Meng-Ju Hsieh, Yi-Fan Tsai, Pen-Chung Yew, Yeh-Ching Chung

View

Not for public 原文請洽圖書室

Code

IM-IIS-11-003

Subject / Author / Abstract

Workload-Aware VM Dynamic Provision Strategy for Cloud Computing
Ching-Chi Lin, Pangfeng Liu, Jan-Jan Wu

View

Not for public 原文請洽圖書室

Institute of Information Science, Academia Sinica

Library

2011 Technical Report