Browsing by Author "Yi, Chih-Wei"

Now showing 1 - 20 of 47

A Crowdsourcing-based Road Anomaly Classification System
(2016)
Wang, Ru-Yu
;
Chuang, Yi-Ta
;
Yi, Chih-Wei
;
資訊工程學系
;
Department of Computer Science
Road networks are the most important facility to the public transportation in modern cities. Governments around the world allocate large amounts of budgets for the pavement maintenance every year. In this paper, we proposed a crowd-sourcing solution to categorize road anomalies into safety related anomalies such as speed bumps and rumble strips, and dangerous anomalies such as bumps and potholes. The proposed system is composed of three parts: a smart probe car crowds (SPC-crowd) that serve as the anomaly data source; cloud servers that are the core for the anomaly classification; and application services that provide various innovative applications to facilitate the pavement maintenance. To support the crowdsourcing procedure, in the SPC-crowd side, we proposed cross-SPC techniques by adopting the underdamped oscillation model (UOM). In the cloud side, a supervised learning classification model was adopted on the anomaly data generated from the SPC-crowd. To validate the proposed system, extensive field trial was performed. The experimental results shown that our system can facilitate the pavement maintenance through the crowdsourcing solution.
A Preliminary Study on SPC-Crowd Pavement Indexing
(2015)
Wu, Tsung-Hsiang
;
Yi, Chih-Wei
;
Chan, Ching-Yao
;
Chang, Ya-Lan
;
Tseng, Chien-Chao
In most developed and developing countries, the maintenance of roadway pavement consumes considerable resources and represents a major item in government budget. However, due to the lack of sustainable and systematic approaches to implement long-term road pavement monitoring and assessment, the cost-effectiveness of performing such tasks is often a concern. In this paper, a crowdsourcing framework is proposed as a solution based on a smart probe car (SPC) system that utilizes mobile sensing technologies to collect environmental roadway data. Built upon the probe-vehicle system, the proposed method not only has the potential to save a significant amount of resources but also makes it possible to acquire real-time pavement information in a large scale.
A Study on Multiple Wearable Sensors for Activity Recognition
(2017-01-01)
Huang, Yu-Chuan
;
Yi, Chih-Wei
;
Peng, Wen-Chih
;
Lin, Hsing-Chen
;
Huang, Ching-Yu
In the past few years, human activity recognition is an active area of machine learning. The possible applications include daily activity monitoring for elders, exercise and fitness workout assistant systems, life style analysis, etc. In this work, tri-axial accelerometers were worn at the right wrist, left wrist and waist to collect motion data for activity recognition. Three supervised machine learning algorithms including random forests, decision trees and support vector machines were implemented to classify daily activities into running, walking, standing, sitting and dining from inertial data. The purposes of this study are to understand how good the machine learning algorithms can achieve and how the wearing location and number of sensors impact the recognition accuracy. Our results showed that the multi-sensors achieve the accuracy of 81%, and dominant hand sensor achieves the accuracy of 80%, which is 7% higher than non-dominant hand sensor.
A Study on Wearable Electrocardiograph
(2015)
Tsai, Hsin-Hsi
;
Chuang, Yi-Ta
;
Yi, Chih-Wei
;
Wang, Yi-Chang
;
資訊工程學系
The emergence of wearable devices based on advances in MEMS and low-energy technologies makes low-cost sustained physiological monitoring possible and brings health care into a new territory. But until now, most wearable devices for physiological monitoring do not have FDA approval and the measured data may have some level of noise. In this work, to understand the feasibility of applying wearable electrocardiograph for heart activity monitoring during jogging, we propose a jogging monitoring system based on wearable devices and provide some experimental results. First, the proposed jogging monitoring system is sketched. Next, the traditional ECG tests and the fundamental ECG morphology are introduced. And then, a QRS detection algorithm is introduced and implemented to process ECG data. Last, we collect ECG data by a wearable device in jogging, and some MLII lead ECG records from the MIT-BIH database are selected for comparison. The accuracy of the implemented QRS detection algorithm against both data sets is provided.
A Two-Layer Hierarchical Framework for Activity Sequence Recognition by Wearable Sensors
(2016)
Chan, Guo-Jing
;
Lin, Dong-Hung
;
Yi, Chih-Wei
;
Tseng, Chien-Chao
;
資訊工程學系
As the aging population grows, the elderly care service has become an important part of the service industry in the aging population era. Activity monitoring is one of the most important services in the field of the elderly care service. In this paper, we proposed a wearable solution to provide an activity monitoring service on elders for caregivers. This service monitors restroom activities, such as washing hands, urinating and defecation. In the proposed solution, wireless motion sensors are wore on elder\'s wrist and waist to measure their body movement. The measured motion data are processed to statistical features and aggregated to cloud servers through gateways. A two-layer hierarchical framework is used for the activity recognition. In the first layer, a preliminary recognition is performed by a supervised Reduced Error Pruning (REP) Tree classifier to detect the transition of the activity. In the second layer, a Variable Order Hidden Markov Model (VOHMM) is proposed to determine the sequence of the activities. The experiment results show that the recognition accuracy is 70 percent. We developed a prototype service App to provide a life log for the recording of the activity sequence. The caregivers can make use of this information to take necessary actions accordingly.
A Unified Analytic Framework Based on Minimum Scan Statistics for Wireless Ad Hoc and Sensor Networks
(2009-09-01)
Yi, Chih-Wei
;
資訊工程學系
;
Department of Computer Science
Due to limitations on transmission power of wireless devices, areas with sparse nodes are decisive to some extreme properties of network topology. In this paper, we assume wireless ad hoc and sensor networks are represented by uniform point processes or Poisson point processes. Asymptotic analyses based on minimum scan statistics are given for some crucial network properties, including coverage of wireless sensor networks, connectivity of wireless ad hoc networks, the largest edge length of geometric structures, and local-minimum-free geographic routing protocols. We derive explicit formulas of minimum scan statistics. By taking the transmission radius as a major parameter, our results are applied to various network problems. This work offers a unified approach to solve various problems and reveals the evolution of network topology. In addition, boundary effects are thoroughly handled.
An optimal algorithm for the minimum disc cover problem
(2008-01-01)
Sun, Min-Te
;
Yi, Chih-Wei
;
Yang, Chuan-Kai
;
Lai, Ten-Hwang
;
資訊工程學系
The minimum disc cover can be used to construct a dominating set on the fly for energy-efficient communications in mobile ad hoc networks, but the approach used to compute the minimum disc cover proposed in previous studies is computationally relatively expensive. In this paper, we show that the disc cover problem is in fact a special case of the general alpha-hull problem. In spite of being a special case, the disc cover problem is not easier than the general alpha-hull problem. In addition to applying the existing alpha-hull algorithm to solve the disc cover problem, we present a simple, yet optimal divide-and-conquer algorithm that constructs the minimum disc cover for arbitrary cases, including those degenerate cases where the alpha-hull approach would fail.
Asymptotic Critical Transmission Radii for Greedy Forward Routing in Wireless Ad Hoc Networks
(2009-05-01)
Wan, Peng-Jun
;
Yi, Chih-Wei
;
Wang, Lixin
;
Yao, Frances
;
Jia, Xiaohua
In wireless ad hoe networks, greedy forward routing is a localized geographic routing algorithm in which one node discards a packet if none of its neighbors is closer to the destination of the packet than itself, or otherwise forwards the packet to the neighbor closest to the destination. If all nodes have the same transmission radii, the critical transmission radius for greedy forward routing is the smallest transmission radius which ensures packets can be delivered by greedy forward routing through any source-destination pair. In this paper, we study asymptotic critical transmission radii of randomly deployed wireless ad hoc networks. Assume network nodes are represented by a Poisson point process of density n over a unit-area convex compact region whose boundary curvature is bounded. We show that the ratio of critical transmission radii to root ln n/pi n is asymptotically almost surely equal to root 1/(2/3 - root 3/2 pi) approximate to 1.6.
Asymptotic Critical Transmission Radius for k-Connectivity in Wireless Ad Hoc Networks
(2010-06-01)
Wan, Peng-Jun
;
Yi, Chih-Wei
;
Wang, Lixin
;
資訊工程學系
;
Department of Computer Science
A range assignment to the nodes in a wireless ad hoc network induces a topology in which there is an edge between two nodes if and only if both of them are within each other's transmission range. The critical transmission radius for k-connectivity is the smallest r such that if all nodes have the transmission radius r, the induced topology is k-connected. In this paper, we study the asymptotic critical transmission radius for k-connectivity in a wireless ad hoc network whose nodes are uniformly and independently distributed in a unit-area square or disk. We provide a precise asymptotic distribution of the critical transmission radius for k-connectivity. In addition, the critical neighbor number for k-connectivity is the smallest integer r such that if every node sets its transmission radius equal to the distance between itself and its l-th nearest neighbor, the induced (symmetric) topology is k-connected. Applying the critical transmission radius for k-connectivity, we can obtain an asymptotic almost sure upper bound on the critical neighbor number for k-connectivity.
Asymptotic Distribution of The Number of Isolated Nodes in Wireless Ad Hoc Networks with Unreliable Nodes and Links
(2006)
Yi, Chih-Wei
;
Wan, Peng-Jun
;
Lin, Kuo-Wei
;
Huang, Chih-Hao
;
資訊工程學系
In randomly-deployed wireless ad hoc networks with reliable nodes and links, vanishment of isolated nodes asymptotically implies connectivity of networks. However, in a realistic system, nodes may become inactive, and links may become down. The inactive nodes and down links cannot take part in routing/relaying and thus may affect the connectivity. In this paper, we study the connectivity of a wireless ad hoc network that is composed of unreliable nodes and links by investigating the distribution of the number of isolated nodes in the network. We assume that the wireless ad hoc network consists of n nodes which are distributed independently and uniformly in a unit-area disk or square. Nodes are active independently with probability 0 < p(1) <= 1, and links are up independently with probability 0 < p(2) <= 1. A node is said to be isolated if it doesn't have an up link to an active node. We show that if all nodes have a maximum transmission radius r(n) = root ln n broken vertical bar xi/pi p(1)p2(n) for some constant xi, then the total number of isolated nodes is asymptotically Poisson with mean e(-xi) and the total number of isolated active nodes is also asymptotically Poisson with mean p(1)e(-xi). In addition, the work can be extended for secure wireless networks which adopt m-composite key predistribution schemes in which a node is said to be isolated if it doesn't have a secure link. Let p denote the probability of the event that two neighbor nodes have a secure link. We show that if all nodes have a maximum transmission radius r(n) = root ln n+xi/pi pn for some constant., then the total number of isolated nodes is asymptotically Poisson with mean
Augmented Reality Assisted Photo Positioning for Mobile Devices
(2013)
Lin, Ju-Yi
;
Yi, Chih-Wei
;
Tseng, Yu-Chee
;
資訊工程學系
;
Department of Computer Science
Recent developments in mobile techniques have enabled a great variety of Location Based Services (LBS). A high positioning accuracy is a fundamental requirement for precision LBS applications, e. g., precise LBS marketing in shopping malls or indoor emergency evacuation services with mobile devices. However, most offerable commercial positioning systems, such as GPS/GNSS and RF-based systems, can not provide positioning accuracy within one meter. In this work, a new positioning approach is proposed for mobile devices, Called Photo Positioning, it can provide a high accuracy positioning service. The "positioning" here means to find the location where a photo was taken by investigating the geometric relations between the images of points of interest (POI) in the photo and their location in the real world based on the principle of photo imaging. To implement a photo positioning system, three major components are needed, including a POI database, a method to recognize and locate POIs in a photo and an algorithm to calculate the position where the photo was taken from the POI information. A positioning algorithm based on the geometric similarity in photo imaging is presented in this work, and a prototype system is developed for Android smartphone platforms. Our experimental results shows that the average positioning error of the proposed photo positioning approach can be as low as 74.34 cm.
Autonomous Light Control by Wireless Sensor and Actuator Networks
(2010-06-01)
Yeh, Lun-Wu
;
Lu, Che-Yen
;
Kou, Chi-Wai
;
Tseng, Yu-Chee
;
Yi, Chih-Wei
Recently, wireless sensor and actuator networks (WSANs) have been widely discussed in many applications. In this paper, we propose an autonomous light control system based on the feedback from light sensors carried by users. Our design focuses on meeting users' preferences and energy efficiency. Both whole and local lighting devices are considered. Users' preferences may depend on their activities and profiles and two requirement models are considered: binary satisfaction and continuous satisfaction models. For controlling whole lighting devices, two decision algorithms are proposed. For controlling local lighting devices, a surface-tracking scheme is proposed. Our solutions are autonomous because, as opposed to existing solutions, they can dynamically adapt to environment changes and do not need to track users' current locations. Simulations and prototyping results are presented to verify the effectiveness of our designs.
Delivery Guarantee of Greedy Routing in Three Dimensional Wireless Networks
(2008)
Wang, Yu
;
Yi, Chih-Wei
;
Li, Fan
;
交大名義發表
;
National Chiao Tung University
In this paper, we investigate how to design greedy routing to guarantee packet delivery in a three-dimensional (313) network. In 2D networks, many position-based routing protocols apply face routing on planar routing structure as a backup method to guarantee packet delivery when greedy routing fails at local minimum. However, in 3D networks, no planar topology can be constructed anymore. Even worse, a recent result [6] showed that there is no deterministic localized routing algorithm that guarantees the delivery of packets in 3D networks. Therefore, we propose to set up the transmission radius large enough to eliminate local minimum in the 3D network. In particular, we study the asymptotic critical transmission radius for greedy routing to ensure the packet delivery in randomly deployed 3D networks. Using similar techniques in [12], we theoretically prove that for a 3D network, formed by nodes that are produced by a Poisson point is process of density n over a convex compact region of unit volume, 3 root 3 beta(0)1n n/4 pi n is asymptotically almost surely (abbreviated by a.a.s.) the threshold of the critical transmission radius for 3D greedy routing, where beta(0) = 3.2. We also conduct extensive simulations to confirm our theoretical results.
Design and Implement a Mobile Badminton Stroke Classification System
(2017-01-01)
Lin, Juyi
;
Chang, Chia-Wei
;
Wang, Chih-Hao
;
Chi, Hong-Chuan
;
Yi, Chih-Wei
The use of the badminton stroke strategy in the evenly matched game is often the key to victory. In this work, a smart racket based on wearable seusors is proposed to collect the data of swing of badminton. A cell phone APP with machine learning techniques is implemented to record stroke types automatically. In each stroke hit event, this prototype system uses Bluetooth earphone to collect the sound for detecting the accuracy time. It uses the data of IMU in each stroke for determining stroke type. Compared to IMU only solution, the system will reduce the false count of stroke hit. Using cloud techniques could record the training and game record in a long period. Overall the accuracy of stroke hit event is almost 100% by using voice print. The data of IMU is classified by Random Forest or SMO. The accuracy for personal model is 95.91 %, and it is 79.32% for general model. We develop a stroke record system which is combined with Wearable sensor, Mobile platform and Cloud service.
Discovering Phase Timing Information of Traffic Light Systems by Stop-Go Shockwaves
(2015-01-01)
Chuang, Yi-Ta
;
Yi, Chih-Wei
;
Tseng, Yu-Chee
;
Nian, Chia-Sheng
;
Ching, Chia-Hao
The cycle lengths and signal transition time of Traffic Light Systems (TLS\'s), or known as the Phase Timing Information (PTI), play a key role in modern transportation systems. However, such information is not always available to the public. In this paper, we propose a crowdsourcing approach to solve this problem by exploiting the stop and go events, abbreviated by SG events, of vehicles on roads happening in front of target traffic lights. The PTI discovery problem is formulated by allowing only part of the vehicles participating in the discovery process. The proposed framework starts with discovering SG events, followed by collapsing these events over multiple signal cycles into one and calculating PTI information through a shockwave technique. The crowdsourcing part may be directly implemented on smartphones. The proposed framework was verified via field trials and simulations. Our simulation results showed that, even with a low penetration rate around 3.8 percent, the root mean square errors of the cycle length, green light and red light signal transition time of a TLS are 0. 04, 1.3 and 5.8 seconds, respectively. The achieved accuracy can be helpful in many PTI-enabled applications.
FetalCare: A Mobile Cardiotocograph System
(2015)
Chang, Chia-Wei
;
Tsai, Hsin-Hsi
;
Yi, Chih-Wei
;
Wang, Yi-Chang
;
Kuo, Jen-Yau
Due to the increasing of elderly primigravidae, the demand of the tocolysis is more than before. In this project, NCTU and ITRI worked together to improve the Maya Tocolytic System at Mackay Memorial Hospital by augmenting remote real-time physiological data transmission and display on smartphones. Doctors can have real-time physiological data to make diagnosis and provide advice remotely by leverage mobile technologies.
File Transfer for Mobile Devices in Heterogeneous Radio Networks
(2010)
Yi, Chih-Wei
;
Yang, Shau-Shiuan
;
Lin, Yi-Bing
;
Chuang, Yi-Ta
;
Liu, Pin-Chuan
In the past, although mobile devices were equipped with multiple radio interfaces, for the sake of power saving, only one was activated for data transmission. The idea of concurrent transmission via multiple radio interfaces has not been seriously studied. However, nowadays, power consumption no longer is a problem in many application scenarios, e. g. VANETs. In this work, we investigate the performance improvement of concurrent file transfer over two heterogeneous radio networks, e. g. WiFi and 3G. The traditional File Transfer Protocol is modified to utilize two heterogeneous radio connections and experiments are executed over the Internet and a 3G data network to measure the latency and average bandwidth. Our results show that it is possible to integrate the bandwidth of both radio networks.
Forwarding Methods in Data Dissemination and Routing Protocols for Vehicular Ad Hoc Networks
(2013-11-01)
Daraghmi, Yousef-Awwad
;
Yi, Chih-Wei
;
Stojmenovic, Ivan
;
資訊工程學系
;
Department of Computer Science
Intermittent connectivity, abrupt changes in network topology and low reception rate are the most important properties that distinguish VANET (vehicular ad hoc networks) from other types of ad hoc networks. To optimize reliability and time criticality metrics in data communication protocols for VANET, novel ideas are needed. In this article, we present a tutorial on methods (at the network layer), encountered in recent literature, for small and large scale routing protocols, and geocasting (broadcasting, data dissemination, and warning delivery) protocols.
G-Constellations: G-Sensor Motion Tracking Systems
(2010)
Yi, Chih-Wei
;
Su, Chao-Min
;
Chai, Wen-Tien
;
Huang, Jiun-Long
;
Chiang, Tsun-Chieh
In most inertial motion tracking systems, motion directions are detected and measured by direction sensors such as magnetometers and gyroscopes. In this paper, we propose a motion tracking system, called g-sensor constellations, in which only g-sensors but no direction sensors are used. The g-sensor constellation is a loose coupling g-sensor system with rigid geometric topology. As few as three g-sensors are needed for motion tracking, including direction detection. The system is easy to be installed. No complicated calibrations are needed and the necessary information is the distances between sensors. The proposed framework can improve the accuracy of dead reckoning systems and help in the analyzing of traffic accidents and developing new human-computer interfaces. In our experiments, a g-sensor constellation composed of three g-sensors, which are located at the vertices of an equilateral triangle with edges of 0.3m and communicate with the processing unit via Bluetooth links, is built to verify the proposed technique.
Group Access Control with Blacklist for Data Dissemination in Mobile Opportunistic Networks
(2013)
Ho, Tzu-Hsin
;
Yi, Chih-Wei
;
Tseng, Chien-Chao
;
資訊工程學系
;
Department of Computer Science
In mobile opportunistic networks, security is a major concern in consequence of store-carry-and-forward data dissemination. To guarantee that messages can be only accessed by authorized users, access control is a necessity. Fuzzy-IBE that has been proposed can provide group access control. In addition, if a sender wants to expose his/her message only to some members of his/her group, Fuzzy-IBE can provide a whitelist mechanism to further indicate which users in the group can access the data. Because overhead linearly increases with respect to the length of the whitelist, this mechanism is not sufficiently scalable as the whitelist becomes larger. In this work, we propose a Fuzzy-IBE based negative access control scheme (NAC) that allows users to selectively exclude specific members from accessing data dynamically. NAC, which does not require intensive calculation in mobile devices, is suitable for mobile opportunistic networks.