Browsing by Author "Tseng, Po-Hsuan"

Now showing 1 - 20 of 31

3-D Interactive Augmented Reality-enhanced Digital Learning Systems for Mobile Devices
(2013)
Feng, Kai-Ten
;
Tseng, Po-Hsuan
;
Chiu, Pei-Shuan
;
Yang, Jia-Lin
;
Chiu, Chun-Jie
With enhanced processing capability of mobile platforms, augmented reality (AR) has been considered a promising technology for achieving enhanced user experiences (UX). Augmented reality is to impose virtual information, e.g., videos and images, on to a live-view digital display. UX on real-world environment via the display can be effectively enhanced with the adoption of interactive AR technology. Enhancement on UX can be beneficial for digital learning systems. There are existing research works based on AR targeting for the design of e-learning systems. However, none of these work focuses on providing three-dimensional (3-D) object modeling for enhanced UX based on interactive AR techniques. In this paper, the 3-D interactive augmented reality-enhanced learning (IARL) systems will be proposed to provide enhanced UX for digital learning. The proposed IARL systems consist of two major components, including the markerless pattern recognition (MPR) for 3-D models and velocity-based object tracking (VOT) algorithms. Realistic implementation of proposed IARL system is conducted on Android-based mobile platforms. UX on digital learning can be greatly improved with the adoption of proposed IARL systems.
A Pitch-Aided Lane Tracking Algorithm for Driver Assistance System with Insufficient Observations
(2012)
Tang, Chi-Wai
;
Feng, Kai-Ten
;
Tseng, Po-Hsuan
;
Chen, Chien-Hua
;
Guo, Jing-Wei
The driver assistance system (DAS) is important to assist drivers with driving safety and efficiency. The lane departure warning system (LDWS), being an essential portion within the DAS, aims at providing warning messages under irregular driving behaviors due to distraction, drowsiness, or driver negligence. As the developments of mobile devices are boosted vigorously, it becomes more feasible to integrate sensing, photographing and displaying components within the LDWS. In this paper, the pitch-aided lane tracking (PALT) algorithm is proposed to detect and track the lanes on the roadways base on the particle filter. A pitch model is proposed to track the lane attributes associated with the pitch angle of a mobile device in order to conquer the effects from bumpy road and loosely-equipped of a mobile device. Moreover, the proposed PALT scheme can compensate the influences of insufficient observations from the display. Numerical results show that proposed PALT scheme can provide better lane-tracking performance compared to both original observations and the model without considering pitch angle.
A predictive location tracking algorithm for mobile devices with deficient signal sources
(2007)
Lin, Yu-Chiun
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電信工程研究所
;
Institute of Communications Engineering
Location estimation and tracking for the mobile devices have attracted a significant amount of attention in recent years. The network-based location estimation schemes have been widely adopted based on the radio signals between the mobile device and the base stations. The location estimators associated with the Kalman filtering techniques are exploited to both acquire location estimation and trajectory tracking for the mobile devices. However, most of the existing schemes become unapplicable due to the insufficiency of signal sources. In this paper, a Predictive Location Tracking (PLT) algorithm is proposed to alleviate this problem. The predictive information obtained from the Kalman filter is employed to provide the additional signal inputs for the location estimators. The proposed PLT scheme can offer persistent accuracy for location tracking of the mobile devices, especially with inadequate signal sources. Numerical results demonstrate that the proposed PLT algorithm can achieve better precision, comparing with other existing schemes, in mobile location estimation and tracking.
A predictive movement based handover algorithm for broadband wireless networks
(2008)
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電信工程研究所
;
Institute of Communications Engineering
The design of an efficient handover technique is considered as one of the crucial topics in the cellular based wireless networks. Feasible handover process can minimize the performance degradation as the mobile device is moving between different base stations. Based on the IEEE 802.16e standard, it is required to provide satisfactory handover performance under a wide range of moving speeds of the mobile device. In this paper, a predictive movement based handover (PMHO) algorithm is proposed. Without incurring excessive overheads as from the conventional location estimation and tracking techniques, the proposed sideline traveling scheme within the PMHO algorithm employs the moving state indicator (MSI) to predict the potential moving characteristics of the mobile device. The handover decision of the PMHO approach is determined based on the MSI values associated with the neighbor base stations. Numerical results show that the proposed PMHO algorithm can effectively reduce both the number of handover and the size of the neighbor scanning set within the mobile device.
An Enhanced Predictive Location Tracking Scheme with Deficient Signal Sources for Wireless Networks
(2009)
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
Location estimation and tracking for the mobile devices have attracted a significant amount of attention in recent years. The location estimators associated with the Kalman filtering techniques are exploited to both acquire location estimation and trajectory tracking for the mobile devices. However, most of the existing schemes become inapplicable for location tracking due to the deficiency of signal sources. In this paper, the enhanced predictive location tracking (EPLT) are proposed to alleviate this problem. The EPLT scheme utilizes the predictive information obtained from the Kalman filter in order to provide the additional signal inputs for the location estimator. Furthermore, the EPLT scheme incorporates the geometric dilution of precision (GDOP) information into the algorithm design. Persistent accuracy for location tracking can be achieved by adopting the proposed EPLT scheme, especially with inadequate signal sources. Numerical results demonstrate that the EPLT algorithm can achieve better precision in comparison with other location tracking schemes.
An unified Kalman tracking technique for wireless location systems
(2007)
Tseng, Po-Hsuan
;
Chen, Chao-Lin
;
Feng, Kai-Ten
;
電信工程研究所
;
Institute of Communications Engineering
Location estimation and tracking for the mobile stations have attracted a significant amount of attention in recent years. In order to offer sufficient precision for wireless location tracking, different techniques have been studied and combined, e.g. the Least Square methods for location estimation associated with the Kalman filters for location tracking. In this paper, an Unified Kalman Tracking (UKT) technique is proposed to provide an integrated algorithm for precise location estimation and tracking. Based on the Time-Of-Arrival (TOA) measurements, a new variable is incorporated as an additional state within the Kalman filtering formulation in order to consider the nonlinear behavior for wireless location estimate. Numerical results illustrate that the proposed UKT algorithm can achieve enhanced accuracy for mobile location tracking, comparing with other existing schemes.
Automatic Hybrid Access Point Deployment for Wireless Localization Systems
(2017-01-01)
Hung, Yun-Ting
;
Feng, Kai-Ten
;
Tseng, Po-Hsuan
;
電機工程學系
;
Department of Electrical and Computer Engineering
Location estimation has received wide attention due to the emerging demand for location-based services (LBSs) in indoor environments. Although positioning algorithms have been rapidly developed, the positioning accuracy has not reached the requirement of indoor LBSs. Indoor positioning methods based on the existing communication systems such as Wi-Fi or Bluetooth have the advantage of lower cost and higher penetration rates, which can provide a sufficient number of signal sources. Unlike the global positioning system where the satellites are well-deployed to provide four or more signal sources and well-conditioned geometric for outdoor devices, the critical limits of indoor positioning are the insufficient signal sources and disunified deployment for access points (APs). To address the problem, we propose an automatic hybrid AP deployment (AHAD) algorithm to provide optimal locations and required numbers of both Wi-Fi APs and BLE APs for achieving higher location estimation accuracy. With the adoption of genetic algorithm, the AHAD scheme can maintain satisfactory Wi-Fi communication quality and fulfill user's budget for AP deployment. Furthermore, a hybrid indoor positioning (HIP) scheme is also proposed based on the combination of Wi-Fi fingerprinting and BLE proximity. Experimental results show that the proposed AHAD algorithm can provide better location estimation accuracy compared to conventional AP deployment based on user instinct.
Cloud Computing Based Mobile Augmented Reality Interactive System
(2014)
Chiu, Pei-Hsuan
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電機工程學系
;
Department of Electrical and Computer Engineering
Augmented reality (AR) technology is mainly composed of feature extraction, feature points matching and 3-D object drawing to impose virtual information onto the real camera. Hand tracking method is designed to realize interactive AR system to further enhance user experience (UX). However, feature extraction, feature points matching, and hand tracking all require high computational cost. To achieve real-time implementation on mobile, we proposed a Mobile Augmented ReAlity Interactive System based on cloud computing (MARAIS). Cloud side is responsible for heavy tasks that need sufficient computational capabilities, and device side performs sensing, detecting and displaying. Device/cloud architecture has been designed to maintain tolerable processing delay with minimum communication overhead. To validate the suitability for real-time interaction, MARAIS is combined with a picture book to realize a digital learning system. Implementation of interactive AR system based on cloud computing eases the requirement of data storage, memory, and computational power at device side.
Compressive Sensing based Location Estimation using Channel Impulse Response Measurements
(2014)
Lin, Yu-Pei
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電機工程學系
;
Department of Electrical and Computer Engineering
Due to the popularity of location-based services in environment with weak GPS signals, indoor location estimation problem have attracted more and more attention in recent years. Among several distance-related measurements, channel impulse response (CIR) reflects multi-path situation between the transmitter and receiver pair and is suitable to describe the characteristic of different positions. Note that CIR, which can be obtained from the inverse Fourier transform of channel frequency response in broadband wireless networks, is supported in most of the commercial standards. In this paper, a novel compressive sensing based location estimation using CIR measurements (CS-CIR) is proposed as well as fingerprinting algorithm. CIR information is collected from each reference point (RP) to access point (AP) and stored in the database. During the on-line stage, the mobile user measures CIR from the AP and compares measured CIR with those CIR values in the database. Note that user position is close to one of the RPs and user position vector is represented as a sparse vector. By applying compressive sensing theory, user position can be recovered by solving l(1)-minimization problem. Simulation result validate that the CS-CIR outperfiffms the K-nearest neighbour method using CIR measurements and conventional received signal strength based methods.
Cooperative Self-Navigation in a Mixed LOS and NLOS Environment
(2014-02-01)
Tseng, Po-Hsuan
;
Ding, Zhi
;
Feng, Kai-Ten
;
電機資訊學士班
;
Undergraduate Honors Program of Electrical Engineering and Computer Science
We investigate the problem of cooperative self-navigation (CSN) for multiple mobile sensors in the mixed line-of-sight (LOS) and nonline-of-sight (NLOS) environment based on measuring time-of-arrival (TOA) from the cooperative sensing. We first derive an optimized recursive Bayesian solution by adopting a multiple model sampling-based importance resampling particle filter for the development of CSN. It can accommodate nonlinear signal model and non-Gaussian position movement under different levels of channel knowledge. We also utilize a Rao-Blackwellization particle filter to split the original problem by tracking the channel condition with a grid-based filter and estimating the position with a particle filter. The CSN with position and channel tracking exhibits advantage over the noncooperative methods by utilizing additional cooperative measurements. It also shows improvement over the methods without channel tracking. Simulation results validate that both schemes can take the advantage of cooperative sensing and channel condition tracking in mixed LOS/NLOS environments, which motivates future research of cooperative gain for navigation and localization in a more general environment.
DEMO PAPER: PARTICLE-BASED AUGMENTED REALITY INTERACTIVE SYSTEM
(2013)
Chiu, Pei-Hsuan
;
Feng, Kai-Ten
;
Tseng, Po-Hsuan
;
電機資訊學士班
;
Undergraduate Honors Program of Electrical Engineering and Computer Science
With enhanced processing capability of mobile platforms, augmented reality (AR) has been considered a promising technology for achieving enhanced user experiences (UX). AR is to impose virtual information onto the real camera view of mobile platforms. In this demonstration, a particle-based augmented reality interactive system (PARIS) is designed to provide interactive AR for enhanced UX. The proposed PARIS consists of two parts, including the markerless pattern recognition (MPR) enabling the projection of recognized 3-D objects onto the real camera view and particle filter based hand tracking algorithms to provide interaction with the 3-D objects. Implementation of PARIS is realized on Android-based mobile platforms, where low computational complexity has been observed for real-time interaction.
Derivation of CRLB for linear least square estimator in wireless location systems
(2012-10-01)
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電機工程學系
;
Department of Electrical and Computer Engineering
Linear estimators, including the well-adopted linear least squares (LLS) estimator, have been extensively utilized for wireless location estimation for their simplicity and closed-form property. However, there exists information lost from the linearization of the location estimator to the nonlinear location estimation, which prevents the linear estimator from approaching the Cramer-Rao lower bound (CRLB). In this paper, the linearized location estimation problem based CRLB (L-CRLB) is derived to provide a portrayal in order to fully characterize the behavior of the linearized location estimator. The relationships between the L-CRLB and the CRLB are obtained and theoretically proven in this paper. Furthermore, the geometric layout between the mobile station (MS) and the base stations (BSs) that can achieve the minimum L-CRLB is also acquired. As can be suggested by the L-CRLB, an LLS location estimator can achieve higher accuracy if the MS is located inside the geometry confined by the BSs compared to the case that the MS is situated outside of the geometric layout. This result will be beneficial to the deployment of BSs or the signal selection schemes targeting for location estimation. Simulation results utilizing the LLS estimator as one of the implementation of the linearized location estimators further validate the theoretical proofs and the effectiveness of the L-CRLB.
Enhanced Area Estimation Algorithms for Indoor Wireless Localization
(2016)
Liu, Heng-Xiu
;
Chen, Bo-An
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
Wang, Tian-Sheng
We enhance the area average (AA) algorithm based on received signal strength (RSS) fingerprinting in Wi-Fi infrastructure by utilizing map information. The area searching (AS) algorithm adopts the concept of Cell-ID method before fingerprinting is proposed to select appropriate areas to reduce the estimation error. The fusion-based RSS distance computation (FRD) scheme can represent the RSS distance more accurate when both 2.4 GHz and 5 GHz are considered. Experiment results validate that the enhanced algorithms achieve above 90% of area estimation accuracy in the area with access points in four different testing environments.
Femto-Assisted Location Estimation in Macro/Femto Heterogeneous Networks
(2013)
Lee, Ke-Ting
;
Tseng, Po-Hsuan
;
Chen, Chien-Hua
;
Feng, Kai-Ten
;
電機工程學系
Location estimation and tracking for mobile stations (MSs) have attracted a significant amount of attention in recent years. In indoor environment where the signals from global positioning system (GPS) are either weak or blocked, signal sources from long term evolution advanced (LTE-A) system can be adopted to provide location estimation for MS. Based on the range signals from both macro base station (mBS) and femto BS (fBS) in LTE-A heterogeneous networks (HetNet), we propose femto-assisted location estimation (FALE) schemes to estimate MS's position especially for indoor environments with insufficient signal inputs from mBSs. Since fBSs are usually deployed by users in their residential or business buildings, the locations of these fBSs generally cannot be known exactly. We depict the imprecise fBSs' positions as belief information to investigate a Bayesian estimation method based on the time difference of arrival (TDOA) measurement, and utilize particle filtering technique to develop the FALE algorithm. Moreover, a simplified FALE (FALE-S) scheme is proposed to reduce the computation cost resulting from the particle filter in original FALE method. Performance evaluation is conducted based on the LTE-A HetNet environments. Compared to conventional scheme, simulation results show that the proposed FALE algorithms can provide better location estimation of MS through the assistance of fBSs.
Fingerprinting-based Dynamic RSS Adjustment for Time-variant Indoor Positioning Systems
(2016)
Chen, Bo-An
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
Wang, Tian-Sheng
;
電機工程學系
Fingerprinting is one of the most reliable technique for indoor positioning. The concept of the fingerprinting is to compare the offline received signal strength (RSS) database and online RSS measurement to find the most similar reference point to represent the user position. However, fingerprinting is sensitive to the environmental changes, where the offline database is required to be updated. In this paper, we discuss the offline and online measurement mismatch when the transmission power is altered by the Wi-Fi access point. We propose dynamic adjustment (DA) algorithm to reduce the influence caused by the power variation. Experimental results validate that the proposed DA algorithm maintains the positioning accuracy under the power adjustment.
GDOP-Assisted Location Estimation Algorithms in Wireless Location Systems
(2008)
Chu, Lin-Chih
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電信工程研究所
;
Institute of Communications Engineering
In recent years, wireless location estimation has attracted a significant amount of attention in different areas. The network-based location estimation schemes have been widely adopted based on the radio signals between the mobile station (MS) and the base stations (BSs). The two-step Least Square (LS) method has been studied in related research to provide efficient location estimation of the MS. However, the algorithm results in inaccurate location estimation under the circumstances with poor geometric dilution of precision (GDOP). In this paper, the GDOP-assisted location estimation (GOLE) schemes are proposed by considering the geometric relationships between the MS and its associated BSs. According to the minimal GDOP criterion, the BSs are fictitiously repositioned and are served as a new set of BSs within the formulation of the two-step LS algorithm. The proposed GOLE schemes can both preserve the computational efficiency from the two-step LS method and obtain precise location estimation under poor GDOP environments. Comparing with other existing schemes, numerical results demonstrate that the proposed GOLE algorithms can achieve better accuracy in wireless location estimation.
Geometry-Assisted Localization Algorithms for Wireless Networks
(2013-04-01)
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電機資訊學士班
;
Undergraduate Honors Program of Electrical Engineering and Computer Science
Linear estimators have been extensively utilized for wireless location estimation for their simplicity and closed form property. In the paper, the class of linear estimator by introducing an additional variable, e. g., the well-adopted linear least squares (LLS) estimator, is discussed. There exists information loss from the linearization of location estimator to the nonlinear location estimation, which prevents the linear estimator from approaching the Cramer-Rao lower bound (CRLB). The linearized location estimation problem-based CRLB (L-CRLB) is derived in this paper to provide a portrayal that can fully characterize the behavior for this type of linearized location estimator. The relationships between the proposed L-CRLB and the conventional CRLB are obtained and theoretically proven in this paper. As suggested by the L-CRLB, higher estimation accuracy can be achieved if the mobile station (MS) is located inside the convex hull of the base stations (BSs) compared to the case that the MS is situated outside of the geometric layout. This result motivates the proposal of geometry-assisted localization (GAL) algorithm in order to consider the geometric effect associated with the linearization loss. Based on the initial estimation, the GAL algorithm fictitiously moves the BSs based on the L-CRLB criteria. Two different implementations, including the GAL with two-step least squares estimator (GAL-TSLS) and the GAL with Kalman filter (GAL-KF), are proposed to consider the situations with and without the adoption of MS's historical estimation. Simulation results show that the GAL-KF scheme can compensate the linearization loss and improve the performance of conventional location estimators.
Hybrid Network/Satellite-Based Location Estimation and Tracking Systems for Wireless Networks
(2009-11-01)
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電信工程研究所
;
Institute of Communications Engineering
Mobile location estimation has attracted much attention in recent years. Location algorithms for mobile stations (MSs) can generally be categorized into network-and satellite-based systems. Both types of systems have their advantages and limitations under different environments (i.e., urban or rural area). In this paper, hybrid location estimation schemes, which combine both the satellite-and the network-based signals, are proposed to provide adaptation to various scenarios for location estimation. By exploiting the fusion algorithm, the proposed fusion-based hybrid (FH) architecture integrates the estimation results that are acquired from both the satellite-and the network-based systems. Two different types of signal selection schemes are adopted within the FH architecture: 1) the fixed set of signal inputs approach and 2) the selective set of signal inputs approach. On the other hand, the unified hybrid architecture employs the proposed hybrid signal-selection scheme and the hybrid least square estimator, which can conduct location estimation within a selected set of signal sources from the heterogeneous networks. The Kalman filtering technique is exploited in the proposed algorithms to both eliminate the measurement noises and to track the trajectories of the MSs. Numerical results demonstrate that the proposed hybrid location schemes can provide accurate location estimation by adapting themselves to different environments.
Hybrid TOA/TDOA Based Unified Kalman Tracking Algorithm for Wireless Networks
(2010)
Chiang, Cheng-Tse
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電機工程學系
;
Department of Electrical and Computer Engineering
Location estimation and tracking for the mobile stations have attracted a significant amount of attention in recent years. Moreover, different types of signal sources are considered available to provide the measurement inputs for location estimation and tracking. In this paper, a hybrid unified Kalman tracking (HUKT) technique is proposed to provide an integrated algorithm for precise location tracking based on both the time-of-arrival (TOA) and time-difference-of-arrival (TDOA) measurements. A new variable is incorporated as an additional state within the Kalman filtering formulation in order to consider the nonlinear behavior for wireless location estimation. Comparing with existing schemes, numerical results illustrate that the proposed HUKT algorithm can achieve enhanced accuracy for mobile location tracking, especially under the environments with insufficient number of signal sources in a single signal path.
Hybrid Unified Kalman Tracking Algorithms for Heterogeneous Wireless Location Systems
(2012-02-01)
Chiang, Cheng-Tse
;
Tseng, Po-Hsuan
;
Feng, Kai-Ten
;
電子物理學系
;
Department of Electrophysics
Location estimation and tracking for mobile stations have attracted a significant amount of attention in recent years. Different types of signal sources are considered available to provide measurement inputs for location estimation and tracking in heterogeneous wireless networks. Various techniques have been studied and combined for location tracking, e. g., the least square methods for location estimation associated with the Kalman filters for location tracking. In this paper, the hybrid unified Kalman tracking (HUKT) technique is proposed to provide an integrated algorithm for precise location tracking based on both time of arrival (TOA) and time difference of arrival (TDOA) measurements. A new variable is incorporated as an additional state within the Kalman filtering formulation to consider the nonlinear behavior in the measurement update process. The relationship between this new variable and the desired location estimate is applied in the state update process of the Kalman filter. Three different designs of hybrid factor are proposed to adaptively adjust the weighting value between the TOA and TDOA measurements. Moreover, similar concepts are also utilized in the design of unified Kalman tracking schemes for pure TOA and TDOA measurement inputs in this paper. Compared with existing schemes, numerical results illustrate that the proposed HUKT algorithm can achieve enhanced accuracy for mobile location tracking, particularly under environments with an insufficient number of measurements in one of the signal paths.