Site

Project

  • Multicarrier Wireless Network
    • Jan. 2011 ~ now
    • The University of Texas at Austin
    • Advisor: Prof. Lili Qiu
    • Wireless multicarrier communication transmits data by spreading them over multiple subcarriers and is widely used today owing to its robustness to multipath fading, high spectrum efficiency, and ease of implementation. We use real measurements to show there is significant frequency diversity in WiFi channels, and propose a series of techniques to explicitly harness such frequency diversity. In particular, we leverage the channel state information (CSI), which captures the SNR on each subcarrier, (i) to map symbols to subcarriers according to their importance, (ii) to effectively recover partially corrupted FEC groups and facilitate FEC decoding, and (iii) to develop MAC-layer FEC to offer different degrees of protection to the symbols according to their error rate at the PHY layer.
  • L-MAC
    • June 2010 ~ August 2010
    • Hamilton Institute
    • Advisor: Prof. Douglas Leith, Prof. David Malone
    • By comining the features of CSMA and TDMA, fully decentralized WLAN MAC schemes have recently been proposed that converge to collision-free schedules. In L-MAC, we introduce learning algorithms that can substantially speed up convergence of collision free operation and develop a decentralized schedule length adaptation scheme that provide long-run fair access to the medium while maintaining collision-free access for arbitrary numbers of stations. I realize the algorithm of L-MAC using Madwifi. The benefit is that only the driver is modified and no additional hardware support is necessary so the scheme can be directly applied to any current machine.
  • Vehicular Network
    • Aug. 2009 ~ May 2010
    • The University of Texas at Austin
    • Advisor: Prof. Lili Qiu
    • The goal is to present a system for enabling high-bandwidth content distribution in vehicular networks. In the system, a vehicle opportunistically communicates with nearby access points to download the content of interest. To fully take advantage of such transient contact with APs, we proactively push content to APs that the vehicles will likely visit in the near future. It includes two components: 1) an algorithm for predicting the APs that will soon be visited by the vehicles, and 2) a series of replication optimization techniques to enable the synergy among Internet connectivity, local wireless connectivity, and vehicular relay connectivity.
  • YushanNet
    • Mar. 2008 ~ July 2009
    • Network Research Lab, Academia Sinica
    • Advisor: Prof. Ling-Jyh Chen
    • http://yushannet.org/index_e.php
    • The objective of this project is to establish a reliable and comprehensive system that can effectively monitor the environment and assist the tracking of tourists in Yushan National Park. As full wireless coverage is impossible to maintain in wilderness environments, network communication in the proposed system will be inevitably intermittent, and therefore the development of this observation system will be very challenging. The focus of YushanNet is to conquer this issue using delay/disruption tolerant techniques. In this way, tourist information can be aggregated effectively during the time when the network is working most efficiently.
  • Energy-efficient Zigbee localization
    • Sept. 2004 ~ Oct. 2006
    • Ubicomp Lab, National Taiwan University
    • Advisor: Prof. Hao-hau Chu
    • It is an energy-aware localization system based on Zigbee radio sensor network. Promising to satisfy an applicationís requirements on positional accuracy, the system tries to minimize its energy consumption. Our method is to adapt the sampling rate to the targetís mobility level. We have created several real testbed deployments and shown that energy saving can be as high as 50%.
  • Collaborative localization
    • Sept. 2004 ~ Oct. 2006
    • Ubicomp Lab, National Taiwan University
    • Advisor: Prof. Hao-hau Chu
    • While research on Wi-Fi indoor localization has demonstrated adequate performance, localization error increases significantly in crowded and dynamic situations due to electromagnetic interferences. This work proposes collaborative localization as an approach to enhance position estimation by leveraging more accurate location information from nearby neighbors within the same cluster and the results have shown 23.4-56.4% accuracy improvement over the well-known system.