About Me
Home
Publications

Xiao Simon LI

alt text 

Xiao (Simon) Li
Ph.D Candidate (4nd Year)
Advisor : Prof. Anna Scaglione
Department of Electrical and Computer Engineering
University of California, Davis

Visiting Student Research Collaborator (VSRC)
Advisor : Prof. H. Vincent Poor
Department of Electrical Engineering
Princeton University

Address :
Electrical and Computer Engineering
University of California, Davis
One Shields Avenue, 2221 Kemper Hall
Davis, California 95616-5294

Email : textrm{eceli}@textrm{ucdavis.edu}
Web : www.ece.ucdavis.edu/~xiaoli

Recent News

  • New paper "Compressive Anomaly Detection for Large Networks" submitted to IEEE GlobalSIP 2013.
  • New paper "Blind Topology Identification for Power Systems" submitted to IEEE SmartGridComm 2013.

Short Bio

I am currently a visiting researcher at Princeton University working with Prof. H. Vincent Poor. I am a 4-th year PHD student at University of California, Davis under the supervision of Prof. Anna Scaglione. My interests lie in the general area of signal processing and optimization for large scale systems and networked data:

  • Distributed optimization and adaptive learning for big data processing and analytics
  • Compressive acquisition and statistical inferences for large data sets
  • Information processing and management for smart grid, and other networked systems

Education

Awards and Honors

PhD Research Overview

Big Data Processing and Analytics: Distributed Algorithms and Adaptive Learning

alt text 

Tremendous attention has been given to distributed and learning algorithms recently, which performs optimizations in a network by having network nodes interact with each other locally and make progress collaboratively through network diffusions. Most algorithms are developed using first order methods, which typically converges slower than second order methods such as Newton algorithm. One side of our studies in this project is to develop fast converging distributed algorithms and provide performance guarantees. On the other hand, we are also interested in studying the fundamental problem of optimum network formation by evaluating the benefits and costs of the links maintained by individual agents, which can further shed light on building trust models among network agents.

  • Xiao Li and A. Scaglione, “Non-Linear Least Squares Estimation via Network Gossiping”, Asilomar Conference on Signals, Systems and Computers 2012, Asilomar, CA, USA.

  • L. Li, Xiao Li, A. Scaglione and J. Manton, “Decentralized Subspace Tracking via Gossiping”, IEEE DCOSS 2009, Marina del Rey, CA, USA.

Sensing and Inference of High Dimensional Data: Theory and Applications

alt text 

Many signals encountered in engineering are high dimensional, sometimes even continuous (i.e., communication signals). Nowadays, the bottleneck of information processing has been shifted from the scarcity of data to the lack of fast machinery that operates at high data rates. Thus, instead of taking raw samples from high dimensional fields, taking advantage of the low dimensional structures of different signals in different domains becomes inevitable. The crux is to analyze the specific data structures and optimize the sensing and acquisition architectures to reduce the number of data points, while preserving the analytics of interest that are consistent with the raw data. Another equally important aspect is the post-processing of these data, which performs various statistical inferences and decision-making for different purposes. So far, we have tackled useful applications in communications such as acquisition of GPS signals and multiuser signals.

  • Xiao Li, A. Rueetschi, A. Scaglione and Y. C. Eldar, “Optimal Sampling Structure for Asynchronous Multi-Access Channels”, IEEE ICASSP 2012, Kyoto, Japan.

  • Xiao Li, A. Rueetschi, Y. C. Eldar and A. Scaglione, “GPS Signals Acquisition via Compressive Multichannel Sampling”, Physical Communication, Special Issue on Compressive Sensing in Communications. [PDF]

  • A. Scaglione and Xiao Li, “Compressed Channel Sensing : Is Restricted Isometry Property the Right Metric?”, IEEE DSP 2011, Corfu, Greece.

Information Processing, Management and Analytics in Smart Energy Systems

alt text 

The vision of Smart Grid is to exploit communication and information technologies to improve the efficiency, reliability and sustainability of power generation, transmission, distribution and consumption by acting upon data gathered from Supervisory Control and Data Acquisition (SCADA) systems, Wide Area Measurement System (WAMS) and Advanced Metering Infrastructure (AMI). We envision a fully decentralized architecture consisting of distributed network agents forming a cloud service, where each agent is a computer or database available in the local control area such as substation, feeder or residential unit, communicating and collaborating with each other to provide optimal decisions for various purposes. One of the key application we studied is demand response for optimal power dispatch, and power system state estimation for wide area monitoring and control.

  • Xiao Li, A. Scaglione and T.-H. Chang, “ A Unified Framework for Phasor Measurement Placement Design in Hybrid State Estimation”, submitted to IEEE Trans. on Power Systems.

  • M. Alizadeh, Xiao Li, Z. Wang, A. Scaglione and R. Melton, “ Demand Side Management in the Smart Grid : Information Processing for the Power Switch”, Signal Processing Magazine, Special Issue on Signal Processing Techniques for Smart Grid. [PDF]

  • Z. Wang, Xiao Li, V. Muthukumar, A. Scaglione, S. Peisert and C. McParland, “ Networked Loads in the Distribution Grid”, APSIPA 2012, Los Angeles, CA, USA.

  • Xiao Li, Z. Wang and A. Scaglione, “Decentralized Data Processing and Management in Smart Grid via Gossiping”, IEEE SAM 2012, New Jersey, USA. (Best Student Paper Contest Finalist)

Master Research

Synchronization and Channel Estimation for Cooperative Communications

alt text 

The concept of distributed MIMO system has been advocated by many researchers, where the sharing of antennas among several single-antenna terminals to cooperatively transmit data is suggested. It has been pointed out that with proper cooperative strategies, the same benefits of centralized MIMO systems can be achieved in a cooperative MIMO system. In this work, we look at the problem of multiple symbol timing estimation at the destination, and more importantly how to compensate the multiple timing errors. We separately investigated the decode-and-forward cooperative system and amplify-and-forward cooperative system, and proposed two general frameworks for re-synchronization with and without feedback from destination, respectively.

  • Xiao Li, C. Xing, Y.-C. Wu and S.C. Chan, “Timing Estimation and Re-synchronization for Amplify-and-Forward Communication Systems”, IEEE Trans. on Signal Processing, vol. 58, no. 4, pp. 2218-2229, Apr 2010. [PDF]

  • Xiao Li, Y.-C. Wu and E. Serpedin, “Timing Synchronization in Decode-and-Forward Cooperative Communication Systems”, IEEE Trans. on Signal Processing, vol. 57, no. 4, pp. 1444-1455, Apr 2009. [PDF]

  • Xiao Li, Y.-C. Wu and E. Serpedin, “Multiple Timing Offsets Compensation in Cooperative Communication Systems”, IEEE DSP 2009, Santorini, Greece, Jul. 2009.

  • Xiao Li, Y.-C. Wu and E. Serpedin, “On Performance Bounds for Timing Estimation under Fading Channels”, IEEE WCNC 2009, Budapest, Hungary, Apr 2009.