Jie Fan, PhD

Student Name: Jie Fan, PhD, SenSIP 

Email: [email protected]

Biography

Jie Fan received his PhD from Arizona State University, co-advised by Dr. Andreas Spanias and Dr. Cihan Tepedelenlioglu. He completed B.E. degree in electrical engineering and the automatization from Huazhong University of Science and Technology, China, in 2014. His research interests include array processing, data recovery, data classification and graph theory. In recent years, his research involves data classification and clustering through graph-based techniques.

<!–
Background and Research Interests:

“Currently, I am working as a Ph.D. candidate in SenSIP lab at ASU.

I received my M.S. degree in electrical engineering-communications from school of electrical, computer and energy engineering at Arizona State University in Oct. 2016. Also the B.S. degree in electrical engineering is received from Huazhong University of Science and Technology (HUST) in May 2014.

In my master thesis, I developed a novel signal recovery algorithm for sensor array with failures. Comparing to conventional approaches based on interpolation and neural networks, our method requires no priori knowledge from the failure conditions and provides high success rate of recovering the missing data.

My current research is mainly abou graphic digital signal processing (GDSP). I focus on applying DSP operators to graphic data applications, such as data classification and data denoising. Those GDSP based methods are very suitable for dataset with complex graphic structure. Comparing to traditional machine learning algorithms, such methods may also provide better performance in success rate of estimation.”

 
–>

Research Work

In recent years, graph-structured datasets and their processing are experiencing a surge in various areas including wireless sensor networks, social networks, image processing and deep learning. Towards this goal, traditional DSP theory has been generalized to graph signal processing (DSPG). The DSPG framework enables applications of fundamental techniques, such as filtering and frequency analysis, to graph signals. Our recent research focuses on improving the accuracy of graph filters for data classification. We implement our algorithms on both real and synthetic datasets.

Research Picture:

 

Publications: 

  1. Jie Fan, Andreas Spanias and Cihan Tepedelenligolu, “A signal recovery method for array processing,” in IASTED International Conference on Modeling, Identification and Control, Feb. 2017
  2. Jie Fan, Tepedelenligolu and Andreas Spanias, “Graph filtering with multiple graph shift matrices,” in ICASSP, 2019. (Submitted)
  3. Uday S. Shanthamallu, Sunil Rao, Abhinav Dixit, Vivek S. Narayanaswamy, Jie Fan, Andreas Spanias, “Introducing machine learning in undergraduate DSP class,” in ICASSP, 2019. (Submitted)

Acknowledgements: 

I acknowledge the SenSIP Center of Electrical Engineering for the excellent advisory and support.