The abundance of large and heterogeneous systems is rendering contemporary data more pervasive, intricate, and with a non-regular structure. With classical techniques facing troubles to deal with the irregular (non-Euclidean) domain where the signals are defined, a popular approach at the heart of graph signal processing (GSP) is to: (i) represent the underlying support via a graph and (ii) exploit the topology of this graph to process the signals at hand. In addition to the irregular structure of the signals, another critical limitation is that the observed data is prone to the presence of perturbations, which, in the context of GSP, may affect not only the observed signals but also the topology of the supporting graph. Ignoring the presence of perturbations, along with the couplings between the errors in the signal and the errors in their support, can drastically hinder estimation performance. While many GSP works have looked at the presence of perturbations in the signals, much fewer have looked at the presence of perturbations in the graph, and almost none at their joint effect. While this is not surprising (GSP is a relatively new field), we expect this to change in the upcoming years. Motivated by the previous discussion, the goal of this thesis is to advance toward a robust GSP paradigm where the algorithms are carefully designed to incorporate the influence of perturbations in the graph signals, the graph support, and both. To do so, we consider different types of perturbations, evaluate their disruptive impact on fundamental GSP tasks, and design robust algorithms to address them.
Network Topology Inference with Sparsity and Laplacian Constraints
Hong Kong University of Science and Technology(香港科技大学);North China University of Technology(北方工业大学);Shenzhen Research Institute of Big Data(深圳大数据研究院);City University of Hong Kong(香港城市大学)
University of Rochester(罗切斯特大学);University of California Irvine(加州大学尔湾分校);University of Minnesota(明尼苏达大学);DEVCOM Army Research Lab.(美国陆军研究开发司令部陆军研究实验室)
Directed Graph Topology Inference via Graph Filter Identification
NVIDIA;Dept. of Signal Theory and Comms.(信号理论与通信系);King Juan Carlos University(胡安卡洛斯国王大学);Dept. of Electrical and Computer Eng.(电气与计算机工程系);Rice University(里士满大学);University of Rochester(罗切斯特大学)
Sequential Inference for Gaussian Processes: A Signal Processing Perspective
Basis Research Institute(基础研究机构);Department of Electrical and Computer Engineering, Stony Brook University(石溪大学电气与计算机工程系);Data Sciences Directorate, Brookhaven National Laboratory(布鲁赫斯国家实验室数据科学总局)
Energy-Efficient Processing and Robust Wireless Cooperative Transmission for Edge Inference
ShanghaiTech University(上海科技大学);Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences(中国科学院上海微系统与信息技术研究所);University of Chinese Academy of Sciences(中国科学院大学);University of Toronto(多伦多大学);University of California at Davis(加州大学戴维斯分校)