Ali H. Sayed
Distributed networks linking sensors and actuators will form the backbone of future data communication and control networks. Applications will range from sensor networks to precision agriculture, environment monitoring, disaster relief management, smart spaces, target localization, as well as medical applications. In all these cases, the distribution of the nodes in the field yields spatial diversity, which should be exploited alongside the temporal dimension in order to enhance the robustness of the processing tasks and improve the probability of signal and event detection. Distributed processing techniques allow for the efficient extraction of temporal and spatial information from data collected at such distributed nodes by relying on local cooperation and data processing. For example, each node in the network could collect noisy observations related to a certain parameter of interest. The nodes would then interact with their neighboring nodes, as dictated by the network topology, in order to estimate the parameter. The objective is to arrive at an estimate that is as reliable as the one that would be obtained if each node had access to the information across the entire network.
In contrast, in the centralized approach to parameter estimation, the data from all nodes would be conveyed to a central processor where they would be fused and the vector of parameters estimated. Such an approach requires sufficient communications resources to transmit the data back and forth between the nodes and the central processor, which would limit the autonomy of the network besides adding a critical point of failure in the network due to the presence of a central node. A centralized solution also limits the ability of the nodes to adapt in real-time to time varying statistical profiles in the data and the environment.
This tutorial describes recent development in distributed processing over adaptive networks. The presentation covers adaptive algorithms that allow neighboring nodes to communicate with each other at every iteration. At each node, estimates exchanged with neighboring nodes are fused and promptly fed into the local adaptation rules. In this way, a truly adaptive network is obtained where the structure as a whole is adaptive, and is able to respond in real-time to the temporal and spatial variations in the statistical profile of the data. Different adaptation or learning rules at the nodes, allied with different cooperation protocols, give rise to adaptive networks and topologies of various complexities and potential. Obviously, the effectiveness of any distributed implementation depends on the modes of cooperation that are allowed among the nodes such as incremental, diffusion or probabilistic diffusion.
The tutorial will develop the concept of adaptive networks and show how to design adaptive strategies for distributed processing. The designs will illustrate three major ways of node collaboration (incremental, diffusion, and probabilistic diffusion). The tutorial will further illustrate how to study the performance of an interconnected network of adaptive nodes that are subject to varying spatial and temporal data profiles. Energy conservation arguments and state-space formulations will be exploited to great effect to evaluate the performance of the individual nodes and the network as a whole.
Ali H. Sayed (S’90-M’92-SM’99-F’01) received the Ph.D. degree in electrical engineering from Stanford University, Stanford, CA, in 1992. He is Professor and Chairman of Electrical Engineering at the University of California, Los Angeles (UCLA). He is also the Principal Investigator of the UCLA Adaptive Systems Laboratory (www.ee.ucla.edu/asl).
He is the author or coauthor of over 290 journals and conference publications and 5 books. He is the author of the textbooks Adaptive Filters (New York: Wiley, 2008), Fundamentals of Adaptive Filtering (New York: Wiley, 2003), the coauthor of the research monograph Indefinite Quadratic Estimation and Control (Philadelphia, PA: SIAM, 1999), and of the graduate-level textbook Linear Estimation (Englewood Cliffs, NJ: Prentice-Hall, 2000). He is also coeditor of Fast Reliable Algorithms for Matrices with Structure (Philadelphia, PA: SIAM, 1999).
He has contributed several articles to engineering and mathematical encyclopedias and handbooks and has served on the program committees of several international meetings. His research interests include adaptive and statistical signal processing, distributed processing, filtering and estimation theories, signal processing for communications, interplays between signal processing and control methodologies, system theory, and fast algorithms for large-scale problems.
Dr. Sayed is a recipient of the 1996 IEEE Donald G. Fink Award, a 2002 Best Paper Award from the IEEE Signal Processing Society, the 2003 Kuwait Prize in Basic Science, the 2005 Frederick E. Terman Award, the 2005 Young Author Best Paper Award from the IEEE Signal Processing Society, and two Best Student Paper awards at international meetings. He has served as Editor-in-Chief of the IEEE Transactions on Signal Processing (2003-2005) and the EURASIP Journal on Advances on Signal Processing (2006-2007). He also served as 2005 Distinguished Lecturer of the IEEE Signal Processing Society, and as General Chairman of the 2008 International Conference on Acoustics, Speech, and Signal Processing (ICASSP). He chairs the Technical Committee on Signal and Systems (SPTM) of the IEEE Signal Processing Society, is Vice-President-Elect (Publications), and sits on the Board of Governors of the same society.