Engineering Awareness
Abstract
Generalized Process Tracking. Defined a rigorous concept of "trackability" of processes in a distributed sensing system. Established fundamental properties of processes and sensing infrastructure that are necessary and sufficient for certain types of trackability to be feasible. Problem addressed and solved: determine the "complexity" of estimating state trajectories of a target process based on a discrete-time sequence of noisy "observations". Conducted a comparative analysis of design methodologies for Agent-Based Systems. Machine Learning complex processes from data: discovery of a new algorithm to learn Hidden Markov Models (HMMs) from typical realizations of the associated stochastic process. The new method is based on the non-negative matrix factorization (NMF) of higher order Markovian statistics and is structurally different from the classical Baum-Welsh and associated approaches. Cognitive Complexification: development of new methods to shape network communications for preventing covert transmissions from hiding behind the statistics of ordinary traffic.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 07, 2010
- Accession Number
- ADA573423
Entities
People
- George Cybenko
- Valentino Crespi
Organizations
- California State University, Los Angeles