An Objective-Based Experimental Design Framework for Signal Processing in the Context of Canonical Expansions Conference Paper uri icon

abstract

  • © 2017 IEEE. This work presents an objective-based experimental design framework for uncertainty reduction in filtering and signal processing when the underlying random processes are expressed in terms of canonical expansions. Canonical expansions are convenient representations for random processes that facilitate finding closed-form solutions for operators (filters). In the proposed experimental design framework, uncertainty is quantified based on the concept of mean objective cost of uncertainty, which measures uncertainty by taking into account the ultimate modeling objective, which is optimal filtering.

author list (cited authors)

  • Dehghannasiri, R., Qian, X., & Dougherty, E. R.

citation count

  • 0

publication date

  • October 2017

publisher