Event-Based Statistical Signal Processing

Authors:
  • YYYasin Yilmaz
    Yasin Yilmaz
  • GMGeorge V. Moustakides
    George V. Moustakides
  • XWXiaodong Wang
    Xiaodong Wang
  • AHAlfred O. Hero
    Alfred O. Hero

Abstract
This chapter analyzes the use of event-based sampling as a means of information transmission for decentralized detection and estimation. Event-based paradigm is an alternative to conventional time-driven systems in control and signal processing. Event-based methods are adaptive to the observed entities, as opposed to the time-driven techniques. In signal processing applications, event-based paradigm is mainly used as a means of nonuniform sampling. Time-encoding machine is a broad event-based sampling concept, in which the signal is compared with a reference signal and sampled at the crossings. Alternatively, in event-based sampling, some predefined events on the signal to be sampled trigger the sampling mechanism; that is, sampling times are determined by the signal and the event space. In signal processing applications, event-based paradigm is mainly used as a means of nonuniform sampling. The theory of signal reconstruction from nonuniform samples applies to event-based sampling.

This publication has 1 reference indexed in Scilit: