CALCULATING DEPENDENT SURVEILLANCE UPDATE RATES 
BY MODELING THE TIME-DEPENDENCE OF INFORMATION VALUE

Stephen Atkins, and R. John Hansman, Jr.


Abstract

The time-dependence of information has historically been managed though
manual procedures, typically developed heuristically, specifying how
frequently or under what conditions aging information should be
refreshed.  In the design of Automatic Dependent Surveillance (ADS)
applications, when aircraft should automatically broadcast state
information (e.g., position, velocity, or intent) must be addressed more
rigorously.  To answer questions about how frequently or under what
conditions aging information should be updated, when the cost of seeking
information must be accounted, a novel model of time-dependent
information value is developed for a class of proceduralized decision
problems, common in aviation.  The model combines elements of classic,
time-invariant information value theory, which is incapable of modeling
the relationship between information value and information age, with
estimation techniques.  The time-dependent information value model is
applied to investigate the rate at which aircraft should broadcast state
information in an ADS environment.  Results are presented for aircraft
flying on parallel trajectories as well as on crossing trajectories.
Results also provide general insight into the time-dependence of
information value and its implications on information management
processes.