Time-critical problem-solving with cached knowledge: a case study in shortwave radio resource allocation

Successful communication between two places through shortwave radio depends on the right choice (sometimes within a very short period) of both frequen cy and power. This choice involves calling on mixtures of precise models an d heuristics about prevailing geophysical conditions, and knowledge of spec ific past experiences. The design and the features of a knowledge-based sys tem that is capable of answering these requirements for a given transmissio n request are described. There are two unusual constituents: a caching stru cture that can accommodate different knowledge representations and time lim its within which answers must be completed, and interpolation between items of knowledge in different cells of a cache. Results are demonstrated for s imple but realistic radio propagation exercises, to illustrate the effectiv eness of caching and interpolation for time-critical knowledge-based comput ations. (C) 1999 Elsevier Science Ltd. All rights reserved.