Kalman filtering of traffic fluctuations for real-time network management

TLDR: In this paper, a version of the Kaiman filter algorithm is adopted to deal with traffic fluctuations in telephone networks, and the overall approach is tested against the real data gathered by the supervisor of the French toll network.

Abstract: Discrete- time stochastic models of sequential traffic records in telephone networks are considered. They appear to be well suited for a use of the Kaiman filter. A version of the filtering algorithm is adopted to deal with traffic fluctuations. The overall approach is tested against the real data gathered by the supervisor of the French toll network. The adaptive traffic estimator is inserted into a simulator of a testbed network operated in the urban zone of Paris. The traffic estimations provided by the filter are used to decide an overflow hierarchy.