Kalman Filtering of Distributed Time Series

TL;DR

This paper adapts Kalman Filtering to handle collections of distributed time series, improving modeling and prediction of natural phenomena across geographical areas, with an application in meteorology.

Contribution

It introduces a novel adaptation of Kalman Filtering for distributed time series, enabling better system state modeling and prediction in spatially correlated data.

Findings

Enhanced accuracy in modeling meteorological data

Effective handling of correlated distributed signals

Demonstrated improvement over traditional methods

Abstract

This paper aims to introduce an application to Kalman Filtering Theory, which is rather unconventional. Recent experiments have shown that many natural phenomena, especially from ecology or meteorology, could be monitored and predicted more accurately when accounting their evolution over some geographical area. Thus, the signals they provide are gathered together into a collection of distributed time series. Despite the common sense, such time series are more or less correlated each other. Instead of processing each time series independently, their collection can constitute the set of measurable states provided by some open system. Modeling and predicting the system states can take benefit from the family of Kalman filtering algorithms. The article describes an adaptation of basic Kalman filter to the context of distributed signals collections and completes with an application coming from Meteorology.