Parallel Kalman filtering for real-time signal identification

Abstract EN

The applicability of Kalman filter (KF) to real-time signal processing problems is limited by the relatively complex mathematical operations necessary in computing Kalman filtering algorithms.

However, with the rapid development of fast processing/memory devices that has offered a new research direction in high-speed real-time, systematic implementation on KF.

Presently, the research trend is to achieve a major improvement in computational speed that will come from the concurrent use of many processor cells.Parallel processing usually makes a major impact in real-time signal identification.

These require high-speed computations, which must be performed on continuous data streams.

This result here in the stimulation of novel architectures for parallel Kalman filter (PKF).

The implementation of the PKF is achieved on a simulated radar signal that works in real-time by using the simulink package.

Key words: Signal identification; Real-time signal processing; Parallel processing; Parallel Kalman filter; Simulink package.