Title :
Transputer parallel implementation algorithms for multisensor Kalman filtering
Author_Institution :
Dept. de Comput. Sci., Quebec Univ., Hull, Que., Canada
Abstract :
The objective of this paper is to analyze some factors influencing the performance of different farmer parallel programs, developed for the hierarchical Kalman filtering. During our research we systematically modified a number of hardware and/or software parameters which characterize the farmer and we have shown that these modifications has a reduced influence to linear speed-up rate
Keywords :
Kalman filters; computational complexity; discrete time filters; filtering theory; hierarchical systems; parallel algorithms; parallel architectures; pipeline processing; sensor fusion; software performance evaluation; farmer parallel programs; hierarchical Kalman filtering; linear speed-up rate; multisensor Kalman filtering; parallel implementation algorithms; transputer; Covariance matrix; Equations; Filtering algorithms; Kalman filters; Noise figure; Noise measurement; Parallel algorithms; Signal processing algorithms; State estimation; Time measurement;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 1997. IMTC/97. Proceedings. Sensing, Processing, Networking., IEEE
Conference_Location :
Ottawa, Ont.
Print_ISBN :
0-7803-3747-6
DOI :
10.1109/IMTC.1997.603938
