论文信息 - A new parallel smoothing algorithm

A new parallel smoothing algorithm

This paper describes a new smoothing algorithm that uses a number of processors to smooth blocks of measurements in parallel with little communication between the processors. This procedure is found to have a total on-line computational complexity slightly higher than that of the non-parallel implementations. However, since several processors are used in parallel, the running time of this algorithm is much smaller than that of a single smoother solution.

[1] C. Striebel,et al. On the maximum likelihood estimates for linear dynamic systems , 1965 .

[2] Gerard L. Meyer,et al. An approach to reliable parallel data processing , 1982, 1982 21st IEEE Conference on Decision and Control.

[3] Thomas Kailath,et al. Square-root algorithms for parallel processing in optimal estimation, , 1979, Autom..

[4] David Q. Mayne,et al. A solution of the smoothing problem for linear dynamic systems , 1966, Autom..