Abstract
We present a rigorous approach to nonlinear estimation of signals in additive white Gaussian noise using the innovations method. The general argument is deliberately chosen to follow the same lines as in the linear case given in a previous note. Although a much heavier use of martingales is necessary here, the parallelism is quite striking.
Original language | American English |
---|---|
Title of host publication | 1973 IEEE Conference on Decision and Control including the 12th Symposium on Adaptive Processes |
State | Published - 1973 |