No label defined (Q3754529)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4001999
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 4001999 |
Statements
1987
0 references
Kalman filtering
0 references
generalized linear systems
0 references
causality
0 references
0.8530353903770447
0 references
0.8196895718574524
0 references
0.7953808903694153
0 references
This paper solves the Kalman filtering problem for generalized linear systems driven by Gaussian noises, by first writing the implicit system in an explicit form (under certain hypotheses) and then imposing a rather strong hypothesis for ''causality'' of the original implicit linear system. The notation throughout the paper is confusing. In the proof of Proposition I, no dimensions are given for \(x_ 1\), \(x_ 2\), or \(v_ 1\), \(v_ 2\), and equation (12) need not be used to obtain the form (3) of the conclusion of the proposition. In the second part of the paper, statistical notation is introduced before it is (if ever) defined. Examples of this are the expression for \(\hat x\) and the ''bar'' notation for mean used in \(\bar y{}_ 2\). (p. 353 line 13). The results are correct, although the assumption of surjectivity of \(L_ 2\) in (12) is rather strong. This assumption implies that the model (3), (4) (5) can actually be written in the usual explicit state space form (without an arbitrary \(x_ 2\) part of the state), which renders this result trivial.
0 references