Tracking Time-Varying Systems: Revisiting Some Basic Trade-offs
1997 (English)In: Proceedings of the 11th IFAC Symposium on System Identification, 1997, Vol. 4, 1759-1764 p.Conference paper (Refereed)
The trade-off between tracking ability and noise sensitivity inadaptive algorithms is a classical research topic. In this paper we will revisit some of the main results. An input/output view of the archetypical algorithms for recursive identification will be taken, thus enabling the links to H∞ and H2 norms. We will shed some light on the H∞ optimality of the LMS algorithm and also show the corresponding merits of the RLS-type algorithms. Furthermore, the role of the distribution of the noise will be illustrated.
Place, publisher, year, edition, pages
1997. Vol. 4, 1759-1764 p.
Recursive identification, Estimation, Kalman filter, RLS, LMS, Noise sensivity, Adaptive algorithms
IdentifiersURN: urn:nbn:se:liu:diva-93780ISBN: 0080425925OAI: oai:DiVA.org:liu-93780DiVA: diva2:627955
11th IFAC Symposium on System Identification, Fukuoka, Japan, July, 1997