DocumentCode
825871
Title
Inverses of finite group systems
Author
Chizeck, Howard J.
Author_Institution
Massachusetts Institute of Technology, Cambridge, MA, USA
Volume
23
Issue
1
fYear
1978
fDate
2/1/1978 12:00:00 AM
Firstpage
66
Lastpage
70
Abstract
Inverse systems are considered for a class of discrete time-invariant systems that include the finite linear sequential circuits (LSC\´s). Invertibility results for finite group homomorphic sequential systems (FGHSS\´s), given by Willsky [8], are extended to include systems with throughput A construction is developed for an
-delay inverse of any FGHSS that is invertible with
delays. This inverse is always a discrete time-invariant system. Necessary and sufficient conditions are given for the inverse to consist of only homomorphic maps when the FGHSS state group is abelian, and only homomorphic and antihomomorphic maps when nonabelian. In the abelian case these conditions are necessary and sufficient for the existence of an inverse system of the specified delay that is itself an FGHSS. Invertible FGHSS\´s can be regarded as a generalization of convolutional encoders, since they include the class of invertible finite LSC\´s as a proper subclass.
-delay inverse of any FGHSS that is invertible with
delays. This inverse is always a discrete time-invariant system. Necessary and sufficient conditions are given for the inverse to consist of only homomorphic maps when the FGHSS state group is abelian, and only homomorphic and antihomomorphic maps when nonabelian. In the abelian case these conditions are necessary and sufficient for the existence of an inverse system of the specified delay that is itself an FGHSS. Invertible FGHSS\´s can be regarded as a generalization of convolutional encoders, since they include the class of invertible finite LSC\´s as a proper subclass.Keywords
Convolutional codes; Group theory; Inverse systems; Linear systems, time-invariant discrete-time; Sequential machines; Context modeling; Information analysis; Information filtering; Information filters; Mathematical model; Maximum likelihood estimation; Minimax techniques; Nonlinear filters; Statistics; Topology;
fLanguage
English
Journal_Title
Automatic Control, IEEE Transactions on
Publisher
ieee
ISSN
0018-9286
Type
jour
DOI
10.1109/TAC.1978.1101695
Filename
1101695
Link To Document