Minimal and canonical rational generator matrices for convolutional codes
The authors review and extend known properties of generator matrices for convolutional codes. In the past a lot of papers were devoted to showing under which conditions a polynomial generator matrix is minimal (i.e. using as few memory elements as possible for an encoder for the code) or canonical (i.e. a generator matrix that has a minimal realisation in a controller canonical form). These properties are tied up with the so-called predictable degree property. Later on these results were generalised to rational generator matrices, however the generalisations were not completely correct and no proofs were given. The present paper tries to complete and to correct the suggested generalisations. The properties of minimality and canonicality are connected with the so-called predictable valuation property which is a natural generalisation of the predictable degree property. Also some other concepts of valuation theory are used and discussed.
- The predictable degree property and minimality in multidimensional convolutional coding
- Matrix fraction descriptions in convolutional coding
- scientific article; zbMATH DE number 1498810 (Why is no real title available?)
- scientific article; zbMATH DE number 820493 (Why is no real title available?)
- State-space realizations of periodic convolutional codes
This page was built for publication: Minimal and canonical rational generator matrices for convolutional codes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3129473)
