By Topic

Enumeration of minimal convolutional encoders (Corresp.)

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)

A general ensemble of all rateK/Nconvolutional encoders with overall constraint length (OCL)nuis considered, and the notion of a minimum overall constraint length (MOCL) encoder is introduced. The noncatastrophic subclass of this class of zero delay MOCL encoders is the class of minimal encoders. Using this fact, upper and lower bounds on the number of rateK/Nminimal encoders are obtained. Results for finite values ofnuand asymptotic results for largenuare obtained. In all cases (except in the limit asK rightarrow inftyfor rateK/(K + 1))the lower bound is a nonzero fraction of the upper bound, which in turn is a nonzero fraction of the total number of encoders. This implies that for an arbitrary fixed rate and for any OCL, a nonzero fraction of encoders is minimal, and hence noncatastrophic. In general, as the rate decreases the fraction of all zero delay encoders that are minimal increases. The upper and lower bounds tend to merge in the low rate case providing a good estimate of the number of minimal encoders. The results in the high rate case are not as satisfactory since the upper and lower bounds are widely separated. The results also indicate that for a given rate the probability of selecting a minimal encoder increases with increasingK. Thus there may be some advantages to considering encoders with an increased number of input and output sequences.

Published in:

Information Theory, IEEE Transactions on  (Volume:23 ,  Issue: 1 )