Title :
Maximin performance of binary-input channels with uncertain noise distributions
Author :
McKellips, Andrew L. ; Verdu, Sergio
Author_Institution :
Dept. of Electr. Eng., Princeton Univ., NJ, USA
fDate :
5/1/1998 12:00:00 AM
Abstract :
We consider uncertainty classes of noise distributions defined by a bound on the divergence with respect to a nominal noise distribution. The noise that maximizes the minimum error probability for binary-input channels is found. The effect of the reduction in uncertainty brought about by knowledge of the signal-to-noise ratio is also studied. The particular class of Gaussian nominal distributions provides an analysis tool for near-Gaussian channels. The asymptotic behavior of the least favorable noise distribution and the resulting error probability are studied in a variety of scenarios, namely: asymptotically small divergence with and without power constraint; asymptotically large divergence with and without power constraint; and asymptotically large signal-to-noise ratio
Keywords :
Gaussian channels; Gaussian distribution; Gaussian noise; error statistics; maximum likelihood detection; minimax techniques; Gaussian nominal distributions; asymptotically large divergence; asymptotically large signal-to-noise ratio; asymptotically small divergence; binary-input channels; error probability; maximin performance; maximum-likelihood detection; minimum error probability; near-Gaussian channels; power constraint; signal distortion; signal-to-noise ratio; uncertain noise distributions; uncertainty classes; Background noise; Crosstalk; Detectors; Error probability; Gaussian noise; Interference constraints; Intersymbol interference; Jamming; Multiple access interference; Signal to noise ratio;
Journal_Title :
Information Theory, IEEE Transactions on