# D we have i x x h x h x x

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ” X = 1 ˆ = Pr X = “?” . Hence, ﬁnding the rate distortion function consists of minimizing the mutual information under ˆ the constraint Pr X = “?” ≤ D . We have ˆ ˆ I (X ; X ) = H (X ) − H (X |X ) ˆ ˆ ˆ ˆ ˆ ˆ = 1 − Pr X = 0 H (X |X = 0) + Pr X = 1 H (X |X = 1) + Pr X = “?” H (X |X = “?”) =0 =0 ≤D ≤1 ≥ 1 − D · 1, where the last inequality is met with equality if PX |X (ˆ|x) = x ˆ 1 − D if x = x ˆ D if x = “?”. ˆ A simple coding scheme would be as follows. The encoder describes only the ﬁrst (1 − D )n bits of X n . The decoder then pads the remaining D n p ositions with the erasure symbol. Rate Distortion Function with Inﬁnite Distortion Problem 3 First, we note that for the computation of the rate distortion function we only need to consider ˆ D ∈ [0, 1/2], since for rate R = 0 we can achieve the distortion D = 1/2 by having X = 0 with probability 1. Note also that achieving ﬁnite distortion requires PX |X (ˆ = 0|x = 1) = 0. Hence, x ˆ ˆ th...
View Full Document

## This note was uploaded on 05/18/2013 for the course EE Informatio taught by Professor Amoslapidoth during the Fall '11 term at Swiss Federal Institute of Technology Zurich.

Ask a homework question - tutors are online