EE Seminar: Dithered Quantization via Orthogonal Transformations and a Cauchy-Schwarz like Inequality

~~
Speaker: Ran Hadad, 
M.Sc. student under the supervision of Prof. Uri Erez

Wednesday, February 10, 2016 at 15:00
Room 011, Kitot Bldg., Faculty of Engineering

Dithered Quantization via Orthogonal Transformations and a Cauchy-Schwarz like Inequality

Abstract

Dithered quantization is a technique used to reduce or eliminate the statistical dependence between the signal and quantization error. This is most often achieved via adding pseudo-random noise prior to quantization.

This work develops a different dithering method, where dithering is accomplished by applying an orthogonal transformation to a vector of samples prior to quantization, and applying its inverse to the output of the quantizer. Focusing on fixed rate uniform scalar quantization, it is shown that for any quantization rate, the proposed architecture approaches second-order independence, i.e., vanishing correlation, as the dimension of the vector of samples jointly processed grows.
 
We also describe another simple result that arose in the course of our work on the main topic. The result is a Cauchy-Schwarz like inequality for the correlation of two univariate functions operating on symmetric bivariate normal random variables. The inequality has some interesting applications beyond its use in the main part of this work. We briefly explore it as a criterion for the identification of a memoryless non-linearity.

 

10 בפברואר 2016, 15:00 
חדר 011, בניין כיתות-חשמל 
אוניברסיטת תל אביב עושה כל מאמץ לכבד זכויות יוצרים. אם בבעלותך זכויות יוצרים בתכנים שנמצאים פה ו/או השימוש
שנעשה בתכנים אלה לדעתך מפר זכויות, נא לפנות בהקדם לכתובת שכאן >>