Introduction to coding theory (CMU, Spr 2010)

February 5, 2010

Lecture 8 summary

Filed under: Lecture summary — Venkat Guruswami @ 3:33 pm

We stated the first JPL bound {R(\delta) \le h(\frac{1}{2} - \sqrt{\delta (1-\delta)})} and outlined the approach of Friedman and Tillich to prove it for linear codes by showing that the dual of a code must have small “essential” covering radius and therefore be large. The method was generalized to non-linear codes by Navon and Samorodnitsky; we will mostly follow their presentation for our proof for the linear case.

We finished most of the (self-contained) proof using Fourier analysis. In particular, we proved the following facts about the Fourier spectrum which we will use to wrap up the proof at the beginning of next lecture:

  1. If {g} is a function on the hypercube and {A} is the adjacency matrix of the hypercube, then {\widehat{Ag}(\alpha) = \hat{g}(\alpha) (n- 2 {\rm wt}(\alpha))}.
  2. Let {f} be a function on the hypercube and for {z \in {\mathbb F}_2^n}, let {f_z} denote the shifted function {f_z(x) = f(z+x)}. Then for every subset {S \subseteq {\mathbb F}_2^n}, the Fourier transform of the function {F= \frac{1}{2^n} \sum_{z\in S} f_z} satisfies

    \displaystyle  \hat{F}(\alpha) = \widehat{1_S}(\alpha) \hat{f}(\alpha) \ ,

    where {1_S} is the characteristic function of the set {S}. (We will use this for {S = C^\perp}.)


Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at

%d bloggers like this: