English

On extracting common random bits from correlated sources on large alphabets

Information Theory 2012-08-30 v1 math.IT

Abstract

Suppose Alice and Bob receive strings X=(X1,...,Xn)X=(X_1,...,X_n) and Y=(Y1,...,Yn)Y=(Y_1,...,Y_n) each uniformly random in [s]n[s]^n but so that XX and YY are correlated . For each symbol ii, we have that Yi=XiY_i = X_i with probability 1\eps1-\eps and otherwise YiY_i is chosen independently and uniformly from [s][s]. Alice and Bob wish to use their respective strings to extract a uniformly chosen common sequence from [s]k[s]^k but without communicating. How well can they do? The trivial strategy of outputting the first kk symbols yields an agreement probability of (1\eps+\eps/s)k(1 - \eps + \eps/s)^k. In a recent work by Bogdanov and Mossel it was shown that in the binary case where s=2s=2 and k=k(\eps)k = k(\eps) is large enough then it is possible to extract kk bits with a better agreement probability rate. In particular, it is possible to achieve agreement probability (k\eps)1/22k\eps/(2(1\eps/2))(k\eps)^{-1/2} \cdot 2^{-k\eps/(2(1 - \eps/2))} using a random construction based on Hamming balls, and this is optimal up to lower order terms. In the current paper we consider the same problem over larger alphabet sizes ss and we show that the agreement probability rate changes dramatically as the alphabet grows. In particular we show no strategy can achieve agreement probability better than (1\eps)k(1+δ(s))k(1-\eps)^k (1+\delta(s))^k where δ(s)0\delta(s) \to 0 as ss \to \infty. We also show that Hamming ball based constructions have {\em much lower} agreement probability rate than the trivial algorithm as ss \to \infty. Our proofs and results are intimately related to subtle properties of hypercontractive inequalities.

Keywords

Cite

@article{arxiv.1208.5946,
  title  = {On extracting common random bits from correlated sources on large alphabets},
  author = {Siu On Chan and Elchanan Mossel and Joe Neeman},
  journal= {arXiv preprint arXiv:1208.5946},
  year   = {2012}
}

Comments

15 pages, 1 figure

R2 v1 2026-06-21T21:56:53.645Z