B2.14

Information Theory
Part II, 2001

A subset C\mathcal{C} of the Hamming space {0,1}n\{0,1\}^{n} of cardinality C=r|\mathcal{C}|=r and with the minimal (Hamming) distance min[d(x,x):x,xC,xx]=δ\min \left[d\left(x, x^{\prime}\right): x, x^{\prime} \in \mathcal{C}, x \neq x^{\prime}\right]=\delta is called an [n,r,δ][n, r, \delta]-code (not necessarily linear). An [n,r,δ][n, r, \delta]-code is called maximal if it is not contained in any [n,r+1,δ][n, r+1, \delta]-code. Prove that an [n,r,δ][n, r, \delta]-code is maximal if and only if for any y{0,1}ny \in\{0,1\}^{n} there exists xCx \in \mathcal{C} such that d(x,y)<δd(x, y)<\delta. Conclude that if there are δ\delta or more changes made in a codeword then the new word is closer to some other codeword than to the original one.

Suppose that a maximal [n,r,δ][n, r, \delta]-code is used for transmitting information via a binary memoryless channel with the error probability pp, and the receiver uses the maximum likelihood decoder. Prove that the probability of erroneous decoding, πerrml\pi_{\mathrm{err}}^{\mathrm{ml}}, obeys the bounds

1b(n,δ1)πerrml1b(n,[(δ1)/2])1-b(n, \delta-1) \leqslant \pi_{\mathrm{err}}^{\mathrm{ml}} \leqslant 1-b(n,[(\delta-1) / 2])

where

b(n,m)=0km(nk)pk(1p)nkb(n, m)=\sum_{0 \leqslant k \leqslant m}\left(\begin{array}{l} n \\ k \end{array}\right) p^{k}(1-p)^{n-k}

is a partial binomial sum and [][\cdot] is the integer part.