Compare with the (best known) asymptotic bounds of Gilbert-Varshamov (lower) and McEliece-Rodemich-Rumsey-Welch (upper).
next
Problem for this talk:
Given d, find the largest k for which there exists a [1000,k,d] binary linear code.Too hard! So we consider the following much easier problem:
Given d, find the largest k for which the existence of a [1000,k,d] binary linear code is not contradicted by linear programming.
Largely worked out.
But how can we interpret the resulting data?
Compare with the (best known) asymptotic bounds of
Gilbert-Varshamov (lower) and McEliece-Rodemich-Rumsey-Welch (upper).
next