A Construction of Nonlinear Codes Which Betters or Equals Known Results for Certain Parameters

A new technique for constructing non-linear codes is presented, which, in at least two cases, yields larger codes of a given length and minimum distance than any previously known code (according to the table in F. J. MacWilliams and N. J. A. Sloane's “The Theory of Error Correcting Codes”, North-Holland, Amsterdam, 1977 ). The technique depends largely on identifying elements of GF(2k) with rows of a particular class of binary matrices. Several examples of the general theorem are given.