Application of deterministic chaos in generation of error-correction block codes

In this paper, we present a new method for constructing error-correction block codes which is based on theory of deterministic chaos. The method is naturally rooted on nonlinear iterative mappings and property of sensitive dependence on initial conditions. We describe the coding process for binary block codes and block codes with continuous amplitude. To justify the applicability of the new method to problem of error-correction, these codes are compared to maximum length binary codes and a class of asymptotically good codes. The results are encouraging: estimated coding gain for binary codes with respect to the union coding bound of the maximum length codes is significant.