Solving the error correcting code problem with parallel hybrid heuristics
Enrique Alba, Francisco Chicano · 2004
Some telecommunication systems can not afford the cost of repeating a corrupted message. Instead, the message should be somewhat "corrected" by the receiver. In these cases an error correcting code is suitable. The problem of finding an error correcting code of n bits and M codewords that corrects a given maximum number of errors is NP-hard. For this reason the problem has been solved in the literature with heuristic techniques such as Simulated Annealing and Genetic Algorithms. In this paper we present a new local search algorithm for the problem: the Repulsion Algorithm. We further use a hybrid between Parallel Genetic Algorithm and this new algorithm to solve the problem, and we compare it against a pure Parallel Genetic Algorithm. The results show that an important improvement is achieved with the inclusion of the Repulsion Algorithm and the parallelism.