As an extension to the paper presented at SPS 2017 (in portuguese), we present a study on the use of kernel adaptive filters for channel equalization in a chaos-based communication system.

Different from the approaches that we have used in previous works, in this proposal, the receiver does not have knowledge of the system used in the transmitter for generating the chaotic signal and encoding the message. Thus, the idea is to use a kernel adaptive filter with two purposes: to equalize linear and nonlinear channels and, at the same time, decode the transmitted message from the chaotic signal. Simulation results indicate that the solution is able to perform these tasks, which may imply in a challenge for the secrecy in chaos-based communication systems argument.

The PDF file can be obtained in this page.