Over the past few years, several methods have been proposed for decontaminating noisy chaotic signals by exploiting the short-term predictability of chaotic signals. This work evaluates the effectiveness, for a differential chaos shift keying (DCSK) telecommunications system, of a noise reduction approach using a deterministic optimization technique. Noise reduction is governed by a cost function which consists of two terms: the first gives the distance between the noisy and enhanced orbits, while the second one checks the dynamics of the cleaned signal. These two terms are weighted by a scalar Γ. The effect of this factor on the noise reduction performance is also studied. Evaluation of the noise performance of a telecommunication system by computer simulation requires a very long simulation time. We propose a computationally-efficient criterion for quantifying the performance improvement of a DCSK system. We show that the noise reduction technique improves the overall noise performance only if the energy per bit-to-noise spectral density (Eb/N0) exceeds a certain threshold. The effect of code length on this threshold level is also evaluated. Finally, the effect of parameter mismatch, which is present in every practical system, is analyzed.
|Number of pages||6|
|Journal||IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications|
|Publication status||Published - Dec 1 2000|
ASJC Scopus subject areas
- Electrical and Electronic Engineering