Chaotic oscillators and complex mapping feed forward networks (CMFFNs) for signal detection in noisy environments

The use of chaotic systems for signal processing applications is limited by the ability to understand and interpret oscillator output results. Currently, phase plane data are used for system study, but neural networks are particularly well suited for this application. The authors have developed a complex-mapping-feedforward-network (CMFFN) that can interpret the phase plane data from chaotic systems. It is shown that this network, in conjunction with a chaotic oscillator, is able to distinguish signals buried in random Gaussian noise. The CMFFN is capable of detecting a signal with a 12-dB signal-to-noise ratio