Blind Estimation of Chaotic Spread Spectrum Sequences by Neural Network

Chaotic spread spectrum sequences have higher complexity than traditional direct spread sequences, but they are difficult to estimate chaotic direct spread sequences blindly. In order to blind estimate it effectively, an improved method is proposed to blind estimate the chaotic spread spectrum sequences based on the neural network. This method takes full advantages of the neural network's nonlinearity and increases the blind signal separation module. The simulation results show that the method does not need to search the synchronization point between the information code and the spreading sequence. Even under the condition of low SNR(signal to noise ratio), the chaotic spread spectrum signal can be effectively separated from the noise background and blind. The original chaotic sequence is also estimated.