Research On Blind Extraction And Optimization Algorithm Of Speech Signal Based On Quantum Chaos

With the rapid development of the Internet,hundreds of millions of sensitive voice data are being transmitted through open and shared networks every second.Therefore,ensuring that the transmission of voice information is safe and effective,and that it is not leaked will become very important.Using chaotic signals with strong randomness is the primary problem in designing a higher security level chaotic masking speech signal encryption method.The high structural complexity and high transmission security of quantum chaos make it possible to design a voice signal encryption method with a higher security level.Based on the highly complex structure of quantum chaos and the non-stationary characteristics of speech signals,this research proposes a speech signal encryption method under the concealment of quantum chaotic signals.This method uses quantum chaos as a masking signal to mask the speech signal to achieve encryption,which greatly enhances the security of the speech signal.Aiming at the problem of extracting and restoring the source signal from the observation signal,a method of extracting and restoring using an improved blind extraction technique is proposed.This method uses the Levi Flight Strategy Algorithm(LFSA)and Greedy Strategy(GSA)to optimize the Quantum Particle Swarm Optimization(QPSO)to improve the randomness of individual particle motion.The optimized quantum particle swarm algorithm is used to obtain the kurtosis of the observation signal,and finally the kurtosis of the signal is used as a fitness function to blindly extract the observation signal to realize the extraction of the speech signal.Simulation experiments show that the use of quantum chaotic signals to mask speech signals is reliable and safe.The quantum particle swarm optimization algorithm(KLG-QPSO)based on the Levi flight strategy greedy algorithm based on kurtosis and the blind based on independent component analysis(ICA)can achieve effective extraction for the source signal.