| Nowadays,with the popularization of computer and network technology,the issue of information security has caused great attention ranging from the government,academic circles to the whole society.The academic community is continually exploring and introducing unconventional new methods for information security.The characteristics of chaos are easy to produce and difficult to decipher,so the application of chaotic sequences will help to protect information and enhance the security of information greatly in the information hiding field.As the easiest way in human communication,speech has been widely noticed in the development of communications technology.With its unique advantages,digital speech communication technology is widely used in recent years.Based on the background of digital technology and the digital chaotic signal,the digital speech signals are masked in order to enhance the reliability and security of signal transmission.Using independent component analysis(ICA)in the blind extraction of speech signal which is the classic one in blind source separation(BSS)algorithm to strengthen the security of information transmission has important theoretical value and practical significance.In this paper,the chaotic system is used to mask the digital speech signal and based on the information hiding technology the secure transmission of information is achieved.Three different dimensions of chaotic signals are put forward as the hidden carrier,the analyzing of the speech signal and chaotic signal are realized by wavelet transform.According to the time-frequency distribution characteristics of the two signals,which energy gathering bands could be extracted,and the hidden of speech signal could successful achieved.Based on the ideal positive definite communication system,the fast independent component analysis(FastICA)algorithm is used in our simulation experiment to separate the speech signal hidden in the chaotic signal blindly in the noisy channel,and the error rate and similarity coefficient compared and analyzed under different SNR conditions to verify the validity of the algorithm.On the basis ofverifying the feasibility of the positive definite system and the current communication situation,focusing on the problem that the receiving end antenna number is less than the number of transmit t end in the communication system,we propose that using Empirical Mode Decomposition algorithm(EMD)to construct virtual receiver arrays and to supplement the missing antenna array element,realizing the converting of an underdetermined system to a positive definite system.The blind extraction of digital speech signals can be carried out in different dimensional chaotic systems.Thus,the validity and universality of the algorithm and the reliability of digital technology in the background of noise are proved in this paper. |
