蒲越, 李国东, 赵静. 基于细胞神经网络混沌特性的音频加密技术应用[J]. 云南大学学报(自然科学版), 2017, 39(4): 539-546. doi: 10.7540/j.ynu.20150737
引用本文: 蒲越, 李国东, 赵静. 基于细胞神经网络混沌特性的音频加密技术应用[J]. 云南大学学报(自然科学版), 2017, 39(4): 539-546. doi: 10.7540/j.ynu.20150737
PU Yue, LI Guo-dong, ZHAO Jing. On application of audio encryption technology based on chaotic of Cellular Neural Network[J]. Journal of Yunnan University: Natural Sciences Edition, 2017, 39(4): 539-546. DOI: 10.7540/j.ynu.20150737
Citation: PU Yue, LI Guo-dong, ZHAO Jing. On application of audio encryption technology based on chaotic of Cellular Neural Network[J]. Journal of Yunnan University: Natural Sciences Edition, 2017, 39(4): 539-546. DOI: 10.7540/j.ynu.20150737

基于细胞神经网络混沌特性的音频加密技术应用

On application of audio encryption technology based on chaotic of Cellular Neural Network

  • 摘要: 细胞神经网络(Cellular Neural Network,简称CNN)是高度的非动力学系统,广泛应用于信息的加密处理等方面.首先将使用者个人设定的密码作为第1个一级密钥;其次利用构造的六阶细胞神经网络混沌系统,根据音频矩阵特征,自发选取混沌序列中与之相应的数作为第2个一级密钥,即达到不同的音频选取不同的密钥的效果,使信息安全性更高;然后将第1个一级密钥和第2个一级密钥按权重叠加生成二级密钥,并根据Logistic混沌映射生成三级密钥并加密原音频文件;最后仿真结果表明,加密后的音频文件波形与原音频文件波形相关性小,抗攻击性强,通过频谱比较,进一步证明了加密效果显著.因为第1个一级密钥是人为设定,第2个一级密钥是根据音频特征自动选取的,所以在避免了易遗失密钥而导致信息无法解开的同时,也避免了他人在没有密码的情况下破解文件的情况,故该算法具有高安全、高敏感等特性.

     

    Abstract: Cellular Neural Network (CNN) is a highly nonlinear system,which is widely used in the encryption of information.Firstly,it sets the user personal password key as aⅠgrade encryption key.Secondly,it uses the structure chaotic systems of six-order Cellular Neural Network according to the characteristics of the audio matrix and spontaneous selection in the chaotic sequence,the corresponding number in the chaotic sequence is chosen as the secondⅠgrade encryption key.Different audio options achieve different key effects,and the security of the information is higher.Then,the first Ⅰgrade encryption key and the second Ⅰgrade encryption key are overlapped to generate the Ⅱ grade encryption key,and the Ⅲ grade encryption key is generated according to the Logistic chaotic map and the original audio file is encrypted.Finally,the simulation results shows that the correlation of encrypted audio waveform is small than the original audio file waveform,strong ability of resistance to attack.Through spectrum comparison, it is further proved that the encryption effect is remarkable.Because the firstⅠgrade encryption key is artificially set,the second Ⅰgrade encryption key is automatically selected according to the feature of the audio,so it not only avoids easy to lose keys and cause information to be untied,but also avoids the case that others break the file without the password.This algorithm has the highly security and sensitive features.

     

/

返回文章
返回