.png){kind=icon}
返回
Speech BioHashing security authentication algorithm based on CNN hyperchaotic map
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-022-12985-y.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The current speech content authentication process directly constructs the extracted biometrics into a binary hashing sequence and stores them in the cloud for hashing matching, so the authentication process is vulnerable to similarity attacks, which may lead to the disclosure of biometrics and hash sequences. In order to solve the above problem, this paper proposes an algorithm that is speech BioHashing security authentication algorithm based on CNN hyperchaotic map. Firstly, the user terminal conducts on linear prediction coefficients (LPC) analysis for the pre-processed speech signal, then extracts the improved Mel frequency cepstral coefficients (MFCC) features to obtains the new feature parameter, and calculates the Euclidean distance of the parameter as the biometric vector. Then, the biometrics and the orthogonal set matrix which is constructed by the improved random Fourier measurement matrix are inner product to form biosafety templates, and the templates are further quantified into BioHashing sequences. Finally, the BioHashing sequences are encrypted by CNN hyperchaotic map encryption algorithm and upload them to the cloud server. The experimental results show that the biosafety template not only provides the safe template for the biometric, but also the encryption algorithm of CNN hyperchaotic map guarantees the safety of the BioHashing sequence. At the same time, the algorithm has better robustness and discrimination, and tamper detection and location can better detect and locate mute attacks and substitution attacks.
Keyword:
Speech content authentication
CNN hyperchaotic map
Biosafety template
BioHashing
Tamper detection and location
期刊
IF:
3
论文数:
1.9W
被引数:
3.2W
机构
引用论文
Facial expression recognition based on improved octave convolution algorithm基于改进的八度卷积算法的面部表情识别.png)
IF0
PREAI
分享收藏Heart sound classification based on improved MFCC features and convolutional recurrent neural networksNEURAL NETWORKS
NEURAL NETWORKSIF6.3
PREAI
分享收藏Robust audio fingerprinting using peak-pair-based hash of non-repeating foreground audio in a real environment在真实环境中使用基于峰值对的非重复前景音频哈希进行鲁棒的音频指纹识别
PREAI
分享收藏分享收藏Color image encryption using orthogonal Latin squares and a new 2D chaotic system基于正交拉丁方和一个新的二维混沌系统的彩色图像加密
PREAI
分享收藏
查看更多内容