端到端语音加密通信的同步信号设计

在语音加密通信过程中，接收方需要通过信号同步来实现精确的解密并恢复出语音信号，而现有方法精确度低且运算量大。为了解决端到端语音加密、解密过程中的同步问题，设计了一种新的基于线性调频信号的同步信号结构。该同步信号是基于带宽经过筛选的线性调频信号产生，避免了因为添加同步信号而使带宽扩展的问题。同步信号结构由两种不同长度的线性调频信号组合而成，不同的组合形式又会产生不同的同步效果。这种结构设计大大提高了同步的精确度和减少了同步运算量。理论和实验证明：该方法可以实现对接收信号帧起始位置进行精确地定位；不会展宽和影响语音信号的频谱；可以透过语音编解码器传输；具有一定的抗噪声性能；运算量比原始结构的同步信号大大减少。

Design of synchronous signal for end-to-end speech encryption communication

In the process of the encrypted speech communication,the receiver needs to synchronize the signal to decrypt the speech precisely and restore the original speech signal. However,the existing methods have low accuracy and cost a large computation. In order to solve the synchronization problem in the process of the end-to-end speech encryption and decryption,this paper designs a new synchronous signal structure based on the linear frequency modulation(LFM) signal. The synchronous signal is the chirp signal whose bandwidth has been filtered already and this method can avoid expanding the speech bandwidth caused by adding the synchronous signal. Synchronous signal structure is composed of two different lengths LFM signal and different combinations will produce different synchronization effect. This design greatly improves the accuracy of synchronization and reduces the amount of synchronous operation.Theoretical analysis and experimental verification show that the method can accurately locate the beginning of the received signal,and the speech signal spectrum will not be broadened and affected.This synchronous signal can be transmitted through the voice codec and has anti-noise performance. Compared with the original structure of the synchronous signal,the computation cost of the scheme is greatly reduced.