The Application Of Audio Information Hiding Technology In Military Communication

Information hiding is a new multi-disciplinary field, and it has an extensive appli-cation prospect in the realm of covert communication and digital watermarking. This thesis focuses on the application of this technology in covert communication, and the audio information hiding algorithms and applications are discussed thoroughly. The main work of this thesis is as follows:The principle and performance of spread spectrum information hiding algorithm are researched, and the methods of spread spectrum modulation and apperceptive scale are improved. Tamed spread spectrum modulation is used to replace direct sequence spread spectrum, and the embedded power of spreading sequence is decided by the critical bands’masking properties, and the cover object’s interference is reduced by normalizing discrete Fourier transform (DFT) coefficients. Experiment results show that the improved algorithm has some advantages including excellent robustness, good transparency and large capacity. The practicality of the improved algorithm is studied, and a covert communication experimental platform under noisy environment is de-signed. Experiments reveal that the algorithm improves the performance of resisting the assaults in transmission channel.The principle and performance of common quantization index modulation (QIM) algorithms are analyzed theoretically, and its disadvantages of fixed quantization step and sensitivity to amplitude scale are pointed out. According to these conclusions, an adaptive QIM algorithm based on masking properties of the human auditory system is proposed. The thought of critical bands partition is introduced to make the quantization steps vary with the critical bands’ masking effect, so the tradeoff between transparency and robustness is achieved. Based on the linearity property of DFT, the data is added to the normalized DFT coefficients to avoid the influence of amplitude scale. The algo-rithm is evaluated in the analog channel environment and its robustness is proved through the results.The robustness of line-spectrum pair (LSP) coefficients and the audible distortion caused by modification are investigated experimentally, and then a concrete algorithm thought which uses these coefficients to hide data is proposed. Combining the tech-niques of reduplicate coding and hiding position adaptive selection, the embedding and extracting methods are discussed in detail. The results show that the algorithm is robust against EVRC compression. Based on this characteristic, a scheme using voice service of military CDMA mobile communication system to hide information is put forward.