中文摘要：

与传统矢量量化不同,索引约束矢量量化在量化过程中通过约束码字索引二进制形式中某一位的值来限定码字的搜索范围。利用了其特殊的码字搜索方法,提出的算法达到了在音频信号中添加水印的目的。将原始音频信号分段,每段进行DCT并提取若干中频系数构成矢量。水印嵌入时根据水印比特信息和预先设定的索引约束位的值找到匹配码字修改各段DCT中频系数。水印提取时利用传统矢量量化方法得到各量化索引值后,提取出各索引值中与嵌入端相同位的比特值即为水印信息。该方法在量化过程中嵌入水印信息,有很好的实时性。实验结果表明,利用该方法嵌入的水印为一种脆弱水印,可用于认证。

英文摘要：

Unlike normal vector quantization,index constrained vector quantization could limit the range of searching codewords by selecting a constrained position in the index in the process of quantization.This paper made use of this special searching method and the proposed algorithm reached the purpose of adding watermark in audio signal.First segmented the original audio signal into several frames,performed DCT transform on every frame and extracted the middle-frequnency coefficients to form vectors.The embedding procedure could be described as finding matching codewords to modify the DCT middle-frequency coefficient vectors in accordance with the watermarking bits and the constrained bits.It adopted normal vector quantization in the extracting procedure,extracted the constrained bits in the same position as the watermarking information.This scheme is real-time because the watermarks can be embeded in the quantization process.The experimental results show that the watermarking is fragile and can be used for authentication.