Investigating the dynamics of watermark features in audio streams

The dynamics of watermark features after embedded into audio streams through digital watermarking techniques are unstable. The audio streams exits as a series of waveform amplitude of sound over the range of information it contains. Within this range, there are variations of the presentation of the stream taken per second and given in hertz. The precision of the stream representations is measured by the number of bits per stream. The fact that the streams bits are high is a sign for data already existing which means that within empty streams additional information can be embedded. In general, added information is described as noise and these audio streams are considered as noise tolerant. Watermarks are embedded into a spatial or transformed domain with the effect that the presentation of some bit streams will be affected. This paper investigates the dynamics of watermarks embedded in an audio stream, the contained file being noise intolerant. The watermark file is embedded in several positions within the audio signal stream by spread spectrum techniques. The most suitable positions for embedding the watermark is identified which ensures a strong and robust watermark as a result.