音频信息隐藏算法的研究与实现-翻译Word格式.docx

1、外文文献资料收集：苏州大学 应用技术学院11 电子班（学号 1116405046）汤珏MULTI-CHANNEL AUDIO INFORMATION HIDING ABSTRACTWe consider a method of hiding many audio channels in one host signal. The purpose of this is to provide a mix that incorporates information on all the channels used to produce it, thereby allowing all, or, at

2、least some channels to be stored in the mix for later use. After providing an overview of some recently published audio water marking schemes in the time and transform domains, we present a method that is based on using a four least signicant bits scheme to embed ve MP3 les into a single 16-bit host

3、 stereo WAV le without incurring any perceptual audio distortions in the host data. The host WAV le is taken to be the nal mix associated with the original data before applying lose MP3 compression or alternatively an arbitrary host audio signal into which other multichannel audio data is hidden. Fu

4、rther, the embedded information can be encrypted and/or the embedding locations randomized on a channel by channel basis depending on the protocol desired by the user. The method is illustrated by providing example m-code for interested readers to investigate and reproduce the results obtained to da

5、te and as a basis for further development.1. INTRODUCTIONThe approach to information hiding discussed in this paper relates to an investigation into the design of intelligent coding algorithms for audio post-production based on a current FP7 research project. The research has and is continuing to be

6、 undertaken jointly between the Audio Research Group, the Digital Media Centre at Dublin Institute of Technology and Tamborine Productions Limited. This includes the use of第 0 页watermarking methods for audio signal authentication and multiple audio signal embedding techniques using computational sol

7、utions based on image information hiding and authentication techniques, e.g.While there are a number of different approaches for hiding a signal in another host signal, inclusive of encryption, or otherwise, there are relatively few published methods that consider a multi-signal to single host signa

8、l approach. In this paper we consider the problem of embedding many audio signals provide as in MP3 formats into a single wave of WAV le under the following conditions: （i） the embedded MP3 audio data can be recovered without any loss of information; （ii） the host signal has no perceptual change in

9、its audio delity. An algorithm is presented based on a time domain method in which ve MP3 les are hidden in a single WAV le using a four least signicant bit scheme which satises these conditions.The method considered has applications in covert information transmission with applications in areas such

10、 as communications security, signal authentication and DRM （Digital Rights Management）, for example. However, the method reported has been developed in the context of an audio post-production product to allows a stereo mix to incorporate information on all or some of the channels used to produce the

11、 same mix, thereby allowing all or selective channels to be embedded in the mix for remixing at a later date and/or for data storage. Current scenarios usually involves a post-production centre having to maintain records of all the data used to generate a mix with the view that this same data may be

12、 required at a later date, thereby necessitating the need for a data base management infrastructure which may require signicant overheads.After providing an overview of recently published works that focuson the area of audio information hiding （Section 2）, we present a time domain solution to the pr

13、oblem in Section 3. This includes prototype m- code for interested readers to test the algorithms for themselves and as a第 16 页basis for developing the methodology further.2. AUDIO INFORMATION HIDINGAudio information hiding orAudio Steganography is based on two specic and distinct approaches which c

14、onsider the development of algorithms using time domain data or transform domain data such as the consine transform, for example. In this section, we provide a brief overview of some of the most recently published material in this area, publications that have been studied by the authors in associati

15、on with the multi-channel information hiding approach considered in this paper.2.1. Audio Hiding in the Time DomainTime domain audio hiding has the advantage of being relative simple to implement and computationally efcient. However the approach does not yield algorithms that are as diverse and robu

16、st as those developed for the transform domain. This is because of the relatively limited number of ways in which data can be manipulated in the time domain subject to the generation of an output that is perceptually compatible with the original audio signal. In the literature survey that follows, it is clear that all the methods conceived are based on some form of data manipulation b