Application of Chaotic Systems for Side-Information Source Coding

Arithmetic coding is a kind of source coding that has been adopted in image compression standards such as JPEG-2000 and H.264/AVC. Recently, research on nonlinear dynamics reveals that finding the best initial condition for iterating a piecewise linear chaotic map to generate a symbolic representation corresponds to a message sequence is equivalent to arithmetic coding. This leads to the possibility of using chaotic systems for data compression. In this project, the researchers are going to extend the use of chaotic systems from standard source coding to distributed or side-information source coding. The outcome of the research will lead to a pioneer work along this direction and a better understanding of this topic. As a result, chaotic systems can be used for compression at a rate lower than the entropy with side information. This leads to significant application potential in video coding and sensor networks.