Methods for Lossless Data Compression скачать в хорошем качестве

Methods for Lossless Data Compression

Methods for Lossless Data Compression, by Lawrence Esswood Abstract: Modern algorithms as well as hardware developments have allowed us to deal with very large quantities of data. The need to store and transport this data can, however, be very costly. Compression is a mechanism with which we can expect to reduce the load on our infrastructure. As such, the development of algorithms that can efficiently compress data is an active area of research. My talk will introduce some basic theory relating to compression and show how they can be put into practice. The talk will describe Arithmetic Coding, a technique for encoding symbols based on their source distribution. A method for estimating a distribution, Prediction By Partial Matching, will then be covered to provide a complete compression algorithm. More information: http://talks.cam.ac.uk/talk/index/56844