An efficient lossless compression using double Huffman minimum variance encoding technique

A Huffman code is a particular type of optimal prefix code that is commonly used for loss-less data compression. The process of finding such a code is known as Huffman coding. The output from Huffman's algorithm can be viewed as a variable-length code table for encoding a source symbol. The algorithm derives this table from the estimated probability or frequency of occurrence for each possible value of the source symbol. In this paper, we present a new approach to measure the performance and redundancy that work on two methods of coding like Huffman coding and Minimum Variance Huffman Coding. After getting the code-word for each symbol, we compress it on the basis of its binary values like 0 and 1 using binary coding. This is applied to both the approaches; this process is called as Double Huffman Coding. Finally we produce a better result than Huffman coding.