huffmanenco

Huffman encoder

Description

example

enco = huffmanenco(sig,dict) encodes input signal sig using the Huffman codes described by input code dictionary dict. sig can have the form of a numeric vector, numeric cell array, or alphanumeric cell array. If sig is a cell array, it must be either a row or a column. dict is an N-by-2 cell array, where N is the number of distinct possible symbols to be encoded. The first column of dict represents the distinct symbols and the second column represents the corresponding codewords. Each codeword is represented as a numeric row vector, and no codeword in dict can be the prefix of any other codeword in dict. You can generate dict using the huffmandict function.

Examples

collapse all

Create unique symbols, and assign probabilities of occurrence to them.

symbols = 1:6; 
p = [.5 .125 .125 .125 .0625 .0625];

Create a Huffman dictionary based on the symbols and their probabilities.

dict = huffmandict(symbols,p);

Generate a vector of random symbols.

sig = randsrc(100,1,[symbols;p]);

Encode the random symbols.

comp = huffmanenco(sig,dict);

Decode the symbols. Verify that the decoded symbols match the original symbols.

dsig = huffmandeco(comp,dict);
isequal(sig,dsig)
ans = logical
   1

Convert the original symbols to a binary symbols, and determine the length of the binary symbols.

binarySig = de2bi(sig);
seqLen = numel(binarySig)
seqLen = 300

Convert the Huffman-encoded symbols, and determine its length.

binaryComp = de2bi(comp);
encodedLen = numel(binaryComp)
encodedLen = 224

Define the alphanumeric symbols in cell array form.

sig = {'a2',44,'a3',55,'a1'}
sig=1×5 cell array
    {'a2'}    {[44]}    {'a3'}    {[55]}    {'a1'}

Define a Huffman dictionary. Codes for signal letters must be numeric.

dict = {'a1',0; 'a2',[1,0]; 'a3',[1,1,0]; 44,[1,1,1,0]; 55,[1,1,1,1]}
dict=5×2 cell array
    {'a1'}    {[       0]}
    {'a2'}    {1x2 double}
    {'a3'}    {1x3 double}
    {[44]}    {1x4 double}
    {[55]}    {1x4 double}

Encode the alphanumeric symbols.

comp = huffmanenco(sig,dict);

Decode the symbols. Verify that the decoded symbols match the original symbols.

dsig = huffmandeco(comp,dict)
dsig=1×5 cell array
    {'a2'}    {[44]}    {'a3'}    {[55]}    {'a1'}

isequal(sig,dsig)
ans = logical
   1

Input Arguments

collapse all

Input signal for compression, specified as a numeric vector, numeric cell array, or an alphanumeric cell array. sig can have the form of a numeric vector, numeric cell array, or alphanumeric cell array. If sig is a cell array, it must be a 1-by-S or S-by-1 cell array, where S is the number of symbols.

Data Types: double | cell

Huffman code dictionary, specified as an N-by-2 cell array. N is the number of distinct possible symbols for the function to encode. The first column of dict represents the distinct symbols and the second column represents the corresponding codewords. Each codeword is represented as a numeric row vector, and no codeword in dict can be the prefix of any other codeword in dict. You can generate dict by using the huffmandict function.

Data Types: double | cell

Output Arguments

collapse all

Encoded signal for input Huffman code dictionary dict, returned as a numeric row vector.

References

[1] Sayood, Khalid. Introduction to Data Compression. 2nd ed. San Francisco: Morgan Kaufmann Publishers, 2000.

Introduced before R2006a