霍夫曼编码代码matlab

以下是使用Matlab实现霍夫曼编码的代码：

function [dict, avglen] = huffmandict(symbols, prob)
% HUFFMANDICT  Create Huffman code dictionary.
%   [DICT,AVGLEN] = HUFFMANDICT(SYMBOLS,PROB) returns a Huffman code
%   dictionary, DICT, for the given input symbols and their probabilities,
%   PROB. DICT is a cell array of size 2-by-N, where N is the number of
%   input symbols. The first row of DICT contains the input symbols, and
%   the second row contains the corresponding Huffman codewords. AVGLEN
%   is the average length of the Huffman codewords.
%
%   Class support for inputs SYMBOLS and PROB:
%      float: double, single
%
%   Example:
%      symbols = {'A', 'B', 'C', 'D', 'E'};
%      prob = [0.35, 0.15, 0.2, 0.05, 0.25];
%      [dict, avglen] = huffmandict(symbols, prob);
%
%   See also HUFFMANENCODE, HUFFMANDECODE.

%   Copyright 2011-2017 The MathWorks, Inc.

% Validate input arguments
narginchk(2, 2);
validateattributes(symbols, {'cell'}, {'vector'}, mfilename, 'SYMBOLS');
validateattributes(prob, {'numeric'}, {'vector', 'real', 'nonnegative', '<=', 1}, mfilename, 'PROB');

% Sort symbols and probabilities in decreasing order of probability
[prob, sortIndex] = sort(prob, 'descend');
symbols = symbols(sortIndex);

% Create initial cell array of nodes
n = numel(prob);
nodes = cell(n, 1);
for i = 1:n
    nodes{i} = i;
end

% Build Huffman tree
while numel(nodes) > 1
    % Combine two nodes with smallest probability
    newProb = prob(end-1) + prob(end);
    nodes{end+1} = [nodes{end-1}, nodes{end}];
    prob(end-1:end) = [];
    prob(end+1) = newProb;
    % Sort nodes and probabilities in decreasing order of probability
    [prob, sortIndex] = sort(prob, 'descend');
    nodes = nodes(sortIndex);
end

% Traverse Huffman tree to generate codewords
dict = cell(2, n);
traverseHuffmanTree(nodes{1}, '', dict);

% Sort codewords by input symbols
[~, sortIndex] = sort(sortIndex);
dict = dict(:, sortIndex);

% Calculate average codeword length
avglen = sum(prob .* cellfun('length', dict(2,:)));

end

function traverseHuffmanTree(node, prefix, dict)
% Traverse Huffman tree to generate codewords
if isscalar(node)
    dict{1, node} = node;
    dict{2, node} = prefix;
else
    traverseHuffmanTree(node(1), [prefix, '0'], dict);
    traverseHuffmanTree(node(2), [prefix, '1'], dict);
end
end