Namespaces
Variants
Actions

Difference between revisions of "Huffman code"

From Encyclopedia of Mathematics
Jump to: navigation, search
(Importing text file)
 
m (tex encoded by computer)
 
Line 1: Line 1:
An algorithm for the construction of optimal (i.e., minimum redundancy) variable-length <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103401.png" />-ary codes (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103402.png" />) for finite memoryless information sources with known statistics was proposed by D. Huffman in 1952 (see [[#References|[a1]]], [[#References|[a2]]]). (The algorithm refines the Morse alphabet, as it associates the shortest codewords to the most probable source letters.)
+
<!--
 +
h1103401.png
 +
$#A+1 = 73 n = 0
 +
$#C+1 = 73 : ~/encyclopedia/old_files/data/H110/H.1100340 Huffman code
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
  
Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103403.png" /> be such a source with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103404.png" /> letters, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103405.png" />, and let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103406.png" /> be the probability that the source emits letter <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103407.png" /> (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103408.png" />). Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h1103409.png" /> be a uniquely decipherable <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034011.png" />-ary code for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034012.png" /> (i.e., a set of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034013.png" /> sequences <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034014.png" /> from an alphabet of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034015.png" /> symbols) and let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034016.png" /> be the length (or number of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034017.png" />-ary symbols) of codeword <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034018.png" /> (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034019.png" />) associated to the letter <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034020.png" />. The difference
+
{{TEX|auto}}
 +
{{TEX|done}}
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034021.png" /></td> </tr></table>
+
An algorithm for the construction of optimal (i.e., minimum redundancy) variable-length  $  D $-
 +
ary codes ( $  D \geq  2 $)
 +
for finite memoryless information sources with known statistics was proposed by D. Huffman in 1952 (see [[#References|[a1]]], [[#References|[a2]]]). (The algorithm refines the Morse alphabet, as it associates the shortest codewords to the most probable source letters.)
  
between the average length <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034022.png" /> of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034023.png" /> and the source entropy (cf. also [[Entropy|Entropy]]) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034024.png" /> is called the redundancy of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034025.png" />. A Huffman code <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034026.png" /> for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034027.png" /> minimizes <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034028.png" /> (hence <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034029.png" />) over the set of all uniquely decipherable codes for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034030.png" />.
+
Let  $  A _ {K} $
 +
be such a source with  $  K $
 +
letters,  $  K \geq  2 $,
 +
and let  $  p _ {i} $
 +
be the probability that the source emits letter  $  a _ {i} $(
 +
$  1 \leq  i \leq  K $).  
 +
Let  $  {\mathcal C} = \{ w _ {1} \dots w _ {K} \} $
 +
be a uniquely decipherable  $  D $-
 +
ary code for $  A _ {K} $(
 +
i.e., a set of  $  K $
 +
sequences  $  w _ {i} $
 +
from an alphabet of  $  D $
 +
symbols) and let  $  l _ {i} $
 +
be the length (or number of  $  D $-
 +
ary symbols) of codeword  $  w _ {i} $(
 +
$  1 \leq  i \leq  K $)
 +
associated to the letter  $  a _ {i} $.  
 +
The difference
  
In the binary case (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034031.png" />), the algorithm consists in the construction of a sequence of "reduced"  sources <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034032.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034033.png" /> is obtained from <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034034.png" /> by merging the two least probable letters of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034035.png" /> into one letter of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034036.png" /> (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034037.png" />) and leaving all other letters unchanged (ties are broken arbitrarily). The probability of the new letter is the sum of the probabilities of the merged letters. The <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034038.png" /> mergings can be represented by a binary code tree in which each letter of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034039.png" /> corresponds to a leaf and each codeword is the sequence of binary labels of the path from root to leaf. (If the code alphabet has <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034040.png" /> letters, each reduced source is obtained by merging the <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034041.png" /> least probable letters of the previous source, except possibly in the first step, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034042.png" /> letters have to be grouped together, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034043.png" /> being the remainder of the division of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034044.png" /> by <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034045.png" />. The resulting <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034046.png" />-ary code tree is labeled and again the codewords are the sequences of labels from root to leaves.)
+
$$
 +
R ( {\mathcal C} ) = l ^ {*} ( {\mathcal C} ) - H ( P )
 +
$$
  
Owing to the arbitrary resolution of possible ties in grouping the least probable sequences at each step, for a given source <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034047.png" /> there may be several distinct Huffman codes, but two such codes are considered (essentially) different only if the corresponding sets <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034048.png" /> of codeword lengths do not coincide. However, the average length of all Huffman codes for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034049.png" /> is the same.
+
between the average length  $  l  ^ {*} ( {\mathcal C} ) = \sum _ {i = 1 }  ^ {K} p _ {i} l _ {i} $
 +
of $  {\mathcal C} $
 +
and the source entropy (cf. also [[Entropy|Entropy]])  $  H ( P ) = - \sum _ {i = 1 }  ^ {K} p _ {i} { \mathop{\rm log} } _ {D} p _ {i} $
 +
is called the redundancy of $  {\mathcal C} $.  
 +
A Huffman code  $  {\mathcal H} $
 +
for  $  A _ {K} $
 +
minimizes  $  R ( {\mathcal C} ) $(
 +
hence  $  l  ^ {*} ( {\mathcal C} ) $)
 +
over the set of all uniquely decipherable codes for $  A _ {K} $.
  
More formally, consider the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034050.png" /> of all <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034051.png" />-order probability distributions (cf. [[Probability distribution|Probability distribution]]) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034052.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034053.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034054.png" /> (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034055.png" />). The informational divergence (also called the [[Kullback–Leibler information|Kullback–Leibler information]]) between two probability distributions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034056.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034057.png" /> in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034058.png" /> is the non-negative quantity <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034059.png" />. <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034060.png" /> vanishes if and only if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034061.png" />, hence it can be considered as an oriented pseudo-distance (cf. also [[Pseudo-metric|Pseudo-metric]]) in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034062.png" />. Now, if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034063.png" /> is the set of codeword lengths of a Huffman code <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034064.png" /> for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034065.png" />, the redundancy <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034066.png" /> is equal to the informational divergence <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034067.png" /> between the source probability distribution <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034068.png" /> and the  "dyadic" probability distribution <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034069.png" />.
+
In the binary case ( $  D = 2 $),  
 +
the algorithm consists in the construction of a sequence of  "reduced" sources  $  \{ A _ {K - 1 }  \dots A _ {2} \} $,
 +
where  $  A _ {j - 1 }  $
 +
is obtained from  $  A _ {j} $
 +
by merging the two least probable letters of  $  A _ {j} $
 +
into one letter of $  A _ {j - 1 }  $(
 +
$  K \geq  j \geq  3 $)  
 +
and leaving all other letters unchanged (ties are broken arbitrarily). The probability of the new letter is the sum of the probabilities of the merged letters. The  $  K - 2 $
 +
mergings can be represented by a binary code tree in which each letter of  $  A _ {K} $
 +
corresponds to a leaf and each codeword is the sequence of binary labels of the path from root to leaf. (If the code alphabet has  $  D > 2 $
 +
letters, each reduced source is obtained by merging the $  D $
 +
least probable letters of the previous source, except possibly in the first step, where  $  2 + R _ {D - 1 }  ( K - 2 ) $
 +
letters have to be grouped together,  $  R _ {a} ( b ) $
 +
being the remainder of the division of  $ b $
 +
by  $ a $.  
 +
The resulting  $  D $-
 +
ary code tree is labeled and again the codewords are the sequences of labels from root to leaves.)
  
The optimality of Huffman codes can now be restated by saying that to any <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034070.png" /> (i.e., to any source <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034071.png" />), the algorithm associates (one of) the dyadic probability distribution(s) "closest" to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034072.png" /> in the sense of the pseudo-metric <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034073.png" />.
+
Owing to the arbitrary resolution of possible ties in grouping the least probable sequences at each step, for a given source  $  A _ {K} $
 +
there may be several distinct Huffman codes, but two such codes are considered (essentially) different only if the corresponding sets $ \{ l _ {1} \dots l _ {K} \} $
 +
of codeword lengths do not coincide. However, the average length of all Huffman codes for  $  A _ {K} $
 +
is the same.
  
In several practical situations the statistics of the source is not perfectly known or varies over time, and the necessity arises of updating the Huffman code when the new(ly known) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/h/h110/h110340/h11034074.png" /> becomes  "closer"  to a dyadic probability distribution different from the previous one (see [[#References|[a3]]], [[#References|[a4]]]).
+
More formally, consider the set  $  {\mathcal P} _ {K} $
 +
of all  $  K $-
 +
order probability distributions (cf. [[Probability distribution|Probability distribution]])  $  P = \{ p _ {1} \dots p _ {K} \} $,
 +
$  \sum _ {i = 1 }  ^ {K} p _ {i} = 1 $,
 +
$  p _ {i} > 0 $(
 +
$  1 \leq  i \leq  K $).
 +
The informational divergence (also called the [[Kullback–Leibler information|Kullback–Leibler information]]) between two probability distributions  $  P $,
 +
$  Q $
 +
in  $  {\mathcal P} _ {K} $
 +
is the non-negative quantity  $  D ( P \mid  Q ) = \sum _ {i = 1 }  ^ {K} p _ {i} { \mathop{\rm log} } ( { {p _ {i} } / {q _ {i} } } ) $.
 +
$  D ( P \mid  Q ) $
 +
vanishes if and only if  $  P \equiv Q $,
 +
hence it can be considered as an oriented pseudo-distance (cf. also [[Pseudo-metric|Pseudo-metric]]) in  $  {\mathcal P} _ {K} $.
 +
Now, if  $  \{ l _ {1} \dots l _ {K} \} $
 +
is the set of codeword lengths of a Huffman code  $  {\mathcal H} $
 +
for  $  A _ {K} $,
 +
the redundancy  $  R ( {\mathcal H} ) $
 +
is equal to the informational divergence  $  D ( P \mid  H ) $
 +
between the source probability distribution  $  P $
 +
and the  "dyadic"  probability distribution  $  H = \{ 2 ^ {- l _ {1} } \dots 2 ^ {- l _ {K} } \} $.
 +
 
 +
The optimality of Huffman codes can now be restated by saying that to any  $  P \in {\mathcal P} _ {K} $(
 +
i.e., to any source  $  A _ {K} $),
 +
the algorithm associates (one of) the dyadic probability distribution(s)  "closest"  to  $  P $
 +
in the sense of the pseudo-metric  $  D ( \cdot \mid  \cdot ) $.
 +
 
 +
In several practical situations the statistics of the source is not perfectly known or varies over time, and the necessity arises of updating the Huffman code when the new(ly known) $  P $
 +
becomes  "closer"  to a dyadic probability distribution different from the previous one (see [[#References|[a3]]], [[#References|[a4]]]).
  
 
See also [[Coding and decoding|Coding and decoding]]; [[Information theory|Information theory]].
 
See also [[Coding and decoding|Coding and decoding]]; [[Information theory|Information theory]].

Latest revision as of 22:11, 5 June 2020


An algorithm for the construction of optimal (i.e., minimum redundancy) variable-length $ D $- ary codes ( $ D \geq 2 $) for finite memoryless information sources with known statistics was proposed by D. Huffman in 1952 (see [a1], [a2]). (The algorithm refines the Morse alphabet, as it associates the shortest codewords to the most probable source letters.)

Let $ A _ {K} $ be such a source with $ K $ letters, $ K \geq 2 $, and let $ p _ {i} $ be the probability that the source emits letter $ a _ {i} $( $ 1 \leq i \leq K $). Let $ {\mathcal C} = \{ w _ {1} \dots w _ {K} \} $ be a uniquely decipherable $ D $- ary code for $ A _ {K} $( i.e., a set of $ K $ sequences $ w _ {i} $ from an alphabet of $ D $ symbols) and let $ l _ {i} $ be the length (or number of $ D $- ary symbols) of codeword $ w _ {i} $( $ 1 \leq i \leq K $) associated to the letter $ a _ {i} $. The difference

$$ R ( {\mathcal C} ) = l ^ {*} ( {\mathcal C} ) - H ( P ) $$

between the average length $ l ^ {*} ( {\mathcal C} ) = \sum _ {i = 1 } ^ {K} p _ {i} l _ {i} $ of $ {\mathcal C} $ and the source entropy (cf. also Entropy) $ H ( P ) = - \sum _ {i = 1 } ^ {K} p _ {i} { \mathop{\rm log} } _ {D} p _ {i} $ is called the redundancy of $ {\mathcal C} $. A Huffman code $ {\mathcal H} $ for $ A _ {K} $ minimizes $ R ( {\mathcal C} ) $( hence $ l ^ {*} ( {\mathcal C} ) $) over the set of all uniquely decipherable codes for $ A _ {K} $.

In the binary case ( $ D = 2 $), the algorithm consists in the construction of a sequence of "reduced" sources $ \{ A _ {K - 1 } \dots A _ {2} \} $, where $ A _ {j - 1 } $ is obtained from $ A _ {j} $ by merging the two least probable letters of $ A _ {j} $ into one letter of $ A _ {j - 1 } $( $ K \geq j \geq 3 $) and leaving all other letters unchanged (ties are broken arbitrarily). The probability of the new letter is the sum of the probabilities of the merged letters. The $ K - 2 $ mergings can be represented by a binary code tree in which each letter of $ A _ {K} $ corresponds to a leaf and each codeword is the sequence of binary labels of the path from root to leaf. (If the code alphabet has $ D > 2 $ letters, each reduced source is obtained by merging the $ D $ least probable letters of the previous source, except possibly in the first step, where $ 2 + R _ {D - 1 } ( K - 2 ) $ letters have to be grouped together, $ R _ {a} ( b ) $ being the remainder of the division of $ b $ by $ a $. The resulting $ D $- ary code tree is labeled and again the codewords are the sequences of labels from root to leaves.)

Owing to the arbitrary resolution of possible ties in grouping the least probable sequences at each step, for a given source $ A _ {K} $ there may be several distinct Huffman codes, but two such codes are considered (essentially) different only if the corresponding sets $ \{ l _ {1} \dots l _ {K} \} $ of codeword lengths do not coincide. However, the average length of all Huffman codes for $ A _ {K} $ is the same.

More formally, consider the set $ {\mathcal P} _ {K} $ of all $ K $- order probability distributions (cf. Probability distribution) $ P = \{ p _ {1} \dots p _ {K} \} $, $ \sum _ {i = 1 } ^ {K} p _ {i} = 1 $, $ p _ {i} > 0 $( $ 1 \leq i \leq K $). The informational divergence (also called the Kullback–Leibler information) between two probability distributions $ P $, $ Q $ in $ {\mathcal P} _ {K} $ is the non-negative quantity $ D ( P \mid Q ) = \sum _ {i = 1 } ^ {K} p _ {i} { \mathop{\rm log} } ( { {p _ {i} } / {q _ {i} } } ) $. $ D ( P \mid Q ) $ vanishes if and only if $ P \equiv Q $, hence it can be considered as an oriented pseudo-distance (cf. also Pseudo-metric) in $ {\mathcal P} _ {K} $. Now, if $ \{ l _ {1} \dots l _ {K} \} $ is the set of codeword lengths of a Huffman code $ {\mathcal H} $ for $ A _ {K} $, the redundancy $ R ( {\mathcal H} ) $ is equal to the informational divergence $ D ( P \mid H ) $ between the source probability distribution $ P $ and the "dyadic" probability distribution $ H = \{ 2 ^ {- l _ {1} } \dots 2 ^ {- l _ {K} } \} $.

The optimality of Huffman codes can now be restated by saying that to any $ P \in {\mathcal P} _ {K} $( i.e., to any source $ A _ {K} $), the algorithm associates (one of) the dyadic probability distribution(s) "closest" to $ P $ in the sense of the pseudo-metric $ D ( \cdot \mid \cdot ) $.

In several practical situations the statistics of the source is not perfectly known or varies over time, and the necessity arises of updating the Huffman code when the new(ly known) $ P $ becomes "closer" to a dyadic probability distribution different from the previous one (see [a3], [a4]).

See also Coding and decoding; Information theory.

References

[a1] D.A. Huffman, "A method for the construction of minimum redundancy codes" Proc. I.R.E. , 40 (1952) pp. 1098–1101
[a2] R. Gallager, "Variations on a theme by Huffman" IEEE Trans. Inform. Theory , IT–24 (1978) pp. 668–674
[a3] G. Longo, G. Galasso, "An application of informational divergence to Huffman codes" IEEE Trans. Inform. Theory , IT–28 (1982) pp. 36–43
[a4] D.A. Lelewer, D.S. Hirschberg, "Data compression" ACM Comput. Surv. , 19 (1987) pp. 261–296
How to Cite This Entry:
Huffman code. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Huffman_code&oldid=18372
This article was adapted from an original article by G. Longo (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article