huffman encoding and compression

I have a data set that I imported to MATLAB using textscan, and now I want to encode/compress it. The data set is 3660x4, I used this script to upload it:

fileID=fopen('filename');

if fileID < 3
    disp('Error: file couldn''t be opened.')
else
    C = textscan(fileID,'%s %s %s %s')
end

And got this:

C = 

      {3660x1 cell}    {3660x1 cell}    {3660x1 cell}    {3660x1 cell}

What I originally did was take each column, round the data (so that the code would work, if I didnt round it, I would get a repmat error), put it through the code, and get the dictionary and encoded data. i would repeat this with every column and I'd end up with 4 different dictionaries.

The thing is that I need each data point to be as precise as possible, so I can't round it. And when I enter a whole column (not rounder) into the script, I get a repmat error:

>> huff.m
Enter Symbols[C{1}]
Enter Probabilities[p]
??? Maximum recursion limit of 500 reached. Use set(0,'RecursionLimit',N)
to change the limit.  Be aware that exceeding your available stack space can
crash MATLAB and/or your computer.

Error in ==> huffmandict>insertMaxVar

What I want to accomplish is to create a single dictionary (not 4), that I would be able to use on the whole data set.

For example, I'd have numbers 0-9, as well as '.', '/'

0 [code]
1 [code]
2 [code]
3 [code]
4 [code]
5 [code]
6 [code]
7 [code]
8 [code]
9 [code]
. [code]
/ [code]

Ex. Let's say '.' is 000 0 is 001 1 is 010 2 is 100 , and I want the code for 1.102 to be 010000010001100

Anyone know how to do this? The huffman script I use is:

symbols = input('Enter Symbols');
x=symbols;

             p = input('Enter Probabilities');

y=p;         
           [dict,avglen] = huffmandict(symbols,p) % Create the dictionary

             entropy=-sum(p.*log2(p))
             I=entropy;         
             efficiency=avglen/I
             temp = dict;
             for i = 1:length(temp)
             temp{i,2} = num2str(temp{i,2});
             end
             temp;
             sig = repmat(input('Signal'),1,1); % Data to encode
  symbols = [x]; 
  p = [y]; 
  dict = huffmandict(symbols,p); 
  hcode = huffmanenco(sig,dict) 
  dhsig = huffmandeco(hcode,dict);