Arithmetic coding

MATLAB implementation of a binary encoding app that makes use of Arithmetic coding. Includes a MATLAB GUI.

MATLAB functions

arit_encoder.m The main script of the program. Parses the alphabet as individual characters from a given char array and again defines the given probability distribution as a symbol-probability Map. Removes all spaces within the message and calls the encoding function. Returns the encoded message and the plaintext message (devoid of spaces)
arithmetic_encode_recursive.m Recursively implements the arithmetic coding algorithm. Takes as parameters the extremes of and interval (a, b), a probability distribution of the input alphabet (p), and a message to encode (msg)
float_mant_to_binary.m Given a float, computes the binary expansion of its fractional part. The conversion is done by taking the fractional part, multiplying it by two and adding its whole part (either a 0 or a 1) to the output expansion. Since the binary expansion might be infinite, the function also takes a max_iter parameter that truncates the expansion after a certain number of iterations, if the conversion has not yet reached an end
val_between_bin_range.m Given two (potentially infinite, but truncated) binary expansions a and b (such that a<b), returns a binary value contained within the interval [a,b)
encoder_gui_exported.m The GUI of the program. Checks for errors in the input values and displays the result of the conversion. Automatically normalizes the input probability vector