I was reading these slides about Source Coding and I have the following questions:
1) Let's consider this part:
Can you explain me step by step the reasons of a, b, c? Which is the general rule to understand if a certain code is Huffman or not?
2) Let's consider the general algorithm for Huffman Coding.
My question is: when we choose the symbols with lowest probabilities and we group them, is their assignment to 0 and 1 arbitrary? Or should we follow a rule (for instance: 0 for the most probable symbol etc)?
3) Let's consider this part about Shannon - Fano Coding:
My question is: once I have got the big table in the picture, should I read the bits (to get codes for a1, a2, etc) only from left to right (as it does) or can I get also a correct code by reading them from right to left?