A New Source Encoding Algorithm Based on Shannon Encoding

Source encoding algorithm has been widely concerned by the researchers, which aims to improve communication efficiency. In this paper, we put forward a new source encoding algorithm to decrease the redundancies between compiled code words in Shannon encoding. With the proposed algorithm, a code word is selected by considering whether the code is the prefix code of the other code words or not. The performance of the new encoding algorithm is investigated for both equal probability of source symbols and non-equal probability condition. Numerical results show that the proposed encoding algorithm can improve the encoding efficiency obviously, especially for the case of great probability differences between source symbols.