Effective adaptive rate allocation for DP-LDPC image transmission system based on source and channel characteristics

Abstract

Image transmission issue with high entropy frames is a great challenge for joint source channel coding (JSCC) system based on double protograph low-density parity-check codes (DP-LDPC) due to its sensitivity to source statistics. Aiming at this question, we proposed an effective rate allocation scheme with the premise of unfixed total rate to allocate the source coding rate and channel coding rate for each frame based on source and channel characteristics. In this scheme, two kinds of objective utility function (OUF) models and corresponding rate allocation methods are provided, including the single objective utility function (SOUF) model with global search method and the multi-objective utility function (MOUF) model with fuzzy logic control (FLC) method. Simulation results indicate that both models are effective for DP-LDPC system with high entropy frames, and the MOUF model with FLC method has better adaptive capacity to time-varying channel and better balance on transmission efficiency, image quality and fairness of allocation. The proposed methods offer some new solutions for high entropy frame problem of DP-LDPC image transmission system.