Joint Source Channel Decoding exploiting 2D source correlation with parameter estimation for image transmission over Rayleigh fading channels

Abstract

This paper investigates the performance of a 2- Dimensional (2D) Joint Source Channel Coding (JSCC) system assisted with parameter estimation for 2D image transmission over an Additive White Gaussian Noise (AWGN) channel and a Rayleigh fading channel. Baum-Welsh Algorithm (BWA) is employed in the proposed 2D JSCC system to estimate the source correlation statistics during channel decoding. The source correlation is then exploited during channel decoding using a Modified Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm. The performance of the 2D JSCC system with the BWA-based parameter estimation technique (2D-JSCC-PET1) is evaluated via image transmission simulations. Two images, each exhibits strong and weak source correlation are considered in the evaluation by measuring the Peak Signal Noise Ratio of the decoded images at the receiver. The proposed 2D-JSCC-PET1 system is compared with various benchmark systems. Simulation results reveal that the 2D-JSCC-PET1 system outperforms the other benchmark systems (performance gain of 4.23 dB over the 2D-JSCC-PET2 system and 6.10 dB over the 2D JSCC system). The proposed system also can perform very close to the ideal 2D JSCC system relying on the assumption of perfect source correlation knowledge at the receiver that shown only 0.88 dB difference in performance gain.