On the robustness of joint source/channel coding for transmission through an ionospheric channel

In this paper, we present the results relative to our work on still image transmission on a noisy channel. The results are obtained, on one hand, from a digital transmission chain developed in the IRCOM-SIC Laboratory of Poitiers (France), and, on the other hand, from a high frequency transmission system (TRILION) which uses the ionospheric channel as transmission support. The purpose of this work is to find an efficient transmission system, in terms of visual quality for the received image. We already know that usual source coding schemes are inappropriate when errors occur during the transmission. Many error correction systems exist but they significantly decrease the compression ratios for a low error resistance. In order to solve this problem we use a vector quantization (VQ) algorithm based on the Kohonen topological maps. Results were obtained for various configurations of transmission chains from still images. We got a minimal loss, in terms of PSNR, of 0.04 dB compared to the original image thanks to a source coding using the Kohonen VQ. Simulations were carried out on a Gaussian channel by using a digital modulation QAM-256 with a bit error rate (BER)of 3.10/sup -3/.