Multi-Pass Decoding for the Robust Transmission of Deep-Space Images

Linear index coding is generally more robust against channel variations as compared to the fixed-to-variable length coding. This paper proposes a novel multi-pass decoding approach to decode linear index coded images. In contrast to the typical one-pass decoding, the proposed scheme harnesses the information recovered in the first decoding pass with the source statistics and utilize it in the subsequent decoding passes. It is demonstrated that the linear index coded image transmitted from deep-space shows an improvement of up to 3.2 dB in terms of reconstructed peak signal to noise ratio by using multi-pass decoding.