An adaptive modulation scheme for image transmission over wireless channels

In this paper, we propose an adaptive modulation scheme for image transmission over fading channels. The proposed scheme partitions the received signal-to-noise (SNR) space as well as the transmitted information based on its importance. Then, it jointly considers the channel condition and the importance of the transmitted bitstream when deciding on the modulation level. More specifically, the scheme partitions the compressed image information into different bit streams with unequal importance based on the properties of the human visual system. Then, it decides on the modulation level that meets a predefined target bit error rate (BER) using an offline lookup table. This predefined BER is based on the SNR level as seen by the receiver and the sensitivity of the transmitted bit stream. Simulation results show that the proposed scheme provides significant improvement in the spectral efficiency when compared to fixed-modulation based systems while maintaining relatively high peak signal-to-noise ratio (PSNR) and acceptable perceptual quality.