Robust satellite image transmission over bandwidth-constrained wireless channels

With ability to eliminate the cliff effect and achieve graceful degradation, analog-like transmission such as SoftCast has become a hot research issue for robust image/video delivery over wireless channels. However, it has not been well studied in the case of bandwidth compression, which usually occurs in bandwidth-constrained wireless environments such as satellite communication scenarios. In this paper, we propose an analoglike robust transmission scheme based on Compressive Sensing (CS) for satellite image delivery over bandwidth-constrained wireless channels. The motivation for integration of CS in the proposed scheme lies in the fact that the simple dropping strategy is not optimal for satellite image transmission when bandwidth is insufficient. Considering high information entropy and rich structure information properties of satellite images, we perform an amplitude offset operation and the block-based CS (BCS) procedure to improve the energy efficiency and meet the bandwidth budget respectively. We analyze the system distortion of the proposed scheme and formulate the distortion minimization problem as a resource allocation problem. Then, we propose an efficient two-step strategy to find the optimal solution for bandwidth and power allocation. The simulation results show that the proposed scheme achieves up to 5.5dB gain over the state-of-the-art transmission schemes for satellite image delivery.