Efficient image transmission over MIMO channel using compressive sensing

In wireless communication, the channel is characterized by multipath propagation due to scattering by different obstacles. Multiple Input Multiple Output (MIMO) channels which consider this multipath propagation offers significant increase in data throughput and link range without additional bandwidth or transmit power. With the availability of high data rates by MIMO channel, images can be transmitted with high reliability. In this paper, both source coding and channel coding techniques of an image are employed. An image is source coded by using Compressive Sensing (CS) algorithm where the input image is converted to numerical data. These numerical data´s are then space time block coded which can be transmitted simultaneously over the four antennas. Then unequal power allocation scheme is employed over those data streams in which each transmitting antenna gets different amount of power according to the data it transmits. The results are simulated using MATLAB. The results show that the unequal power allocation provides significant quality improvement than the conventional equal power allocation technique. The results also reveal that spatial diversity can be achieved at a higher rate using Compressive Sensing than with JPEG and JPEG 2000 compression algorithms.