Error correction for video streaming to enhance MPEG performance quality

摘要

For real-time transmission, still image and video sources require wide bandwidths and large storage memory space. Therefore, compression techniques are useful and must be applied to overcome the load variation problem during the transmission and storage. To reconstruct images or video from compressed data without any disruption or losses, two important objective image quality metrics, the peak-signal-to-noise ratio (PSNR) as well as the structural similarity index measure (SSIM) will be used. PSNR value increased and approaches infinity as the Mean Square Error (MSE) approaches zero; thus a higher image quality resulted from a higher PSNR value and a lower value of MSE. At the other end of scale, a higher numerical differences between images resulted from a small value of PSNR and a large value of MSE[1]. The SSIM is the other quality metric which used to measure the similarity between two images. The paper simulates a real channel with a protection signal to noise ratio (snr), and take into account the Group of Pictures (GoP) structure. Also, the paper studies the impact of a real noise which generated to reconstruct the modulated video with a lower size to safe in storage and an optimal resolution to reach the suitable quality. The simulation results show that the Quadrature Amplitude Modulation is the best modulation in terms of the performance of video streaming and sending image over wireless channel. To increase the performance, a constrained Reed Solomon Protection technique can be used, by increasing redundancies bits the performance increase but the delay also increased.