In order to improve the spatial resolution of image, a framework of a blind single-image super resolution (SR) reconstruction method based on gaussian point spread function (PSF) is proposed. in the lowe-resolution (LR) imaging model, the processes of gaussian blur, down-sampling and noise are considered. through an error-parameter analyzing method, the parameters of gaussian PSF are estimated automatically. utilizing the estimated PSF, high resolution (HR) image is reconstructed through iterative back projection (IBP) algorithm. experiment is performed on a simulated LR image, and the result justify the effectiveness of the proposed algorithm. the parameters of gaussina PSF of the imaging system can be estimated with high accuracy. the importance of the PSF estimation in SR reconstruction is also demonstrated in an experimental way. when the parameters of the PSF are estimated accurately, the SR reconstructed image has higher peak signal to noise ratio (PSNR) and better visual effect.