Robust Image Transmission over Noisy Channel Using Independent Component Analysis

Independent Component Analysis (ICA) is the decomposition technique of a random vector of data into linear components which are “independent as possible.” Involves finding a suitable representation of multivariate data for computational and conceptual simplicity, the representation is often sought as a linear transformation of the original data. The linear transformation methods include Principal Component Analysis (PCA), Factor Analysis, and Projection Pursuit. Here attempt to transmit similar dimension multiple images as a single linear transformed image using Independent Component Analysis (ICA), Gaussian noise is added into linearly transformed image. We try to retrieve the original images one by one from noisy transformed image. The analysis is made by varying noise variance against peak signal to noise ratio (PSNR) with the original image. Our demonstrated work is highly useful in reducing bandwidth over the channel.