In a communication system, there are different challenges to be faced in transmitting and receiving information. One of the major challenges is to use the bandwidth effectively. A recently developed technique known as compressive sensing provides a solution for using the bandwidth effectively by transmitting the samples of a compressed analog signal. For compressive sampling, the analog has to be converted into sparse, for which an apt transform must be used. Next, the analog must be sampled using a basis function. This paper proposes a novel analog to information converter where a cascaded transform is used to convert the image signal into sparse and where the sparse signal is compressed using a Gaussian random matrix. For signal recovery, an orthogonal matching pursuit is used. The performance of the proposed algorithm was measured both qualitatively and quantitatively, and the results demonstrated that the proposed algorithm is effective with all types of images.

Analog to information convertor using cascaded transform and Gaussian random matrix

Transmission of high-frequency signals require a channel with very high bandwidth. But effective use of bandwidth always remains as an important criterion in communication. So, compression of the message signal is important before transmission. Many transforms are used such as discrete Fourier transform, discrete cosine transform, etc. The objective of this paper is to compress the given image using a cascaded transform and a novel technique known as compressive sensing.

T. Vigneswaran

Papers

A Novel Approach to Compress an Image Using Cascaded Transform and Compressive Sensing