Compressed Sensing: An approach to real time DTMF signaling system

Abstract

The central aim of this paper is to provide a Compressed Sensing (CS) approach to implement a multifrequency signaling system so as to address its specific real time constraints and challenges. Here, the implementation is illustrated by using real time Dual Tone Multi-Frequency (DTMF) signaling system. DTMF is a signaling scheme that is developed by utilizing voice frequency tones. In this paper, an attempt is made to apply the theory of CS to a real time DTMF signaling system. Here the CS methodology is applied to the source model that generates DTMF signals. To work with practical systems, modifications are made to the CS framework by adding noise to the measurements. This paper proposes a simple and effective sparse signal recovery algorithm that uses linear algebra theory to convert an underdetermined system to an overdetermined system. The methodology extends to a variety of other situations and higher dimensions.