Simultaneous Determination of Modulation Types and Signal-to-Noise Ratios Using Feature-Based Approach

This paper presents a low-complexity technique for simultaneous determination of modulation types and signal-to-noise ratios (SNRs) in wireless communication systems. The proposed approach exploits the extracted features of patterns observed in signals’ asynchronous amplitudes histograms, for the simultaneous determination of these quantities using support vector machine. Features extraction has been performed by a well-known technique called principal component analysis which is used to extract the most significant features before being supplied to the artificial intelligent system. Simulations for three commonly-used modulation types have been conducted under real-world channel conditions. The results conclude that the presented approach can accurately identify the modulation types with 99.83% accuracy despite the existence of real-world channel impairments. Furthermore, the algorithm is capable of SNRs estimation over a broad range of 0–30 dB with average estimation error of 0.79 dB. The proposed paper exploits the simplicity of generating asynchronous amplitudes histograms to enable cost-effective and reduced-complexity implementation in cognitive wireless systems.