A fast automatic modulation recognition algorithm and its implementation in a spectrum monitoring application

Abstract

An algorithm based on an efficient decision-tree approach for performing real-time automatic modulation recognition is presented. The algorithm is computationally efficient and can discriminate between various digital and analog signal modulation, delivering reliable decisions using very short observation times. Communications formats, such as AMPS, GSM and US digital cellular, can also be recognized. The implementation of this algorithm in a practical spectrum monitoring system is discussed. Measurements using off-the-air signals, and computer simulations, indicate that the system performs well in different surveillance scenarios.