Blind Symbol Rate Estimation and Testbed Implementation for Linearly Modulated Signals

Abstract

Blind symbol or baud rate estimation is one of the most important parts of the blind signal detection process. Most of the blind symbol rate estimation algorithms provided in the literature either estimate symbol rate from the baseband signal or require prior knowledge of transmitted signal parameters. Moreover, these algorithms had been mostly limited to theoretical studies. In this paper, a cyclic correlation based blind symbol rate estimation is considered for linearly modulated signals. The estimation is performed directly from an intermediate frequency (IF) signal and without having prior knowledge of transmitted signal parameters. The algorithm is implemented on a National Instruments (NI) hardware to measure its performance in a realistic environment. Theoretical and experimental results are provided and compared with existing methods.