Blind Cyclostationarity-Based Symbol Period Estimation for FSK Signals

In this letter, we propose an algorithm for blind estimation of the symbol period of frequency-shift-keying (FSK) signals affected by fading channels. The algorithm exploits the property of the second-order cyclic moment magnitude to exhibit a change in the peak pattern at a delay equal to the symbol period, which we prove analytically. It has the advantage that it requires neither channel estimation nor carrier recovery, while providing a good performance.