Cyclostationary joint phase and timing estimation for staggered modulations

The paper presents a cyclostationary approach to the non-data-aided (NDA) phase and timing estimation problem for staggered modulations. The problem is addressed under the low-SNR unconditional maximum likelihood (UML) framework, and modulation schemes, such as offset quadrature amplitude modulation (OQAM) and minimum shift keying (MSK), are considered. It is found that not only the timing parameter but also the phase, can be jointly obtained from the asymptotic UML cost function based on spectral line generation after a second-order non-linearity.