Diffuse correlation spectroscopy (DCS) is an optical technique that allows for non-invasive measurements of tissue perfusion, and is often used for neuromonitoring applications. However, a major challenge of DCS is low SNR for deep tissue measurements. Recent works have demonstrated the potential for SPAD arrays to provide significant SNR increases by averaging autocorrelation signals from individual speckles. To fully exploit the potential of SPAD array signals, in this work we present alternative methods to recovering the desired blood flow signal and demonstrate their effectiveness in reducing the required sampling time or improving computation speed.