Performance of two-stage massive correlator architecture for fast acquisition of GPS signals

A typical operation of GPS receivers assumes a search of the satellites visible on the sky by synchronizing locally generated replica with the transmitted pseudo-random noise (PRN) code sequence. This synchronization is initially performed by finding the highest correlation between the incoming signal and replica, a process known as “acquisition”. Highest correlation is observed as a correlator peak response which is compared with a certain threshold to identify availability and values of unknown code phase and frequency of a residual carrier modulation. If the peak is not detected or the decision is wrong then acquisition stage should be repeated many times which is quite a time consuming task. State-of-the-art advanced receivers use massive correlators which parallelize the acquisition process. While massive correlators improve significantly the sensitivity of the receivers, so-called multiple peak selection approach provides an opportunity to save computations by sharing tasks between the massive correlator and validation system [1]. In this paper we study a performance of two stage correlator consisting of massive and supplementary implementations. The massive correlator is not making firm decisions each time it finds a peak, but provides several possible options (a limited set of highest peaks) to supplementary system.