Period estimation with linear complexity of sparse time varying point processes

For sparse time-varying point processes we present a period estimator, which is relevant for many applications such as time synchronization of wireless networks in harsh environments. Compared to state-of-the-art methods that are available only for stationary processes, the proposed estimator has a significantly reduced computational complexity while maintaining the same estimation accuracy. The novel approach of estimating the period in time domain, allows to analytically derive the lower bound on the mean square error (MSE), which is presented for the first time. In numerical analysis we show that the proposed algorithm attains the bound.