Downlink Synchronization Techniques for Heterogeneous Cellular Networks

Base stations (BS) in a cellular network send unique synchronization signals based on their physical layer cell-identity, in every frame. To synchronize with a BS, a mobile terminal (MT) needs to identify the following synchronization parameters-starting time of the frame, relative frequency offset and the cell-identity of the corresponding BS. In a heterogeneous cellular network, due to the increased density of low-powered base stations, the MT receives signals from multiple BS in its vicinity. We consider the downlink synchronization problem where the MT tries to recover the synchronization parameters of all the BS in its neighborhood, using which, MT can subsequently choose the suitable BS for its connection. In any given scenario, the number of BS in the neighborhood of MT is relatively small compared to the total number of BS in the entire network. Exploiting this sparseness, we present a two-stage synchronization approach where the first stage identifies the frame timing using an approximation to the maximum likelihood detector and the second stage recovers the cell identities and the corresponding frequency offsets of all the neighborhood BS using block sparse signal recovery framework of compressive sensing. We analytically and numerically study the recovery guarantees of our proposed techniques and establish their superior performance over existing successive interference cancellation and matched filtering approaches.