SUMMARYSystem synchronization and channel estimation are two key problems of IEEE 802.16e Orthogonal Frequency Division Multiplex Access (OFDMA) downlink system. According to the standard, system synchronization is completed through the cell search procedure, which includes two steps: frame detection and cell-ID identification. First, based on the special time-domain (TD) and frequency-domain (FD) structures of the preamble, we propose two novel frame detection ms: the TD algorithm and the FD algorithm. Their robustness performance is nearly the same, and the complexity of the FD algorithm is only 1 31 of that of the TD algorithm. Then, an efficient cell-ID identification algorithm is proposed, in which the necessary correlation length can be dramatically shortened even without any a priori channel knowledge. Thereby, it greatly reduces the cell search time. As for the problem of channel estimation, a cluster-based channel estimation method is proposed. It is based on the special pilot distribution of IEEE 802.16e and utilizes the continuity of channel variations to mitigate the impact of time-variant channel. System simulation results show that the proposed algorithms are both robust and efficient.