Intelligent reflecting surface (IRS) can bring significant performance enhancement for wireless communication systems by reconfiguring wireless channels via passive signal reflection. However, such performance improvement generally relies on the knowledge of channel state information (CSI) for IRS-associated links. Prior IRS channel estimation strategies mainly estimate IRS-cascaded channels based on the excessive pilot signals received at the users/base station (BS) with time-varying IRS reflections, which, however, are not compatible with the existing channel training/estimation protocol for cellular networks. To address this issue, we propose in this paper a new channel estimation scheme for IRS-assisted communication systems based on the received signal power measured at the user, which is practically attainable without the need of changing the current protocol. Specifically, due to the lack of signal phase information in power measurements, the autocorrelation matrix of the BS-IRS-user cascaded channel is estimated by solving equivalent matrix-rank-minimization problems. Simulation results are provided to verify the effectiveness of the proposed channel estimation algorithm as well as the IRS passive reflection design based on the estimated channel autocorrelation matrix.