Spectrum sensing is an important issue in cognitive radio (CR) networks. In a periodic sensing frame structure, the spectrum sensing performance and the protection to the primary network improves while the sensing time increases. However, the longer the sensing time, the lower the throughput the CR network can achieve. In this paper, we formulate an optimization problem to balance the sensing-throughput tradeoff for a CR network with wideband spectrum sensing, where the combined aggregate throughput for all the two possible scenarios (instead of only one scenario used in previous related works) of the CR network is maximized under a given interference imposed on the primary network. To reduce the computational complexity, we further present a two-stage iterative algorithm to solve the optimization problem. Simulation results show that the achievable aggregate throughput of the proposed approach is much higher than those of the previous related works.