This paper investigates the energy-efficient power loading with intercarrier and intersymbol interference considerations for OFDM-based cognitive systems. The objective is to maximize the energy efficiency (EE) as well as to balance the tradeoff between intercarrier interference (ICI) and intersymbol interference (ISI) by jointly optimizing the subcarrier bandwidth and power allocation in a mobile scenario, under the power budget and the interference constraint. First, the primal problem is converted into a convex optimization problem by fractional programming. Then, the Lagrange dual function and the subgradient method are adopted to achieve the optimal power allocation and the golden section method is employed to search for the optimal subcarrier number (i.e, subcarrier bandwidth). Numerical results show that the proposed algorithm can realize ICI control by choosing an optimal subcarrier number, and simultaneously the EE can be significantly improved by 139%.