In cognitive radio networks, primary user emulation (PUE) attack is a denial-of-service (DoS) attack on secondary users. It means that a malicious attacker sends primary-user-like signals to jam certain spectrum channels during the spectrum sensing period. Sensing the attacker's signal, the legitimate secondary user will regard these channels are used by the primary users, and give up using these attacked channels. In this paper, the interaction between the PUE attacker and the secondary user is modeled as a constant sum differential game which is called PUE attak game. The secondary user's objective is to find the optimal sensing strategy so as to maximize its overall channel usability, while the attacker's objective is to minimize the secondary user's overall channel usability. The Nash equilibrium solution of this PUE attack game is deprived, and the optimal anti-PUE attack strategy is obtained. Numerical results demonstrate the trajectories of the secondary user's optimal channel sensing strategies over time, and also shows that: by following the differential game solution, the secondary user can always optimize its channel usability when confronting PUE attacks.