This paper investigates the problem of synchronization in complex dynamical networks. Based on the stability theory for impulsive differential equations, an impulsive control scheme is proposed to achieve impulsive synchronization for complex dynamical networks with unknown coupling. The synchronization strategy considers the influence of all nodes in the dynamical network and the effect intensity of every node to network synchronization relies on its weight in the network. For practical problems, by choosing appropriately the weights of the nodes in the network, network synchronization can be achieved by only a few useful nodes. Simulated examples are provided by using the chaotic Chua system as nodes of the dynamical network, and the effectiveness of the proposed impulsive control are demonstrated.