In this paper, the filtering issues are investigated for event-based Wireless Network Control Systems (WNCSs) with User Datagram Protocol (UDP). Because the optimal estimation of WCNSs with packet loss in both control and observation channels via UDP cannot be applied in practice due to the exponential increase of the computation, and in order to reduce the energy consumption of wireless communication nodes, an efficient recursive filter is developed for the event-based WNCSs with UDP. The state estimation and the filtering error covariance are derived from the interacting multiple model (IMM) methods, and the filter gain is determined by minimising an upper bound of the filtering error covariance. It is proved that the stability of the minimum upper bound of the filtering error covariance is only related to the actual arrival rate of the observation packet, and the minimum upper bound converges to the error covariance of the optimal estimation of the corresponding system with Transmission Control Protocol (TCP) if both the trigger threshold and the control input tend to zero. Numerical examples are given to verify the effectiveness and feasibility of the theoretical results.