In this paper, a new framework for sequential Bayesian estimation in a sensor network by using both the received data and the information conveyed by missing data due to per-sensor censoring is proposed. In this framework, each local sensor maintains a Kalman Filter (KF) and the Fusion Center (FC) runs a particle filter (PF) to track the system state. Informative measurements are selected by the per-sensor censoring process executed at the sensors at each time. Though the less informative measurements are not sent to the FC, their absence still conveys some information, and the proposed scheme exploits such information from the missing message to achieve better inference performance. Numerical examples are provided to support the theoretical results.