We study a linear discrete-time partially observed system perturbed by a white noise. The observations are transmitted to the controller via communication channels with irregular transmission times. Various measurement signals may incur independent delays, arrive at the estimator out of order, and be lost or corrupted. The estimator is given a dynamic control over the sensors and administrates producing the messages to be sent from sensors. The minimum variance state estimate and the optimal sensor control strategy are obtained. Ideas of model predictive control are applied to derive a nonoptimal but implementable in real time method for sensor control.