This paper analyzes the exponential stability of a discrete-time linear plant in feedback control over a communication network with N sensor nodes, dynamic quantization, large communication delays, variable sampling intervals, and round-robin scheduling. The closed-loop system is modeled as a switched system with multiple-ordered time-varying delays and bounded disturbances. We propose a time-triggered zooming algorithm implemented at the sensors that preserves exponential stability of the closed-loop system. A direct Lyapunov approach is presented for initialization of the zoom variable. The proposed framework can be applied to the plants with polytopic type uncertainties. The effectiveness of the method is illustrated on cart-pendulum and quadruple-tank processes.