This paper is concerned with closed-loop stability analysis and dynamic quantization parameters design of discrete-time networked control systems (NCSs) with noise disturbance. First, based on the Lyapunov function, two state invariant regions of estimation error system and plant system are constructed. The quadratically attractive conditions of the two systems are obtained by using the proposed invariant region sequence, respectively. Second, a quantized control strategy is presented by dynamically adjusting the scaling parameters of logarithmic quantizers. Moreover, the connection of quadratic stability between the plant system and the estimation error system is also given.