An Adaptive Two-Level Quantizer for Networked Control Systems

Abstract

This brief proposes a novel adaptive two-level quantizer, which generates 1-b per sample and can be used for a class of industrial networked control systems (NCSs). The adaptive quantizer is introduced to automatically estimate the input of the quantizer using equivalent control-based sliding mode control. The stability condition of the proposed quantizer is derived analytically with the help of a sliding-mode analysis. The effectiveness of the proposed quantized NCS is illustrated experimentally by designing a quantized feedback controller to control the position of a dc motor, where signals with 1-b per sample are utilized to interconnect the controller with the plant. Simulation results from a laboratory prototype illustrate that the proposed quantizer with the feedback controller controls the system effectively, introduces less quantization noise, and minimizes the data rate.