Networked control systems (NCSs) have been one of the main research fields in recent years . In modern control systems , information is shared through networks . These networks present many new challenges , including variable time-delay , limited capacity and packet loss . The performance of networked control systems is affected by time-varying transmission delays . These delays may lead to poor performance and even instability of the networked control system . Usual solutions to this problem include storing arriving packets in a buffer at the cost of increased constant delay . Buffer can transform variable delay to fixed delay and regulates packets queue . If the plant is stable and delay is constant , we can use smith predictor to overcome delay problems . Because of limited bandwidth , fixed buffer is not always appropriate . For optimal use of network bandwidth , adaptive buffer is used in this thesis . The size of buffer is based on delay variation . We propose a novel approach to adapt buffer size and its content . When the buffer size needs to be increased , the packets will be predicted and when it needs to be reduced some packets will be dropped . Simulation results show the effectiveness of the proposed approach . Key Words: Networked control systems , Adaptive control , Time delay , Model based predictor
