Joint optimization of communication and controller components of wireless networked control systems

• Published in: 2015 IEEE International Conference on Communications (ICC) pp. 6487 - 6493
• Main Authors: Sadi, Yalcin, Coleri Ergen, Sinem
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2015
• Subjects: Control systems; Delays; energy minimization; Modulation; networked control system; Optimization; Power demand; stability; Wireless communication
• ISSN: 1550-3607; 1938-1883
• DOI: 10.1109/ICC.2015.7249358

Designing communication system for wireless networked control systems requires overcoming the additional challenge of maintaining a guaranteed performance for the control system in the presence of wireless network induced imperfections including packet error, delay, sampling and quantization errors compared to traditional wireless sensor networks. The joint optimization of controller and communication systems encompassing efficient abstractions of each system and taking into account all wireless induced imperfections, the parameters of the wireless communication system including the transmission power, rate and scheduling and the parameters of the control system including the sampling period has been studied for the objective of minimizing the average power consumption of the network and the MQAM modulation scheme. In this paper, we extend the joint optimization problem for a generalized power cost function that represents many power-related objectives including minimization of total power consumption of the network and minimization of maximum power consumption among the nodes in the network and for any modulation scheme that satisfies certain properties including MQAM and MFSK. The optimization problem is formulated as a Mixed-Integer Programming problem thus difficult to solve for the global optimum. However, upon determining the optimality conditions for the optimization variables, the problem reduces to an Integer Programming problem for which we propose an optimal fast enumeration algorithm. Simulations demonstrate that the proposed optimal solution method outperforms the traditional separate design of control and communication systems.