Delay-Reliability Model of Industrial WSN for Networked Control Systems

Abstract

In a so-called `Smart Factory', sensors, actuators, and a processing logic are interconnected via wireless communication. A popular class of industrial processes is Networked Control Systems (NCS), where the sensor, controller, and actuator of a control system are distributed over a network. Wireless brings several benefits to NCS but affects their performance. This aspect is particularly critical, as NCS pose stringent delay and reliability requirements to data packets in order to fulfil a desired Quality of Control (QoC). Industrial Wireless Sensor Networks (IWSN) is a candidate communication technology to haul NCS traffic. IWSN, however, suffer from packet loss caused by the harsh industrial environment. The characterization of the impact of delay and packet loss on the QoC of NCS is a challenging task, as it requires the analysis of mutually dependent random processes. We tackle this investigation deriving a delay-reliability model for IWSN based on the Loop Success Probability, a metric that associates the network performance to the QoC of the NCS. Initially, the effect of Loop Success Probability on QoC is evaluated, then, it is mathematically related to the end-to-end delays of IWSN packets. The model provides a connection between IWSN parameters and QoC and is used to define their operating regions. Via measurements of an IWSN testbed and a simulated NCS, we prove the validity of the proposed model.