Accounting for Failures in Delay Analysis for WirelessHART Networks

WirelessHART networks are gaining ground as a real-time communication infrastructure in industrial wireless control systems. Because wireless communication is often susceptible to transmission failures in industrial environments, it is essential to account for failures in the delay analysis for realtime flows between sensors and actuators in process control. WirelessHART networks handle transmission failures through retransmissions using dedicated and shared time slots through different paths in the routing graphs. While these mechanisms for handling transmission failures are critical for process control requiring reliable communication, they introduce substantial challenges to worst-case end-to-end delay analysis for real-time flows. This paper presents the first worst-case end-to-end delay analysis for periodic real-time flows in a WirelessHART network that takes into account transmission failures. The delay bounds can be used to quickly assess the schedulability of real-time flows for industrial wireless control applications with stringent requirements on both high reliability and network latency. Simulations based on the topologies of a wireless sensor network testbed consisting of 69 TelosB motes indicate that our analysis provides safe upper bounds of the end-to-end delays of real-time flows at an acceptable level of pessimism. Type of Report: Other Department of Computer Science & Engineering Washington University in St. Louis Campus Box 1045 St. Louis, MO 63130 ph: (314) 935-6160 Accounting for Failures in Delay Analysis for WirelessHART Networks Abusayeed Saifullah, Paras Babu Tiwari, Bo Li, Chenyang Lu, Yixin Chen Department of Computer Science and Engineering Washington University in St. Louis {saifullah, pbtiwari, b.li, lu, ychen25}@wustl.edu Abstract—WirelessHART networks are gaining ground as a real-time communication infrastructure in industrial wireless control systems. Because wireless communication is often susceptible to transmission failures in industrial environments, it is essential to account for failures in the delay analysis for realtime flows between sensors and actuators in process control. WirelessHART networks handle transmission failures through retransmissions using dedicated and shared time slots through different paths in the routing graphs. While these mechanisms for handling transmission failures are critical for process control requiring reliable communication, they introduce substantial challenges to worst-case end-to-end delay analysis for real-time flows. This paper presents the first worst-case end-to-end delay analysis for periodic real-time flows in a WirelessHART network that takes into account transmission failures. The delay bounds can be used to quickly assess the schedulability of real-time flows for industrial wireless control applications with stringent requirements on both high reliability and network latency. Simulations based on the topologies of a wireless sensor network testbed consisting of 69 TelosB motes indicate that our analysis provides safe upper bounds of the end-to-end delays of real-time flows at an acceptable level of pessimism.WirelessHART networks are gaining ground as a real-time communication infrastructure in industrial wireless control systems. Because wireless communication is often susceptible to transmission failures in industrial environments, it is essential to account for failures in the delay analysis for realtime flows between sensors and actuators in process control. WirelessHART networks handle transmission failures through retransmissions using dedicated and shared time slots through different paths in the routing graphs. While these mechanisms for handling transmission failures are critical for process control requiring reliable communication, they introduce substantial challenges to worst-case end-to-end delay analysis for real-time flows. This paper presents the first worst-case end-to-end delay analysis for periodic real-time flows in a WirelessHART network that takes into account transmission failures. The delay bounds can be used to quickly assess the schedulability of real-time flows for industrial wireless control applications with stringent requirements on both high reliability and network latency. Simulations based on the topologies of a wireless sensor network testbed consisting of 69 TelosB motes indicate that our analysis provides safe upper bounds of the end-to-end delays of real-time flows at an acceptable level of pessimism.