TMO-NanoQ+: A Real-Time Kernel for Sensor Networks Supporting Time-Triggered and Message-Triggered Tasks

Abstract

As the range of sensor network's applicability is getting extended to applications related to military operations or detecting radioactivity, characterization of timeliness-guaranteed sensor network operations is required at design stage. In this paper, a new real-time sensor network modeling scheme by using the well-known TMO model is proposed. And for this modeling scheme, a new realtime sensor network kernel; named TMO-NanoQ+; that supports executions of time-triggered and message-triggered real-time tasks, has been developed based on the ETRI 's (Electronics and Telecommunications Research Institute, Korea) NanoQ+ kernel. Development of the TMO-NanoQ+ has been done by modifying the existing timer interrupt module to support a FIFO-based power-aware scheduling and an EDF/LLF scheduling. And for message-triggered tasks, a distributed IPC scheme that supports logical multicast channels over RF communications has been also developed