Event-driven operating systems such as TinyOS are the preferred choice for wireless sensor networks. Alternative designs such as MANTIS following a classical multi-threaded approach are also available. Event-based systems are generally more energy efficient than multi-threaded systems. However, multi-threaded systems are more capable than event-based systems of supporting time critical tasks as task preemption is supported. Timeliness can be traded for energy efficiency by choosing the appropriate operating system. In our recent work we have shown that the multi-threaded system MANTIS can be modified to be as energy efficient as TinyOS. As a result, the modified MANTIS can be used to fit both sensor network design goals of energy efficiency and timeliness. This solution is not considered optimal as most existing sensor network applications and software libraries are developed for TinyOS. Therefore, we present a TinyOS modification that adds preemption while retaining the existing TinyOS structure and features.
The capabilities of a sensor network are strongly influenced by the operating system used on the sensor nodes. In general, two different sensor network operating system types are currently considered: event driven and multi-threaded. It is commonly assumed that event driven operating systems are more suited to sensor networks as they use less memory and processing resources. However, if factors other than resource usage are considered important, a multi-threaded system might be preferred. This paper compares the resource needs of multi-threaded and event driven sensor network operating systems. The resources considered are memory usage and power consumption. Additionally, the event handling capabilities of event driven and multi-threaded operating systems are analyzed and compared. The results presented in this paper show that for a number of application areas a thread-based sensor network operating system is feasible and preferable.
Abstract. Event-driven operating systems such as TinyOS are the preferred choice for wireless sensor networks. Alternative designs following a classical multi-threaded approach are also available. A popular implementation of such a multi-threaded sensor network operating system is MANTIS. The event-based TinyOS is more energy efficient than the multi-threaded MANTIS system. However, MANTIS is more capable than TinyOS of supporting time critical tasks as task preemption is supported. Thus, timeliness can be traded for energy efficiency by choosing the appropriate operating system. In this paper we present a MANTIS kernel modification that enables MANTIS to be as power-efficient as TinyOS. Results from an experimental analysis demonstrate that the modified MANTIS can be used to fit both sensor network design goals of energy efficiency and timeliness.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.