Title :
A Light-weighted Operating System with Deadlock Prevention Strategy for Wireless Sensor Nodes
Author :
Wang, Chunping ; Wu, Wei
Author_Institution :
China Sch. of Comput. Sci., Beihang Univ., Beijing
Abstract :
To overcome the flaws on scheduling reliability in current operating systems for wireless sensor nodes, we presented VROS: a light-weighted and multitasking operating system based on message queue scheduling. In VROS, message is not only used as a means of communication among tasks, but also an access method of software and hardware resources. Each resource access within a task is implemented by binding relevant messages and their data process with the task. With the resource access method, a deadlock prevention strategy is designed which is achieved by attacking the circular wait condition. The experiments on the macro benchmark of surge application show that the deadlock caused by resource competitions were effectively prevented, CPU active time in VROS was nearly identical with TinyOS and SOS, and the data transporting delay was rarely affected.
Keywords :
multiprogramming; scheduling; telecommunication computing; telecommunication network management; wireless sensor networks; SOS; TinyOS; VROS; data transporting delay; deadlock prevention strategy; lightweighted operating system; message queue scheduling; multitasking operating system; scheduling reliability; wireless sensor nodes; Application software; Hardware; Intelligent sensors; Magnetic sensors; Operating systems; Peer to peer computing; Resource management; Sensor systems; System recovery; Wireless sensor networks; Deadlock Prevention; Light-weighted; Operating System; Wireless Sensor Networks;
Conference_Titel :
Communications and Mobile Computing, 2009. CMC '09. WRI International Conference on
Conference_Location :
Yunnan
Print_ISBN :
978-0-7695-3501-2
DOI :
10.1109/CMC.2009.197