| With developments of sensor techniques, embedded systems, network techniques, wireless communication and distributed computation, wireless sensor networks have become an hot spot of research. Compared with traditional computation device, wireless networked sensors have the following characteristics: small physical size and low power consumption, robust operation, concurrency-intensive operation, and so on. The design of system software and tools in sensor systems leads to some research challenges.The traditional operating systems are not suitable for wireless networked sensors directly. Therefore, a new embedded operating system is required to manage and operate these kinds of devices. According to the design requirement and analysis of wireless sensor network systems, the existing embedded operating systems are evaluated and classified. Based on these works, a new wireless sensor network operating system called ARTs-SNOS is designed systematically, including its architecture and the resource management strategies as a focus of this work. The models of memory management, I/O management and interruption management are given. First, a sequence search algorithm is provided and it can work very well in a system with limited memory. Then, the system uses a simple strategy based on static index table to avoid chaos of node equipment management. Furthermore, in order to reduce the time of interrupt mask the system adopts an interruption executive model of front-back segment.The data compression subsystem of ARTs-SNOS is to deal with the compression of sensor data. Based on the study of PMC_mean algorithm, a new real-time sensor data compression algorithm called PMC_RT is proposed. The PMC_RT algorithm is good for the compression of little fluctuation data. For validation of data compression algorithms, a few simulation experiments have been done and the results demonstrate that PMC_RT algorithm is efficient in compressing the sensor data in sensor network systems. |
PDF Full Text Request