Design And Implementation Of Scheduling Scheme For Wireless Sensor Networks Node

With the development of computer, communications and other high-tech materials, wireless sensors and wireless sensor network technology is applied more widely. The demand of its real-time performance, reliability and durability is increasing. Because of the limit (such as size, computing power, and energy) of the wireless sensor networks, its hardware resources is scarcer than the traditional embedded devices. How to make the task scheduled reasonably and efficiently is an important research topic in wireless sensor networks, as well as the focus of discussion in this article.The paper analyses and compares some famous task scheduling algorithm, points out some drawbacks of them, such as no urgency indicators, failure to guarantee the timeliness; unobvious prioritization, failure to fix increasingly complex application environments. For this reason, we propose a new scheduling scheme, which considers the research experience on embedded operating systems and sensor network, combined with the development sensor network applications and diverse, distributed collaborative tasks, and resources limited characteristics. I design and implements the new scheduling scheme based on the analysis on the existing scheduling algorithm. This scheme performs much better than existing scheduling scheme in timeliness and reliability.The innovation of this paper is:1.Priorities for different tasks based on the timeliness and importance, used the concept of micro-processes, the task is divided into three types:micro-processes, threads, and the background.2.Research on different scheduling mechanisms. According to the type of task, a preemptive and non-preemptive combined scheduling method is introduced in this paper. This methods is a good way to enhance flexibility of the system, and enables sensor networks to adapt to a more complex environment.3.Simplifying a series of judgments for real-time scheduling, which is more suitable for the characteristics of wireless sensor networks. This scheme algorithm is implemented on AVRStudio simulation platform and MicaZ node. The results show that the scheme performs better than original scheduling scheme in real-time and reliability when running in wireless sensor networks.