Yibin Li, Shandong University, China
From: Wireless Sensor Systems, Session 6a (Resource management 2) Sensor in scarce resource management, energy harvesting cloud computing, 18 - 19 June 2012, London, UK
19 June 2012 Communications channel
The scheduling of Wireless Sensor Network (WSN) node with energy harvesting is fundamentally different from battery-based node. However, reconfigurable HW has different power consumption modes compared with processor-based implementation. Therefore, adaptive nscheduling strategy is needed for reconfigurable HW under energy-harvesting environment.
In this paper, a novel scheduling is proposed to schedule hardware reconfiguration. For WSN implementation, the proposed strategy aims to achieve higher efficiency with less computing requirement. In the proposed strategy, a task window is built first. Then, inside the task window, the executing time for each task is tallied. Only the most used task in future (MUF) is implemented using reconfigurable HW. An HW-reconfigurable WSN node is presented with related design flow to demonstrate the proposed scheduling strategy. Four test cases are chosen to evaluate our strategy. The execution of test cases is divided into blocks of 256 bit as ation. Three other strategies (All Hardware, All Software, Random) are adopted to implement the evaluation. The results from the experiments show that more than 50% energy cost can be saved if the proposed method is implemented.