基于远程动态可重构的WSN节点研究与实现

发布时间：2018-05-20 11:57

本文选题：无线传感网 + FPGA　；参考：《计算机研究与发展》2014年01期

【摘要】：作为近年来的一个研究热点,无线传感网(wireless sensor network,WSN)是一项可能改变现有工作与生活方式的技术.一个无线传感网系统可能由几十个到上万个节点组成.每个节点在保证具有足够的信息处理和通信能力的前提下,还需要满足低功耗和低成本的要求.从高效能计算的角度看,以FPGA为代表的可重构硬件已经被证明是提高系统计算效能的重要方式.传统上,在嵌入式系统中,FPGA被认为并不适用于低功耗设计.然而,FPGA的可重构不仅包括静态可重构还包括远程的部分动态可重构(partial dynamic reconfiguration,PDR).通过对芯片的某一特定区域的时分复用,在所需芯片面积减小的情况下,芯片的功耗可以大大减少.除此之外,WSN部署后的维护和功能的更新也可以通过FPGA芯片的远程PDR来实现.描述了基于远程硬件PDR技术的WSN节点.通过对几个典型算法(IIR,GPSR,SHA-2,FFT)的PDR实现,得出节点的功耗、面积与存储消耗,并与软件实现方式进行了比较.实验结果表明,通过采用PDR技术,在计算时间减少的情况下,所需的芯片面积减少27%,运行时功耗最多可减少60%(829mW).
[Abstract]:As a research hotspot in recent years, Wireless sensor Network (WSNs) is a technology that may change the existing work and life style. A wireless sensor network system may consist of dozens to tens of thousands of nodes. Each node needs to meet the requirements of low power consumption and low cost while ensuring sufficient information processing and communication capability. From the point of view of high performance computing, the reconfigurable hardware represented by FPGA has been proved to be an important way to improve the computing efficiency of the system. Traditionally, FPGA is not considered suitable for low power design in embedded systems. However, the reconfiguration of FPGA includes not only static reconfiguration but also remote partial dynamic reconfiguration. By time division multiplexing in a specific area of the chip, the power consumption of the chip can be greatly reduced when the required chip area is reduced. In addition, the maintenance and function update of FPGA after deployment can also be realized by remote PDR of FPGA chip. The WSN node based on remote hardware PDR technology is described. Through the PDR implementation of several typical algorithms, the power consumption, area and storage consumption of the node are obtained, and the results are compared with the software implementation. The experimental results show that by using PDR technology, the area of the chip is reduced by 27% and the power consumption can be reduced by up to 60 ~ 829mW ~ (-1) when the computing time is reduced.
【作者单位】：山东大学计算机科学与技术学院;
【基金】：国家国际科技合作专项(2011DFR20090) 海洋公益性行业科研专项(201205036-04)
【分类号】：TP212.91

【参考文献】