NeTS: Small: Design of an Efficient Backscatter Communication Stack for Sensor Devices

NeTS：小型：传感器设备的高效反向散射通信堆栈设计

• 批准号: 1217606
• 负责人: Deepak Ganesan
• 金额: $ 45万
• 依托单位: University of Massachusetts Amherst
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-07-01 至 2016-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1217606&HistoricalAwards=false
• 关键词: NeTS; Small; Design; Efficient; Backscatter

This proposal seeks to perform fundamental cross-disciplinary cross-layered research into the design of a network stack for sensor devices that use backscatter communication. Backscatter communication is particularly attractive for single-hop sensor network deployments since communication is powered by the reader, thereby resulting in tremendous reduction in power consumption compared to devices with active radios (e.g. 802.15.4). Despite these benefits, protocols designed for backscatter are ill-designed for sensor workloads. The existing backscatter communication stack, EPC Gen 2, is optimized for supportinglarge numbers of simple passive tags, that each need to transfer a small amount of data (their unique identifier). While such a model may be appropriate when RFIDs are deployed in retail stores, sensors require a different model where there are a few or moderate number of tags which need to transfer a significant amount of sensed data during contact events with a reader.This proposes outlines a fundamentally new backscatter communication stack that is designed for next-generation sensors that use passive rather than active radios. The stack will be build on an extensive measurement-driven study of backscatter communication including an understanding of channel quality metrics, link asymmetry, spatial and temporal propagation, and harvesting characteristics. Based on this study, we will design a network stack for backscatter communication that replaces the EPC Gen 2 stack with novel methods for mobility detection, rate adaptation, channel selection, bulk transfer, and energy management.Our research will directly impact the design, development, and deployment of next-generation sensor networks that can operate for longer durations or use smaller power sources. This can directly translate into lower total operational costs by lowering the overhead of replacing batteries. The research and teaching that will flow from the project will impact students across the Five Colleges, of which UMass Amherst is a part. The Five Colleges includes two all-women's colleges, Smith and Mount Holyoke, whose students can enroll in courses taught at UMass because of the consortium's cross-registration policy.

该提案旨在对使用反向散射通信的传感器设备的网络堆栈设计进行基础的跨学科跨层次研究。反向散射通信对于单跳传感器网络部署特别有吸引力，因为通信由阅读器供电，从而与具有有源无线电（例如 802.15.4）的设备相比，功耗大大降低。尽管有这些好处，但为反向散射设计的协议对于传感器工作负载来说设计不当。现有的反向散射通信堆栈 EPC Gen 2 经过优化，可支持大量简单的无源标签，每个标签都需要传输少量数据（其唯一标识符）。虽然当 RFID 部署在零售商店时，这种模型可能是合适的，但传感器需要不同的模型，其中有少量或中等数量的标签，需要在与阅读器的接触事件期间传输大量感测数据。本文提出了概述一种全新的反向散射通信堆栈，专为使用无源而非有源无线电的下一代传感器而设计。该堆栈将建立在对反向散射通信的广泛测量驱动研究的基础上，包括对信道质量指标、链路不对称性、空间和时间传播以及收集特性的理解。基于这项研究，我们将设计一个用于反向散射通信的网络堆栈，用移动性检测、速率自适应、信道选择、批量传输和能量管理的新方法取代 EPC Gen 2 堆栈。我们的研究将直接影响设计、开发，以及部署可以运行更长时间或使用更小的电源的下一代传感器网络。通过降低更换电池的开销，这可以直接转化为更低的总运营成本。该项目的研究和教学将影响五个学院的学生，麻省大学阿默斯特分校是其中的一部分。五所学院包括两所女子学院：史密斯学院和曼荷莲学院，由于该联盟的交叉注册政策，这些学院的学生可以报名参加麻省大学教授的课程。