Low Power Sensing, Communication, and Localization Using Emerging Radio Technologies

该项目的目的是探索新兴无线电技术的适用性，例如新引入的标准，即IEEE 802.15.4z超宽带(UWB)物理层，以及其他技术，如2.4 GHz LoRa无线电技术，用于网络通信协议、安全测距、物体和运动的定位和被动感知。这项研究的目的是利用这些无线电的改进性能，与研究非常充分的IEEE 802.15.4/BLE无线电相比，在恶劣的环境中提供高度可靠的服务。这项研究中使用的一种高使用率技术UWB具有这样的特性，它能够提供高数据速率、准确的时间同步和良好的多径衰落恢复能力，以及能够在利用不太拥挤的频谱的同时进行精细测距和定位。多亏了来自UWB脉冲无线电的容易获得的信道状态信息，被动感知由物体和人类的移动引起的环境变化也是可能的。对于2.4 GHz的LoRa无线电，与BLE和Zigbee/802.15.4相比，性能的改善来自于其远距离传输。除了利用改进的功能外，这项工作还将寻求利用可以在ARM Cortex-M系列微控制器上实施的轻量级机器学习方法。不同的方法和创新的设计将寻求优化通信协议的操作，例如通过了解最佳链路和通信调度，提高对对象、人类及其活动的被动(无设备/非合作)感知的准确性，以及改进技术的测距和定位性能。直到最近，物联网部署通常依赖蓝牙、蓝牙低能耗、IEEE 802.15.4或WiFi无线电进行通信和本地化。在有许多其他设备占用2.4 GHz频率的环境中，这些技术的性能可能会受到很大影响。此外，这些窄带无线电在封闭的室内环境中受到多径衰落的严重影响，这可能导致分组丢失的高概率和定位(米数量级)的显著不准确。UWB是本PHD项目中使用的主要技术之一，由于其固有的对多径衰落的免疫力，是现有窄带技术的潜在良好替代方案。由于脉冲无线电还提供纳秒时间分辨率，因此可以清楚地识别第一个多径分量，从而允许准确的测距和可靠的通信。在原型和实验方面，这项工作将利用商用现成无线电以及物联网应用程序中常用的网络协议的开源实现，例如Zephy项目或Contiki-NG开源项目提供的框架。这项工作将与东芝的单位接触，以开发所生产的技术。该项目属于EPSRC工程研究领域，“信息和通信技术网络和分布式系统”。

The aims of this project is to explore the applicability of emerging radio technologies such as newly introduced standards i.e. IEEE 802.15.4z Ultrawideband (UWB) PHY and other technologies such as 2.4GHz LoRa radio technologies for use with network communications protocols, secure ranging, localisation and passive sensing of objects and movements. The research aims leverage the improved properties of these radios to provide highly reliable services in harsh environments compared to the very well researched IEEE 802.15.4/BLE radios. One of the highly used technologies used in this research UWB has characteristics that allow it to provide high data rates, accurate time synchronisation and good resilience to multi-path fading as well as capable of fine ranging and localisation whilst making use of the a less crowded spectrum. Thanks to the readily available channel state information from the UWB impulse radios, passive sensing of environmental changes caused by movement of objects and humans is also possible. For 2.4GHz LoRa radios, the improved performance can come from its long range transmissions compared to BLE and Zigbee/802.15.4. Apart from leveraging the improved features, the work would look to exploit lightweight machine learning approaches that can be implemented on ARM Cortex-M family of microcontrollers. Different methods and innovative designs would look to optimise the operation of communication protocols for example by learning about best links and communication schedules, enhance accuracy of passive (device free/non-cooperative) sensing of objects, humans and their activities as well as improving the ranging and localisation performance of the technologies. Up until recently, Internet of Things deployments often rely on Bluetooth, Bluetooth Low Energy, IEEE 802.15.4 or WiFi radios for communication and localisation. The performance of these technologies can suffer greatly in environments where there are many other devices occupying the 2.4GHz frequency. Additionally these narrowband radios suffer greatly from multi-path fading in enclosed indoor environments which can lead to high probability of packet loss and significant inaccuracies for localisation (order of metres).UWB being one the main technologies used in this PhD project, is a potentially good alternative to already available narrowband technologies in that it is inherently immune to multipath fading. As an impulse radio is also provides nanosecond time resolution so the first multi-path component can be clearly identified thus allowing for accurate ranging and reliable communication. For prototyping and experimentation, the work will make use of commercial off the shelf radios along with open-source implementations of network protocols commonly used in IoT applications such as those frameworks provided by the Zephyr project or Contiki-NG open source projects. This work will engage with Toshiba units for exploitation of the produced technologies. This project falls within the EPSRC Engineering research Area, "ICT Networks and Distributed Systems".