Polymorphic radios: a new design paradigm for ultra-low power communication

Polymorphic radios: a new design paradigm for ultra-low power communication
复制标题

多态无线电:超低功耗通信的新设计范例

DOI:
10.1145/3230543.3230571
复制
发表时间:
2018
期刊:
ACM SIGCOMM
影响因子:
--
通讯作者:
Ganesan, Deepak
Ganesan, Deepak
中科院分区:
--
文献类型:
--
作者:
Rostami, Mohammad;Gummeson, Jeremy;Kiaghadi, Ali;Ganesan, Deepak

文献摘要

相似文献

占空比已经成为优化低功率无线电功耗的主要方法,特别是对于以小突发零星传输的传感器。但对于需要将数据传输到云端以执行复杂机器学习模型的可穿戴传感器(如伊穆斯、麦克风和成像器)的高速率传感器数据的应用,负载循环不太适合。我们认为,如果我们能够利用短距离设置中的信道动态特性,那么优化低功耗无线电的空间很大。然而,我们在设计在μ W和mW之间的功率电平下有效的无线电时面临挑战,以利用良好信号强度的周期并灵活地处理由身体运动引起的高度动态信道。为了实现这一目标,我们提出了无线电多态性,无线电架构与紧密集成的无源和有源组件,使我们能够把高信道动态我们的优势。我们在网络堆栈中以各种方式利用被动模式,从最大限度地减少数据传输和控制开销到改进速率选择和实现信道感知的机会传输。我们在一个完整的硬件-软件原型Morpho中实例化了我们的设计,并在不同的场景和应用中展示了效率的提高。
Duty-cycling has emerged as the predominant method for optimizing power consumption of low-power radios, particularly for sensors that transmit sporadically in small bursts. But duty-cycling is a poor fit for applications involving high-rate sensor data from wearable sensors such as IMUs, microphones, and imagers that need to stream data to the cloud to execute sophisticated machine learning models.We argue that there is significant room to optimize low-power radios if we can take advantage of channel dynamics in short-range settings. However, we face challenges in designing radios that are efficient at power levels betweenμWs and mWs to take advantage of periods of good signal strength and nimble to deal with highly dynamic channels resulting from body movements. To achieve this, we propose radio polymorphism, a radio architecture with tightly integrated passive and active components that allows us to turn high channel dynamics to our advantage. We leverage passive modes in myriad ways within the network stack, from minimizing data transfer and control overheads to improving rate selection and enabling channel-aware opportunistic transmission. We instantiate our design in a full hardware-software prototype, Morpho, and demonstrate up to an order of improvement in efficiency across diverse scenarios and applications.