基于压缩采样和能量自治的物联网无线感知接口研究

Intelligent tags with multi-parameter sensing are the key elements in the context of the Internet of Things (IoT), potentially with a wide spectrum of applications such as in wearable healthcare and food safety. Concurrent design of ultra-low-power (ULP) circuit，system and communication has attracted research attention, aiming for the realization of ultra-tinny IoT devices under energy-autonomy constraint. The objective of this project is to explore design space and to develop ULP chipset for sub-100uW powered IoT devices, through trade-offs for communication, circuit and system architectures. Smart power management scheme with adaptive energy-scavenging, self-aware sleep and burst mode will be exploited to facilitate energy autonomous operation. Compressed sampling based analog-to-information interface will be investigated that not only simplifies the sensor front-end but also reduces raw data to be transmitted thus significantly save energy. As a proof of concept, a self-powered sensor tag with UWB/UHF asymmetric communication and compressed sampling interface will be demonstrated in a 65 nm (or below) CMOS process.

多参数和多传感器集成的智能感知节点是物联网的核心器件，将广泛用于可穿戴健康医疗与食品安全等领域。通过电路、系统和通信链路的超低功耗协同设计和优化，实现能量自治的智能无线感知，是物联网领域的一项核心技术并成为新的国际研究热点。本项目研究100uW及以下超低功耗物联网无线感知接口电路和系统的设计方法，并通过自适应无线能量收集、自知驱动的系统休眠和突发模式交替工作等智慧能源管理方法实现系统的能量自治。项目将深入研究对传感器模拟信号的压缩采样技术原理和信号调理，减少原始数据的处理和传输能量负担，并采用脉冲超宽频（UWB）结合超高频（UHF）非对称链路的短距离无线通信技术，大大提高传感信号从采样到传输的总体能量效率，实现物联网无线感知接口的能量自治，并在65nm或以下CMOS技术节点的工艺上进行电路验证。本项目所述超低功耗方法适用于感知稀疏性强的自然信号，对物联网应用意义重大。

面向物联网应用对集成电路与系统高能效、智能化的需求，本项目研究了100μW及以下超低功耗物联网无线感知接口电路和系统的设计方法，重点研究了基于时域信号稀疏表达的编码与调制方法、休眠和突发交替模式的高能效通信协议、基于压缩采样与信号调理的多通道传感接口、“算法-系统-电路“协同设计方法等关键技术。项目提出了基于脉冲超宽频（UWB）结合超高频（UHF）非对称链路的无线感知架构，实现了能量自治的通信-传感一体化系统与芯片；项目以心电信号采集和处理为例，采用跨层次优化方法，实现了超低功耗可穿戴心电信号监护片上系统芯片；项目以多电极腔内心电监测为例，设计了一款集成于智能导管的压缩采样专用集成电路。本项目执行了预期的各项研究计划，产出了数篇高水平的学术论文，完成了系统仿真与优化模型框架、核心低功耗电路模块的设计与仿真、片上系统芯片的流片测试，并在健康物联网、广域物联网等典型应用场景中进行了应用验证。项目所突破的关键技术对物联网智能芯片与系统将发挥积极的推动作用。