Energy-Efficient Integrated Circuits and Systems to Bridge the Gap between Power and Performance in Wireless Sensor Networks

Abstract

The vision of the Internet of Things (IoT) is to create an intelligent and invisible worldwide network of interconnected objects that can be sensed, controlled, and programmed. As we move towards ubiquitous deployment of IoT devices and in order to enable 50 billion connected nodes as predicted by technology leaders, ultra-low power consumption of wireless sensor nodes becomes critical. This is while existing low power approaches in wireless integrated circuits design often sacrifice performance metrics, such as noise, receiver sensitivity, transmitter efficiency, etc. which is not desirable for many IoT applications. Motivated by the ultra-low power requirements of IoT applications, this research presents low power circuit and system design methodologies to bridge the gap between performance and ultra-low power consumption in wireless sensor nodes for future IoT applications.

This work introduces an array of innovative design approaches at the RF and analog circuit level, as well as system architectures for wireless networks that enable ultra-low power consumption at the IoT sensor node, while providing high sensitivity and power efficiency values. This includes circuit designs for a long-range RF transmitter and receiver, along with front-end amplifiers for sensor read-out circuits. The proposed receiver and transmitter are designed to allow data rate agile, narrowband, and long-range IoT communications for Low Power Wide Area Networks (LPWAN) applications. More specifically, the receiver that operates at the Multi-Use Radio Service (MURS) frequency band, consumes only 152µW to achieve a sensitivity of -99dBm at 5kbps, making it capable of communication ranges on the order of tens of kilometers. Another contribution of this work is a novel transmitter architecture.

The transmitter utilizes an efficient system architecture with low power design methods, both in baseband data generation and RF circuits. It delivers a peak efficiency of 41% at a peak output power of 0dBm at 5kbps BPSK modulated data transmission, and also enables 16QAM OFDM transmission with a data rate of 384kbps, making it a suitable solution for remote IoT connectivity in multipath rich environments. In addition to the circuit level implementation of the presented ultra-low power MURS band transceiver, a narrowband transmission scheme along with a system level design and link budget analysis in the MURS band is also presented. The system analysis and transmission scheme in the MURS band shows that it can serve as an ultra-low power and long-range alternative for LPWAN applications.

Finally, an ultra-low power low-noise health monitoring analog front-end (AFE) is developed and presented in this work, that demonstrates the feasibility of < 100 nW AFE for continuous ECG monitoring applications. The low-noise 68nW AFE was also integrated on a self-powered physiological monitoring System on Chip (SoC) that was used to capture ECG bio-signals from human subjects.