Periodic Reporting for period 1 - UBIGIoT (Ultra-Low Design-Effort, Energy-Efficient and Battery-Indifferent Sensor Node for the Green Internet of Things)
Reporting period: 2022-12-01 to 2024-11-30
Relentless advancements in circuits for battery-less silicon systems have recently enabled new levels of sensor node miniaturization, cost, and device lifespan, removing the limitations imposed by batteries.
The UBIGIoT project intends to develop a comprehensive and energy-autonomous sensor node able to monitor parameters (i.e. temperature, humidity, acceleration, etc.) and transfer wireless data. Multiple types of ambient energy sources can be converted to usable power. Design strategies for the electronic circuit connected to the energy harvester terminals will be developed with the aim of maximizing the energy extracted under real operating conditions. The power management circuits proposed in the literature for the maximization of the harvested energy are developed under the assumption of steady-state conditions of the ambient source (constant wind, sinusoidal accelerations, etc.), while the proposed research activity will tackle the challenge of maximizing the extracted energy under real life sporadic conditions of ambient sources.
The energy source addressed in this first part of the project is Radio Frequency (RF) energy provided by conventional Wi-Fi transmitters. The adoption of the 802.11b Wi-Fi standard is preferred to BLE, LTE, and 5G since these standards do not have widely distributed infrastructure in most of the indoor spaces (e.g. home, offices, and gyms), which makes rapid low-cost deployment quite difficult Being the available RF harvested power in the deep sub-µW/mm2 range, peak power budgets are restricted to the 1-µW range or less at millimeter-scale form factors, requiring aggressive system power reductions. At the battery-less system level, wireless transmission (e.g. Wi-Fi) generally sets the limit to further system power reductions in view of their dominant peak power. The milliwatt-range power consumption of state-of-the-art conventional Wi-Fi transmitters was recently reduced to tens of µWs through backscattering architectures. Indeed, the latter suppress gigahertz-range PLL and power amplifier by properly reflecting an incident wave from a tone generator. Complete PLL elimination has recently reduced Wi-Fi transmitter power to µWs, thanks to eventdriven temperature-compensated ring oscillator-based local oscillators.
The goal of the research activity was to enable a new class of Wi-Fi backscattering transmitters with power in the sub-µW range by removing the local oscillator power while simultaneously for the first time in the literature reusing the incoming RF signal for backscattering communications, harvesting, and sensor-less position/motion sensing.
The UBIGIoT project intends to develop a comprehensive and energy-autonomous sensor node able to monitor parameters (i.e. temperature, humidity, acceleration, etc.) and transfer wireless data. Multiple types of ambient energy sources can be converted to usable power. Design strategies for the electronic circuit connected to the energy harvester terminals will be developed with the aim of maximizing the energy extracted under real operating conditions. The power management circuits proposed in the literature for the maximization of the harvested energy are developed under the assumption of steady-state conditions of the ambient source (constant wind, sinusoidal accelerations, etc.), while the proposed research activity will tackle the challenge of maximizing the extracted energy under real life sporadic conditions of ambient sources.
The energy source addressed in this first part of the project is Radio Frequency (RF) energy provided by conventional Wi-Fi transmitters. The adoption of the 802.11b Wi-Fi standard is preferred to BLE, LTE, and 5G since these standards do not have widely distributed infrastructure in most of the indoor spaces (e.g. home, offices, and gyms), which makes rapid low-cost deployment quite difficult Being the available RF harvested power in the deep sub-µW/mm2 range, peak power budgets are restricted to the 1-µW range or less at millimeter-scale form factors, requiring aggressive system power reductions. At the battery-less system level, wireless transmission (e.g. Wi-Fi) generally sets the limit to further system power reductions in view of their dominant peak power. The milliwatt-range power consumption of state-of-the-art conventional Wi-Fi transmitters was recently reduced to tens of µWs through backscattering architectures. Indeed, the latter suppress gigahertz-range PLL and power amplifier by properly reflecting an incident wave from a tone generator. Complete PLL elimination has recently reduced Wi-Fi transmitter power to µWs, thanks to eventdriven temperature-compensated ring oscillator-based local oscillators.
The goal of the research activity was to enable a new class of Wi-Fi backscattering transmitters with power in the sub-µW range by removing the local oscillator power while simultaneously for the first time in the literature reusing the incoming RF signal for backscattering communications, harvesting, and sensor-less position/motion sensing.
The outcome of the research activity is the first sub-microwatt 802.11b backscattering transmitter to enable the reuse of the same incident wave for three purposes: RF harvesting, backscattering communications, and position/motion sensing. The activity has been carried out in strong cooperation with the other partners and, in particular, with the National University of Singapore.
The prototype, fabricated in a standard 180-nm CMOS technology and assembled on a custom PCB, has been presented and experimentally validated under all process corner wafers (corners not considered in the prior art). Its architecture reuses the same incident wave for RF harvesting, backscattering communications, clock extraction, and position/motion sensing. Such reuse removes the battery, any explicit physical harvester, any power-hungry on-chip local oscillator, and off-chip motion sensor (e.g. MEMS) for aggressive miniaturization, unrestricted device lifespan, low cost, and low maintenance cost for ubiquitous adoption. The microwatt power wall for Wi-Fi transmitters is broken for the first time via local oscillator elimination, as achieved by extracting its frequency through second-order intermodulation of a two-tone incident wave. The two-tone scheme also enables a cumulative harvesting/transmission/sensing sensitivity down to Pmin ∼ −19 dBm. Position/motion sensing is enabled by using the harvested voltage as a proxy for the received signal strength (RSS), allowing to sense the chip location with respect to the tone generator(s) shared across tags in indoor neighborhoods.
The prototype, fabricated in a standard 180-nm CMOS technology and assembled on a custom PCB, has been presented and experimentally validated under all process corner wafers (corners not considered in the prior art). Its architecture reuses the same incident wave for RF harvesting, backscattering communications, clock extraction, and position/motion sensing. Such reuse removes the battery, any explicit physical harvester, any power-hungry on-chip local oscillator, and off-chip motion sensor (e.g. MEMS) for aggressive miniaturization, unrestricted device lifespan, low cost, and low maintenance cost for ubiquitous adoption. The microwatt power wall for Wi-Fi transmitters is broken for the first time via local oscillator elimination, as achieved by extracting its frequency through second-order intermodulation of a two-tone incident wave. The two-tone scheme also enables a cumulative harvesting/transmission/sensing sensitivity down to Pmin ∼ −19 dBm. Position/motion sensing is enabled by using the harvested voltage as a proxy for the received signal strength (RSS), allowing to sense the chip location with respect to the tone generator(s) shared across tags in indoor neighborhoods.
Compared to previous arts adopting intermodulation-based clock extraction circuitry, the proposed solution does not require any filters, any noisy and power-hungry injection-locked ring oscillators, and any limiting amplifiers. Thanks to its simplicity clock extraction circuit allows, hence, to aggressively reduce the power consumption down to a few hundreds of (or less) nanowatts. In addition, the circuit is digitally adjustable to always meet the jitter/phase noise requirements of the adopted 802.11b Wi-Fi standard against process corner variations.
The two-tone incident wave is also exploited, for the first time, for Wi-Fi backscattering communication. Compared to single-tone (i.e. from PID) backscattering radios the adoption of two input tones gives an improvement of more than 3 dB on the signal-to-noise ratio (SNR) at the receiver end and, hence, an increased working tag-to-receiver distance range (up to 25 m) within standard compliant BER ≤ 10-5.
Inexpensive rectification and amplification of the received signal strength (RSS) voltage is obtained via a novel and efficient crossed-NMOS Dickson charge pump exploited for the RF energy harvesting operation and fully battery-less. Thanks to this topology that mixes the benefits of cross-coupled charge pumps (e.g. low-input sensitivity) with Dickson topologies (e.g. higher output current and higher efficiency levels), the best-in-class 1-V sensitivity is achieved with an input power down to -20.3 dBm. The 43% peak power conversion efficiency (PCE) is reached with the lowest input power compared to similar works dealing with the 2.4-GHz band RF-to-dc converters. The harvested voltage is used as a proxy for RSS to quantify the chip location with respect to the two-tone generator(s). The achieved resolution, also under interference signals, goes down to 16 cm, which is 3× better than the previous RSS indicator (RSSI) and channel state information (CSI) position detection systems.
Finally, the adopted clock extraction and two-tone backscattering communication open the road to the adaptability of the proposed battery-less tag to other standards such as BLE, LTE, and 5G.
The two-tone incident wave is also exploited, for the first time, for Wi-Fi backscattering communication. Compared to single-tone (i.e. from PID) backscattering radios the adoption of two input tones gives an improvement of more than 3 dB on the signal-to-noise ratio (SNR) at the receiver end and, hence, an increased working tag-to-receiver distance range (up to 25 m) within standard compliant BER ≤ 10-5.
Inexpensive rectification and amplification of the received signal strength (RSS) voltage is obtained via a novel and efficient crossed-NMOS Dickson charge pump exploited for the RF energy harvesting operation and fully battery-less. Thanks to this topology that mixes the benefits of cross-coupled charge pumps (e.g. low-input sensitivity) with Dickson topologies (e.g. higher output current and higher efficiency levels), the best-in-class 1-V sensitivity is achieved with an input power down to -20.3 dBm. The 43% peak power conversion efficiency (PCE) is reached with the lowest input power compared to similar works dealing with the 2.4-GHz band RF-to-dc converters. The harvested voltage is used as a proxy for RSS to quantify the chip location with respect to the two-tone generator(s). The achieved resolution, also under interference signals, goes down to 16 cm, which is 3× better than the previous RSS indicator (RSSI) and channel state information (CSI) position detection systems.
Finally, the adopted clock extraction and two-tone backscattering communication open the road to the adaptability of the proposed battery-less tag to other standards such as BLE, LTE, and 5G.