The ULPIoT project strived successfully in increasing at most the level of automation in the design flow. The main IC analog building blocks have been implemented by means of a fully synthesizable design approach. Several test-chip have been taped-out in order to validate the innovative ideas thought test-chip measurements.
The main ULPIoT breakthroughs are the first fully synthesizable Digital-to-analog Converters (DACs) able to be designed with a fully automated digital design flow and the first pW-range wake-up oscillator for IoT sensor nodes able to operate from 0.3V to 1.8V avoidings the traditional need of additional power-hungry voltage regulation.
The proposed DACs significantly reduce the design effort compared to conventional analog design styles as they are based on a digital standard cells approach. This enables digital-like shrinkage across CMOS generations and hence low area at down-scaled technologies, as well as operation down to near-threshold voltages.
Three different versions have been proposed. The first DAC with a nominal resolution of 12-bit exhibits a graceful degradation under-voltage/frequency over scaling; The others are 16-bit and 12-bit versions pointing-out respectively only performance and area reduction.
All these characteristics make the proposed solutions highly suitable for the IoT nodes.
Slow oscillators that periodically wake up the sensor nodes are fundamental building blocks in the emerging Internet of Things (IoT). The proposed oscillator can work stand-alone, with no further voltage or current reference, and with a low-frequency sensitivity to the supply voltage. Thus, the actual power consumed by the always-on oscillator, and hence by the entire system is drastically reduced.
Another ULPIoT outcome is based on the automatized design approach employed to conceive analog comparison using only standard cells (logic gate). Also, digital-based operational amplifiers have been conceived.
Moreover, fully synthesizable Current- and Voltage-input ADCs architectures have been designed and tested.
As a further remark, the suitability of the employed methodologies for the processing in Deep Learning application has been explored.
The ULPIoT results have been presented in the premier conferences of “Circuit and System” (CAS) and “Solid State Circuit” (SSC) Societies (i.e. ISCAS and VLSI Symposium). The respective more extended works have been also published in top journal papers as Transaction of CAS and Journal of SSC.
As further dissemination activities, MSCA events have been held (
https://sites.google.com/view/ulpiot/dissemination/seminars(se abrirá en una nueva ventana)). Moreover, many technical seminars and tutorials at international conferences have been also held (
https://sites.google.com/view/ulpiot/dissemination/tutorials(se abrirá en una nueva ventana)).
As result of the scientific quality of the ULPIoT results, several international scientific presses have reported the value of the research outcomes developed in Singapore (as shown in the up-to-date ULPIoT’s press review:
https://sites.google.com/view/ulpiot/press(se abrirá en una nueva ventana)).
Moreover, a Semiconductor company (i.e. STMicroelectronics) has already expressed interest in ULPIoT results for the development of commercial and concrete IoT applications.
All the above-mentioned achievements are certainly proof of a remarkable scientific value of the ULPIoT project and commitment of the MSCA Fellow.