Impact of Front-End Wearout Mechanisms on the Performance of a Ring Oscillator-Based Thermal Sensor

Abstract

This work studies how process/environmental parameters and front-end wearout mechanisms affect the performance of a FinFET-based thermal sensor which can be applied for on-chip temperature monitoring and temperature tracking for healthcare. This work has considered process/environmental parameters, such as gate length, supply voltage (VDD), bank capacitance, and temperature, and front-end wearout mechanisms, including bias temperature instability (BTI), hot carrier injection (HCI), and random telegraph noise (RTN). The impact of wearout mechanisms on each module was checked, and it was found that the ring oscillator is the most sensitive part. It was found that a larger gate length and lower VDD cause lower digital output values (indicating a lower operating frequency) and less power consumption. Wearout causes the sensor’s digital output values to decrease, and it causes more deviation in the digital output because of wearout induced parameter variations. It was found that linear recalibration is not a perfect solution for wearout induced parameter deviations of the digital output.