Fig. 5: Basic metrics, reconfigurability and performance evaluation. | Nature Communications

Fig. 5: Basic metrics, reconfigurability and performance evaluation.

From: Demonstration of high-reconfigurability and low-power strong physical unclonable function empowered by FeFET cycle-to-cycle variation and charge-domain computing

Fig. 5

a A response vector is generated by inputting multiple challenge vectors in our experimental measurements. b Entropy and HW show high uniformity. c HDinter and HDintra (i.e., BER) results highlight the high uniqueness and repeatability of our PUF. d The measured HDreconfig indicates near-ideal uniqueness across reconfiguration cycles. e The correlation matrix of the responses over 10 reconfiguration cycles further demonstrates good reconfigurability. The diagonal entries are plotted as 0 for visual clarity. f Layout of the proposed 2FeFET-1C cell with the MOM capacitor integrated on top of FeFETs via the BEOL process. The cell area is 1 μm2 using 200 nm × 100 nm FeFETs and a 0.5fF capacitor at 28 nm node. g The readout energy peaks when half of the PUF cells output \({V}_{{{{\rm{DD}}}}}\) due to the maximum charge redistribution. Increasing challenge vector length, \({V}_{{{{\rm{DD}}}}}\) or \({V}_{{{{\rm{R}}}}}\) all leads to higher energy consumption. Source data are provided as a Source Data file.

Back to article page